release 2015.02.28

It handles encodings better, for example for 'http://www.npo.nl/nos-journaal/14-02-2015/POW_00942207'

AUTHORS

@@ -112,3 +112,4 @@ Frans de Jonge
 Robin de Rooij
 Ryan Schmidt
 Leslie P. Polzer
+Duncan Keall

Makefile

@@ -2,6 +2,7 @@ all: youtube-dl README.md CONTRIBUTING.md README.txt youtube-dl.1 youtube-dl.bas
 
 clean:
 	rm -rf youtube-dl.1.temp.md youtube-dl.1 youtube-dl.bash-completion README.txt MANIFEST build/ dist/ .coverage cover/ youtube-dl.tar.gz youtube-dl.zsh youtube-dl.fish *.dump *.part *.info.json *.mp4 *.flv *.mp3 *.avi CONTRIBUTING.md.tmp youtube-dl youtube-dl.exe
+	find -name "*.pyc" -delete
 
 PREFIX ?= /usr/local
 BINDIR ?= $(PREFIX)/bin

README.md

@@ -139,6 +139,8 @@ which means you can modify it, redistribute it or use it however you like.
                                      dislike_count <? 50 & description" .
     --no-playlist                    If the URL refers to a video and a
                                      playlist, download only the video.
+    --yes-playlist                   If the URL refers to a video and a
+                                     playlist, download the playlist.
     --age-limit YEARS                download only videos suitable for the given
                                      age
     --download-archive FILE          Download only videos not listed in the
@@ -406,6 +408,8 @@ which means you can modify it, redistribute it or use it however you like.
                                      downloading, similar to find's -exec
                                      syntax. Example: --exec 'adb push {}
                                      /sdcard/Music/ && rm {}'
+    --convert-subtitles FORMAT       Convert the subtitles to other format
+                                     (currently supported: srt|ass|vtt)
 
 # CONFIGURATION
 

docs/supportedsites.md

@@ -17,6 +17,7 @@
  - **AdultSwim**
  - **Aftenposten**
  - **Aftonbladet**
+ - **AirMozilla**
  - **AlJazeera**
  - **Allocine**
  - **AlphaPorno**
@@ -209,6 +210,7 @@
  - **Jove**
  - **jpopsuki.tv**
  - **Jukebox**
+ - **Kaltura**
  - **Kankan**
  - **Karaoketv**
  - **keek**
@@ -220,6 +222,9 @@
  - **Ku6**
  - **la7.tv**
  - **Laola1Tv**
+ - **Letv**
+ - **LetvPlaylist**
+ - **LetvTv**
  - **lifenews**: LIFE | NEWS
  - **LiveLeak**
  - **livestream**
@@ -304,6 +309,7 @@
  - **Nuvid**
  - **NYTimes**
  - **ocw.mit.edu**
+ - **Odnoklassniki**
  - **OktoberfestTV**
  - **on.aol.com**
  - **Ooyala**
@@ -330,6 +336,7 @@
  - **PornoXO**
  - **PromptFile**
  - **prosiebensat1**: ProSiebenSat.1 Digital
+ - **Puls4**
  - **Pyvideo**
  - **QuickVid**
  - **R7**
@@ -408,7 +415,7 @@
  - **StreamCZ**
  - **StreetVoice**
  - **SunPorno**
- - **SVTPlay**
+ - **SVTPlay**: SVT Play and Öppet arkiv
  - **SWRMediathek**
  - **Syfy**
  - **SztvHu**

test/test_utils.py

@@ -246,6 +246,7 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(parse_duration('2.5 hours'), 9000)
         self.assertEqual(parse_duration('02:03:04'), 7384)
         self.assertEqual(parse_duration('01:02:03:04'), 93784)
+        self.assertEqual(parse_duration('1 hour 3 minutes'), 3780)
 
     def test_fix_xml_ampersands(self):
         self.assertEqual(

youtube_dl/YoutubeDL.py

@@ -1300,17 +1300,18 @@ class YoutubeDL(object):
             # subtitles download errors are already managed as troubles in relevant IE
             # that way it will silently go on when used with unsupporting IE
             subtitles = info_dict['requested_subtitles']
+            ie = self.get_info_extractor(info_dict['extractor_key'])
             for sub_lang, sub_info in subtitles.items():
                 sub_format = sub_info['ext']
                 if sub_info.get('data') is not None:
                     sub_data = sub_info['data']
                 else:
                     try:
-                        uf = self.urlopen(sub_info['url'])
-                        sub_data = uf.read().decode('utf-8')
-                    except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
+                        sub_data = ie._download_webpage(
+                            sub_info['url'], info_dict['id'], note=False)
+                    except ExtractorError as err:
                         self.report_warning('Unable to download subtitle for "%s": %s' %
-                                            (sub_lang, compat_str(err)))
+                                            (sub_lang, compat_str(err.cause)))
                         continue
                 try:
                     sub_filename = subtitles_filename(filename, sub_lang, sub_format)

youtube_dl/__init__.py

@@ -170,6 +170,9 @@ def _real_main(argv=None):
     if opts.recodevideo is not None:
         if opts.recodevideo not in ['mp4', 'flv', 'webm', 'ogg', 'mkv']:
             parser.error('invalid video recode format specified')
+    if opts.convertsubtitles is not None:
+        if opts.convertsubtitles not in ['srt', 'vtt', 'ass']:
+            parser.error('invalid subtitle format specified')
 
     if opts.date is not None:
         date = DateRange.day(opts.date)
@@ -223,6 +226,11 @@ def _real_main(argv=None):
             'key': 'FFmpegVideoConvertor',
             'preferedformat': opts.recodevideo,
         })
+    if opts.convertsubtitles:
+        postprocessors.append({
+            'key': 'FFmpegSubtitlesConvertor',
+            'format': opts.convertsubtitles,
+        })
     if opts.embedsubtitles:
         postprocessors.append({
             'key': 'FFmpegEmbedSubtitle',

youtube_dl/downloader/f4m.py

@@ -11,6 +11,7 @@ from .common import FileDownloader
 from .http import HttpFD
 from ..compat import (
     compat_urlparse,
+    compat_urllib_error,
 )
 from ..utils import (
     struct_pack,
@@ -121,7 +122,8 @@ class FlvReader(io.BytesIO):
 
         self.read_unsigned_int()  # BootstrapinfoVersion
         # Profile,Live,Update,Reserved
-        self.read(1)
+        flags = self.read_unsigned_char()
+        live = flags & 0x20 != 0
         # time scale
         self.read_unsigned_int()
         # CurrentMediaTime
@@ -160,6 +162,7 @@ class FlvReader(io.BytesIO):
         return {
             'segments': segments,
             'fragments': fragments,
+            'live': live,
         }
 
     def read_bootstrap_info(self):
@@ -182,6 +185,10 @@ def build_fragments_list(boot_info):
     for segment, fragments_count in segment_run_table['segment_run']:
         for _ in range(fragments_count):
             res.append((segment, next(fragments_counter)))
+
+    if boot_info['live']:
+        res = res[-2:]
+
     return res
 
