release 2014.02.19

This is a pseudo extractor that can be used to quickly look up test URLs, or test without the test harness.

.gitignore

@@ -23,6 +23,8 @@ updates_key.pem
 *.vtt
 *.flv
 *.mp4
+*.m4a
+*.m4v
 *.part
 test/testdata
 .tox

README.md

@@ -16,109 +16,145 @@ which means you can modify it, redistribute it or use it however you like.
 # OPTIONS
     -h, --help                       print this help text and exit
     --version                        print program version and exit
-    -U, --update               update this program to latest version. Make sure
+    -U, --update                     update this program to latest version. Make
-                               that you have sufficient permissions (run with
+                                     sure that you have sufficient permissions
-                               sudo if needed)
+                                     (run with sudo if needed)
-    -i, --ignore-errors        continue on download errors, for example to to
+    -i, --ignore-errors              continue on download errors, for example to
-                               skip unavailable videos in a playlist
+                                     to skip unavailable videos in a playlist
     --abort-on-error                 Abort downloading of further videos (in the
-                               playlist or the command line) if an error occurs
+                                     playlist or the command line) if an error
+                                     occurs
     --dump-user-agent                display the current browser identification
     --user-agent UA                  specify a custom user agent
-    --referer REF              specify a custom referer, use if the video access
+    --referer REF                    specify a custom referer, use if the video
-                               is restricted to one domain
+                                     access is restricted to one domain
-    --list-extractors          List all supported extractors and the URLs they
+    --list-extractors                List all supported extractors and the URLs
-                               would handle
+                                     they would handle
-    --extractor-descriptions   Output descriptions of all supported extractors
+    --extractor-descriptions         Output descriptions of all supported
-    --proxy URL                Use the specified HTTP/HTTPS proxy. Pass in an
+                                     extractors
-                               empty string (--proxy "") for direct connection
+    --proxy URL                      Use the specified HTTP/HTTPS proxy. Pass in
+                                     an empty string (--proxy "") for direct
+                                     connection
     --no-check-certificate           Suppress HTTPS certificate validation.
-    --cache-dir DIR            Location in the filesystem where youtube-dl can
+    --cache-dir DIR                  Location in the filesystem where youtube-dl
-                               store downloaded information permanently. By
+                                     can store some downloaded information
-                               default $XDG_CACHE_HOME/youtube-dl or ~/.cache
+                                     permanently. By default $XDG_CACHE_HOME
-                               /youtube-dl .
+                                     /youtube-dl or ~/.cache/youtube-dl . At the
+                                     moment, only YouTube player files (for
+                                     videos with obfuscated signatures) are
+                                     cached, but that may change.
     --no-cache-dir                   Disable filesystem caching
-    --bidi-workaround          Work around terminals that lack bidirectional
+    --socket-timeout None            Time to wait before giving up, in seconds
-                               text support. Requires fribidi executable in PATH
+    --bidi-workaround                Work around terminals that lack
+                                     bidirectional text support. Requires bidiv
+                                     or fribidi executable in PATH
+    --default-search PREFIX          Use this prefix for unqualified URLs. For
+                                     example "gvsearch2:" downloads two videos
+                                     from google videos for  youtube-dl "large
+                                     apple". By default (with value "auto")
+                                     youtube-dl guesses.
+    --ignore-config                  Do not read configuration files. When given
+                                     in the global configuration file /etc
+                                     /youtube-dl.conf: do not read the user
+                                     configuration in ~/.config/youtube-dl.conf
+                                     (%APPDATA%/youtube-dl/config.txt on
+                                     Windows)
 
 ## Video Selection:
     --playlist-start NUMBER          playlist video to start at (default is 1)
     --playlist-end NUMBER            playlist video to end at (default is last)
-    --match-title REGEX        download only matching titles (regex or caseless
+    --match-title REGEX              download only matching titles (regex or
-                               sub-string)
+                                     caseless sub-string)
     --reject-title REGEX             skip download for matching titles (regex or
                                      caseless sub-string)
     --max-downloads NUMBER           Abort after downloading NUMBER files
-    --min-filesize SIZE        Do not download any videos smaller than SIZE
+    --min-filesize SIZE              Do not download any videos smaller than
+                                     SIZE (e.g. 50k or 44.6m)
+    --max-filesize SIZE              Do not download any videos larger than SIZE
                                      (e.g. 50k or 44.6m)
-    --max-filesize SIZE        Do not download any videos larger than SIZE (e.g.
-                               50k or 44.6m)
     --date DATE                      download only videos uploaded in this date
-    --datebefore DATE          download only videos uploaded before this date
+    --datebefore DATE                download only videos uploaded on or before
-    --dateafter DATE           download only videos uploaded after this date
+                                     this date (i.e. inclusive)
-    --min-views COUNT          Do not download any videos with less than COUNT
+    --dateafter DATE                 download only videos uploaded on or after
-                               views
+                                     this date (i.e. inclusive)
-    --max-views COUNT          Do not download any videos with more than COUNT
+    --min-views COUNT                Do not download any videos with less than
-                               views
+                                     COUNT views
+    --max-views COUNT                Do not download any videos with more than
+                                     COUNT views
     --no-playlist                    download only the currently playing video
-    --age-limit YEARS          download only videos suitable for the given age
+    --age-limit YEARS                download only videos suitable for the given
-    --download-archive FILE    Download only videos not listed in the archive
+                                     age
-                               file. Record the IDs of all downloaded videos in
+    --download-archive FILE          Download only videos not listed in the
-                               it.
+                                     archive file. Record the IDs of all
+                                     downloaded videos in it.
+    --include-ads                    Download advertisements as well
+                                     (experimental)
+    --youtube-include-dash-manifest  Try to download the DASH manifest on
+                                     YouTube videos (experimental)
 
 ## Download Options:
-    -r, --rate-limit LIMIT     maximum download rate in bytes per second (e.g.
+    -r, --rate-limit LIMIT           maximum download rate in bytes per second
-                               50K or 4.2M)
+                                     (e.g. 50K or 4.2M)
     -R, --retries RETRIES            number of retries (default is 10)
     --buffer-size SIZE               size of download buffer (e.g. 1024 or 16K)
                                      (default is 1024)
-    --no-resize-buffer         do not automatically adjust the buffer size. By
+    --no-resize-buffer               do not automatically adjust the buffer
-                               default, the buffer size is automatically resized
+                                     size. By default, the buffer size is
-                               from an initial value of SIZE.
+                                     automatically resized from an initial value
+                                     of SIZE.
 
 ## Filesystem Options:
     -t, --title                      use title in file name (default)
     --id                             use only video ID in file name
     -l, --literal                    [deprecated] alias of --title
     -A, --auto-number                number downloaded files starting from 00000
-    -o, --output TEMPLATE      output filename template. Use %(title)s to get
+    -o, --output TEMPLATE            output filename template. Use %(title)s to
-                               the title, %(uploader)s for the uploader name,
+                                     get the title, %(uploader)s for the
-                               %(uploader_id)s for the uploader nickname if
+                                     uploader name, %(uploader_id)s for the
-                               different, %(autonumber)s to get an automatically
+                                     uploader nickname if different,
-                               incremented number, %(ext)s for the filename
+                                     %(autonumber)s to get an automatically
-                               extension, %(format)s for the format description
+                                     incremented number, %(ext)s for the
-                               (like "22 - 1280x720" or "HD"),%(format_id)s for
+                                     filename extension, %(format)s for the
-                               the unique id of the format (like Youtube's
+                                     format description (like "22 - 1280x720" or
-                               itags: "137"),%(upload_date)s for the upload date
+                                     "HD"), %(format_id)s for the unique id of
+                                     the format (like Youtube's itags: "137"),
+                                     %(upload_date)s for the upload date
                                      (YYYYMMDD), %(extractor)s for the provider
-                               (youtube, metacafe, etc), %(id)s for the video id
+                                     (youtube, metacafe, etc), %(id)s for the
-                               , %(playlist)s for the playlist the video is in,
+                                     video id, %(playlist)s for the playlist the
-                               %(playlist_index)s for the position in the
+                                     video is in, %(playlist_index)s for the
-                               playlist and %% for a literal percent. Use - to
+                                     position in the playlist and %% for a
-                               output to stdout. Can also be used to download to
+                                     literal percent. Use - to output to stdout.
-                               a different directory, for example with -o '/my/d
+                                     Can also be used to download to a different
-                               ownloads/%(uploader)s/%(title)s-%(id)s.%(ext)s' .
+                                     directory, for example with -o '/my/downloa
-    --autonumber-size NUMBER   Specifies the number of digits in %(autonumber)s
+                                     ds/%(uploader)s/%(title)s-%(id)s.%(ext)s' .
-                               when it is present in output filename template or
+    --autonumber-size NUMBER         Specifies the number of digits in
-                               --auto-number option is given
+                                     %(autonumber)s when it is present in output
-    --restrict-filenames       Restrict filenames to only ASCII characters, and
+                                     filename template or --auto-number option
-                               avoid "&" and spaces in filenames
+                                     is given
-    -a, --batch-file FILE      file containing URLs to download ('-' for stdin)
+    --restrict-filenames             Restrict filenames to only ASCII
+                                     characters, and avoid "&" and spaces in
+                                     filenames
+    -a, --batch-file FILE            file containing URLs to download ('-' for
+                                     stdin)
     --load-info FILE                 json file containing the video information
-                               (created with the "--write-json" option
+                                     (created with the "--write-json" option)
     -w, --no-overwrites              do not overwrite files
-    -c, --continue             force resume of partially downloaded files. By
+    -c, --continue                   force resume of partially downloaded files.
-                               default, youtube-dl will resume downloads if
+                                     By default, youtube-dl will resume
-                               possible.
+                                     downloads if possible.
-    --no-continue              do not resume partially downloaded files (restart
+    --no-continue                    do not resume partially downloaded files
-                               from beginning)
+                                     (restart from beginning)
-    --cookies FILE             file to read cookies from and dump cookie jar in
+    --cookies FILE                   file to read cookies from and dump cookie
+                                     jar in
     --no-part                        do not use .part files
-    --no-mtime                 do not use the Last-modified header to set the
+    --no-mtime                       do not use the Last-modified header to set
-                               file modification time
+                                     the file modification time
-    --write-description        write video description to a .description file
+    --write-description              write video description to a .description
+                                     file
     --write-info-json                write video metadata to a .info.json file
-    --write-annotations        write video annotations to a .annotation file
+    --write-annotations              write video annotations to a .annotation
+                                     file
     --write-thumbnail                write thumbnail image to disk
 
 ## Verbosity / Simulation Options:
@@ -139,57 +175,74 @@ which means you can modify it, redistribute it or use it however you like.
     --no-progress                    do not print progress bar
     --console-title                  display progress in console titlebar
     -v, --verbose                    print various debugging information
-    --dump-intermediate-pages  print downloaded pages to debug problems(very
+    --dump-intermediate-pages        print downloaded pages to debug problems
-                               verbose)
+                                     (very verbose)
-    --write-pages              Write downloaded intermediary pages to files in
+    --write-pages                    Write downloaded intermediary pages to
-                               the current directory to debug problems
+                                     files in the current directory to debug
+                                     problems
+    --print-traffic                  Display sent and read HTTP traffic
 
 ## Video Format Options:
     -f, --format FORMAT              video format code, specify the order of
-                               preference using slashes: "-f 22/17/18". "-f mp4"
+                                     preference using slashes: "-f 22/17/18".
-                               and "-f flv" are also supported
+                                     "-f mp4" and "-f flv" are also supported.
+                                     You can also use the special names "best",
+                                     "bestaudio", "worst", and "worstaudio". By
+                                     default, youtube-dl will pick the best
+                                     quality.
     --all-formats                    download all available video formats
-    --prefer-free-formats      prefer free video formats unless a specific one
+    --prefer-free-formats            prefer free video formats unless a specific
-                               is requested
+                                     one is requested
     --max-quality FORMAT             highest quality format to download
-    -F, --list-formats         list all available formats (currently youtube
+    -F, --list-formats               list all available formats
-                               only)
 
 ## Subtitle Options:
     --write-sub                      write subtitle file
-    --write-auto-sub           write automatic subtitle file (youtube only)
+    --write-auto-sub                 write automatic subtitle file (youtube
-    --all-subs                 downloads all the available subtitles of the
-                               video
-    --list-subs                lists all available subtitles for the video
-    --sub-format FORMAT        subtitle format (default=srt) ([sbv/vtt] youtube
                                      only)
-    --sub-lang LANGS           languages of the subtitles to download (optional)
+    --all-subs                       downloads all the available subtitles of
-                               separated by commas, use IETF language tags like
+                                     the video
-                               'en,pt'
+    --list-subs                      lists all available subtitles for the video
+    --sub-format FORMAT              subtitle format (default=srt) ([sbv/vtt]
+                                     youtube only)
+    --sub-lang LANGS                 languages of the subtitles to download
+                                     (optional) separated by commas, use IETF
+                                     language tags like 'en,pt'
 
 ## Authentication Options:
     -u, --username USERNAME          account username
     -p, --password PASSWORD          account password
     -n, --netrc                      use .netrc authentication data
-    --video-password PASSWORD  video password (vimeo only)
+    --video-password PASSWORD        video password (vimeo, smotri)
 
 ## Post-processing Options:
-    -x, --extract-audio        convert video files to audio-only files (requires
+    -x, --extract-audio              convert video files to audio-only files
-                               ffmpeg or avconv and ffprobe or avprobe)
+                                     (requires ffmpeg or avconv and ffprobe or
-    --audio-format FORMAT      "best", "aac", "vorbis", "mp3", "m4a", "opus", or
+                                     avprobe)
-                               "wav"; best by default
+    --audio-format FORMAT            "best", "aac", "vorbis", "mp3", "m4a",
-    --audio-quality QUALITY    ffmpeg/avconv audio quality specification, insert
+                                     "opus", or "wav"; best by default
-                               a value between 0 (better) and 9 (worse) for VBR
+    --audio-quality QUALITY          ffmpeg/avconv audio quality specification,
-                               or a specific bitrate like 128K (default 5)
+                                     insert a value between 0 (better) and 9
-    --recode-video FORMAT      Encode the video to another format if necessary
+                                     (worse) for VBR or a specific bitrate like
-                               (currently supported: mp4|flv|ogg|webm)
+                                     128K (default 5)
-    -k, --keep-video           keeps the video file on disk after the post-
+    --recode-video FORMAT            Encode the video to another format if
-                               processing; the video is erased by default
+                                     necessary (currently supported:
-    --no-post-overwrites       do not overwrite post-processed files; the post-
+                                     mp4|flv|ogg|webm)
-                               processed files are overwritten by default
+    -k, --keep-video                 keeps the video file on disk after the
+                                     post-processing; the video is erased by
+                                     default
+    --no-post-overwrites             do not overwrite post-processed files; the
+                                     post-processed files are overwritten by
+                                     default
     --embed-subs                     embed subtitles in the video (only for mp4
                                      videos)
-    --add-metadata             add metadata to the files
+    --add-metadata                   write metadata to the video file
+    --xattrs                         write metadata to the video file's xattrs
+                                     (using dublin core and xdg standards)
+    --prefer-avconv                  Prefer avconv over ffmpeg for running the
+                                     postprocessors (default)
+    --prefer-ffmpeg                  Prefer ffmpeg over avconv for running the
+                                     postprocessors
 
 # CONFIGURATION
 
@@ -228,9 +281,14 @@ Videos can be filtered by their upload date using the options `--date`, `--dateb
  
 Examples:
 
-	$ youtube-dl --dateafter now-6months #will only download the videos uploaded in the last 6 months
+    # Download only the videos uploaded in the last 6 months
-	$ youtube-dl --date 19700101 #will only download the videos uploaded in January 1, 1970
+    $ youtube-dl --dateafter now-6months
-	$ youtube-dl --dateafter 20000101 --datebefore 20100101 #will only download the videos uploaded between 2000 and 2010
+
+    # Download only the videos uploaded on January 1, 1970
+    $ youtube-dl --date 19700101
+
+    $ # will only download the videos uploaded in the 200x decade
+    $ youtube-dl --dateafter 20000101 --datebefore 20091231
 
 # FAQ
 
@@ -275,11 +333,31 @@ Since June 2012 (#342) youtube-dl is packed as an executable zipfile, simply unz
 
 To run the exe you need to install first the [Microsoft Visual C++ 2008 Redistributable Package](http://www.microsoft.com/en-us/download/details.aspx?id=29).
 
-# COPYRIGHT
+# DEVELOPER INSTRUCTIONS
 
-youtube-dl is released into the public domain by the copyright holders.
+Most users do not need to build youtube-dl and can [download the builds](http://rg3.github.io/youtube-dl/download.html) or get them from their distribution.
 
-This README file was originally written by Daniel Bolton (<https://github.com/dbbolton>) and is likewise released into the public domain.
+To run youtube-dl as a developer, you don't need to build anything either. Simply execute
+
+    python -m youtube_dl
+
+To run the test, simply invoke your favorite test runner, or execute a test file directly; any of the following work:
+
+    python -m unittest discover
+    python test/test_download.py
+    nosetests
+
+If you want to create a build of youtube-dl yourself, you'll need
+
+* python
+* make
+* pandoc
+* zip
+* nosetests
+
+### Adding support for a new site
+
+If you want to add support for a new site, copy *any* [recently modified](https://github.com/rg3/youtube-dl/commits/master/youtube_dl/extractor) file in `youtube_dl/extractor`, add an import in [`youtube_dl/extractor/__init__.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/__init__.py). Have a look at [`youtube_dl/common/extractor/common.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py) for possible helper methods and a [detailed description of what your extractor should return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L38). Don't forget to run the tests with `python test/test_download.py Test_Download.test_YourExtractor`! For a detailed tutorial, refer to [this blog post](http://filippo.io/add-support-for-a-new-video-site-to-youtube-dl/).
 
 # BUGS
 
@@ -309,7 +387,7 @@ Site support requests must contain an example URL. An example URL is a URL you m
 
 ###  Are you using the latest version?
 
-Before reporting any issue, type youtube-dl -U. This should report that you're up-to-date. Ábout 20% of the reports we receive are already fixed, but people are using outdated versions. This goes for feature requests as well.
+Before reporting any issue, type youtube-dl -U. This should report that you're up-to-date. About 20% of the reports we receive are already fixed, but people are using outdated versions. This goes for feature requests as well.
 
 ###  Is the issue already documented?
 
@@ -334,3 +412,13 @@ In particular, every site support request issue should only pertain to services
 ###  Is anyone going to need the feature?
 
 Only post features that you (or an incapicated friend you can personally talk to) require. Do not post features because they seem like a good idea. If they are really useful, they will be requested by someone who requires them.
+
+###  Is your question about youtube-dl?
+
+It may sound strange, but some bug reports we receive are completely unrelated to youtube-dl and relate to a different or even the reporter's own application. Please make sure that you are actually using youtube-dl. If you are using a UI for youtube-dl, report the bug to the maintainer of the actual application providing the UI. On the other hand, if your UI for youtube-dl fails in some way you believe is related to youtube-dl, by all means, go ahead and report the bug.
+
+# COPYRIGHT
+
+youtube-dl is released into the public domain by the copyright holders.
+
+This README file was originally written by Daniel Bolton (<https://github.com/dbbolton>) and is likewise released into the public domain.

devscripts/bash-completion.in

@@ -6,7 +6,7 @@ __youtube_dl()
     prev="${COMP_WORDS[COMP_CWORD-1]}"
     opts="{{flags}}"
     keywords=":ytfavorites :ytrecommended :ytsubscriptions :ytwatchlater :ythistory"
-    fileopts="-a|--batch-file|--download-archive|--cookies"
+    fileopts="-a|--batch-file|--download-archive|--cookies|--load-info"
     diropts="--cache-dir"
 
     if [[ ${prev} =~ ${fileopts} ]]; then

devscripts/check-porn.py

@@ -3,6 +3,9 @@
 """
 This script employs a VERY basic heuristic ('porn' in webpage.lower()) to check
 if we are not 'age_limit' tagging some porn site
+
+A second approach implemented relies on a list of porn domains, to activate it
+pass the list filename as the only argument
 """
 
 # Allow direct execution
@@ -11,9 +14,17 @@ import sys
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import get_testcases
+from youtube_dl.utils import compat_urllib_parse_urlparse
 from youtube_dl.utils import compat_urllib_request
 
+if len(sys.argv) > 1:
+    METHOD = 'LIST'
+    LIST = open(sys.argv[1]).read().decode('utf8').strip()
+else:
+    METHOD = 'EURISTIC'
+
 for test in get_testcases():
+    if METHOD == 'EURISTIC':
         try:
             webpage = compat_urllib_request.urlopen(test['url'], timeout=10).read()
         except:
@@ -22,14 +33,23 @@ for test in get_testcases():
 
         webpage = webpage.decode('utf8', 'replace')
 
-    if 'porn' in webpage.lower() and ('info_dict' not in test
+        RESULT = 'porn' in webpage.lower()
-                                      or 'age_limit' not in test['info_dict']
+
+    elif METHOD == 'LIST':
+        domain = compat_urllib_parse_urlparse(test['url']).netloc
+        if not domain:
+            print('\nFail: {0}'.format(test['name']))
+            continue
+        domain = '.'.join(domain.split('.')[-2:])
+
+        RESULT = ('.' + domain + '\n' in LIST or '\n' + domain + '\n' in LIST)
+
+    if RESULT and ('info_dict' not in test or 'age_limit' not in test['info_dict']
                    or test['info_dict']['age_limit'] != 18):
         print('\nPotential missing age_limit check: {0}'.format(test['name']))
 
-    elif 'porn' not in webpage.lower() and ('info_dict' in test and
+    elif not RESULT and ('info_dict' in test and 'age_limit' in test['info_dict']
-                                            'age_limit' in test['info_dict'] and
+                         and test['info_dict']['age_limit'] == 18):
-                                            test['info_dict']['age_limit'] == 18):
         print('\nPotential false negative: {0}'.format(test['name']))
 
     else:

devscripts/gh-pages/update-feed.py

@@ -1,56 +1,76 @@
 #!/usr/bin/env python3
 
 import datetime
-
+import io
+import json
 import textwrap
 
-import json
 
 atom_template = textwrap.dedent("""\
-								<?xml version='1.0' encoding='utf-8'?>
+    <?xml version="1.0" encoding="utf-8"?>
-								<atom:feed xmlns:atom="http://www.w3.org/2005/Atom">
+    <feed xmlns="http://www.w3.org/2005/Atom">
-									<atom:title>youtube-dl releases</atom:title>
+        <link rel="self" href="http://rg3.github.io/youtube-dl/update/releases.atom" />
-									<atom:id>youtube-dl-updates-feed</atom:id>
+        <title>youtube-dl releases</title>
-									<atom:updated>@TIMESTAMP@</atom:updated>
+        <id>https://yt-dl.org/feed/youtube-dl-updates-feed</id>
+        <updated>@TIMESTAMP@</updated>
         @ENTRIES@
-								</atom:feed>""")
+    </feed>""")
 
 entry_template = textwrap.dedent("""
-								<atom:entry>
+    <entry>
-									<atom:id>youtube-dl-@VERSION@</atom:id>
+        <id>https://yt-dl.org/feed/youtube-dl-updates-feed/youtube-dl-@VERSION@</id>
-									<atom:title>New version @VERSION@</atom:title>
+        <title>New version @VERSION@</title>
-									<atom:link href="http://rg3.github.io/youtube-dl" />
+        <link href="http://rg3.github.io/youtube-dl" />
-									<atom:content type="xhtml">
+        <content type="xhtml">
             <div xmlns="http://www.w3.org/1999/xhtml">
                 Downloads available at <a href="https://yt-dl.org/downloads/@VERSION@/">https://yt-dl.org/downloads/@VERSION@/</a>
             </div>
-									</atom:content>
+        </content>
-									<atom:author>
+        <author>
-										<atom:name>The youtube-dl maintainers</atom:name>
+            <name>The youtube-dl maintainers</name>
-									</atom:author>
+        </author>
-									<atom:updated>@TIMESTAMP@</atom:updated>
+        <updated>@TIMESTAMP@</updated>
-								</atom:entry>
+    </entry>
     """)
 
 now = datetime.datetime.now()
-now_iso = now.isoformat()
+now_iso = now.isoformat() + 'Z'
 
 atom_template = atom_template.replace('@TIMESTAMP@', now_iso)
 
-entries=[]
-
 versions_info = json.load(open('update/versions.json'))
 versions = list(versions_info['versions'].keys())
 versions.sort()
 
+entries = []
 for v in versions:
-	entry = entry_template.replace('@TIMESTAMP@',v.replace('.','-'))
+    fields = v.split('.')
+    year, month, day = map(int, fields[:3])
+    faked = 0
+    patchlevel = 0
+    while True:
+        try:
+            datetime.date(year, month, day)
+        except ValueError:
+            day -= 1
+            faked += 1
+            assert day > 0
+            continue
+        break
+    if len(fields) >= 4:
+        try:
+            patchlevel = int(fields[3])
+        except ValueError:
+            patchlevel = 1
+    timestamp = '%04d-%02d-%02dT00:%02d:%02dZ' % (year, month, day, faked, patchlevel)
+
+    entry = entry_template.replace('@TIMESTAMP@', timestamp)
     entry = entry.replace('@VERSION@', v)
     entries.append(entry)
 
 entries_str = textwrap.indent(''.join(entries), '\t')
 atom_template = atom_template.replace('@ENTRIES@', entries_str)
 
-with open('update/releases.atom','w',encoding='utf-8') as atom_file:
+with io.open('update/releases.atom', 'w', encoding='utf-8') as atom_file:
     atom_file.write(atom_template)
 

devscripts/make_readme.py

@@ -1,10 +1,14 @@
+import io
 import sys
 import re
 
 README_FILE = 'README.md'
 helptext = sys.stdin.read()
 
-with open(README_FILE) as f:
+if isinstance(helptext, bytes):
+    helptext = helptext.decode('utf-8')
+
+with io.open(README_FILE, encoding='utf-8') as f:
     oldreadme = f.read()
 
 header = oldreadme[:oldreadme.index('# OPTIONS')]
@@ -14,7 +18,7 @@ options = helptext[helptext.index('  General Options:')+19:]
 options = re.sub(r'^  (\w.+)$', r'## \1', options, flags=re.M)
 options = '# OPTIONS\n' + options + '\n'
 
-with open(README_FILE, 'w') as f:
+with io.open(README_FILE, 'w', encoding='utf-8') as f:
     f.write(header)
     f.write(options)
     f.write(footer)

devscripts/release.sh

@@ -24,6 +24,8 @@ if [ -z "$1" ]; then echo "ERROR: specify version number like this: $0 1994.09.0
 version="$1"
 if [ ! -z "`git tag | grep "$version"`" ]; then echo 'ERROR: version already present'; exit 1; fi
 if [ ! -z "`git status --porcelain | grep -v CHANGELOG`" ]; then echo 'ERROR: the working directory is not clean; commit or stash changes'; exit 1; fi
+useless_files=$(find youtube_dl -type f -not -name '*.py')
+if [ ! -z "$useless_files" ]; then echo "ERROR: Non-.py files in youtube_dl: $useless_files"; exit 1; fi
 if [ ! -f "updates_key.pem" ]; then echo 'ERROR: updates_key.pem missing'; exit 1; fi
 
 /bin/echo -e "\n### First of all, testing..."

setup.py

@@ -3,7 +3,9 @@
 
 from __future__ import print_function
 
+import os.path
 import pkg_resources
+import warnings
 import sys
 
 try:
@@ -44,12 +46,24 @@ py2exe_params = {
 if len(sys.argv) >= 2 and sys.argv[1] == 'py2exe':
     params = py2exe_params
 else:
-    params = {
+    files_spec = [
-        'data_files': [  # Installing system-wide would require sudo...
         ('etc/bash_completion.d', ['youtube-dl.bash-completion']),
         ('share/doc/youtube_dl', ['README.txt']),
         ('share/man/man1', ['youtube-dl.1'])
     ]
+    root = os.path.dirname(os.path.abspath(__file__))
+    data_files = []
+    for dirname, files in files_spec:
+        resfiles = []
+        for fn in files:
+            if not os.path.exists(fn):
+                warnings.warn('Skipping file %s since it is not present. Type  make  to build all automatically generated files.' % fn)
+            else:
+                resfiles.append(fn)
+        data_files.append((dirname, resfiles))
+
+    params = {
+        'data_files': data_files,
     }
     if setuptools_available:
         params['entry_points'] = {'console_scripts': ['youtube-dl = youtube_dl:main']}
@@ -71,7 +85,10 @@ setup(
     author_email='ytdl@yt-dl.org',
     maintainer='Philipp Hagemeister',
     maintainer_email='phihag@phihag.de',
-    packages=['youtube_dl', 'youtube_dl.extractor'],
+    packages=[
+        'youtube_dl',
+        'youtube_dl.extractor', 'youtube_dl.downloader',
+        'youtube_dl.postprocessor'],
 
     # Provokes warning on most systems (why?!)
     # test_suite = 'nose.collector',

test/test_YoutubeDL.py

@@ -1,5 +1,7 @@
 #!/usr/bin/env python
 
+from __future__ import unicode_literals
+
 # Allow direct execution
 import os
 import sys
@@ -8,6 +10,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 from test.helper import FakeYDL
 from youtube_dl import YoutubeDL
+from youtube_dl.extractor import YoutubeIE
 
 
 class YDL(FakeYDL):
@@ -29,105 +32,191 @@ class TestFormatSelection(unittest.TestCase):
         ydl = YDL()
         ydl.params['prefer_free_formats'] = True
         formats = [
-            {u'ext': u'webm', u'height': 460},
+            {'ext': 'webm', 'height': 460},
-            {u'ext': u'mp4',  u'height': 460},
+            {'ext': 'mp4',  'height': 460},
         ]
-        info_dict = {u'formats': formats, u'extractor': u'test'}
+        info_dict = {'formats': formats, 'extractor': 'test'}
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'ext'], u'webm')
+        self.assertEqual(downloaded['ext'], 'webm')
 
         # Different resolution => download best quality (mp4)
         ydl = YDL()
         ydl.params['prefer_free_formats'] = True
         formats = [
-            {u'ext': u'webm', u'height': 720},
+            {'ext': 'webm', 'height': 720},
-            {u'ext': u'mp4', u'height': 1080},
+            {'ext': 'mp4', 'height': 1080},
         ]
-        info_dict[u'formats'] = formats
+        info_dict['formats'] = formats
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'ext'], u'mp4')
+        self.assertEqual(downloaded['ext'], 'mp4')
 
-        # No prefer_free_formats => keep original formats order
+        # No prefer_free_formats => prefer mp4 and flv for greater compatibilty
         ydl = YDL()
         ydl.params['prefer_free_formats'] = False
         formats = [
-            {u'ext': u'webm', u'height': 720},
+            {'ext': 'webm', 'height': 720},
-            {u'ext': u'flv', u'height': 720},
+            {'ext': 'mp4', 'height': 720},
+            {'ext': 'flv', 'height': 720},
         ]
-        info_dict[u'formats'] = formats
+        info_dict['formats'] = formats
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'ext'], u'flv')
+        self.assertEqual(downloaded['ext'], 'mp4')
+
+        ydl = YDL()
+        ydl.params['prefer_free_formats'] = False
+        formats = [
+            {'ext': 'flv', 'height': 720},
+            {'ext': 'webm', 'height': 720},
+        ]
+        info_dict['formats'] = formats
+        yie = YoutubeIE(ydl)
+        yie._sort_formats(info_dict['formats'])
+        ydl.process_ie_result(info_dict)
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['ext'], 'flv')
 
     def test_format_limit(self):
         formats = [
-            {u'format_id': u'meh', u'url': u'http://example.com/meh'},
+            {'format_id': 'meh', 'url': 'http://example.com/meh', 'preference': 1},
-            {u'format_id': u'good', u'url': u'http://example.com/good'},
+            {'format_id': 'good', 'url': 'http://example.com/good', 'preference': 2},
-            {u'format_id': u'great', u'url': u'http://example.com/great'},
+            {'format_id': 'great', 'url': 'http://example.com/great', 'preference': 3},
-            {u'format_id': u'excellent', u'url': u'http://example.com/exc'},
+            {'format_id': 'excellent', 'url': 'http://example.com/exc', 'preference': 4},
         ]
         info_dict = {
-            u'formats': formats, u'extractor': u'test', 'id': 'testvid'}
+            'formats': formats, 'extractor': 'test', 'id': 'testvid'}
 
         ydl = YDL()
         ydl.process_ie_result(info_dict)
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'format_id'], u'excellent')
+        self.assertEqual(downloaded['format_id'], 'excellent')
 
         ydl = YDL({'format_limit': 'good'})
         assert ydl.params['format_limit'] == 'good'
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'format_id'], u'good')
+        self.assertEqual(downloaded['format_id'], 'good')
 
         ydl = YDL({'format_limit': 'great', 'format': 'all'})
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
-        self.assertEqual(ydl.downloaded_info_dicts[0][u'format_id'], u'meh')
+        self.assertEqual(ydl.downloaded_info_dicts[0]['format_id'], 'meh')
-        self.assertEqual(ydl.downloaded_info_dicts[1][u'format_id'], u'good')
+        self.assertEqual(ydl.downloaded_info_dicts[1]['format_id'], 'good')
-        self.assertEqual(ydl.downloaded_info_dicts[2][u'format_id'], u'great')
+        self.assertEqual(ydl.downloaded_info_dicts[2]['format_id'], 'great')
         self.assertTrue('3' in ydl.msgs[0])
 
         ydl = YDL()
         ydl.params['format_limit'] = 'excellent'
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded[u'format_id'], u'excellent')
+        self.assertEqual(downloaded['format_id'], 'excellent')
 
     def test_format_selection(self):
         formats = [
-            {u'format_id': u'35', u'ext': u'mp4'},
+            {'format_id': '35', 'ext': 'mp4', 'preference': 1},
-            {u'format_id': u'45', u'ext': u'webm'},
+            {'format_id': '45', 'ext': 'webm', 'preference': 2},
-            {u'format_id': u'47', u'ext': u'webm'},
+            {'format_id': '47', 'ext': 'webm', 'preference': 3},
-            {u'format_id': u'2', u'ext': u'flv'},
+            {'format_id': '2', 'ext': 'flv', 'preference': 4},
         ]
-        info_dict = {u'formats': formats, u'extractor': u'test'}
+        info_dict = {'formats': formats, 'extractor': 'test'}
 
-        ydl = YDL({'format': u'20/47'})
+        ydl = YDL({'format': '20/47'})
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'47')
+        self.assertEqual(downloaded['format_id'], '47')
 
-        ydl = YDL({'format': u'20/71/worst'})
+        ydl = YDL({'format': '20/71/worst'})
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'35')
+        self.assertEqual(downloaded['format_id'], '35')
 
         ydl = YDL()
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'2')
+        self.assertEqual(downloaded['format_id'], '2')
 
-        ydl = YDL({'format': u'webm/mp4'})
+        ydl = YDL({'format': 'webm/mp4'})
-        ydl.process_ie_result(info_dict)
+        ydl.process_ie_result(info_dict.copy())
         downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'47')
+        self.assertEqual(downloaded['format_id'], '47')
 
-        ydl = YDL({'format': u'3gp/40/mp4'})
+        ydl = YDL({'format': '3gp/40/mp4'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], '35')
+
+    def test_format_selection_audio(self):
+        formats = [
+            {'format_id': 'audio-low', 'ext': 'webm', 'preference': 1, 'vcodec': 'none'},
+            {'format_id': 'audio-mid', 'ext': 'webm', 'preference': 2, 'vcodec': 'none'},
+            {'format_id': 'audio-high', 'ext': 'flv', 'preference': 3, 'vcodec': 'none'},
+            {'format_id': 'vid', 'ext': 'mp4', 'preference': 4},
+        ]
+        info_dict = {'formats': formats, 'extractor': 'test'}
+
+        ydl = YDL({'format': 'bestaudio'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'audio-high')
+
+        ydl = YDL({'format': 'worstaudio'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'audio-low')
+
+        formats = [
+            {'format_id': 'vid-low', 'ext': 'mp4', 'preference': 1},
+            {'format_id': 'vid-high', 'ext': 'mp4', 'preference': 2},
+        ]
+        info_dict = {'formats': formats, 'extractor': 'test'}
+
+        ydl = YDL({'format': 'bestaudio/worstaudio/best'})
+        ydl.process_ie_result(info_dict.copy())
+        downloaded = ydl.downloaded_info_dicts[0]
+        self.assertEqual(downloaded['format_id'], 'vid-high')
+
+    def test_youtube_format_selection(self):
+        order = [
+            '38', '37', '46', '22', '45', '35', '44', '18', '34', '43', '6', '5', '36', '17', '13',
+            # Apple HTTP Live Streaming
+            '96', '95', '94', '93', '92', '132', '151',
+            # 3D
+            '85', '84', '102', '83', '101', '82', '100',
+            # Dash video
+            '138', '137', '248', '136', '247', '135', '246',
+            '245', '244', '134', '243', '133', '242', '160',
+            # Dash audio
+            '141', '172', '140', '139', '171',
+        ]
+
+        for f1id, f2id in zip(order, order[1:]):
+            f1 = YoutubeIE._formats[f1id].copy()
+            f1['format_id'] = f1id
+            f2 = YoutubeIE._formats[f2id].copy()
+            f2['format_id'] = f2id
+
+            info_dict = {'formats': [f1, f2], 'extractor': 'youtube'}
+            ydl = YDL()
+            yie = YoutubeIE(ydl)
+            yie._sort_formats(info_dict['formats'])
             ydl.process_ie_result(info_dict)
             downloaded = ydl.downloaded_info_dicts[0]
-        self.assertEqual(downloaded['format_id'], u'35')
+            self.assertEqual(downloaded['format_id'], f1id)
+
+            info_dict = {'formats': [f2, f1], 'extractor': 'youtube'}
+            ydl = YDL()
+            yie = YoutubeIE(ydl)
+            yie._sort_formats(info_dict['formats'])
+            ydl.process_ie_result(info_dict)
+            downloaded = ydl.downloaded_info_dicts[0]
+            self.assertEqual(downloaded['format_id'], f1id)
 
     def test_add_extra_info(self):
         test_dict = {
@@ -143,17 +232,17 @@ class TestFormatSelection(unittest.TestCase):
 
     def test_prepare_filename(self):
         info = {
-            u'id': u'1234',
+            'id': '1234',
-            u'ext': u'mp4',
+            'ext': 'mp4',
-            u'width': None,
+            'width': None,
         }
         def fname(templ):
             ydl = YoutubeDL({'outtmpl': templ})
             return ydl.prepare_filename(info)
-        self.assertEqual(fname(u'%(id)s.%(ext)s'), u'1234.mp4')
+        self.assertEqual(fname('%(id)s.%(ext)s'), '1234.mp4')
-        self.assertEqual(fname(u'%(id)s-%(width)s.%(ext)s'), u'1234-NA.mp4')
+        self.assertEqual(fname('%(id)s-%(width)s.%(ext)s'), '1234-NA.mp4')
         # Replace missing fields with 'NA'
-        self.assertEqual(fname(u'%(uploader_date)s-%(id)s.%(ext)s'), u'NA-1234.mp4')
+        self.assertEqual(fname('%(uploader_date)s-%(id)s.%(ext)s'), 'NA-1234.mp4')
 
 
 if __name__ == '__main__':

test/test_all_urls.py

@@ -1,5 +1,7 @@
 #!/usr/bin/env python
 
+from __future__ import unicode_literals
+
 # Allow direct execution
 import os
 import sys
@@ -13,6 +15,7 @@ from youtube_dl.extractor import (
     FacebookIE,
     gen_extractors,
     JustinTVIE,
+    PBSIE,
     YoutubeIE,
 )
 
@@ -29,18 +32,20 @@ class TestAllURLsMatching(unittest.TestCase):
 
     def test_youtube_playlist_matching(self):
         assertPlaylist = lambda url: self.assertMatch(url, ['youtube:playlist'])
-        assertPlaylist(u'ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
+        assertPlaylist('ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
-        assertPlaylist(u'UUBABnxM4Ar9ten8Mdjj1j0Q') #585
+        assertPlaylist('UUBABnxM4Ar9ten8Mdjj1j0Q') #585
-        assertPlaylist(u'PL63F0C78739B09958')
+        assertPlaylist('PL63F0C78739B09958')
-        assertPlaylist(u'https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
+        assertPlaylist('https://www.youtube.com/playlist?list=UUBABnxM4Ar9ten8Mdjj1j0Q')
-        assertPlaylist(u'https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
+        assertPlaylist('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
-        assertPlaylist(u'https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
+        assertPlaylist('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
-        assertPlaylist(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012') #668
+        assertPlaylist('https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012') #668
-        self.assertFalse('youtube:playlist' in self.matching_ies(u'PLtS2H6bU1M'))
+        self.assertFalse('youtube:playlist' in self.matching_ies('PLtS2H6bU1M'))
+        # Top tracks
+        assertPlaylist('https://www.youtube.com/playlist?list=MCUS.20142101')
 
     def test_youtube_matching(self):
-        self.assertTrue(YoutubeIE.suitable(u'PLtS2H6bU1M'))
+        self.assertTrue(YoutubeIE.suitable('PLtS2H6bU1M'))
-        self.assertFalse(YoutubeIE.suitable(u'https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
+        self.assertFalse(YoutubeIE.suitable('https://www.youtube.com/watch?v=AV6J6_AeFEQ&playnext=1&list=PL4023E734DA416012')) #668
         self.assertMatch('http://youtu.be/BaW_jenozKc', ['youtube'])
         self.assertMatch('http://www.youtube.com/v/BaW_jenozKc', ['youtube'])
         self.assertMatch('https://youtube.googleapis.com/v/BaW_jenozKc', ['youtube'])
@@ -63,6 +68,9 @@ class TestAllURLsMatching(unittest.TestCase):
     def test_youtube_show_matching(self):
         self.assertMatch('http://www.youtube.com/show/airdisasters', ['youtube:show'])
 
+    def test_youtube_truncated(self):
+        self.assertMatch('http://www.youtube.com/watch?', ['youtube:truncated_url'])
+
     def test_justin_tv_channelid_matching(self):
         self.assertTrue(JustinTVIE.suitable(u"justin.tv/vanillatv"))
         self.assertTrue(JustinTVIE.suitable(u"twitch.tv/vanillatv"))
@@ -80,7 +88,7 @@ class TestAllURLsMatching(unittest.TestCase):
         self.assertTrue(JustinTVIE.suitable(u"http://www.twitch.tv/tsm_theoddone/c/2349361"))
 
     def test_youtube_extract(self):
-        assertExtractId = lambda url, id: self.assertEqual(YoutubeIE()._extract_id(url), id)
+        assertExtractId = lambda url, id: self.assertEqual(YoutubeIE.extract_id(url), id)
         assertExtractId('http://www.youtube.com/watch?&v=BaW_jenozKc', 'BaW_jenozKc')
         assertExtractId('https://www.youtube.com/watch?&v=BaW_jenozKc', 'BaW_jenozKc')
         assertExtractId('https://www.youtube.com/watch?feature=player_embedded&v=BaW_jenozKc', 'BaW_jenozKc')
@@ -89,7 +97,7 @@ class TestAllURLsMatching(unittest.TestCase):
         assertExtractId('BaW_jenozKc', 'BaW_jenozKc')
 
     def test_facebook_matching(self):
-        self.assertTrue(FacebookIE.suitable(u'https://www.facebook.com/Shiniknoh#!/photo.php?v=10153317450565268'))
+        self.assertTrue(FacebookIE.suitable('https://www.facebook.com/Shiniknoh#!/photo.php?v=10153317450565268'))
 
     def test_no_duplicates(self):
         ies = gen_extractors()
@@ -113,10 +121,20 @@ class TestAllURLsMatching(unittest.TestCase):
     def test_vimeo_matching(self):
         self.assertMatch('http://vimeo.com/channels/tributes', ['vimeo:channel'])
         self.assertMatch('http://vimeo.com/user7108434', ['vimeo:user'])
+        self.assertMatch('http://vimeo.com/user7108434/videos', ['vimeo:user'])
+        self.assertMatch('https://vimeo.com/user21297594/review/75524534/3c257a1b5d', ['vimeo:review'])
 
     # https://github.com/rg3/youtube-dl/issues/1930
     def test_soundcloud_not_matching_sets(self):
         self.assertMatch('http://soundcloud.com/floex/sets/gone-ep', ['soundcloud:set'])
 
+    def test_tumblr(self):
+        self.assertMatch('http://tatianamaslanydaily.tumblr.com/post/54196191430/orphan-black-dvd-extra-behind-the-scenes', ['Tumblr'])
+        self.assertMatch('http://tatianamaslanydaily.tumblr.com/post/54196191430', ['Tumblr'])
+
+    def test_pbs(self):
+        # https://github.com/rg3/youtube-dl/issues/2350
+        self.assertMatch('http://video.pbs.org/viralplayer/2365173446/', ['PBS'])
+
 if __name__ == '__main__':
     unittest.main()

test/test_download.py

@@ -22,6 +22,7 @@ import socket
 
 import youtube_dl.YoutubeDL
 from youtube_dl.utils import (
+    compat_http_client,
     compat_str,
     compat_urllib_error,
     compat_HTTPError,
@@ -90,7 +91,7 @@ def generator(test_case):
         def _hook(status):
             if status['status'] == 'finished':
                 finished_hook_called.add(status['filename'])
-        ydl.fd.add_progress_hook(_hook)
+        ydl.add_progress_hook(_hook)
 
         def get_tc_filename(tc):
             return tc.get('file') or ydl.prepare_filename(tc.get('info_dict', {}))
@@ -110,7 +111,7 @@ def generator(test_case):
                     ydl.download([test_case['url']])
                 except (DownloadError, ExtractorError) as err:
                     # Check if the exception is not a network related one
-                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
+                    if not err.exc_info[0] in (compat_urllib_error.URLError, socket.timeout, UnavailableVideoError, compat_http_client.BadStatusLine) or (err.exc_info[0] == compat_HTTPError and err.exc_info[1].code == 503):
                         raise
 
                     if try_num == RETRIES:
@@ -148,7 +149,7 @@ def generator(test_case):
                     for key, value in info_dict.items()
                     if value and key in ('title', 'description', 'uploader', 'upload_date', 'uploader_id', 'location'))
                 if not all(key in tc.get('info_dict', {}).keys() for key in test_info_dict.keys()):
-                    sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=2) + u'\n')
+                    sys.stderr.write(u'\n"info_dict": ' + json.dumps(test_info_dict, ensure_ascii=False, indent=4) + u'\n')
 
                 # Check for the presence of mandatory fields
                 for key in ('id', 'url', 'title', 'ext'):

test/test_playlists.py

@@ -1,6 +1,7 @@
 #!/usr/bin/env python
 # encoding: utf-8
 
+from __future__ import unicode_literals
 
 # Allow direct execution
 import os
@@ -12,6 +13,7 @@ from test.helper import FakeYDL
 
 
 from youtube_dl.extractor import (
+    AcademicEarthCourseIE,
     DailymotionPlaylistIE,
     DailymotionUserIE,
     VimeoChannelIE,
@@ -26,7 +28,14 @@ from youtube_dl.extractor import (
     BambuserChannelIE,
     BandcampAlbumIE,
     SmotriCommunityIE,
-    SmotriUserIE
+    SmotriUserIE,
+    IviCompilationIE,
+    ImdbListIE,
+    KhanAcademyIE,
+    EveryonesMixtapeIE,
+    RutubeChannelIE,
+    GoogleSearchIE,
+    GenericIE,
 )
 
 
@@ -40,23 +49,23 @@ class TestPlaylists(unittest.TestCase):
         ie = DailymotionPlaylistIE(dl)
         result = ie.extract('http://www.dailymotion.com/playlist/xv4bw_nqtv_sport/1#video=xl8v3q')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'SPORT')
+        self.assertEqual(result['title'], 'SPORT')
         self.assertTrue(len(result['entries']) > 20)
 
     def test_dailymotion_user(self):
         dl = FakeYDL()
         ie = DailymotionUserIE(dl)
-        result = ie.extract('http://www.dailymotion.com/user/generation-quoi/')
+        result = ie.extract('https://www.dailymotion.com/user/nqtv')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Génération Quoi')
+        self.assertEqual(result['title'], 'Rémi Gaillard')
-        self.assertTrue(len(result['entries']) >= 26)
+        self.assertTrue(len(result['entries']) >= 100)
 
     def test_vimeo_channel(self):
         dl = FakeYDL()
         ie = VimeoChannelIE(dl)
         result = ie.extract('http://vimeo.com/channels/tributes')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Vimeo Tributes')
+        self.assertEqual(result['title'], 'Vimeo Tributes')
         self.assertTrue(len(result['entries']) > 24)
 
     def test_vimeo_user(self):
@@ -64,7 +73,7 @@ class TestPlaylists(unittest.TestCase):
         ie = VimeoUserIE(dl)
         result = ie.extract('http://vimeo.com/nkistudio/videos')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Nki')
+        self.assertEqual(result['title'], 'Nki')
         self.assertTrue(len(result['entries']) > 65)
 
     def test_vimeo_album(self):
@@ -72,7 +81,7 @@ class TestPlaylists(unittest.TestCase):
         ie = VimeoAlbumIE(dl)
         result = ie.extract('http://vimeo.com/album/2632481')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Staff Favorites: November 2013')
+        self.assertEqual(result['title'], 'Staff Favorites: November 2013')
         self.assertTrue(len(result['entries']) > 12)
 
     def test_vimeo_groups(self):
@@ -80,7 +89,7 @@ class TestPlaylists(unittest.TestCase):
         ie = VimeoGroupsIE(dl)
         result = ie.extract('http://vimeo.com/groups/rolexawards')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Rolex Awards for Enterprise')
+        self.assertEqual(result['title'], 'Rolex Awards for Enterprise')
         self.assertTrue(len(result['entries']) > 72)
 
     def test_ustream_channel(self):
@@ -88,7 +97,7 @@ class TestPlaylists(unittest.TestCase):
         ie = UstreamChannelIE(dl)
         result = ie.extract('http://www.ustream.tv/channel/young-americans-for-liberty')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'5124905')
+        self.assertEqual(result['id'], '5124905')
         self.assertTrue(len(result['entries']) >= 11)
 
     def test_soundcloud_set(self):
@@ -96,7 +105,7 @@ class TestPlaylists(unittest.TestCase):
         ie = SoundcloudSetIE(dl)
         result = ie.extract('https://soundcloud.com/the-concept-band/sets/the-royal-concept-ep')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'The Royal Concept EP')
+        self.assertEqual(result['title'], 'The Royal Concept EP')
         self.assertTrue(len(result['entries']) >= 6)
 
     def test_soundcloud_user(self):
@@ -104,7 +113,7 @@ class TestPlaylists(unittest.TestCase):
         ie = SoundcloudUserIE(dl)
         result = ie.extract('https://soundcloud.com/the-concept-band')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'9615865')
+        self.assertEqual(result['id'], '9615865')
         self.assertTrue(len(result['entries']) >= 12)
 
     def test_livestream_event(self):
@@ -112,7 +121,7 @@ class TestPlaylists(unittest.TestCase):
         ie = LivestreamIE(dl)
         result = ie.extract('http://new.livestream.com/tedx/cityenglish')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'TEDCity2.0 (English)')
+        self.assertEqual(result['title'], 'TEDCity2.0 (English)')
         self.assertTrue(len(result['entries']) >= 4)
 
     def test_nhl_videocenter(self):
@@ -120,8 +129,8 @@ class TestPlaylists(unittest.TestCase):
         ie = NHLVideocenterIE(dl)
         result = ie.extract('http://video.canucks.nhl.com/videocenter/console?catid=999')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'999')
+        self.assertEqual(result['id'], '999')
-        self.assertEqual(result['title'], u'Highlights')
+        self.assertEqual(result['title'], 'Highlights')
         self.assertEqual(len(result['entries']), 12)
 
     def test_bambuser_channel(self):
@@ -129,7 +138,7 @@ class TestPlaylists(unittest.TestCase):
         ie = BambuserChannelIE(dl)
         result = ie.extract('http://bambuser.com/channel/pixelversity')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'pixelversity')
+        self.assertEqual(result['title'], 'pixelversity')
         self.assertTrue(len(result['entries']) >= 60)
 
     def test_bandcamp_album(self):
@@ -137,7 +146,7 @@ class TestPlaylists(unittest.TestCase):
         ie = BandcampAlbumIE(dl)
         result = ie.extract('http://mpallante.bandcamp.com/album/nightmare-night-ep')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['title'], u'Nightmare Night EP')
+        self.assertEqual(result['title'], 'Nightmare Night EP')
         self.assertTrue(len(result['entries']) >= 4)
         
     def test_smotri_community(self):
@@ -145,8 +154,8 @@ class TestPlaylists(unittest.TestCase):
         ie = SmotriCommunityIE(dl)
         result = ie.extract('http://smotri.com/community/video/kommuna')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'kommuna')
+        self.assertEqual(result['id'], 'kommuna')
-        self.assertEqual(result['title'], u'КПРФ')
+        self.assertEqual(result['title'], 'КПРФ')
         self.assertTrue(len(result['entries']) >= 4)
         
     def test_smotri_user(self):
@@ -154,9 +163,92 @@ class TestPlaylists(unittest.TestCase):
         ie = SmotriUserIE(dl)
         result = ie.extract('http://smotri.com/user/inspector')
         self.assertIsPlaylist(result)
-        self.assertEqual(result['id'], u'inspector')
+        self.assertEqual(result['id'], 'inspector')
-        self.assertEqual(result['title'], u'Inspector')
+        self.assertEqual(result['title'], 'Inspector')
         self.assertTrue(len(result['entries']) >= 9)
 
+    def test_AcademicEarthCourse(self):
+        dl = FakeYDL()
+        ie = AcademicEarthCourseIE(dl)
+        result = ie.extract('http://academicearth.org/courses/building-dynamic-websites/')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'building-dynamic-websites')
+        self.assertEqual(result['title'], 'Building Dynamic Websites')
+        self.assertEqual(result['description'], u"Today's websites are increasingly dynamic. Pages are no longer static HTML files but instead generated by scripts and database calls. User interfaces are more seamless, with technologies like Ajax replacing traditional page reloads. This course teaches students how to build dynamic websites with Ajax and with Linux, Apache, MySQL, and PHP (LAMP), one of today's most popular frameworks. Students learn how to set up domain names with DNS, how to structure pages with XHTML and CSS, how to program in JavaScript and PHP, how to configure Apache and MySQL, how to design and query databases with SQL, how to use Ajax with both XML and JSON, and how to build mashups. The course explores issues of security, scalability, and cross-browser support and also discusses enterprise-level deployments of websites, including third-party hosting, virtualization, colocation in data centers, firewalling, and load-balancing.")
+        self.assertEqual(len(result['entries']), 10)
+        
+    def test_ivi_compilation(self):
+        dl = FakeYDL()
+        ie = IviCompilationIE(dl)
+        result = ie.extract('http://www.ivi.ru/watch/dezhurnyi_angel')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'dezhurnyi_angel')
+        self.assertEqual(result['title'], 'Дежурный ангел (2010 - 2012)')
+        self.assertTrue(len(result['entries']) >= 36)
+        
+    def test_ivi_compilation_season(self):
+        dl = FakeYDL()
+        ie = IviCompilationIE(dl)
+        result = ie.extract('http://www.ivi.ru/watch/dezhurnyi_angel/season2')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'dezhurnyi_angel/season2')
+        self.assertEqual(result['title'], 'Дежурный ангел (2010 - 2012) 2 сезон')
+        self.assertTrue(len(result['entries']) >= 20)
+        
+    def test_imdb_list(self):
+        dl = FakeYDL()
+        ie = ImdbListIE(dl)
+        result = ie.extract('http://www.imdb.com/list/JFs9NWw6XI0')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'JFs9NWw6XI0')
+        self.assertEqual(result['title'], 'March 23, 2012 Releases')
+        self.assertEqual(len(result['entries']), 7)
+
+    def test_khanacademy_topic(self):
+        dl = FakeYDL()
+        ie = KhanAcademyIE(dl)
+        result = ie.extract('https://www.khanacademy.org/math/applied-math/cryptography')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'cryptography')
+        self.assertEqual(result['title'], 'Journey into cryptography')
+        self.assertEqual(result['description'], 'How have humans protected their secret messages through history? What has changed today?')
+        self.assertTrue(len(result['entries']) >= 3)
+
+    def test_EveryonesMixtape(self):
+        dl = FakeYDL()
+        ie = EveryonesMixtapeIE(dl)
+        result = ie.extract('http://everyonesmixtape.com/#/mix/m7m0jJAbMQi')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'm7m0jJAbMQi')
+        self.assertEqual(result['title'], 'Driving')
+        self.assertEqual(len(result['entries']), 24)
+        
+    def test_rutube_channel(self):
+        dl = FakeYDL()
+        ie = RutubeChannelIE(dl)
+        result = ie.extract('http://rutube.ru/tags/video/1409')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], '1409')
+        self.assertTrue(len(result['entries']) >= 34)
+
+    def test_multiple_brightcove_videos(self):
+        # https://github.com/rg3/youtube-dl/issues/2283
+        dl = FakeYDL()
+        ie = GenericIE(dl)
+        result = ie.extract('http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'always-never-nuclear-command-and-control')
+        self.assertEqual(result['title'], 'Always/Never: A Little-Seen Movie About Nuclear Command and Control : The New Yorker')
+        self.assertEqual(len(result['entries']), 3)
+
+    def test_GoogleSearch(self):
+        dl = FakeYDL()
+        ie = GoogleSearchIE(dl)
+        result = ie.extract('gvsearch15:python language')
+        self.assertIsPlaylist(result)
+        self.assertEqual(result['id'], 'python language')
+        self.assertEqual(result['title'], 'python language')
+        self.assertTrue(len(result['entries']) == 15)
+
 if __name__ == '__main__':
     unittest.main()

test/test_subtitles.py

@@ -10,9 +10,11 @@ from test.helper import FakeYDL, md5
 
 
 from youtube_dl.extractor import (
+    BlipTVIE,
     YoutubeIE,
     DailymotionIE,
     TEDIE,
+    VimeoIE,
 )
 