 
@@ -246,6 +253,38 @@ class F4mFD(FileDownloader):
             self.report_error('Unsupported DRM')
         return media
 
+    def _get_bootstrap_from_url(self, bootstrap_url):
+        bootstrap = self.ydl.urlopen(bootstrap_url).read()
+        return read_bootstrap_info(bootstrap)
+
+    def _update_live_fragments(self, bootstrap_url, latest_fragment):
+        fragments_list = []
+        retries = 30
+        while (not fragments_list) and (retries > 0):
+            boot_info = self._get_bootstrap_from_url(bootstrap_url)
+            fragments_list = build_fragments_list(boot_info)
+            fragments_list = [f for f in fragments_list if f[1] > latest_fragment]
+            if not fragments_list:
+                # Retry after a while
+                time.sleep(5.0)
+                retries -= 1
+
+        if not fragments_list:
+            self.report_error('Failed to update fragments')
+
+        return fragments_list
+
+    def _parse_bootstrap_node(self, node, base_url):
+        if node.text is None:
+            bootstrap_url = compat_urlparse.urljoin(
+                base_url, node.attrib['url'])
+            boot_info = self._get_bootstrap_from_url(bootstrap_url)
+        else:
+            bootstrap_url = None
+            bootstrap = base64.b64decode(node.text)
+            boot_info = read_bootstrap_info(bootstrap)
+        return (boot_info, bootstrap_url)
+
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
         requested_bitrate = info_dict.get('tbr')
@@ -265,18 +304,13 @@ class F4mFD(FileDownloader):
 
         base_url = compat_urlparse.urljoin(man_url, media.attrib['url'])
         bootstrap_node = doc.find(_add_ns('bootstrapInfo'))
-        if bootstrap_node.text is None:
-            bootstrap_url = compat_urlparse.urljoin(
-                base_url, bootstrap_node.attrib['url'])
-            bootstrap = self.ydl.urlopen(bootstrap_url).read()
-        else:
-            bootstrap = base64.b64decode(bootstrap_node.text)
+        boot_info, bootstrap_url = self._parse_bootstrap_node(bootstrap_node, base_url)
+        live = boot_info['live']
         metadata_node = media.find(_add_ns('metadata'))
         if metadata_node is not None:
             metadata = base64.b64decode(metadata_node.text)
         else:
             metadata = None
-        boot_info = read_bootstrap_info(bootstrap)
 
         fragments_list = build_fragments_list(boot_info)
         if self.params.get('test', False):
@@ -301,7 +335,8 @@ class F4mFD(FileDownloader):
         (dest_stream, tmpfilename) = sanitize_open(tmpfilename, 'wb')
 
         write_flv_header(dest_stream)
-        write_metadata_tag(dest_stream, metadata)
+        if not live:
+            write_metadata_tag(dest_stream, metadata)
 
         # This dict stores the download progress, it's updated by the progress
         # hook
@@ -348,24 +383,45 @@ class F4mFD(FileDownloader):
         http_dl.add_progress_hook(frag_progress_hook)
 
         frags_filenames = []
-        for (seg_i, frag_i) in fragments_list:
+        while fragments_list:
+            seg_i, frag_i = fragments_list.pop(0)
             name = 'Seg%d-Frag%d' % (seg_i, frag_i)
             url = base_url + name
             if akamai_pv:
                 url += '?' + akamai_pv.strip(';')
             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)
+            try:
+                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
+                if live:
+                    os.remove(frag_filename)
+                else:
+                    frags_filenames.append(frag_filename)
+            except (compat_urllib_error.HTTPError, ) as err:
+                if live and (err.code == 404 or err.code == 410):
+                    # We didn't keep up with the live window. Continue
+                    # with the next available fragment.
+                    msg = 'Fragment %d unavailable' % frag_i
+                    self.report_warning(msg)
+                    fragments_list = []
+                else:
+                    raise
+
+            if not fragments_list and live and bootstrap_url:
+                fragments_list = self._update_live_fragments(bootstrap_url, frag_i)
+                total_frags += len(fragments_list)
+                if fragments_list and (fragments_list[0][1] > frag_i + 1):
+                    msg = 'Missed %d fragments' % (fragments_list[0][1] - (frag_i + 1))
+                    self.report_warning(msg)
 
         dest_stream.close()
 

youtube_dl/downloader/rtmp.py

@@ -119,7 +119,9 @@ class RtmpFD(FileDownloader):
         # 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]
+        basic_args = [
+            'rtmpdump', '--verbose', '-r', url,
+            '-o', encodeFilename(tmpfilename, True)]
         if player_url is not None:
             basic_args += ['--swfVfy', player_url]
         if page_url is not None:

youtube_dl/extractor/__init__.py

@@ -8,6 +8,7 @@ from .adobetv import AdobeTVIE
 from .adultswim import AdultSwimIE
 from .aftenposten import AftenpostenIE
 from .aftonbladet import AftonbladetIE
+from .airmozilla import AirMozillaIE
 from .aljazeera import AlJazeeraIE
 from .alphaporno import AlphaPornoIE
 from .anitube import AnitubeIE
@@ -226,6 +227,7 @@ from .jeuxvideo import JeuxVideoIE
 from .jove import JoveIE
 from .jukebox import JukeboxIE
 from .jpopsukitv import JpopsukiIE
+from .kaltura import KalturaIE
 from .kankan import KankanIE
 from .karaoketv import KaraoketvIE
 from .keezmovies import KeezMoviesIE
@@ -237,6 +239,11 @@ from .krasview import KrasViewIE
 from .ku6 import Ku6IE
 from .la7 import LA7IE
 from .laola1tv import Laola1TvIE
+from .letv import (
+    LetvIE,
+    LetvTvIE,
+    LetvPlaylistIE
+)
 from .lifenews import LifeNewsIE
 from .liveleak import LiveLeakIE
 from .livestream import (
@@ -339,6 +346,7 @@ from .ntvde import NTVDeIE
 from .ntvru import NTVRuIE
 from .nytimes import NYTimesIE
 from .nuvid import NuvidIE
+from .odnoklassniki import OdnoklassnikiIE
 from .oktoberfesttv import OktoberfestTVIE
 from .ooyala import OoyalaIE
 from .openfilm import OpenFilmIE
@@ -366,6 +374,7 @@ from .pornotube import PornotubeIE
 from .pornoxo import PornoXOIE
 from .promptfile import PromptFileIE
 from .prosiebensat1 import ProSiebenSat1IE
+from .puls4 import Puls4IE
 from .pyvideo import PyvideoIE
 from .quickvid import QuickVidIE
 from .r7 import R7IE