 
@@ -36,10 +38,6 @@ class TestYoutubeSubtitles(BaseTestSubtitles):
     url = 'QRS8MkLhQmM'
     IE = YoutubeIE
 
-    def getSubtitles(self):
-        info_dict = self.getInfoDict()
-        return info_dict[0]['subtitles']
-
     def test_youtube_no_writesubtitles(self):
         self.DL.params['writesubtitles'] = False
         subtitles = self.getSubtitles()
@@ -171,13 +169,13 @@ class TestTedSubtitles(BaseTestSubtitles):
     def test_subtitles(self):
         self.DL.params['writesubtitles'] = True
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['en']), '2154f31ff9b9f89a0aa671537559c21d')
+        self.assertEqual(md5(subtitles['en']), '4262c1665ff928a2dada178f62cb8d14')
 
     def test_subtitles_lang(self):
         self.DL.params['writesubtitles'] = True
         self.DL.params['subtitleslangs'] = ['fr']
         subtitles = self.getSubtitles()
-        self.assertEqual(md5(subtitles['fr']), '7616cbc6df20ec2c1204083c83871cf6')
+        self.assertEqual(md5(subtitles['fr']), '66a63f7f42c97a50f8c0e90bc7797bb5')
 
     def test_allsubtitles(self):
         self.DL.params['writesubtitles'] = True
@@ -206,5 +204,80 @@ class TestTedSubtitles(BaseTestSubtitles):
         for lang in langs:
             self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
 
+
+class TestBlipTVSubtitles(BaseTestSubtitles):
+    url = 'http://blip.tv/a/a-6603250'
+    IE = BlipTVIE
+
+    def test_list_subtitles(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['listsubtitles'] = True
+        info_dict = self.getInfoDict()
+        self.assertEqual(info_dict, None)
+
+    def test_allsubtitles(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['en']))
+        self.assertEqual(md5(subtitles['en']), '5b75c300af65fe4476dff79478bb93e4')
+
+
+class TestVimeoSubtitles(BaseTestSubtitles):
+    url = 'http://vimeo.com/76979871'
+    IE = VimeoIE
+
+    def test_no_writesubtitles(self):
+        subtitles = self.getSubtitles()
+        self.assertEqual(subtitles, None)
+
+    def test_subtitles(self):
+        self.DL.params['writesubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['en']), '8062383cf4dec168fc40a088aa6d5888')
+
+    def test_subtitles_lang(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['subtitleslangs'] = ['fr']
+        subtitles = self.getSubtitles()
+        self.assertEqual(md5(subtitles['fr']), 'b6191146a6c5d3a452244d853fde6dc8')
+
+    def test_allsubtitles(self):
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(set(subtitles.keys()), set(['de', 'en', 'es', 'fr']))
+
+    def test_list_subtitles(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['listsubtitles'] = True
+        info_dict = self.getInfoDict()
+        self.assertEqual(info_dict, None)
+
+    def test_automatic_captions(self):
+        self.DL.expect_warning(u'Automatic Captions not supported by this server')
+        self.DL.params['writeautomaticsub'] = True
+        self.DL.params['subtitleslang'] = ['en']
+        subtitles = self.getSubtitles()
+        self.assertTrue(len(subtitles.keys()) == 0)
+
+    def test_nosubtitles(self):
+        self.DL.expect_warning(u'video doesn\'t have subtitles')
+        self.url = 'http://vimeo.com/56015672'
+        self.DL.params['writesubtitles'] = True
+        self.DL.params['allsubtitles'] = True
+        subtitles = self.getSubtitles()
+        self.assertEqual(len(subtitles), 0)
+
+    def test_multiple_langs(self):
+        self.DL.params['writesubtitles'] = True
+        langs = ['es', 'fr', 'de']
+        self.DL.params['subtitleslangs'] = langs
+        subtitles = self.getSubtitles()
+        for lang in langs:
+            self.assertTrue(subtitles.get(lang) is not None, u'Subtitles for \'%s\' not extracted' % lang)
+
+
 if __name__ == '__main__':
     unittest.main()

test/test_unicode_literals.py

@@ -0,0 +1,47 @@
+from __future__ import unicode_literals
+
+import io
+import os
+import re
+import unittest
+
+rootDir = os.path.dirname(os.path.dirname(os.path.abspath(__file__)))
+
+IGNORED_FILES = [
+    'setup.py',  # http://bugs.python.org/issue13943
+]
+
+
+class TestUnicodeLiterals(unittest.TestCase):
+    def test_all_files(self):
+        print('Skipping this test (not yet fully implemented)')
+        return
+
+        for dirpath, _, filenames in os.walk(rootDir):
+            for basename in filenames:
+                if not basename.endswith('.py'):
+                    continue
+                if basename in IGNORED_FILES:
+                    continue
+
+                fn = os.path.join(dirpath, basename)
+                with io.open(fn, encoding='utf-8') as inf:
+                    code = inf.read()
+
+                if "'" not in code and '"' not in code:
+                    continue
+                imps = 'from __future__ import unicode_literals'
+                self.assertTrue(
+                    imps in code,
+                    ' %s  missing in %s' % (imps, fn))
+
+                m = re.search(r'(?<=\s)u[\'"](?!\)|,|$)', code)
+                if m is not None:
+                    self.assertTrue(
+                        m is None,
+                        'u present in %s, around %s' % (
+                            fn, code[m.start() - 10:m.end() + 10]))
+
+
+if __name__ == '__main__':
+    unittest.main()

test/test_utils.py

@@ -13,20 +13,25 @@ import xml.etree.ElementTree
 
 #from youtube_dl.utils import htmlentity_transform
 from youtube_dl.utils import (
-    timeconvert,
-    sanitize_filename,
-    unescapeHTML,
-    orderedSet,
     DateRange,
-    unified_strdate,
-    find_xpath_attr,
-    get_meta_content,
-    xpath_with_ns,
-    smuggle_url,
-    unsmuggle_url,
-    shell_quote,
     encodeFilename,
+    find_xpath_attr,
+    fix_xml_ampersands,
+    get_meta_content,
+    orderedSet,
+    PagedList,
+    parse_duration,
+    sanitize_filename,
+    shell_quote,
+    smuggle_url,
     str_to_int,
+    struct_unpack,
+    timeconvert,
+    unescapeHTML,
+    unified_strdate,
+    unsmuggle_url,
+    url_basename,
+    xpath_with_ns,
 )
 
 if sys.version_info < (3, 0):
@@ -123,6 +128,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unified_strdate('8/7/2009'), '20090708')
         self.assertEqual(unified_strdate('Dec 14, 2012'), '20121214')
         self.assertEqual(unified_strdate('2012/10/11 01:56:38 +0000'), '20121011')
+        self.assertEqual(unified_strdate('1968-12-10'), '19681210')
 
     def test_find_xpath_attr(self):
         testxml = u'''<root>
@@ -181,6 +187,68 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(str_to_int('123,456'), 123456)
         self.assertEqual(str_to_int('123.456'), 123456)
 
+    def test_url_basename(self):
+        self.assertEqual(url_basename(u'http://foo.de/'), u'')
+        self.assertEqual(url_basename(u'http://foo.de/bar/baz'), u'baz')
+        self.assertEqual(url_basename(u'http://foo.de/bar/baz?x=y'), u'baz')
+        self.assertEqual(url_basename(u'http://foo.de/bar/baz#x=y'), u'baz')
+        self.assertEqual(url_basename(u'http://foo.de/bar/baz/'), u'baz')
+        self.assertEqual(
+            url_basename(u'http://media.w3.org/2010/05/sintel/trailer.mp4'),
+            u'trailer.mp4')
+
+    def test_parse_duration(self):
+        self.assertEqual(parse_duration(None), None)
+        self.assertEqual(parse_duration('1'), 1)
+        self.assertEqual(parse_duration('1337:12'), 80232)
+        self.assertEqual(parse_duration('9:12:43'), 33163)
+        self.assertEqual(parse_duration('12:00'), 720)
+        self.assertEqual(parse_duration('00:01:01'), 61)
+        self.assertEqual(parse_duration('x:y'), None)
+        self.assertEqual(parse_duration('3h11m53s'), 11513)
+        self.assertEqual(parse_duration('62m45s'), 3765)
+        self.assertEqual(parse_duration('6m59s'), 419)
+        self.assertEqual(parse_duration('49s'), 49)
+        self.assertEqual(parse_duration('0h0m0s'), 0)
+        self.assertEqual(parse_duration('0m0s'), 0)
+        self.assertEqual(parse_duration('0s'), 0)
+
+    def test_fix_xml_ampersands(self):
+        self.assertEqual(
+            fix_xml_ampersands('"&x=y&z=a'), '"&amp;x=y&amp;z=a')
+        self.assertEqual(
+            fix_xml_ampersands('"&amp;x=y&wrong;&z=a'),
+            '"&amp;x=y&amp;wrong;&amp;z=a')
+        self.assertEqual(
+            fix_xml_ampersands('&amp;&apos;&gt;&lt;&quot;'),
+            '&amp;&apos;&gt;&lt;&quot;')
+        self.assertEqual(
+            fix_xml_ampersands('&#1234;&#x1abC;'), '&#1234;&#x1abC;')
+        self.assertEqual(fix_xml_ampersands('&#&#'), '&amp;#&amp;#')
+
+    def test_paged_list(self):
+        def testPL(size, pagesize, sliceargs, expected):
+            def get_page(pagenum):
+                firstid = pagenum * pagesize
+                upto = min(size, pagenum * pagesize + pagesize)
+                for i in range(firstid, upto):
+                    yield i
+
+            pl = PagedList(get_page, pagesize)
+            got = pl.getslice(*sliceargs)
+            self.assertEqual(got, expected)
+
+        testPL(5, 2, (), [0, 1, 2, 3, 4])
+        testPL(5, 2, (1,), [1, 2, 3, 4])
+        testPL(5, 2, (2,), [2, 3, 4])
+        testPL(5, 2, (4,), [4])
+        testPL(5, 2, (0, 3), [0, 1, 2])
+        testPL(5, 2, (1, 4), [1, 2, 3])
+        testPL(5, 2, (2, 99), [2, 3, 4])
+        testPL(5, 2, (20, 99), [])
+
+    def test_struct_unpack(self):
+        self.assertEqual(struct_unpack(u'!B', b'\x00'), (0,))
 
 if __name__ == '__main__':
     unittest.main()

test/test_youtube_lists.py

@@ -30,7 +30,7 @@ class TestYoutubeLists(unittest.TestCase):
         result = ie.extract('https://www.youtube.com/playlist?list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
         self.assertIsPlaylist(result)
         self.assertEqual(result['title'], 'ytdl test PL')
-        ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']]
+        ytie_results = [YoutubeIE().extract_id(url['url']) for url in result['entries']]
         self.assertEqual(ytie_results, [ 'bV9L5Ht9LgY', 'FXxLjLQi3Fg', 'tU3Bgo5qJZE'])
 
     def test_youtube_playlist_noplaylist(self):
@@ -39,7 +39,7 @@ class TestYoutubeLists(unittest.TestCase):
         ie = YoutubePlaylistIE(dl)
         result = ie.extract('https://www.youtube.com/watch?v=FXxLjLQi3Fg&list=PLwiyx1dc3P2JR9N8gQaQN_BCvlSlap7re')
         self.assertEqual(result['_type'], 'url')
-        self.assertEqual(YoutubeIE()._extract_id(result['url']), 'FXxLjLQi3Fg')
+        self.assertEqual(YoutubeIE().extract_id(result['url']), 'FXxLjLQi3Fg')
 
     def test_issue_673(self):
         dl = FakeYDL()
@@ -59,7 +59,7 @@ class TestYoutubeLists(unittest.TestCase):
         dl = FakeYDL()
         ie = YoutubePlaylistIE(dl)
         result = ie.extract('https://www.youtube.com/playlist?list=PLwP_SiAcdui0KVebT0mU9Apz359a4ubsC')
-        ytie_results = [YoutubeIE()._extract_id(url['url']) for url in result['entries']]
+        ytie_results = [YoutubeIE().extract_id(url['url']) for url in result['entries']]
         self.assertFalse('pElCt5oNDuI' in ytie_results)
         self.assertFalse('KdPEApIVdWM' in ytie_results)
         
@@ -76,9 +76,9 @@ class TestYoutubeLists(unittest.TestCase):
         # TODO find a > 100 (paginating?) videos course
         result = ie.extract('https://www.youtube.com/course?list=ECUl4u3cNGP61MdtwGTqZA0MreSaDybji8')
         entries = result['entries']
-        self.assertEqual(YoutubeIE()._extract_id(entries[0]['url']), 'j9WZyLZCBzs')
+        self.assertEqual(YoutubeIE().extract_id(entries[0]['url']), 'j9WZyLZCBzs')
         self.assertEqual(len(entries), 25)
-        self.assertEqual(YoutubeIE()._extract_id(entries[-1]['url']), 'rYefUsYuEp0')
+        self.assertEqual(YoutubeIE().extract_id(entries[-1]['url']), 'rYefUsYuEp0')
 
     def test_youtube_channel(self):
         dl = FakeYDL()
@@ -117,10 +117,17 @@ class TestYoutubeLists(unittest.TestCase):
         original_video = entries[0]
         self.assertEqual(original_video['id'], 'rjFaenf1T-Y')
 
+    def test_youtube_toptracks(self):
+        dl = FakeYDL()
+        ie = YoutubePlaylistIE(dl)
+        result = ie.extract('https://www.youtube.com/playlist?list=MCUS')
+        entries = result['entries']
+        self.assertEqual(len(entries), 100)
+
     def test_youtube_toplist(self):
         dl = FakeYDL()
         ie = YoutubeTopListIE(dl)
-        result = ie.extract('yttoplist:music:Top Tracks')
+        result = ie.extract('yttoplist:music:Trending')
         entries = result['entries']
         self.assertTrue(len(entries) >= 5)
 

test/test_youtube_signature.py

@@ -28,10 +28,10 @@ _TESTS = [
         u'3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@',
     ),
     (
-        u'https://s.ytimg.com/yts/swfbin/watch_as3-vflg5GhxU.swf',
+        u'https://s.ytimg.com/yts/jsbin/html5player-vfle-mVwz.js',
-        u'swf',
+        u'js',
-        82,
+        90,
-        u':/.-,+*)=\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBAzyxw>utsrqponmlkjihgfedcba987654321'
+        u']\\[@?>=<;:/.-,+*)(\'&%$#"hZYXWVUTSRQPONMLKJIHGFEDCBAzyxwvutsrqponmlkjiagfedcb39876',
     ),
 ]
 

youtube-dl.plugin.zsh

@@ -0,0 +1,24 @@
+# This allows the youtube-dl command to be installed in ZSH using antigen.
+# Antigen is a bundle manager. It allows you to enhance the functionality of
+# your zsh session by installing bundles and themes easily.
+
+# Antigen documentation:
+# http://antigen.sharats.me/
+# https://github.com/zsh-users/antigen
+
+# Install youtube-dl:
+# antigen bundle rg3/youtube-dl
+# Bundles installed by antigen are available for use immediately.
+
+# Update youtube-dl (and all other antigen bundles):
+# antigen update
+
+# The antigen command will download the git repository to a folder and then
+# execute an enabling script (this file). The complete process for loading the
+# code is documented here:
+# https://github.com/zsh-users/antigen#notes-on-writing-plugins
+
+# This specific script just aliases youtube-dl to the python script that this
+# library provides. This requires updating the PYTHONPATH to ensure that the
+# full set of code can be located.
+alias youtube-dl="PYTHONPATH=$(dirname $0) $(dirname $0)/bin/youtube-dl"

youtube_dl/FileDownloader.py

@@ -1,724 +1,12 @@
-import os
+# Legacy file for backwards compatibility, use youtube_dl.downloader instead!
-import re
+from .downloader import FileDownloader as RealFileDownloader
-import subprocess
+from .downloader import get_suitable_downloader
-import sys
-import time
-
-from .utils import (
-    compat_urllib_error,
-    compat_urllib_request,
-    ContentTooShortError,
-    determine_ext,
-    encodeFilename,
-    format_bytes,
-    sanitize_open,
-    timeconvert,
-)
-
-
-class FileDownloader(object):
-    """File Downloader class.
-
-    File downloader objects are the ones responsible of downloading the
-    actual video file and writing it to disk.
-
-    File downloaders accept a lot of parameters. In order not to saturate
-    the object constructor with arguments, it receives a dictionary of
-    options instead.
-
-    Available options:
-
-    verbose:           Print additional info to stdout.
-    quiet:             Do not print messages to stdout.
-    ratelimit:         Download speed limit, in bytes/sec.
-    retries:           Number of times to retry for HTTP error 5xx
-    buffersize:        Size of download buffer in bytes.
-    noresizebuffer:    Do not automatically resize the download buffer.
-    continuedl:        Try to continue downloads if possible.
-    noprogress:        Do not print the progress bar.
-    logtostderr:       Log messages to stderr instead of stdout.
-    consoletitle:      Display progress in console window's titlebar.
-    nopart:            Do not use temporary .part files.
-    updatetime:        Use the Last-modified header to set output file timestamps.
-    test:              Download only first bytes to test the downloader.
-    min_filesize:      Skip files smaller than this size
-    max_filesize:      Skip files larger than this size
-    """
-
-    params = None
-
-    def __init__(self, ydl, params):
-        """Create a FileDownloader object with the given options."""
-        self.ydl = ydl
-        self._progress_hooks = []
-        self.params = params
-
-    @staticmethod
-    def format_seconds(seconds):
-        (mins, secs) = divmod(seconds, 60)
-        (hours, mins) = divmod(mins, 60)
-        if hours > 99:
-            return '--:--:--'
-        if hours == 0:
-            return '%02d:%02d' % (mins, secs)
-        else:
-            return '%02d:%02d:%02d' % (hours, mins, secs)
-
-    @staticmethod
-    def calc_percent(byte_counter, data_len):
-        if data_len is None:
-            return None
-        return float(byte_counter) / float(data_len) * 100.0
-
-    @staticmethod
-    def format_percent(percent):
-        if percent is None:
-            return '---.-%'
-        return '%6s' % ('%3.1f%%' % percent)
-
-    @staticmethod
-    def calc_eta(start, now, total, current):
-        if total is None:
-            return None
-        dif = now - start
-        if current == 0 or dif < 0.001: # One millisecond
-            return None
-        rate = float(current) / dif
-        return int((float(total) - float(current)) / rate)
-
-    @staticmethod
-    def format_eta(eta):
-        if eta is None:
-            return '--:--'
-        return FileDownloader.format_seconds(eta)
-
-    @staticmethod
-    def calc_speed(start, now, bytes):
-        dif = now - start
-        if bytes == 0 or dif < 0.001: # One millisecond
-            return None
-        return float(bytes) / dif
-
-    @staticmethod
-    def format_speed(speed):
-        if speed is None:
-            return '%10s' % '---b/s'
-        return '%10s' % ('%s/s' % format_bytes(speed))
-
-    @staticmethod
-    def best_block_size(elapsed_time, bytes):
-        new_min = max(bytes / 2.0, 1.0)
-        new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
-        if elapsed_time < 0.001:
-            return int(new_max)
-        rate = bytes / elapsed_time
-        if rate > new_max:
-            return int(new_max)
-        if rate < new_min:
-            return int(new_min)
-        return int(rate)
-
-    @staticmethod
-    def parse_bytes(bytestr):
-        """Parse a string indicating a byte quantity into an integer."""
-        matchobj = re.match(r'(?i)^(\d+(?:\.\d+)?)([kMGTPEZY]?)$', bytestr)
-        if matchobj is None:
-            return None
-        number = float(matchobj.group(1))
-        multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower())
-        return int(round(number * multiplier))
-
-    def to_screen(self, *args, **kargs):
-        self.ydl.to_screen(*args, **kargs)
-
-    def to_stderr(self, message):
-        self.ydl.to_screen(message)
-
-    def to_console_title(self, message):
-        self.ydl.to_console_title(message)
-
-    def trouble(self, *args, **kargs):
-        self.ydl.trouble(*args, **kargs)
-
-    def report_warning(self, *args, **kargs):
-        self.ydl.report_warning(*args, **kargs)
-
-    def report_error(self, *args, **kargs):
-        self.ydl.report_error(*args, **kargs)
-
-    def slow_down(self, start_time, byte_counter):
-        """Sleep if the download speed is over the rate limit."""
-        rate_limit = self.params.get('ratelimit', None)
-        if rate_limit is None or byte_counter == 0:
-            return
-        now = time.time()
-        elapsed = now - start_time
-        if elapsed <= 0.0:
-            return
-        speed = float(byte_counter) / elapsed
-        if speed > rate_limit:
-            time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)
-
-    def temp_name(self, filename):
-        """Returns a temporary filename for the given filename."""
-        if self.params.get('nopart', False) or filename == u'-' or \
-                (os.path.exists(encodeFilename(filename)) and not os.path.isfile(encodeFilename(filename))):
-            return filename
-        return filename + u'.part'
-
-    def undo_temp_name(self, filename):
-        if filename.endswith(u'.part'):
-            return filename[:-len(u'.part')]
-        return filename
-
-    def try_rename(self, old_filename, new_filename):
-        try:
-            if old_filename == new_filename:
-                return
-            os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
-        except (IOError, OSError):
-            self.report_error(u'unable to rename file')
-
-    def try_utime(self, filename, last_modified_hdr):
-        """Try to set the last-modified time of the given file."""
-        if last_modified_hdr is None:
-            return
-        if not os.path.isfile(encodeFilename(filename)):
-            return
-        timestr = last_modified_hdr
-        if timestr is None:
-            return
-        filetime = timeconvert(timestr)
-        if filetime is None:
-            return filetime
-        # Ignore obviously invalid dates
-        if filetime == 0:
-            return
-        try:
-            os.utime(filename, (time.time(), filetime))
-        except:
-            pass
-        return filetime
-
-    def report_destination(self, filename):
-        """Report destination filename."""
-        self.to_screen(u'[download] Destination: ' + filename)
-
-    def _report_progress_status(self, msg, is_last_line=False):
-        fullmsg = u'[download] ' + msg
-        if self.params.get('progress_with_newline', False):
-            self.to_screen(fullmsg)
-        else:
-            if os.name == 'nt':
-                prev_len = getattr(self, '_report_progress_prev_line_length',
-                                   0)
-                if prev_len > len(fullmsg):
-                    fullmsg += u' ' * (prev_len - len(fullmsg))
-                self._report_progress_prev_line_length = len(fullmsg)
-                clear_line = u'\r'
-            else:
-                clear_line = (u'\r\x1b[K' if sys.stderr.isatty() else u'\r')
-            self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)
-        self.to_console_title(u'youtube-dl ' + msg)
-
-    def report_progress(self, percent, data_len_str, speed, eta):
-        """Report download progress."""
-        if self.params.get('noprogress', False):
-            return
-        if eta is not None:
-            eta_str = self.format_eta(eta)
-        else:
-            eta_str = 'Unknown ETA'
-        if percent is not None:
-            percent_str = self.format_percent(percent)
-        else:
-            percent_str = 'Unknown %'
-        speed_str = self.format_speed(speed)
-
-        msg = (u'%s of %s at %s ETA %s' %
-               (percent_str, data_len_str, speed_str, eta_str))
-        self._report_progress_status(msg)
-
-    def report_progress_live_stream(self, downloaded_data_len, speed, elapsed):
-        if self.params.get('noprogress', False):
-            return
-        downloaded_str = format_bytes(downloaded_data_len)
-        speed_str = self.format_speed(speed)
-        elapsed_str = FileDownloader.format_seconds(elapsed)
-        msg = u'%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)
-        self._report_progress_status(msg)
-
-    def report_finish(self, data_len_str, tot_time):
-        """Report download finished."""
-        if self.params.get('noprogress', False):
-            self.to_screen(u'[download] Download completed')
-        else:
-            self._report_progress_status(
-                (u'100%% of %s in %s' %
-                 (data_len_str, self.format_seconds(tot_time))),
-                is_last_line=True)
-
-    def report_resuming_byte(self, resume_len):
-        """Report attempt to resume at given byte."""
-        self.to_screen(u'[download] Resuming download at byte %s' % resume_len)
-
-    def report_retry(self, count, retries):
-        """Report retry in case of HTTP error 5xx"""
-        self.to_screen(u'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))
-
-    def report_file_already_downloaded(self, file_name):
-        """Report file has already been fully downloaded."""
-        try:
-            self.to_screen(u'[download] %s has already been downloaded' % file_name)
-        except UnicodeEncodeError:
-            self.to_screen(u'[download] The file has already been downloaded')
-
-    def report_unable_to_resume(self):
-        """Report it was impossible to resume download."""
-        self.to_screen(u'[download] Unable to resume')
-
-    def _download_with_rtmpdump(self, filename, url, player_url, page_url, play_path, tc_url, live, conn):
-        def run_rtmpdump(args):
-            start = time.time()
-            resume_percent = None
-            resume_downloaded_data_len = None
-            proc = subprocess.Popen(args, stderr=subprocess.PIPE)
-            cursor_in_new_line = True
-            proc_stderr_closed = False
-            while not proc_stderr_closed:
-                # read line from stderr
-                line = u''
-                while True:
-                    char = proc.stderr.read(1)
-                    if not char:
-                        proc_stderr_closed = True
-                        break
-                    if char in [b'\r', b'\n']:
-                        break
-                    line += char.decode('ascii', 'replace')
-                if not line:
-                    # proc_stderr_closed is True
-                    continue
-                mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec \(([0-9]{1,2}\.[0-9])%\)', line)
-                if mobj:
-                    downloaded_data_len = int(float(mobj.group(1))*1024)
-                    percent = float(mobj.group(2))
-                    if not resume_percent:
-                        resume_percent = percent
-                        resume_downloaded_data_len = downloaded_data_len
-                    eta = self.calc_eta(start, time.time(), 100-resume_percent, percent-resume_percent)
-                    speed = self.calc_speed(start, time.time(), downloaded_data_len-resume_downloaded_data_len)
-                    data_len = None
-                    if percent > 0:
-                        data_len = int(downloaded_data_len * 100 / percent)
-                    data_len_str = u'~' + format_bytes(data_len)
-                    self.report_progress(percent, data_len_str, speed, eta)
-                    cursor_in_new_line = False
-                    self._hook_progress({
-                        'downloaded_bytes': downloaded_data_len,
-                        'total_bytes': data_len,
-                        'tmpfilename': tmpfilename,
-                        'filename': filename,
-                        'status': 'downloading',
-                        'eta': eta,
-                        'speed': speed,
-                    })
-                else:
-                    # no percent for live streams
-                    mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec', line)
-                    if mobj:
-                        downloaded_data_len = int(float(mobj.group(1))*1024)
-                        time_now = time.time()
-                        speed = self.calc_speed(start, time_now, downloaded_data_len)
-                        self.report_progress_live_stream(downloaded_data_len, speed, time_now - start)
-                        cursor_in_new_line = False
-                        self._hook_progress({
-                            'downloaded_bytes': downloaded_data_len,
-                            'tmpfilename': tmpfilename,
-                            'filename': filename,
-                            'status': 'downloading',
-                            'speed': speed,
-                        })
-                    elif self.params.get('verbose', False):
-                        if not cursor_in_new_line:
-                            self.to_screen(u'')
-                        cursor_in_new_line = True
-                        self.to_screen(u'[rtmpdump] '+line)
-            proc.wait()
-            if not cursor_in_new_line:
-                self.to_screen(u'')
-            return proc.returncode
-
-        self.report_destination(filename)
-        tmpfilename = self.temp_name(filename)
-        test = self.params.get('test', False)
-
-        # Check for rtmpdump first
-        try:
-            subprocess.call(['rtmpdump', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
-        except (OSError, IOError):
-            self.report_error(u'RTMP download detected but "rtmpdump" could not be run')
-            return False
-
-        # Download using rtmpdump. rtmpdump returns exit code 2 when
-        # the connection was interrumpted and resuming appears to be
-        # possible. This is part of rtmpdump's normal usage, AFAIK.
-        basic_args = ['rtmpdump', '--verbose', '-r', url, '-o', tmpfilename]
-        if player_url is not None:
-            basic_args += ['--swfVfy', player_url]
-        if page_url is not None:
-            basic_args += ['--pageUrl', page_url]
-        if play_path is not None:
-            basic_args += ['--playpath', play_path]
-        if tc_url is not None:
-            basic_args += ['--tcUrl', url]
-        if test:
-            basic_args += ['--stop', '1']
-        if live:
-            basic_args += ['--live']
-        if conn:
-            basic_args += ['--conn', conn]
-        args = basic_args + [[], ['--resume', '--skip', '1']][self.params.get('continuedl', False)]
-
-        if sys.platform == 'win32' and sys.version_info < (3, 0):
-            # Windows subprocess module does not actually support Unicode
-            # on Python 2.x
-            # See http://stackoverflow.com/a/9951851/35070
-            subprocess_encoding = sys.getfilesystemencoding()
-            args = [a.encode(subprocess_encoding, 'ignore') for a in args]
-        else:
-            subprocess_encoding = None
-
-        if self.params.get('verbose', False):
-            if subprocess_encoding:
-                str_args = [
-                    a.decode(subprocess_encoding) if isinstance(a, bytes) else a
-                    for a in args]
-            else:
-                str_args = args
-            try:
-                import pipes
-                shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
-            except ImportError:
-                shell_quote = repr
-            self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(str_args))
-
-        retval = run_rtmpdump(args)
-
-        while (retval == 2 or retval == 1) and not test:
-            prevsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'[rtmpdump] %s bytes' % prevsize)
-            time.sleep(5.0) # This seems to be needed
-            retval = run_rtmpdump(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
-            cursize = os.path.getsize(encodeFilename(tmpfilename))
-            if prevsize == cursize and retval == 1:
-                break
-             # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
-            if prevsize == cursize and retval == 2 and cursize > 1024:
-                self.to_screen(u'[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
-                retval = 0
-                break
-        if retval == 0 or (test and retval == 2):
-            fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'[rtmpdump] %s bytes' % fsize)
-            self.try_rename(tmpfilename, filename)
-            self._hook_progress({
-                'downloaded_bytes': fsize,
-                'total_bytes': fsize,
-                'filename': filename,
-                'status': 'finished',
-            })
-            return True
-        else:
-            self.to_stderr(u"\n")
-            self.report_error(u'rtmpdump exited with code %d' % retval)
-            return False
-
-    def _download_with_mplayer(self, filename, url):
-        self.report_destination(filename)
-        tmpfilename = self.temp_name(filename)
-
-        args = ['mplayer', '-really-quiet', '-vo', 'null', '-vc', 'dummy', '-dumpstream', '-dumpfile', tmpfilename, url]
-        # Check for mplayer first
-        try:
-            subprocess.call(['mplayer', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
-        except (OSError, IOError):
-            self.report_error(u'MMS or RTSP download detected but "%s" could not be run' % args[0] )
-            return False
-
-        # Download using mplayer. 
-        retval = subprocess.call(args)
-        if retval == 0:
-            fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
-            self.try_rename(tmpfilename, filename)
-            self._hook_progress({
-                'downloaded_bytes': fsize,
-                'total_bytes': fsize,
-                'filename': filename,
-                'status': 'finished',
-            })
-            return True
-        else:
-            self.to_stderr(u"\n")
-            self.report_error(u'mplayer exited with code %d' % retval)
-            return False
-
-    def _download_m3u8_with_ffmpeg(self, filename, url):
-        self.report_destination(filename)
-        tmpfilename = self.temp_name(filename)
-
-        args = ['-y', '-i', url, '-f', 'mp4', '-c', 'copy',
-            '-bsf:a', 'aac_adtstoasc', tmpfilename]
-
-        for program in ['avconv', 'ffmpeg']:
-            try:
-                subprocess.call([program, '-version'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
-                break
-            except (OSError, IOError):
-                pass
-        else:
-            self.report_error(u'm3u8 download detected but ffmpeg or avconv could not be found')
-        cmd = [program] + args
-
-        retval = subprocess.call(cmd)
-        if retval == 0:
-            fsize = os.path.getsize(encodeFilename(tmpfilename))
-            self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
-            self.try_rename(tmpfilename, filename)
-            self._hook_progress({
-                'downloaded_bytes': fsize,
-                'total_bytes': fsize,
-                'filename': filename,
-                'status': 'finished',
-            })
-            return True
-        else:
-            self.to_stderr(u"\n")
-            self.report_error(u'ffmpeg exited with code %d' % retval)
-            return False
 
 
+# This class reproduces the old behaviour of FileDownloader
+class FileDownloader(RealFileDownloader):
     def _do_download(self, filename, info_dict):
-        url = info_dict['url']
+        real_fd = get_suitable_downloader(info_dict)(self.ydl, self.params)
-
-        # Check file already present
-        if self.params.get('continuedl', False) and os.path.isfile(encodeFilename(filename)) and not self.params.get('nopart', False):
-            self.report_file_already_downloaded(filename)
-            self._hook_progress({
-                'filename': filename,
-                'status': 'finished',
-                'total_bytes': os.path.getsize(encodeFilename(filename)),
-            })
-            return True
-
-        # Attempt to download using rtmpdump
-        if url.startswith('rtmp'):
-            return self._download_with_rtmpdump(filename, url,
-                                                info_dict.get('player_url', None),
-                                                info_dict.get('page_url', None),
-                                                info_dict.get('play_path', None),
-                                                info_dict.get('tc_url', None),
-                                                info_dict.get('rtmp_live', False),
-                                                info_dict.get('rtmp_conn', None))
-
-        # Attempt to download using mplayer
-        if url.startswith('mms') or url.startswith('rtsp'):
-            return self._download_with_mplayer(filename, url)
-
-        # m3u8 manifest are downloaded with ffmpeg
-        if determine_ext(url) == u'm3u8':
-            return self._download_m3u8_with_ffmpeg(filename, url)
-
-        tmpfilename = self.temp_name(filename)
-        stream = None
-
-        # Do not include the Accept-Encoding header
-        headers = {'Youtubedl-no-compression': 'True'}
-        if 'user_agent' in info_dict:
-            headers['Youtubedl-user-agent'] = info_dict['user_agent']
-        basic_request = compat_urllib_request.Request(url, None, headers)
-        request = compat_urllib_request.Request(url, None, headers)
-
-        if self.params.get('test', False):
-            request.add_header('Range','bytes=0-10240')
-
-        # Establish possible resume length
-        if os.path.isfile(encodeFilename(tmpfilename)):
-            resume_len = os.path.getsize(encodeFilename(tmpfilename))
-        else:
-            resume_len = 0
-
-        open_mode = 'wb'
-        if resume_len != 0:
-            if self.params.get('continuedl', False):
-                self.report_resuming_byte(resume_len)
-                request.add_header('Range','bytes=%d-' % resume_len)
-                open_mode = 'ab'
-            else:
-                resume_len = 0
-
-        count = 0
-        retries = self.params.get('retries', 0)
-        while count <= retries:
-            # Establish connection
-            try:
-                if count == 0 and 'urlhandle' in info_dict:
-                    data = info_dict['urlhandle']
-                data = compat_urllib_request.urlopen(request)
-                break
-            except (compat_urllib_error.HTTPError, ) as err:
-                if (err.code < 500 or err.code >= 600) and err.code != 416:
-                    # Unexpected HTTP error
-                    raise
-                elif err.code == 416:
-                    # Unable to resume (requested range not satisfiable)
-                    try:
-                        # Open the connection again without the range header
-                        data = compat_urllib_request.urlopen(basic_request)
-                        content_length = data.info()['Content-Length']
-                    except (compat_urllib_error.HTTPError, ) as err:
-                        if err.code < 500 or err.code >= 600:
-                            raise
-                    else:
-                        # Examine the reported length
-                        if (content_length is not None and
-                                (resume_len - 100 < int(content_length) < resume_len + 100)):
-                            # The file had already been fully downloaded.
-                            # Explanation to the above condition: in issue #175 it was revealed that
-                            # YouTube sometimes adds or removes a few bytes from the end of the file,
-                            # changing the file size slightly and causing problems for some users. So
-                            # I decided to implement a suggested change and consider the file
-                            # completely downloaded if the file size differs less than 100 bytes from
-                            # the one in the hard drive.
-                            self.report_file_already_downloaded(filename)
-                            self.try_rename(tmpfilename, filename)
-                            self._hook_progress({
-                                'filename': filename,
-                                'status': 'finished',
-                            })
-                            return True
-                        else:
-                            # The length does not match, we start the download over
-                            self.report_unable_to_resume()
-                            open_mode = 'wb'
-                            break
-            # Retry
-            count += 1
-            if count <= retries:
-                self.report_retry(count, retries)
-
-        if count > retries:
-            self.report_error(u'giving up after %s retries' % retries)
-            return False
-
-        data_len = data.info().get('Content-length', None)
-        if data_len is not None:
-            data_len = int(data_len) + resume_len
-            min_data_len = self.params.get("min_filesize", None)
-            max_data_len =  self.params.get("max_filesize", None)
-            if min_data_len is not None and data_len < min_data_len:
-                self.to_screen(u'\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len, min_data_len))
-                return False
-            if max_data_len is not None and data_len > max_data_len:
-                self.to_screen(u'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
-                return False
-
-        data_len_str = format_bytes(data_len)
-        byte_counter = 0 + resume_len
-        block_size = self.params.get('buffersize', 1024)
-        start = time.time()
-        while True:
-            # Download and write
-            before = time.time()
-            data_block = data.read(block_size)
-            after = time.time()
-            if len(data_block) == 0:
-                break
-            byte_counter += len(data_block)
-
-            # Open file just in time
-            if stream is None:
-                try:
-                    (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode)
-                    assert stream is not None
-                    filename = self.undo_temp_name(tmpfilename)
-                    self.report_destination(filename)
-                except (OSError, IOError) as err:
-                    self.report_error(u'unable to open for writing: %s' % str(err))
-                    return False
-            try:
-                stream.write(data_block)
-            except (IOError, OSError) as err:
-                self.to_stderr(u"\n")
-                self.report_error(u'unable to write data: %s' % str(err))
-                return False
-            if not self.params.get('noresizebuffer', False):
-                block_size = self.best_block_size(after - before, len(data_block))
-
-            # Progress message
-            speed = self.calc_speed(start, time.time(), byte_counter - resume_len)
-            if data_len is None:
-                eta = percent = None
-            else:
-                percent = self.calc_percent(byte_counter, data_len)
-                eta = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
-            self.report_progress(percent, data_len_str, speed, eta)
-
-            self._hook_progress({
-                'downloaded_bytes': byte_counter,
-                'total_bytes': data_len,
-                'tmpfilename': tmpfilename,
-                'filename': filename,
-                'status': 'downloading',
-                'eta': eta,
-                'speed': speed,
-            })
-
-            # Apply rate limit
-            self.slow_down(start, byte_counter - resume_len)
-
-        if stream is None:
-            self.to_stderr(u"\n")
-            self.report_error(u'Did not get any data blocks')
-            return False
-        stream.close()
-        self.report_finish(data_len_str, (time.time() - start))
-        if data_len is not None and byte_counter != data_len:
-            raise ContentTooShortError(byte_counter, int(data_len))
-        self.try_rename(tmpfilename, filename)
-
-        # Update file modification time
-        if self.params.get('updatetime', True):
-            info_dict['filetime'] = self.try_utime(filename, data.info().get('last-modified', None))
-
-        self._hook_progress({
-            'downloaded_bytes': byte_counter,
-            'total_bytes': byte_counter,
-            'filename': filename,
-            'status': 'finished',
-        })
-
-        return True
-
-    def _hook_progress(self, status):
         for ph in self._progress_hooks:
-            ph(status)
+            real_fd.add_progress_hook(ph)
-
+        return real_fd.download(filename, info_dict)
-    def add_progress_hook(self, ph):
-        """ ph gets called on download progress, with a dictionary with the entries
-        * filename: The final filename
-        * status: One of "downloading" and "finished"
-
-        It can also have some of the following entries:
-
-        * downloaded_bytes: Bytes on disks
-        * total_bytes: Total bytes, None if unknown
-        * tmpfilename: The filename we're currently writing to
-        * eta: The estimated time in seconds, None if unknown
-        * speed: The download speed in bytes/second, None if unknown
-
-        Hooks are guaranteed to be called at least once (with status "finished")
-        if the download is successful.
-        """
-        self._progress_hooks.append(ph)

youtube_dl/YoutubeDL.py

@@ -1,7 +1,7 @@
 #!/usr/bin/env python
 # -*- coding: utf-8 -*-
 
-from __future__ import absolute_import
+from __future__ import absolute_import, unicode_literals
 
 import collections
 import errno
@@ -39,6 +39,7 @@ from .utils import (
     locked_file,
     make_HTTPS_handler,
     MaxDownloadsReached,
+    PagedList,
     PostProcessingError,
     platform_name,
     preferredencoding,
@@ -47,12 +48,15 @@ from .utils import (
     subtitles_filename,
     takewhile_inclusive,
     UnavailableVideoError,
+    url_basename,
     write_json_file,
     write_string,
     YoutubeDLHandler,
+    prepend_extension,
 )
 from .extractor import get_info_extractor, gen_extractors
-from .FileDownloader import FileDownloader
+from .downloader import get_suitable_downloader
+from .postprocessor import FFmpegMergerPP
 from .version import __version__
 
 
@@ -147,11 +151,19 @@ class YoutubeDL(object):
     socket_timeout:    Time to wait for unresponsive hosts, in seconds
     bidi_workaround:   Work around buggy terminals without bidirectional text
                        support, using fridibi
+    debug_printtraffic:Print out sent and received HTTP traffic
+    include_ads:       Download ads as well
+    default_search:    Prepend this string if an input url is not valid.
+                       'auto' for elaborate guessing
 
     The following parameters are not used by YoutubeDL itself, they are used by
     the FileDownloader:
     nopart, updatetime, buffersize, ratelimit, min_filesize, max_filesize, test,
     noresizebuffer, retries, continuedl, noprogress, consoletitle
+
+    The following options are used by the post processors:
+    prefer_ffmpeg:     If True, use ffmpeg instead of avconv if both are available,
+                       otherwise prefer avconv.
     """
 
     params = None
@@ -163,6 +175,8 @@ class YoutubeDL(object):
 
     def __init__(self, params=None):
         """Create a FileDownloader object with the given options."""
+        if params is None:
+            params = {}
         self._ies = []
         self._ies_instances = {}
         self._pps = []
@@ -171,7 +185,7 @@ class YoutubeDL(object):
         self._num_downloads = 0
         self._screen_file = [sys.stdout, sys.stderr][params.get('logtostderr', False)]
         self._err_file = sys.stderr
-        self.params = {} if params is None else params
+        self.params = params
 
         if params.get('bidi_workaround', False):
             try:
@@ -182,15 +196,21 @@ class YoutubeDL(object):
                     width_args = []
                 else:
                     width_args = ['-w', str(width)]
-                self._fribidi = subprocess.Popen(
+                sp_kwargs = dict(
-                    ['fribidi', '-c', 'UTF-8'] + width_args,
                     stdin=subprocess.PIPE,
                     stdout=slave,
                     stderr=self._err_file)
-                self._fribidi_channel = os.fdopen(master, 'rb')
+                try:
+                    self._output_process = subprocess.Popen(
+                        ['bidiv'] + width_args, **sp_kwargs
+                    )
+                except OSError:
+                    self._output_process = subprocess.Popen(
+                        ['fribidi', '-c', 'UTF-8'] + width_args, **sp_kwargs)
+                self._output_channel = os.fdopen(master, 'rb')
             except OSError as ose:
                 if ose.errno == 2:
-                    self.report_warning(u'Could not find fribidi executable, ignoring --bidi-workaround . Make sure that  fribidi  is an executable file in one of the directories in your $PATH.')
+                    self.report_warning('Could not find fribidi executable, ignoring --bidi-workaround . Make sure that  fribidi  is an executable file in one of the directories in your $PATH.')
                 else:
                     raise
 
@@ -199,15 +219,13 @@ class YoutubeDL(object):
                 and not params['restrictfilenames']):
             # On Python 3, the Unicode filesystem API will throw errors (#1474)
             self.report_warning(
-                u'Assuming --restrict-filenames since file system encoding '
+                'Assuming --restrict-filenames since file system encoding '
-                u'cannot encode all charactes. '
+                'cannot encode all charactes. '
-                u'Set the LC_ALL environment variable to fix this.')
+                'Set the LC_ALL environment variable to fix this.')
             self.params['restrictfilenames'] = True
 
-        self.fd = FileDownloader(self, self.params)
-
         if '%(stitle)s' in self.params.get('outtmpl', ''):
-            self.report_warning(u'%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
+            self.report_warning('%(stitle)s is deprecated. Use the %(title)s and the --restrict-filenames flag(which also secures %(uploader)s et al) instead.')
 
         self._setup_opener()
 
@@ -241,17 +259,22 @@ class YoutubeDL(object):
         self._pps.append(pp)
         pp.set_downloader(self)
 
+    def add_progress_hook(self, ph):
+        """Add the progress hook (currently only for the file downloader)"""
+        self._progress_hooks.append(ph)
+
     def _bidi_workaround(self, message):
-        if not hasattr(self, '_fribidi_channel'):
+        if not hasattr(self, '_output_channel'):
             return message
 
-        assert type(message) == type(u'')
+        assert hasattr(self, '_output_process')
-        line_count = message.count(u'\n') + 1
+        assert type(message) == type('')
-        self._fribidi.stdin.write((message + u'\n').encode('utf-8'))
+        line_count = message.count('\n') + 1
-        self._fribidi.stdin.flush()
+        self._output_process.stdin.write((message + '\n').encode('utf-8'))
-        res = u''.join(self._fribidi_channel.readline().decode('utf-8')
+        self._output_process.stdin.flush()
+        res = ''.join(self._output_channel.readline().decode('utf-8')
                        for _ in range(line_count))
-        return res[:-len(u'\n')]
+        return res[:-len('\n')]
 
     def to_screen(self, message, skip_eol=False):
         """Print message to stdout if not in quiet mode."""
@@ -263,19 +286,19 @@ class YoutubeDL(object):
             self.params['logger'].debug(message)
         elif not check_quiet or not self.params.get('quiet', False):
             message = self._bidi_workaround(message)
-            terminator = [u'\n', u''][skip_eol]
+            terminator = ['\n', ''][skip_eol]
             output = message + terminator
 
             write_string(output, self._screen_file)
 
     def to_stderr(self, message):
         """Print message to stderr."""
-        assert type(message) == type(u'')
+        assert type(message) == type('')
         if self.params.get('logger'):
             self.params['logger'].error(message)
         else:
             message = self._bidi_workaround(message)
-            output = message + u'\n'
+            output = message + '\n'
             write_string(output, self._err_file)
 
     def to_console_title(self, message):
@@ -286,21 +309,21 @@ class YoutubeDL(object):
             # already of type unicode()
             ctypes.windll.kernel32.SetConsoleTitleW(ctypes.c_wchar_p(message))
         elif 'TERM' in os.environ:
-            write_string(u'\033]0;%s\007' % message, self._screen_file)
+            write_string('\033]0;%s\007' % message, self._screen_file)
 
     def save_console_title(self):
         if not self.params.get('consoletitle', False):
             return
         if 'TERM' in os.environ:
             # Save the title on stack
-            write_string(u'\033[22;0t', self._screen_file)
+            write_string('\033[22;0t', self._screen_file)
 
     def restore_console_title(self):
         if not self.params.get('consoletitle', False):
             return
         if 'TERM' in os.environ:
             # Restore the title from stack
-            write_string(u'\033[23;0t', self._screen_file)
+            write_string('\033[23;0t', self._screen_file)
 
     def __enter__(self):
         self.save_console_title()
@@ -326,13 +349,13 @@ class YoutubeDL(object):
         if self.params.get('verbose'):
             if tb is None:
                 if sys.exc_info()[0]:  # if .trouble has been called from an except block
-                    tb = u''
+                    tb = ''
                     if hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
-                        tb += u''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
+                        tb += ''.join(traceback.format_exception(*sys.exc_info()[1].exc_info))
                     tb += compat_str(traceback.format_exc())
                 else:
                     tb_data = traceback.format_list(traceback.extract_stack())
-                    tb = u''.join(tb_data)
+                    tb = ''.join(tb_data)
             self.to_stderr(tb)
         if not self.params.get('ignoreerrors', False):
             if sys.exc_info()[0] and hasattr(sys.exc_info()[1], 'exc_info') and sys.exc_info()[1].exc_info[0]:
@@ -348,10 +371,10 @@ class YoutubeDL(object):
         If stderr is a tty file the 'WARNING:' will be colored
         '''
         if self._err_file.isatty() and os.name != 'nt':
-            _msg_header = u'\033[0;33mWARNING:\033[0m'
+            _msg_header = '\033[0;33mWARNING:\033[0m'
         else:
-            _msg_header = u'WARNING:'
+            _msg_header = 'WARNING:'
-        warning_message = u'%s %s' % (_msg_header, message)
+        warning_message = '%s %s' % (_msg_header, message)
         self.to_stderr(warning_message)
 
     def report_error(self, message, tb=None):
@@ -360,22 +383,18 @@ class YoutubeDL(object):
         in red if stderr is a tty file.
         '''
         if self._err_file.isatty() and os.name != 'nt':
-            _msg_header = u'\033[0;31mERROR:\033[0m'
+            _msg_header = '\033[0;31mERROR:\033[0m'
         else:
-            _msg_header = u'ERROR:'
+            _msg_header = 'ERROR:'
-        error_message = u'%s %s' % (_msg_header, message)
+        error_message = '%s %s' % (_msg_header, message)
         self.trouble(error_message, tb)
 
     def report_file_already_downloaded(self, file_name):
         """Report file has already been fully downloaded."""
         try:
-            self.to_screen(u'[download] %s has already been downloaded' % file_name)
+            self.to_screen('[download] %s has already been downloaded' % file_name)
         except UnicodeEncodeError:
-            self.to_screen(u'[download] The file has already been downloaded')
+            self.to_screen('[download] The file has already been downloaded')
-
-    def increment_downloads(self):
-        """Increment the ordinal that assigns a number to each file."""
-        self._num_downloads += 1
 
     def prepare_filename(self, info_dict):
         """Generate the output filename."""
@@ -386,61 +405,61 @@ class YoutubeDL(object):
             autonumber_size = self.params.get('autonumber_size')
             if autonumber_size is None:
                 autonumber_size = 5
-            autonumber_templ = u'%0' + str(autonumber_size) + u'd'
+            autonumber_templ = '%0' + str(autonumber_size) + 'd'
             template_dict['autonumber'] = autonumber_templ % self._num_downloads
             if template_dict.get('playlist_index') is not None:
-                template_dict['playlist_index'] = u'%05d' % template_dict['playlist_index']
+                template_dict['playlist_index'] = '%05d' % template_dict['playlist_index']
 
             sanitize = lambda k, v: sanitize_filename(
                 compat_str(v),
                 restricted=self.params.get('restrictfilenames'),
-                is_id=(k == u'id'))
+                is_id=(k == 'id'))
             template_dict = dict((k, sanitize(k, v))
                                  for k, v in template_dict.items()
                                  if v is not None)
-            template_dict = collections.defaultdict(lambda: u'NA', template_dict)
+            template_dict = collections.defaultdict(lambda: 'NA', template_dict)
 
             tmpl = os.path.expanduser(self.params['outtmpl'])
             filename = tmpl % template_dict
             return filename
         except ValueError as err:
-            self.report_error(u'Error in output template: ' + str(err) + u' (encoding: ' + repr(preferredencoding()) + ')')
+            self.report_error('Error in output template: ' + str(err) + ' (encoding: ' + repr(preferredencoding()) + ')')
             return None
 
     def _match_entry(self, info_dict):
         """ Returns None iff the file should be downloaded """
 
-        video_title = info_dict.get('title', info_dict.get('id', u'video'))
+        video_title = info_dict.get('title', info_dict.get('id', 'video'))
         if 'title' in info_dict:
             # This can happen when we're just evaluating the playlist
             title = info_dict['title']
             matchtitle = self.params.get('matchtitle', False)
             if matchtitle:
                 if not re.search(matchtitle, title, re.IGNORECASE):
-                    return u'"' + title + '" title did not match pattern "' + matchtitle + '"'
+                    return '"' + title + '" title did not match pattern "' + matchtitle + '"'
             rejecttitle = self.params.get('rejecttitle', False)
             if rejecttitle:
                 if re.search(rejecttitle, title, re.IGNORECASE):
-                    return u'"' + title + '" title matched reject pattern "' + rejecttitle + '"'
+                    return '"' + title + '" title matched reject pattern "' + rejecttitle + '"'
         date = info_dict.get('upload_date', None)
         if date is not None:
             dateRange = self.params.get('daterange', DateRange())
             if date not in dateRange:
-                return u'%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
+                return '%s upload date is not in range %s' % (date_from_str(date).isoformat(), dateRange)
         view_count = info_dict.get('view_count', None)
         if view_count is not None:
             min_views = self.params.get('min_views')
             if min_views is not None and view_count < min_views:
-                return u'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
+                return 'Skipping %s, because it has not reached minimum view count (%d/%d)' % (video_title, view_count, min_views)
             max_views = self.params.get('max_views')
             if max_views is not None and view_count > max_views:
-                return u'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
+                return 'Skipping %s, because it has exceeded the maximum view count (%d/%d)' % (video_title, view_count, max_views)
         age_limit = self.params.get('age_limit')
         if age_limit is not None:
             if age_limit < info_dict.get('age_limit', 0):
-                return u'Skipping "' + title + '" because it is age restricted'
+                return 'Skipping "' + title + '" because it is age restricted'
         if self.in_download_archive(info_dict):
-            return u'%s has already been recorded in archive' % video_title
+            return '%s has already been recorded in archive' % video_title
         return None
 
     @staticmethod
@@ -467,8 +486,8 @@ class YoutubeDL(object):
                 continue
 
             if not ie.working():
-                self.report_warning(u'The program functionality for this site has been marked as broken, '
+                self.report_warning('The program functionality for this site has been marked as broken, '
-                                    u'and will probably not work.')
+                                    'and will probably not work.')
 
             try:
                 ie_result = ie.extract(url)
@@ -484,6 +503,7 @@ class YoutubeDL(object):
                     {
                         'extractor': ie.IE_NAME,
                         'webpage_url': url,
+                        'webpage_url_basename': url_basename(url),
                         'extractor_key': ie.ie_key(),
                     })
                 if process:
@@ -493,6 +513,8 @@ class YoutubeDL(object):
             except ExtractorError as de: # An error we somewhat expected
                 self.report_error(compat_str(de), de.format_traceback())
                 break
+            except MaxDownloadsReached:
+                raise
             except Exception as e:
                 if self.params.get('ignoreerrors', False):
                     self.report_error(compat_str(e), tb=compat_str(traceback.format_exc()))
@@ -500,7 +522,7 @@ class YoutubeDL(object):
                 else:
                     raise
         else:
-            self.report_error(u'no suitable InfoExtractor: %s' % url)
+            self.report_error('no suitable InfoExtractor: %s' % url)
 
     def process_ie_result(self, ie_result, download=True, extra_info={}):
         """
@@ -531,7 +553,7 @@ class YoutubeDL(object):
             def make_result(embedded_info):
                 new_result = ie_result.copy()
                 for f in ('_type', 'url', 'ext', 'player_url', 'formats',
-                          'entries', 'urlhandle', 'ie_key', 'duration',
+                          'entries', 'ie_key', 'duration',
                           'subtitles', 'annotations', 'format',
                           'thumbnail', 'thumbnails'):
                     if f in new_result:
@@ -551,37 +573,46 @@ class YoutubeDL(object):
         elif result_type == 'playlist':
             # We process each entry in the playlist
             playlist = ie_result.get('title', None) or ie_result.get('id', None)
-            self.to_screen(u'[download] Downloading playlist: %s' % playlist)
+            self.to_screen('[download] Downloading playlist: %s' % playlist)
 
             playlist_results = []
 
-            n_all_entries = len(ie_result['entries'])
             playliststart = self.params.get('playliststart', 1) - 1
             playlistend = self.params.get('playlistend', None)
             # For backwards compatibility, interpret -1 as whole list
             if playlistend == -1:
                 playlistend = None
 
+            if isinstance(ie_result['entries'], list):
+                n_all_entries = len(ie_result['entries'])
                 entries = ie_result['entries'][playliststart:playlistend]
                 n_entries = len(entries)
-
                 self.to_screen(
-                u"[%s] playlist '%s': Collected %d video ids (downloading %d of them)" %
+                    "[%s] playlist %s: Collected %d video ids (downloading %d of them)" %
                     (ie_result['extractor'], playlist, n_all_entries, n_entries))
+            else:
+                assert isinstance(ie_result['entries'], PagedList)
+                entries = ie_result['entries'].getslice(
+                    playliststart, playlistend)
+                n_entries = len(entries)
+                self.to_screen(
+                    "[%s] playlist %s: Downloading %d videos" %
+                    (ie_result['extractor'], playlist, n_entries))
 
             for i, entry in enumerate(entries, 1):
-                self.to_screen(u'[download] Downloading video #%s of %s' % (i, n_entries))
+                self.to_screen('[download] Downloading video #%s of %s' % (i, n_entries))
                 extra = {
                     'playlist': playlist,
                     'playlist_index': i + playliststart,
                     'extractor': ie_result['extractor'],
                     'webpage_url': ie_result['webpage_url'],
+                    'webpage_url_basename': url_basename(ie_result['webpage_url']),
                     'extractor_key': ie_result['extractor_key'],
                 }
 
                 reason = self._match_entry(entry)
                 if reason is not None:
-                    self.to_screen(u'[download] ' + reason)
+                    self.to_screen('[download] ' + reason)
                     continue
 
                 entry_result = self.process_ie_result(entry,
@@ -596,6 +627,7 @@ class YoutubeDL(object):
                     {
                         'extractor': ie_result['extractor'],
                         'webpage_url': ie_result['webpage_url'],
+                        'webpage_url_basename': url_basename(ie_result['webpage_url']),
                         'extractor_key': ie_result['extractor_key'],
                     })
                 return r
@@ -612,8 +644,20 @@ class YoutubeDL(object):
             return available_formats[-1]
         elif format_spec == 'worst':
             return available_formats[0]
+        elif format_spec == 'bestaudio':
+            audio_formats = [
+                f for f in available_formats
+                if f.get('vcodec') == 'none']
+            if audio_formats:
+                return audio_formats[-1]
+        elif format_spec == 'worstaudio':
+            audio_formats = [
+                f for f in available_formats
+                if f.get('vcodec') == 'none']
+            if audio_formats:
+                return audio_formats[0]
         else:
-            extensions = [u'mp4', u'flv', u'webm', u'3gp']
+            extensions = ['mp4', 'flv', 'webm', '3gp']
             if format_spec in extensions:
                 filter_f = lambda f: f['ext'] == format_spec
             else:
@@ -632,7 +676,7 @@ class YoutubeDL(object):
             info_dict['playlist_index'] = None
 
         # This extractors handle format selection themselves
-        if info_dict['extractor'] in [u'youtube', u'Youku']:
+        if info_dict['extractor'] in ['Youku']:
             if download:
                 self.process_info(info_dict)
             return info_dict
@@ -649,35 +693,34 @@ class YoutubeDL(object):
             if format.get('format_id') is None:
                 format['format_id'] = compat_str(i)
             if format.get('format') is None:
-                format['format'] = u'{id} - {res}{note}'.format(
+                format['format'] = '{id} - {res}{note}'.format(
                     id=format['format_id'],
                     res=self.format_resolution(format),
-                    note=u' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
+                    note=' ({0})'.format(format['format_note']) if format.get('format_note') is not None else '',
                 )
             # Automatically determine file extension if missing
             if 'ext' not in format:
                 format['ext'] = determine_ext(format['url'])
 
-        if self.params.get('listformats', None):
-            self.list_formats(info_dict)
-            return
-
         format_limit = self.params.get('format_limit', None)
         if format_limit:
             formats = list(takewhile_inclusive(
                 lambda f: f['format_id'] != format_limit, formats
             ))
-        if self.params.get('prefer_free_formats'):
-            def _free_formats_key(f):
-                try:
-                    ext_ord = [u'flv', u'mp4', u'webm'].index(f['ext'])
-                except ValueError:
-                    ext_ord = -1
-                # We only compare the extension if they have the same height and width
-                return (f.get('height'), f.get('width'), ext_ord)
-            formats = sorted(formats, key=_free_formats_key)
 
-        req_format = self.params.get('format', 'best')
+        # TODO Central sorting goes here
+
+        if formats[0] is not info_dict:
+            # only set the 'formats' fields if the original info_dict list them
+            # otherwise we end up with a circular reference, the first (and unique)
+            # element in the 'formats' field in info_dict is info_dict itself,
+            # wich can't be exported to json
+            info_dict['formats'] = formats
+        if self.params.get('listformats', None):
+            self.list_formats(info_dict)
+            return
+
+        req_format = self.params.get('format')
         if req_format is None:
             req_format = 'best'
         formats_to_download = []
@@ -685,21 +728,35 @@ class YoutubeDL(object):
         if req_format in ('-1', 'all'):
             formats_to_download = formats
         else:
-            # We can accept formats requestd in the format: 34/5/best, we pick
+            # We can accept formats requested in the format: 34/5/best, we pick
             # the first that is available, starting from left
             req_formats = req_format.split('/')
             for rf in req_formats:
+                if re.match(r'.+?\+.+?', rf) is not None:
+                    # Two formats have been requested like '137+139'
+                    format_1, format_2 = rf.split('+')
+                    formats_info = (self.select_format(format_1, formats),
+                        self.select_format(format_2, formats))
+                    if all(formats_info):
+                        selected_format = {
+                            'requested_formats': formats_info,
+                            'format': rf,
+                            'ext': formats_info[0]['ext'],
+                        }
+                    else:
+                        selected_format = None
+                else:
                     selected_format = self.select_format(rf, formats)
                 if selected_format is not None:
                     formats_to_download = [selected_format]
                     break
         if not formats_to_download:
-            raise ExtractorError(u'requested format not available',
+            raise ExtractorError('requested format not available',
                                  expected=True)
 
         if download:
             if len(formats_to_download) > 1:
-                self.to_screen(u'[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
+                self.to_screen('[info] %s: downloading video in %s formats' % (info_dict['id'], len(formats_to_download)))
             for format in formats_to_download:
                 new_info = dict(info_dict)
                 new_info.update(format)
@@ -712,12 +769,15 @@ class YoutubeDL(object):
         """Process a single resolved IE result."""
 
         assert info_dict.get('_type', 'video') == 'video'
-        #We increment the download the download count here to match the previous behaviour.
+
-        self.increment_downloads()
+        max_downloads = self.params.get('max_downloads')
+        if max_downloads is not None:
+            if self._num_downloads >= int(max_downloads):
+                raise MaxDownloadsReached()
 
         info_dict['fulltitle'] = info_dict['title']
         if len(info_dict['title']) > 200:
-            info_dict['title'] = info_dict['title'][:197] + u'...'
+            info_dict['title'] = info_dict['title'][:197] + '...'
 