youtube_dl/extractor/airmozilla.py

@@ -0,0 +1,74 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    parse_duration,
+    parse_iso8601,
+)
+
+
+class AirMozillaIE(InfoExtractor):
+    _VALID_URL = r'https?://air\.mozilla\.org/(?P<id>[0-9a-z-]+)/?'
+    _TEST = {
+        'url': 'https://air.mozilla.org/privacy-lab-a-meetup-for-privacy-minded-people-in-san-francisco/',
+        'md5': '2e3e7486ba5d180e829d453875b9b8bf',
+        'info_dict': {
+            'id': '6x4q2w',
+            'ext': 'mp4',
+            'title': 'Privacy Lab - a meetup for privacy minded people in San Francisco',
+            'thumbnail': 're:https://\w+\.cloudfront\.net/6x4q2w/poster\.jpg\?t=\d+',
+            'description': 'Brings together privacy professionals and others interested in privacy at for-profits, non-profits, and NGOs in an effort to contribute to the state of the ecosystem...',
+            'timestamp': 1422487800,
+            'upload_date': '20150128',
+            'location': 'SFO Commons',
+            'duration': 3780,
+            'view_count': int,
+            'categories': ['Main'],
+        }
+    }
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        video_id = self._html_search_regex(r'//vid.ly/(.*?)/embed', webpage, 'id')
+
+        embed_script = self._download_webpage('https://vid.ly/{0}/embed'.format(video_id), video_id)
+        jwconfig = self._search_regex(r'\svar jwconfig = (\{.*?\});\s', embed_script, 'metadata')
+        metadata = self._parse_json(jwconfig, video_id)
+
+        formats = [{
+            'url': source['file'],
+            'ext': source['type'],
+            'format_id': self._search_regex(r'&format=(.*)$', source['file'], 'video format'),
+            'format': source['label'],
+            'height': int(source['label'].rstrip('p')),
+        } for source in metadata['playlist'][0]['sources']]
+        self._sort_formats(formats)
+
+        view_count = int_or_none(self._html_search_regex(
+            r'Views since archived: ([0-9]+)',
+            webpage, 'view count', fatal=False))
+        timestamp = parse_iso8601(self._html_search_regex(
+            r'<time datetime="(.*?)"', webpage, 'timestamp', fatal=False))
+        duration = parse_duration(self._search_regex(
+            r'Duration:\s*(\d+\s*hours?\s*\d+\s*minutes?)',
+            webpage, 'duration', fatal=False))
+
+        return {
+            'id': video_id,
+            'title': self._og_search_title(webpage),
+            'formats': formats,
+            'url': self._og_search_url(webpage),
+            'display_id': display_id,
+            'thumbnail': metadata['playlist'][0].get('image'),
+            'description': self._og_search_description(webpage),
+            'timestamp': timestamp,
+            'location': self._html_search_regex(r'Location: (.*)', webpage, 'location', default=None),
+            'duration': duration,
+            'view_count': view_count,
+            'categories': re.findall(r'<a href=".*?" class="channel">(.*?)</a>', webpage),
+        }

youtube_dl/extractor/comedycentral.py

@@ -250,6 +250,8 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
                 })
                 self._sort_formats(formats)
 
+            subtitles = self._extract_subtitles(cdoc, guid)
+
             virtual_id = show_name + ' ' + epTitle + ' part ' + compat_str(part_num + 1)
             entries.append({
                 'id': guid,
@@ -260,6 +262,7 @@ class ComedyCentralShowsIE(MTVServicesInfoExtractor):
                 'duration': duration,
                 'thumbnail': thumbnail,
                 'description': description,
+                'subtitles': subtitles,
             })
 
         return {

youtube_dl/extractor/common.py

@@ -921,39 +921,57 @@ class InfoExtractor(object):
 
         formats = []
         rtmp_count = 0
-        for video in smil.findall('./body/switch/video'):
-            src = video.get('src')
-            if not src:
-                continue
-            bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
-            width = int_or_none(video.get('width'))
-            height = int_or_none(video.get('height'))
-            proto = video.get('proto')
-            if not proto:
-                if base:
-                    if base.startswith('rtmp'):
-                        proto = 'rtmp'
-                    elif base.startswith('http'):
-                        proto = 'http'
-            ext = video.get('ext')
-            if proto == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(src, video_id, ext))
-            elif proto == 'rtmp':
-                rtmp_count += 1
-                streamer = video.get('streamer') or base
-                formats.append({
-                    'url': streamer,
-                    'play_path': src,
-                    'ext': 'flv',
-                    'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate),
-                    'tbr': bitrate,
-                    'width': width,
-                    'height': height,
-                })
+        if smil.findall('./body/seq/video'):
+            video = smil.findall('./body/seq/video')[0]
+            fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count)
+            formats.extend(fmts)
+        else:
+            for video in smil.findall('./body/switch/video'):
+                fmts, rtmp_count = self._parse_smil_video(video, video_id, base, rtmp_count)
+                formats.extend(fmts)
+
         self._sort_formats(formats)
 
         return formats
 
+    def _parse_smil_video(self, video, video_id, base, rtmp_count):
+        src = video.get('src')
+        if not src:
+            return ([], rtmp_count)
+        bitrate = int_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000)
+        width = int_or_none(video.get('width'))
+        height = int_or_none(video.get('height'))
+        proto = video.get('proto')
+        if not proto:
+            if base:
+                if base.startswith('rtmp'):
+                    proto = 'rtmp'
+                elif base.startswith('http'):
+                    proto = 'http'
+        ext = video.get('ext')
+        if proto == 'm3u8':
+            return (self._extract_m3u8_formats(src, video_id, ext), rtmp_count)
+        elif proto == 'rtmp':
+            rtmp_count += 1
+            streamer = video.get('streamer') or base
+            return ([{
+                'url': streamer,
+                'play_path': src,
+                'ext': 'flv',
+                'format_id': 'rtmp-%d' % (rtmp_count if bitrate is None else bitrate),
+                'tbr': bitrate,
+                'width': width,
+                'height': height,
+            }], rtmp_count)
+        elif proto.startswith('http'):
+            return ([{
+                'url': base + src,
+                'ext': ext or 'flv',
+                'tbr': bitrate,
+                'width': width,
+                'height': height,
+            }], rtmp_count)
+
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()

youtube_dl/extractor/eporner.py

@@ -35,10 +35,7 @@ class EpornerIE(InfoExtractor):
         title = self._html_search_regex(
             r'<title>(.*?) - EPORNER', webpage, 'title')
 
-        redirect_code = self._html_search_regex(
-            r'<script type="text/javascript" src="/config5/%s/([a-f\d]+)/">' % video_id,
-            webpage, 'redirect_code')
-        redirect_url = 'http://www.eporner.com/config5/%s/%s' % (video_id, redirect_code)
+        redirect_url = 'http://www.eporner.com/config5/%s' % video_id
         player_code = self._download_webpage(
             redirect_url, display_id, note='Downloading player config')
 
@@ -69,5 +66,5 @@ class EpornerIE(InfoExtractor):
             'duration': duration,
             'view_count': view_count,
             'formats': formats,
-            'age_limit': self._rta_search(webpage),
+            'age_limit': 18,
         }

youtube_dl/extractor/escapist.py

@@ -3,15 +3,18 @@ from __future__ import unicode_literals
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse,
+    compat_urllib_request,
 )
 from ..utils import (
     ExtractorError,
     js_to_json,
+    parse_duration,
 )
 
 
 class EscapistIE(InfoExtractor):
     _VALID_URL = r'https?://?(www\.)?escapistmagazine\.com/videos/view/[^/?#]+/(?P<id>[0-9]+)-[^/?#]*(?:$|[?#])'
+    _USER_AGENT = 'Mozilla/5.0 (Windows NT 6.1; WOW64; Trident/7.0; rv:11.0) like Gecko'
     _TEST = {
         'url': 'http://www.escapistmagazine.com/videos/view/the-escapist-presents/6618-Breaking-Down-Baldurs-Gate',
         'md5': 'ab3a706c681efca53f0a35f1415cf0d1',
@@ -23,12 +26,15 @@ class EscapistIE(InfoExtractor):
             'uploader': 'The Escapist Presents',
             'title': "Breaking Down Baldur's Gate",
             'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 264,
         }
     }
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
+        webpage_req = compat_urllib_request.Request(url)
+        webpage_req.add_header('User-Agent', self._USER_AGENT)
+        webpage = self._download_webpage(webpage_req, video_id)
 
         uploader_id = self._html_search_regex(
             r"<h1\s+class='headline'>\s*<a\s+href='/videos/view/(.*?)'",
@@ -37,6 +43,7 @@ class EscapistIE(InfoExtractor):
             r"<h1\s+class='headline'>(.*?)</a>",
             webpage, 'uploader', fatal=False)
         description = self._html_search_meta('description', webpage)
+        duration = parse_duration(self._html_search_meta('duration', webpage))
 
         raw_title = self._html_search_meta('title', webpage, fatal=True)
         title = raw_title.partition(' : ')[2]
@@ -44,31 +51,42 @@ class EscapistIE(InfoExtractor):
         config_url = compat_urllib_parse.unquote(self._html_search_regex(
             r'''(?x)
             (?:
-                <param\s+name="flashvars"\s+value="config=|
+                <param\s+name="flashvars".*?\s+value="config=|
                 flashvars=&quot;config=
             )
-            ([^"&]+)
+            (https?://[^"&]+)
             ''',
             webpage, 'config URL'))
 
         formats = []
+        ad_formats = []
 
-        def _add_format(name, cfgurl, quality):
+        def _add_format(name, cfg_url, quality):
+            cfg_req = compat_urllib_request.Request(cfg_url)
+            cfg_req.add_header('User-Agent', self._USER_AGENT)
             config = self._download_json(
-                cfgurl, video_id,
+                cfg_req, video_id,
                 'Downloading ' + name + ' configuration',
                 'Unable to download ' + name + ' configuration',
                 transform_source=js_to_json)
 
             playlist = config['playlist']
-            video_url = next(
-                p['url'] for p in playlist
-                if p.get('eventCategory') == 'Video')
-            formats.append({
-                'url': video_url,
-                'format_id': name,
-                'quality': quality,
-            })
+            for p in playlist:
+                if p.get('eventCategory') == 'Video':
+                    ar = formats
+                elif p.get('eventCategory') == 'Video Postroll':
+                    ar = ad_formats
+                else:
+                    continue
+
+                ar.append({
+                    'url': p['url'],
+                    'format_id': name,
+                    'quality': quality,
+                    'http_headers': {
+                        'User-Agent': self._USER_AGENT,
+                    },
+                })
 
         _add_format('normal', config_url, quality=0)
         hq_url = (config_url +
@@ -77,10 +95,12 @@ class EscapistIE(InfoExtractor):
             _add_format('hq', hq_url, quality=1)
         except ExtractorError:
             pass  # That's fine, we'll just use normal quality
-
         self._sort_formats(formats)
 
-        return {
+        if '/escapist/sales-marketing/' in formats[-1]['url']:
+            raise ExtractorError('This IP address has been blocked by The Escapist', expected=True)
+
+        res = {
             'id': video_id,
             'formats': formats,
             'uploader': uploader,
@@ -88,4 +108,21 @@ class EscapistIE(InfoExtractor):
             'title': title,
             'thumbnail': self._og_search_thumbnail(webpage),
             'description': description,
+            'duration': duration,
         }
+
+        if self._downloader.params.get('include_ads') and ad_formats:
+            self._sort_formats(ad_formats)
+            ad_res = {
+                'id': '%s-ad' % video_id,
+                'title': '%s (Postroll)' % title,
+                'formats': ad_formats,
+            }
+            return {
+                '_type': 'playlist',
+                'entries': [res, ad_res],
+                'title': title,
+                'id': video_id,
+            }
+
+        return res

youtube_dl/extractor/gamestar.py

@@ -1,6 +1,8 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
@@ -31,7 +33,7 @@ class GameStarIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         og_title = self._og_search_title(webpage)
-        title = og_title.replace(' - Video bei GameStar.de', '').strip()
+        title = re.sub(r'\s*- Video (bei|-) GameStar\.de$', '', og_title)
 
         url = 'http://gamestar.de/_misc/videos/portal/getVideoUrl.cfm?premium=0&videoId=' + video_id
 

youtube_dl/extractor/generic.py

@@ -557,6 +557,18 @@ class GenericIE(InfoExtractor):
                 'title': 'EP3S5 - Bon Appétit - Baqueira Mi Corazon !',
             }
         },
+        # Kaltura embed
+        {
+            'url': 'http://www.monumentalnetwork.com/videos/john-carlson-postgame-2-25-15',
+            'info_dict': {
+                'id': '1_eergr3h1',
+                'ext': 'mp4',
+                'upload_date': '20150226',
+                'uploader_id': 'MonumentalSports-Kaltura@perfectsensedigital.com',
+                'timestamp': int,
+                'title': 'John Carlson Postgame 2/25/15',
+            },
+        },
     ]
 
     def report_following_redirect(self, new_url):
@@ -1113,6 +1125,12 @@ class GenericIE(InfoExtractor):
         if mobj is not None:
             return self.url_result(mobj.group('url'), 'Zapiks')
 