         # Keep for backwards compatibility
         info_dict['stitle'] = info_dict['title']
@@ -727,13 +787,10 @@ class YoutubeDL(object):
 
         reason = self._match_entry(info_dict)
         if reason is not None:
-            self.to_screen(u'[download] ' + reason)
+            self.to_screen('[download] ' + reason)
             return
 
-        max_downloads = self.params.get('max_downloads')
+        self._num_downloads += 1
-        if max_downloads is not None:
-            if self._num_downloads > int(max_downloads):
-                raise MaxDownloadsReached()
 
         filename = self.prepare_filename(info_dict)
 
@@ -744,7 +801,7 @@ class YoutubeDL(object):
             self.to_stdout(info_dict['id'])
         if self.params.get('forceurl', False):
             # For RTMP URLs, also include the playpath
-            self.to_stdout(info_dict['url'] + info_dict.get('play_path', u''))
+            self.to_stdout(info_dict['url'] + info_dict.get('play_path', ''))
         if self.params.get('forcethumbnail', False) and info_dict.get('thumbnail') is not None:
             self.to_stdout(info_dict['thumbnail'])
         if self.params.get('forcedescription', False) and info_dict.get('description') is not None:
@@ -771,37 +828,37 @@ class YoutubeDL(object):
             if dn != '' and not os.path.exists(dn):
                 os.makedirs(dn)
         except (OSError, IOError) as err:
-            self.report_error(u'unable to create directory ' + compat_str(err))
+            self.report_error('unable to create directory ' + compat_str(err))
             return
 
         if self.params.get('writedescription', False):
-            descfn = filename + u'.description'
+            descfn = filename + '.description'
             if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(descfn)):
-                self.to_screen(u'[info] Video description is already present')
+                self.to_screen('[info] Video description is already present')
             else:
                 try:
-                    self.to_screen(u'[info] Writing video description to: ' + descfn)
+                    self.to_screen('[info] Writing video description to: ' + descfn)
                     with io.open(encodeFilename(descfn), 'w', encoding='utf-8') as descfile:
                         descfile.write(info_dict['description'])
                 except (KeyError, TypeError):
-                    self.report_warning(u'There\'s no description to write.')
+                    self.report_warning('There\'s no description to write.')
                 except (OSError, IOError):
-                    self.report_error(u'Cannot write description file ' + descfn)
+                    self.report_error('Cannot write description file ' + descfn)
                     return
 
         if self.params.get('writeannotations', False):
-            annofn = filename + u'.annotations.xml'
+            annofn = filename + '.annotations.xml'
             if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(annofn)):
-                self.to_screen(u'[info] Video annotations are already present')
+                self.to_screen('[info] Video annotations are already present')
             else:
                 try:
-                    self.to_screen(u'[info] Writing video annotations to: ' + annofn)
+                    self.to_screen('[info] Writing video annotations to: ' + annofn)
                     with io.open(encodeFilename(annofn), 'w', encoding='utf-8') as annofile:
                         annofile.write(info_dict['annotations'])
                 except (KeyError, TypeError):
-                    self.report_warning(u'There are no annotations to write.')
+                    self.report_warning('There are no annotations to write.')
                 except (OSError, IOError):
-                    self.report_error(u'Cannot write annotations file: ' + annofn)
+                    self.report_error('Cannot write annotations file: ' + annofn)
                     return
 
         subtitles_are_requested = any([self.params.get('writesubtitles', False),
@@ -819,46 +876,45 @@ class YoutubeDL(object):
                 try:
                     sub_filename = subtitles_filename(filename, sub_lang, sub_format)
                     if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(sub_filename)):
-                        self.to_screen(u'[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
+                        self.to_screen('[info] Video subtitle %s.%s is already_present' % (sub_lang, sub_format))
                     else:
-                        self.to_screen(u'[info] Writing video subtitles to: ' + sub_filename)
+                        self.to_screen('[info] Writing video subtitles to: ' + sub_filename)
                         with io.open(encodeFilename(sub_filename), 'w', encoding='utf-8') as subfile:
                                 subfile.write(sub)
                 except (OSError, IOError):
-                    self.report_error(u'Cannot write subtitles file ' + descfn)
+                    self.report_error('Cannot write subtitles file ' + descfn)
                     return
 
         if self.params.get('writeinfojson', False):
-            infofn = os.path.splitext(filename)[0] + u'.info.json'
+            infofn = os.path.splitext(filename)[0] + '.info.json'
             if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(infofn)):
-                self.to_screen(u'[info] Video description metadata is already present')
+                self.to_screen('[info] Video description metadata is already present')
             else:
-                self.to_screen(u'[info] Writing video description metadata as JSON to: ' + infofn)
+                self.to_screen('[info] Writing video description metadata as JSON to: ' + infofn)
                 try:
-                    json_info_dict = dict((k, v) for k, v in info_dict.items() if not k in ['urlhandle'])
+                    write_json_file(info_dict, encodeFilename(infofn))
-                    write_json_file(json_info_dict, encodeFilename(infofn))
                 except (OSError, IOError):
-                    self.report_error(u'Cannot write metadata to JSON file ' + infofn)
+                    self.report_error('Cannot write metadata to JSON file ' + infofn)
                     return
 
         if self.params.get('writethumbnail', False):
             if info_dict.get('thumbnail') is not None:
-                thumb_format = determine_ext(info_dict['thumbnail'], u'jpg')
+                thumb_format = determine_ext(info_dict['thumbnail'], 'jpg')
-                thumb_filename = os.path.splitext(filename)[0] + u'.' + thumb_format
+                thumb_filename = os.path.splitext(filename)[0] + '.' + thumb_format
                 if self.params.get('nooverwrites', False) and os.path.exists(encodeFilename(thumb_filename)):
-                    self.to_screen(u'[%s] %s: Thumbnail is already present' %
+                    self.to_screen('[%s] %s: Thumbnail is already present' %
                                    (info_dict['extractor'], info_dict['id']))
                 else:
-                    self.to_screen(u'[%s] %s: Downloading thumbnail ...' %
+                    self.to_screen('[%s] %s: Downloading thumbnail ...' %
                                    (info_dict['extractor'], info_dict['id']))
                     try:
                         uf = compat_urllib_request.urlopen(info_dict['thumbnail'])
                         with open(thumb_filename, 'wb') as thumbf:
                             shutil.copyfileobj(uf, thumbf)
-                        self.to_screen(u'[%s] %s: Writing thumbnail to: %s' %
+                        self.to_screen('[%s] %s: Writing thumbnail to: %s' %
                             (info_dict['extractor'], info_dict['id'], thumb_filename))
                     except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                        self.report_warning(u'Unable to download thumbnail "%s": %s' %
+                        self.report_warning('Unable to download thumbnail "%s": %s' %
                             (info_dict['thumbnail'], compat_str(err)))
 
         if not self.params.get('skip_download', False):
@@ -866,21 +922,49 @@ class YoutubeDL(object):
                 success = True
             else:
                 try:
-                    success = self.fd._do_download(filename, info_dict)
+                    def dl(name, info):
+                        fd = get_suitable_downloader(info)(self, self.params)
+                        for ph in self._progress_hooks:
+                            fd.add_progress_hook(ph)
+                        return fd.download(name, info)
+                    if info_dict.get('requested_formats') is not None:
+                        downloaded = []
+                        success = True
+                        merger = FFmpegMergerPP(self)
+                        if not merger._get_executable():
+                            postprocessors = []
+                            self.report_warning('You have requested multiple '
+                                'formats but ffmpeg or avconv are not installed.'
+                                ' The formats won\'t be merged')
+                        else:
+                            postprocessors = [merger]
+                        for f in info_dict['requested_formats']:
+                            new_info = dict(info_dict)
+                            new_info.update(f)
+                            fname = self.prepare_filename(new_info)
+                            fname = prepend_extension(fname, 'f%s' % f['format_id'])
+                            downloaded.append(fname)
+                            partial_success = dl(fname, new_info)
+                            success = success and partial_success
+                        info_dict['__postprocessors'] = postprocessors
+                        info_dict['__files_to_merge'] = downloaded
+                    else:
+                        # Just a single file
+                        success = dl(filename, info_dict)
                 except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                    self.report_error(u'unable to download video data: %s' % str(err))
+                    self.report_error('unable to download video data: %s' % str(err))
                     return
                 except (OSError, IOError) as err:
                     raise UnavailableVideoError(err)
                 except (ContentTooShortError, ) as err:
-                    self.report_error(u'content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
+                    self.report_error('content too short (expected %s bytes and served %s)' % (err.expected, err.downloaded))
                     return
 
             if success:
                 try:
                     self.post_process(filename, info_dict)
                 except (PostProcessingError) as err:
-                    self.report_error(u'postprocessing: %s' % str(err))
+                    self.report_error('postprocessing: %s' % str(err))
                     return
 
         self.record_download_archive(info_dict)
@@ -897,9 +981,9 @@ class YoutubeDL(object):
                 #It also downloads the videos
                 self.extract_info(url)
             except UnavailableVideoError:
-                self.report_error(u'unable to download video')
+                self.report_error('unable to download video')
             except MaxDownloadsReached:
-                self.to_screen(u'[info] Maximum number of downloaded files reached.')
+                self.to_screen('[info] Maximum number of downloaded files reached.')
                 raise
 
         return self._download_retcode
@@ -912,7 +996,7 @@ class YoutubeDL(object):
         except DownloadError:
             webpage_url = info.get('webpage_url')
             if webpage_url is not None:
-                self.report_warning(u'The info failed to download, trying with "%s"' % webpage_url)
+                self.report_warning('The info failed to download, trying with "%s"' % webpage_url)
                 return self.download([webpage_url])
             else:
                 raise
@@ -923,7 +1007,11 @@ class YoutubeDL(object):
         info = dict(ie_info)
         info['filepath'] = filename
         keep_video = None
-        for pp in self._pps:
+        pps_chain = []
+        if ie_info.get('__postprocessors') is not None:
+            pps_chain.extend(ie_info['__postprocessors'])
+        pps_chain.extend(self._pps)
+        for pp in pps_chain:
             try:
                 keep_video_wish, new_info = pp.run(info)
                 if keep_video_wish is not None:
@@ -936,10 +1024,10 @@ class YoutubeDL(object):
                 self.report_error(e.msg)
         if keep_video is False and not self.params.get('keepvideo', False):
             try:
-                self.to_screen(u'Deleting original file %s (pass -k to keep)' % filename)
+                self.to_screen('Deleting original file %s (pass -k to keep)' % filename)
                 os.remove(encodeFilename(filename))
             except (IOError, OSError):
-                self.report_warning(u'Unable to remove downloaded video file')
+                self.report_warning('Unable to remove downloaded video file')
 
     def _make_archive_id(self, info_dict):
         # Future-proof against any change in case
@@ -950,7 +1038,7 @@ class YoutubeDL(object):
                 extractor = info_dict.get('ie_key')  # key in a playlist
         if extractor is None:
             return None  # Incomplete video information
-        return extractor.lower() + u' ' + info_dict['id']
+        return extractor.lower() + ' ' + info_dict['id']
 
     def in_download_archive(self, info_dict):
         fn = self.params.get('download_archive')
@@ -978,53 +1066,72 @@ class YoutubeDL(object):
         vid_id = self._make_archive_id(info_dict)
         assert vid_id
         with locked_file(fn, 'a', encoding='utf-8') as archive_file:
-            archive_file.write(vid_id + u'\n')
+            archive_file.write(vid_id + '\n')
 
     @staticmethod
     def format_resolution(format, default='unknown'):
         if format.get('vcodec') == 'none':
             return 'audio only'
-        if format.get('_resolution') is not None:
+        if format.get('resolution') is not None:
-            return format['_resolution']
+            return format['resolution']
         if format.get('height') is not None:
             if format.get('width') is not None:
-                res = u'%sx%s' % (format['width'], format['height'])
+                res = '%sx%s' % (format['width'], format['height'])
             else:
-                res = u'%sp' % format['height']
+                res = '%sp' % format['height']
+        elif format.get('width') is not None:
+            res = '?x%d' % format['width']
         else:
             res = default
         return res
 
     def list_formats(self, info_dict):
         def format_note(fdict):
-            res = u''
+            res = ''
+            if fdict.get('ext') in ['f4f', 'f4m']:
+                res += '(unsupported) '
             if fdict.get('format_note') is not None:
-                res += fdict['format_note'] + u' '
+                res += fdict['format_note'] + ' '
+            if fdict.get('tbr') is not None:
+                res += '%4dk ' % fdict['tbr']
+            if fdict.get('container') is not None:
+                if res:
+                    res += ', '
+                res += '%s container' % fdict['container']
             if (fdict.get('vcodec') is not None and
                     fdict.get('vcodec') != 'none'):
-                res += u'%-5s' % fdict['vcodec']
+                if res:
-            elif fdict.get('vbr') is not None:
+                    res += ', '
-                res += u'video'
+                res += fdict['vcodec']
                 if fdict.get('vbr') is not None:
-                res += u'@%4dk' % fdict['vbr']
+                    res += '@'
+            elif fdict.get('vbr') is not None and fdict.get('abr') is not None:
+                res += 'video@'
+            if fdict.get('vbr') is not None:
+                res += '%4dk' % fdict['vbr']
             if fdict.get('acodec') is not None:
                 if res:
-                    res += u', '
+                    res += ', '
-                res += u'%-5s' % fdict['acodec']
+                if fdict['acodec'] == 'none':
+                    res += 'video only'
+                else:
+                    res += '%-5s' % fdict['acodec']
             elif fdict.get('abr') is not None:
                 if res:
-                    res += u', '
+                    res += ', '
                 res += 'audio'
             if fdict.get('abr') is not None:
-                res += u'@%3dk' % fdict['abr']
+                res += '@%3dk' % fdict['abr']
+            if fdict.get('asr') is not None:
+                res += ' (%5dHz)' % fdict['asr']
             if fdict.get('filesize') is not None:
                 if res:
-                    res += u', '
+                    res += ', '
                 res += format_bytes(fdict['filesize'])
             return res
 
         def line(format, idlen=20):
-            return ((u'%-' + compat_str(idlen + 1) + u's%-10s%-12s%s') % (
+            return (('%-' + compat_str(idlen + 1) + 's%-10s%-12s%s') % (
                 format['format_id'],
                 format['ext'],
                 self.format_resolution(format),
@@ -1032,7 +1139,7 @@ class YoutubeDL(object):
             ))
 
         formats = info_dict.get('formats', [info_dict])
-        idlen = max(len(u'format code'),
+        idlen = max(len('format code'),
                     max(len(f['format_id']) for f in formats))
         formats_s = [line(f, idlen) for f in formats]
         if len(formats) > 1:
@@ -1040,10 +1147,10 @@ class YoutubeDL(object):
             formats_s[-1] += (' ' if format_note(formats[-1]) else '') + '(best)'
 
         header_line = line({
-            'format_id': u'format code', 'ext': u'extension',
+            'format_id': 'format code', 'ext': 'extension',
-            '_resolution': u'resolution', 'format_note': u'note'}, idlen=idlen)
+            'resolution': 'resolution', 'format_note': 'note'}, idlen=idlen)
-        self.to_screen(u'[info] Available formats for %s:\n%s\n%s' %
+        self.to_screen('[info] Available formats for %s:\n%s\n%s' %
-                       (info_dict['id'], header_line, u"\n".join(formats_s)))
+                       (info_dict['id'], header_line, '\n'.join(formats_s)))
 
     def urlopen(self, req):
         """ Start an HTTP download """
@@ -1052,7 +1159,7 @@ class YoutubeDL(object):
     def print_debug_header(self):
         if not self.params.get('verbose'):
             return
-        write_string(u'[debug] youtube-dl version ' + __version__ + u'\n')
+        write_string('[debug] youtube-dl version ' + __version__ + '\n')
         try:
             sp = subprocess.Popen(
                 ['git', 'rev-parse', '--short', 'HEAD'],
@@ -1061,20 +1168,20 @@ class YoutubeDL(object):
             out, err = sp.communicate()
             out = out.decode().strip()
             if re.match('[0-9a-f]+', out):
-                write_string(u'[debug] Git HEAD: ' + out + u'\n')
+                write_string('[debug] Git HEAD: ' + out + '\n')
         except:
             try:
                 sys.exc_clear()
             except:
                 pass
-        write_string(u'[debug] Python version %s - %s' %
+        write_string('[debug] Python version %s - %s' %
-                     (platform.python_version(), platform_name()) + u'\n')
+                     (platform.python_version(), platform_name()) + '\n')
 
         proxy_map = {}
         for handler in self._opener.handlers:
             if hasattr(handler, 'proxies'):
                 proxy_map.update(handler.proxies)
-        write_string(u'[debug] Proxy map: ' + compat_str(proxy_map) + u'\n')
+        write_string('[debug] Proxy map: ' + compat_str(proxy_map) + '\n')
 
     def _setup_opener(self):
         timeout_val = self.params.get('socket_timeout')
@@ -1104,10 +1211,13 @@ class YoutubeDL(object):
             if 'http' in proxies and 'https' not in proxies:
                 proxies['https'] = proxies['http']
         proxy_handler = compat_urllib_request.ProxyHandler(proxies)
+
+        debuglevel = 1 if self.params.get('debug_printtraffic') else 0
         https_handler = make_HTTPS_handler(
-            self.params.get('nocheckcertificate', False))
+            self.params.get('nocheckcertificate', False), debuglevel=debuglevel)
+        ydlh = YoutubeDLHandler(debuglevel=debuglevel)
         opener = compat_urllib_request.build_opener(
-            https_handler, proxy_handler, cookie_processor, YoutubeDLHandler())
+            https_handler, proxy_handler, cookie_processor, ydlh)
         # Delete the default user-agent header, which would otherwise apply in
         # cases where our custom HTTP handler doesn't come into play
         # (See https://github.com/rg3/youtube-dl/issues/1309 for details)

youtube_dl/__init__.py

@@ -38,12 +38,21 @@ __authors__  = (
     'Takuya Tsuchida',
     'Sergey M.',
     'Michael Orlitzky',
+    'Chris Gahan',
+    'Saimadhav Heblikar',
+    'Mike Col',
+    'Oleg Prutz',
+    'pulpe',
+    'Andreas Schmitz',
+    'Michael Kaiser',
+    'Niklas Laxström',
 )
 
 __license__ = 'Public Domain'
 
 import codecs
 import getpass
+import locale
 import optparse
 import os
 import random
@@ -56,7 +65,6 @@ from .utils import (
     compat_print,
     DateRange,
     decodeOption,
-    determine_ext,
     get_term_width,
     DownloadError,
     get_cachedir,
@@ -74,11 +82,12 @@ from .FileDownloader import (
 from .extractor import gen_extractors
 from .version import __version__
 from .YoutubeDL import YoutubeDL
-from .PostProcessor import (
+from .postprocessor import (
     FFmpegMetadataPP,
     FFmpegVideoConvertor,
     FFmpegExtractAudioPP,
     FFmpegEmbedSubtitlePP,
+    XAttrMetadataPP,
 )
 
 
@@ -96,6 +105,43 @@ def parseOpts(overrideArguments=None):
             optionf.close()
         return res
 
+    def _readUserConf():
+        xdg_config_home = os.environ.get('XDG_CONFIG_HOME')
+        if xdg_config_home:
+            userConfFile = os.path.join(xdg_config_home, 'youtube-dl', 'config')
+            if not os.path.isfile(userConfFile):
+                userConfFile = os.path.join(xdg_config_home, 'youtube-dl.conf')
+        else:
+            userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl', 'config')
+            if not os.path.isfile(userConfFile):
+                userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf')
+        userConf = _readOptions(userConfFile, None)
+
+        if userConf is None:
+            appdata_dir = os.environ.get('appdata')
+            if appdata_dir:
+                userConf = _readOptions(
+                    os.path.join(appdata_dir, 'youtube-dl', 'config'),
+                    default=None)
+                if userConf is None:
+                    userConf = _readOptions(
+                        os.path.join(appdata_dir, 'youtube-dl', 'config.txt'),
+                        default=None)
+
+        if userConf is None:
+            userConf = _readOptions(
+                os.path.join(os.path.expanduser('~'), 'youtube-dl.conf'),
+                default=None)
+        if userConf is None:
+            userConf = _readOptions(
+                os.path.join(os.path.expanduser('~'), 'youtube-dl.conf.txt'),
+                default=None)
+
+        if userConf is None:
+            userConf = []
+
+        return userConf
+
     def _format_option_string(option):
         ''' ('-o', '--option') -> -o, --format METAVAR'''
 
@@ -186,16 +232,23 @@ def parseOpts(overrideArguments=None):
     general.add_option('--no-check-certificate', action='store_true', dest='no_check_certificate', default=False, help='Suppress HTTPS certificate validation.')
     general.add_option(
         '--cache-dir', dest='cachedir', default=get_cachedir(), metavar='DIR',
-        help='Location in the filesystem where youtube-dl can store downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl .')
+        help='Location in the filesystem where youtube-dl can store some downloaded information permanently. By default $XDG_CACHE_HOME/youtube-dl or ~/.cache/youtube-dl . At the moment, only YouTube player files (for videos with obfuscated signatures) are cached, but that may change.')
     general.add_option(
         '--no-cache-dir', action='store_const', const=None, dest='cachedir',
         help='Disable filesystem caching')
     general.add_option(
         '--socket-timeout', dest='socket_timeout',
-        type=float, default=None, help=optparse.SUPPRESS_HELP)
+        type=float, default=None, help=u'Time to wait before giving up, in seconds')
     general.add_option(
         '--bidi-workaround', dest='bidi_workaround', action='store_true',
-        help=u'Work around terminals that lack bidirectional text support. Requires fribidi executable in PATH')
+        help=u'Work around terminals that lack bidirectional text support. Requires bidiv or fribidi executable in PATH')
+    general.add_option('--default-search',
+            dest='default_search', metavar='PREFIX',
+            help='Use this prefix for unqualified URLs. For example "gvsearch2:" downloads two videos from google videos for  youtube-dl "large apple". By default (with value "auto") youtube-dl guesses.')
+    general.add_option(
+        '--ignore-config',
+        action='store_true',
+        help='Do not read configuration files. When given in the global configuration file /etc/youtube-dl.conf: do not read the user configuration in ~/.config/youtube-dl.conf (%APPDATA%/youtube-dl/config.txt on Windows)')
 
 
     selection.add_option(
@@ -214,8 +267,12 @@ def parseOpts(overrideArguments=None):
     selection.add_option('--min-filesize', metavar='SIZE', dest='min_filesize', help="Do not download any videos smaller than SIZE (e.g. 50k or 44.6m)", default=None)
     selection.add_option('--max-filesize', metavar='SIZE', dest='max_filesize', help="Do not download any videos larger than SIZE (e.g. 50k or 44.6m)", default=None)
     selection.add_option('--date', metavar='DATE', dest='date', help='download only videos uploaded in this date', default=None)
-    selection.add_option('--datebefore', metavar='DATE', dest='datebefore', help='download only videos uploaded before this date', default=None)
+    selection.add_option(
-    selection.add_option('--dateafter', metavar='DATE', dest='dateafter', help='download only videos uploaded after this date', default=None)
+        '--datebefore', metavar='DATE', dest='datebefore', default=None,
+        help='download only videos uploaded on or before this date (i.e. inclusive)')
+    selection.add_option(
+        '--dateafter', metavar='DATE', dest='dateafter', default=None,
+        help='download only videos uploaded on or after this date (i.e. inclusive)')
     selection.add_option(
         '--min-views', metavar='COUNT', dest='min_views',
         default=None, type=int,
@@ -231,7 +288,14 @@ def parseOpts(overrideArguments=None):
     selection.add_option('--download-archive', metavar='FILE',
                          dest='download_archive',
                          help='Download only videos not listed in the archive file. Record the IDs of all downloaded videos in it.')
-
+    selection.add_option(
+        '--include-ads', dest='include_ads',
+        action='store_true',
+        help='Download advertisements as well (experimental)')
+    selection.add_option(
+        '--youtube-include-dash-manifest', action='store_true',
+        dest='youtube_include_dash_manifest', default=False,
+        help='Try to download the DASH manifest on YouTube videos (experimental)')
 
     authentication.add_option('-u', '--username',
             dest='username', metavar='USERNAME', help='account username')
@@ -240,12 +304,12 @@ def parseOpts(overrideArguments=None):
     authentication.add_option('-n', '--netrc',
             action='store_true', dest='usenetrc', help='use .netrc authentication data', default=False)
     authentication.add_option('--video-password',
-            dest='videopassword', metavar='PASSWORD', help='video password (vimeo only)')
+            dest='videopassword', metavar='PASSWORD', help='video password (vimeo, smotri)')
 
 
     video_format.add_option('-f', '--format',
-            action='store', dest='format', metavar='FORMAT', default='best',
+            action='store', dest='format', metavar='FORMAT', default=None,
-            help='video format code, specify the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported')
+            help='video format code, specify the order of preference using slashes: "-f 22/17/18". "-f mp4" and "-f flv" are also supported. You can also use the special names "best", "bestaudio", "worst", and "worstaudio". By default, youtube-dl will pick the best quality.')
     video_format.add_option('--all-formats',
             action='store_const', dest='format', help='download all available video formats', const='all')
     video_format.add_option('--prefer-free-formats',
@@ -253,7 +317,7 @@ def parseOpts(overrideArguments=None):
     video_format.add_option('--max-quality',
             action='store', dest='format_limit', metavar='FORMAT', help='highest quality format to download')
     video_format.add_option('-F', '--list-formats',
-            action='store_true', dest='listformats', help='list all available formats (currently youtube only)')
+            action='store_true', dest='listformats', help='list all available formats')
 
     subtitles.add_option('--write-sub', '--write-srt',
             action='store_true', dest='writesubtitles',
@@ -334,6 +398,9 @@ def parseOpts(overrideArguments=None):
     verbosity.add_option('--youtube-print-sig-code',
             action='store_true', dest='youtube_print_sig_code', default=False,
             help=optparse.SUPPRESS_HELP)
+    verbosity.add_option('--print-traffic',
+            dest='debug_printtraffic', action='store_true', default=False,
+            help='Display sent and read HTTP traffic')
 
 
     filesystem.add_option('-t', '--title',
@@ -369,7 +436,7 @@ def parseOpts(overrideArguments=None):
             dest='batchfile', metavar='FILE', help='file containing URLs to download (\'-\' for stdin)')
     filesystem.add_option('--load-info',
             dest='load_info_filename', metavar='FILE',
-            help='json file containing the video information (created with the "--write-json" option')
+            help='json file containing the video information (created with the "--write-json" option)')
     filesystem.add_option('-w', '--no-overwrites',
             action='store_true', dest='nooverwrites', help='do not overwrite files', default=False)
     filesystem.add_option('-c', '--continue',
@@ -413,7 +480,13 @@ def parseOpts(overrideArguments=None):
     postproc.add_option('--embed-subs', action='store_true', dest='embedsubtitles', default=False,
             help='embed subtitles in the video (only for mp4 videos)')
     postproc.add_option('--add-metadata', action='store_true', dest='addmetadata', default=False,
-            help='add metadata to the files')
+            help='write metadata to the video file')
+    postproc.add_option('--xattrs', action='store_true', dest='xattrs', default=False,
+            help='write metadata to the video file\'s xattrs (using dublin core and xdg standards)')
+    postproc.add_option('--prefer-avconv', action='store_false', dest='prefer_ffmpeg',
+        help='Prefer avconv over ffmpeg for running the postprocessors (default)')
+    postproc.add_option('--prefer-ffmpeg', action='store_true', dest='prefer_ffmpeg',
+        help='Prefer ffmpeg over avconv for running the postprocessors')
 
 
     parser.add_option_group(general)
@@ -431,49 +504,25 @@ def parseOpts(overrideArguments=None):
         if opts.verbose:
             write_string(u'[debug] Override config: ' + repr(overrideArguments) + '\n')
     else:
-        systemConf = _readOptions('/etc/youtube-dl.conf')
-
-        xdg_config_home = os.environ.get('XDG_CONFIG_HOME')
-        if xdg_config_home:
-            userConfFile = os.path.join(xdg_config_home, 'youtube-dl', 'config')
-            if not os.path.isfile(userConfFile):
-                userConfFile = os.path.join(xdg_config_home, 'youtube-dl.conf')
-        else:
-            userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl', 'config')
-            if not os.path.isfile(userConfFile):
-                userConfFile = os.path.join(os.path.expanduser('~'), '.config', 'youtube-dl.conf')
-        userConf = _readOptions(userConfFile, None)
-
-        if userConf is None:
-            appdata_dir = os.environ.get('appdata')
-            if appdata_dir:
-                userConf = _readOptions(
-                    os.path.join(appdata_dir, 'youtube-dl', 'config'),
-                    default=None)
-                if userConf is None:
-                    userConf = _readOptions(
-                        os.path.join(appdata_dir, 'youtube-dl', 'config.txt'),
-                        default=None)
-
-        if userConf is None:
-            userConf = _readOptions(
-                os.path.join(os.path.expanduser('~'), 'youtube-dl.conf'),
-                default=None)
-        if userConf is None:
-            userConf = _readOptions(
-                os.path.join(os.path.expanduser('~'), 'youtube-dl.conf.txt'),
-                default=None)
-
-        if userConf is None:
-            userConf = []
-
         commandLineConf = sys.argv[1:]
+        if '--ignore-config' in commandLineConf:
+            systemConf = []
+            userConf = []
+        else:
+            systemConf = _readOptions('/etc/youtube-dl.conf')
+            if '--ignore-config' in systemConf:
+                userConf = []
+            else:
+                userConf = _readUserConf()
         argv = systemConf + userConf + commandLineConf
+
         opts, args = parser.parse_args(argv)
         if opts.verbose:
             write_string(u'[debug] System config: ' + repr(_hide_login_info(systemConf)) + '\n')
             write_string(u'[debug] User config: ' + repr(_hide_login_info(userConf)) + '\n')
             write_string(u'[debug] Command-line args: ' + repr(_hide_login_info(commandLineConf)) + '\n')
+            write_string(u'[debug] Encodings: locale %r, fs %r, out %r, pref: %r\n' %
+                         (locale.getpreferredencoding(), sys.getfilesystemencoding(), sys.stdout.encoding, preferredencoding()))
 
     return parser, opts, args
 
@@ -518,6 +567,8 @@ def _real_main(argv=None):
             sys.exit(u'ERROR: batch file could not be read')
     all_urls = batchurls + args
     all_urls = [url.strip() for url in all_urls]
+    _enc = preferredencoding()
+    all_urls = [url.decode(_enc, 'ignore') if isinstance(url, bytes) else url for url in all_urls]
 
     extractors = gen_extractors()
 
@@ -525,7 +576,6 @@ def _real_main(argv=None):
         for ie in sorted(extractors, key=lambda ie: ie.IE_NAME.lower()):
             compat_print(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else ''))
             matchedUrls = [url for url in all_urls if ie.suitable(url)]
-            all_urls = [url for url in all_urls if url not in matchedUrls]
             for mu in matchedUrls:
                 compat_print(u'  ' + mu)
         sys.exit(0)
@@ -598,6 +648,12 @@ def _real_main(argv=None):
         date = DateRange.day(opts.date)
     else:
         date = DateRange(opts.dateafter, opts.datebefore)
+    if opts.default_search not in ('auto', None) and ':' not in opts.default_search:
+        parser.error(u'--default-search invalid; did you forget a colon (:) at the end?')
+
+    # Do not download videos when there are audio-only formats
+    if opts.extractaudio and not opts.keepvideo and opts.format is None:
+        opts.format = 'bestaudio/best'
 
     # --all-sub automatically sets --write-sub if --write-auto-sub is not given
     # this was the old behaviour if only --all-sub was given.
@@ -622,6 +678,7 @@ def _real_main(argv=None):
                      u' template'.format(outtmpl))
 
     any_printing = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson
+    download_archive_fn = os.path.expanduser(opts.download_archive) if opts.download_archive is not None else opts.download_archive
 
     ydl_opts = {
         'usenetrc': opts.usenetrc,
@@ -689,12 +746,17 @@ def _real_main(argv=None):
         'cachedir': opts.cachedir,
         'youtube_print_sig_code': opts.youtube_print_sig_code,
         'age_limit': opts.age_limit,
-        'download_archive': opts.download_archive,
+        'download_archive': download_archive_fn,
         'cookiefile': opts.cookiefile,
         'nocheckcertificate': opts.no_check_certificate,
         'proxy': opts.proxy,
         'socket_timeout': opts.socket_timeout,
         'bidi_workaround': opts.bidi_workaround,
+        'debug_printtraffic': opts.debug_printtraffic,
+        'prefer_ffmpeg': opts.prefer_ffmpeg,
+        'include_ads': opts.include_ads,
+        'default_search': opts.default_search,
+        'youtube_include_dash_manifest': opts.youtube_include_dash_manifest,
     }
 
     with YoutubeDL(ydl_opts) as ydl:
@@ -711,6 +773,8 @@ def _real_main(argv=None):
             ydl.add_post_processor(FFmpegVideoConvertor(preferedformat=opts.recodevideo))
         if opts.embedsubtitles:
             ydl.add_post_processor(FFmpegEmbedSubtitlePP(subtitlesformat=opts.subtitlesformat))
+        if opts.xattrs:
+            ydl.add_post_processor(XAttrMetadataPP())
 
         # Update version
         if opts.update_self:

youtube_dl/aes.py

@@ -1,4 +1,4 @@
-__all__ = ['aes_encrypt', 'key_expansion', 'aes_ctr_decrypt', 'aes_decrypt_text']
+__all__ = ['aes_encrypt', 'key_expansion', 'aes_ctr_decrypt', 'aes_cbc_decrypt', 'aes_decrypt_text']
 
 import base64
 from math import ceil
@@ -32,6 +32,31 @@ def aes_ctr_decrypt(data, key, counter):
     
     return decrypted_data
 
+def aes_cbc_decrypt(data, key, iv):
+    """
+    Decrypt with aes in CBC mode
+    
+    @param {int[]} data        cipher
+    @param {int[]} key         16/24/32-Byte cipher key
+    @param {int[]} iv          16-Byte IV
+    @returns {int[]}           decrypted data
+    """
+    expanded_key = key_expansion(key)
+    block_count = int(ceil(float(len(data)) / BLOCK_SIZE_BYTES))
+    
+    decrypted_data=[]
+    previous_cipher_block = iv
+    for i in range(block_count):
+        block = data[i*BLOCK_SIZE_BYTES : (i+1)*BLOCK_SIZE_BYTES]
+        block += [0]*(BLOCK_SIZE_BYTES - len(block))
+        
+        decrypted_block = aes_decrypt(block, expanded_key)
+        decrypted_data += xor(decrypted_block, previous_cipher_block)
+        previous_cipher_block = block
+    decrypted_data = decrypted_data[:len(data)]
+    
+    return decrypted_data
+
 def key_expansion(data):
     """
     Generate key schedule
@@ -86,6 +111,26 @@ def aes_encrypt(data, expanded_key):
 
     return data
 
+def aes_decrypt(data, expanded_key):
+    """
+    Decrypt one block with aes
+    
+    @param {int[]} data          16-Byte cipher
+    @param {int[]} expanded_key  176/208/240-Byte expanded key
+    @returns {int[]}             16-Byte state
+    """
+    rounds = len(expanded_key) // BLOCK_SIZE_BYTES - 1
+    
+    for i in range(rounds, 0, -1):
+        data = xor(data, expanded_key[i*BLOCK_SIZE_BYTES : (i+1)*BLOCK_SIZE_BYTES])
+        if i != rounds:
+            data = mix_columns_inv(data)
+        data = shift_rows_inv(data)
+        data = sub_bytes_inv(data)
+    data = xor(data, expanded_key[:BLOCK_SIZE_BYTES])
+    
+    return data
+
 def aes_decrypt_text(data, password, key_size_bytes):
     """
     Decrypt text
@@ -139,14 +184,69 @@ SBOX = (0x63, 0x7C, 0x77, 0x7B, 0xF2, 0x6B, 0x6F, 0xC5, 0x30, 0x01, 0x67, 0x2B,
         0x70, 0x3E, 0xB5, 0x66, 0x48, 0x03, 0xF6, 0x0E, 0x61, 0x35, 0x57, 0xB9, 0x86, 0xC1, 0x1D, 0x9E,
         0xE1, 0xF8, 0x98, 0x11, 0x69, 0xD9, 0x8E, 0x94, 0x9B, 0x1E, 0x87, 0xE9, 0xCE, 0x55, 0x28, 0xDF,
         0x8C, 0xA1, 0x89, 0x0D, 0xBF, 0xE6, 0x42, 0x68, 0x41, 0x99, 0x2D, 0x0F, 0xB0, 0x54, 0xBB, 0x16)
-MIX_COLUMN_MATRIX = ((2,3,1,1),
+SBOX_INV = (0x52, 0x09, 0x6a, 0xd5, 0x30, 0x36, 0xa5, 0x38, 0xbf, 0x40, 0xa3, 0x9e, 0x81, 0xf3, 0xd7, 0xfb,
-                     (1,2,3,1),
+            0x7c, 0xe3, 0x39, 0x82, 0x9b, 0x2f, 0xff, 0x87, 0x34, 0x8e, 0x43, 0x44, 0xc4, 0xde, 0xe9, 0xcb,
-                     (1,1,2,3),
+            0x54, 0x7b, 0x94, 0x32, 0xa6, 0xc2, 0x23, 0x3d, 0xee, 0x4c, 0x95, 0x0b, 0x42, 0xfa, 0xc3, 0x4e,
-                     (3,1,1,2))
+            0x08, 0x2e, 0xa1, 0x66, 0x28, 0xd9, 0x24, 0xb2, 0x76, 0x5b, 0xa2, 0x49, 0x6d, 0x8b, 0xd1, 0x25,
+            0x72, 0xf8, 0xf6, 0x64, 0x86, 0x68, 0x98, 0x16, 0xd4, 0xa4, 0x5c, 0xcc, 0x5d, 0x65, 0xb6, 0x92,
+            0x6c, 0x70, 0x48, 0x50, 0xfd, 0xed, 0xb9, 0xda, 0x5e, 0x15, 0x46, 0x57, 0xa7, 0x8d, 0x9d, 0x84,
+            0x90, 0xd8, 0xab, 0x00, 0x8c, 0xbc, 0xd3, 0x0a, 0xf7, 0xe4, 0x58, 0x05, 0xb8, 0xb3, 0x45, 0x06,
+            0xd0, 0x2c, 0x1e, 0x8f, 0xca, 0x3f, 0x0f, 0x02, 0xc1, 0xaf, 0xbd, 0x03, 0x01, 0x13, 0x8a, 0x6b,
+            0x3a, 0x91, 0x11, 0x41, 0x4f, 0x67, 0xdc, 0xea, 0x97, 0xf2, 0xcf, 0xce, 0xf0, 0xb4, 0xe6, 0x73,
+            0x96, 0xac, 0x74, 0x22, 0xe7, 0xad, 0x35, 0x85, 0xe2, 0xf9, 0x37, 0xe8, 0x1c, 0x75, 0xdf, 0x6e,
+            0x47, 0xf1, 0x1a, 0x71, 0x1d, 0x29, 0xc5, 0x89, 0x6f, 0xb7, 0x62, 0x0e, 0xaa, 0x18, 0xbe, 0x1b,
+            0xfc, 0x56, 0x3e, 0x4b, 0xc6, 0xd2, 0x79, 0x20, 0x9a, 0xdb, 0xc0, 0xfe, 0x78, 0xcd, 0x5a, 0xf4,
+            0x1f, 0xdd, 0xa8, 0x33, 0x88, 0x07, 0xc7, 0x31, 0xb1, 0x12, 0x10, 0x59, 0x27, 0x80, 0xec, 0x5f,
+            0x60, 0x51, 0x7f, 0xa9, 0x19, 0xb5, 0x4a, 0x0d, 0x2d, 0xe5, 0x7a, 0x9f, 0x93, 0xc9, 0x9c, 0xef,
+            0xa0, 0xe0, 0x3b, 0x4d, 0xae, 0x2a, 0xf5, 0xb0, 0xc8, 0xeb, 0xbb, 0x3c, 0x83, 0x53, 0x99, 0x61,
+            0x17, 0x2b, 0x04, 0x7e, 0xba, 0x77, 0xd6, 0x26, 0xe1, 0x69, 0x14, 0x63, 0x55, 0x21, 0x0c, 0x7d)
+MIX_COLUMN_MATRIX = ((0x2,0x3,0x1,0x1),
+                     (0x1,0x2,0x3,0x1),
+                     (0x1,0x1,0x2,0x3),
+                     (0x3,0x1,0x1,0x2))
+MIX_COLUMN_MATRIX_INV = ((0xE,0xB,0xD,0x9),
+                         (0x9,0xE,0xB,0xD),
+                         (0xD,0x9,0xE,0xB),
+                         (0xB,0xD,0x9,0xE))
+RIJNDAEL_EXP_TABLE = (0x01, 0x03, 0x05, 0x0F, 0x11, 0x33, 0x55, 0xFF, 0x1A, 0x2E, 0x72, 0x96, 0xA1, 0xF8, 0x13, 0x35,
+                      0x5F, 0xE1, 0x38, 0x48, 0xD8, 0x73, 0x95, 0xA4, 0xF7, 0x02, 0x06, 0x0A, 0x1E, 0x22, 0x66, 0xAA,
+                      0xE5, 0x34, 0x5C, 0xE4, 0x37, 0x59, 0xEB, 0x26, 0x6A, 0xBE, 0xD9, 0x70, 0x90, 0xAB, 0xE6, 0x31,
+                      0x53, 0xF5, 0x04, 0x0C, 0x14, 0x3C, 0x44, 0xCC, 0x4F, 0xD1, 0x68, 0xB8, 0xD3, 0x6E, 0xB2, 0xCD,
+                      0x4C, 0xD4, 0x67, 0xA9, 0xE0, 0x3B, 0x4D, 0xD7, 0x62, 0xA6, 0xF1, 0x08, 0x18, 0x28, 0x78, 0x88,
+                      0x83, 0x9E, 0xB9, 0xD0, 0x6B, 0xBD, 0xDC, 0x7F, 0x81, 0x98, 0xB3, 0xCE, 0x49, 0xDB, 0x76, 0x9A,
+                      0xB5, 0xC4, 0x57, 0xF9, 0x10, 0x30, 0x50, 0xF0, 0x0B, 0x1D, 0x27, 0x69, 0xBB, 0xD6, 0x61, 0xA3,
+                      0xFE, 0x19, 0x2B, 0x7D, 0x87, 0x92, 0xAD, 0xEC, 0x2F, 0x71, 0x93, 0xAE, 0xE9, 0x20, 0x60, 0xA0,
+                      0xFB, 0x16, 0x3A, 0x4E, 0xD2, 0x6D, 0xB7, 0xC2, 0x5D, 0xE7, 0x32, 0x56, 0xFA, 0x15, 0x3F, 0x41,
+                      0xC3, 0x5E, 0xE2, 0x3D, 0x47, 0xC9, 0x40, 0xC0, 0x5B, 0xED, 0x2C, 0x74, 0x9C, 0xBF, 0xDA, 0x75,
+                      0x9F, 0xBA, 0xD5, 0x64, 0xAC, 0xEF, 0x2A, 0x7E, 0x82, 0x9D, 0xBC, 0xDF, 0x7A, 0x8E, 0x89, 0x80,
+                      0x9B, 0xB6, 0xC1, 0x58, 0xE8, 0x23, 0x65, 0xAF, 0xEA, 0x25, 0x6F, 0xB1, 0xC8, 0x43, 0xC5, 0x54,
+                      0xFC, 0x1F, 0x21, 0x63, 0xA5, 0xF4, 0x07, 0x09, 0x1B, 0x2D, 0x77, 0x99, 0xB0, 0xCB, 0x46, 0xCA,
+                      0x45, 0xCF, 0x4A, 0xDE, 0x79, 0x8B, 0x86, 0x91, 0xA8, 0xE3, 0x3E, 0x42, 0xC6, 0x51, 0xF3, 0x0E,
+                      0x12, 0x36, 0x5A, 0xEE, 0x29, 0x7B, 0x8D, 0x8C, 0x8F, 0x8A, 0x85, 0x94, 0xA7, 0xF2, 0x0D, 0x17,
+                      0x39, 0x4B, 0xDD, 0x7C, 0x84, 0x97, 0xA2, 0xFD, 0x1C, 0x24, 0x6C, 0xB4, 0xC7, 0x52, 0xF6, 0x01)
+RIJNDAEL_LOG_TABLE = (0x00, 0x00, 0x19, 0x01, 0x32, 0x02, 0x1a, 0xc6, 0x4b, 0xc7, 0x1b, 0x68, 0x33, 0xee, 0xdf, 0x03,
+                      0x64, 0x04, 0xe0, 0x0e, 0x34, 0x8d, 0x81, 0xef, 0x4c, 0x71, 0x08, 0xc8, 0xf8, 0x69, 0x1c, 0xc1,
+                      0x7d, 0xc2, 0x1d, 0xb5, 0xf9, 0xb9, 0x27, 0x6a, 0x4d, 0xe4, 0xa6, 0x72, 0x9a, 0xc9, 0x09, 0x78,
+                      0x65, 0x2f, 0x8a, 0x05, 0x21, 0x0f, 0xe1, 0x24, 0x12, 0xf0, 0x82, 0x45, 0x35, 0x93, 0xda, 0x8e,
+                      0x96, 0x8f, 0xdb, 0xbd, 0x36, 0xd0, 0xce, 0x94, 0x13, 0x5c, 0xd2, 0xf1, 0x40, 0x46, 0x83, 0x38,
+                      0x66, 0xdd, 0xfd, 0x30, 0xbf, 0x06, 0x8b, 0x62, 0xb3, 0x25, 0xe2, 0x98, 0x22, 0x88, 0x91, 0x10,
+                      0x7e, 0x6e, 0x48, 0xc3, 0xa3, 0xb6, 0x1e, 0x42, 0x3a, 0x6b, 0x28, 0x54, 0xfa, 0x85, 0x3d, 0xba,
+                      0x2b, 0x79, 0x0a, 0x15, 0x9b, 0x9f, 0x5e, 0xca, 0x4e, 0xd4, 0xac, 0xe5, 0xf3, 0x73, 0xa7, 0x57,
+                      0xaf, 0x58, 0xa8, 0x50, 0xf4, 0xea, 0xd6, 0x74, 0x4f, 0xae, 0xe9, 0xd5, 0xe7, 0xe6, 0xad, 0xe8,
+                      0x2c, 0xd7, 0x75, 0x7a, 0xeb, 0x16, 0x0b, 0xf5, 0x59, 0xcb, 0x5f, 0xb0, 0x9c, 0xa9, 0x51, 0xa0,
+                      0x7f, 0x0c, 0xf6, 0x6f, 0x17, 0xc4, 0x49, 0xec, 0xd8, 0x43, 0x1f, 0x2d, 0xa4, 0x76, 0x7b, 0xb7,
+                      0xcc, 0xbb, 0x3e, 0x5a, 0xfb, 0x60, 0xb1, 0x86, 0x3b, 0x52, 0xa1, 0x6c, 0xaa, 0x55, 0x29, 0x9d,
+                      0x97, 0xb2, 0x87, 0x90, 0x61, 0xbe, 0xdc, 0xfc, 0xbc, 0x95, 0xcf, 0xcd, 0x37, 0x3f, 0x5b, 0xd1,
+                      0x53, 0x39, 0x84, 0x3c, 0x41, 0xa2, 0x6d, 0x47, 0x14, 0x2a, 0x9e, 0x5d, 0x56, 0xf2, 0xd3, 0xab,
+                      0x44, 0x11, 0x92, 0xd9, 0x23, 0x20, 0x2e, 0x89, 0xb4, 0x7c, 0xb8, 0x26, 0x77, 0x99, 0xe3, 0xa5,
+                      0x67, 0x4a, 0xed, 0xde, 0xc5, 0x31, 0xfe, 0x18, 0x0d, 0x63, 0x8c, 0x80, 0xc0, 0xf7, 0x70, 0x07)
 
 def sub_bytes(data):
     return [SBOX[x] for x in data]
 
+def sub_bytes_inv(data):
+    return [SBOX_INV[x] for x in data]
+
 def rotate(data):
     return data[1:] + [data[0]]
 
@@ -160,30 +260,31 @@ def key_schedule_core(data, rcon_iteration):
 def xor(data1, data2):
     return [x^y for x, y in zip(data1, data2)]
 
-def mix_column(data):
+def rijndael_mul(a, b):
+    if(a==0 or b==0):
+        return 0
+    return RIJNDAEL_EXP_TABLE[(RIJNDAEL_LOG_TABLE[a] + RIJNDAEL_LOG_TABLE[b]) % 0xFF]
+
+def mix_column(data, matrix):
     data_mixed = []
     for row in range(4):
         mixed = 0
         for column in range(4):
-            addend = data[column]
+            # xor is (+) and (-)
-            if MIX_COLUMN_MATRIX[row][column] in (2,3):
+            mixed ^= rijndael_mul(data[column], matrix[row][column])
-                addend <<= 1
-                if addend > 0xff:
-                    addend &= 0xff
-                    addend ^= 0x1b
-                if MIX_COLUMN_MATRIX[row][column] == 3:
-                    addend ^= data[column]
-            mixed ^= addend & 0xff
         data_mixed.append(mixed)
     return data_mixed
 
-def mix_columns(data):
+def mix_columns(data, matrix=MIX_COLUMN_MATRIX):
     data_mixed = []
     for i in range(4):
         column = data[i*4 : (i+1)*4]
-        data_mixed += mix_column(column)
+        data_mixed += mix_column(column, matrix)
     return data_mixed
 
+def mix_columns_inv(data):
+    return mix_columns(data, MIX_COLUMN_MATRIX_INV)
+
 def shift_rows(data):
     data_shifted = []
     for column in range(4):
@@ -191,6 +292,13 @@ def shift_rows(data):
             data_shifted.append( data[((column + row) & 0b11) * 4 + row] )
     return data_shifted
 