+        # Look for Kaltura embeds
+        mobj = re.search(
+            r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?'wid'\s*:\s*'_?(?P<partner_id>[^']+)',.*?'entry_id'\s*:\s*'(?P<id>[^']+)',", webpage)
+        if mobj is not None:
+            return self.url_result('kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(), 'Kaltura')
+
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
                 return True
@@ -1208,7 +1226,9 @@ class GenericIE(InfoExtractor):
             return entries[0]
         else:
             for num, e in enumerate(entries, start=1):
-                e['title'] = '%s (%d)' % (e['title'], num)
+                # 'url' results don't have a title
+                if e.get('title') is not None:
+                    e['title'] = '%s (%d)' % (e['title'], num)
             return {
                 '_type': 'playlist',
                 'entries': entries,

youtube_dl/extractor/kaltura.py

@@ -0,0 +1,138 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_urllib_parse
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+)
+
+
+class KalturaIE(InfoExtractor):
+    _VALID_URL = r'''(?x)
+    (?:kaltura:|
+       https?://(:?(?:www|cdnapisec)\.)?kaltura\.com/index\.php/kwidget/(?:[^/]+/)*?wid/_
+    )(?P<partner_id>\d+)
+    (?::|
+       /(?:[^/]+/)*?entry_id/
+    )(?P<id>[0-9a-z_]+)'''
+    _API_BASE = 'http://cdnapi.kaltura.com/api_v3/index.php?'
+    _TESTS = [
+        {
+            'url': 'kaltura:269692:1_1jc2y3e4',
+            'md5': '3adcbdb3dcc02d647539e53f284ba171',
+            'info_dict': {
+                'id': '1_1jc2y3e4',
+                'ext': 'mp4',
+                'title': 'Track 4',
+                'upload_date': '20131219',
+                'uploader_id': 'mlundberg@wolfgangsvault.com',
+                'description': 'The Allman Brothers Band, 12/16/1981',
+                'thumbnail': 're:^https?://.*/thumbnail/.*',
+                'timestamp': int,
+            },
+        },
+        {
+            'url': 'http://www.kaltura.com/index.php/kwidget/cache_st/1300318621/wid/_269692/uiconf_id/3873291/entry_id/1_1jc2y3e4',
+            'only_matching': True,
+        },
+        {
+            'url': 'https://cdnapisec.kaltura.com/index.php/kwidget/wid/_557781/uiconf_id/22845202/entry_id/1_plr1syf3',
+            'only_matching': True,
+        },
+    ]
+
+    def _kaltura_api_call(self, video_id, actions, *args, **kwargs):
+        params = actions[0]
+        if len(actions) > 1:
+            for i, a in enumerate(actions[1:], start=1):
+                for k, v in a.items():
+                    params['%d:%s' % (i, k)] = v
+
+        query = compat_urllib_parse.urlencode(params)
+        url = self._API_BASE + query
+        data = self._download_json(url, video_id, *args, **kwargs)
+
+        status = data if len(actions) == 1 else data[0]
+        if status.get('objectType') == 'KalturaAPIException':
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, status['message']))
+
+        return data
+
+    def _get_kaltura_signature(self, video_id, partner_id):
+        actions = [{
+            'apiVersion': '3.1',
+            'expiry': 86400,
+            'format': 1,
+            'service': 'session',
+            'action': 'startWidgetSession',
+            'widgetId': '_%s' % partner_id,
+        }]
+        return self._kaltura_api_call(
+            video_id, actions, note='Downloading Kaltura signature')['ks']
+
+    def _get_video_info(self, video_id, partner_id):
+        signature = self._get_kaltura_signature(video_id, partner_id)
+        actions = [
+            {
+                'action': 'null',
+                'apiVersion': '3.1.5',
+                'clientTag': 'kdp:v3.8.5',
+                'format': 1,  # JSON, 2 = XML, 3 = PHP
+                'service': 'multirequest',
+                'ks': signature,
+            },
+            {
+                'action': 'get',
+                'entryId': video_id,
+                'service': 'baseentry',
+                'version': '-1',
+            },
+            {
+                'action': 'getContextData',
+                'contextDataParams:objectType': 'KalturaEntryContextDataParams',
+                'contextDataParams:referrer': 'http://www.kaltura.com/',
+                'contextDataParams:streamerType': 'http',
+                'entryId': video_id,
+                'service': 'baseentry',
+            },
+        ]
+        return self._kaltura_api_call(
+            video_id, actions, note='Downloading video info JSON')
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        partner_id, entry_id = mobj.group('partner_id'), mobj.group('id')
+
+        info, source_data = self._get_video_info(entry_id, partner_id)
+
+        formats = [{
+            'format_id': '%(fileExt)s-%(bitrate)s' % f,
+            'ext': f['fileExt'],
+            'tbr': f['bitrate'],
+            'fps': f.get('frameRate'),
+            'filesize_approx': int_or_none(f.get('size'), invscale=1024),
+            'container': f.get('containerFormat'),
+            'vcodec': f.get('videoCodecId'),
+            'height': f.get('height'),
+            'width': f.get('width'),
+            'url': '%s/flavorId/%s' % (info['dataUrl'], f['id']),
+        } for f in source_data['flavorAssets']]
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': info['name'],
+            'formats': formats,
+            'description': info.get('description'),
+            'thumbnail': info.get('thumbnailUrl'),
+            'duration': info.get('duration'),
+            'timestamp': info.get('createdAt'),
+            'uploader_id': info.get('userId'),
+            'view_count': info.get('plays'),
+        }

youtube_dl/extractor/laola1tv.py

@@ -27,8 +27,6 @@ class Laola1TvIE(InfoExtractor):
         }
     }
 
-    _BROKEN = True  # Not really - extractor works fine, but f4m downloader does not support live streams yet.
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
@@ -57,11 +55,7 @@ class Laola1TvIE(InfoExtractor):
         title = xpath_text(hd_doc, './/video/title', fatal=True)
         flash_url = xpath_text(hd_doc, './/video/url', fatal=True)
         uploader = xpath_text(hd_doc, './/video/meta_organistation')
-
         is_live = xpath_text(hd_doc, './/video/islive') == 'true'
-        if is_live:
-            raise ExtractorError(
-                'Live streams are not supported by the f4m downloader.')
 
         categories = xpath_text(hd_doc, './/video/meta_sports')
         if categories:

youtube_dl/extractor/letv.py

@@ -0,0 +1,190 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import datetime
+import re
+import time
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urlparse,
+    compat_urllib_parse,
+)
+from ..utils import (
+    determine_ext,
+    ExtractorError,
+    parse_iso8601,
+)
+
+
+class LetvIE(InfoExtractor):
+    _VALID_URL = r'http://www\.letv\.com/ptv/vplay/(?P<id>\d+).html'
+
+    _TESTS = [{
+        'url': 'http://www.letv.com/ptv/vplay/22005890.html',
+        'md5': 'cab23bd68d5a8db9be31c9a222c1e8df',
+        'info_dict': {
+            'id': '22005890',
+            'ext': 'mp4',
+            'title': '第87届奥斯卡颁奖礼完美落幕 《鸟人》成最大赢家',
+            'timestamp': 1424747397,
+            'upload_date': '20150224',
+            'description': 'md5:a9cb175fd753e2962176b7beca21a47c',
+        }
+    }, {
+        'url': 'http://www.letv.com/ptv/vplay/1415246.html',
+        'info_dict': {
+            'id': '1415246',
+            'ext': 'mp4',
+            'title': '美人天下01',
+            'description': 'md5:f88573d9d7225ada1359eaf0dbf8bcda',
+        },
+        'expected_warnings': [
+            'publish time'
+        ]
+    }]
+    # http://www.letv.com/ptv/vplay/1118082.html
+    # This video is available only in Mainland China
+
+    @staticmethod
+    def urshift(val, n):
+        return val >> n if val >= 0 else (val + 0x100000000) >> n
+
+    # ror() and calc_time_key() are reversed from a embedded swf file in KLetvPlayer.swf
+    def ror(self, param1, param2):
+        _loc3_ = 0
+        while _loc3_ < param2:
+            param1 = self.urshift(param1, 1) + ((param1 & 1) << 31)
+            _loc3_ += 1
+        return param1
+
+    def calc_time_key(self, param1):
+        _loc2_ = 773625421
+        _loc3_ = self.ror(param1, _loc2_ % 13)
+        _loc3_ = _loc3_ ^ _loc2_
+        _loc3_ = self.ror(_loc3_, _loc2_ % 17)
+        return _loc3_
+
+    def _real_extract(self, url):
+        media_id = self._match_id(url)
+        page = self._download_webpage(url, media_id)
+        params = {
+            'id': media_id,
+            'platid': 1,
+            'splatid': 101,
+            'format': 1,
+            'tkey': self.calc_time_key(int(time.time())),
+            'domain': 'www.letv.com'
+        }
+        play_json = self._download_json(
+            'http://api.letv.com/mms/out/video/playJson?' + compat_urllib_parse.urlencode(params),
+            media_id, 'playJson data')
+
+        # Check for errors
+        playstatus = play_json['playstatus']
+        if playstatus['status'] == 0:
+            flag = playstatus['flag']
+            if flag == 1:
+                msg = 'Country %s auth error' % playstatus['country']
+            else:
+                msg = 'Generic error. flag = %d' % flag
+            raise ExtractorError(msg, expected=True)
+
+        playurl = play_json['playurl']
+
+        formats = ['350', '1000', '1300', '720p', '1080p']
+        dispatch = playurl['dispatch']
+
+        urls = []
+        for format_id in formats:
+            if format_id in dispatch:
+                media_url = playurl['domain'][0] + dispatch[format_id][0]
+
+                # Mimic what flvxz.com do
+                url_parts = list(compat_urlparse.urlparse(media_url))
+                qs = dict(compat_urlparse.parse_qs(url_parts[4]))
+                qs.update({
+                    'platid': '14',
+                    'splatid': '1401',
+                    'tss': 'no',
+                    'retry': 1
+                })
+                url_parts[4] = compat_urllib_parse.urlencode(qs)
+                media_url = compat_urlparse.urlunparse(url_parts)
+
+                url_info_dict = {
+                    'url': media_url,
+                    'ext': determine_ext(dispatch[format_id][1])
+                }
+
+                if format_id[-1:] == 'p':
+                    url_info_dict['height'] = format_id[:-1]
+
+                urls.append(url_info_dict)
+
+        publish_time = parse_iso8601(self._html_search_regex(
+            r'发布时间&nbsp;([^<>]+) ', page, 'publish time', fatal=False),
+            delimiter=' ', timezone=datetime.timedelta(hours=8))
+        description = self._html_search_meta('description', page, fatal=False)
+
+        return {
+            'id': media_id,
+            'formats': urls,
+            'title': playurl['title'],
+            'thumbnail': playurl['pic'],
+            'description': description,
+            'timestamp': publish_time,
+        }
+
+
+class LetvTvIE(InfoExtractor):
+    _VALID_URL = r'http://www.letv.com/tv/(?P<id>\d+).html'
+    _TESTS = [{
+        'url': 'http://www.letv.com/tv/46177.html',
+        'info_dict': {
+            'id': '46177',
+            'title': '美人天下',
+            'description': 'md5:395666ff41b44080396e59570dbac01c'
+        },
+        'playlist_count': 35
+    }]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+        page = self._download_webpage(url, playlist_id)
+
+        media_urls = list(set(re.findall(
+            r'http://www.letv.com/ptv/vplay/\d+.html', page)))
+        entries = [self.url_result(media_url, ie='Letv')
+                   for media_url in media_urls]
+
+        title = self._html_search_meta('keywords', page,
+                                       fatal=False).split('，')[0]
+        description = self._html_search_meta('description', page, fatal=False)
+
+        return self.playlist_result(entries, playlist_id, playlist_title=title,
+                                    playlist_description=description)
+
+
+class LetvPlaylistIE(LetvTvIE):
+    _VALID_URL = r'http://tv.letv.com/[a-z]+/(?P<id>[a-z]+)/index.s?html'
+    _TESTS = [{
+        'url': 'http://tv.letv.com/izt/wuzetian/index.html',
+        'info_dict': {
+            'id': 'wuzetian',
+            'title': '武媚娘传奇',
+            'description': 'md5:e12499475ab3d50219e5bba00b3cb248'
+        },
+        # This playlist contains some extra videos other than the drama itself
+        'playlist_mincount': 96
+    }, {
+        'url': 'http://tv.letv.com/pzt/lswjzzjc/index.shtml',
+        'info_dict': {
+            'id': 'lswjzzjc',
+            # The title should be "劲舞青春", but I can't find a simple way to
+            # determine the playlist title
+            'title': '乐视午间自制剧场',
+            'description': 'md5:b1eef244f45589a7b5b1af9ff25a4489'
+        },
+        'playlist_mincount': 7
+    }]