+def shift_rows_inv(data):
+    data_shifted = []
+    for column in range(4):
+        for row in range(4):
+            data_shifted.append( data[((column - row) & 0b11) * 4 + row] )
+    return data_shifted
+
 def inc(data):
     data = data[:] # copy
     for i in range(len(data)-1,-1,-1):

youtube_dl/downloader/__init__.py

@@ -0,0 +1,29 @@
+from __future__ import unicode_literals
+
+from .common import FileDownloader
+from .hls import HlsFD
+from .http import HttpFD
+from .mplayer import MplayerFD
+from .rtmp import RtmpFD
+from .f4m import F4mFD
+
+from ..utils import (
+    determine_ext,
+)
+
+
+def get_suitable_downloader(info_dict):
+    """Get the downloader class that can handle the info dict."""
+    url = info_dict['url']
+    protocol = info_dict.get('protocol')
+
+    if url.startswith('rtmp'):
+        return RtmpFD
+    if (protocol == 'm3u8') or (protocol is None and determine_ext(url) == 'm3u8'):
+        return HlsFD
+    if url.startswith('mms') or url.startswith('rtsp'):
+        return MplayerFD
+    if determine_ext(url) == 'f4m':
+        return F4mFD
+    else:
+        return HttpFD

youtube_dl/downloader/common.py

@@ -0,0 +1,316 @@
+import os
+import re
+import sys
+import time
+
+from ..utils import (
+    encodeFilename,
+    timeconvert,
+    format_bytes,
+)
+
+
+class FileDownloader(object):
+    """File Downloader class.
+
+    File downloader objects are the ones responsible of downloading the
+    actual video file and writing it to disk.
+
+    File downloaders accept a lot of parameters. In order not to saturate
+    the object constructor with arguments, it receives a dictionary of
+    options instead.
+
+    Available options:
+
+    verbose:           Print additional info to stdout.
+    quiet:             Do not print messages to stdout.
+    ratelimit:         Download speed limit, in bytes/sec.
+    retries:           Number of times to retry for HTTP error 5xx
+    buffersize:        Size of download buffer in bytes.
+    noresizebuffer:    Do not automatically resize the download buffer.
+    continuedl:        Try to continue downloads if possible.
+    noprogress:        Do not print the progress bar.
+    logtostderr:       Log messages to stderr instead of stdout.
+    consoletitle:      Display progress in console window's titlebar.
+    nopart:            Do not use temporary .part files.
+    updatetime:        Use the Last-modified header to set output file timestamps.
+    test:              Download only first bytes to test the downloader.
+    min_filesize:      Skip files smaller than this size
+    max_filesize:      Skip files larger than this size
+
+    Subclasses of this one must re-define the real_download method.
+    """
+
+    params = None
+
+    def __init__(self, ydl, params):
+        """Create a FileDownloader object with the given options."""
+        self.ydl = ydl
+        self._progress_hooks = []
+        self.params = params
+
+    @staticmethod
+    def format_seconds(seconds):
+        (mins, secs) = divmod(seconds, 60)
+        (hours, mins) = divmod(mins, 60)
+        if hours > 99:
+            return '--:--:--'
+        if hours == 0:
+            return '%02d:%02d' % (mins, secs)
+        else:
+            return '%02d:%02d:%02d' % (hours, mins, secs)
+
+    @staticmethod
+    def calc_percent(byte_counter, data_len):
+        if data_len is None:
+            return None
+        return float(byte_counter) / float(data_len) * 100.0
+
+    @staticmethod
+    def format_percent(percent):
+        if percent is None:
+            return '---.-%'
+        return '%6s' % ('%3.1f%%' % percent)
+
+    @staticmethod
+    def calc_eta(start, now, total, current):
+        if total is None:
+            return None
+        dif = now - start
+        if current == 0 or dif < 0.001: # One millisecond
+            return None
+        rate = float(current) / dif
+        return int((float(total) - float(current)) / rate)
+
+    @staticmethod
+    def format_eta(eta):
+        if eta is None:
+            return '--:--'
+        return FileDownloader.format_seconds(eta)
+
+    @staticmethod
+    def calc_speed(start, now, bytes):
+        dif = now - start
+        if bytes == 0 or dif < 0.001: # One millisecond
+            return None
+        return float(bytes) / dif
+
+    @staticmethod
+    def format_speed(speed):
+        if speed is None:
+            return '%10s' % '---b/s'
+        return '%10s' % ('%s/s' % format_bytes(speed))
+
+    @staticmethod
+    def best_block_size(elapsed_time, bytes):
+        new_min = max(bytes / 2.0, 1.0)
+        new_max = min(max(bytes * 2.0, 1.0), 4194304) # Do not surpass 4 MB
+        if elapsed_time < 0.001:
+            return int(new_max)
+        rate = bytes / elapsed_time
+        if rate > new_max:
+            return int(new_max)
+        if rate < new_min:
+            return int(new_min)
+        return int(rate)
+
+    @staticmethod
+    def parse_bytes(bytestr):
+        """Parse a string indicating a byte quantity into an integer."""
+        matchobj = re.match(r'(?i)^(\d+(?:\.\d+)?)([kMGTPEZY]?)$', bytestr)
+        if matchobj is None:
+            return None
+        number = float(matchobj.group(1))
+        multiplier = 1024.0 ** 'bkmgtpezy'.index(matchobj.group(2).lower())
+        return int(round(number * multiplier))
+
+    def to_screen(self, *args, **kargs):
+        self.ydl.to_screen(*args, **kargs)
+
+    def to_stderr(self, message):
+        self.ydl.to_screen(message)
+
+    def to_console_title(self, message):
+        self.ydl.to_console_title(message)
+
+    def trouble(self, *args, **kargs):
+        self.ydl.trouble(*args, **kargs)
+
+    def report_warning(self, *args, **kargs):
+        self.ydl.report_warning(*args, **kargs)
+
+    def report_error(self, *args, **kargs):
+        self.ydl.report_error(*args, **kargs)
+
+    def slow_down(self, start_time, byte_counter):
+        """Sleep if the download speed is over the rate limit."""
+        rate_limit = self.params.get('ratelimit', None)
+        if rate_limit is None or byte_counter == 0:
+            return
+        now = time.time()
+        elapsed = now - start_time
+        if elapsed <= 0.0:
+            return
+        speed = float(byte_counter) / elapsed
+        if speed > rate_limit:
+            time.sleep((byte_counter - rate_limit * (now - start_time)) / rate_limit)
+
+    def temp_name(self, filename):
+        """Returns a temporary filename for the given filename."""
+        if self.params.get('nopart', False) or filename == u'-' or \
+                (os.path.exists(encodeFilename(filename)) and not os.path.isfile(encodeFilename(filename))):
+            return filename
+        return filename + u'.part'
+
+    def undo_temp_name(self, filename):
+        if filename.endswith(u'.part'):
+            return filename[:-len(u'.part')]
+        return filename
+
+    def try_rename(self, old_filename, new_filename):
+        try:
+            if old_filename == new_filename:
+                return
+            os.rename(encodeFilename(old_filename), encodeFilename(new_filename))
+        except (IOError, OSError) as err:
+            self.report_error(u'unable to rename file: %s' % str(err))
+
+    def try_utime(self, filename, last_modified_hdr):
+        """Try to set the last-modified time of the given file."""
+        if last_modified_hdr is None:
+            return
+        if not os.path.isfile(encodeFilename(filename)):
+            return
+        timestr = last_modified_hdr
+        if timestr is None:
+            return
+        filetime = timeconvert(timestr)
+        if filetime is None:
+            return filetime
+        # Ignore obviously invalid dates
+        if filetime == 0:
+            return
+        try:
+            os.utime(filename, (time.time(), filetime))
+        except:
+            pass
+        return filetime
+
+    def report_destination(self, filename):
+        """Report destination filename."""
+        self.to_screen(u'[download] Destination: ' + filename)
+
+    def _report_progress_status(self, msg, is_last_line=False):
+        fullmsg = u'[download] ' + msg
+        if self.params.get('progress_with_newline', False):
+            self.to_screen(fullmsg)
+        else:
+            if os.name == 'nt':
+                prev_len = getattr(self, '_report_progress_prev_line_length',
+                                   0)
+                if prev_len > len(fullmsg):
+                    fullmsg += u' ' * (prev_len - len(fullmsg))
+                self._report_progress_prev_line_length = len(fullmsg)
+                clear_line = u'\r'
+            else:
+                clear_line = (u'\r\x1b[K' if sys.stderr.isatty() else u'\r')
+            self.to_screen(clear_line + fullmsg, skip_eol=not is_last_line)
+        self.to_console_title(u'youtube-dl ' + msg)
+
+    def report_progress(self, percent, data_len_str, speed, eta):
+        """Report download progress."""
+        if self.params.get('noprogress', False):
+            return
+        if eta is not None:
+            eta_str = self.format_eta(eta)
+        else:
+            eta_str = 'Unknown ETA'
+        if percent is not None:
+            percent_str = self.format_percent(percent)
+        else:
+            percent_str = 'Unknown %'
+        speed_str = self.format_speed(speed)
+
+        msg = (u'%s of %s at %s ETA %s' %
+               (percent_str, data_len_str, speed_str, eta_str))
+        self._report_progress_status(msg)
+
+    def report_progress_live_stream(self, downloaded_data_len, speed, elapsed):
+        if self.params.get('noprogress', False):
+            return
+        downloaded_str = format_bytes(downloaded_data_len)
+        speed_str = self.format_speed(speed)
+        elapsed_str = FileDownloader.format_seconds(elapsed)
+        msg = u'%s at %s (%s)' % (downloaded_str, speed_str, elapsed_str)
+        self._report_progress_status(msg)
+
+    def report_finish(self, data_len_str, tot_time):
+        """Report download finished."""
+        if self.params.get('noprogress', False):
+            self.to_screen(u'[download] Download completed')
+        else:
+            self._report_progress_status(
+                (u'100%% of %s in %s' %
+                 (data_len_str, self.format_seconds(tot_time))),
+                is_last_line=True)
+
+    def report_resuming_byte(self, resume_len):
+        """Report attempt to resume at given byte."""
+        self.to_screen(u'[download] Resuming download at byte %s' % resume_len)
+
+    def report_retry(self, count, retries):
+        """Report retry in case of HTTP error 5xx"""
+        self.to_screen(u'[download] Got server HTTP error. Retrying (attempt %d of %d)...' % (count, retries))
+
+    def report_file_already_downloaded(self, file_name):
+        """Report file has already been fully downloaded."""
+        try:
+            self.to_screen(u'[download] %s has already been downloaded' % file_name)
+        except UnicodeEncodeError:
+            self.to_screen(u'[download] The file has already been downloaded')
+
+    def report_unable_to_resume(self):
+        """Report it was impossible to resume download."""
+        self.to_screen(u'[download] Unable to resume')
+
+    def download(self, filename, info_dict):
+        """Download to a filename using the info from info_dict
+        Return True on success and False otherwise
+        """
+        # Check file already present
+        if self.params.get('continuedl', False) and os.path.isfile(encodeFilename(filename)) and not self.params.get('nopart', False):
+            self.report_file_already_downloaded(filename)
+            self._hook_progress({
+                'filename': filename,
+                'status': 'finished',
+                'total_bytes': os.path.getsize(encodeFilename(filename)),
+            })
+            return True
+
+        return self.real_download(filename, info_dict)
+
+    def real_download(self, filename, info_dict):
+        """Real download process. Redefine in subclasses."""
+        raise NotImplementedError(u'This method must be implemented by sublcasses')
+
+    def _hook_progress(self, status):
+        for ph in self._progress_hooks:
+            ph(status)
+
+    def add_progress_hook(self, ph):
+        """ ph gets called on download progress, with a dictionary with the entries
+        * filename: The final filename
+        * status: One of "downloading" and "finished"
+
+        It can also have some of the following entries:
+
+        * downloaded_bytes: Bytes on disks
+        * total_bytes: Total bytes, None if unknown
+        * tmpfilename: The filename we're currently writing to
+        * eta: The estimated time in seconds, None if unknown
+        * speed: The download speed in bytes/second, None if unknown
+
+        Hooks are guaranteed to be called at least once (with status "finished")
+        if the download is successful.
+        """
+        self._progress_hooks.append(ph)

youtube_dl/downloader/f4m.py

@@ -0,0 +1,315 @@
+from __future__ import unicode_literals
+
+import base64
+import io
+import itertools
+import os
+import time
+import xml.etree.ElementTree as etree
+
+from .common import FileDownloader
+from .http import HttpFD
+from ..utils import (
+    struct_pack,
+    struct_unpack,
+    compat_urllib_request,
+    compat_urlparse,
+    format_bytes,
+    encodeFilename,
+    sanitize_open,
+)
+
+
+class FlvReader(io.BytesIO):
+    """
+    Reader for Flv files
+    The file format is documented in https://www.adobe.com/devnet/f4v.html
+    """
+
+    # Utility functions for reading numbers and strings
+    def read_unsigned_long_long(self):
+        return struct_unpack('!Q', self.read(8))[0]
+
+    def read_unsigned_int(self):
+        return struct_unpack('!I', self.read(4))[0]
+
+    def read_unsigned_char(self):
+        return struct_unpack('!B', self.read(1))[0]
+
+    def read_string(self):
+        res = b''
+        while True:
+            char = self.read(1)
+            if char == b'\x00':
+                break
+            res += char
+        return res
+
+    def read_box_info(self):
+        """
+        Read a box and return the info as a tuple: (box_size, box_type, box_data)
+        """
+        real_size = size = self.read_unsigned_int()
+        box_type = self.read(4)
+        header_end = 8
+        if size == 1:
+            real_size = self.read_unsigned_long_long()
+            header_end = 16
+        return real_size, box_type, self.read(real_size-header_end)
+
+    def read_asrt(self):
+        # version
+        self.read_unsigned_char()
+        # flags
+        self.read(3)
+        quality_entry_count = self.read_unsigned_char()
+        # QualityEntryCount
+        for i in range(quality_entry_count):
+            self.read_string()
+
+        segment_run_count = self.read_unsigned_int()
+        segments = []
+        for i in range(segment_run_count):
+            first_segment = self.read_unsigned_int()
+            fragments_per_segment = self.read_unsigned_int()
+            segments.append((first_segment, fragments_per_segment))
+
+        return {
+            'segment_run': segments,
+        }
+
+    def read_afrt(self):
+        # version
+        self.read_unsigned_char()
+        # flags
+        self.read(3)
+        # time scale
+        self.read_unsigned_int()
+
+        quality_entry_count = self.read_unsigned_char()
+        # QualitySegmentUrlModifiers
+        for i in range(quality_entry_count):
+            self.read_string()
+
+        fragments_count = self.read_unsigned_int()
+        fragments = []
+        for i in range(fragments_count):
+            first = self.read_unsigned_int()
+            first_ts = self.read_unsigned_long_long()
+            duration = self.read_unsigned_int()
+            if duration == 0:
+                discontinuity_indicator = self.read_unsigned_char()
+            else:
+                discontinuity_indicator = None
+            fragments.append({
+                'first': first,
+                'ts': first_ts,
+                'duration': duration,
+                'discontinuity_indicator': discontinuity_indicator,
+            })
+
+        return {
+            'fragments': fragments,
+        }
+
+    def read_abst(self):
+        # version
+        self.read_unsigned_char()
+        # flags
+        self.read(3)
+        # BootstrapinfoVersion
+        bootstrap_info_version = self.read_unsigned_int()
+        # Profile,Live,Update,Reserved
+        self.read(1)
+        # time scale
+        self.read_unsigned_int()
+        # CurrentMediaTime
+        self.read_unsigned_long_long()
+        # SmpteTimeCodeOffset
+        self.read_unsigned_long_long()
+        # MovieIdentifier
+        movie_identifier = self.read_string()
+        server_count = self.read_unsigned_char()
+        # ServerEntryTable
+        for i in range(server_count):
+            self.read_string()
+        quality_count = self.read_unsigned_char()
+        # QualityEntryTable
+        for i in range(server_count):
+            self.read_string()
+        # DrmData
+        self.read_string()
+        # MetaData
+        self.read_string()
+
+        segments_count = self.read_unsigned_char()
+        segments = []
+        for i in range(segments_count):
+            box_size, box_type, box_data = self.read_box_info()
+            assert box_type == b'asrt'
+            segment = FlvReader(box_data).read_asrt()
+            segments.append(segment)
+        fragments_run_count = self.read_unsigned_char()
+        fragments = []
+        for i in range(fragments_run_count):
+            box_size, box_type, box_data = self.read_box_info()
+            assert box_type == b'afrt'
+            fragments.append(FlvReader(box_data).read_afrt())
+
+        return {
+            'segments': segments,
+            'fragments': fragments,
+        }
+
+    def read_bootstrap_info(self):
+        total_size, box_type, box_data = self.read_box_info()
+        assert box_type == b'abst'
+        return FlvReader(box_data).read_abst()
+
+
+def read_bootstrap_info(bootstrap_bytes):
+    return FlvReader(bootstrap_bytes).read_bootstrap_info()
+
+
+def build_fragments_list(boot_info):
+    """ Return a list of (segment, fragment) for each fragment in the video """
+    res = []
+    segment_run_table = boot_info['segments'][0]
+    # I've only found videos with one segment
+    segment_run_entry = segment_run_table['segment_run'][0]
+    n_frags = segment_run_entry[1]
+    fragment_run_entry_table = boot_info['fragments'][0]['fragments']
+    first_frag_number = fragment_run_entry_table[0]['first']
+    for (i, frag_number) in zip(range(1, n_frags+1), itertools.count(first_frag_number)):
+        res.append((1, frag_number))
+    return res
+
+
+def write_flv_header(stream, metadata):
+    """Writes the FLV header and the metadata to stream"""
+    # FLV header
+    stream.write(b'FLV\x01')
+    stream.write(b'\x05')
+    stream.write(b'\x00\x00\x00\x09')
+    # FLV File body
+    stream.write(b'\x00\x00\x00\x00')
+    # FLVTAG
+    # Script data
+    stream.write(b'\x12')
+    # Size of the metadata with 3 bytes
+    stream.write(struct_pack('!L', len(metadata))[1:])
+    stream.write(b'\x00\x00\x00\x00\x00\x00\x00')
+    stream.write(metadata)
+    # Magic numbers extracted from the output files produced by AdobeHDS.php
+    #(https://github.com/K-S-V/Scripts)
+    stream.write(b'\x00\x00\x01\x73')
+
+
+def _add_ns(prop):
+    return '{http://ns.adobe.com/f4m/1.0}%s' % prop
+
+
+class HttpQuietDownloader(HttpFD):
+    def to_screen(self, *args, **kargs):
+        pass
+
+
+class F4mFD(FileDownloader):
+    """
+    A downloader for f4m manifests or AdobeHDS.
+    """
+
+    def real_download(self, filename, info_dict):
+        man_url = info_dict['url']
+        self.to_screen('[download] Downloading f4m manifest')
+        manifest = self.ydl.urlopen(man_url).read()
+        self.report_destination(filename)
+        http_dl = HttpQuietDownloader(self.ydl,
+            {
+                'continuedl': True,
+                'quiet': True,
+                'noprogress': True,
+                'test': self.params.get('test', False),
+            })
+
+        doc = etree.fromstring(manifest)
+        formats = [(int(f.attrib.get('bitrate', -1)), f) for f in doc.findall(_add_ns('media'))]
+        formats = sorted(formats, key=lambda f: f[0])
+        rate, media = formats[-1]
+        base_url = compat_urlparse.urljoin(man_url, media.attrib['url'])
+        bootstrap = base64.b64decode(doc.find(_add_ns('bootstrapInfo')).text)
+        metadata = base64.b64decode(media.find(_add_ns('metadata')).text)
+        boot_info = read_bootstrap_info(bootstrap)
+        fragments_list = build_fragments_list(boot_info)
+        if self.params.get('test', False):
+            # We only download the first fragment
+            fragments_list = fragments_list[:1]
+        total_frags = len(fragments_list)
+
+        tmpfilename = self.temp_name(filename)
+        (dest_stream, tmpfilename) = sanitize_open(tmpfilename, 'wb')
+        write_flv_header(dest_stream, metadata)
+
+        # This dict stores the download progress, it's updated by the progress
+        # hook
+        state = {
+            'downloaded_bytes': 0,
+            'frag_counter': 0,
+        }
+        start = time.time()
+
+        def frag_progress_hook(status):
+            frag_total_bytes = status.get('total_bytes', 0)
+            estimated_size = (state['downloaded_bytes'] +
+                (total_frags - state['frag_counter']) * frag_total_bytes)
+            if status['status'] == 'finished':
+                state['downloaded_bytes'] += frag_total_bytes
+                state['frag_counter'] += 1
+                progress = self.calc_percent(state['frag_counter'], total_frags)
+                byte_counter = state['downloaded_bytes']
+            else:
+                frag_downloaded_bytes = status['downloaded_bytes']
+                byte_counter = state['downloaded_bytes'] + frag_downloaded_bytes
+                frag_progress = self.calc_percent(frag_downloaded_bytes,
+                    frag_total_bytes)
+                progress = self.calc_percent(state['frag_counter'], total_frags)
+                progress += frag_progress / float(total_frags)
+
+            eta = self.calc_eta(start, time.time(), estimated_size, byte_counter)
+            self.report_progress(progress, format_bytes(estimated_size),
+                status.get('speed'), eta)
+        http_dl.add_progress_hook(frag_progress_hook)
+
+        frags_filenames = []
+        for (seg_i, frag_i) in fragments_list:
+            name = 'Seg%d-Frag%d' % (seg_i, frag_i)
+            url = base_url + name
+            frag_filename = '%s-%s' % (tmpfilename, name)
+            success = http_dl.download(frag_filename, {'url': url})
+            if not success:
+                return False
+            with open(frag_filename, 'rb') as down:
+                down_data = down.read()
+                reader = FlvReader(down_data)
+                while True:
+                    _, box_type, box_data = reader.read_box_info()
+                    if box_type == b'mdat':
+                        dest_stream.write(box_data)
+                        break
+            frags_filenames.append(frag_filename)
+
+        self.report_finish(format_bytes(state['downloaded_bytes']), time.time() - start)
+
+        self.try_rename(tmpfilename, filename)
+        for frag_file in frags_filenames:
+            os.remove(frag_file)
+
+        fsize = os.path.getsize(encodeFilename(filename))
+        self._hook_progress({
+            'downloaded_bytes': fsize,
+            'total_bytes': fsize,
+            'filename': filename,
+            'status': 'finished',
+        })
+
+        return True

youtube_dl/downloader/hls.py

@@ -0,0 +1,44 @@
+import os
+import subprocess
+
+from .common import FileDownloader
+from ..utils import (
+    encodeFilename,
+)
+
+
+class HlsFD(FileDownloader):
+    def real_download(self, filename, info_dict):
+        url = info_dict['url']
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+
+        args = ['-y', '-i', url, '-f', 'mp4', '-c', 'copy',
+            '-bsf:a', 'aac_adtstoasc', tmpfilename]
+
+        for program in ['avconv', 'ffmpeg']:
+            try:
+                subprocess.call([program, '-version'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+                break
+            except (OSError, IOError):
+                pass
+        else:
+            self.report_error(u'm3u8 download detected but ffmpeg or avconv could not be found')
+        cmd = [program] + args
+
+        retval = subprocess.call(cmd)
+        if retval == 0:
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen(u'\r[%s] %s bytes' % (cmd[0], fsize))
+            self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
+            return True
+        else:
+            self.to_stderr(u"\n")
+            self.report_error(u'ffmpeg exited with code %d' % retval)
+            return False

youtube_dl/downloader/http.py

@@ -0,0 +1,186 @@
+import os
+import time
+
+from .common import FileDownloader
+from ..utils import (
+    compat_urllib_request,
+    compat_urllib_error,
+    ContentTooShortError,
+
+    encodeFilename,
+    sanitize_open,
+    format_bytes,
+)
+
+
+class HttpFD(FileDownloader):
+    def real_download(self, filename, info_dict):
+        url = info_dict['url']
+        tmpfilename = self.temp_name(filename)
+        stream = None
+
+        # Do not include the Accept-Encoding header
+        headers = {'Youtubedl-no-compression': 'True'}
+        if 'user_agent' in info_dict:
+            headers['Youtubedl-user-agent'] = info_dict['user_agent']
+        basic_request = compat_urllib_request.Request(url, None, headers)
+        request = compat_urllib_request.Request(url, None, headers)
+
+        if self.params.get('test', False):
+            request.add_header('Range', 'bytes=0-10240')
+
+        # Establish possible resume length
+        if os.path.isfile(encodeFilename(tmpfilename)):
+            resume_len = os.path.getsize(encodeFilename(tmpfilename))
+        else:
+            resume_len = 0
+
+        open_mode = 'wb'
+        if resume_len != 0:
+            if self.params.get('continuedl', False):
+                self.report_resuming_byte(resume_len)
+                request.add_header('Range', 'bytes=%d-' % resume_len)
+                open_mode = 'ab'
+            else:
+                resume_len = 0
+
+        count = 0
+        retries = self.params.get('retries', 0)
+        while count <= retries:
+            # Establish connection
+            try:
+                data = compat_urllib_request.urlopen(request)
+                break
+            except (compat_urllib_error.HTTPError, ) as err:
+                if (err.code < 500 or err.code >= 600) and err.code != 416:
+                    # Unexpected HTTP error
+                    raise
+                elif err.code == 416:
+                    # Unable to resume (requested range not satisfiable)
+                    try:
+                        # Open the connection again without the range header
+                        data = compat_urllib_request.urlopen(basic_request)
+                        content_length = data.info()['Content-Length']
+                    except (compat_urllib_error.HTTPError, ) as err:
+                        if err.code < 500 or err.code >= 600:
+                            raise
+                    else:
+                        # Examine the reported length
+                        if (content_length is not None and
+                                (resume_len - 100 < int(content_length) < resume_len + 100)):
+                            # The file had already been fully downloaded.
+                            # Explanation to the above condition: in issue #175 it was revealed that
+                            # YouTube sometimes adds or removes a few bytes from the end of the file,
+                            # changing the file size slightly and causing problems for some users. So
+                            # I decided to implement a suggested change and consider the file
+                            # completely downloaded if the file size differs less than 100 bytes from
+                            # the one in the hard drive.
+                            self.report_file_already_downloaded(filename)
+                            self.try_rename(tmpfilename, filename)
+                            self._hook_progress({
+                                'filename': filename,
+                                'status': 'finished',
+                            })
+                            return True
+                        else:
+                            # The length does not match, we start the download over
+                            self.report_unable_to_resume()
+                            open_mode = 'wb'
+                            break
+            # Retry
+            count += 1
+            if count <= retries:
+                self.report_retry(count, retries)
+
+        if count > retries:
+            self.report_error(u'giving up after %s retries' % retries)
+            return False
+
+        data_len = data.info().get('Content-length', None)
+        if data_len is not None:
+            data_len = int(data_len) + resume_len
+            min_data_len = self.params.get("min_filesize", None)
+            max_data_len = self.params.get("max_filesize", None)
+            if min_data_len is not None and data_len < min_data_len:
+                self.to_screen(u'\r[download] File is smaller than min-filesize (%s bytes < %s bytes). Aborting.' % (data_len, min_data_len))
+                return False
+            if max_data_len is not None and data_len > max_data_len:
+                self.to_screen(u'\r[download] File is larger than max-filesize (%s bytes > %s bytes). Aborting.' % (data_len, max_data_len))
+                return False
+
+        data_len_str = format_bytes(data_len)
+        byte_counter = 0 + resume_len
+        block_size = self.params.get('buffersize', 1024)
+        start = time.time()
+        while True:
+            # Download and write
+            before = time.time()
+            data_block = data.read(block_size)
+            after = time.time()
+            if len(data_block) == 0:
+                break
+            byte_counter += len(data_block)
+
+            # Open file just in time
+            if stream is None:
+                try:
+                    (stream, tmpfilename) = sanitize_open(tmpfilename, open_mode)
+                    assert stream is not None
+                    filename = self.undo_temp_name(tmpfilename)
+                    self.report_destination(filename)
+                except (OSError, IOError) as err:
+                    self.report_error(u'unable to open for writing: %s' % str(err))
+                    return False
+            try:
+                stream.write(data_block)
+            except (IOError, OSError) as err:
+                self.to_stderr(u"\n")
+                self.report_error(u'unable to write data: %s' % str(err))
+                return False
+            if not self.params.get('noresizebuffer', False):
+                block_size = self.best_block_size(after - before, len(data_block))
+
+            # Progress message
+            speed = self.calc_speed(start, time.time(), byte_counter - resume_len)
+            if data_len is None:
+                eta = percent = None
+            else:
+                percent = self.calc_percent(byte_counter, data_len)
+                eta = self.calc_eta(start, time.time(), data_len - resume_len, byte_counter - resume_len)
+            self.report_progress(percent, data_len_str, speed, eta)
+
+            self._hook_progress({
+                'downloaded_bytes': byte_counter,
+                'total_bytes': data_len,
+                'tmpfilename': tmpfilename,
+                'filename': filename,
+                'status': 'downloading',
+                'eta': eta,
+                'speed': speed,
+            })
+
+            # Apply rate limit
+            self.slow_down(start, byte_counter - resume_len)
+
+        if stream is None:
+            self.to_stderr(u"\n")
+            self.report_error(u'Did not get any data blocks')
+            return False
+        stream.close()
+        self.report_finish(data_len_str, (time.time() - start))
+        if data_len is not None and byte_counter != data_len:
+            raise ContentTooShortError(byte_counter, int(data_len))
+        self.try_rename(tmpfilename, filename)
+
+        # Update file modification time
+        if self.params.get('updatetime', True):
+            info_dict['filetime'] = self.try_utime(filename, data.info().get('last-modified', None))
+
+        self._hook_progress({
+            'downloaded_bytes': byte_counter,
+            'total_bytes': byte_counter,
+            'filename': filename,
+            'status': 'finished',
+        })
+
+        return True

youtube_dl/downloader/mplayer.py

@@ -0,0 +1,40 @@
+import os
+import subprocess
+
+from .common import FileDownloader
+from ..utils import (
+    encodeFilename,
+)
+
+
+class MplayerFD(FileDownloader):
+    def real_download(self, filename, info_dict):
+        url = info_dict['url']
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+
+        args = ['mplayer', '-really-quiet', '-vo', 'null', '-vc', 'dummy', '-dumpstream', '-dumpfile', tmpfilename, url]
+        # Check for mplayer first
+        try:
+            subprocess.call(['mplayer', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+        except (OSError, IOError):
+            self.report_error(u'MMS or RTSP download detected but "%s" could not be run' % args[0])
+            return False
+
+        # Download using mplayer.
+        retval = subprocess.call(args)
+        if retval == 0:
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen(u'\r[%s] %s bytes' % (args[0], fsize))
+            self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
+            return True
+        else:
+            self.to_stderr(u"\n")
+            self.report_error(u'mplayer exited with code %d' % retval)
+            return False

youtube_dl/downloader/rtmp.py

@@ -0,0 +1,184 @@
+import os
+import re
+import subprocess
+import sys
+import time
+
+from .common import FileDownloader
+from ..utils import (
+    encodeFilename,
+    format_bytes,
+)
+
+
+class RtmpFD(FileDownloader):
+    def real_download(self, filename, info_dict):
+        def run_rtmpdump(args):
+            start = time.time()
+            resume_percent = None
+            resume_downloaded_data_len = None
+            proc = subprocess.Popen(args, stderr=subprocess.PIPE)
+            cursor_in_new_line = True
+            proc_stderr_closed = False
+            while not proc_stderr_closed:
+                # read line from stderr
+                line = u''
+                while True:
+                    char = proc.stderr.read(1)
+                    if not char:
+                        proc_stderr_closed = True
+                        break
+                    if char in [b'\r', b'\n']:
+                        break
+                    line += char.decode('ascii', 'replace')
+                if not line:
+                    # proc_stderr_closed is True
+                    continue
+                mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec \(([0-9]{1,2}\.[0-9])%\)', line)
+                if mobj:
+                    downloaded_data_len = int(float(mobj.group(1))*1024)
+                    percent = float(mobj.group(2))
+                    if not resume_percent:
+                        resume_percent = percent
+                        resume_downloaded_data_len = downloaded_data_len
+                    eta = self.calc_eta(start, time.time(), 100-resume_percent, percent-resume_percent)
+                    speed = self.calc_speed(start, time.time(), downloaded_data_len-resume_downloaded_data_len)
+                    data_len = None
+                    if percent > 0:
+                        data_len = int(downloaded_data_len * 100 / percent)
+                    data_len_str = u'~' + format_bytes(data_len)
+                    self.report_progress(percent, data_len_str, speed, eta)
+                    cursor_in_new_line = False
+                    self._hook_progress({
+                        'downloaded_bytes': downloaded_data_len,
+                        'total_bytes': data_len,
+                        'tmpfilename': tmpfilename,
+                        'filename': filename,
+                        'status': 'downloading',
+                        'eta': eta,
+                        'speed': speed,
+                    })
+                else:
+                    # no percent for live streams
+                    mobj = re.search(r'([0-9]+\.[0-9]{3}) kB / [0-9]+\.[0-9]{2} sec', line)
+                    if mobj:
+                        downloaded_data_len = int(float(mobj.group(1))*1024)
+                        time_now = time.time()
+                        speed = self.calc_speed(start, time_now, downloaded_data_len)
+                        self.report_progress_live_stream(downloaded_data_len, speed, time_now - start)
+                        cursor_in_new_line = False
+                        self._hook_progress({
+                            'downloaded_bytes': downloaded_data_len,
+                            'tmpfilename': tmpfilename,
+                            'filename': filename,
+                            'status': 'downloading',
+                            'speed': speed,
+                        })
+                    elif self.params.get('verbose', False):
+                        if not cursor_in_new_line:
+                            self.to_screen(u'')
+                        cursor_in_new_line = True
+                        self.to_screen(u'[rtmpdump] '+line)
+            proc.wait()
+            if not cursor_in_new_line:
+                self.to_screen(u'')
+            return proc.returncode
+
+        url = info_dict['url']
+        player_url = info_dict.get('player_url', None)
+        page_url = info_dict.get('page_url', None)
+        app = info_dict.get('app', None)
+        play_path = info_dict.get('play_path', None)
+        tc_url = info_dict.get('tc_url', None)
+        flash_version = info_dict.get('flash_version', None)
+        live = info_dict.get('rtmp_live', False)
+        conn = info_dict.get('rtmp_conn', None)
+
+        self.report_destination(filename)
+        tmpfilename = self.temp_name(filename)
+        test = self.params.get('test', False)
+
+        # Check for rtmpdump first
+        try:
+            subprocess.call(['rtmpdump', '-h'], stdout=(open(os.path.devnull, 'w')), stderr=subprocess.STDOUT)
+        except (OSError, IOError):
+            self.report_error(u'RTMP download detected but "rtmpdump" could not be run')
+            return False
+
+        # Download using rtmpdump. rtmpdump returns exit code 2 when
+        # the connection was interrumpted and resuming appears to be
+        # possible. This is part of rtmpdump's normal usage, AFAIK.
+        basic_args = ['rtmpdump', '--verbose', '-r', url, '-o', tmpfilename]
+        if player_url is not None:
+            basic_args += ['--swfVfy', player_url]
+        if page_url is not None:
+            basic_args += ['--pageUrl', page_url]
+        if app is not None:
+            basic_args += ['--app', app]
+        if play_path is not None:
+            basic_args += ['--playpath', play_path]
+        if tc_url is not None:
+            basic_args += ['--tcUrl', url]
+        if test:
+            basic_args += ['--stop', '1']
+        if flash_version is not None:
+            basic_args += ['--flashVer', flash_version]
+        if live:
+            basic_args += ['--live']
+        if conn:
+            basic_args += ['--conn', conn]
+        args = basic_args + [[], ['--resume', '--skip', '1']][self.params.get('continuedl', False)]
+
+        if sys.platform == 'win32' and sys.version_info < (3, 0):
+            # Windows subprocess module does not actually support Unicode
+            # on Python 2.x
+            # See http://stackoverflow.com/a/9951851/35070
+            subprocess_encoding = sys.getfilesystemencoding()
+            args = [a.encode(subprocess_encoding, 'ignore') for a in args]
+        else:
+            subprocess_encoding = None
+
+        if self.params.get('verbose', False):
+            if subprocess_encoding:
+                str_args = [
+                    a.decode(subprocess_encoding) if isinstance(a, bytes) else a
+                    for a in args]
+            else:
+                str_args = args
+            try:
+                import pipes
+                shell_quote = lambda args: ' '.join(map(pipes.quote, str_args))
+            except ImportError:
+                shell_quote = repr
+            self.to_screen(u'[debug] rtmpdump command line: ' + shell_quote(str_args))
+
+        retval = run_rtmpdump(args)
+
+        while (retval == 2 or retval == 1) and not test:
+            prevsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen(u'[rtmpdump] %s bytes' % prevsize)
+            time.sleep(5.0) # This seems to be needed
+            retval = run_rtmpdump(basic_args + ['-e'] + [[], ['-k', '1']][retval == 1])
+            cursize = os.path.getsize(encodeFilename(tmpfilename))
+            if prevsize == cursize and retval == 1:
+                break
+             # Some rtmp streams seem abort after ~ 99.8%. Don't complain for those
+            if prevsize == cursize and retval == 2 and cursize > 1024:
+                self.to_screen(u'[rtmpdump] Could not download the whole video. This can happen for some advertisements.')
+                retval = 0
+                break
+        if retval == 0 or (test and retval == 2):
+            fsize = os.path.getsize(encodeFilename(tmpfilename))
+            self.to_screen(u'[rtmpdump] %s bytes' % fsize)
+            self.try_rename(tmpfilename, filename)
+            self._hook_progress({
+                'downloaded_bytes': fsize,
+                'total_bytes': fsize,
+                'filename': filename,
+                'status': 'finished',
+            })
+            return True
+        else:
+            self.to_stderr(u"\n")
+            self.report_error(u'rtmpdump exited with code %d' % retval)
+            return False

youtube_dl/extractor/__init__.py

@@ -1,6 +1,8 @@
-from .appletrailers import AppleTrailersIE
+from .academicearth import AcademicEarthCourseIE
 from .addanime import AddAnimeIE
 from .anitube import AnitubeIE
+from .aparat import AparatIE
+from .appletrailers import AppleTrailersIE
 from .archiveorg import ArchiveOrgIE
 from .ard import ARDIE
 from .arte import (
@@ -13,6 +15,7 @@ from .arte import (
 from .auengine import AUEngineIE
 from .bambuser import BambuserIE, BambuserChannelIE
 from .bandcamp import BandcampIE, BandcampAlbumIE
+from .bbccouk import BBCCoUkIE
 from .blinkx import BlinkxIE
 from .bliptv import BlipTVIE, BlipTVUserIE
 from .bloomberg import BloombergIE
@@ -23,14 +26,21 @@ from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
 from .cbs import CBSIE
 from .channel9 import Channel9IE
+from .chilloutzone import ChilloutzoneIE
 from .cinemassacre import CinemassacreIE
 from .clipfish import ClipfishIE
+from .cliphunter import CliphunterIE
 from .clipsyndicate import ClipsyndicateIE
-from .cnn import CNNIE
+from .cmt import CMTIE
+from .cnn import (
+    CNNIE,
+    CNNBlogsIE,
+)
 from .collegehumor import CollegeHumorIE
 from .comedycentral import ComedyCentralIE, ComedyCentralShowsIE
 from .condenast import CondeNastIE
 from .criterion import CriterionIE
+from .crunchyroll import CrunchyrollIE
 from .cspan import CSpanIE
 from .d8 import D8IE
 from .dailymotion import (
@@ -43,27 +53,37 @@ from .depositfiles import DepositFilesIE
 from .dotsub import DotsubIE
 from .dreisat import DreiSatIE
 from .defense import DefenseGouvFrIE
+from .discovery import DiscoveryIE
+from .dropbox import DropboxIE
 from .ebaumsworld import EbaumsWorldIE
 from .ehow import EHowIE
 from .eighttracks import EightTracksIE
 from .eitb import EitbIE
+from .elpais import ElPaisIE
 from .escapist import EscapistIE
+from .everyonesmixtape import EveryonesMixtapeIE
 from .exfm import ExfmIE
 from .extremetube import ExtremeTubeIE
 from .facebook import FacebookIE
 from .faz import FazIE
+from .firstpost import FirstpostIE
+from .firsttv import FirstTVIE
 from .fktv import (
     FKTVIE,
     FKTVPosteckeIE,
 )
 from .flickr import FlickrIE
+from .fourtube import FourTubeIE
+from .franceinter import FranceInterIE
 from .francetv import (
     PluzzIE,
     FranceTvInfoIE,
     FranceTVIE,
-    GenerationQuoiIE
+    GenerationQuoiIE,
+    CultureboxIE,
 )
 from .freesound import FreesoundIE
+from .freespeech import FreespeechIE
 from .funnyordie import FunnyOrDieIE
 from .gamekings import GamekingsIE
 from .gamespot import GameSpotIE
@@ -72,31 +92,59 @@ from .generic import GenericIE
 from .googleplus import GooglePlusIE
 from .googlesearch import GoogleSearchIE
 from .hark import HarkIE
+from .helsinki import HelsinkiIE
 from .hotnewhiphop import HotNewHipHopIE
 from .howcast import HowcastIE
+from .huffpost import HuffPostIE
 from .hypem import HypemIE
 from .ign import IGNIE, OneUPIE
-from .imdb import ImdbIE
+from .imdb import (
+    ImdbIE,
+    ImdbListIE
+)
 from .ina import InaIE
 from .infoq import InfoQIE
 from .instagram import InstagramIE
 from .internetvideoarchive import InternetVideoArchiveIE
+from .iprima import IPrimaIE
+from .ivi import (
+    IviIE,
+    IviCompilationIE
+)
+from .jadorecettepub import JadoreCettePubIE
 from .jeuxvideo import JeuxVideoIE
 from .jukebox import JukeboxIE
 from .justintv import JustinTVIE
+from .jpopsukitv import JpopsukiIE
 from .kankan import KankanIE
 from .keezmovies import KeezMoviesIE
+from .khanacademy import KhanAcademyIE
 from .kickstarter import KickStarterIE
 from .keek import KeekIE
+from .kontrtube import KontrTubeIE
+from .la7 import LA7IE
+from .lifenews import LifeNewsIE
 from .liveleak import LiveLeakIE
 from .livestream import LivestreamIE, LivestreamOriginalIE
+from .lynda import (
+    LyndaIE,
+    LyndaCourseIE
+)
+from .m6 import M6IE
+from .macgamestore import MacGameStoreIE
+from .malemotion import MalemotionIE
 from .mdr import MDRIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mit import TechTVMITIE, MITIE
 from .mixcloud import MixcloudIE
+from .mpora import MporaIE
 from .mofosex import MofosexIE
-from .mtv import MTVIE
+from .mooshare import MooshareIE
+from .mtv import (
+    MTVIE,
+    MTVIggyIE,
+)
 from .muzu import MuzuTVIE
 from .myspace import MySpaceIE
 from .myspass import MySpassIE
@@ -104,11 +152,16 @@ from .myvideo import MyVideoIE
 from .naver import NaverIE
 from .nba import NBAIE
 from .nbc import NBCNewsIE
+from .ndr import NDRIE
 from .ndtv import NDTVIE
 from .newgrounds import NewgroundsIE
+from .nfb import NFBIE
 from .nhl import NHLIE, NHLVideocenterIE
 from .niconico import NiconicoIE
 from .ninegag import NineGagIE
+from .normalboots import NormalbootsIE
+from .novamov import NovamovIE
+from .nowness import NownessIE
 from .nowvideo import NowVideoIE
 from .ooyala import OoyalaIE
 from .orf import ORFIE
@@ -127,7 +180,13 @@ from .ro220 import Ro220IE
 from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
 from .rtlnow import RTLnowIE
-from .rutube import RutubeIE
+from .rutube import (
+    RutubeIE,
+    RutubeChannelIE,
+    RutubeMovieIE,
+    RutubePersonIE,
+)
+from .servingsys import ServingSysIE
 from .sina import SinaIE
 from .slashdot import SlashdotIE
 from .slideshare import SlideshareIE
@@ -146,17 +205,22 @@ from .southparkstudios import (
 from .space import SpaceIE
 from .spankwire import SpankwireIE
 from .spiegel import SpiegelIE
+from .spike import SpikeIE
 from .stanfordoc import StanfordOpenClassroomIE
 from .statigram import StatigramIE
 from .steam import SteamIE
 from .streamcloud import StreamcloudIE
+from .streamcz import StreamCZIE
+from .syfy import SyfyIE
 from .sztvhu import SztvHuIE
 from .teamcoco import TeamcocoIE
 from .techtalks import TechTalksIE
 from .ted import TEDIE
+from .testurl import TestURLIE
 from .tf1 import TF1IE
 from .theplatform import ThePlatformIE
 from .thisav import ThisAVIE
+from .tinypic import TinyPicIE
 from .toutv import TouTvIE
 from .traileraddict import TrailerAddictIE
 from .trilulilu import TriluliluIE
@@ -170,6 +234,7 @@ from .ustream import UstreamIE, UstreamChannelIE
 from .vbox7 import Vbox7IE
 from .veehd import VeeHDIE
 from .veoh import VeohIE
+from .vesti import VestiIE
 from .vevo import VevoIE
 from .vice import ViceIE
 from .viddler import ViddlerIE
@@ -182,12 +247,13 @@ from .vimeo import (
     VimeoUserIE,
     VimeoAlbumIE,
     VimeoGroupsIE,
+    VimeoReviewIE,
 )
 from .vine import VineIE
 from .viki import VikiIE
 from .vk import VKIE
+from .vube import VubeIE
 from .wat import WatIE
-from .websurg import WeBSurgIE
 from .weibo import WeiboIE
 from .wimp import WimpIE
 from .wistia import WistiaIE

youtube_dl/extractor/academicearth.py

@@ -0,0 +1,32 @@
+from __future__ import unicode_literals
+import re
+
+from .common import InfoExtractor
+
+
+class AcademicEarthCourseIE(InfoExtractor):
+    _VALID_URL = r'^https?://(?:www\.)?academicearth\.org/(?:courses|playlists)/(?P<id>[^?#/]+)'
+    IE_NAME = 'AcademicEarth:Course'
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        playlist_id = m.group('id')
+
+        webpage = self._download_webpage(url, playlist_id)
+        title = self._html_search_regex(
+            r'<h1 class="playlist-name">(.*?)</h1>', webpage, u'title')
+        description = self._html_search_regex(
+            r'<p class="excerpt">(.*?)</p>',
+            webpage, u'description', fatal=False)
+        urls = re.findall(
+            r'<h3 class="lecture-title"><a target="_blank" href="([^"]+)">',
+            webpage)
+        entries = [self.url_result(u) for u in urls]
+
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'title': title,
+            'description': description,
+            'entries': entries,
+        }

youtube_dl/extractor/aparat.py

@@ -0,0 +1,56 @@
+#coding: utf-8
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    HEADRequest,
+)
+
+
+class AparatIE(InfoExtractor):
+    _VALID_URL = r'^https?://(?:www\.)?aparat\.com/(?:v/|video/video/embed/videohash/)(?P<id>[a-zA-Z0-9]+)'
+
+    _TEST = {
+        u'url': u'http://www.aparat.com/v/wP8On',
+        u'file': u'wP8On.mp4',
+        u'md5': u'6714e0af7e0d875c5a39c4dc4ab46ad1',
+        u'info_dict': {
+            u"title": u"تیم گلکسی 11 - زومیت",
+        },
+        #u'skip': u'Extremely unreliable',
+    }
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        video_id = m.group('id')
+
+        # Note: There is an easier-to-parse configuration at
+        # http://www.aparat.com/video/video/config/videohash/%video_id
+        # but the URL in there does not work
+        embed_url = (u'http://www.aparat.com/video/video/embed/videohash/' +
+                     video_id + u'/vt/frame')
+        webpage = self._download_webpage(embed_url, video_id)
+
+        video_urls = re.findall(r'fileList\[[0-9]+\]\s*=\s*"([^"]+)"', webpage)
+        for i, video_url in enumerate(video_urls):
+            req = HEADRequest(video_url)
+            res = self._request_webpage(
+                req, video_id, note=u'Testing video URL %d' % i, errnote=False)
+            if res:
+                break
+        else:
+            raise ExtractorError(u'No working video URLs found')
+
+        title = self._search_regex(r'\s+title:\s*"([^"]+)"', webpage, u'title')
+        thumbnail = self._search_regex(
+            r'\s+image:\s*"([^"]+)"', webpage, u'thumbnail', fatal=False)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+            'ext': 'mp4',
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/appletrailers.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -11,46 +13,46 @@ from ..utils import (
 class AppleTrailersIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?trailers\.apple\.com/trailers/(?P<company>[^/]+)/(?P<movie>[^/]+)'
     _TEST = {
-        u"url": u"http://trailers.apple.com/trailers/wb/manofsteel/",
+        "url": "http://trailers.apple.com/trailers/wb/manofsteel/",
-        u"playlist": [
+        "playlist": [
             {
-                u"file": u"manofsteel-trailer4.mov",
+                "file": "manofsteel-trailer4.mov",
-                u"md5": u"d97a8e575432dbcb81b7c3acb741f8a8",
+                "md5": "d97a8e575432dbcb81b7c3acb741f8a8",
-                u"info_dict": {
+                "info_dict": {
-                    u"duration": 111,
+                    "duration": 111,
-                    u"title": u"Trailer 4",
+                    "title": "Trailer 4",
-                    u"upload_date": u"20130523",
+                    "upload_date": "20130523",
-                    u"uploader_id": u"wb",
+                    "uploader_id": "wb",
                 },
             },
             {
-                u"file": u"manofsteel-trailer3.mov",
+                "file": "manofsteel-trailer3.mov",
-                u"md5": u"b8017b7131b721fb4e8d6f49e1df908c",
+                "md5": "b8017b7131b721fb4e8d6f49e1df908c",
-                u"info_dict": {
+                "info_dict": {
-                    u"duration": 182,
+                    "duration": 182,
-                    u"title": u"Trailer 3",
+                    "title": "Trailer 3",
-                    u"upload_date": u"20130417",
+                    "upload_date": "20130417",
-                    u"uploader_id": u"wb",
+                    "uploader_id": "wb",
                 },
             },
             {
-                u"file": u"manofsteel-trailer.mov",
+                "file": "manofsteel-trailer.mov",
-                u"md5": u"d0f1e1150989b9924679b441f3404d48",
+                "md5": "d0f1e1150989b9924679b441f3404d48",
-                u"info_dict": {
+                "info_dict": {
-                    u"duration": 148,
+                    "duration": 148,
-                    u"title": u"Trailer",
+                    "title": "Trailer",
-                    u"upload_date": u"20121212",
+                    "upload_date": "20121212",
-                    u"uploader_id": u"wb",
+                    "uploader_id": "wb",
                 },
             },
             {
-                u"file": u"manofsteel-teaser.mov",
+                "file": "manofsteel-teaser.mov",
-                u"md5": u"5fe08795b943eb2e757fa95cb6def1cb",
+                "md5": "5fe08795b943eb2e757fa95cb6def1cb",
-                u"info_dict": {
+                "info_dict": {
-                    u"duration": 93,
+                    "duration": 93,
-                    u"title": u"Teaser",
+                    "title": "Teaser",
-                    u"upload_date": u"20120721",
+                    "upload_date": "20120721",
-                    u"uploader_id": u"wb",
+                    "uploader_id": "wb",
                 },
             }
         ]
@@ -110,7 +112,8 @@ class AppleTrailersIE(InfoExtractor):
                     'width': format['width'],
                     'height': int(format['height']),
                 })
-            formats = sorted(formats, key=lambda f: (f['height'], f['width']))
+
+            self._sort_formats(formats)
 
             playlist.append({
                 '_type': 'video',

youtube_dl/extractor/archiveorg.py

@@ -1,9 +1,10 @@
+from __future__ import unicode_literals
+
 import json
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
     unified_strdate,
 )
 
@@ -13,23 +14,22 @@ class ArchiveOrgIE(InfoExtractor):
     IE_DESC = 'archive.org videos'
     _VALID_URL = r'(?:https?://)?(?:www\.)?archive\.org/details/(?P<id>[^?/]+)(?:[?].*)?$'
     _TEST = {
-        u"url": u"http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect",
+        "url": "http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect",
-        u'file': u'XD300-23_68HighlightsAResearchCntAugHumanIntellect.ogv',
+        'file': 'XD300-23_68HighlightsAResearchCntAugHumanIntellect.ogv',
-        u'md5': u'8af1d4cf447933ed3c7f4871162602db',
+        'md5': '8af1d4cf447933ed3c7f4871162602db',
-        u'info_dict': {
+        'info_dict': {
-            u"title": u"1968 Demo - FJCC Conference Presentation Reel #1",
+            "title": "1968 Demo - FJCC Conference Presentation Reel #1",
-            u"description": u"Reel 1 of 3: Also known as the \"Mother of All Demos\", Doug Engelbart's presentation at the Fall Joint Computer Conference in San Francisco, December 9, 1968 titled \"A Research Center for Augmenting Human Intellect.\" For this presentation, Doug and his team astonished the audience by not only relating their research, but demonstrating it live. This was the debut of the mouse, interactive computing, hypermedia, computer supported software engineering, video teleconferencing, etc. See also <a href=\"http://dougengelbart.org/firsts/dougs-1968-demo.html\" rel=\"nofollow\">Doug's 1968 Demo page</a> for more background, highlights, links, and the detailed paper published in this conference proceedings. Filmed on 3 reels: Reel 1 | <a href=\"http://www.archive.org/details/XD300-24_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 2</a> | <a href=\"http://www.archive.org/details/XD300-25_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 3</a>",
+            "description": "Reel 1 of 3: Also known as the \"Mother of All Demos\", Doug Engelbart's presentation at the Fall Joint Computer Conference in San Francisco, December 9, 1968 titled \"A Research Center for Augmenting Human Intellect.\" For this presentation, Doug and his team astonished the audience by not only relating their research, but demonstrating it live. This was the debut of the mouse, interactive computing, hypermedia, computer supported software engineering, video teleconferencing, etc. See also <a href=\"http://dougengelbart.org/firsts/dougs-1968-demo.html\" rel=\"nofollow\">Doug's 1968 Demo page</a> for more background, highlights, links, and the detailed paper published in this conference proceedings. Filmed on 3 reels: Reel 1 | <a href=\"http://www.archive.org/details/XD300-24_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 2</a> | <a href=\"http://www.archive.org/details/XD300-25_68HighlightsAResearchCntAugHumanIntellect\" rel=\"nofollow\">Reel 3</a>",
-            u"upload_date": u"19681210",
+            "upload_date": "19681210",
-            u"uploader": u"SRI International"
+            "uploader": "SRI International"
         }
     }
 
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
 
-        json_url = url + (u'?' if u'?' in url else '&') + u'output=json'
+        json_url = url + ('?' if '?' in url else '&') + 'output=json'
         json_data = self._download_webpage(json_url, video_id)
         data = json.loads(json_data)
 
@@ -38,16 +38,16 @@ class ArchiveOrgIE(InfoExtractor):
         uploader = data['metadata']['creator'][0]
         upload_date = unified_strdate(data['metadata']['date'][0])
 
-        formats = [{
+        formats = [
+            {
                 'format': fdata['format'],
                 'url': 'http://' + data['server'] + data['dir'] + fn,
                 'file_size': int(fdata['size']),
             }
             for fn, fdata in data['files'].items()
             if 'Video' in fdata['format']]
-        formats.sort(key=lambda fdata: fdata['file_size'])
+
-        for f in formats:
+        self._sort_formats(formats)
-            f['ext'] = determine_ext(f['url'])
 
         return {
             '_type': 'video',

youtube_dl/extractor/ard.py

@@ -1,22 +1,28 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
+    determine_ext,
     ExtractorError,
 )
 
+
 class ARDIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
+    _VALID_URL = r'^https?://(?:(?:www\.)?ardmediathek\.de|mediathek\.daserste\.de)/(?:.*/)(?P<video_id>[^/\?]+)(?:\?.*)?'
-    _TITLE = r'<h1(?: class="boxTopHeadline")?>(?P<title>.*)</h1>'
+
-    _MEDIA_STREAM = r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)'
     _TEST = {
-        u'url': u'http://www.ardmediathek.de/das-erste/tagesschau-in-100-sek?documentId=14077640',
+        'url': 'http://www.ardmediathek.de/das-erste/guenther-jauch/edward-snowden-im-interview-held-oder-verraeter?documentId=19288786',
-        u'file': u'14077640.mp4',
+        'file': '19288786.mp4',
-        u'md5': u'6ca8824255460c787376353f9e20bbd8',
+        'md5': '515bf47ce209fb3f5a61b7aad364634c',
-        u'info_dict': {
+        'info_dict': {
-            u"title": u"11.04.2013 09:23 Uhr - Tagesschau in 100 Sekunden"
+            'title': 'Edward Snowden im Interview - Held oder Verräter?',
+            'description': 'Edward Snowden hat alles aufs Spiel gesetzt, um die weltweite \xdcberwachung durch die Geheimdienste zu enttarnen. Nun stellt sich der ehemalige NSA-Mitarbeiter erstmals weltweit in einem TV-Interview den Fragen eines NDR-Journalisten. Die Sendung vom Sonntagabend.',
+            'thumbnail': 'http://www.ardmediathek.de/ard/servlet/contentblob/19/28/87/90/19288790/bild/2250037',
         },
-        u'skip': u'Requires rtmpdump'
+        'skip': 'Blocked outside of Germany',
     }
 
     def _real_extract(self, url):
@@ -29,26 +35,49 @@ class ARDIE(InfoExtractor):
         else:
             video_id = m.group('video_id')
 
-        # determine title and media streams from webpage
+        webpage = self._download_webpage(url, video_id)
-        html = self._download_webpage(url, video_id)
+
-        title = re.search(self._TITLE, html).group('title')
+        title = self._html_search_regex(
-        streams = [mo.groupdict() for mo in re.finditer(self._MEDIA_STREAM, html)]
+            r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>', webpage, 'title')
+        description = self._html_search_meta(
+            'dcterms.abstract', webpage, 'description')
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        streams = [
+            mo.groupdict()
+            for mo in re.finditer(
+                r'mediaCollection\.addMediaStream\((?P<media_type>\d+), (?P<quality>\d+), "(?P<rtmp_url>[^"]*)", "(?P<video_url>[^"]*)", "[^"]*"\)', webpage)]
         if not streams:
-            assert '"fsk"' in html
+            if '"fsk"' in webpage:
-            raise ExtractorError(u'This video is only available after 8:00 pm')
+                raise ExtractorError('This video is only available after 20:00')
 
-        # choose default media type and highest quality for now
+        formats = []
-        stream = max([s for s in streams if int(s["media_type"]) == 0],
+        for s in streams:
-                     key=lambda s: int(s["quality"]))
+            format = {
-
+                'quality': int(s['quality']),
-        # there's two possibilities: RTMP stream or HTTP download
+            }
-        info = {'id': video_id, 'title': title, 'ext': 'mp4'}
+            if s.get('rtmp_url'):
-        if stream['rtmp_url']:
+                format['protocol'] = 'rtmp'
-            self.to_screen(u'RTMP download detected')
+                format['url'] = s['rtmp_url']
-            assert stream['video_url'].startswith('mp4:')
+                format['playpath'] = s['video_url']
-            info["url"] = stream["rtmp_url"]
-            info["play_path"] = stream['video_url']
             else:
-            assert stream["video_url"].endswith('.mp4')
+                format['url'] = s['video_url']
-            info["url"] = stream["video_url"]
+
-        return [info]
+            quality_name = self._search_regex(
+                r'[,.]([a-zA-Z0-9_-]+),?\.mp4', format['url'],
+                'quality name', default='NA')
+            format['format_id'] = '%s-%s-%s-%s' % (
+                determine_ext(format['url']), quality_name, s['media_type'],
+                s['quality'])
+
+            formats.append(format)
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'formats': formats,
+            'thumbnail': thumbnail,
+        }

youtube_dl/extractor/arte.py

@@ -1,4 +1,6 @@
 # encoding: utf-8
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -22,7 +24,7 @@ class ArteTvIE(InfoExtractor):
     _LIVEWEB_URL = r'(?:http://)?liveweb\.arte\.tv/(?P<lang>fr|de)/(?P<subpage>.+?)/(?P<name>.+)'
     _LIVE_URL = r'index-[0-9]+\.html$'
 
-    IE_NAME = u'arte.tv'
+    IE_NAME = 'arte.tv'
 
     @classmethod
     def suitable(cls, url):
@@ -37,7 +39,7 @@ class ArteTvIE(InfoExtractor):
     #         r'src="(.*?/videothek_js.*?\.js)',
     #         0,
     #         [
-    #             (1, 'url', u'Invalid URL: %s' % url)
+    #             (1, 'url', 'Invalid URL: %s' % url)
     #         ]
     #     )
     #     http_host = url.split('/')[2]
@@ -49,12 +51,12 @@ class ArteTvIE(InfoExtractor):
     #             '(rtmp://.*?)\'',
     #         re.DOTALL,
     #         [
-    #             (1, 'path',   u'could not extract video path: %s' % url),
+    #             (1, 'path',   'could not extract video path: %s' % url),
-    #             (2, 'player', u'could not extract video player: %s' % url),
+    #             (2, 'player', 'could not extract video player: %s' % url),
-    #             (3, 'url',    u'could not extract video url: %s' % url)
+    #             (3, 'url',    'could not extract video url: %s' % url)
     #         ]
     #     )
-    #     video_url = u'%s/%s' % (info.get('url'), info.get('path'))
+    #     video_url = '%s/%s' % (info.get('url'), info.get('path'))
 
     def _real_extract(self, url):
         mobj = re.match(self._VIDEOS_URL, url)
@@ -107,9 +109,9 @@ class ArteTvIE(InfoExtractor):
     def _extract_liveweb(self, url, name, lang):
         """Extract form http://liveweb.arte.tv/"""
         webpage = self._download_webpage(url, name)
-        video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, u'event id')
+        video_id = self._search_regex(r'eventId=(\d+?)("|&)', webpage, 'event id')
         config_doc = self._download_xml('http://download.liveweb.arte.tv/o21/liveweb/events/event-%s.xml' % video_id,
-                                            video_id, u'Downloading information')
+                                            video_id, 'Downloading information')
         event_doc = config_doc.find('event')
         url_node = event_doc.find('video').find('urlHd')
         if url_node is None:
@@ -124,7 +126,7 @@ class ArteTvIE(InfoExtractor):
 
 
 class ArteTVPlus7IE(InfoExtractor):
-    IE_NAME = u'arte.tv:+7'
+    IE_NAME = 'arte.tv:+7'
     _VALID_URL = r'https?://www\.arte.tv/guide/(?P<lang>fr|de)/(?:(?:sendungen|emissions)/)?(?P<id>.*?)/(?P<name>.*?)(\?.*)?'
 