youtube_dl/extractor/lynda.py

@@ -18,7 +18,7 @@ from ..utils import (
 class LyndaIE(InfoExtractor):
     IE_NAME = 'lynda'
     IE_DESC = 'lynda.com videos'
-    _VALID_URL = r'https?://www\.lynda\.com/[^/]+/[^/]+/\d+/(\d+)-\d\.html'
+    _VALID_URL = r'https?://www\.lynda\.com/(?:[^/]+/[^/]+/\d+|player/embed)/(\d+)'
     _LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
     _NETRC_MACHINE = 'lynda'
 
@@ -27,7 +27,7 @@ class LyndaIE(InfoExtractor):
 
     ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
         'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
         'info_dict': {
@@ -36,7 +36,10 @@ class LyndaIE(InfoExtractor):
             'title': 'Using the exercise files',
             'duration': 68
         }
-    }
+    }, {
+        'url': 'https://www.lynda.com/player/embed/133770?tr=foo=1;bar=g;fizz=rt&fs=0',
+        'only_matching': True,
+    }]
 
     def _real_initialize(self):
         self._login()
@@ -152,7 +155,7 @@ class LyndaIE(InfoExtractor):
                 continue
             appear_time = m_current.group('timecode')
             disappear_time = m_next.group('timecode')
-            text = seq_current['Caption']
+            text = seq_current['Caption'].lstrip()
             srt += '%s\r\n%s --> %s\r\n%s' % (str(pos), appear_time, disappear_time, text)
         if srt:
             return srt

youtube_dl/extractor/mitele.py

@@ -18,7 +18,7 @@ class MiTeleIE(InfoExtractor):
     IE_NAME = 'mitele.es'
     _VALID_URL = r'http://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
         'md5': '6a75fe9d0d3275bead0cb683c616fddb',
         'info_dict': {
@@ -29,7 +29,7 @@ class MiTeleIE(InfoExtractor):
             'display_id': 'programa-144',
             'duration': 2913,
         },
-    }
+    }]
 
     def _real_extract(self, url):
         episode = self._match_id(url)

youtube_dl/extractor/mpora.py

@@ -5,7 +5,7 @@ from ..utils import int_or_none
 
 
 class MporaIE(InfoExtractor):
-    _VALID_URL = r'https?://(www\.)?mpora\.(?:com|de)/videos/(?P<id>[^?#/]+)'
+    _VALID_URL = r'https?://(?:www\.)?mpora\.(?:com|de)/videos/(?P<id>[^?#/]+)'
     IE_NAME = 'MPORA'
 
     _TEST = {
@@ -25,7 +25,9 @@ class MporaIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         data_json = self._search_regex(
-            r"new FM\.Player\('[^']+',\s*(\{.*?)\).player;", webpage, 'json')
+            [r"new FM\.Player\('[^']+',\s*(\{.*?)\).player;",
+             r"new\s+FM\.Kaltura\.Player\('[^']+'\s*,\s*({.+?})\);"],
+            webpage, 'json')
         data = self._parse_json(data_json, video_id)
 
         uploader = data['info_overlay'].get('username')

youtube_dl/extractor/musicvault.py

@@ -3,17 +3,13 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import (
-    parse_duration,
-    unified_strdate,
-)
 
 
 class MusicVaultIE(InfoExtractor):
     _VALID_URL = r'https?://www\.musicvault\.com/(?P<uploader_id>[^/?#]*)/video/(?P<display_id>[^/?#]*)_(?P<id>[0-9]+)\.html'
     _TEST = {
         'url': 'http://www.musicvault.com/the-allman-brothers-band/video/straight-from-the-heart_1010863.html',
-        'md5': '2cdbb3ae75f7fb3519821507d2fb3c15',
+        'md5': '3adcbdb3dcc02d647539e53f284ba171',
         'info_dict': {
             'id': '1010863',
             'ext': 'mp4',
@@ -22,9 +18,10 @@ class MusicVaultIE(InfoExtractor):
             'duration': 244,
             'uploader': 'The Allman Brothers Band',
             'thumbnail': 're:^https?://.*/thumbnail/.*',
-            'upload_date': '19811216',
+            'upload_date': '20131219',
             'location': 'Capitol Theatre (Passaic, NJ)',
             'description': 'Listen to The Allman Brothers Band perform Straight from the Heart at Capitol Theatre (Passaic, NJ) on Dec 16, 1981',
+            'timestamp': int,
         }
     }
 
@@ -43,34 +40,24 @@ class MusicVaultIE(InfoExtractor):
             r'<h1.*?>(.*?)</h1>', data_div, 'uploader', fatal=False)
         title = self._html_search_regex(
             r'<h2.*?>(.*?)</h2>', data_div, 'title')
-        upload_date = unified_strdate(self._html_search_regex(
-            r'<h3.*?>(.*?)</h3>', data_div, 'uploader', fatal=False))
         location = self._html_search_regex(
             r'<h4.*?>(.*?)</h4>', data_div, 'location', fatal=False)
 