     @classmethod
@@ -207,7 +209,7 @@ class ArteTVPlus7IE(InfoExtractor):
             if bitrate is not None:
                 quality += '-%d' % bitrate
             if format_info.get('versionCode') is not None:
-                format_id = u'%s-%s' % (quality, format_info['versionCode'])
+                format_id = '%s-%s' % (quality, format_info['versionCode'])
             else:
                 format_id = quality
             info = {
@@ -216,7 +218,7 @@ class ArteTVPlus7IE(InfoExtractor):
                 'width': format_info.get('width'),
                 'height': height,
             }
-            if format_info['mediaType'] == u'rtmp':
+            if format_info['mediaType'] == 'rtmp':
                 info['url'] = format_info['streamer']
                 info['play_path'] = 'mp4:' + format_info['url']
                 info['ext'] = 'flv'
@@ -231,27 +233,27 @@ class ArteTVPlus7IE(InfoExtractor):
 
 # It also uses the arte_vp_url url from the webpage to extract the information
 class ArteTVCreativeIE(ArteTVPlus7IE):
-    IE_NAME = u'arte.tv:creative'
+    IE_NAME = 'arte.tv:creative'
     _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de)/magazine?/(?P<id>.+)'
 
     _TEST = {
-        u'url': u'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
+        'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
-        u'file': u'050489-002.mp4',
+        'file': '050489-002.mp4',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'Agentur Amateur / Agence Amateur #2 : Corporate Design',
+            'title': 'Agentur Amateur / Agence Amateur #2 : Corporate Design',
         },
     }
 
 
 class ArteTVFutureIE(ArteTVPlus7IE):
-    IE_NAME = u'arte.tv:future'
+    IE_NAME = 'arte.tv:future'
     _VALID_URL = r'https?://future\.arte\.tv/(?P<lang>fr|de)/(thema|sujet)/.*?#article-anchor-(?P<id>\d+)'
 
     _TEST = {
-        u'url': u'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
+        'url': 'http://future.arte.tv/fr/sujet/info-sciences#article-anchor-7081',
-        u'file': u'050940-003.mp4',
+        'file': '050940-003.mp4',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'Les champignons au secours de la planète',
+            'title': 'Les champignons au secours de la planète',
         },
     }
 
@@ -263,23 +265,9 @@ class ArteTVFutureIE(ArteTVPlus7IE):
 
 
 class ArteTVDDCIE(ArteTVPlus7IE):
-    IE_NAME = u'arte.tv:ddc'
+    IE_NAME = 'arte.tv:ddc'
     _VALID_URL = r'http?://ddc\.arte\.tv/(?P<lang>emission|folge)/(?P<id>.+)'
 
-    _TEST = {
-        u'url': u'http://ddc.arte.tv/folge/neues-aus-mauretanien',
-        u'file': u'049881-009_PLUS7-D.flv',
-        u'info_dict': {
-            u'title': u'Mit offenen Karten',
-            u'description': u'md5:57929b0eaeddeb8a0c983f58e9ebd3b6',
-            u'upload_date': u'20131207',
-        },
-        u'params': {
-            # rtmp download
-            u'skip_download': True,
-        },
-    }
-
     def _real_extract(self, url):
         video_id, lang = self._extract_url_info(url)
         if lang == 'folge':

youtube_dl/extractor/auengine.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -7,13 +9,14 @@ from ..utils import (
     ExtractorError,
 )
 
+
 class AUEngineIE(InfoExtractor):
     _TEST = {
-        u'url': u'http://auengine.com/embed.php?file=lfvlytY6&w=650&h=370',
+        'url': 'http://auengine.com/embed.php?file=lfvlytY6&w=650&h=370',
-        u'file': u'lfvlytY6.mp4',
+        'file': 'lfvlytY6.mp4',
-        u'md5': u'48972bdbcf1a3a2f5533e62425b41d4f',
+        'md5': '48972bdbcf1a3a2f5533e62425b41d4f',
-        u'info_dict': {
+        'info_dict': {
-            u"title": u"[Commie]The Legend of the Legendary Heroes - 03 - Replication Eye (Alpha Stigma)[F9410F5A]"
+            'title': '[Commie]The Legend of the Legendary Heroes - 03 - Replication Eye (Alpha Stigma)[F9410F5A]'
         }
     }
     _VALID_URL = r'(?:http://)?(?:www\.)?auengine\.com/embed\.php\?.*?file=([^&]+).*?'
@@ -23,7 +26,7 @@ class AUEngineIE(InfoExtractor):
         video_id = mobj.group(1)
         webpage = self._download_webpage(url, video_id)
         title = self._html_search_regex(r'<title>(?P<title>.+?)</title>',
-                webpage, u'title')
+                webpage, 'title')
         title = title.strip()
         links = re.findall(r'\s(?:file|url):\s*["\']([^\'"]+)["\']', webpage)
         links = map(compat_urllib_parse.unquote, links)
@@ -37,7 +40,7 @@ class AUEngineIE(InfoExtractor):
                 video_url = link
         if not video_url:
             raise ExtractorError(u'Could not find video URL')
-        ext = u'.' + determine_ext(video_url)
+        ext = '.' + determine_ext(video_url)
         if ext == title[-len(ext):]:
             title = title[:-len(ext)]
 

youtube_dl/extractor/bambuser.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 import itertools
@@ -9,26 +11,26 @@ from ..utils import (
 
 
 class BambuserIE(InfoExtractor):
-    IE_NAME = u'bambuser'
+    IE_NAME = 'bambuser'
     _VALID_URL = r'https?://bambuser\.com/v/(?P<id>\d+)'
     _API_KEY = '005f64509e19a868399060af746a00aa'
 
     _TEST = {
-        u'url': u'http://bambuser.com/v/4050584',
+        'url': 'http://bambuser.com/v/4050584',
         # MD5 seems to be flaky, see https://travis-ci.org/rg3/youtube-dl/jobs/14051016#L388
-        #u'md5': u'fba8f7693e48fd4e8641b3fd5539a641',
+        #u'md5': 'fba8f7693e48fd4e8641b3fd5539a641',
-        u'info_dict': {
+        'info_dict': {
-            u'id': u'4050584',
+            'id': '4050584',
-            u'ext': u'flv',
+            'ext': 'flv',
-            u'title': u'Education engineering days - lightning talks',
+            'title': 'Education engineering days - lightning talks',
-            u'duration': 3741,
+            'duration': 3741,
-            u'uploader': u'pixelversity',
+            'uploader': 'pixelversity',
-            u'uploader_id': u'344706',
+            'uploader_id': '344706',
         },
-        u'params': {
+        'params': {
             # It doesn't respect the 'Range' header, it would download the whole video
             # caused the travis builds to fail: https://travis-ci.org/rg3/youtube-dl/jobs/14493845#L59
-            u'skip_download': True,
+            'skip_download': True,
         },
     }
 
@@ -53,7 +55,7 @@ class BambuserIE(InfoExtractor):
 
 
 class BambuserChannelIE(InfoExtractor):
-    IE_NAME = u'bambuser:channel'
+    IE_NAME = 'bambuser:channel'
     _VALID_URL = r'https?://bambuser\.com/channel/(?P<user>.*?)(?:/|#|\?|$)'
     # The maximum number we can get with each request
     _STEP = 50
@@ -72,7 +74,7 @@ class BambuserChannelIE(InfoExtractor):
             # Without setting this header, we wouldn't get any result
             req.add_header('Referer', 'http://bambuser.com/channel/%s' % user)
             info_json = self._download_webpage(req, user,
-                u'Downloading page %d' % i)
+                'Downloading page %d' % i)
             results = json.loads(info_json)['result']
             if len(results) == 0:
                 break

youtube_dl/extractor/bandcamp.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import json
 import re
 
@@ -10,16 +12,16 @@ from ..utils import (
 
 
 class BandcampIE(InfoExtractor):
-    IE_NAME = u'Bandcamp'
     _VALID_URL = r'http://.*?\.bandcamp\.com/track/(?P<title>.*)'
     _TESTS = [{
-        u'url': u'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
+        'url': 'http://youtube-dl.bandcamp.com/track/youtube-dl-test-song',
-        u'file': u'1812978515.mp3',
+        'file': '1812978515.mp3',
-        u'md5': u'cdeb30cdae1921719a3cbcab696ef53c',
+        'md5': 'c557841d5e50261777a6585648adf439',
-        u'info_dict': {
+        'info_dict': {
-            u"title": u"youtube-dl test song \"'/\\\u00e4\u21ad"
+            "title": "youtube-dl  \"'/\\\u00e4\u21ad - youtube-dl test song \"'/\\\u00e4\u21ad",
+            "duration": 10,
         },
-        u'skip': u'There is a limit of 200 free downloads / month for the test song'
+        '_skip': 'There is a limit of 200 free downloads / month for the test song'
     }]
 
     def _real_extract(self, url):
@@ -33,82 +35,95 @@ class BandcampIE(InfoExtractor):
             if m_trackinfo:
                 json_code = m_trackinfo.group(1)
                 data = json.loads(json_code)
+                d = data[0]
 
-            for d in data:
+                duration = int(round(d['duration']))
-                formats = [{
+                formats = []
-                    'format_id': 'format_id',
+                for format_id, format_url in d['file'].items():
+                    ext, _, abr_str = format_id.partition('-')
+
+                    formats.append({
+                        'format_id': format_id,
                         'url': format_url,
-                    'ext': format_id.partition('-')[0]
+                        'ext': format_id.partition('-')[0],
-                } for format_id, format_url in sorted(d['file'].items())]
+                        'vcodec': 'none',
+                        'acodec': format_id.partition('-')[0],
+                        'abr': int(format_id.partition('-')[2]),
+                    })
+
+                self._sort_formats(formats)
+
                 return {
                     'id': compat_str(d['id']),
                     'title': d['title'],
                     'formats': formats,
+                    'duration': duration,
                 }
             else:
-            raise ExtractorError(u'No free songs found')
+                raise ExtractorError('No free songs found')
 
         download_link = m_download.group(1)
-        id = re.search(r'var TralbumData = {(.*?)id: (?P<id>\d*?)$', 
+        video_id = re.search(
+            r'var TralbumData = {(.*?)id: (?P<id>\d*?)$',
             webpage, re.MULTILINE | re.DOTALL).group('id')
 
-        download_webpage = self._download_webpage(download_link, id,
+        download_webpage = self._download_webpage(download_link, video_id,
                                                   'Downloading free downloads page')
         # We get the dictionary of the track from some javascrip code
         info = re.search(r'items: (.*?),$',
                          download_webpage, re.MULTILINE).group(1)
         info = json.loads(info)[0]
         # We pick mp3-320 for now, until format selection can be easily implemented.
-        mp3_info = info[u'downloads'][u'mp3-320']
+        mp3_info = info['downloads']['mp3-320']
         # If we try to use this url it says the link has expired
-        initial_url = mp3_info[u'url']
+        initial_url = mp3_info['url']
         re_url = r'(?P<server>http://(.*?)\.bandcamp\.com)/download/track\?enc=mp3-320&fsig=(?P<fsig>.*?)&id=(?P<id>.*?)&ts=(?P<ts>.*)$'
         m_url = re.match(re_url, initial_url)
         #We build the url we will use to get the final track url
         # This url is build in Bandcamp in the script download_bunde_*.js
-        request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), id, m_url.group('ts'))
+        request_url = '%s/statdownload/track?enc=mp3-320&fsig=%s&id=%s&ts=%s&.rand=665028774616&.vrs=1' % (m_url.group('server'), m_url.group('fsig'), video_id, m_url.group('ts'))
-        final_url_webpage = self._download_webpage(request_url, id, 'Requesting download url')
+        final_url_webpage = self._download_webpage(request_url, video_id, 'Requesting download url')
         # If we could correctly generate the .rand field the url would be
         #in the "download_url" key
         final_url = re.search(r'"retry_url":"(.*?)"', final_url_webpage).group(1)
 
-        track_info = {'id':id,
+        return {
-                      'title' : info[u'title'],
+            'id': video_id,
+            'title': info['title'],
             'ext': 'mp3',
+            'vcodec': 'none',
             'url': final_url,
-                      'thumbnail' : info[u'thumb_url'],
+            'thumbnail': info.get('thumb_url'),
-                      'uploader' :  info[u'artist']
+            'uploader': info.get('artist'),
         }
 
-        return [track_info]
-
 
 class BandcampAlbumIE(InfoExtractor):
-    IE_NAME = u'Bandcamp:album'
+    IE_NAME = 'Bandcamp:album'
     _VALID_URL = r'http://.*?\.bandcamp\.com/album/(?P<title>.*)'
 
     _TEST = {
-        u'url': u'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
+        'url': 'http://blazo.bandcamp.com/album/jazz-format-mixtape-vol-1',
-        u'playlist': [
+        'playlist': [
             {
-                u'file': u'1353101989.mp3',
+                'file': '1353101989.mp3',
-                u'md5': u'39bc1eded3476e927c724321ddf116cf',
+                'md5': '39bc1eded3476e927c724321ddf116cf',
-                u'info_dict': {
+                'info_dict': {
-                    u'title': u'Intro',
+                    'title': 'Intro',
                 }
             },
             {
-                u'file': u'38097443.mp3',
+                'file': '38097443.mp3',
-                u'md5': u'1a2c32e2691474643e912cc6cd4bffaa',
+                'md5': '1a2c32e2691474643e912cc6cd4bffaa',
-                u'info_dict': {
+                'info_dict': {
-                    u'title': u'Kero One - Keep It Alive (Blazo remix)',
+                    'title': 'Kero One - Keep It Alive (Blazo remix)',
                 }
             },
         ],
-        u'params': {
+        'params': {
-            u'playlistend': 2
+            'playlistend': 2
         },
-        u'skip': u'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
+        'skip': 'Bancamp imposes download limits. See test_playlists:test_bandcamp_album for the playlist test'
     }
 
     def _real_extract(self, url):
@@ -117,11 +132,11 @@ class BandcampAlbumIE(InfoExtractor):
         webpage = self._download_webpage(url, title)
         tracks_paths = re.findall(r'<a href="(.*?)" itemprop="url">', webpage)
         if not tracks_paths:
-            raise ExtractorError(u'The page doesn\'t contain any track')
+            raise ExtractorError('The page doesn\'t contain any tracks')
         entries = [
             self.url_result(compat_urlparse.urljoin(url, t_path), ie=BandcampIE.ie_key())
             for t_path in tracks_paths]
-        title = self._search_regex(r'album_title : "(.*?)"', webpage, u'title')
+        title = self._search_regex(r'album_title : "(.*?)"', webpage, 'title')
         return {
             '_type': 'playlist',
             'title': title,

youtube_dl/extractor/bbccouk.py

@@ -0,0 +1,218 @@
+from __future__ import unicode_literals
+
+import re
+
+from .subtitles import SubtitlesInfoExtractor
+from ..utils import ExtractorError
+
+
+class BBCCoUkIE(SubtitlesInfoExtractor):
+    IE_NAME = 'bbc.co.uk'
+    IE_DESC = 'BBC iPlayer'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:programmes|iplayer/episode)/(?P<id>[\da-z]{8})'
+
+    _TESTS = [
+        {
+            'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
+            'info_dict': {
+                'id': 'b039d07m',
+                'ext': 'flv',
+                'title': 'Kaleidoscope: Leonard Cohen',
+                'description': 'md5:db4755d7a665ae72343779f7dacb402c',
+                'duration': 1740,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            }
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b00yng5w/The_Man_in_Black_Series_3_The_Printed_Name/',
+            'info_dict': {
+                'id': 'b00yng1d',
+                'ext': 'flv',
+                'title': 'The Man in Black: Series 3: The Printed Name',
+                'description': "Mark Gatiss introduces Nicholas Pierpan's chilling tale of a writer's devilish pact with a mysterious man. Stars Ewan Bailey.",
+                'duration': 1800,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Episode is no longer available on BBC iPlayer Radio',
+        },
+        {
+            'url': 'http://www.bbc.co.uk/iplayer/episode/b03vhd1f/The_Voice_UK_Series_3_Blind_Auditions_5/',
+            'info_dict': {
+                'id': 'b00yng1d',
+                'ext': 'flv',
+                'title': 'The Voice UK: Series 3: Blind Auditions 5',
+                'description': "Emma Willis and Marvin Humes present the fifth set of blind auditions in the singing competition, as the coaches continue to build their teams based on voice alone.",
+                'duration': 5100,
+            },
+            'params': {
+                # rtmp download
+                'skip_download': True,
+            },
+            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
+        }
+    ]
+
+    def _extract_asx_playlist(self, connection, programme_id):
+        asx = self._download_xml(connection.get('href'), programme_id, 'Downloading ASX playlist')
+        return [ref.get('href') for ref in asx.findall('./Entry/ref')]
+
+    def _extract_connection(self, connection, programme_id):
+        formats = []
+        protocol = connection.get('protocol')
+        supplier = connection.get('supplier')
+        if protocol == 'http':
+            href = connection.get('href')
+            # ASX playlist
+            if supplier == 'asx':
+                for i, ref in enumerate(self._extract_asx_playlist(connection, programme_id)):
+                    formats.append({
+                        'url': ref,
+                        'format_id': 'ref%s_%s' % (i, supplier),
+                    })
+            # Direct link
+            else:
+                formats.append({
+                    'url': href,
+                    'format_id': supplier,
+                })
+        elif protocol == 'rtmp':
+            application = connection.get('application', 'ondemand')
+            auth_string = connection.get('authString')
+            identifier = connection.get('identifier')
+            server = connection.get('server')
+            formats.append({
+                'url': '%s://%s/%s?%s' % (protocol, server, application, auth_string),
+                'play_path': identifier,
+                'app': '%s?%s' % (application, auth_string),
+                'page_url': 'http://www.bbc.co.uk',
+                'player_url': 'http://www.bbc.co.uk/emp/releases/iplayer/revisions/617463_618125_4/617463_618125_4_emp.swf',
+                'rtmp_live': False,
+                'ext': 'flv',
+                'format_id': supplier,
+            })
+        return formats
+
+    def _extract_items(self, playlist):
+        return playlist.findall('./{http://bbc.co.uk/2008/emp/playlist}item')
+
+    def _extract_medias(self, media_selection):
+        return media_selection.findall('./{http://bbc.co.uk/2008/mp/mediaselection}media')
+
+    def _extract_connections(self, media):
+        return media.findall('./{http://bbc.co.uk/2008/mp/mediaselection}connection')
+
+    def _extract_video(self, media, programme_id):
+        formats = []
+        vbr = int(media.get('bitrate'))
+        vcodec = media.get('encoding')
+        service = media.get('service')
+        width = int(media.get('width'))
+        height = int(media.get('height'))
+        file_size = int(media.get('media_file_size'))
+        for connection in self._extract_connections(media):
+            conn_formats = self._extract_connection(connection, programme_id)
+            for format in conn_formats:
+                format.update({
+                    'format_id': '%s_%s' % (service, format['format_id']),
+                    'width': width,
+                    'height': height,
+                    'vbr': vbr,
+                    'vcodec': vcodec,
+                    'filesize': file_size,
+                })
+            formats.extend(conn_formats)
+        return formats
+
+    def _extract_audio(self, media, programme_id):
+        formats = []
+        abr = int(media.get('bitrate'))
+        acodec = media.get('encoding')
+        service = media.get('service')
+        for connection in self._extract_connections(media):
+            conn_formats = self._extract_connection(connection, programme_id)
+            for format in conn_formats:
+                format.update({
+                    'format_id': '%s_%s' % (service, format['format_id']),
+                    'abr': abr,
+                    'acodec': acodec,
+                })
+            formats.extend(conn_formats)
+        return formats
+
+    def _extract_captions(self, media, programme_id):
+        subtitles = {}
+        for connection in self._extract_connections(media):
+            captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
+            lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
+            ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
+            srt = ''
+            for pos, p in enumerate(ps):
+                srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'),
+                                                          p.text.strip() if p.text is not None else '')
+            subtitles[lang] = srt
+        return subtitles
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        group_id = mobj.group('id')
+
+        playlist = self._download_xml('http://www.bbc.co.uk/iplayer/playlist/%s' % group_id, group_id,
+            'Downloading playlist XML')
+
+        no_items = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}noItems')
+        if no_items is not None:
+            reason = no_items.get('reason')
+            if reason == 'preAvailability':
+                msg = 'Episode %s is not yet available' % group_id
+            elif reason == 'postAvailability':
+                msg = 'Episode %s is no longer available' % group_id
+            else:
+                msg = 'Episode %s is not available: %s' % (group_id, reason)
+            raise ExtractorError(msg, expected=True)
+
+        formats = []
+        subtitles = None
+
+        for item in self._extract_items(playlist):
+            kind = item.get('kind')
+            if kind != 'programme' and kind != 'radioProgramme':
+                continue
+            title = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}title').text
+            description = playlist.find('./{http://bbc.co.uk/2008/emp/playlist}summary').text
+
+            programme_id = item.get('identifier')
+            duration = int(item.get('duration'))
+
+            media_selection = self._download_xml(
+                'http://open.live.bbc.co.uk/mediaselector/5/select/version/2.0/mediaset/pc/vpid/%s'  % programme_id,
+                programme_id, 'Downloading media selection XML')
+
+            for media in self._extract_medias(media_selection):
+                kind = media.get('kind')
+                if kind == 'audio':
+                    formats.extend(self._extract_audio(media, programme_id))
+                elif kind == 'video':
+                    formats.extend(self._extract_video(media, programme_id))
+                elif kind == 'captions':
+                    subtitles = self._extract_captions(media, programme_id)
+
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(programme_id, subtitles)
+            return
+
+        self._sort_formats(formats)
+
+        return {
+            'id': programme_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'formats': formats,
+            'subtitles': subtitles,
+        }

youtube_dl/extractor/blinkx.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import datetime
 import json
 import re
@@ -9,20 +11,20 @@ from ..utils import (
 
 
 class BlinkxIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://(?:www\.)blinkx\.com/ce/|blinkx:)(?P<id>[^?]+)'
+    _VALID_URL = r'^(?:https?://(?:www\.)blinkx\.com/#?ce/|blinkx:)(?P<id>[^?]+)'
-    _IE_NAME = u'blinkx'
+    IE_NAME = 'blinkx'
 
     _TEST = {
-        u'url': u'http://www.blinkx.com/ce/8aQUy7GVFYgFzpKhT0oqsilwOGFRVXk3R1ZGWWdGenBLaFQwb3FzaWx3OGFRVXk3R1ZGWWdGenB',
+        'url': 'http://www.blinkx.com/ce/8aQUy7GVFYgFzpKhT0oqsilwOGFRVXk3R1ZGWWdGenBLaFQwb3FzaWx3OGFRVXk3R1ZGWWdGenB',
-        u'file': u'8aQUy7GV.mp4',
+        'file': '8aQUy7GV.mp4',
-        u'md5': u'2e9a07364af40163a908edbf10bb2492',
+        'md5': '2e9a07364af40163a908edbf10bb2492',
-        u'info_dict': {
+        'info_dict': {
-            u"title": u"Police Car Rolls Away",
+            "title": "Police Car Rolls Away",
-            u"uploader": u"stupidvideos.com",
+            "uploader": "stupidvideos.com",
-            u"upload_date": u"20131215",
+            "upload_date": "20131215",
-            u"description": u"A police car gently rolls away from a fight. Maybe it felt weird being around a confrontation and just had to get out of there!",
+            "description": "A police car gently rolls away from a fight. Maybe it felt weird being around a confrontation and just had to get out of there!",
-            u"duration": 14.886,
+            "duration": 14.886,
-            u"thumbnails": [{
+            "thumbnails": [{
                 "width": 100,
                 "height": 76,
                 "url": "http://cdn.blinkx.com/stream/b/41/StupidVideos/20131215/1873969261/1873969261_tn_0.jpg",
@@ -30,17 +32,17 @@ class BlinkxIE(InfoExtractor):
         },
     }
 
-    def _real_extract(self, url):
+    def _real_extract(self, rl):
-        m = re.match(self._VALID_URL, url)
+        m = re.match(self._VALID_URL, rl)
         video_id = m.group('id')
         display_id = video_id[:8]
 
         api_url = (u'https://apib4.blinkx.com/api.php?action=play_video&' +
-                   u'video=%s' % video_id)
+                   'video=%s' % video_id)
         data_json = self._download_webpage(api_url, display_id)
         data = json.loads(data_json)['api']['results'][0]
         dt = datetime.datetime.fromtimestamp(data['pubdate_epoch'])
-        upload_date = dt.strftime('%Y%m%d')
+        pload_date = dt.strftime('%Y%m%d')
 
         duration = None
         thumbnails = []
@@ -54,12 +56,17 @@ class BlinkxIE(InfoExtractor):
                 })
             elif m['type'] == 'original':
                 duration = m['d']
+            elif m['type'] == 'youtube':
+                yt_id = m['link']
+                self.to_screen(u'Youtube video detected: %s' % yt_id)
+                return self.url_result(yt_id, 'Youtube', video_id=yt_id)
             elif m['type'] in ('flv', 'mp4'):
                 vcodec = remove_start(m['vcodec'], 'ff')
                 acodec = remove_start(m['acodec'], 'ff')
+                tbr = (int(m['vbr']) + int(m['abr'])) // 1000
                 format_id = (u'%s-%sk-%s' %
                              (vcodec,
-                              (int(m['vbr']) + int(m['abr'])) // 1000,
+                              tbr,
                               m['w']))
                 formats.append({
                     'format_id': format_id,
@@ -68,10 +75,12 @@ class BlinkxIE(InfoExtractor):
                     'acodec': acodec,
                     'abr': int(m['abr']) // 1000,
                     'vbr': int(m['vbr']) // 1000,
+                    'tbr': tbr,
                     'width': int(m['w']),
                     'height': int(m['h']),
                 })
-        formats.sort(key=lambda f: (f['width'], f['vbr'], f['abr']))
+
+        self._sort_formats(formats)
 
         return {
             'id': display_id,
@@ -79,7 +88,7 @@ class BlinkxIE(InfoExtractor):
             'title': data['title'],
             'formats': formats,
             'uploader': data['channel_name'],
-            'upload_date': upload_date,
+            'upload_date': pload_date,
             'description': data.get('description'),
             'thumbnails': thumbnails,
             'duration': duration,

youtube_dl/extractor/bliptv.py

@@ -1,157 +1,145 @@
+from __future__ import unicode_literals
+
 import datetime
-import json
-import os
 import re
-import socket
 
 from .common import InfoExtractor
+from .subtitles import SubtitlesInfoExtractor
 from ..utils import (
-    compat_http_client,
-    compat_parse_qs,
     compat_str,
-    compat_urllib_error,
-    compat_urllib_parse_urlparse,
     compat_urllib_request,
 
-    ExtractorError,
     unescapeHTML,
 )
 
 
-class BlipTVIE(InfoExtractor):
+class BlipTVIE(SubtitlesInfoExtractor):
     """Information extractor for blip.tv"""
 
-    _VALID_URL = r'^(?:https?://)?(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(.+)$'
+    _VALID_URL = r'https?://(?:\w+\.)?blip\.tv/((.+/)|(play/)|(api\.swf#))(?P<presumptive_id>.+)$'
-    _URL_EXT = r'^.*\.([a-z0-9]+)$'
-    IE_NAME = u'blip.tv'
-    _TEST = {
-        u'url': u'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
-        u'file': u'5779306.m4v',
-        u'md5': u'80baf1ec5c3d2019037c1c707d676b9f',
-        u'info_dict': {
-            u"upload_date": u"20111205", 
-            u"description": u"md5:9bc31f227219cde65e47eeec8d2dc596", 
-            u"uploader": u"Comic Book Resources - CBR TV", 
-            u"title": u"CBR EXCLUSIVE: \"Gotham City Imposters\" Bats VS Jokerz Short 3"
-        }
-    }
 
-    def report_direct_download(self, title):
+    _TESTS = [{
-        """Report information extraction."""
+        'url': 'http://blip.tv/cbr/cbr-exclusive-gotham-city-imposters-bats-vs-jokerz-short-3-5796352',
-        self.to_screen(u'%s: Direct download detected' % title)
+        'md5': 'c6934ad0b6acf2bd920720ec888eb812',
+        'info_dict': {
+            'id': '5779306',
+            'ext': 'mov',
+            'upload_date': '20111205',
+            'description': 'md5:9bc31f227219cde65e47eeec8d2dc596',
+            'uploader': 'Comic Book Resources - CBR TV',
+            'title': 'CBR EXCLUSIVE: "Gotham City Imposters" Bats VS Jokerz Short 3',
+        }
+    }, {
+        # https://github.com/rg3/youtube-dl/pull/2274
+        'note': 'Video with subtitles',
+        'url': 'http://blip.tv/play/h6Uag5OEVgI.html',
+        'md5': '309f9d25b820b086ca163ffac8031806',
+        'info_dict': {
+            'id': '6586561',
+            'ext': 'mp4',
+            'uploader': 'Red vs. Blue',
+            'description': 'One-Zero-One',
+            'upload_date': '20130614',
+            'title': 'Red vs. Blue Season 11 Episode 1',
+        }
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
+        presumptive_id = mobj.group('presumptive_id')
-            raise ExtractorError(u'Invalid URL: %s' % url)
 
         # See https://github.com/rg3/youtube-dl/issues/857
-        api_mobj = re.match(r'http://a\.blip\.tv/api\.swf#(?P<video_id>[\d\w]+)', url)
+        embed_mobj = re.match(r'https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', url)
-        if api_mobj is not None:
+        if embed_mobj:
-            url = 'http://blip.tv/play/g_%s' % api_mobj.group('video_id')
+            info_url = 'http://blip.tv/play/%s.x?p=1' % embed_mobj.group(1)
-        urlp = compat_urllib_parse_urlparse(url)
+            info_page = self._download_webpage(info_url, embed_mobj.group(1))
-        if urlp.path.startswith('/play/'):
+            video_id = self._search_regex(
-            response = self._request_webpage(url, None, False)
+                r'data-episode-id="([0-9]+)', info_page, 'video_id')
-            redirecturl = response.geturl()
+            return self.url_result('http://blip.tv/a/a-' + video_id, 'BlipTV')
-            rurlp = compat_urllib_parse_urlparse(redirecturl)
-            file_id = compat_parse_qs(rurlp.fragment)['file'][0].rpartition('/')[2]
-            url = 'http://blip.tv/a/a-' + file_id
-            return self._real_extract(url)
         
-
+        cchar = '&' if '?' in url else '?'
-        if '?' in url:
-            cchar = '&'
-        else:
-            cchar = '?'
         json_url = url + cchar + 'skin=json&version=2&no_wrap=1'
         request = compat_urllib_request.Request(json_url)
         request.add_header('User-Agent', 'iTunes/10.6.1')
-        self.report_extraction(mobj.group(1))
-        info = None
-        urlh = self._request_webpage(request, None, False,
-            u'unable to download video info webpage')
-        if urlh.headers.get('Content-Type', '').startswith('video/'): # Direct download
-            basename = url.split('/')[-1]
-            title,ext = os.path.splitext(basename)
-            title = title.decode('UTF-8')
-            ext = ext.replace('.', '')
-            self.report_direct_download(title)
-            info = {
-                'id': title,
-                'url': url,
-                'uploader': None,
-                'upload_date': None,
-                'title': title,
-                'ext': ext,
-                'urlhandle': urlh
-            }
-        if info is None: # Regular URL
-            try:
-                json_code_bytes = urlh.read()
-                json_code = json_code_bytes.decode('utf-8')
-            except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-                raise ExtractorError(u'Unable to read video info webpage: %s' % compat_str(err))
 
-            try:
+        json_data = self._download_json(request, video_id=presumptive_id)
-                json_data = json.loads(json_code)
+
         if 'Post' in json_data:
             data = json_data['Post']
         else:
             data = json_data
 
+        video_id = compat_str(data['item_id'])
         upload_date = datetime.datetime.strptime(data['datestamp'], '%m-%d-%y %H:%M%p').strftime('%Y%m%d')
+        subtitles = {}
+        formats = []
         if 'additionalMedia' in data:
-                    formats = sorted(data['additionalMedia'], key=lambda f: int(f['media_height']))
+            for f in data['additionalMedia']:
-                    best_format = formats[-1]
+                if f.get('file_type_srt') == 1:
-                    video_url = best_format['url']
+                    LANGS = {
+                        'english': 'en',
+                    }
+                    lang = f['role'].rpartition('-')[-1].strip().lower()
+                    langcode = LANGS.get(lang, lang)
+                    subtitles[langcode] = f['url']
+                    continue
+                if not int(f['media_width']):  # filter m3u8
+                    continue
+                formats.append({
+                    'url': f['url'],
+                    'format_id': f['role'],
+                    'width': int(f['media_width']),
+                    'height': int(f['media_height']),
+                })
         else:
-                    video_url = data['media']['url']
+            formats.append({
-                umobj = re.match(self._URL_EXT, video_url)
+                'url': data['media']['url'],
-                if umobj is None:
+                'width': int(data['media']['width']),
-                    raise ValueError('Can not determine filename extension')
+                'height': int(data['media']['height']),
-                ext = umobj.group(1)
+            })
+        self._sort_formats(formats)
 
-                info = {
+        # subtitles
-                    'id': compat_str(data['item_id']),
+        video_subtitles = self.extract_subtitles(video_id, subtitles)
-                    'url': video_url,
+        if self._downloader.params.get('listsubtitles', False):
+            self._list_available_subtitles(video_id, subtitles)
+            return
+
+        return {
+            'id': video_id,
             'uploader': data['display_name'],
             'upload_date': upload_date,
             'title': data['title'],
-                    'ext': ext,
-                    'format': data['media']['mimeType'],
             'thumbnail': data['thumbnailUrl'],
             'description': data['description'],
-                    'player_url': data['embedUrl'],
             'user_agent': 'iTunes/10.6.1',
+            'formats': formats,
+            'subtitles': video_subtitles,
         }
-            except (ValueError,KeyError) as err:
-                raise ExtractorError(u'Unable to parse video information: %s' % repr(err))
 
-        return [info]
+    def _download_subtitle_url(self, sub_lang, url):
+        # For some weird reason, blip.tv serves a video instead of subtitles
+        # when we request with a common UA
+        req = compat_urllib_request.Request(url)
+        req.add_header('Youtubedl-user-agent', 'youtube-dl')
+        return self._download_webpage(req, None, note=False)
 
 
 class BlipTVUserIE(InfoExtractor):
-    """Information Extractor for blip.tv users."""
-
     _VALID_URL = r'(?:(?:(?:https?://)?(?:\w+\.)?blip\.tv/)|bliptvuser:)([^/]+)/*$'
     _PAGE_SIZE = 12
-    IE_NAME = u'blip.tv:user'
+    IE_NAME = 'blip.tv:user'
 
     def _real_extract(self, url):
-        # Extract username
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
-
         username = mobj.group(1)
 
         page_base = 'http://m.blip.tv/pr/show_get_full_episode_list?users_id=%s&lite=0&esi=1'
 
-        page = self._download_webpage(url, username, u'Downloading user page')
+        page = self._download_webpage(url, username, 'Downloading user page')
         mobj = re.search(r'data-users-id="([^"]+)"', page)
         page_base = page_base % mobj.group(1)
 
-
         # Download video ids using BlipTV Ajax calls. Result size per
         # query is limited (currently to 12 videos) so we need to query
         # page by page until there are no video ids - it means we got
@@ -162,8 +150,8 @@ class BlipTVUserIE(InfoExtractor):
 
         while True:
             url = page_base + "&page=" + str(pagenum)
-            page = self._download_webpage(url, username,
+            page = self._download_webpage(
-                                          u'Downloading video ids from page %d' % pagenum)
+                url, username, 'Downloading video ids from page %d' % pagenum)
 
             # Extract video identifiers
             ids_in_page = []
@@ -185,6 +173,6 @@ class BlipTVUserIE(InfoExtractor):
 
             pagenum += 1
 
-        urls = [u'http://blip.tv/%s' % video_id for video_id in video_ids]
+        urls = ['http://blip.tv/%s' % video_id for video_id in video_ids]
         url_entries = [self.url_result(vurl, 'BlipTV') for vurl in urls]
         return [self.playlist_result(url_entries, playlist_title=username)]

youtube_dl/extractor/bloomberg.py

@@ -1,6 +1,7 @@
 import re
 
 from .common import InfoExtractor
+from .ooyala import OoyalaIE
 
 
 class BloombergIE(InfoExtractor):
@@ -23,5 +24,7 @@ class BloombergIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         name = mobj.group('name')
         webpage = self._download_webpage(url, name)
-        ooyala_url = self._og_search_video_url(webpage)
+        embed_code = self._search_regex(
-        return self.url_result(ooyala_url, ie='Ooyala')
+            r'<source src="https?://[^/]+/[^/]+/[^/]+/([^/]+)', webpage,
+            'embed code')
+        return OoyalaIE._build_url_result(embed_code)

youtube_dl/extractor/breakcom.py

@@ -1,18 +1,20 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
 from .common import InfoExtractor
-from ..utils import determine_ext
 
 
 class BreakIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?break\.com/video/([^/]+)'
+    _VALID_URL = r'http://(?:www\.)?break\.com/video/([^/]+)'
     _TEST = {
-        u'url': u'http://www.break.com/video/when-girls-act-like-guys-2468056',
+        'url': 'http://www.break.com/video/when-girls-act-like-guys-2468056',
-        u'file': u'2468056.mp4',
+        'md5': 'a3513fb1547fba4fb6cfac1bffc6c46b',
-        u'md5': u'a3513fb1547fba4fb6cfac1bffc6c46b',
+        'info_dict': {
-        u'info_dict': {
+            'id': '2468056',
-            u"title": u"When Girls Act Like D-Bags"
+            'ext': 'mp4',
+            'title': 'When Girls Act Like D-Bags',
         }
     }
 
@@ -22,17 +24,16 @@ class BreakIE(InfoExtractor):
         embed_url = 'http://www.break.com/embed/%s' % video_id
         webpage = self._download_webpage(embed_url, video_id)
         info_json = self._search_regex(r'var embedVars = ({.*?});', webpage,
-                                       u'info json', flags=re.DOTALL)
+                                       'info json', flags=re.DOTALL)
         info = json.loads(info_json)
         video_url = info['videoUri']
         m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', video_url)
         if m_youtube is not None:
             return self.url_result(m_youtube.group(1), 'Youtube')
         final_url = video_url + '?' + info['AuthToken']
-        return [{
+        return {
             'id': video_id,
             'url': final_url,
-            'ext':       determine_ext(final_url),
             'title': info['contentName'],
             'thumbnail': info['thumbUri'],
-        }]
+        }

youtube_dl/extractor/brightcove.py

@@ -1,4 +1,5 @@
 # encoding: utf-8
+from __future__ import unicode_literals
 
 import re
 import json
@@ -8,65 +9,68 @@ from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
     find_xpath_attr,
+    fix_xml_ampersands,
     compat_urlparse,
     compat_str,
     compat_urllib_request,
+    compat_parse_qs,
 
     ExtractorError,
+    unsmuggle_url,
+    unescapeHTML,
 )
 
 
 class BrightcoveIE(InfoExtractor):
     _VALID_URL = r'https?://.*brightcove\.com/(services|viewer).*\?(?P<query>.*)'
     _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
-    _PLAYLIST_URL_TEMPLATE = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s'
 
     _TESTS = [
         {
             # From http://www.8tv.cat/8aldia/videos/xavier-sala-i-martin-aquesta-tarda-a-8-al-dia/
-            u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
+            'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1654948606001&flashID=myExperience&%40videoPlayer=2371591881001',
-            u'file': u'2371591881001.mp4',
+            'file': '2371591881001.mp4',
-            u'md5': u'8eccab865181d29ec2958f32a6a754f5',
+            'md5': '5423e113865d26e40624dce2e4b45d95',
-            u'note': u'Test Brightcove downloads and detection in GenericIE',
+            'note': 'Test Brightcove downloads and detection in GenericIE',
-            u'info_dict': {
+            'info_dict': {
-                u'title': u'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
+                'title': 'Xavier Sala i Martín: “Un banc que no presta és un banc zombi que no serveix per a res”',
-                u'uploader': u'8TV',
+                'uploader': '8TV',
-                u'description': u'md5:a950cc4285c43e44d763d036710cd9cd',
+                'description': 'md5:a950cc4285c43e44d763d036710cd9cd',
             }
         },
         {
             # From http://medianetwork.oracle.com/video/player/1785452137001
-            u'url': u'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
+            'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=1217746023001&flashID=myPlayer&%40videoPlayer=1785452137001',
-            u'file': u'1785452137001.flv',
+            'file': '1785452137001.flv',
-            u'info_dict': {
+            'info_dict': {
-                u'title': u'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
+                'title': 'JVMLS 2012: Arrays 2.0 - Opportunities and Challenges',
-                u'description': u'John Rose speaks at the JVM Language Summit, August 1, 2012.',
+                'description': 'John Rose speaks at the JVM Language Summit, August 1, 2012.',
-                u'uploader': u'Oracle',
+                'uploader': 'Oracle',
             },
         },
         {
             # From http://mashable.com/2013/10/26/thermoelectric-bracelet-lets-you-control-your-body-temperature/
-            u'url': u'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001',
+            'url': 'http://c.brightcove.com/services/viewer/federated_f9?&playerID=1265504713001&publisherID=AQ%7E%7E%2CAAABBzUwv1E%7E%2CxP-xFHVUstiMFlNYfvF4G9yFnNaqCw_9&videoID=2750934548001',
-            u'info_dict': {
+            'info_dict': {
-                u'id': u'2750934548001',
+                'id': '2750934548001',
-                u'ext': u'mp4',
+                'ext': 'mp4',
-                u'title': u'This Bracelet Acts as a Personal Thermostat',
+                'title': 'This Bracelet Acts as a Personal Thermostat',
-                u'description': u'md5:547b78c64f4112766ccf4e151c20b6a0',
+                'description': 'md5:547b78c64f4112766ccf4e151c20b6a0',
-                u'uploader': u'Mashable',
+                'uploader': 'Mashable',
             },
         },
         {
             # test that the default referer works
             # from http://national.ballet.ca/interact/video/Lost_in_Motion_II/
-            u'url': u'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001',
+            'url': 'http://link.brightcove.com/services/player/bcpid756015033001?bckey=AQ~~,AAAApYJi_Ck~,GxhXCegT1Dp39ilhXuxMJxasUhVNZiil&bctid=2878862109001',
-            u'info_dict': {
+            'info_dict': {
-                u'id': u'2878862109001',
+                'id': '2878862109001',
-                u'ext': u'mp4',
+                'ext': 'mp4',
-                u'title': u'Lost in Motion II',
+                'title': 'Lost in Motion II',
-                u'description': u'md5:363109c02998fee92ec02211bd8000df',
+                'description': 'md5:363109c02998fee92ec02211bd8000df',
-                u'uploader': u'National Ballet of Canada',
+                'uploader': 'National Ballet of Canada',
-            },
             },
+        }
     ]
 
     @classmethod
@@ -80,18 +84,34 @@ class BrightcoveIE(InfoExtractor):
         object_str = re.sub(r'(<param name="[^"]+" value="[^"]+")>',
                             lambda m: m.group(1) + '/>', object_str)
         # Fix up some stupid XML, see https://github.com/rg3/youtube-dl/issues/1608
-        object_str = object_str.replace(u'<--', u'<!--')
+        object_str = object_str.replace('<--', '<!--')
+        object_str = fix_xml_ampersands(object_str)
 
         object_doc = xml.etree.ElementTree.fromstring(object_str)
-        assert u'BrightcoveExperience' in object_doc.attrib['class']
+
-        params = {'flashID': object_doc.attrib['id'],
+        fv_el = find_xpath_attr(object_doc, './param', 'name', 'flashVars')
-                  'playerID': find_xpath_attr(object_doc, './param', 'name', 'playerID').attrib['value'],
+        if fv_el is not None:
-                  }
+            flashvars = dict(
+                (k, v[0])
+                for k, v in compat_parse_qs(fv_el.attrib['value']).items())
+        else:
+            flashvars = {}
+
         def find_param(name):
+            if name in flashvars:
+                return flashvars[name]
             node = find_xpath_attr(object_doc, './param', 'name', name)
             if node is not None:
                 return node.attrib['value']
             return None
+
+        params = {}
+
+        playerID = find_param('playerID')
+        if playerID is None:
+            raise ExtractorError('Cannot find player ID')
+        params['playerID'] = playerID
+
         playerKey = find_param('playerKey')
         # Not all pages define this value
         if playerKey is not None:
@@ -108,18 +128,32 @@ class BrightcoveIE(InfoExtractor):
 
     @classmethod
     def _extract_brightcove_url(cls, webpage):
-        """Try to extract the brightcove url from the wepbage, returns None
+        """Try to extract the brightcove url from the webpage, returns None
         if it can't be found
         """
-        m_brightcove = re.search(
+        urls = cls._extract_brightcove_urls(webpage)
-            r'<object[^>]+?class=([\'"])[^>]*?BrightcoveExperience.*?\1.+?</object>',
+        return urls[0] if urls else None
-            webpage, re.DOTALL)
+
-        if m_brightcove is not None:
+    @classmethod
-            return cls._build_brighcove_url(m_brightcove.group())
+    def _extract_brightcove_urls(cls, webpage):
-        else:
+        """Return a list of all Brightcove URLs from the webpage """
-            return None
+
+        url_m = re.search(r'<meta\s+property="og:video"\s+content="(http://c.brightcove.com/[^"]+)"', webpage)
+        if url_m:
+            return [unescapeHTML(url_m.group(1))]
+
+        matches = re.findall(
+            r'''(?sx)<object
+            (?:
+                [^>]+?class=[\'"][^>]*?BrightcoveExperience.*?[\'"] |
+                [^>]*?>\s*<param\s+name="movie"\s+value="https?://[^/]*brightcove\.com/
+            ).+?</object>''',
+            webpage)
+        return [cls._build_brighcove_url(m) for m in matches]
 
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
+
         # Change the 'videoId' and others field to '@videoPlayer'
         url = re.sub(r'(?<=[?&])(videoI(d|D)|bctid)', '%40videoPlayer', url)
         # Change bckey (used by bcove.me urls) to playerKey
@@ -130,9 +164,10 @@ class BrightcoveIE(InfoExtractor):
 
         videoPlayer = query.get('@videoPlayer')
         if videoPlayer:
-            return self._get_video_info(videoPlayer[0], query_str, query,
             # We set the original url as the default 'Referer' header
-                referer=url)
+            referer = smuggled_data.get('Referer', url)
+            return self._get_video_info(
+                videoPlayer[0], query_str, query, referer=referer)
         else:
             player_key = query['playerKey']
             return self._get_playlist_info(player_key[0])
@@ -151,16 +186,18 @@ class BrightcoveIE(InfoExtractor):
         info = self._search_regex(r'var experienceJSON = ({.*?});', webpage, 'json')
         info = json.loads(info)['data']
         video_info = info['programmedContent']['videoPlayer']['mediaDTO']
+        video_info['_youtubedl_adServerURL'] = info.get('adServerURL')
 
         return self._extract_video_info(video_info)
 
     def _get_playlist_info(self, player_key):
-        playlist_info = self._download_webpage(self._PLAYLIST_URL_TEMPLATE % player_key,
+        info_url = 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=%s' % player_key
-                                               player_key, u'Downloading playlist information')
+        playlist_info = self._download_webpage(
+            info_url, player_key, 'Downloading playlist information')
 
         json_data = json.loads(playlist_info)
         if 'videoList' not in json_data:
-            raise ExtractorError(u'Empty playlist')
+            raise ExtractorError('Empty playlist')
         playlist_info = json_data['videoList']
         videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
 
@@ -170,7 +207,7 @@ class BrightcoveIE(InfoExtractor):
     def _extract_video_info(self, video_info):
         info = {
             'id': compat_str(video_info['id']),
-            'title': video_info['displayName'],
+            'title': video_info['displayName'].strip(),
             'description': video_info.get('shortDescription'),
             'thumbnail': video_info.get('videoStillURL') or video_info.get('thumbnailURL'),
             'uploader': video_info.get('publisherName'),
@@ -188,6 +225,23 @@ class BrightcoveIE(InfoExtractor):
             info.update({
                 'url': video_info['FLVFullLengthURL'],
             })
+
+        if self._downloader.params.get('include_ads', False):
+            adServerURL = video_info.get('_youtubedl_adServerURL')
+            if adServerURL:
+                ad_info = {
+                    '_type': 'url',
+                    'url': adServerURL,
+                }
+                if 'url' in info:
+                    return {
+                        '_type': 'playlist',
+                        'title': info['title'],
+                        'entries': [ad_info, info],
+                    }
                 else:
-            raise ExtractorError(u'Unable to extract video url for %s' % info['id'])
+                    return ad_info
+
+        if 'url' not in info and not info.get('formats'):
+            raise ExtractorError('Unable to extract video url for %s' % info['id'])
         return info

youtube_dl/extractor/c56.py

@@ -1,21 +1,21 @@
 # coding: utf-8
+from __future__ import unicode_literals
 
 import re
 import json
 
 from .common import InfoExtractor
-from ..utils import determine_ext
+
 
 class C56IE(InfoExtractor):
     _VALID_URL = r'https?://((www|player)\.)?56\.com/(.+?/)?(v_|(play_album.+-))(?P<textid>.+?)\.(html|swf)'
-    IE_NAME = u'56.com'
+    IE_NAME = '56.com'
-
     _TEST = {
-        u'url': u'http://www.56.com/u39/v_OTM0NDA3MTY.html',
+        'url': 'http://www.56.com/u39/v_OTM0NDA3MTY.html',
-        u'file': u'93440716.flv',
+        'file': '93440716.flv',
-        u'md5': u'e59995ac63d0457783ea05f93f12a866',
+        'md5': 'e59995ac63d0457783ea05f93f12a866',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'网事知多少 第32期：车怒',
+            'title': '网事知多少 第32期：车怒',
         },
     }
 
@@ -23,14 +23,18 @@ class C56IE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
         text_id = mobj.group('textid')
         info_page = self._download_webpage('http://vxml.56.com/json/%s/' % text_id,
-                                           text_id, u'Downloading video info')
+                                           text_id, 'Downloading video info')
         info = json.loads(info_page)['info']
-        best_format = sorted(info['rfiles'], key=lambda f: int(f['filesize']))[-1]
+        formats = [{
-        video_url = best_format['url']
+            'format_id': f['type'],
+            'filesize': int(f['filesize']),
+            'url': f['url']
+        } for f in info['rfiles']]
+        self._sort_formats(formats)
 
-        return {'id': info['vid'],
+        return {
+            'id': info['vid'],
             'title': info['Subject'],
-                'url': video_url,
+            'formats': formats,
-                'ext': determine_ext(video_url),
             'thumbnail': info.get('bimg') or info.get('img'),
         }

youtube_dl/extractor/channel9.py

@@ -1,4 +1,4 @@
-# encoding: utf-8
+from __future__ import unicode_literals
 
 import re
 
@@ -13,36 +13,38 @@ class Channel9IE(InfoExtractor):
     meta Search.PageType from web page HTML rather than URL itself, as it is
     not always possible to do.
     '''
-    IE_DESC = u'Channel 9'
+    IE_DESC = 'Channel 9'
-    IE_NAME = u'channel9'
+    IE_NAME = 'channel9'
-    _VALID_URL = r'^https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+)/?'
+    _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+)/?'
 
     _TESTS = [
         {
-            u'url': u'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
+            'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
-            u'file': u'Events_TechEd_Australia_2013_KOS002.mp4',
+            'md5': 'bbd75296ba47916b754e73c3a4bbdf10',
-            u'md5': u'bbd75296ba47916b754e73c3a4bbdf10',
+            'info_dict': {
-            u'info_dict': {
+                'id': 'Events/TechEd/Australia/2013/KOS002',
-                u'title': u'Developer Kick-Off Session: Stuff We Love',
+                'ext': 'mp4',
-                u'description': u'md5:c08d72240b7c87fcecafe2692f80e35f',
+                'title': 'Developer Kick-Off Session: Stuff We Love',
-                u'duration': 4576,
+                'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
-                u'thumbnail': u'http://media.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg',
+                'duration': 4576,
-                u'session_code': u'KOS002',
+                'thumbnail': 'http://media.ch9.ms/ch9/9d51/03902f2d-fc97-4d3c-b195-0bfe15a19d51/KOS002_220.jpg',
-                u'session_day': u'Day 1',
+                'session_code': 'KOS002',
-                u'session_room': u'Arena 1A',
+                'session_day': 'Day 1',
-                u'session_speakers': [ u'Ed Blankenship', u'Andrew Coates', u'Brady Gaster', u'Patrick Klug', u'Mads Kristensen' ],
+                'session_room': 'Arena 1A',
+                'session_speakers': [ 'Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug', 'Mads Kristensen' ],
             },
         },
         {
-            u'url': u'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
+            'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
-            u'file': u'posts_Self-service-BI-with-Power-BI-nuclear-testing.mp4',
+            'md5': 'b43ee4529d111bc37ba7ee4f34813e68',
-            u'md5': u'b43ee4529d111bc37ba7ee4f34813e68',
+            'info_dict': {
-            u'info_dict': {
+                'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing',
-                u'title': u'Self-service BI with Power BI - nuclear testing',
+                'ext': 'mp4',
-                u'description': u'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
+                'title': 'Self-service BI with Power BI - nuclear testing',
-                u'duration': 1540,
+                'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
-                u'thumbnail': u'http://media.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg',
+                'duration': 1540,
-                u'authors': [ u'Mike Wilmot' ],
+                'thumbnail': 'http://media.ch9.ms/ch9/87e1/0300391f-a455-4c72-bec3-4422f19287e1/selfservicenuk_512.jpg',
+                'authors': [ 'Mike Wilmot' ],
             },
         }
     ]
@@ -60,7 +62,7 @@ class Channel9IE(InfoExtractor):
             return 0
         units = m.group('units')
         try:
-            exponent = [u'B', u'KB', u'MB', u'GB', u'TB', u'PB', u'EB', u'ZB', u'YB'].index(units.upper())
+            exponent = ['B', 'KB', 'MB', 'GB', 'TB', 'PB', 'EB', 'ZB', 'YB'].index(units.upper())
         except ValueError:
             return 0
         size = float(m.group('size'))
@@ -76,21 +78,25 @@ class Channel9IE(InfoExtractor):
             </div>)?                                                # File size part may be missing
         '''
         # Extract known formats
-        formats = [{'url': x.group('url'),
+        formats = [{
+            'url': x.group('url'),
             'format_id': x.group('quality'),
             'format_note': x.group('note'),
             'format': '%s (%s)' % (x.group('quality'), x.group('note')),
             'filesize': self._restore_bytes(x.group('filesize')), # File size is approximate
+            'preference': self._known_formats.index(x.group('quality')),
+            'vcodec': 'none' if x.group('note') == 'Audio only' else None,
         } for x in list(re.finditer(FORMAT_REGEX, html)) if x.group('quality') in self._known_formats]
-        # Sort according to known formats list
+
-        formats.sort(key=lambda fmt: self._known_formats.index(fmt['format_id']))
+        self._sort_formats(formats)
+
         return formats
 
     def _extract_title(self, html):
-        title = self._html_search_meta(u'title', html, u'title')
+        title = self._html_search_meta('title', html, 'title')
         if title is None:           
             title = self._og_search_title(html)
-            TITLE_SUFFIX = u' (Channel 9)'
+            TITLE_SUFFIX = ' (Channel 9)'
             if title is not None and title.endswith(TITLE_SUFFIX):
                 title = title[:-len(TITLE_SUFFIX)]
         return title
@@ -106,7 +112,7 @@ class Channel9IE(InfoExtractor):
         m = re.search(DESCRIPTION_REGEX, html)
         if m is not None:
             return m.group('description')
-        return self._html_search_meta(u'description', html, u'description')
+        return self._html_search_meta('description', html, 'description')
 
     def _extract_duration(self, html):
         m = re.search(r'data-video_duration="(?P<hours>\d{2}):(?P<minutes>\d{2}):(?P<seconds>\d{2})"', html)
@@ -168,7 +174,7 @@ class Channel9IE(InfoExtractor):
 
         # Nothing to download
         if len(formats) == 0 and slides is None and zip_ is None:
-            self._downloader.report_warning(u'None of recording, slides or zip are available for %s' % content_path)
+            self._downloader.report_warning('None of recording, slides or zip are available for %s' % content_path)
             return
 
         # Extract meta
@@ -240,7 +246,7 @@ class Channel9IE(InfoExtractor):
         return contents
 
     def _extract_list(self, content_path):
-        rss = self._download_xml(self._RSS_URL % content_path, content_path, u'Downloading RSS')
+        rss = self._download_xml(self._RSS_URL % content_path, content_path, 'Downloading RSS')
         entries = [self.url_result(session_url.text, 'Channel9')
                    for session_url in rss.findall('./channel/item/link')]
         title_text = rss.find('./channel/title').text
@@ -250,11 +256,11 @@ class Channel9IE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         content_path = mobj.group('contentpath')
 
-        webpage = self._download_webpage(url, content_path, u'Downloading web page')
+        webpage = self._download_webpage(url, content_path, 'Downloading web page')
 
         page_type_m = re.search(r'<meta name="Search.PageType" content="(?P<pagetype>[^"]+)"/>', webpage)
         if page_type_m is None:
-            raise ExtractorError(u'Search.PageType not found, don\'t know how to process this page', expected=True)
+            raise ExtractorError('Search.PageType not found, don\'t know how to process this page', expected=True)
 
         page_type = page_type_m.group('pagetype')
         if page_type == 'List':         # List page, may contain list of 'item'-like objects
@@ -264,4 +270,4 @@ class Channel9IE(InfoExtractor):
         elif page_type == 'Session':    # Event session page, may contain downloadable content
             return self._extract_session(webpage, content_path)
         else:
-            raise ExtractorError(u'Unexpected Search.PageType %s' % page_type, expected=True)
+            raise ExtractorError('Unexpected Search.PageType %s' % page_type, expected=True)

youtube_dl/extractor/chilloutzone.py

@@ -0,0 +1,97 @@
+from __future__ import unicode_literals
+
+import re
+import base64
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    clean_html,
+    ExtractorError
+)
+
+
+class ChilloutzoneIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?chilloutzone\.net/video/(?P<id>[\w|-]+)\.html'
+    _TESTS = [{
+        'url': 'http://www.chilloutzone.net/video/enemene-meck-alle-katzen-weg.html',
+        'md5': 'a76f3457e813ea0037e5244f509e66d1',
+        'info_dict': {
+            'id': 'enemene-meck-alle-katzen-weg',
+            'ext': 'mp4',
+            'title': 'Enemene Meck - Alle Katzen weg',
+            'description': 'Ist das der Umkehrschluss des Niesenden Panda-Babys?',
+        },
+    }, {
+        'note': 'Video hosted at YouTube',
+        'url': 'http://www.chilloutzone.net/video/eine-sekunde-bevor.html',
+        'info_dict': {
+            'id': '1YVQaAgHyRU',
+            'ext': 'mp4',
+            'title': '16 Photos Taken 1 Second Before Disaster',
+            'description': 'md5:58a8fcf6a459fe0a08f54140f0ad1814',
+            'uploader': 'BuzzFeedVideo',
+            'uploader_id': 'BuzzFeedVideo',
+            'upload_date': '20131105',
+        },
+    }, {
+        'note': 'Video hosted at Vimeo',
+        'url': 'http://www.chilloutzone.net/video/icon-blending.html',
+        'md5': '2645c678b8dc4fefcc0e1b60db18dac1',
+        'info_dict': {
+            'id': '85523671',
+            'ext': 'mp4',
+            'title': 'The Sunday Times - Icons',
+            'description': 'md5:3e1c0dc6047498d6728dcdaad0891762',
+            'uploader': 'Us',
+            'uploader_id': 'usfilms',
+            'upload_date': '20140131'
+        },
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        base64_video_info = self._html_search_regex(
+            r'var cozVidData = "(.+?)";', webpage, 'video data')
+        decoded_video_info = base64.b64decode(base64_video_info).decode("utf-8")
+        video_info_dict = json.loads(decoded_video_info)
+
+        # get video information from dict
+        video_url = video_info_dict['mediaUrl']
+        description = clean_html(video_info_dict.get('description'))
+        title = video_info_dict['title']
+        native_platform = video_info_dict['nativePlatform']
+        native_video_id = video_info_dict['nativeVideoId']
+        source_priority = video_info_dict['sourcePriority']
+
+        # If nativePlatform is None a fallback mechanism is used (i.e. youtube embed)
+        if native_platform is None:
+            youtube_url = self._html_search_regex(
+                r'<iframe.* src="((?:https?:)?//(?:[^.]+\.)?youtube\.com/.+?)"',
+                webpage, 'fallback video URL', default=None)
+            if youtube_url is not None:
+                return self.url_result(youtube_url, ie='Youtube')
+
+        # Non Fallback: Decide to use native source (e.g. youtube or vimeo) or
+        # the own CDN
+        if source_priority == 'native':
+            if native_platform == 'youtube':
+                return self.url_result(native_video_id, ie='Youtube')
+            if native_platform == 'vimeo':
+                return self.url_result(
+                    'http://vimeo.com/' + native_video_id, ie='Vimeo')
+
+        if not video_url:
+            raise ExtractorError('No video found')
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'ext': 'mp4',
+            'title': title,
+            'description': description,
+        }

youtube_dl/extractor/cliphunter.py

@@ -0,0 +1,56 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+translation_table = {
+    'a': 'h', 'd': 'e', 'e': 'v', 'f': 'o', 'g': 'f', 'i': 'd', 'l': 'n',
+    'm': 'a', 'n': 'm', 'p': 'u', 'q': 't', 'r': 's', 'v': 'p', 'x': 'r',
+    'y': 'l', 'z': 'i',
+    '$': ':', '&': '.', '(': '=', '^': '&', '=': '/',
+}
+
+
+class CliphunterIE(InfoExtractor):
+    IE_NAME = 'cliphunter'
+
+    _VALID_URL = r'''(?x)http://(?:www\.)?cliphunter\.com/w/
+        (?P<id>[0-9]+)/
+        (?P<seo>.+?)(?:$|[#\?])
+    '''
+    _TEST = {
+        'url': 'http://www.cliphunter.com/w/1012420/Fun_Jynx_Maze_solo',
+        'file': '1012420.flv',
+        'md5': '15e7740f30428abf70f4223478dc1225',
+        'info_dict': {
+            'title': 'Fun Jynx Maze solo',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        pl_fiji = self._search_regex(
+            r'pl_fiji = \'([^\']+)\'', webpage, 'video data')
+        pl_c_qual = self._search_regex(
+            r'pl_c_qual = "(.)"', webpage, 'video quality')
+        video_title = self._search_regex(
+            r'mediaTitle = "([^"]+)"', webpage, 'title')
+
+        video_url = ''.join(translation_table.get(c, c) for c in pl_fiji)
+
+        formats = [{
+            'url': video_url,
+            'format_id': pl_c_qual,
+        }]
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'formats': formats,
+        }

youtube_dl/extractor/clipsyndicate.py

@@ -3,7 +3,7 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     find_xpath_attr,
-    fix_xml_all_ampersand,
+    fix_xml_ampersands
 )
 
 
@@ -33,7 +33,7 @@ class ClipsyndicateIE(InfoExtractor):
         pdoc = self._download_xml(
             'http://eplayer.clipsyndicate.com/osmf/playlist?%s' % flvars,
             video_id, u'Downloading video info',
-            transform_source=fix_xml_all_ampersand) 
+            transform_source=fix_xml_ampersands)
 
         track_doc = pdoc.find('trackList/track')
         def find_param(name):

youtube_dl/extractor/cmt.py

@@ -0,0 +1,19 @@
+from .mtv import MTVIE
+
+class CMTIE(MTVIE):
+    IE_NAME = u'cmt.com'
+    _VALID_URL = r'https?://www\.cmt\.com/videos/.+?/(?P<videoid>[^/]+)\.jhtml'
+    _FEED_URL = 'http://www.cmt.com/sitewide/apps/player/embed/rss/'
+
+    _TESTS = [
+        {
+            u'url': u'http://www.cmt.com/videos/garth-brooks/989124/the-call-featuring-trisha-yearwood.jhtml#artist=30061',
+            u'md5': u'e6b7ef3c4c45bbfae88061799bbba6c2',
+            u'info_dict': {
+                u'id': u'989124',
+                u'ext': u'mp4',
+                u'title': u'Garth Brooks - "The Call (featuring Trisha Yearwood)"',
+                u'description': u'Blame It All On My Roots',
+            },
+        },
+    ]

youtube_dl/extractor/cnn.py

@@ -1,7 +1,13 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
-from ..utils import determine_ext
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    url_basename,
+)
 
 
 class CNNIE(InfoExtractor):
@@ -9,21 +15,24 @@ class CNNIE(InfoExtractor):
         (?P<path>.+?/(?P<title>[^/]+?)(?:\.cnn|(?=&)))'''
 
     _TESTS = [{
-        u'url': u'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
+        'url': 'http://edition.cnn.com/video/?/video/sports/2013/06/09/nadal-1-on-1.cnn',
-        u'file': u'sports_2013_06_09_nadal-1-on-1.cnn.mp4',
+        'file': 'sports_2013_06_09_nadal-1-on-1.cnn.mp4',
-        u'md5': u'3e6121ea48df7e2259fe73a0628605c4',
+        'md5': '3e6121ea48df7e2259fe73a0628605c4',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'Nadal wins 8th French Open title',
+            'title': 'Nadal wins 8th French Open title',
-            u'description': u'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
+            'description': 'World Sport\'s Amanda Davies chats with 2013 French Open champion Rafael Nadal.',
+            'duration': 135,
+            'upload_date': '20130609',
         },
     },
     {
-        u"url": u"http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29",
+        "url": "http://edition.cnn.com/video/?/video/us/2013/08/21/sot-student-gives-epic-speech.georgia-institute-of-technology&utm_source=feedburner&utm_medium=feed&utm_campaign=Feed%3A+rss%2Fcnn_topstories+%28RSS%3A+Top+Stories%29",
-        u"file": u"us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
+        "file": "us_2013_08_21_sot-student-gives-epic-speech.georgia-institute-of-technology.mp4",
-        u"md5": u"b5cc60c60a3477d185af8f19a2a26f4e",
+        "md5": "b5cc60c60a3477d185af8f19a2a26f4e",
-        u"info_dict": {
+        "info_dict": {
-            u"title": "Student's epic speech stuns new freshmen",
+            "title": "Student's epic speech stuns new freshmen",
-            u"description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\""
+            "description": "A Georgia Tech student welcomes the incoming freshmen with an epic speech backed by music from \"2001: A Space Odyssey.\"",
+            "upload_date": "20130821",
         }
     }]
 
@@ -31,26 +40,87 @@ class CNNIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         path = mobj.group('path')
         page_title = mobj.group('title')
-        info_url = u'http://cnn.com/video/data/3.0/%s/index.xml' % path
+        info_url = 'http://cnn.com/video/data/3.0/%s/index.xml' % path
         info = self._download_xml(info_url, page_title)
 
         formats = []
+        rex = re.compile(r'''(?x)
+            (?P<width>[0-9]+)x(?P<height>[0-9]+)
+            (?:_(?P<bitrate>[0-9]+)k)?
+        ''')
         for f in info.findall('files/file'):
-            mf = re.match(r'(\d+)x(\d+)(?:_(.*)k)?',f.attrib['bitrate'])
+            video_url = 'http://ht.cdn.turner.com/cnn/big%s' % (f.text.strip())
-            if mf is not None:
+            fdct = {
-                formats.append((int(mf.group(1)), int(mf.group(2)), int(mf.group(3) or 0), f.text))
+                'format_id': f.attrib['bitrate'],
-        formats = sorted(formats)
+                'url': video_url,
-        (_,_,_, video_path) = formats[-1]
+            }
-        video_url = 'http://ht.cdn.turner.com/cnn/big%s' % video_path
+
+            mf = rex.match(f.attrib['bitrate'])
+            if mf:
+                fdct['width'] = int(mf.group('width'))
+                fdct['height'] = int(mf.group('height'))
+                fdct['tbr'] = int_or_none(mf.group('bitrate'))
+            else:
+                mf = rex.search(f.text)
+                if mf:
+                    fdct['width'] = int(mf.group('width'))
+                    fdct['height'] = int(mf.group('height'))
+                    fdct['tbr'] = int_or_none(mf.group('bitrate'))
+                else:
+                    mi = re.match(r'ios_(audio|[0-9]+)$', f.attrib['bitrate'])
+                    if mi:
+                        if mi.group(1) == 'audio':
+                            fdct['vcodec'] = 'none'
+                            fdct['ext'] = 'm4a'
+                        else:
+                            fdct['tbr'] = int(mi.group(1))
+
+            formats.append(fdct)
+
+        self._sort_formats(formats)
 
         thumbnails = sorted([((int(t.attrib['height']),int(t.attrib['width'])), t.text) for t in info.findall('images/image')])
         thumbs_dict = [{'resolution': res, 'url': t_url} for (res, t_url) in thumbnails]
 
-        return {'id': info.attrib['id'],
+        metas_el = info.find('metas')
+        upload_date = (
+            metas_el.attrib.get('version') if metas_el is not None else None)
+
+        duration_el = info.find('length')
+        duration = parse_duration(duration_el.text)
+
+        return {
+            'id': info.attrib['id'],
             'title': info.find('headline').text,
-                'url': video_url,
+            'formats': formats,
-                'ext': determine_ext(video_url),
             'thumbnail': thumbnails[-1][1],
             'thumbnails': thumbs_dict,
             'description': info.find('description').text,
+            'duration': duration,
+            'upload_date': upload_date,
+        }
+
+
+class CNNBlogsIE(InfoExtractor):
+    _VALID_URL = r'https?://[^\.]+\.blogs\.cnn\.com/.+'
+    _TEST = {
+        'url': 'http://reliablesources.blogs.cnn.com/2014/02/09/criminalizing-journalism/',
+        'md5': '3e56f97b0b6ffb4b79f4ea0749551084',
+        'info_dict': {
+            'id': 'bestoftv/2014/02/09/criminalizing-journalism.cnn',
+            'ext': 'mp4',
+            'title': 'Criminalizing journalism?',
+            'description': 'Glenn Greenwald responds to comments made this week on Capitol Hill that journalists could be criminal accessories.',
+            'upload_date': '20140209',
+        },
+        'add_ie': ['CNN'],
+    }
+
+    def _real_extract(self, url):
+        webpage = self._download_webpage(url, url_basename(url))
+        cnn_url = self._html_search_regex(r'data-url="(.+?)"', webpage, 'cnn url')
+        return {
+            '_type': 'url',
+            'url': cnn_url,
+            'ie_key': CNNIE.ie_key(),
         }

youtube_dl/extractor/collegehumor.py

@@ -1,82 +1,99 @@
+from __future__ import unicode_literals
+
+import json
 import re
 
 from .common import InfoExtractor
-from ..utils import (
+from ..utils import int_or_none
-    compat_urllib_parse_urlparse,
-    determine_ext,
-
-    ExtractorError,
-)
 
 
 class CollegeHumorIE(InfoExtractor):
     _VALID_URL = r'^(?:https?://)?(?:www\.)?collegehumor\.com/(video|embed|e)/(?P<videoid>[0-9]+)/?(?P<shorttitle>.*)$'
 
     _TESTS = [{
-        u'url': u'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
+        'url': 'http://www.collegehumor.com/video/6902724/comic-con-cosplay-catastrophe',
-        u'file': u'6902724.mp4',
+        'md5': 'dcc0f5c1c8be98dc33889a191f4c26bd',
-        u'md5': u'1264c12ad95dca142a9f0bf7968105a0',
+        'info_dict': {
-        u'info_dict': {
+            'id': '6902724',
-            u'title': u'Comic-Con Cosplay Catastrophe',
+            'ext': 'mp4',
-            u'description': u'Fans get creative this year at San Diego.  Too creative.  And yes, that\'s really Joss Whedon.',
+            'title': 'Comic-Con Cosplay Catastrophe',
+            'description': 'Fans get creative this year',
+            'age_limit': 13,
         },
     },
     {
-        u'url': u'http://www.collegehumor.com/video/3505939/font-conference',
+        'url': 'http://www.collegehumor.com/video/3505939/font-conference',
-        u'file': u'3505939.mp4',
+        'md5': '72fa701d8ef38664a4dbb9e2ab721816',
-        u'md5': u'c51ca16b82bb456a4397987791a835f5',
+        'info_dict': {
-        u'info_dict': {
+            'id': '3505939',
-            u'title': u'Font Conference',
+            'ext': 'mp4',
-            u'description': u'This video wasn\'t long enough, so we made it double-spaced.',
+            'title': 'Font Conference',
+            'description': 'This video wasn\'t long enough,',
+            'age_limit': 10,
+            'duration': 179,
         },
-    }]
+    },
+    # embedded youtube video
+    {
+        'url': 'http://www.collegehumor.com/embed/6950457',
+        'info_dict': {
+            'id': 'W5gMp3ZjYg4',
+            'ext': 'mp4',
+            'title': 'Funny Dogs Protecting Babies Compilation 2014 [NEW HD]',
+            'uploader': 'Funnyplox TV',
+            'uploader_id': 'funnyploxtv',
+            'description': 'md5:7ded37421526d54afdf005e25bc2b7a3',
+            'upload_date': '20140128',
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': ['Youtube'],
+    },
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
         video_id = mobj.group('videoid')
 
-        info = {
+        jsonUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id + '.json'
-            'id': video_id,
+        data = json.loads(self._download_webpage(
-            'uploader': None,
+            jsonUrl, video_id, 'Downloading info JSON'))
-            'upload_date': None,
+        vdata = data['video']
+        if vdata.get('youtubeId') is not None:
+            return {
+                '_type': 'url',
+                'url': vdata['youtubeId'],
+                'ie_key': 'Youtube',
             }
 
-        self.report_extraction(video_id)
+        AGE_LIMITS = {'nc17': 18, 'r': 18, 'pg13': 13, 'pg': 10, 'g': 0}
-        xmlUrl = 'http://www.collegehumor.com/moogaloop/video/' + video_id
+        rating = vdata.get('rating')
-        mdoc = self._download_xml(xmlUrl, video_id,
+        if rating:
-                                         u'Downloading info XML',
+            age_limit = AGE_LIMITS.get(rating.lower())
-                                         u'Unable to download video info XML')
-
-        try:
-            videoNode = mdoc.findall('./video')[0]
-            youtubeIdNode = videoNode.find('./youtubeID')
-            if youtubeIdNode is not None:
-                return self.url_result(youtubeIdNode.text, 'Youtube')
-            info['description'] = videoNode.findall('./description')[0].text
-            info['title'] = videoNode.findall('./caption')[0].text
-            info['thumbnail'] = videoNode.findall('./thumbnail')[0].text
-            next_url = videoNode.findall('./file')[0].text
-        except IndexError:
-            raise ExtractorError(u'Invalid metadata XML file')
-
-        if next_url.endswith(u'manifest.f4m'):
-            manifest_url = next_url + '?hdcore=2.10.3'
-            adoc = self._download_xml(manifest_url, video_id,
-                                         u'Downloading XML manifest',
-                                         u'Unable to download video info XML')
-
-            try:
-                video_id = adoc.findall('./{http://ns.adobe.com/f4m/1.0}id')[0].text
-            except IndexError:
-                raise ExtractorError(u'Invalid manifest file')
-            url_pr = compat_urllib_parse_urlparse(info['thumbnail'])
-            info['url'] = url_pr.scheme + '://' + url_pr.netloc + video_id[:-2].replace('.csmil','').replace(',','')
-            info['ext'] = 'mp4'
         else:
-            # Old-style direct links
+            age_limit = None  # None = No idea
-            info['url'] = next_url
-            info['ext'] = determine_ext(info['url'])
 
-        return info
+        PREFS = {'high_quality': 2, 'low_quality': 0}
+        formats = []
+        for format_key in ('mp4', 'webm'):
+            for qname, qurl in vdata.get(format_key, {}).items():
+                formats.append({
+                    'format_id': format_key + '_' + qname,
+                    'url': qurl,
+                    'format': format_key,
+                    'preference': PREFS.get(qname),
+                })
+        self._sort_formats(formats)
+
+        duration = int_or_none(vdata.get('duration'), 1000)
+
+        return {
+            'id': video_id,
+            'title': vdata['title'],
+            'description': vdata.get('description'),
+            'thumbnail': vdata.get('thumbnail'),
+            'formats': formats,
+            'age_limit': age_limit,
+            'duration': duration,
+        }

youtube_dl/extractor/comedycentral.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -12,31 +14,25 @@ from ..utils import (
 
 
 class ComedyCentralIE(MTVServicesInfoExtractor):
-    _VALID_URL = r'https?://(?:www.)?comedycentral.com/(video-clips|episodes|cc-studios)/(?P<title>.*)'
+    _VALID_URL = r'''(?x)https?://(?:www\.)?comedycentral\.com/
-    _FEED_URL = u'http://comedycentral.com/feeds/mrss/'
+        (video-clips|episodes|cc-studios|video-collections)
+        /(?P<title>.*)'''
+    _FEED_URL = 'http://comedycentral.com/feeds/mrss/'
 
     _TEST = {
-        u'url': u'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
+        'url': 'http://www.comedycentral.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother',
-        u'md5': u'4167875aae411f903b751a21f357f1ee',
+        'md5': '4167875aae411f903b751a21f357f1ee',
-        u'info_dict': {
+        'info_dict': {
-            u'id': u'cef0cbb3-e776-4bc9-b62e-8016deccb354',
+            'id': 'cef0cbb3-e776-4bc9-b62e-8016deccb354',
-            u'ext': u'mp4',
+            'ext': 'mp4',
-            u'title': u'Uncensored - Greg Fitzsimmons - Too Good of a Mother',
+            'title': 'CC:Stand-Up|Greg Fitzsimmons: Life on Stage|Uncensored - Too Good of a Mother',
-            u'description': u'After a certain point, breastfeeding becomes c**kblocking.',
+            'description': 'After a certain point, breastfeeding becomes c**kblocking.',
         },
     }
 
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        title = mobj.group('title')
-        webpage = self._download_webpage(url, title)
-        mgid = self._search_regex(r'data-mgid="(?P<mgid>mgid:.*?)"',
-                                  webpage, u'mgid')
-        return self._get_videos_info(mgid)
-
 
 class ComedyCentralShowsIE(InfoExtractor):
-    IE_DESC = u'The Daily Show / Colbert Report'
+    IE_DESC = 'The Daily Show / Colbert Report'
     # urls can be abbreviations like :thedailyshow or :colbert
     # urls for episodes like:
     # or urls for clips like: http://www.thedailyshow.com/watch/mon-december-10-2012/any-given-gun-day
@@ -53,14 +49,14 @@ class ComedyCentralShowsIE(InfoExtractor):
                               extended-interviews/(?P<interID>[0-9]+)/playlist_tds_extended_(?P<interview_title>.*?)/.*?)))
                      $"""
     _TEST = {
-        u'url': u'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
+        'url': 'http://www.thedailyshow.com/watch/thu-december-13-2012/kristen-stewart',
-        u'file': u'422212.mp4',
+        'file': '422212.mp4',
-        u'md5': u'4e2f5cb088a83cd8cdb7756132f9739d',
+        'md5': '4e2f5cb088a83cd8cdb7756132f9739d',
-        u'info_dict': {
+        'info_dict': {
-            u"upload_date": u"20121214", 
+            "upload_date": "20121214",
-            u"description": u"Kristen Stewart", 
+            "description": "Kristen Stewart",
-            u"uploader": u"thedailyshow", 
+            "uploader": "thedailyshow",
-            u"title": u"thedailyshow-kristen-stewart part 1"
+            "title": "thedailyshow-kristen-stewart part 1"
         }
     }
 
@@ -90,22 +86,22 @@ class ComedyCentralShowsIE(InfoExtractor):
 
     @staticmethod
     def _transform_rtmp_url(rtmp_video_url):
-        m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp.comedystor/.*)$', rtmp_video_url)
+        m = re.match(r'^rtmpe?://.*?/(?P<finalid>gsp\.comedystor/.*)$', rtmp_video_url)
         if not m:
-            raise ExtractorError(u'Cannot transform RTMP url')
+            raise ExtractorError('Cannot transform RTMP url')
         base = 'http://mtvnmobile.vo.llnwd.net/kip0/_pxn=1+_pxI0=Ripod-h264+_pxL0=undefined+_pxM0=+_pxK=18639+_pxE=mp4/44620/mtvnorigin/'
         return base + m.group('finalid')
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url, re.VERBOSE)
         if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
+            raise ExtractorError('Invalid URL: %s' % url)
 
         if mobj.group('shortname'):
             if mobj.group('shortname') in ('tds', 'thedailyshow'):
-                url = u'http://www.thedailyshow.com/full-episodes/'
+                url = 'http://www.thedailyshow.com/full-episodes/'
             else:
-                url = u'http://www.colbertnation.com/full-episodes/'
+                url = 'http://www.colbertnation.com/full-episodes/'
             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
             assert mobj is not None
 
@@ -131,9 +127,9 @@ class ComedyCentralShowsIE(InfoExtractor):
             url = htmlHandle.geturl()
             mobj = re.match(self._VALID_URL, url, re.VERBOSE)
             if mobj is None:
-                raise ExtractorError(u'Invalid redirected URL: ' + url)
+                raise ExtractorError('Invalid redirected URL: ' + url)
             if mobj.group('episode') == '':
-                raise ExtractorError(u'Redirected URL is still not specific: ' + url)
+                raise ExtractorError('Redirected URL is still not specific: ' + url)
             epTitle = mobj.group('episode')
 
         mMovieParams = re.findall('(?:<param name="movie" value="|var url = ")(http://media.mtvnservices.com/([^"]*(?:episode|video).*?:.*?))"', webpage)
@@ -145,15 +141,15 @@ class ComedyCentralShowsIE(InfoExtractor):
 
             altMovieParams = re.findall('data-mgid="([^"]*(?:episode|video).*?:.*?)"', webpage)
             if len(altMovieParams) == 0:
-                raise ExtractorError(u'unable to find Flash URL in webpage ' + url)
+                raise ExtractorError('unable to find Flash URL in webpage ' + url)
             else:
                 mMovieParams = [("http://media.mtvnservices.com/" + altMovieParams[0], altMovieParams[0])]
 
         uri = mMovieParams[0][1]
         indexUrl = 'http://shadow.comedycentral.com/feeds/video_player/mrss/?' + compat_urllib_parse.urlencode({'uri': uri})
         idoc = self._download_xml(indexUrl, epTitle,
-                                          u'Downloading show index',
+                                          'Downloading show index',
-                                          u'unable to download episode index')
+                                          'unable to download episode index')
 
         results = []
 
@@ -168,7 +164,7 @@ class ComedyCentralShowsIE(InfoExtractor):
             configUrl = ('http://www.comedycentral.com/global/feeds/entertainment/media/mediaGenEntertainment.jhtml?' +
                         compat_urllib_parse.urlencode({'uri': mediaId}))
             cdoc = self._download_xml(configUrl, epTitle,
-                                               u'Downloading configuration for %s' % shortMediaId)
+                                               'Downloading configuration for %s' % shortMediaId)
 
             turls = []
             for rendition in cdoc.findall('.//rendition'):
@@ -176,7 +172,7 @@ class ComedyCentralShowsIE(InfoExtractor):
                 turls.append(finfo)
 
             if len(turls) == 0:
-                self._downloader.report_error(u'unable to download ' + mediaId + ': No videos found')
+                self._downloader.report_error('unable to download ' + mediaId + ': No videos found')
                 continue
 
             formats = []
@@ -190,7 +186,7 @@ class ComedyCentralShowsIE(InfoExtractor):
                     'width': w,
                 })
 
-            effTitle = showId + u'-' + epTitle + u' part ' + compat_str(partNum+1)
+            effTitle = showId + '-' + epTitle + ' part ' + compat_str(partNum+1)
             results.append({
                 'id': shortMediaId,
                 'formats': formats,

youtube_dl/extractor/common.py

@@ -1,4 +1,6 @@
 import base64
+import hashlib
+import json
 import os
 import re
 import socket
@@ -9,6 +11,7 @@ import xml.etree.ElementTree
 from ..utils import (
     compat_http_client,
     compat_urllib_error,
+    compat_urllib_parse_urlparse,
     compat_str,
 
     clean_html,
@@ -18,6 +21,7 @@ from ..utils import (
     sanitize_filename,
     unescapeHTML,
 )
+_NO_DEFAULT = object()
 
 
 class InfoExtractor(object):
@@ -36,10 +40,12 @@ class InfoExtractor(object):
     id:             Video identifier.
     title:          Video title, unescaped.
 
-    Additionally, it must contain either a formats entry or url and ext:
+    Additionally, it must contain either a formats entry or a url one:
 
-    formats:        A list of dictionaries for each format available, it must
+    formats:        A list of dictionaries for each format available, ordered
-                    be ordered from worst to best quality. Potential fields:
+                    from worst to best quality.
+
+                    Potential fields:
                     * url        Mandatory. The URL of the video file
                     * ext        Will be calculated from url if missing
                     * format     A human-readable description of the format
@@ -47,23 +53,38 @@ class InfoExtractor(object):
                                  Calculated from the format_id, width, height.
                                  and format_note fields if missing.
                     * format_id  A short description of the format
-                                 ("mp4_h264_opus" or "19")
+                                 ("mp4_h264_opus" or "19").
+                                Technically optional, but strongly recommended.
                     * format_note Additional info about the format
                                  ("3D" or "DASH video")
                     * width      Width of the video, if known
                     * height     Height of the video, if known
+                    * resolution Textual description of width and height
+                    * tbr        Average bitrate of audio and video in KBit/s
                     * abr        Average audio bitrate in KBit/s
                     * acodec     Name of the audio codec in use
+                    * asr        Audio sampling rate in Hertz
                     * vbr        Average video bitrate in KBit/s
                     * vcodec     Name of the video codec in use
+                    * container  Name of the container format
                     * filesize   The number of bytes, if known in advance
                     * player_url SWF Player URL (used for rtmpdump).
+                    * protocol   The protocol that will be used for the actual
+                                 download, lower-case.
+                                 "http", "https", "rtsp", "rtmp", "m3u8" or so.
+                    * preference Order number of this format. If this field is
+                                 present and not None, the formats get sorted
+                                 by this field.
+                                 -1 for default (order by other properties),
+                                 -2 or smaller for less than default.
+                    * quality    Order number of the video quality of this
+                                 format, irrespective of the file format.
+                                 -1 for default (order by other properties),
+                                 -2 or smaller for less than default.
     url:            Final video URL.
     ext:            Video filename extension.
     format:         The video format, defaults to ext (used for --get-format)
     player_url:     SWF Player URL (used for rtmpdump).
-    urlhandle:      [internal] The urlHandle to be used to download the file,
-                    like returned by urllib.request.urlopen
 
     The following fields are optional:
 
@@ -169,6 +190,8 @@ class InfoExtractor(object):
         try:
             return self._downloader.urlopen(url_or_request)
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+            if errnote is False:
+                return False
             if errnote is None:
                 errnote = u'Unable to download webpage'
             errmsg = u'%s: %s' % (errnote, compat_str(err))
@@ -199,6 +222,8 @@ class InfoExtractor(object):
                           webpage_bytes[:1024])
             if m:
                 encoding = m.group(1).decode('ascii')
+            elif webpage_bytes.startswith(b'\xff\xfe'):
+                encoding = 'utf-16'
             else:
                 encoding = 'utf-8'
         if self._downloader.params.get('dump_intermediate_pages', False):
@@ -214,6 +239,9 @@ class InfoExtractor(object):
                 url = url_or_request.get_full_url()
             except AttributeError:
                 url = url_or_request
+            if len(url) > 200:
+                h = u'___' + hashlib.md5(url.encode('utf-8')).hexdigest()
+                url = url[:200 - len(h)] + h
             raw_filename = ('%s_%s.dump' % (video_id, url))
             filename = sanitize_filename(raw_filename, restricted=True)
             self.to_screen(u'Saving request to ' + filename)
@@ -241,6 +269,23 @@ class InfoExtractor(object):
             xml_string = transform_source(xml_string)
         return xml.etree.ElementTree.fromstring(xml_string.encode('utf-8'))
 
+    def _download_json(self, url_or_request, video_id,
+                       note=u'Downloading JSON metadata',
+                       errnote=u'Unable to download JSON metadata',
+                       transform_source=None):
+        json_string = self._download_webpage(url_or_request, video_id, note, errnote)
+        if transform_source:
+            json_string = transform_source(json_string)
+        try:
+            return json.loads(json_string)
+        except ValueError as ve:
+            raise ExtractorError('Failed to download JSON', cause=ve)
+
+    def report_warning(self, msg, video_id=None):
+        idstr = u'' if video_id is None else u'%s: ' % video_id
+        self._downloader.report_warning(
+            u'[%s] %s%s' % (self.IE_NAME, idstr, msg))
+
     def to_screen(self, msg):
         """Print msg to screen, prefixing it with '[ie_name]'"""
         self._downloader.to_screen(u'[%s] %s' % (self.IE_NAME, msg))
@@ -262,7 +307,8 @@ class InfoExtractor(object):
         self.to_screen(u'Logging in')
 
     #Methods for following #608
-    def url_result(self, url, ie=None, video_id=None):
+    @staticmethod
+    def url_result(url, ie=None, video_id=None):
         """Returns a url that points to a page that should be processed"""
         #TODO: ie should be the class used for getting the info
         video_info = {'_type': 'url',
@@ -271,7 +317,8 @@ class InfoExtractor(object):
         if video_id is not None:
             video_info['id'] = video_id
         return video_info
-    def playlist_result(self, entries, playlist_id=None, playlist_title=None):
+    @staticmethod
+    def playlist_result(entries, playlist_id=None, playlist_title=None):
         """Returns a playlist"""
         video_info = {'_type': 'playlist',
                       'entries': entries}
@@ -281,7 +328,7 @@ class InfoExtractor(object):
             video_info['title'] = playlist_title
         return video_info
 
-    def _search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
+    def _search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0):
         """
         Perform a regex search on the given string, using a single or a list of
         patterns returning the first matching group.
@@ -295,7 +342,7 @@ class InfoExtractor(object):
                 mobj = re.search(p, string, flags)
                 if mobj: break
 
-        if sys.stderr.isatty() and os.name != 'nt':
+        if os.name != 'nt' and sys.stderr.isatty():
             _name = u'\033[0;34m%s\033[0m' % name
         else:
             _name = name
@@ -303,7 +350,7 @@ class InfoExtractor(object):
         if mobj:
             # return the first matching group
             return next(g for g in mobj.groups() if g is not None)
-        elif default is not None:
+        elif default is not _NO_DEFAULT:
             return default
         elif fatal:
             raise RegexNotFoundError(u'Unable to extract %s' % _name)
@@ -312,7 +359,7 @@ class InfoExtractor(object):
                 u'please report this issue on http://yt-dl.org/bug' % _name)
             return None
 
-    def _html_search_regex(self, pattern, string, name, default=None, fatal=True, flags=0):
+    def _html_search_regex(self, pattern, string, name, default=_NO_DEFAULT, fatal=True, flags=0):
         """
         Like _search_regex, but strips HTML tags and unescapes entities.
         """
@@ -355,8 +402,8 @@ class InfoExtractor(object):
     # Helper functions for extracting OpenGraph info
     @staticmethod
     def _og_regexes(prop):
-        content_re = r'content=(?:"([^>]+?)"|\'(.+?)\')'
+        content_re = r'content=(?:"([^>]+?)"|\'([^>]+?)\')'
-        property_re = r'property=[\'"]og:%s[\'"]' % re.escape(prop)
+        property_re = r'(?:name|property)=[\'"]og:%s[\'"]' % re.escape(prop)
         template = r'<meta[^>]+?%s[^>]+?%s'
         return [
             template % (property_re, content_re),
@@ -421,6 +468,65 @@ class InfoExtractor(object):
         }
         return RATING_TABLE.get(rating.lower(), None)
 
+    def _twitter_search_player(self, html):
+        return self._html_search_meta('twitter:player', html,
+            'twitter card player')
+
+    def _sort_formats(self, formats):
+        if not formats:
+            raise ExtractorError(u'No video formats found')
+
+        def _formats_key(f):
+            # TODO remove the following workaround
+            from ..utils import determine_ext
+            if not f.get('ext') and 'url' in f:
+                f['ext'] = determine_ext(f['url'])
+
+            preference = f.get('preference')
+            if preference is None:
+                proto = f.get('protocol')
+                if proto is None:
+                    proto = compat_urllib_parse_urlparse(f.get('url', '')).scheme
+
+                preference = 0 if proto in ['http', 'https'] else -0.1
+                if f.get('ext') in ['f4f', 'f4m']:  # Not yet supported
+                    preference -= 0.5
+
+            if f.get('vcodec') == 'none':  # audio only
+                if self._downloader.params.get('prefer_free_formats'):
+                    ORDER = [u'aac', u'mp3', u'm4a', u'webm', u'ogg', u'opus']
+                else:
+                    ORDER = [u'webm', u'opus', u'ogg', u'mp3', u'aac', u'm4a']
+                ext_preference = 0
+                try:
+                    audio_ext_preference = ORDER.index(f['ext'])
+                except ValueError:
+                    audio_ext_preference = -1
+            else:
+                if self._downloader.params.get('prefer_free_formats'):
+                    ORDER = [u'flv', u'mp4', u'webm']
+                else:
+                    ORDER = [u'webm', u'flv', u'mp4']
+                try:
+                    ext_preference = ORDER.index(f['ext'])
+                except ValueError:
+                    ext_preference = -1
+                audio_ext_preference = 0
+
+            return (
+                preference,
+                f.get('quality') if f.get('quality') is not None else -1,
+                f.get('height') if f.get('height') is not None else -1,
+                f.get('width') if f.get('width') is not None else -1,
+                ext_preference,
+                f.get('tbr') if f.get('tbr') is not None else -1,
+                f.get('vbr') if f.get('vbr') is not None else -1,
+                f.get('abr') if f.get('abr') is not None else -1,
+                audio_ext_preference,
+                f.get('filesize') if f.get('filesize') is not None else -1,
+                f.get('format_id'),
+            )
+        formats.sort(key=_formats_key)
 
 
 class SearchInfoExtractor(InfoExtractor):

youtube_dl/extractor/condenast.py

@@ -1,4 +1,5 @@
 # coding: utf-8
+from __future__ import unicode_literals
 
 import re
 import json
@@ -20,30 +21,31 @@ class CondeNastIE(InfoExtractor):
 
     # The keys are the supported sites and the values are the name to be shown
     # to the user and in the extractor description.
-    _SITES = {'wired': u'WIRED',
+    _SITES = {
-              'gq': u'GQ',
+        'wired': 'WIRED',
-              'vogue': u'Vogue',
+        'gq': 'GQ',
-              'glamour': u'Glamour',
+        'vogue': 'Vogue',
-              'wmagazine': u'W Magazine',
+        'glamour': 'Glamour',
-              'vanityfair': u'Vanity Fair',
+        'wmagazine': 'W Magazine',
+        'vanityfair': 'Vanity Fair',
     }
 
-    _VALID_URL = r'http://(video|www).(?P<site>%s).com/(?P<type>watch|series|video)/(?P<id>.+)' % '|'.join(_SITES.keys())
+    _VALID_URL = r'http://(video|www)\.(?P<site>%s)\.com/(?P<type>watch|series|video)/(?P<id>.+)' % '|'.join(_SITES.keys())
-    IE_DESC = u'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
+    IE_DESC = 'Condé Nast media group: %s' % ', '.join(sorted(_SITES.values()))
 
     _TEST = {
-        u'url': u'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
+        'url': 'http://video.wired.com/watch/3d-printed-speakers-lit-with-led',
-        u'file': u'5171b343c2b4c00dd0c1ccb3.mp4',
+        'file': '5171b343c2b4c00dd0c1ccb3.mp4',
-        u'md5': u'1921f713ed48aabd715691f774c451f7',
+        'md5': '1921f713ed48aabd715691f774c451f7',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'3D Printed Speakers Lit With LED',
+            'title': '3D Printed Speakers Lit With LED',
-            u'description': u'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
+            'description': 'Check out these beautiful 3D printed LED speakers.  You can\'t actually buy them, but LumiGeek is working on a board that will let you make you\'re own.',
         }
     }
 
     def _extract_series(self, url, webpage):
         title = self._html_search_regex(r'<div class="cne-series-info">.*?<h1>(.+?)</h1>',
-                                        webpage, u'series title', flags=re.DOTALL)
+                                        webpage, 'series title', flags=re.DOTALL)
         url_object = compat_urllib_parse_urlparse(url)
         base_url = '%s://%s' % (url_object.scheme, url_object.netloc)
         m_paths = re.finditer(r'<p class="cne-thumb-title">.*?<a href="(/watch/.+?)["\?]',
@@ -57,35 +59,37 @@ class CondeNastIE(InfoExtractor):
         description = self._html_search_regex([r'<div class="cne-video-description">(.+?)</div>',
                                                r'<div class="video-post-content">(.+?)</div>',
                                                ],
-                                              webpage, u'description',
+                                              webpage, 'description',
                                               fatal=False, flags=re.DOTALL)
         params = self._search_regex(r'var params = {(.+?)}[;,]', webpage,
-                                    u'player params', flags=re.DOTALL)
+                                    'player params', flags=re.DOTALL)
-        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, u'video id')
+        video_id = self._search_regex(r'videoId: [\'"](.+?)[\'"]', params, 'video id')
-        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, u'player id')
+        player_id = self._search_regex(r'playerId: [\'"](.+?)[\'"]', params, 'player id')
-        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, u'target')
+        target = self._search_regex(r'target: [\'"](.+?)[\'"]', params, 'target')
         data = compat_urllib_parse.urlencode({'videoId': video_id,
                                               'playerId': player_id,
                                               'target': target,
                                               })
         base_info_url = self._search_regex(r'url = [\'"](.+?)[\'"][,;]',
-                                           webpage, u'base info url',
+                                           webpage, 'base info url',
                                            default='http://player.cnevids.com/player/loader.js?')
         info_url = base_info_url + data
         info_page = self._download_webpage(info_url, video_id,
-                                           u'Downloading video info')
+                                           'Downloading video info')
-        video_info = self._search_regex(r'var video = ({.+?});', info_page, u'video info')
+        video_info = self._search_regex(r'var video = ({.+?});', info_page, 'video info')
         video_info = json.loads(video_info)
 
-        def _formats_sort_key(f):
+        formats = [{
-            type_ord = 1 if f['type'] == 'video/mp4' else 0
+            'format_id': '%s-%s' % (fdata['type'].split('/')[-1], fdata['quality']),
-            quality_ord = 1 if f['quality'] == 'high' else 0
+            'url': fdata['src'],
-            return (quality_ord, type_ord)
+            'ext': fdata['type'].split('/')[-1],
-        best_format = sorted(video_info['sources'][0], key=_formats_sort_key)[-1]
+            'quality': 1 if fdata['quality'] == 'high' else 0,
+        } for fdata in video_info['sources'][0]]
+        self._sort_formats(formats)
 
-        return {'id': video_id,
+        return {
-                'url': best_format['src'],
+            'id': video_id,
-                'ext': best_format['type'].split('/')[-1],
+            'formats': formats,
             'title': video_info['title'],
             'thumbnail': video_info['poster_frame'],
             'description': description,

youtube_dl/extractor/crunchyroll.py

@@ -0,0 +1,178 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re, base64, zlib
+from hashlib import sha1
+from math import pow, sqrt, floor
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    compat_urllib_parse,
+    compat_urllib_request,
+    bytes_to_intlist,
+    intlist_to_bytes,
+    unified_strdate,
+    clean_html,
+)
+from ..aes import (
+    aes_cbc_decrypt,
+    inc,
+)
+
+class CrunchyrollIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://)?(?:(?P<prefix>www|m)\.)?(?P<url>crunchyroll\.com/(?:[^/]*/[^/?&]*?|media/\?id=)(?P<video_id>[0-9]+))(?:[/?&]|$)'
+    _TESTS = [{
+        'url': 'http://www.crunchyroll.com/wanna-be-the-strongest-in-the-world/episode-1-an-idol-wrestler-is-born-645513',
+        'file': '645513.flv',
+        #'md5': 'b1639fd6ddfaa43788c85f6d1dddd412',
+        'info_dict': {
+            'title': 'Wanna be the Strongest in the World Episode 1 – An Idol-Wrestler is Born!',
+            'description': 'md5:2d17137920c64f2f49981a7797d275ef',
+            'thumbnail': 'http://img1.ak.crunchyroll.com/i/spire1-tmb/20c6b5e10f1a47b10516877d3c039cae1380951166_full.jpg',
+            'uploader': 'Yomiuri Telecasting Corporation (YTV)',
+            'upload_date': '20131013',
+        },
+        'params': {
+            # rtmp
+            'skip_download': True,
+        },
+    }]
+
+    _FORMAT_IDS = {
+        '360': ('60', '106'),
+        '480': ('61', '106'),
+        '720': ('62', '106'),
+        '1080': ('80', '108'),
+    }
+
+    def _decrypt_subtitles(self, data, iv, id):
+        data = bytes_to_intlist(data)
+        iv = bytes_to_intlist(iv)
+        id = int(id)
+
+        def obfuscate_key_aux(count, modulo, start):
+            output = list(start)
+            for _ in range(count):
+                output.append(output[-1] + output[-2])
+            # cut off start values
+            output = output[2:]
+            output = list(map(lambda x: x % modulo + 33, output))
+            return output
+
+        def obfuscate_key(key):
+            num1 = int(floor(pow(2, 25) * sqrt(6.9)))
+            num2 = (num1 ^ key) << 5
+            num3 = key ^ num1
+            num4 = num3 ^ (num3 >> 3) ^ num2
+            prefix = intlist_to_bytes(obfuscate_key_aux(20, 97, (1, 2)))
+            shaHash = bytes_to_intlist(sha1(prefix + str(num4).encode('ascii')).digest())
+            # Extend 160 Bit hash to 256 Bit
+            return shaHash + [0] * 12
+        
+        key = obfuscate_key(id)
+        class Counter:
+            __value = iv
+            def next_value(self):
+                temp = self.__value
+                self.__value = inc(self.__value)
+                return temp
+        decrypted_data = intlist_to_bytes(aes_cbc_decrypt(data, key, iv))
+        return zlib.decompress(decrypted_data)
+
+    def _convert_subtitles_to_srt(self, subtitles):
+        i=1
+        output = ''
+        for start, end, text in re.findall(r'<event [^>]*?start="([^"]+)" [^>]*?end="([^"]+)" [^>]*?text="([^"]+)"[^>]*?>', subtitles):
+            start = start.replace('.', ',')
+            end = end.replace('.', ',')
+            text = clean_html(text)
+            text = text.replace('\\N', '\n')
+            if not text:
+                continue
+            output += '%d\n%s --> %s\n%s\n\n' % (i, start, end, text)
+            i+=1
+        return output
+
+    def _real_extract(self,url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('video_id')
+
+        if mobj.group('prefix') == 'm':
+            mobile_webpage = self._download_webpage(url, video_id, 'Downloading mobile webpage')
+            webpage_url = self._search_regex(r'<link rel="canonical" href="([^"]+)" />', mobile_webpage, 'webpage_url')
+        else:
+            webpage_url = 'http://www.' + mobj.group('url')
+
+        webpage = self._download_webpage(webpage_url, video_id, 'Downloading webpage')
+        note_m = self._html_search_regex(r'<div class="showmedia-trailer-notice">(.+?)</div>', webpage, 'trailer-notice', default='')
+        if note_m:
+            raise ExtractorError(note_m)
+
+        video_title = self._html_search_regex(r'<h1[^>]*>(.+?)</h1>', webpage, 'video_title', flags=re.DOTALL)
+        video_title = re.sub(r' {2,}', ' ', video_title)
+        video_description = self._html_search_regex(r'"description":"([^"]+)', webpage, 'video_description', default='')
+        if not video_description:
+            video_description = None
+        video_upload_date = self._html_search_regex(r'<div>Availability for free users:(.+?)</div>', webpage, 'video_upload_date', fatal=False, flags=re.DOTALL)
+        if video_upload_date:
+            video_upload_date = unified_strdate(video_upload_date)
+        video_uploader = self._html_search_regex(r'<div>\s*Publisher:(.+?)</div>', webpage, 'video_uploader', fatal=False, flags=re.DOTALL)
+
+        playerdata_url = compat_urllib_parse.unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
+        playerdata_req = compat_urllib_request.Request(playerdata_url)
+        playerdata_req.data = compat_urllib_parse.urlencode({'current_page': webpage_url})
+        playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
+        
+        stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
+        video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
+
+        formats = []
+        for fmt in re.findall(r'\?p([0-9]{3,4})=1', webpage):
+            stream_quality, stream_format = self._FORMAT_IDS[fmt]
+            video_format = fmt+'p'
+            streamdata_req = compat_urllib_request.Request('http://www.crunchyroll.com/xml/')
+            # urlencode doesn't work!
+            streamdata_req.data = 'req=RpcApiVideoEncode%5FGetStreamInfo&video%5Fencode%5Fquality='+stream_quality+'&media%5Fid='+stream_id+'&video%5Fformat='+stream_format
+            streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
+            streamdata_req.add_header('Content-Length', str(len(streamdata_req.data)))
+            streamdata = self._download_webpage(streamdata_req, video_id, note='Downloading media info for '+video_format)
+            video_url = self._search_regex(r'<host>([^<]+)', streamdata, 'video_url')
+            video_play_path = self._search_regex(r'<file>([^<]+)', streamdata, 'video_play_path')
+            formats.append({
+                'url': video_url,
+                'play_path':   video_play_path,
+                'ext': 'flv',
+                'format': video_format,
+                'format_id': video_format,
+            })
+
+        subtitles = {}
+        for sub_id, sub_name in re.findall(r'\?ssid=([0-9]+)" title="([^"]+)', webpage):
+            sub_page = self._download_webpage('http://www.crunchyroll.com/xml/?req=RpcApiSubtitle_GetXml&subtitle_script_id='+sub_id,\
+                                              video_id, note='Downloading subtitles for '+sub_name)
+            id = self._search_regex(r'id=\'([0-9]+)', sub_page, 'subtitle_id', fatal=False)
+            iv = self._search_regex(r'<iv>([^<]+)', sub_page, 'subtitle_iv', fatal=False)
+            data = self._search_regex(r'<data>([^<]+)', sub_page, 'subtitle_data', fatal=False)
+            if not id or not iv or not data:
+                continue
+            id = int(id)
+            iv = base64.b64decode(iv)
+            data = base64.b64decode(data)
+
+            subtitle = self._decrypt_subtitles(data, iv, id).decode('utf-8')
+            lang_code = self._search_regex(r'lang_code=\'([^\']+)', subtitle, 'subtitle_lang_code', fatal=False)
+            if not lang_code:
+                continue
+            subtitles[lang_code] = self._convert_subtitles_to_srt(subtitle)
+
+        return {
+            'id':          video_id,
+            'title':       video_title,
+            'description': video_description,
+            'thumbnail':   video_thumbnail,
+            'uploader':    video_uploader,
+            'upload_date': video_upload_date,
+            'subtitles':   subtitles,
+            'formats':     formats,
+        }

youtube_dl/extractor/cspan.py

@@ -1,51 +1,60 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_urllib_parse,
+    unescapeHTML,
+    find_xpath_attr,
 )
 
+
 class CSpanIE(InfoExtractor):
-    _VALID_URL = r'http://www\.c-spanvideo\.org/program/(.*)'
+    _VALID_URL = r'http://(?:www\.)?c-span\.org/video/\?(?P<id>\d+)'
+    IE_DESC = 'C-SPAN'
     _TEST = {
-        u'url': u'http://www.c-spanvideo.org/program/HolderonV',
+        'url': 'http://www.c-span.org/video/?313572-1/HolderonV',
-        u'file': u'315139.flv',
+        'md5': '8e44ce11f0f725527daccc453f553eb0',
-        u'md5': u'74a623266956f69e4df0068ab6c80fe4',
+        'info_dict': {
-        u'info_dict': {
+            'id': '315139',
-            u"title": u"Attorney General Eric Holder on Voting Rights Act Decision"
+            'ext': 'mp4',
+            'title': 'Attorney General Eric Holder on Voting Rights Act Decision',
+            'description': 'Attorney General Eric Holder spoke to reporters following the Supreme Court decision in Shelby County v. Holder in which the court ruled that the preclearance provisions of the Voting Rights Act could not be enforced until Congress established new guidelines for review.',
         },
-        u'skip': u'Requires rtmpdump'
+        'skip': 'Regularly fails on travis, for unknown reasons',
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        prog_name = mobj.group(1)
+        page_id = mobj.group('id')
-        webpage = self._download_webpage(url, prog_name)
+        webpage = self._download_webpage(url, page_id)
-        video_id = self._search_regex(r'programid=(.*?)&', webpage, 'video id')
+        video_id = self._search_regex(r'data-progid=\'(\d+)\'>', webpage, 'video id')
-        data = compat_urllib_parse.urlencode({'programid': video_id,
-                                              'dynamic':'1'})
-        info_url = 'http://www.c-spanvideo.org/common/services/flashXml.php?' + data
-        video_info = self._download_webpage(info_url, video_id, u'Downloading video info')
 
-        self.report_extraction(video_id)
+        description = self._html_search_regex(
+            [
+                # The full description
+                r'<div class=\'expandable\'>(.*?)<a href=\'#\'',
+                # If the description is small enough the other div is not
+                # present, otherwise this is a stripped version
+                r'<p class=\'initial\'>(.*?)</p>'
+            ],
+            webpage, 'description', flags=re.DOTALL)
 
-        title = self._html_search_regex(r'<string name="title">(.*?)</string>',
+        info_url = 'http://c-spanvideo.org/videoLibrary/assets/player/ajax-player.php?os=android&html5=program&id=' + video_id
-                                        video_info, 'title')
+        data = self._download_json(info_url, video_id)
-        description = self._html_search_regex(r'<meta (?:property="og:|name=")description" content="(.*?)"',
-                                              webpage, 'description',
-                                              flags=re.MULTILINE|re.DOTALL)
 
-        url = self._search_regex(r'<string name="URL">(.*?)</string>',
+        url = unescapeHTML(data['video']['files'][0]['path']['#text'])
-                                 video_info, 'video url')
-        url = url.replace('$(protocol)', 'rtmp').replace('$(port)', '443')
-        path = self._search_regex(r'<string name="path">(.*?)</string>',
-                            video_info, 'rtmp play path')
 
-        return {'id': video_id,
+        doc = self._download_xml('http://www.c-span.org/common/services/flashXml.php?programid=' + video_id,
-                'title': title,
+            video_id)
-                'ext': 'flv',
+
+        def find_string(s):
+            return find_xpath_attr(doc, './/string', 'name', s).text
+
+        return {
+            'id': video_id,
+            'title': find_string('title'),
             'url': url,
-                'play_path': path,
             'description': description,
-                'thumbnail': self._og_search_thumbnail(webpage),
+            'thumbnail': find_string('poster'),
         }

youtube_dl/extractor/d8.py

@@ -1,22 +1,25 @@
 # encoding: utf-8
+from __future__ import unicode_literals
+
 from .canalplus import CanalplusIE
 
 
 class D8IE(CanalplusIE):
     _VALID_URL = r'https?://www\.d8\.tv/.*?/(?P<path>.*)'
     _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/d8/%s'
-    IE_NAME = u'd8.tv'
+    IE_NAME = 'd8.tv'
 
     _TEST = {
-        u'url': u'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html',
+        'url': 'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html',
-        u'file': u'966289.flv',
+        'file': '966289.flv',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'Campagne intime - Documentaire exceptionnel',
+            'title': 'Campagne intime - Documentaire exceptionnel',
-            u'description': u'md5:d2643b799fb190846ae09c61e59a859f',
+            'description': 'md5:d2643b799fb190846ae09c61e59a859f',
-            u'upload_date': u'20131108',
+            'upload_date': '20131108',
         },
-        u'params': {
+        'params': {
             # rtmp
-            u'skip_download': True,
+            'skip_download': True,
         },
+        'skip': 'videos get deleted after a while',
     }

youtube_dl/extractor/defense.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -5,15 +7,14 @@ from .common import InfoExtractor
 
 
 class DefenseGouvFrIE(InfoExtractor):
-    _IE_NAME = 'defense.gouv.fr'
+    IE_NAME = 'defense.gouv.fr'
     _VALID_URL = (r'http://.*?\.defense\.gouv\.fr/layout/set/'
         r'ligthboxvideo/base-de-medias/webtv/(.*)')
 
     _TEST = {
-        u'url': (u'http://www.defense.gouv.fr/layout/set/ligthboxvideo/'
+        'url': 'http://www.defense.gouv.fr/layout/set/ligthboxvideo/base-de-medias/webtv/attaque-chimique-syrienne-du-21-aout-2013-1',
-        u'base-de-medias/webtv/attaque-chimique-syrienne-du-21-aout-2013-1'),
+        'file': '11213.mp4',
-        u'file': u'11213.mp4',
+        'md5': '75bba6124da7e63d2d60b5244ec9430c',
-        u'md5': u'75bba6124da7e63d2d60b5244ec9430c',
         "info_dict": {
             "title": "attaque-chimique-syrienne-du-21-aout-2013-1"
         }

youtube_dl/extractor/discovery.py

@@ -0,0 +1,46 @@
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class DiscoveryIE(InfoExtractor):
+    _VALID_URL = r'http://dsc\.discovery\.com\/[a-zA-Z0-9\-]*/[a-zA-Z0-9\-]*/videos/(?P<id>[a-zA-Z0-9\-]*)(.htm)?'
+    _TEST = {
+        'url': 'http://dsc.discovery.com/tv-shows/mythbusters/videos/mission-impossible-outtakes.htm',
+        'file': '614784.mp4',
+        'md5': 'e12614f9ee303a6ccef415cb0793eba2',
+        'info_dict': {
+            'title': 'MythBusters: Mission Impossible Outtakes',
+            'description': ('Watch Jamie Hyneman and Adam Savage practice being'
+                ' each other -- to the point of confusing Jamie\'s dog -- and '
+                'don\'t miss Adam moon-walking as Jamie ... behind Jamie\'s'
+                ' back.'),
+            'duration': 156,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+
+        video_list_json = self._search_regex(r'var videoListJSON = ({.*?});',
+            webpage, 'video list', flags=re.DOTALL)
+        video_list = json.loads(video_list_json)
+        info = video_list['clips'][0]
+        formats = []
+        for f in info['mp4']:
+            formats.append(
+                {'url': f['src'], r'ext': r'mp4', 'tbr': int(f['bitrate'][:-1])})
+
+        return {
+            'id': info['contentId'],
+            'title': video_list['name'],
+            'formats': formats,
+            'description': info['videoCaption'],
+            'thumbnail': info.get('videoStillURL') or info.get('thumbnailURL'),
+            'duration': info['duration'],
+        }

youtube_dl/extractor/dotsub.py

@@ -1,34 +1,35 @@
+from __future__ import unicode_literals
+
 import re
-import json
 import time
 
 from .common import InfoExtractor
 
 
 class DotsubIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?dotsub\.com/view/([^/]+)'
+    _VALID_URL = r'http://(?:www\.)?dotsub\.com/view/(?P<id>[^/]+)'
     _TEST = {
-        u'url': u'http://dotsub.com/view/aed3b8b2-1889-4df5-ae63-ad85f5572f27',
+        'url': 'http://dotsub.com/view/aed3b8b2-1889-4df5-ae63-ad85f5572f27',
-        u'file': u'aed3b8b2-1889-4df5-ae63-ad85f5572f27.flv',
+        'md5': '0914d4d69605090f623b7ac329fea66e',
-        u'md5': u'0914d4d69605090f623b7ac329fea66e',
+        'info_dict': {
-        u'info_dict': {
+            'id': 'aed3b8b2-1889-4df5-ae63-ad85f5572f27',
-            u"title": u"Pyramids of Waste (2010), AKA The Lightbulb Conspiracy - Planned obsolescence documentary",
+            'ext': 'flv',
-            u"uploader": u"4v4l0n42",
+            'title': 'Pyramids of Waste (2010), AKA The Lightbulb Conspiracy - Planned obsolescence documentary',
-            u'description': u'Pyramids of Waste (2010) also known as "The lightbulb conspiracy" is a documentary about how our economic system based on consumerism  and planned obsolescence is breaking our planet down.\r\n\r\nSolutions to this can be found at:\r\nhttp://robotswillstealyourjob.com\r\nhttp://www.federicopistono.org\r\n\r\nhttp://opensourceecology.org\r\nhttp://thezeitgeistmovement.com',
+            'uploader': '4v4l0n42',
-            u'thumbnail': u'http://dotsub.com/media/aed3b8b2-1889-4df5-ae63-ad85f5572f27/p',
+            'description': 'Pyramids of Waste (2010) also known as "The lightbulb conspiracy" is a documentary about how our economic system based on consumerism  and planned obsolescence is breaking our planet down.\r\n\r\nSolutions to this can be found at:\r\nhttp://robotswillstealyourjob.com\r\nhttp://www.federicopistono.org\r\n\r\nhttp://opensourceecology.org\r\nhttp://thezeitgeistmovement.com',
-            u'upload_date': u'20101213',
+            'thumbnail': 'http://dotsub.com/media/aed3b8b2-1889-4df5-ae63-ad85f5572f27/p',
+            'upload_date': '20101213',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
+        video_id = mobj.group('id')
-        info_url = "https://dotsub.com/api/media/%s/metadata" %(video_id)
+        info_url = "https://dotsub.com/api/media/%s/metadata" % video_id
-        webpage = self._download_webpage(info_url, video_id)
+        info = self._download_json(info_url, video_id)
-        info = json.loads(webpage)
         date = time.gmtime(info['dateCreated']/1000) # The timestamp is in miliseconds
 
-        return [{
+        return {
             'id': video_id,
             'url': info['mediaURI'],
             'ext': 'flv',
@@ -37,5 +38,5 @@ class DotsubIE(InfoExtractor):
             'description': info['description'],
             'uploader': info['user'],
             'view_count': info['numberOfViews'],
-            'upload_date': u'%04i%02i%02i' % (date.tm_year, date.tm_mon, date.tm_mday),
+            'upload_date': '%04i%02i%02i' % (date.tm_year, date.tm_mon, date.tm_mday),
-        }]
+        }

youtube_dl/extractor/dreisat.py

@@ -4,18 +4,17 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
     unified_strdate,
 )
 
 
 class DreiSatIE(InfoExtractor):
     IE_NAME = '3sat'
-    _VALID_URL = r'(?:http://)?(?:www\.)?3sat\.de/mediathek/index\.php\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
+    _VALID_URL = r'(?:http://)?(?:www\.)?3sat\.de/mediathek/(?:index\.php)?\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)$'
     _TEST = {
         u"url": u"http://www.3sat.de/mediathek/index.php?obj=36983",
-        u'file': u'36983.webm',
+        u'file': u'36983.mp4',
-        u'md5': u'57c97d0469d71cf874f6815aa2b7c944',
+        u'md5': u'9dcfe344732808dbfcc901537973c922',
         u'info_dict': {
             u"title": u"Kaffeeland Schweiz",
             u"description": u"Über 80 Kaffeeröstereien liefern in der Schweiz das Getränk, in das das Land so vernarrt ist: Mehr als 1000 Tassen trinkt ein Schweizer pro Jahr. SCHWEIZWEIT nimmt die Kaffeekultur unter die...", 
@@ -52,18 +51,12 @@ class DreiSatIE(InfoExtractor):
             'width': int(fe.find('./width').text),
             'height': int(fe.find('./height').text),
             'url': fe.find('./url').text,
-            'ext': determine_ext(fe.find('./url').text),
             'filesize': int(fe.find('./filesize').text),
             'video_bitrate': int(fe.find('./videoBitrate').text),
-            '3sat_qualityname': fe.find('./quality').text,
         } for fe in format_els
             if not fe.find('./url').text.startswith('http://www.metafilegenerator.de/')]
 
-        def _sortkey(format):
+        self._sort_formats(formats)
-            qidx = ['low', 'med', 'high', 'veryhigh'].index(format['3sat_qualityname'])
-            prefer_http = 1 if 'rtmp' in format['url'] else 0
-            return (qidx, prefer_http, format['video_bitrate'])
-        formats.sort(key=_sortkey)
 
         return {
             '_type': 'video',

youtube_dl/extractor/dropbox.py

@@ -0,0 +1,32 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import os.path
+import re
+
+from .common import InfoExtractor
+
+
+class DropboxIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?dropbox[.]com/s/(?P<id>[a-zA-Z0-9]{15})/(?P<title>[^?#]*)'
+    _TEST = {
+        'url': 'https://www.dropbox.com/s/0qr9sai2veej4f8/THE_DOCTOR_GAMES.mp4',
+        'md5': '8ae17c51172fb7f93bdd6a214cc8c896',
+        'info_dict': {
+            'id': '0qr9sai2veej4f8',
+            'ext': 'mp4',
+            'title': 'THE_DOCTOR_GAMES'
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        title = os.path.splitext(mobj.group('title'))[0]
+        video_url = url + '?dl=1'
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': video_url,
+        }

youtube_dl/extractor/elpais.py

@@ -0,0 +1,58 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import unified_strdate
+
+
+class ElPaisIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:[^.]+\.)?elpais\.com/.*/(?P<id>[^/#?]+)\.html(?:$|[?#])'
+    IE_DESC = 'El País'
+
+    _TEST = {
+        'url': 'http://blogs.elpais.com/la-voz-de-inaki/2014/02/tiempo-nuevo-recetas-viejas.html',
+        'md5': '98406f301f19562170ec071b83433d55',
+        'info_dict': {
+            'id': 'tiempo-nuevo-recetas-viejas',
+            'ext': 'mp4',
+            'title': 'Tiempo nuevo, recetas viejas',
+            'description': 'De lunes a viernes, a partir de las ocho de la mañana, Iñaki Gabilondo nos cuenta su visión de la actualidad nacional e internacional.',
+            'upload_date': '20140206',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        prefix = self._html_search_regex(
+            r'var url_cache = "([^"]+)";', webpage, 'URL prefix')
+        video_suffix = self._search_regex(
+            r"URLMediaFile = url_cache \+ '([^']+)'", webpage, 'video URL')
+        video_url = prefix + video_suffix
+        thumbnail_suffix = self._search_regex(
+            r"URLMediaStill = url_cache \+ '([^']+)'", webpage, 'thumbnail URL',
+            fatal=False)
+        thumbnail = (
+            None if thumbnail_suffix is None
+            else prefix + thumbnail_suffix)
+        title = self._html_search_regex(
+            '<h2 class="entry-header entry-title.*?>(.*?)</h2>',
+            webpage, 'title')
+        date_str = self._search_regex(
+            r'<p class="date-header date-int updated"\s+title="([^"]+)">',
+            webpage, 'upload date', fatal=False)
+        upload_date = (None if date_str is None else unified_strdate(date_str))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'description': self._og_search_description(webpage),
+            'thumbnail': thumbnail,
+            'upload_date': upload_date,
+        }

youtube_dl/extractor/escapist.py

@@ -1,9 +1,9 @@
-import json
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_str,
     compat_urllib_parse,
 
     ExtractorError,
@@ -11,70 +11,68 @@ from ..utils import (
 
 
 class EscapistIE(InfoExtractor):
-    _VALID_URL = r'^https?://?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<episode>[^/?]+)[/?]?.*$'
+    _VALID_URL = r'^https?://?(www\.)?escapistmagazine\.com/videos/view/(?P<showname>[^/]+)/(?P<id>[0-9]+)-'
     _TEST = {
-        u'url': u'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate',
+        'url': 'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate',
-        u'file': u'6618-Breaking-Down-Baldurs-Gate.mp4',
+        'md5': 'ab3a706c681efca53f0a35f1415cf0d1',
-        u'md5': u'ab3a706c681efca53f0a35f1415cf0d1',
+        'info_dict': {
-        u'info_dict': {
+            'id': '6618',
-            u"description": u"Baldur's Gate: Original, Modded or Enhanced Edition? I'll break down what you can expect from the new Baldur's Gate: Enhanced Edition.", 
+            'ext': 'mp4',
-            u"uploader": u"the-escapist-presents", 
+            'description': "Baldur's Gate: Original, Modded or Enhanced Edition? I'll break down what you can expect from the new Baldur's Gate: Enhanced Edition.",
-            u"title": u"Breaking Down Baldur's Gate"
+            'uploader': 'the-escapist-presents',
+            'title': "Breaking Down Baldur's Gate",
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         showName = mobj.group('showname')
-        videoId = mobj.group('episode')
+        video_id = mobj.group('id')
 
-        self.report_extraction(videoId)
+        self.report_extraction(video_id)
-        webpage = self._download_webpage(url, videoId)
+        webpage = self._download_webpage(url, video_id)
 
         videoDesc = self._html_search_regex(
             r'<meta name="description" content="([^"]*)"',
-            webpage, u'description', fatal=False)
+            webpage, 'description', fatal=False)
 
         playerUrl = self._og_search_video_url(webpage, name=u'player URL')
 
         title = self._html_search_regex(
             r'<meta name="title" content="([^"]*)"',
-            webpage, u'title').split(' : ')[-1]
+            webpage, 'title').split(' : ')[-1]
 
-        configUrl = self._search_regex('config=(.*)$', playerUrl, u'config URL')
+        configUrl = self._search_regex('config=(.*)$', playerUrl, 'config URL')
         configUrl = compat_urllib_parse.unquote(configUrl)
 
         formats = []
 
-        def _add_format(name, cfgurl):
+        def _add_format(name, cfgurl, quality):
-            configJSON = self._download_webpage(
+            config = self._download_json(
-                cfgurl, videoId,
+                cfgurl, video_id,
-                u'Downloading ' + name + ' configuration',
+                'Downloading ' + name + ' configuration',
-                u'Unable to download ' + name + ' configuration')
+                'Unable to download ' + name + ' configuration',
+                transform_source=lambda s: s.replace("'", '"'))
 
-            # Technically, it's JavaScript, not JSON
-            configJSON = configJSON.replace("'", '"')
-
-            try:
-                config = json.loads(configJSON)
-            except (ValueError,) as err:
-                raise ExtractorError(u'Invalid JSON in configuration file: ' + compat_str(err))
             playlist = config['playlist']
             formats.append({
                 'url': playlist[1]['url'],
                 'format_id': name,
+                'quality': quality,
             })
 
-        _add_format(u'normal', configUrl)
+        _add_format('normal', configUrl, quality=0)
         hq_url = (configUrl +
                   ('&hq=1' if '?' in configUrl else configUrl + '?hq=1'))
         try:
-            _add_format(u'hq', hq_url)
+            _add_format('hq', hq_url, quality=1)
         except ExtractorError:
             pass  # That's fine, we'll just use normal quality
 
+        self._sort_formats(formats)
+
         return {
-            'id': videoId,
+            'id': video_id,
             'formats': formats,
             'uploader': showName,
             'title': title,

youtube_dl/extractor/everyonesmixtape.py

@@ -0,0 +1,69 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+    ExtractorError,
+)
+
+
+class EveryonesMixtapeIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?everyonesmixtape\.com/#/mix/(?P<id>[0-9a-zA-Z]+)(?:/(?P<songnr>[0-9]))?$'
+
+    _TEST = {
+        'url': 'http://everyonesmixtape.com/#/mix/m7m0jJAbMQi/5',
+        'file': '5bfseWNmlds.mp4',
+        "info_dict": {
+            "title": "Passion Pit - \"Sleepyhead\" (Official Music Video)",
+            "uploader": "FKR.TV",
+            "uploader_id": "frenchkissrecords",
+            "description": "Music video for \"Sleepyhead\" from Passion Pit's debut EP Chunk Of Change.\nBuy on iTunes: https://itunes.apple.com/us/album/chunk-of-change-ep/id300087641\n\nDirected by The Wilderness.\n\nhttp://www.passionpitmusic.com\nhttp://www.frenchkissrecords.com",
+            "upload_date": "20081015"
+        },
+        'params': {
+            'skip_download': True,  # This is simply YouTube
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('id')
+
+        pllist_url = 'http://everyonesmixtape.com/mixtape.php?a=getMixes&u=-1&linked=%s&explore=' % playlist_id
+        pllist_req = compat_urllib_request.Request(pllist_url)
+        pllist_req.add_header('X-Requested-With', 'XMLHttpRequest')
+
+        playlist_list = self._download_json(
+            pllist_req, playlist_id, note='Downloading playlist metadata')
+        try:
+            playlist_no = next(playlist['id']
+                               for playlist in playlist_list
+                               if playlist['code'] == playlist_id)
+        except StopIteration:
+            raise ExtractorError('Playlist id not found')
+
+        pl_url = 'http://everyonesmixtape.com/mixtape.php?a=getMix&id=%s&userId=null&code=' % playlist_no
+        pl_req = compat_urllib_request.Request(pl_url)
+        pl_req.add_header('X-Requested-With', 'XMLHttpRequest')
+        playlist = self._download_json(
+            pl_req, playlist_id, note='Downloading playlist info')
+
+        entries = [{
+            '_type': 'url',
+            'url': t['url'],
+            'title': t['title'],
+        } for t in playlist['tracks']]
+
+        if mobj.group('songnr'):
+            songnr = int(mobj.group('songnr')) - 1
+            return entries[songnr]
+
+        playlist_title = playlist['mixData']['name']
+        return {
+            '_type': 'playlist',
+            'id': playlist_id,
+            'title': playlist_title,
+            'entries': entries,
+        }

youtube_dl/extractor/exfm.py

@@ -1,56 +1,58 @@
+from __future__ import unicode_literals
+
 import re
-import json
 
 from .common import InfoExtractor
 
 
 class ExfmIE(InfoExtractor):
-    IE_NAME = u'exfm'
+    IE_NAME = 'exfm'
-    IE_DESC = u'ex.fm'
+    IE_DESC = 'ex.fm'
-    _VALID_URL = r'(?:http://)?(?:www\.)?ex\.fm/song/([^/]+)'
+    _VALID_URL = r'http://(?:www\.)?ex\.fm/song/(?P<id>[^/]+)'
-    _SOUNDCLOUD_URL = r'(?:http://)?(?:www\.)?api\.soundcloud\.com/tracks/([^/]+)/stream'
+    _SOUNDCLOUD_URL = r'http://(?:www\.)?api\.soundcloud\.com/tracks/([^/]+)/stream'
     _TESTS = [
         {
-            u'url': u'http://ex.fm/song/eh359',
+            'url': 'http://ex.fm/song/eh359',
-            u'file': u'44216187.mp3',
+            'md5': 'e45513df5631e6d760970b14cc0c11e7',
-            u'md5': u'e45513df5631e6d760970b14cc0c11e7',
+            'info_dict': {
-            u'info_dict': {
+                'id': '44216187',
-                u"title": u"Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive",
+                'ext': 'mp3',
-                u"uploader": u"deadjournalist",
+                'title': 'Test House "Love Is Not Enough" (Extended Mix) DeadJournalist Exclusive',
-                u'upload_date': u'20120424',
+                'uploader': 'deadjournalist',
-                u'description': u'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive',
+                'upload_date': '20120424',
+                'description': 'Test House \"Love Is Not Enough\" (Extended Mix) DeadJournalist Exclusive',
             },
-            u'note': u'Soundcloud song',
+            'note': 'Soundcloud song',
-            u'skip': u'The site is down too often',
+            'skip': 'The site is down too often',
         },
         {
-            u'url': u'http://ex.fm/song/wddt8',
+            'url': 'http://ex.fm/song/wddt8',
-            u'file': u'wddt8.mp3',
+            'md5': '966bd70741ac5b8570d8e45bfaed3643',
-            u'md5': u'966bd70741ac5b8570d8e45bfaed3643',
+            'info_dict': {
-            u'info_dict': {
+                'id': 'wddt8',
-                u'title': u'Safe and Sound',
+                'ext': 'mp3',
-                u'uploader': u'Capital Cities',
+                'title': 'Safe and Sound',
+                'uploader': 'Capital Cities',
             },
-            u'skip': u'The site is down too often',
+            'skip': 'The site is down too often',
         },
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        song_id = mobj.group(1)
+        song_id = mobj.group('id')
-        info_url = "http://ex.fm/api/v3/song/%s" %(song_id)
+        info_url = "http://ex.fm/api/v3/song/%s" % song_id
-        webpage = self._download_webpage(info_url, song_id)
+        info = self._download_json(info_url, song_id)['song']
-        info = json.loads(webpage)
+        song_url = info['url']
-        song_url = info['song']['url']
         if re.match(self._SOUNDCLOUD_URL, song_url) is not None:
             self.to_screen('Soundcloud song detected')
             return self.url_result(song_url.replace('/stream', ''), 'Soundcloud')
-        return [{
+        return {
             'id': song_id,
             'url': song_url,
             'ext': 'mp3',
-            'title':       info['song']['title'],
+            'title': info['title'],
-            'thumbnail':   info['song']['image']['large'],
+            'thumbnail': info['image']['large'],
-            'uploader':    info['song']['artist'],
+            'uploader': info['artist'],
-            'view_count':  info['song']['loved_count'],
+            'view_count': info['loved_count'],
-        }]
+        }

youtube_dl/extractor/facebook.py

@@ -17,7 +17,12 @@ from ..utils import (
 class FacebookIE(InfoExtractor):
     """Information Extractor for Facebook"""
 
-    _VALID_URL = r'^(?:https?://)?(?:\w+\.)?facebook\.com/(?:[^#?]*#!/)?(?:video/video|photo)\.php\?(?:.*?)v=(?P<ID>\d+)(?:.*)'
+    _VALID_URL = r'''(?x)
+        (?:https?://)?(?:\w+\.)?facebook\.com/
+        (?:[^#?]*\#!/)?
+        (?:video/video\.php|photo\.php|video/embed)\?(?:.*?)
+        (?:v|video_id)=(?P<id>[0-9]+)
+        (?:.*)'''
     _LOGIN_URL = 'https://www.facebook.com/login.php?next=http%3A%2F%2Ffacebook.com%2Fhome.php&login_attempt=1'
     _CHECKPOINT_URL = 'https://www.facebook.com/checkpoint/?next=http%3A%2F%2Ffacebook.com%2Fhome.php&_fb_noscript=1'
     _NETRC_MACHINE = 'facebook'
@@ -90,7 +95,7 @@ class FacebookIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         if mobj is None:
             raise ExtractorError(u'Invalid URL: %s' % url)
-        video_id = mobj.group('ID')
+        video_id = mobj.group('id')
 
         url = 'https://www.facebook.com/video/video.php?v=%s' % video_id
         webpage = self._download_webpage(url, video_id)

youtube_dl/extractor/firstpost.py

@@ -0,0 +1,38 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class FirstpostIE(InfoExtractor):
+    IE_NAME = 'Firstpost.com'
+    _VALID_URL = r'http://(?:www\.)?firstpost\.com/[^/]+/.*-(?P<id>[0-9]+)\.html'
+
+    _TEST = {
+        'url': 'http://www.firstpost.com/india/india-to-launch-indigenous-aircraft-carrier-monday-1025403.html',
+        'md5': 'ee9114957692f01fb1263ed87039112a',
+        'info_dict': {
+            'id': '1025403',
+            'ext': 'mp4',
+            'title': 'India to launch indigenous aircraft carrier INS Vikrant today',
+            'description': 'Its flight deck is over twice the size of a football field, its power unit can light up the entire Kochi city and the cabling is enough to cover the distance between here to Delhi.',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        video_url = self._html_search_regex(
+            r'<div.*?name="div_video".*?flashvars="([^"]+)">',
+            webpage, 'video URL')
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': self._og_search_title(webpage),
+            'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+        }

youtube_dl/extractor/firsttv.py

@@ -0,0 +1,60 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import int_or_none
+
+
+class FirstTVIE(InfoExtractor):
+    IE_NAME = 'firsttv'
+    IE_DESC = 'Видеоархив - Первый канал'
+    _VALID_URL = r'http://(?:www\.)?1tv\.ru/videoarchive/(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'http://www.1tv.ru/videoarchive/73390',
+        'md5': '3de6390cf0cca4a5eae1d1d83895e5ad',
+        'info_dict': {
+            'id': '73390',
+            'ext': 'mp4',
+            'title': 'Олимпийские канатные дороги',
+            'description': 'md5:cc730d2bf4215463e37fff6a1e277b13',
+            'thumbnail': 'http://img1.1tv.ru/imgsize640x360/PR20140210114657.JPG',
+            'duration': 149,
+        },
+        'skip': 'Only works from Russia',
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id, 'Downloading page')
+
+        video_url = self._html_search_regex(
+            r'''(?s)jwplayer\('flashvideoportal_1'\)\.setup\({.*?'file': '([^']+)'.*?}\);''', webpage, 'video URL')
+
+        title = self._html_search_regex(
+            r'<div class="tv_translation">\s*<h1><a href="[^"]+">([^<]*)</a>', webpage, 'title')
+        description = self._html_search_regex(
+            r'<div class="descr">\s*<div>&nbsp;</div>\s*<p>([^<]*)</p></div>', webpage, 'description', fatal=False)
+
+        thumbnail = self._og_search_thumbnail(webpage)
+        duration = self._og_search_property('video:duration', webpage, 'video duration', fatal=False)
+
+        like_count = self._html_search_regex(r'title="Понравилось".*?/></label> \[(\d+)\]',
+            webpage, 'like count', fatal=False)
+        dislike_count = self._html_search_regex(r'title="Не понравилось".*?/></label> \[(\d+)\]',
+            webpage, 'dislike count', fatal=False)
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'thumbnail': thumbnail,
+            'title': title,
+            'description': description,
+            'duration': int_or_none(duration),
+            'like_count': int_or_none(like_count),
+            'dislike_count': int_or_none(dislike_count),
+        }

youtube_dl/extractor/flickr.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -11,13 +13,13 @@ class FlickrIE(InfoExtractor):
     """Information Extractor for Flickr videos"""
     _VALID_URL = r'(?:https?://)?(?:www\.|secure\.)?flickr\.com/photos/(?P<uploader_id>[\w\-_@]+)/(?P<id>\d+).*'
     _TEST = {
-        u'url': u'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
+        'url': 'http://www.flickr.com/photos/forestwander-nature-pictures/5645318632/in/photostream/',
-        u'file': u'5645318632.mp4',
+        'file': '5645318632.mp4',
-        u'md5': u'6fdc01adbc89d72fc9c4f15b4a4ba87b',
+        'md5': '6fdc01adbc89d72fc9c4f15b4a4ba87b',
-        u'info_dict': {
+        'info_dict': {
-            u"description": u"Waterfalls in the Springtime at Dark Hollow Waterfalls. These are located just off of Skyline Drive in Virginia. They are only about 6/10 of a mile hike but it is a pretty steep hill and a good climb back up.", 
+            "description": "Waterfalls in the Springtime at Dark Hollow Waterfalls. These are located just off of Skyline Drive in Virginia. They are only about 6/10 of a mile hike but it is a pretty steep hill and a good climb back up.", 
-            u"uploader_id": u"forestwander-nature-pictures", 
+            "uploader_id": "forestwander-nature-pictures", 
-            u"title": u"Dark Hollow Waterfalls"
+            "title": "Dark Hollow Waterfalls"
         }
     }
 
@@ -29,13 +31,13 @@ class FlickrIE(InfoExtractor):
         webpage_url = 'http://www.flickr.com/photos/' + video_uploader_id + '/' + video_id
         webpage = self._download_webpage(webpage_url, video_id)
 
-        secret = self._search_regex(r"photo_secret: '(\w+)'", webpage, u'secret')
+        secret = self._search_regex(r"photo_secret: '(\w+)'", webpage, 'secret')
 
         first_url = 'https://secure.flickr.com/apps/video/video_mtl_xml.gne?v=x&photo_id=' + video_id + '&secret=' + secret + '&bitrate=700&target=_self'
         first_xml = self._download_webpage(first_url, video_id, 'Downloading first data webpage')
 
         node_id = self._html_search_regex(r'<Item id="id">(\d+-\d+)</Item>',
-            first_xml, u'node_id')
+            first_xml, 'node_id')
 
         second_url = 'https://secure.flickr.com/video_playlist.gne?node_id=' + node_id + '&tech=flash&mode=playlist&bitrate=700&secret=' + secret + '&rd=video.yahoo.com&noad=1'
         second_xml = self._download_webpage(second_url, video_id, 'Downloading second data webpage')
@@ -44,7 +46,7 @@ class FlickrIE(InfoExtractor):
 
         mobj = re.search(r'<STREAM APP="(.+?)" FULLPATH="(.+?)"', second_xml)
         if mobj is None:
-            raise ExtractorError(u'Unable to extract video url')
+            raise ExtractorError('Unable to extract video url')
         video_url = mobj.group(1) + unescapeHTML(mobj.group(2))
 
         return [{

youtube_dl/extractor/fourtube.py

@@ -0,0 +1,95 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+    unified_strdate,
+    str_to_int,
+    parse_duration,
+)
+from youtube_dl.utils import clean_html
+
+
+class FourTubeIE(InfoExtractor):
+    IE_NAME = '4tube'
+    _VALID_URL = r'https?://(?:www\.)?4tube\.com/videos/(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'http://www.4tube.com/videos/209733/hot-babe-holly-michaels-gets-her-ass-stuffed-by-black',
+        'md5': '6516c8ac63b03de06bc8eac14362db4f',
+        'info_dict': {
+            'id': '209733',
+            'ext': 'mp4',
+            'title': 'Hot Babe Holly Michaels gets her ass stuffed by black',
+            'uploader': 'WCP Club',
+            'uploader_id': 'wcp-club',
+            'upload_date': '20131031',
+            'duration': 583,
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        video_id = mobj.group('id')
+        webpage_url = 'http://www.4tube.com/videos/' + video_id
+        webpage = self._download_webpage(webpage_url, video_id)
+
+        self.report_extraction(video_id)
+
+        playlist_json = self._html_search_regex(r'var playerConfigPlaylist\s+=\s+([^;]+)', webpage, 'Playlist')
+        media_id = self._search_regex(r'idMedia:\s*(\d+)', playlist_json, 'Media Id')
+        sources = self._search_regex(r'sources:\s*\[([^\]]*)\]', playlist_json, 'Sources').split(',')
+        title = self._search_regex(r'title:\s*"([^"]*)', playlist_json, 'Title')
+        thumbnail_url = self._search_regex(r'image:\s*"([^"]*)', playlist_json, 'Thumbnail', fatal=False)
+
+        uploader_str = self._search_regex(r'<span>Uploaded by</span>(.*?)<span>', webpage, 'uploader', fatal=False)
+        mobj = re.search(r'<a href="/sites/(?P<id>[^"]+)"><strong>(?P<name>[^<]+)</strong></a>', uploader_str)
+        (uploader, uploader_id) = (mobj.group('name'), mobj.group('id')) if mobj else (clean_html(uploader_str), None)
+
+        upload_date = None
+        view_count = None
+        duration = None
+        description = self._html_search_meta('description', webpage, 'description')
+        if description:
+            upload_date = self._search_regex(r'Published Date: (\d{2} [a-zA-Z]{3} \d{4})', description, 'upload date',
+                fatal=False)
+            if upload_date:
+                upload_date = unified_strdate(upload_date)
+            view_count = self._search_regex(r'Views: ([\d,\.]+)', description, 'view count', fatal=False)
+            if view_count:
+                view_count = str_to_int(view_count)
+            duration = parse_duration(self._search_regex(r'Length: (\d+m\d+s)', description, 'duration', fatal=False))
+
+        token_url = "http://tkn.4tube.com/{0}/desktop/{1}".format(media_id, "+".join(sources))
+        headers = {
+                b'Content-Type': b'application/x-www-form-urlencoded',
+                b'Origin': b'http://www.4tube.com',
+                }
+        token_req = compat_urllib_request.Request(token_url, b'{}', headers)
+        tokens = self._download_json(token_req, video_id)
+
+        formats = [{
+            'url': tokens[format]['token'],
+            'format_id': format + 'p',
+            'resolution': format + 'p',
+            'quality': int(format),
+            } for format in sources]
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'thumbnail': thumbnail_url,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'upload_date': upload_date,
+            'view_count': view_count,
+            'duration': duration,
+            'age_limit': 18,
+            'webpage_url': webpage_url,
+        }

youtube_dl/extractor/franceinter.py

@@ -0,0 +1,38 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class FranceInterIE(InfoExtractor):
+    _VALID_URL = r'http://(?:www\.)?franceinter\.fr/player/reecouter\?play=(?P<id>[0-9]{6})'
+    _TEST = {
+        'url': 'http://www.franceinter.fr/player/reecouter?play=793962',
+        'file': '793962.mp3',
+        'md5': '4764932e466e6f6c79c317d2e74f6884',
+        "info_dict": {
+            "title": "L’Histoire dans les jeux vidéo",
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+        title = self._html_search_regex(
+            r'<span class="roll_overflow">(.*?)</span></h1>', webpage, 'title')
+        path = self._search_regex(
+            r'&urlAOD=(.*?)&startTime', webpage, 'video url')
+        video_url = 'http://www.franceinter.fr/' + path
+
+        return {
+            'id': video_id,
+            'formats': [{
+                'url': video_url,
+                'vcodec': 'none',
+            }],
+            'title': title,
+        }

youtube_dl/extractor/francetv.py

@@ -1,4 +1,7 @@
 # encoding: utf-8
+
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -30,7 +33,7 @@ class FranceTVBaseInfoExtractor(InfoExtractor):
 
 
 class PluzzIE(FranceTVBaseInfoExtractor):
-    IE_NAME = u'pluzz.francetv.fr'
+    IE_NAME = 'pluzz.francetv.fr'
     _VALID_URL = r'https?://pluzz\.francetv\.fr/videos/(.*?)\.html'
 
     # Can't use tests, videos expire in 7 days
@@ -44,17 +47,17 @@ class PluzzIE(FranceTVBaseInfoExtractor):
 
 
 class FranceTvInfoIE(FranceTVBaseInfoExtractor):
-    IE_NAME = u'francetvinfo.fr'
+    IE_NAME = 'francetvinfo.fr'
     _VALID_URL = r'https?://www\.francetvinfo\.fr/replay.*/(?P<title>.+)\.html'
 
     _TEST = {
-        u'url': u'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
+        'url': 'http://www.francetvinfo.fr/replay-jt/france-3/soir-3/jt-grand-soir-3-lundi-26-aout-2013_393427.html',
-        u'file': u'84981923.mp4',
+        'file': '84981923.mp4',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'Soir 3',
+            'title': 'Soir 3',
         },
-        u'params': {
+        'params': {
-            u'skip_download': True,
+            'skip_download': True,
         },
     }
 
@@ -62,13 +65,13 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         page_title = mobj.group('title')
         webpage = self._download_webpage(url, page_title)
-        video_id = self._search_regex(r'id-video=(\d+?)"', webpage, u'video id')
+        video_id = self._search_regex(r'id-video=(\d+?)[@"]', webpage, 'video id')
         return self._extract_video(video_id)
 
 
 class FranceTVIE(FranceTVBaseInfoExtractor):
-    IE_NAME = u'francetv'
+    IE_NAME = 'francetv'
-    IE_DESC = u'France 2, 3, 4, 5 and Ô'
+    IE_DESC = 'France 2, 3, 4, 5 and Ô'
     _VALID_URL = r'''(?x)https?://www\.france[2345o]\.fr/
         (?:
             emissions/.*?/(videos|emissions)/(?P<id>[^/?]+)
@@ -78,73 +81,73 @@ class FranceTVIE(FranceTVBaseInfoExtractor):
     _TESTS = [
         # france2
         {
-            u'url': u'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
+            'url': 'http://www.france2.fr/emissions/13h15-le-samedi-le-dimanche/videos/75540104',
-            u'file': u'75540104.mp4',
+            'file': '75540104.mp4',
-            u'info_dict': {
+            'info_dict': {
-                u'title': u'13h15, le samedi...',
+                'title': '13h15, le samedi...',
-                u'description': u'md5:2e5b58ba7a2d3692b35c792be081a03d',
+                'description': 'md5:2e5b58ba7a2d3692b35c792be081a03d',
             },
-            u'params': {
+            'params': {
                 # m3u8 download
-                u'skip_download': True,
+                'skip_download': True,
             },
         },
         # france3
         {
-            u'url': u'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575',
+            'url': 'http://www.france3.fr/emissions/pieces-a-conviction/diffusions/13-11-2013_145575',
-            u'info_dict': {
+            'info_dict': {
-                u'id': u'000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au',
+                'id': '000702326_CAPP_PicesconvictionExtrait313022013_120220131722_Au',
-                u'ext': u'flv',
+                'ext': 'flv',
-                u'title': u'Le scandale du prix des médicaments',
+                'title': 'Le scandale du prix des médicaments',
-                u'description': u'md5:1384089fbee2f04fc6c9de025ee2e9ce',
+                'description': 'md5:1384089fbee2f04fc6c9de025ee2e9ce',
             },
-            u'params': {
+            'params': {
                 # rtmp download
-                u'skip_download': True,
+                'skip_download': True,
             },
         },
         # france4
         {
-            u'url': u'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
+            'url': 'http://www.france4.fr/emissions/hero-corp/videos/rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
-            u'info_dict': {
+            'info_dict': {
-                u'id': u'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
+                'id': 'rhozet_herocorp_bonus_1_20131106_1923_06112013172108_F4',
-                u'ext': u'flv',
+                'ext': 'flv',
-                u'title': u'Hero Corp Making of - Extrait 1',
+                'title': 'Hero Corp Making of - Extrait 1',
-                u'description': u'md5:c87d54871b1790679aec1197e73d650a',
+                'description': 'md5:c87d54871b1790679aec1197e73d650a',
             },
-            u'params': {
+            'params': {
                 # rtmp download
-                u'skip_download': True,
+                'skip_download': True,
             },
         },
         # france5
         {
-            u'url': u'http://www.france5.fr/emissions/c-a-dire/videos/92837968',
+            'url': 'http://www.france5.fr/emissions/c-a-dire/videos/92837968',
-            u'info_dict': {
+            'info_dict': {
-                u'id': u'92837968',
+                'id': '92837968',
-                u'ext': u'mp4',
+                'ext': 'mp4',
-                u'title': u'C à dire ?!',
+                'title': 'C à dire ?!',
-                u'description': u'md5:fb1db1cbad784dcce7c7a7bd177c8e2f',
+                'description': 'md5:fb1db1cbad784dcce7c7a7bd177c8e2f',
             },
-            u'params': {
+            'params': {
                 # m3u8 download
-                u'skip_download': True,
+                'skip_download': True,
             },
         },
         # franceo
         {
-            u'url': u'http://www.franceo.fr/jt/info-afrique/04-12-2013',
+            'url': 'http://www.franceo.fr/jt/info-afrique/04-12-2013',
-            u'info_dict': {
+            'info_dict': {
-                u'id': u'92327925',
+                'id': '92327925',
-                u'ext': u'mp4',
+                'ext': 'mp4',
-                u'title': u'Infô-Afrique',
+                'title': 'Infô-Afrique',
-                u'description': u'md5:ebf346da789428841bee0fd2a935ea55',
+                'description': 'md5:ebf346da789428841bee0fd2a935ea55',
             },
-            u'params': {
+            'params': {
                 # m3u8 download
-                u'skip_download': True,
+                'skip_download': True,
             },
-            u'skip': u'The id changes frequently',
+            'skip': 'The id changes frequently',
         },
     ]
 
@@ -160,27 +163,28 @@ class FranceTVIE(FranceTVBaseInfoExtractor):
                  '\.fr/\?id-video=([^"/&]+)'),
                 (r'<a class="video" id="ftv_player_(.+?)"'),
             ]
-            video_id = self._html_search_regex(id_res, webpage, u'video ID')
+            video_id = self._html_search_regex(id_res, webpage, 'video ID')
         else:
             video_id = mobj.group('id')
         return self._extract_video(video_id)
 
 
 class GenerationQuoiIE(InfoExtractor):
-    IE_NAME = u'france2.fr:generation-quoi'
+    IE_NAME = 'france2.fr:generation-quoi'
     _VALID_URL = r'https?://generation-quoi\.france2\.fr/portrait/(?P<name>.*)(\?|$)'
 
     _TEST = {
-        u'url': u'http://generation-quoi.france2.fr/portrait/garde-a-vous',
+        'url': 'http://generation-quoi.france2.fr/portrait/garde-a-vous',
-        u'file': u'k7FJX8VBcvvLmX4wA5Q.mp4',
+        'file': 'k7FJX8VBcvvLmX4wA5Q.mp4',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'Génération Quoi - Garde à Vous',
+            'title': 'Génération Quoi - Garde à Vous',
-            u'uploader': u'Génération Quoi',
+            'uploader': 'Génération Quoi',
         },
-        u'params': {
+        'params': {
             # It uses Dailymotion
-            u'skip_download': True,
+            'skip_download': True,
         },
+        'skip': 'Only available from France',
     }
 
     def _real_extract(self, url):
@@ -191,3 +195,29 @@ class GenerationQuoiIE(InfoExtractor):
         info = json.loads(info_json)
         return self.url_result('http://www.dailymotion.com/video/%s' % info['id'],
             ie='Dailymotion')
+
+
+class CultureboxIE(FranceTVBaseInfoExtractor):
+    IE_NAME = 'culturebox.francetvinfo.fr'
+    _VALID_URL = r'https?://culturebox\.francetvinfo\.fr/(?P<name>.*?)(\?|$)'
+
+    _TEST = {
+        'url': 'http://culturebox.francetvinfo.fr/einstein-on-the-beach-au-theatre-du-chatelet-146813',
+        'info_dict': {
+            'id': 'EV_6785',
+            'ext': 'mp4',
+            'title': 'Einstein on the beach au Théâtre du Châtelet',
+            'description': 'md5:9ce2888b1efefc617b5e58b3f6200eeb',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        name = mobj.group('name')
+        webpage = self._download_webpage(url, name)
+        video_id = self._search_regex(r'"http://videos\.francetv\.fr/video/(.*?)"', webpage, 'video id')
+        return self._extract_video(video_id)

youtube_dl/extractor/freesound.py

@@ -1,18 +1,21 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
-from ..utils import determine_ext
+
 
 class FreesoundIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://)?(?:www\.)?freesound\.org/people/([^/]+)/sounds/(?P<id>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?freesound\.org/people/([^/]+)/sounds/(?P<id>[^/]+)'
     _TEST = {
-        u'url': u'http://www.freesound.org/people/miklovan/sounds/194503/',
+        'url': 'http://www.freesound.org/people/miklovan/sounds/194503/',
-        u'file': u'194503.mp3',
+        'md5': '12280ceb42c81f19a515c745eae07650',
-        u'md5': u'12280ceb42c81f19a515c745eae07650',
+        'info_dict': {
-        u'info_dict': {
+            'id': '194503',
-            u"title": u"gulls in the city.wav",
+            'ext': 'mp3',
-            u"uploader" : u"miklovan",
+            'title': 'gulls in the city.wav',
-            u'description': u'the sounds of seagulls in the city',
+            'uploader': 'miklovan',
+            'description': 'the sounds of seagulls in the city',
         }
     }
 
@@ -20,17 +23,17 @@ class FreesoundIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         music_id = mobj.group('id')
         webpage = self._download_webpage(url, music_id)
-        title = self._html_search_regex(r'<div id="single_sample_header">.*?<a href="#">(.+?)</a>',
+        title = self._html_search_regex(
+            r'<div id="single_sample_header">.*?<a href="#">(.+?)</a>',
             webpage, 'music title', flags=re.DOTALL)
-        music_url = self._og_search_property('audio', webpage, 'music url')
+        description = self._html_search_regex(
-        description = self._html_search_regex(r'<div id="sound_description">(.*?)</div>',
+            r'<div id="sound_description">(.*?)</div>', webpage, 'description',
-                                webpage, 'description', fatal=False, flags=re.DOTALL)
+            fatal=False, flags=re.DOTALL)
 
-        return [{
+        return {
             'id': music_id,
             'title': title,
-            'url':      music_url,
+            'url': self._og_search_property('audio', webpage, 'music url'),
             'uploader': self._og_search_property('audio:artist', webpage, 'music uploader'),
-            'ext':      determine_ext(music_url),
             'description': description,
-        }]
+        }

youtube_dl/extractor/freespeech.py

@@ -0,0 +1,37 @@
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+
+
+class FreespeechIE(InfoExtractor):
+    IE_NAME = 'freespeech.org'
+    _VALID_URL = r'https://www\.freespeech\.org/video/(?P<title>.+)'
+    _TEST = {
+        'add_ie': ['Youtube'],
+        'url': 'https://www.freespeech.org/video/obama-romney-campaign-colorado-ahead-debate-0',
+        'info_dict': {
+            'id': 'poKsVCZ64uU',
+            'ext': 'mp4',
+            'title': 'Obama, Romney Campaign in Colorado Ahead of Debate',
+            'description': 'Obama, Romney Campaign in Colorado Ahead of Debate',
+            'uploader': 'freespeechtv',
+            'uploader_id': 'freespeechtv',
+            'upload_date': '20121002',
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        title = mobj.group('title')
+        webpage = self._download_webpage(url, title)
+        info_json = self._search_regex(r'jQuery.extend\(Drupal.settings, ({.*?})\);', webpage, 'info')
+        info = json.loads(info_json)
+
+        return {
+            '_type': 'url',
+            'url': info['jw_player']['basic_video_node_player']['file'],
+            'ie_key': 'Youtube',
+        }

youtube_dl/extractor/funnyordie.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -6,13 +8,16 @@ from .common import InfoExtractor
 class FunnyOrDieIE(InfoExtractor):
     _VALID_URL = r'^(?:https?://)?(?:www\.)?funnyordie\.com/videos/(?P<id>[0-9a-f]+)/.*$'
     _TEST = {
-        u'url': u'http://www.funnyordie.com/videos/0732f586d7/heart-shaped-box-literal-video-version',
+        'url': 'http://www.funnyordie.com/videos/0732f586d7/heart-shaped-box-literal-video-version',
-        u'file': u'0732f586d7.mp4',
+        'file': '0732f586d7.mp4',
-        u'md5': u'f647e9e90064b53b6e046e75d0241fbd',
+        'md5': 'f647e9e90064b53b6e046e75d0241fbd',
-        u'info_dict': {
+        'info_dict': {
-            u"description": u"Lyrics changed to match the video. Spoken cameo by Obscurus Lupa (from ThatGuyWithTheGlasses.com). Based on a concept by Dustin McLean (DustFilms.com). Performed, edited, and written by David A. Scott.", 
+            'description': ('Lyrics changed to match the video. Spoken cameo '
-            u"title": u"Heart-Shaped Box: Literal Video Version"
+                'by Obscurus Lupa (from ThatGuyWithTheGlasses.com). Based on a '
-        }
+                'concept by Dustin McLean (DustFilms.com). Performed, edited, '
+                'and written by David A. Scott.'),
+            'title': 'Heart-Shaped Box: Literal Video Version',
+        },
     }
 
     def _real_extract(self, url):
@@ -23,13 +28,12 @@ class FunnyOrDieIE(InfoExtractor):
 
         video_url = self._search_regex(
             [r'type="video/mp4" src="(.*?)"', r'src="([^>]*?)" type=\'video/mp4\''],
-            webpage, u'video URL', flags=re.DOTALL)
+            webpage, 'video URL', flags=re.DOTALL)
 
-        info = {
+        return {
             'id': video_id,
             'url': video_url,
             'ext': 'mp4',
             'title': self._og_search_title(webpage),
             'description': self._og_search_description(webpage),
         }
-        return [info]

youtube_dl/extractor/gamespot.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -13,12 +15,12 @@ from ..utils import (
 class GameSpotIE(InfoExtractor):
     _VALID_URL = r'(?:http://)?(?:www\.)?gamespot\.com/.*-(?P<page_id>\d+)/?'
     _TEST = {
-        u"url": u"http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
+        "url": "http://www.gamespot.com/arma-iii/videos/arma-iii-community-guide-sitrep-i-6410818/",
-        u"file": u"gs-2300-6410818.mp4",
+        "file": "gs-2300-6410818.mp4",
-        u"md5": u"b2a30deaa8654fcccd43713a6b6a4825",
+        "md5": "b2a30deaa8654fcccd43713a6b6a4825",
-        u"info_dict": {
+        "info_dict": {
-            u"title": u"Arma 3 - Community Guide: SITREP I",
+            "title": "Arma 3 - Community Guide: SITREP I",
-            u'description': u'Check out this video where some of the basics of Arma 3 is explained.',
+            'description': 'Check out this video where some of the basics of Arma 3 is explained.',
         }
     }
 
@@ -26,7 +28,7 @@ class GameSpotIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         page_id = mobj.group('page_id')
         webpage = self._download_webpage(url, page_id)
-        data_video_json = self._search_regex(r'data-video=\'(.*?)\'', webpage, u'data video')
+        data_video_json = self._search_regex(r'data-video=["\'](.*?)["\']', webpage, 'data video')
         data_video = json.loads(unescapeHTML(data_video_json))
 
         # Transform the manifest url to a link to the mp4 files
@@ -34,7 +36,7 @@ class GameSpotIE(InfoExtractor):
         f4m_url = data_video['videoStreams']['f4m_stream']
         f4m_path = compat_urlparse.urlparse(f4m_url).path
         QUALITIES_RE = r'((,\d+)+,?)'
-        qualities = self._search_regex(QUALITIES_RE, f4m_path, u'qualities').strip(',').split(',')
+        qualities = self._search_regex(QUALITIES_RE, f4m_path, 'qualities').strip(',').split(',')
         http_path = f4m_path[1:].split('/', 1)[1]
         http_template = re.sub(QUALITIES_RE, r'%s', http_path)
         http_template = http_template.replace('.csmil/manifest.f4m', '')

youtube_dl/extractor/gametrailers.py

@@ -1,4 +1,4 @@
-import re
+from __future__ import unicode_literals
 
 from .mtv import MTVServicesInfoExtractor
 
@@ -6,22 +6,14 @@ from .mtv import MTVServicesInfoExtractor
 class GametrailersIE(MTVServicesInfoExtractor):
     _VALID_URL = r'http://www\.gametrailers\.com/(?P<type>videos|reviews|full-episodes)/(?P<id>.*?)/(?P<title>.*)'
     _TEST = {
-        u'url': u'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer',
+        'url': 'http://www.gametrailers.com/videos/zbvr8i/mirror-s-edge-2-e3-2013--debut-trailer',
-        u'file': u'70e9a5d7-cf25-4a10-9104-6f3e7342ae0d.mp4',
+        'md5': '4c8e67681a0ea7ec241e8c09b3ea8cf7',
-        u'md5': u'4c8e67681a0ea7ec241e8c09b3ea8cf7',
+        'info_dict': {
-        u'info_dict': {
+            'id': '70e9a5d7-cf25-4a10-9104-6f3e7342ae0d',
-            u'title': u'E3 2013: Debut Trailer',
+            'ext': 'mp4',
-            u'description': u'Faith is back!  Check out the World Premiere trailer for Mirror\'s Edge 2 straight from the EA Press Conference at E3 2013!',
+            'title': 'E3 2013: Debut Trailer',
+            'description': 'Faith is back!  Check out the World Premiere trailer for Mirror\'s Edge 2 straight from the EA Press Conference at E3 2013!',
         },
     }
 
     _FEED_URL = 'http://www.gametrailers.com/feeds/mrss'
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-        mgid = self._search_regex([r'data-video="(?P<mgid>mgid:.*?)"',
-                                   r'data-contentId=\'(?P<mgid>mgid:.*?)\''],
-                                  webpage, u'mgid')
-        return self._get_videos_info(mgid)

youtube_dl/extractor/generic.py

@@ -1,9 +1,12 @@
 # encoding: utf-8
 
+from __future__ import unicode_literals
+
 import os
 import re
 
 from .common import InfoExtractor
+from .youtube import YoutubeIE
 from ..utils import (
     compat_urllib_error,
     compat_urllib_parse,
@@ -11,64 +14,90 @@ from ..utils import (
     compat_urlparse,
 
     ExtractorError,
+    HEADRequest,
     smuggle_url,
     unescapeHTML,
+    unified_strdate,
+    url_basename,
 )
 from .brightcove import BrightcoveIE
+from .ooyala import OoyalaIE
 
 
 class GenericIE(InfoExtractor):
-    IE_DESC = u'Generic downloader that works on some sites'
+    IE_DESC = 'Generic downloader that works on some sites'
     _VALID_URL = r'.*'
-    IE_NAME = u'generic'
+    IE_NAME = 'generic'
     _TESTS = [
         {
-            u'url': u'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
+            'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
-            u'file': u'13601338388002.mp4',
+            'file': '13601338388002.mp4',
-            u'md5': u'6e15c93721d7ec9e9ca3fdbf07982cfd',
+            'md5': '6e15c93721d7ec9e9ca3fdbf07982cfd',
-            u'info_dict': {
+            'info_dict': {
-                u"uploader": u"www.hodiho.fr",
+                'uploader': 'www.hodiho.fr',
-                u"title": u"R\u00e9gis plante sa Jeep"
+                'title': 'R\u00e9gis plante sa Jeep',
-            }
-        },
-        # embedded vimeo video
-        {
-            u'add_ie': ['Vimeo'],
-            u'url': u'http://skillsmatter.com/podcast/home/move-semanticsperfect-forwarding-and-rvalue-references',
-            u'file': u'22444065.mp4',
-            u'md5': u'2903896e23df39722c33f015af0666e2',
-            u'info_dict': {
-                u'title': u'ACCU 2011: Move Semantics,Perfect Forwarding, and Rvalue references- Scott Meyers- 13/04/2011',
-                u"uploader_id": u"skillsmatter",
-                u"uploader": u"Skills Matter",
             }
         },
         # bandcamp page with custom domain
         {
-            u'add_ie': ['Bandcamp'],
+            'add_ie': ['Bandcamp'],
-            u'url': u'http://bronyrock.com/track/the-pony-mash',
+            'url': 'http://bronyrock.com/track/the-pony-mash',
-            u'file': u'3235767654.mp3',
+            'file': '3235767654.mp3',
-            u'info_dict': {
+            'info_dict': {
-                u'title': u'The Pony Mash',
+                'title': 'The Pony Mash',
-                u'uploader': u'M_Pallante',
+                'uploader': 'M_Pallante',
             },
-            u'skip': u'There is a limit of 200 free downloads / month for the test song',
+            'skip': 'There is a limit of 200 free downloads / month for the test song',
         },
         # embedded brightcove video
         # it also tests brightcove videos that need to set the 'Referer' in the
         # http requests
         {
-            u'add_ie': ['Brightcove'],
+            'add_ie': ['Brightcove'],
-            u'url': u'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
+            'url': 'http://www.bfmtv.com/video/bfmbusiness/cours-bourse/cours-bourse-l-analyse-technique-154522/',
-            u'info_dict': {
+            'info_dict': {
-                u'id': u'2765128793001',
+                'id': '2765128793001',
-                u'ext': u'mp4',
+                'ext': 'mp4',
-                u'title': u'Le cours de bourse : l’analyse technique',
+                'title': 'Le cours de bourse : l’analyse technique',
-                u'description': u'md5:7e9ad046e968cb2d1114004aba466fd9',
+                'description': 'md5:7e9ad046e968cb2d1114004aba466fd9',
-                u'uploader': u'BFM BUSINESS',
+                'uploader': 'BFM BUSINESS',
             },
-            u'params': {
+            'params': {
-                u'skip_download': True,
+                'skip_download': True,
+            },
+        },
+        {
+            # https://github.com/rg3/youtube-dl/issues/2253
+            'url': 'http://bcove.me/i6nfkrc3',
+            'file': '3101154703001.mp4',
+            'md5': '0ba9446db037002366bab3b3eb30c88c',
+            'info_dict': {
+                'title': 'Still no power',
+                'uploader': 'thestar.com',
+                'description': 'Mississauga resident David Farmer is still out of power as a result of the ice storm a month ago. To keep the house warm, Farmer cuts wood from his property for a wood burning stove downstairs.',
+            },
+            'add_ie': ['Brightcove'],
+        },
+        # Direct link to a video
+        {
+            'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
+            'file': 'trailer.mp4',
+            'md5': '67d406c2bcb6af27fa886f31aa934bbe',
+            'info_dict': {
+                'id': 'trailer',
+                'title': 'trailer',
+                'upload_date': '20100513',
+            }
+        },
+        # ooyala video
+        {
+            'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
+            'file': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ.mp4',
+            'md5': '5644c6ca5d5782c1d0d350dad9bd840c',
+            'info_dict': {
+                'id': 'BwY2RxaTrTkslxOfcan0UCf0YqyvWysJ',
+                'ext': 'mp4',
+                'title': '2cc213299525360.mov',  # that's what we get
             },
         },
     ]
@@ -76,30 +105,27 @@ class GenericIE(InfoExtractor):
     def report_download_webpage(self, video_id):
         """Report webpage download."""
         if not self._downloader.params.get('test', False):
-            self._downloader.report_warning(u'Falling back on generic information extractor.')
+            self._downloader.report_warning('Falling back on generic information extractor.')
         super(GenericIE, self).report_download_webpage(video_id)
 
     def report_following_redirect(self, new_url):
         """Report information extraction."""
-        self._downloader.to_screen(u'[redirect] Following redirect to %s' % new_url)
+        self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
 
-    def _test_redirect(self, url):
+    def _send_head(self, url):
         """Check if it is a redirect, like url shorteners, in case return the new url."""
-        class HeadRequest(compat_urllib_request.Request):
-            def get_method(self):
-                return "HEAD"
 
         class HEADRedirectHandler(compat_urllib_request.HTTPRedirectHandler):
             """
             Subclass the HTTPRedirectHandler to make it use our
-            HeadRequest also on the redirected URL
+            HEADRequest also on the redirected URL
             """
             def redirect_request(self, req, fp, code, msg, headers, newurl):
                 if code in (301, 302, 303, 307):
                     newurl = newurl.replace(' ', '%20')
                     newheaders = dict((k,v) for k,v in req.headers.items()
                                       if k.lower() not in ("content-length", "content-type"))
-                    return HeadRequest(newurl,
+                    return HEADRequest(newurl,
                                        headers=newheaders,
                                        origin_req_host=req.get_origin_req_host(),
                                        unverifiable=True)
@@ -128,38 +154,68 @@ class GenericIE(InfoExtractor):
                         compat_urllib_request.HTTPErrorProcessor, compat_urllib_request.HTTPSHandler]:
             opener.add_handler(handler())
 
-        response = opener.open(HeadRequest(url))
+        response = opener.open(HEADRequest(url))
         if response is None:
-            raise ExtractorError(u'Invalid URL protocol')
+            raise ExtractorError('Invalid URL protocol')
-        new_url = response.geturl()
+        return response
-
-        if url == new_url:
-            return False
-
-        self.report_following_redirect(new_url)
-        return new_url
 
     def _real_extract(self, url):
         parsed_url = compat_urlparse.urlparse(url)
         if not parsed_url.scheme:
+            default_search = self._downloader.params.get('default_search')
+            if default_search is None:
+                default_search = 'auto'
+
+            if default_search == 'auto':
+                if '/' in url:
                     self._downloader.report_warning('The url doesn\'t specify the protocol, trying with http')
                     return self.url_result('http://' + url)
+                else:
+                    return self.url_result('ytsearch:' + url)
+            else:
+                assert ':' in default_search
+                return self.url_result(default_search + url)
+        video_id = os.path.splitext(url.split('/')[-1])[0]
+
+        self.to_screen('%s: Requesting header' % video_id)
 
         try:
-            new_url = self._test_redirect(url)
+            response = self._send_head(url)
-            if new_url:
+
-                return [self.url_result(new_url)]
+            # Check for redirect
+            new_url = response.geturl()
+            if url != new_url:
+                self.report_following_redirect(new_url)
+                return self.url_result(new_url)
+
+            # Check for direct link to a video
+            content_type = response.headers.get('Content-Type', '')
+            m = re.match(r'^(?P<type>audio|video|application(?=/ogg$))/(?P<format_id>.+)$', content_type)
+            if m:
+                upload_date = response.headers.get('Last-Modified')
+                if upload_date:
+                    upload_date = unified_strdate(upload_date)
+                return {
+                    'id': video_id,
+                    'title': os.path.splitext(url_basename(url))[0],
+                    'formats': [{
+                        'format_id': m.group('format_id'),
+                        'url': url,
+                        'vcodec': 'none' if m.group('type') == 'audio' else None
+                    }],
+                    'upload_date': upload_date,
+                }
+
         except compat_urllib_error.HTTPError:
             # This may be a stupid server that doesn't like HEAD, our UA, or so
             pass
 
-        video_id = url.split('/')[-1]
         try:
             webpage = self._download_webpage(url, video_id)
         except ValueError:
             # since this is the last-resort InfoExtractor, if
             # this error is thrown, it'll be thrown here
-            raise ExtractorError(u'Failed to download URL: %s' % url)
+            raise ExtractorError('Failed to download URL: %s' % url)
 
         self.report_extraction(video_id)
 
@@ -170,30 +226,50 @@ class GenericIE(InfoExtractor):
         #   Video Title - Tagline | Site Name
         # and so on and so forth; it's just not practical
         video_title = self._html_search_regex(
-            r'(?s)<title>(.*?)</title>', webpage, u'video title',
+            r'(?s)<title>(.*?)</title>', webpage, 'video title',
-            default=u'video')
+            default='video')
 
         # video uploader is domain name
         video_uploader = self._search_regex(
-            r'^(?:https?://)?([^/]*)/.*', url, u'video uploader')
+            r'^(?:https?://)?([^/]*)/.*', url, 'video uploader')
 
         # Look for BrightCove:
-        bc_url = BrightcoveIE._extract_brightcove_url(webpage)
+        bc_urls = BrightcoveIE._extract_brightcove_urls(webpage)
-        if bc_url is not None:
+        if bc_urls:
-            self.to_screen(u'Brightcove video detected.')
+            self.to_screen('Brightcove video detected.')
-            return self.url_result(bc_url, 'Brightcove')
+            entries = [{
+                '_type': 'url',
+                'url': smuggle_url(bc_url, {'Referer': url}),
+                'ie_key': 'Brightcove'
+            } for bc_url in bc_urls]
 
-        # Look for embedded Vimeo player
+            return {
+                '_type': 'playlist',
+                'title': video_title,
+                'id': video_id,
+                'entries': entries,
+            }
+
+        # Look for embedded (iframe) Vimeo player
         mobj = re.search(
-            r'<iframe[^>]+?src="(https?://player.vimeo.com/video/.+?)"', webpage)
+            r'<iframe[^>]+?src="((?:https?:)?//player\.vimeo\.com/video/.+?)"', webpage)
         if mobj:
             player_url = unescapeHTML(mobj.group(1))
             surl = smuggle_url(player_url, {'Referer': url})
             return self.url_result(surl, 'Vimeo')
 
+        # Look for embedded (swf embed) Vimeo player
+        mobj = re.search(
+            r'<embed[^>]+?src="(https?://(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
+        if mobj:
+            return self.url_result(mobj.group(1), 'Vimeo')
+
         # Look for embedded YouTube player
-        matches = re.findall(
+        matches = re.findall(r'''(?x)
-            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube\.com/embed/.+?)\1', webpage)
+            (?:<iframe[^>]+?src=|embedSWF\(\s*)
+            (["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube\.com/
+                (?:embed|v)/.+?)
+            \1''', webpage)
         if matches:
             urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube')
                      for tuppl in matches]
@@ -223,16 +299,12 @@ class GenericIE(InfoExtractor):
             }
 
         # Look for embedded blip.tv player
-        mobj = re.search(r'<meta\s[^>]*https?://api.blip.tv/\w+/redirect/\w+/(\d+)', webpage)
+        mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
         if mobj:
-            return self.url_result('http://blip.tv/seo/-'+mobj.group(1), 'BlipTV')
+            return self.url_result('http://blip.tv/a/a-'+mobj.group(1), 'BlipTV')
-        mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*https?://(?:\w+\.)?blip.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', webpage)
+        mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*(https?://(?:\w+\.)?blip\.tv/(?:play/|api\.swf#)[a-zA-Z0-9]+)', webpage)
         if mobj:
-            player_url = 'http://blip.tv/play/%s.x?p=1' % mobj.group(1)
+            return self.url_result(mobj.group(1), 'BlipTV')
-            player_page = self._download_webpage(player_url, mobj.group(1))
-            blip_video_id = self._search_regex(r'data-episode-id="(\d+)', player_page, u'blip_video_id', fatal=False)
-            if blip_video_id:
-                return self.url_result('http://blip.tv/seo/-'+blip_video_id, 'BlipTV')
 
         # Look for Bandcamp pages with custom domain
         mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
@@ -241,14 +313,56 @@ class GenericIE(InfoExtractor):
             # Don't set the extractor because it can be a track url or an album
             return self.url_result(burl)
 
+        # Look for embedded Vevo player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'))
+
+        # Look for Ooyala videos
+        mobj = re.search(r'player.ooyala.com/[^"?]+\?[^"]*?(?:embedCode|ec)=([^"&]+)', webpage)
+        if mobj is not None:
+            return OoyalaIE._build_url_result(mobj.group(1))
+
+        # Look for Aparat videos
+        mobj = re.search(r'<iframe src="(http://www\.aparat\.com/video/[^"]+)"', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group(1), 'Aparat')
+
+        # Look for MPORA videos
+        mobj = re.search(r'<iframe .*?src="(http://mpora\.(?:com|de)/videos/[^"]+)"', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group(1), 'Mpora')
+
+        # Look for embedded Novamov player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>http://(?:(?:embed|www)\.)?novamov\.com/embed\.php.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Novamov')
+
+        # Look for embedded Facebook player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'Facebook')
+
+        # Look for embedded Huffington Post player
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed\.live\.huffingtonpost\.com/.+?)\1', webpage)
+        if mobj is not None:
+            return self.url_result(mobj.group('url'), 'HuffPost')
+
         # Start with something easy: JW Player in SWFObject
         mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
+        if mobj is None:
+            # Look for gorilla-vid style embedding
+            mobj = re.search(r'(?s)(?:jw_plugins|JWPlayerOptions).*?file\s*:\s*["\'](.*?)["\']', webpage)
         if mobj is None:
             # Broaden the search a little bit
             mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
         if mobj is None:
             # Broaden the search a little bit: JWPlayer JS loader
-            mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http[^\'"]*)', webpage)
+            mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http(?![^\'"]+\.[0-9]+[\'"])[^\'"]+)["\']', webpage)
         if mobj is None:
             # Try to find twitter cards info
             mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
@@ -263,17 +377,21 @@ class GenericIE(InfoExtractor):
             # HTML5 video
             mobj = re.search(r'<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage, flags=re.DOTALL)
         if mobj is None:
-            raise ExtractorError(u'Unsupported URL: %s' % url)
+            raise ExtractorError('Unsupported URL: %s' % url)
 
         # It's possible that one of the regexes
         # matched, but returned an empty group:
         if mobj.group(1) is None:
-            raise ExtractorError(u'Did not find a valid video URL at %s' % url)
+            raise ExtractorError('Did not find a valid video URL at %s' % url)
 
         video_url = mobj.group(1)
         video_url = compat_urlparse.urljoin(url, video_url)
         video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
 
+        # Sometimes, jwplayer extraction will result in a YouTube URL
+        if YoutubeIE.suitable(video_url):
+            return self.url_result(video_url, 'Youtube')
+
         # here's a fun little line of code for you:
         video_id = os.path.splitext(video_id)[0]
 

youtube_dl/extractor/googleplus.py

@@ -1,4 +1,5 @@
 # coding: utf-8
+from __future__ import unicode_literals
 
 import datetime
 import re
@@ -10,32 +11,28 @@ from ..utils import (
 
 
 class GooglePlusIE(InfoExtractor):
-    IE_DESC = u'Google Plus'
+    IE_DESC = 'Google Plus'
-    _VALID_URL = r'(?:https://)?plus\.google\.com/(?:[^/]+/)*?posts/(\w+)'
+    _VALID_URL = r'https://plus\.google\.com/(?:[^/]+/)*?posts/(?P<id>\w+)'
-    IE_NAME = u'plus.google'
+    IE_NAME = 'plus.google'
     _TEST = {
-        u"url": u"https://plus.google.com/u/0/108897254135232129896/posts/ZButuJc6CtH",
+        'url': 'https://plus.google.com/u/0/108897254135232129896/posts/ZButuJc6CtH',
-        u"file": u"ZButuJc6CtH.flv",
+        'info_dict': {
-        u"info_dict": {
+            'id': 'ZButuJc6CtH',
-            u"upload_date": u"20120613",
+            'ext': 'flv',
-            u"uploader": u"井上ヨシマサ",
+            'upload_date': '20120613',
-            u"title": u"嘆きの天使 降臨"
+            'uploader': '井上ヨシマサ',
+            'title': '嘆きの天使 降臨',
         }
     }
 
     def _real_extract(self, url):
         # Extract id from URL
         mobj = re.match(self._VALID_URL, url)
-        if mobj is None:
-            raise ExtractorError(u'Invalid URL: %s' % url)
 
-        post_url = mobj.group(0)
+        video_id = mobj.group('id')
-        video_id = mobj.group(1)
-
-        video_extension = 'flv'
 
         # Step 1, Retrieve post webpage to extract further information
-        webpage = self._download_webpage(post_url, video_id, u'Downloading entry webpage')
+        webpage = self._download_webpage(url, video_id, 'Downloading entry webpage')
 
         self.report_extraction(video_id)
 
@@ -43,7 +40,7 @@ class GooglePlusIE(InfoExtractor):
         upload_date = self._html_search_regex(
             r'''(?x)<a.+?class="o-U-s\s[^"]+"\s+style="display:\s*none"\s*>
                     ([0-9]{4}-[0-9]{2}-[0-9]{2})</a>''',
-            webpage, u'upload date', fatal=False, flags=re.VERBOSE)
+            webpage, 'upload date', fatal=False, flags=re.VERBOSE)
         if upload_date:
             # Convert timestring to a format suitable for filename
             upload_date = datetime.datetime.strptime(upload_date, "%Y-%m-%d")
@@ -51,28 +48,27 @@ class GooglePlusIE(InfoExtractor):
 
         # Extract uploader
         uploader = self._html_search_regex(r'rel\="author".*?>(.*?)</a>',
-            webpage, u'uploader', fatal=False)
+            webpage, 'uploader', fatal=False)
 
         # Extract title
         # Get the first line for title
         video_title = self._html_search_regex(r'<meta name\=\"Description\" content\=\"(.*?)[\n<"]',
-            webpage, 'title', default=u'NA')
+            webpage, 'title', default='NA')
 
         # Step 2, Simulate clicking the image box to launch video
         DOMAIN = 'https://plus.google.com/'
         video_page = self._search_regex(r'<a href="((?:%s)?photos/.*?)"' % re.escape(DOMAIN),
-            webpage, u'video page URL')
+            webpage, 'video page URL')
         if not video_page.startswith(DOMAIN):
             video_page = DOMAIN + video_page
 
-        webpage = self._download_webpage(video_page, video_id, u'Downloading video page')
+        webpage = self._download_webpage(video_page, video_id, 'Downloading video page')
 
-        # Extract video links on video page
+        # Extract video links all sizes
-        """Extract video links of all sizes"""
         pattern = r'\d+,\d+,(\d+),"(http\://redirector\.googlevideo\.com.*?)"'
         mobj = re.findall(pattern, webpage)
         if len(mobj) == 0:
-            raise ExtractorError(u'Unable to extract video links')
+            raise ExtractorError('Unable to extract video links')
 
         # Sort in resolution
         links = sorted(mobj)
@@ -87,12 +83,11 @@ class GooglePlusIE(InfoExtractor):
         except AttributeError: # Python 3
             video_url = bytes(video_url, 'ascii').decode('unicode-escape')
 
-
+        return {
-        return [{
             'id': video_id,
             'url': video_url,
             'uploader': uploader,
             'upload_date': upload_date,
             'title': video_title,
-            'ext':      video_extension,
+            'ext': 'flv',
-        }]
+        }

youtube_dl/extractor/googlesearch.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import itertools
 import re
 
@@ -8,32 +10,42 @@ from ..utils import (
 
 
 class GoogleSearchIE(SearchInfoExtractor):
-    IE_DESC = u'Google Video search'
+    IE_DESC = 'Google Video search'
-    _MORE_PAGES_INDICATOR = r'id="pnnext" class="pn"'
     _MAX_RESULTS = 1000
-    IE_NAME = u'video.google:search'
+    IE_NAME = 'video.google:search'
     _SEARCH_KEY = 'gvsearch'
 
     def _get_n_results(self, query, n):
         """Get a specified number of results for a query"""
 
+        entries = []
         res = {
             '_type': 'playlist',
             'id': query,
-            'entries': []
+            'title': query,
         }
 
-        for pagenum in itertools.count(1):
+        for pagenum in itertools.count():
-            result_url = u'http://www.google.com/search?tbm=vid&q=%s&start=%s&hl=en' % (compat_urllib_parse.quote_plus(query), pagenum*10)
+            result_url = (
-            webpage = self._download_webpage(result_url, u'gvsearch:' + query,
+                'http://www.google.com/search?tbm=vid&q=%s&start=%s&hl=en'
-                                             note='Downloading result page ' + str(pagenum))
+                % (compat_urllib_parse.quote_plus(query), pagenum * 10))
 
-            for mobj in re.finditer(r'<h3 class="r"><a href="([^"]+)"', webpage):
+            webpage = self._download_webpage(
-                e = {
+                result_url, 'gvsearch:' + query,
+                note='Downloading result page ' + str(pagenum + 1))
+
+            for hit_idx, mobj in enumerate(re.finditer(
+                    r'<h3 class="r"><a href="([^"]+)"', webpage)):
+
+                # Skip playlists
+                if not re.search(r'id="vidthumb%d"' % (hit_idx + 1), webpage):
+                    continue
+
+                entries.append({
                     '_type': 'url',
                     'url': mobj.group(1)
-                }
+                })
-                res['entries'].append(e)
 
-            if (pagenum * 10 > n) or not re.search(self._MORE_PAGES_INDICATOR, webpage):
+            if (len(entries) >= n) or not re.search(r'class="pn" id="pnnext"', webpage):
+                res['entries'] = entries[:n]
                 return res

youtube_dl/extractor/helsinki.py

@@ -0,0 +1,62 @@
+# -*- coding: utf-8 -*-
+
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class HelsinkiIE(InfoExtractor):
+    IE_DESC = 'helsinki.fi'
+    _VALID_URL = r'https?://video\.helsinki\.fi/Arkisto/flash\.php\?id=(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://video.helsinki.fi/Arkisto/flash.php?id=20258',
+        'info_dict': {
+            'id': '20258',
+            'ext': 'mp4',
+            'title': 'Tietotekniikkafoorumi-iltapäivä',
+            'description': 'md5:f5c904224d43c133225130fe156a5ee0',
+        },
+        'params': {
+            'skip_download': True,  # RTMP
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        webpage = self._download_webpage(url, video_id)
+        formats = []
+
+        mobj = re.search(r'file=((\w+):[^&]+)', webpage)
+        if mobj:
+            formats.append({
+                'ext': mobj.group(2),
+                'play_path': mobj.group(1),
+                'url': 'rtmp://flashvideo.it.helsinki.fi/vod/',
+                'player_url': 'http://video.helsinki.fi/player.swf',
+                'format_note': 'sd',
+                'quality': 0,
+            })
+
+        mobj = re.search(r'hd\.file=((\w+):[^&]+)', webpage)
+        if mobj:
+            formats.append({
+                'ext': mobj.group(2),
+                'play_path': mobj.group(1),
+                'url': 'rtmp://flashvideo.it.helsinki.fi/vod/',
+                'player_url': 'http://video.helsinki.fi/player.swf',
+                'format_note': 'hd',
+                'quality': 1,
+            })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage).replace('Video: ', ''),
+            'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'formats': formats,
+        }

youtube_dl/extractor/hotnewhiphop.py

@@ -1,17 +1,25 @@
+from __future__ import unicode_literals
+
 import re
 import base64
 
 from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_parse,
+    compat_urllib_request,
+    ExtractorError,
+    HEADRequest,
+)
 
 
 class HotNewHipHopIE(InfoExtractor):
-    _VALID_URL = r'http://www\.hotnewhiphop.com/.*\.(?P<id>.*)\.html'
+    _VALID_URL = r'http://www\.hotnewhiphop\.com/.*\.(?P<id>.*)\.html'
     _TEST = {
-        u'url': u"http://www.hotnewhiphop.com/freddie-gibbs-lay-it-down-song.1435540.html",
+        'url': 'http://www.hotnewhiphop.com/freddie-gibbs-lay-it-down-song.1435540.html',
-        u'file': u'1435540.mp3',
+        'file': '1435540.mp3',
-        u'md5': u'2c2cd2f76ef11a9b3b581e8b232f3d96',
+        'md5': '2c2cd2f76ef11a9b3b581e8b232f3d96',
-        u'info_dict': {
+        'info_dict': {
-            u"title": u'Freddie Gibbs "Lay It Down"'
+            'title': 'Freddie Gibbs - Lay It Down'
         }
     }
 
@@ -21,24 +29,41 @@ class HotNewHipHopIE(InfoExtractor):
 
         webpage_src = self._download_webpage(url, video_id)
 
-        video_url_base64 = self._search_regex(r'data-path="(.*?)"',
+        video_url_base64 = self._search_regex(
-            webpage_src, u'video URL', fatal=False)
+            r'data-path="(.*?)"', webpage_src, u'video URL', fatal=False)
 
-        if video_url_base64 == None:
+        if video_url_base64 is None:
-            video_url = self._search_regex(r'"contentUrl" content="(.*?)"', webpage_src,
+            video_url = self._search_regex(
-                u'video URL')
+                r'"contentUrl" content="(.*?)"', webpage_src, u'video URL')
             return self.url_result(video_url, ie='Youtube')
 
-        video_url = base64.b64decode(video_url_base64).decode('utf-8')
+        reqdata = compat_urllib_parse.urlencode([
+            ('mediaType', 's'),
+            ('mediaId', video_id),
+        ])
+        r = compat_urllib_request.Request(
+            'http://www.hotnewhiphop.com/ajax/media/getActions/', data=reqdata)
+        r.add_header('Content-Type', 'application/x-www-form-urlencoded')
+        mkd = self._download_json(
+            r, video_id, note='Requesting media key',
+            errnote='Could not download media key')
+        if 'mediaKey' not in mkd:
+            raise ExtractorError('Did not get a media key')
 
-        video_title = self._html_search_regex(r"<title>(.*)</title>",
+        redirect_url = base64.b64decode(video_url_base64).decode('utf-8')
-            webpage_src, u'title')
+        redirect_req = HEADRequest(redirect_url)
+        req = self._request_webpage(
+            redirect_req, video_id,
+            note='Resolving final URL', errnote='Could not resolve final URL')
+        video_url = req.geturl()
+        if video_url.endswith('.html'):
+            raise ExtractorError('Redirect failed')
 
-        results = [{
+        video_title = self._og_search_title(webpage_src).strip()
+
+        return {
             'id': video_id,
             'url': video_url,
             'title': video_title,
             'thumbnail': self._og_search_thumbnail(webpage_src),
-                    'ext' : 'mp3',
+        }
-                    }]
-        return results

youtube_dl/extractor/howcast.py

@@ -1,17 +1,20 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 
 
 class HowcastIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://)?(?:www\.)?howcast\.com/videos/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?howcast\.com/videos/(?P<id>\d+)'
     _TEST = {
-        u'url': u'http://www.howcast.com/videos/390161-How-to-Tie-a-Square-Knot-Properly',
+        'url': 'http://www.howcast.com/videos/390161-How-to-Tie-a-Square-Knot-Properly',
-        u'file': u'390161.mp4',
+        'md5': '8b743df908c42f60cf6496586c7f12c3',
-        u'md5': u'8b743df908c42f60cf6496586c7f12c3',
+        'info_dict': {
-        u'info_dict': {
+            'id': '390161',
-            u"description": u"The square knot, also known as the reef knot, is one of the oldest, most basic knots to tie, and can be used in many different ways. Here's the proper way to tie a square knot.", 
+            'ext': 'mp4',
-            u"title": u"How to Tie a Square Knot Properly"
+            'description': 'The square knot, also known as the reef knot, is one of the oldest, most basic knots to tie, and can be used in many different ways. Here\'s the proper way to tie a square knot.', 
+            'title': 'How to Tie a Square Knot Properly',
         }
     }
 
@@ -24,22 +27,15 @@ class HowcastIE(InfoExtractor):
         self.report_extraction(video_id)
 
         video_url = self._search_regex(r'\'?file\'?: "(http://mobile-media\.howcast\.com/[0-9]+\.mp4)',
-            webpage, u'video URL')
+            webpage, 'video URL')
-
-        video_title = self._html_search_regex(r'<meta content=(?:"([^"]+)"|\'([^\']+)\') property=\'og:title\'',
-            webpage, u'title')
 
         video_description = self._html_search_regex(r'<meta content=(?:"([^"]+)"|\'([^\']+)\') name=\'description\'',
-            webpage, u'description', fatal=False)
+            webpage, 'description', fatal=False)
 
-        thumbnail = self._html_search_regex(r'<meta content=\'(.+?)\' property=\'og:image\'',
+        return {
-            webpage, u'thumbnail', fatal=False)
-
-        return [{
             'id': video_id,
             'url': video_url,
-            'ext':      'mp4',
+            'title': self._og_search_title(webpage),
-            'title':    video_title,
             'description': video_description,
-            'thumbnail': thumbnail,
+            'thumbnail': self._og_search_thumbnail(webpage),
-        }]
+        }

youtube_dl/extractor/huffpost.py

@@ -0,0 +1,82 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    parse_duration,
+    unified_strdate,
+)
+
+
+class HuffPostIE(InfoExtractor):
+    IE_DESC = 'Huffington Post'
+    _VALID_URL = r'''(?x)
+        https?://(embed\.)?live\.huffingtonpost\.com/
+        (?:
+            r/segment/[^/]+/|
+            HPLEmbedPlayer/\?segmentId=
+        )
+        (?P<id>[0-9a-f]+)'''
+
+    _TEST = {
+        'url': 'http://live.huffingtonpost.com/r/segment/legalese-it/52dd3e4b02a7602131000677',
+        'file': '52dd3e4b02a7602131000677.mp4',
+        'md5': '55f5e8981c1c80a64706a44b74833de8',
+        'info_dict': {
+            'title': 'Legalese It! with @MikeSacksHP',
+            'description': 'This week on Legalese It, Mike talks to David Bosco about his new book on the ICC, "Rough Justice," he also discusses the Virginia AG\'s historic stance on gay marriage, the execution of Edgar Tamayo, the ICC\'s delay of Kenya\'s President and more.  ',
+            'duration': 1549,
+            'upload_date': '20140124',
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        api_url = 'http://embed.live.huffingtonpost.com/api/segments/%s.json' % video_id
+        data = self._download_json(api_url, video_id)['data']
+
+        video_title = data['title']
+        duration = parse_duration(data['running_time'])
+        upload_date = unified_strdate(data['schedule']['starts_at'])
+        description = data.get('description')
+
+        thumbnails = []
+        for url in data['images'].values():
+            m = re.match('.*-([0-9]+x[0-9]+)\.', url)
+            if not m:
+                continue
+            thumbnails.append({
+                'url': url,
+                'resolution': m.group(1),
+            })
+
+        formats = [{
+            'format': key,
+            'format_id': key.replace('/', '.'),
+            'ext': 'mp4',
+            'url': url,
+            'vcodec': 'none' if key.startswith('audio/') else None,
+        } for key, url in data['sources']['live'].items()]
+        if data.get('fivemin_id'):
+            fid = data['fivemin_id']
+            fcat = str(int(fid) // 100 + 1)
+            furl = 'http://avideos.5min.com/2/' + fcat[-3:] + '/' + fcat + '/' + fid + '.mp4'
+            formats.append({
+                'format': 'fivemin',
+                'url': furl,
+                'preference': 1,
+            })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'description': description,
+            'formats': formats,
+            'duration': duration,
+            'upload_date': upload_date,
+            'thumbnails': thumbnails,
+        }

youtube_dl/extractor/imdb.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 import json
 
@@ -9,25 +11,25 @@ from ..utils import (
 
 
 class ImdbIE(InfoExtractor):
-    IE_NAME = u'imdb'
+    IE_NAME = 'imdb'
-    IE_DESC = u'Internet Movie Database trailers'
+    IE_DESC = 'Internet Movie Database trailers'
-    _VALID_URL = r'http://www\.imdb\.com/video/imdb/vi(?P<id>\d+)'
+    _VALID_URL = r'http://(?:www|m)\.imdb\.com/video/imdb/vi(?P<id>\d+)'
 
     _TEST = {
-        u'url': u'http://www.imdb.com/video/imdb/vi2524815897',
+        'url': 'http://www.imdb.com/video/imdb/vi2524815897',
-        u'md5': u'9f34fa777ade3a6e57a054fdbcb3a068',
+        'md5': '9f34fa777ade3a6e57a054fdbcb3a068',
-        u'info_dict': {
+        'info_dict': {
-            u'id': u'2524815897',
+            'id': '2524815897',
-            u'ext': u'mp4',
+            'ext': 'mp4',
-            u'title': u'Ice Age: Continental Drift Trailer (No. 2) - IMDb',
+            'title': 'Ice Age: Continental Drift Trailer (No. 2) - IMDb',
-            u'description': u'md5:9061c2219254e5d14e03c25c98e96a81',
+            'description': 'md5:9061c2219254e5d14e03c25c98e96a81',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
-        webpage = self._download_webpage(url,video_id)
+        webpage = self._download_webpage('http://www.imdb.com/video/imdb/vi%s' % video_id, video_id)
         descr = get_element_by_attribute('itemprop', 'description', webpage)
         available_formats = re.findall(
             r'case \'(?P<f_id>.*?)\' :$\s+url = \'(?P<path>.*?)\'', webpage,
@@ -37,10 +39,10 @@ class ImdbIE(InfoExtractor):
             f_path = f_path.strip()
             format_page = self._download_webpage(
                 compat_urlparse.urljoin(url, f_path),
-                u'Downloading info for %s format' % f_id)
+                'Downloading info for %s format' % f_id)
             json_data = self._search_regex(
                 r'<script[^>]+class="imdb-player-data"[^>]*?>(.*?)</script>',
-                format_page, u'json data', flags=re.DOTALL)
+                format_page, 'json data', flags=re.DOTALL)
             info = json.loads(json_data)
             format_info = info['videoPlayerObject']['video']
             formats.append({
@@ -55,3 +57,23 @@ class ImdbIE(InfoExtractor):
             'description': descr,
             'thumbnail': format_info['slate'],
         }
+
+
+class ImdbListIE(InfoExtractor):
+    IE_NAME = 'imdb:list'
+    IE_DESC = 'Internet Movie Database lists'
+    _VALID_URL = r'http://www\.imdb\.com/list/(?P<id>[\da-zA-Z_-]{11})'
+    
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        list_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, list_id)
+        entries = [
+            self.url_result('http://www.imdb.com' + m, 'Imdb')
+            for m in re.findall(r'href="(/video/imdb/vi[^"]+)"\s+data-type="playlist"', webpage)]
+
+        list_title = self._html_search_regex(
+            r'<h1 class="header">(.*?)</h1>', webpage, 'list title')
+
+        return self.playlist_result(entries, list_id, list_title)

youtube_dl/extractor/ina.py

@@ -1,17 +1,20 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 
 
 class InaIE(InfoExtractor):
-    """Information Extractor for Ina.fr"""
+    _VALID_URL = r'http://(?:www\.)?ina\.fr/video/(?P<id>I?[A-Z0-9]+)'
-    _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I?[A-F0-9]+)/.*'
     _TEST = {
-        u'url': u'www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
+        'url': 'http://www.ina.fr/video/I12055569/francois-hollande-je-crois-que-c-est-clair-video.html',
-        u'file': u'I12055569.mp4',
+        'md5': 'a667021bf2b41f8dc6049479d9bb38a3',
-        u'md5': u'a667021bf2b41f8dc6049479d9bb38a3',
+        'info_dict': {
-        u'info_dict': {
+            'id': 'I12055569',
-            u"title": u"Fran\u00e7ois Hollande \"Je crois que c'est clair\""
+            'ext': 'mp4',
+            'title': 'François Hollande "Je crois que c\'est clair"',
         }
     }
 
@@ -20,20 +23,14 @@ class InaIE(InfoExtractor):
 
         video_id = mobj.group('id')
         mrss_url = 'http://player.ina.fr/notices/%s.mrss' % video_id
-        video_extension = 'mp4'
+        info_doc = self._download_xml(mrss_url, video_id)
-        webpage = self._download_webpage(mrss_url, video_id)
 
         self.report_extraction(video_id)
 
-        video_url = self._html_search_regex(r'<media:player url="(?P<mp4url>http://mp4.ina.fr/[^"]+\.mp4)',
+        video_url = info_doc.find('.//{http://search.yahoo.com/mrss/}player').attrib['url']
-            webpage, u'video URL')
 
-        video_title = self._search_regex(r'<title><!\[CDATA\[(?P<titre>.*?)]]></title>',
+        return {
-            webpage, u'title')
-
-        return [{
             'id': video_id,
             'url': video_url,
-            'ext':      video_extension,
+            'title': info_doc.find('.//title').text,
-            'title':    video_title,
+        }
-        }]

youtube_dl/extractor/infoq.py

@@ -1,62 +1,55 @@
+from __future__ import unicode_literals
+
 import base64
 import re
 
 from .common import InfoExtractor
 from ..utils import (
     compat_urllib_parse,
-
-    ExtractorError,
 )
 
 
 class InfoQIE(InfoExtractor):
-    _VALID_URL = r'^(?:https?://)?(?:www\.)?infoq\.com/[^/]+/[^/]+$'
+    _VALID_URL = r'https?://(?:www\.)?infoq\.com/[^/]+/(?P<id>[^/]+)$'
     _TEST = {
-        u"name": u"InfoQ",
+        "name": "InfoQ",
-        u"url": u"http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things",
+        "url": "http://www.infoq.com/presentations/A-Few-of-My-Favorite-Python-Things",
-        u"file": u"12-jan-pythonthings.mp4",
+        "file": "12-jan-pythonthings.mp4",
-        u"info_dict": {
+        "info_dict": {
-            u"description": u"Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.",
+            "description": "Mike Pirnat presents some tips and tricks, standard libraries and third party packages that make programming in Python a richer experience.",
-            u"title": u"A Few of My Favorite [Python] Things"
+            "title": "A Few of My Favorite [Python] Things",
+        },
+        "params": {
+            "skip_download": True,
         },
-        u"params": {
-            u"skip_download": True
-        }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
 
-        webpage = self._download_webpage(url, video_id=url)
+        webpage = self._download_webpage(url, video_id)
-        self.report_extraction(url)
 
         # Extract video URL
-        mobj = re.search(r"jsclassref ?= ?'([^']*)'", webpage)
+        encoded_id = self._search_regex(r"jsclassref ?= ?'([^']*)'", webpage, 'encoded id')
-        if mobj is None:
+        real_id = compat_urllib_parse.unquote(base64.b64decode(encoded_id.encode('ascii')).decode('utf-8'))
-            raise ExtractorError(u'Unable to extract video url')
-        real_id = compat_urllib_parse.unquote(base64.b64decode(mobj.group(1).encode('ascii')).decode('utf-8'))
         video_url = 'rtmpe://video.infoq.com/cfx/st/' + real_id
 
         # Extract title
         video_title = self._search_regex(r'contentTitle = "(.*?)";',
-            webpage, u'title')
+            webpage, 'title')
 
         # Extract description
         video_description = self._html_search_regex(r'<meta name="description" content="(.*)"(?:\s*/)?>',
-            webpage, u'description', fatal=False)
+            webpage, 'description', fatal=False)
 
         video_filename = video_url.split('/')[-1]
         video_id, extension = video_filename.split('.')
 
-        info = {
+        return {
             'id': video_id,
             'url': video_url,
-            'uploader': None,
-            'upload_date': None,
             'title': video_title,
             'ext': extension,  # Extension is always(?) mp4, but seems to be flv
-            'thumbnail': None,
             'description': video_description,
         }
-
-        return [info]

youtube_dl/extractor/instagram.py

@@ -1,35 +1,39 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
 
+
 class InstagramIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?instagram\.com/p/(.*?)/'
+    _VALID_URL = r'http://instagram\.com/p/(?P<id>.*?)/'
     _TEST = {
-        u'url': u'http://instagram.com/p/aye83DjauH/?foo=bar#abc',
+        'url': 'http://instagram.com/p/aye83DjauH/?foo=bar#abc',
-        u'file': u'aye83DjauH.mp4',
+        'md5': '0d2da106a9d2631273e192b372806516',
-        u'md5': u'0d2da106a9d2631273e192b372806516',
+        'info_dict': {
-        u'info_dict': {
+            'id': 'aye83DjauH',
-            u"uploader_id": u"naomipq", 
+            'ext': 'mp4',
-            u"title": u"Video by naomipq",
+            'uploader_id': 'naomipq',
-            u'description': u'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
+            'title': 'Video by naomipq',
+            'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
         }
     }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(1)
+        video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
         uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
-            webpage, u'uploader id', fatal=False)
+            webpage, 'uploader id', fatal=False)
-        desc = self._search_regex(r'"caption":"(.*?)"', webpage, u'description',
+        desc = self._search_regex(r'"caption":"(.*?)"', webpage, 'description',
             fatal=False)
 
-        return [{
+        return {
             'id': video_id,
             'url': self._og_search_video_url(webpage, secure=False),
             'ext': 'mp4',
-            'title':     u'Video by %s' % uploader_id,
+            'title': 'Video by %s' % uploader_id,
             'thumbnail': self._og_search_thumbnail(webpage),
             'uploader_id': uploader_id,
             'description': desc,
-        }]
+        }

youtube_dl/extractor/internetvideoarchive.py

@@ -5,7 +5,6 @@ from ..utils import (
     compat_urlparse,
     compat_urllib_parse,
     xpath_with_ns,
-    determine_ext,
 )
 
 
@@ -63,13 +62,17 @@ class InternetVideoArchiveIE(InfoExtractor):
         for content in item.findall(_bp('media:group/media:content')):
             attr = content.attrib
             f_url = attr['url']
+            width = int(attr['width'])
+            bitrate = int(attr['bitrate'])
+            format_id = '%d-%dk' % (width, bitrate)
             formats.append({
+                'format_id': format_id,
                 'url': f_url,
-                'ext': determine_ext(f_url),
+                'width': width,
-                'width': int(attr['width']),
+                'tbr': bitrate,
-                'bitrate': int(attr['bitrate']),
             })
-        formats = sorted(formats, key=lambda f: f['bitrate'])
+
+        self._sort_formats(formats)
 
         return {
             'id': video_id,

youtube_dl/extractor/iprima.py

@@ -0,0 +1,85 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
+import re
+from random import random
+from math import floor
+
+from .common import InfoExtractor
+from ..utils import compat_urllib_request
+
+
+class IPrimaIE(InfoExtractor):
+    _VALID_URL = r'https?://play\.iprima\.cz/(?P<videogroup>.+)/(?P<videoid>.+)'
+
+    _TESTS = [{
+        'url': 'http://play.iprima.cz/particka/particka-92',
+        'info_dict': {
+            'id': '39152',
+            'ext': 'flv',
+            'title': 'Partička (92)',
+            'description': 'md5:3740fda51464da35a2d4d0670b8e4fd6',
+            'thumbnail': 'http://play.iprima.cz/sites/default/files/image_crops/image_620x349/3/491483_particka-92_image_620x349.jpg',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('videoid')
+
+        webpage = self._download_webpage(url, video_id)
+
+        player_url = 'http://embed.livebox.cz/iprimaplay/player-embed-v2.js?__tok%s__=%s' % (
+                         floor(random()*1073741824),
+                         floor(random()*1073741824))
+
+        req = compat_urllib_request.Request(player_url)
+        req.add_header('Referer', url)
+        playerpage = self._download_webpage(req, video_id)
+
+        base_url = ''.join(re.findall(r"embed\['stream'\] = '(.+?)'.+'(\?auth=)'.+'(.+?)';", playerpage)[1])
+
+        zoneGEO = self._html_search_regex(r'"zoneGEO":(.+?),', webpage, 'zoneGEO')
+
+        if zoneGEO != '0':
+            base_url = base_url.replace('token', 'token_'+zoneGEO)
+
+        formats = []
+        for format_id in ['lq', 'hq', 'hd']:
+            filename = self._html_search_regex(r'"%s_id":(.+?),' % format_id, webpage, 'filename')
+
+            if filename == 'null':
+                continue
+
+            real_id = self._search_regex(r'Prima-[0-9]{10}-([0-9]+)_', filename, 'real video id')
+
+            if format_id == 'lq':
+                quality = 0
+            elif format_id == 'hq':
+                quality = 1
+            elif format_id == 'hd':
+                quality = 2
+                filename = 'hq/'+filename
+
+            formats.append({
+                'format_id': format_id,
+                'url': base_url,
+                'quality': quality,
+                'play_path': 'mp4:'+filename.replace('"', '')[:-4],
+                'rtmp_live': True,
+                'ext': 'flv',
+            })
+
+        self._sort_formats(formats)
+
+        return {
+            'id': real_id,
+            'title': self._og_search_title(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'formats': formats,
+            'description': self._og_search_description(webpage),
+        }

youtube_dl/extractor/ivi.py

@@ -0,0 +1,160 @@
+# encoding: utf-8
+from __future__ import unicode_literals
+
+import re
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    compat_urllib_request,
+    ExtractorError,
+)
+
+
+class IviIE(InfoExtractor):
+    IE_DESC = 'ivi.ru'
+    IE_NAME = 'ivi'
+    _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch(?:/(?P<compilationid>[^/]+))?/(?P<videoid>\d+)'
+
+    _TESTS = [
+        # Single movie
+        {
+            'url': 'http://www.ivi.ru/watch/53141',
+            'md5': '6ff5be2254e796ed346251d117196cf4',
+            'info_dict': {
+                'id': '53141',
+                'ext': 'mp4',
+                'title': 'Иван Васильевич меняет профессию',
+                'description': 'md5:b924063ea1677c8fe343d8a72ac2195f',
+                'duration': 5498,
+                'thumbnail': 'http://thumbs.ivi.ru/f20.vcp.digitalaccess.ru/contents/d/1/c3c885163a082c29bceeb7b5a267a6.jpg',
+            },
+            'skip': 'Only works from Russia',
+        },
+        # Serial's serie
+        {
+            'url': 'http://www.ivi.ru/watch/dezhurnyi_angel/74791',
+            'md5': '3e6cc9a848c1d2ebcc6476444967baa9',
+            'info_dict': {
+                'id': '74791',
+                'ext': 'mp4',
+                'title': 'Дежурный ангел - 1 серия',
+                'duration': 2490,
+                'thumbnail': 'http://thumbs.ivi.ru/f7.vcp.digitalaccess.ru/contents/8/e/bc2f6c2b6e5d291152fdd32c059141.jpg',
+            },
+            'skip': 'Only works from Russia',
+         }
+    ]
+
+    # Sorted by quality
+    _known_formats = ['MP4-low-mobile', 'MP4-mobile', 'FLV-lo', 'MP4-lo', 'FLV-hi', 'MP4-hi', 'MP4-SHQ']
+
+    # Sorted by size
+    _known_thumbnails = ['Thumb-120x90', 'Thumb-160', 'Thumb-640x480']
+
+    def _extract_description(self, html):
+        m = re.search(r'<meta name="description" content="(?P<description>[^"]+)"/>', html)
+        return m.group('description') if m is not None else None
+
+    def _extract_comment_count(self, html):
+        m = re.search('(?s)<a href="#" id="view-comments" class="action-button dim gradient">\s*Комментарии:\s*(?P<commentcount>\d+)\s*</a>', html)
+        return int(m.group('commentcount')) if m is not None else 0
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('videoid')
+
+        api_url = 'http://api.digitalaccess.ru/api/json/'
+
+        data = {'method': 'da.content.get',
+                'params': [video_id, {'site': 's183',
+                                      'referrer': 'http://www.ivi.ru/watch/%s' % video_id,
+                                      'contentid': video_id
+                                      }
+                           ]
+                }
+
+        request = compat_urllib_request.Request(api_url, json.dumps(data))
+
+        video_json_page = self._download_webpage(request, video_id, 'Downloading video JSON')
+        video_json = json.loads(video_json_page)
+
+        if 'error' in video_json:
+            error = video_json['error']
+            if error['origin'] == 'NoRedisValidData':
+                raise ExtractorError('Video %s does not exist' % video_id, expected=True)
+            raise ExtractorError('Unable to download video %s: %s' % (video_id, error['message']), expected=True)
+
+        result = video_json['result']
+
+        formats = [{
+            'url': x['url'],
+            'format_id': x['content_format'],
+            'preference': self._known_formats.index(x['content_format']),
+        } for x in result['files'] if x['content_format'] in self._known_formats]
+
+        self._sort_formats(formats)
+
+        if not formats:
+            raise ExtractorError('No media links available for %s' % video_id)
+
+        duration = result['duration']
+        compilation = result['compilation']
+        title = result['title']
+
+        title = '%s - %s' % (compilation, title) if compilation is not None else title  
+
+        previews = result['preview']
+        previews.sort(key=lambda fmt: self._known_thumbnails.index(fmt['content_format']))
+        thumbnail = previews[-1]['url'] if len(previews) > 0 else None
+
+        video_page = self._download_webpage(url, video_id, 'Downloading video page')
+        description = self._extract_description(video_page)
+        comment_count = self._extract_comment_count(video_page)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'description': description,
+            'duration': duration,
+            'comment_count': comment_count,
+            'formats': formats,
+        }
+
+
+class IviCompilationIE(InfoExtractor):
+    IE_DESC = 'ivi.ru compilations'
+    IE_NAME = 'ivi:compilation'
+    _VALID_URL = r'https?://(?:www\.)?ivi\.ru/watch/(?!\d+)(?P<compilationid>[a-z\d_-]+)(?:/season(?P<seasonid>\d+))?$'
+
+    def _extract_entries(self, html, compilation_id):
+        return [self.url_result('http://www.ivi.ru/watch/%s/%s' % (compilation_id, serie), 'Ivi')
+                for serie in re.findall(r'<strong><a href="/watch/%s/(\d+)">(?:[^<]+)</a></strong>' % compilation_id, html)]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        compilation_id = mobj.group('compilationid')
+        season_id = mobj.group('seasonid')
+
+        if season_id is not None: # Season link
+            season_page = self._download_webpage(url, compilation_id, 'Downloading season %s web page' % season_id)
+            playlist_id = '%s/season%s' % (compilation_id, season_id)
+            playlist_title = self._html_search_meta('title', season_page, 'title')
+            entries = self._extract_entries(season_page, compilation_id)
+        else: # Compilation link            
+            compilation_page = self._download_webpage(url, compilation_id, 'Downloading compilation web page')
+            playlist_id = compilation_id
+            playlist_title = self._html_search_meta('title', compilation_page, 'title')
+            seasons = re.findall(r'<a href="/watch/%s/season(\d+)">[^<]+</a>' % compilation_id, compilation_page)
+            if len(seasons) == 0: # No seasons in this compilation
+                entries = self._extract_entries(compilation_page, compilation_id)
+            else:
+                entries = []
+                for season_id in seasons:
+                    season_page = self._download_webpage(
+                        'http://www.ivi.ru/watch/%s/season%s' % (compilation_id, season_id),
+                        compilation_id, 'Downloading season %s web page' % season_id)
+                    entries.extend(self._extract_entries(season_page, compilation_id))
+
+        return self.playlist_result(entries, playlist_id, playlist_title)

youtube_dl/extractor/jadorecettepub.py

@@ -0,0 +1,48 @@
+# coding: utf-8
+
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from .youtube import YoutubeIE
+
+
+class JadoreCettePubIE(InfoExtractor):
+    _VALID_URL = r'http://(?:www\.)?jadorecettepub\.com/[0-9]{4}/[0-9]{2}/(?P<id>.*?)\.html'
+
+    _TEST = {
+        'url': 'http://www.jadorecettepub.com/2010/12/star-wars-massacre-par-les-japonais.html',
+        'md5': '401286a06067c70b44076044b66515de',
+        'info_dict': {
+            'id': 'jLMja3tr7a4',
+            'ext': 'mp4',
+            'title': 'La pire utilisation de Star Wars',
+            'description': "Jadorecettepub.com vous a gratifié de plusieurs pubs géniales utilisant Star Wars et Dark Vador plus particulièrement... Mais l'heure est venue de vous proposer une version totalement massacrée, venue du Japon.  Quand les Japonais détruisent l'image de Star Wars pour vendre du thon en boite, ça promet...",
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        title = self._html_search_regex(
+            r'<span style="font-size: x-large;"><b>(.*?)</b></span>',
+            webpage, 'title')
+        description = self._html_search_regex(
+            r'(?s)<div id="fb-root">(.*?)<script>', webpage, 'description',
+            fatal=False)
+        real_url = self._search_regex(
+            r'\[/postlink\](.*)endofvid', webpage, 'video URL')
+        video_id = YoutubeIE.extract_id(real_url)
+
+        return {
+            '_type': 'url_transparent',
+            'url': real_url,
+            'id': video_id,
+            'title': title,
+            'description': description,
+        }
+

youtube_dl/extractor/jeuxvideo.py

@@ -1,5 +1,7 @@
 # coding: utf-8
 
+from __future__ import unicode_literals
+
 import json
 import re
 
@@ -10,12 +12,13 @@ class JeuxVideoIE(InfoExtractor):
     _VALID_URL = r'http://.*?\.jeuxvideo\.com/.*/(.*?)-\d+\.htm'
 
     _TEST = {
-        u'url': u'http://www.jeuxvideo.com/reportages-videos-jeux/0004/00046170/tearaway-playstation-vita-gc-2013-tearaway-nous-presente-ses-papiers-d-identite-00115182.htm',
+        'url': 'http://www.jeuxvideo.com/reportages-videos-jeux/0004/00046170/tearaway-playstation-vita-gc-2013-tearaway-nous-presente-ses-papiers-d-identite-00115182.htm',
-        u'file': u'5182.mp4',
+        'md5': '046e491afb32a8aaac1f44dd4ddd54ee',
-        u'md5': u'046e491afb32a8aaac1f44dd4ddd54ee',
+        'info_dict': {
-        u'info_dict': {
+            'id': '5182',
-            u'title': u'GC 2013 : Tearaway nous présente ses papiers d\'identité',
+            'ext': 'mp4',
-            u'description': u'Lorsque les développeurs de LittleBigPlanet proposent un nouveau titre, on ne peut que s\'attendre à un résultat original et fort attrayant.\n',
+            'title': 'GC 2013 : Tearaway nous présente ses papiers d\'identité',
+            'description': 'Lorsque les développeurs de LittleBigPlanet proposent un nouveau titre, on ne peut que s\'attendre à un résultat original et fort attrayant.\n',
         },
     }
 
@@ -25,14 +28,14 @@ class JeuxVideoIE(InfoExtractor):
         webpage = self._download_webpage(url, title)
         xml_link = self._html_search_regex(
             r'<param name="flashvars" value="config=(.*?)" />',
-            webpage, u'config URL')
+            webpage, 'config URL')
         
         video_id = self._search_regex(
             r'http://www\.jeuxvideo\.com/config/\w+/\d+/(.*?)/\d+_player\.xml',
-            xml_link, u'video ID')
+            xml_link, 'video ID')
 
         config = self._download_xml(
-            xml_link, title, u'Downloading XML config')
+            xml_link, title, 'Downloading XML config')
         info_json = config.find('format.json').text
         info = json.loads(info_json)['versions'][0]
         

youtube_dl/extractor/jpopsukitv.py

@@ -0,0 +1,73 @@
+# coding=utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    unified_strdate,
+)
+
+
+class JpopsukiIE(InfoExtractor):
+    IE_NAME = 'jpopsuki.tv'
+    _VALID_URL = r'https?://(?:www\.)?jpopsuki\.tv/video/(.*?)/(?P<id>\S+)'
+
+    _TEST = {
+        'url': 'http://www.jpopsuki.tv/video/ayumi-hamasaki---evolution/00be659d23b0b40508169cdee4545771',
+        'md5': '88018c0c1a9b1387940e90ec9e7e198e',
+        'file': '00be659d23b0b40508169cdee4545771.mp4',
+        'info_dict': {
+            'id': '00be659d23b0b40508169cdee4545771',
+            'title': 'ayumi hamasaki - evolution',
+            'description': 'Release date: 2001.01.31\r\n浜崎あゆみ - evolution',
+            'thumbnail': 'http://www.jpopsuki.tv/cache/89722c74d2a2ebe58bcac65321c115b2.jpg',
+            'uploader': 'plama_chan',
+            'uploader_id': '404',
+            'upload_date': '20121101'
+        }
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_url = 'http://www.jpopsuki.tv' + self._html_search_regex(
+            r'<source src="(.*?)" type', webpage, 'video url')
+
+        video_title = self._og_search_title(webpage)
+        description = self._og_search_description(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+        uploader = self._html_search_regex(
+            r'<li>from: <a href="/user/view/user/(.*?)/uid/',
+            webpage, 'video uploader', fatal=False)
+        uploader_id = self._html_search_regex(
+            r'<li>from: <a href="/user/view/user/\S*?/uid/(\d*)',
+            webpage, 'video uploader_id', fatal=False)
+        upload_date = self._html_search_regex(
+            r'<li>uploaded: (.*?)</li>', webpage, 'video upload_date',
+            fatal=False)
+        if upload_date is not None:
+            upload_date = unified_strdate(upload_date)
+        view_count_str = self._html_search_regex(
+            r'<li>Hits: ([0-9]+?)</li>', webpage, 'video view_count',
+            fatal=False)
+        comment_count_str = self._html_search_regex(
+            r'<h2>([0-9]+?) comments</h2>', webpage, 'video comment_count',
+            fatal=False)
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': video_title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'upload_date': upload_date,
+            'view_count': int_or_none(view_count_str),
+            'comment_count': int_or_none(comment_count_str),
+        }

youtube_dl/extractor/kankan.py

@@ -1,21 +1,24 @@
+from __future__ import unicode_literals
+
 import re
 import hashlib
 
 from .common import InfoExtractor
-from ..utils import determine_ext
 
 _md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
 
+
 class KankanIE(InfoExtractor):
     _VALID_URL = r'https?://(?:.*?\.)?kankan\.com/.+?/(?P<id>\d+)\.shtml'
     
     _TEST = {
-        u'url': u'http://yinyue.kankan.com/vod/48/48863.shtml',
+        'url': 'http://yinyue.kankan.com/vod/48/48863.shtml',
-        u'file': u'48863.flv',
+        'file': '48863.flv',
-        u'md5': u'29aca1e47ae68fc28804aca89f29507e',
+        'md5': '29aca1e47ae68fc28804aca89f29507e',
-        u'info_dict': {
+        'info_dict': {
-            u'title': u'Ready To Go',
+            'title': 'Ready To Go',
         },
+        'skip': 'Only available from China',
     }
 
     def _real_extract(self, url):
@@ -23,22 +26,23 @@ class KankanIE(InfoExtractor):
         video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
 
-        title = self._search_regex(r'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, u'video title')
+        title = self._search_regex(r'(?:G_TITLE=|G_MOVIE_TITLE = )[\'"](.+?)[\'"]', webpage, 'video title')
         surls = re.search(r'surls:\[\'.+?\'\]|lurl:\'.+?\.flv\'', webpage).group(0)
         gcids = re.findall(r"http://.+?/.+?/(.+?)/", surls)
         gcid = gcids[-1]
 
-        video_info_page = self._download_webpage('http://p2s.cl.kankan.com/getCdnresource_flv?gcid=%s' % gcid,
+        info_url = 'http://p2s.cl.kankan.com/getCdnresource_flv?gcid=%s' % gcid
-                                                 video_id, u'Downloading video url info')
+        video_info_page = self._download_webpage(
-        ip = self._search_regex(r'ip:"(.+?)"', video_info_page, u'video url ip')
+            info_url, video_id, 'Downloading video url info')
-        path = self._search_regex(r'path:"(.+?)"', video_info_page, u'video url path')
+        ip = self._search_regex(r'ip:"(.+?)"', video_info_page, 'video url ip')
-        param1 = self._search_regex(r'param1:(\d+)', video_info_page, u'param1')
+        path = self._search_regex(r'path:"(.+?)"', video_info_page, 'video url path')
-        param2 = self._search_regex(r'param2:(\d+)', video_info_page, u'param2')
+        param1 = self._search_regex(r'param1:(\d+)', video_info_page, 'param1')
+        param2 = self._search_regex(r'param2:(\d+)', video_info_page, 'param2')
         key = _md5('xl_mp43651' + param1 + param2)
         video_url = 'http://%s%s?key=%s&key1=%s' % (ip, path, key, param2)
 
-        return {'id': video_id,
+        return {
+            'id': video_id,
             'title': title,
             'url': video_url,
-                'ext': determine_ext(video_url),
         }

youtube_dl/extractor/keek.py

@@ -1,3 +1,5 @@
+from __future__ import unicode_literals
+
 import re
 
 from .common import InfoExtractor
@@ -5,36 +7,34 @@ from .common import InfoExtractor
 
 class KeekIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?keek\.com/(?:!|\w+/keeks/)(?P<videoID>\w+)'
-    IE_NAME = u'keek'
+    IE_NAME = 'keek'
     _TEST = {
-        u'url': u'https://www.keek.com/ytdl/keeks/NODfbab',
+        'url': 'https://www.keek.com/ytdl/keeks/NODfbab',
-        u'file': u'NODfbab.mp4',
+        'file': 'NODfbab.mp4',
-        u'md5': u'9b0636f8c0f7614afa4ea5e4c6e57e83',
+        'md5': '9b0636f8c0f7614afa4ea5e4c6e57e83',
-        u'info_dict': {
+        'info_dict': {
-            u"uploader": u"ytdl", 
+            'uploader': 'ytdl',
-            u"title": u"test chars: \"'/\\\u00e4<>This is a test video for youtube-dl.For more information, contact phihag@phihag.de ."
+            'title': 'test chars: "\'/\\\u00e4<>This is a test video for youtube-dl.For more information, contact phihag@phihag.de .',
-        }
+        },
     }
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
         video_id = m.group('videoID')
 
-        video_url = u'http://cdn.keek.com/keek/video/%s' % video_id
+        video_url = 'http://cdn.keek.com/keek/video/%s' % video_id
-        thumbnail = u'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id
+        thumbnail = 'http://cdn.keek.com/keek/thumbnail/%s/w100/h75' % video_id
         webpage = self._download_webpage(url, video_id)
 
-        video_title = self._og_search_title(webpage)
+        uploader = self._html_search_regex(
+            r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>',
+            webpage, 'uploader', fatal=False)
 
-        uploader = self._html_search_regex(r'<div class="user-name-and-bio">[\S\s]+?<h2>(?P<uploader>.+?)</h2>',
+        return {
-            webpage, u'uploader', fatal=False)
-
-        info = {
             'id': video_id,
             'url': video_url,
             'ext': 'mp4',
-                'title': video_title,
+            'title': self._og_search_title(webpage),
             'thumbnail': thumbnail,
             'uploader': uploader
         }
-        return [info]

youtube_dl/extractor/khanacademy.py

@@ -0,0 +1,71 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    unified_strdate,
+)
+
+
+class KhanAcademyIE(InfoExtractor):
+    _VALID_URL = r'^https?://(?:www\.)?khanacademy\.org/(?P<key>[^/]+)/(?:[^/]+/){,2}(?P<id>[^?#/]+)(?:$|[?#])'
+    IE_NAME = 'KhanAcademy'
+
+    _TEST = {
+        'url': 'http://www.khanacademy.org/video/one-time-pad',
+        'file': 'one-time-pad.mp4',
+        'md5': '7021db7f2d47d4fff89b13177cb1e8f4',
+        'info_dict': {
+            'title': 'The one-time pad',
+            'description': 'The perfect cipher',
+            'duration': 176,
+            'uploader': 'Brit Cruise',
+            'upload_date': '20120411',
+        }
+    }
+
+    def _real_extract(self, url):
+        m = re.match(self._VALID_URL, url)
+        video_id = m.group('id')
+
+        if m.group('key') == 'video':
+            data = self._download_json(
+                'http://api.khanacademy.org/api/v1/videos/' + video_id,
+                video_id, 'Downloading video info')
+
+            upload_date = unified_strdate(data['date_added'])
+            uploader = ', '.join(data['author_names'])
+            return {
+                '_type': 'url_transparent',
+                'url': data['url'],
+                'id': video_id,
+                'title': data['title'],
+                'thumbnail': data['image_url'],
+                'duration': data['duration'],
+                'description': data['description'],
+                'uploader': uploader,
+                'upload_date': upload_date,
+            }
+        else:
+            # topic
+            data = self._download_json(
+                'http://api.khanacademy.org/api/v1/topic/' + video_id,
+                video_id, 'Downloading topic info')
+
+            entries = [
+                {
+                    '_type': 'url',
+                    'url': c['url'],
+                    'id': c['id'],
+                    'title': c['title'],
+                }
+                for c in data['children'] if c['kind'] in ('Video', 'Topic')]
+
+            return {
+                '_type': 'playlist',
+                'id': video_id,
+                'title': data['title'],
+                'description': data['description'],
+                'entries': entries,
+            }