-        duration = parse_duration(self._html_search_meta('duration', webpage))
-
-        VIDEO_URL_TEMPLATE = 'http://cdnapi.kaltura.com/p/%(uid)s/sp/%(wid)s/playManifest/entryId/%(entry_id)s/format/url/protocol/http'
         kaltura_id = self._search_regex(
             r'<div id="video-detail-player" data-kaltura-id="([^"]+)"',
             webpage, 'kaltura ID')
-        video_url = VIDEO_URL_TEMPLATE % {
-            'entry_id': kaltura_id,
-            'wid': self._search_regex(r'/wid/_([0-9]+)/', webpage, 'wid'),
-            'uid': self._search_regex(r'uiconf_id/([0-9]+)/', webpage, 'uid'),
-        }
+        wid = self._search_regex(r'/wid/_([0-9]+)/', webpage, 'wid')
 
         return {
             'id': mobj.group('id'),
-            'url': video_url,
-            'ext': 'mp4',
+            '_type': 'url_transparent',
+            'url': 'kaltura:%s:%s' % (wid, kaltura_id),
+            'ie_key': 'Kaltura',
             'display_id': display_id,
             'uploader_id': mobj.group('uploader_id'),
             'thumbnail': thumbnail,
             'description': self._html_search_meta('description', webpage),
-            'upload_date': upload_date,
             'location': location,
             'title': title,
             'uploader': uploader,
-            'duration': duration,
         }

youtube_dl/extractor/nrk.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     float_or_none,
@@ -158,7 +159,9 @@ class NRKTVIE(InfoExtractor):
     def _get_subtitles(self, subtitlesurl, video_id, baseurl):
         url = "%s%s" % (baseurl, subtitlesurl)
         self._debug_print('%s: Subtitle url: %s' % (video_id, url))
-        captions = self._download_xml(url, video_id, 'Downloading subtitles')
+        captions = self._download_xml(
+            url, video_id, 'Downloading subtitles',
+            transform_source=lambda s: s.replace(r'<br />', '\r\n'))
         lang = captions.get('lang', 'no')
         ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/ns/ttml}'))
         srt = ''
@@ -167,8 +170,7 @@ class NRKTVIE(InfoExtractor):
             duration = parse_duration(p.get('dur'))
             starttime = self._seconds2str(begin)
             endtime = self._seconds2str(begin + duration)
-            text = '\n'.join(p.itertext())
-            srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), starttime, endtime, text)
+            srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (compat_str(pos), starttime, endtime, p.text)
         return {lang: [
             {'ext': 'ttml', 'url': url},
             {'ext': 'srt', 'data': srt},

youtube_dl/extractor/odnoklassniki.py

@@ -0,0 +1,85 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    unified_strdate,
+    int_or_none,
+    qualities,
+)
+
+
+class OdnoklassnikiIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:odnoklassniki|ok)\.ru/(?:video|web-api/video/moviePlayer)/(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://ok.ru/video/20079905452',
+        'md5': '8e24ad2da6f387948e7a7d44eb8668fe',
+        'info_dict': {
+            'id': '20079905452',
+            'ext': 'mp4',
+            'title': 'Культура меняет нас (прекрасный ролик!))',
+            'duration': 100,
+            'upload_date': '20141207',
+            'uploader_id': '330537914540',
+            'uploader': 'Виталий Добровольский',
+            'like_count': int,
+            'age_limit': 0,
+        },
+    }, {
+        'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        player = self._parse_json(
+            self._search_regex(
+                r"OKVideo\.start\(({.+?})\s*,\s*'VideoAutoplay_player'", webpage, 'player'),
+            video_id)
+
+        metadata = self._parse_json(player['flashvars']['metadata'], video_id)
+
+        movie = metadata['movie']
+        title = movie['title']
+        thumbnail = movie.get('poster')
+        duration = int_or_none(movie.get('duration'))
+
+        author = metadata.get('author', {})
+        uploader_id = author.get('id')
+        uploader = author.get('name')
+
+        upload_date = unified_strdate(self._html_search_meta(
+            'ya:ovs:upload_date', webpage, 'upload date'))
+
+        age_limit = None
+        adult = self._html_search_meta(
+            'ya:ovs:adult', webpage, 'age limit')
+        if adult:
+            age_limit = 18 if adult == 'true' else 0
+
+        like_count = int_or_none(metadata.get('likeCount'))
+
+        quality = qualities(('mobile', 'lowest', 'low', 'sd', 'hd'))
+
+        formats = [{
+            'url': f['url'],
+            'ext': 'mp4',
+            'format_id': f['name'],
+            'quality': quality(f['name']),
+        } for f in metadata['videos']]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'upload_date': upload_date,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'like_count': like_count,
+            'age_limit': age_limit,
+            'formats': formats,
+        }

youtube_dl/extractor/puls4.py

@@ -0,0 +1,88 @@
+# -*- coding: utf-8 -*-
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    unified_strdate,
+    int_or_none,
+)
+
+
+class Puls4IE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?puls4\.com/video/[^/]+/play/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://www.puls4.com/video/pro-und-contra/play/2716816',
+        'md5': '49f6a6629747eeec43cef6a46b5df81d',
+        'info_dict': {
+            'id': '2716816',
+            'ext': 'mp4',
+            'title': 'Pro und Contra vom 23.02.2015',
+            'description': 'md5:293e44634d9477a67122489994675db6',
+            'duration': 2989,
+            'upload_date': '20150224',
+            'uploader': 'PULS_4',
+        },
+        'skip': 'Only works from Germany',
+    }, {
+        'url': 'http://www.puls4.com/video/kult-spielfilme/play/1298106',
+        'md5': '6a48316c8903ece8dab9b9a7bf7a59ec',
+        'info_dict': {
+            'id': '1298106',
+            'ext': 'mp4',
+            'title': 'Lucky Fritz',
+        },
+        'skip': 'Only works from Germany',
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        error_message = self._html_search_regex(
+            r'<div class="message-error">(.+?)</div>',
+            webpage, 'error message', default=None)
+        if error_message:
+            raise ExtractorError(
+                '%s returned error: %s' % (self.IE_NAME, error_message), expected=True)
+
+        real_url = self._html_search_regex(
+            r'\"fsk-button\".+?href=\"([^"]+)',
+            webpage, 'fsk_button', default=None)
+        if real_url:
+            webpage = self._download_webpage(real_url, video_id)
+
+        player = self._search_regex(
+            r'p4_video_player(?:_iframe)?\("video_\d+_container"\s*,(.+?)\);\s*\}',
+            webpage, 'player')
+
+        player_json = self._parse_json(
+            '[%s]' % player, video_id,
+            transform_source=lambda s: s.replace('undefined,', ''))
+
+        formats = None
+        result = None
+
+        for v in player_json:
+            if isinstance(v, list) and not formats:
+                formats = [{
+                    'url': f['url'],
+                    'format': 'hd' if f.get('hd') else 'sd',
+                    'width': int_or_none(f.get('size_x')),
+                    'height': int_or_none(f.get('size_y')),
+                    'tbr': int_or_none(f.get('bitrate')),
+                } for f in v]
+                self._sort_formats(formats)
+            elif isinstance(v, dict) and not result:
+                result = {
+                    'id': video_id,
+                    'title': v['videopartname'].strip(),
+                    'description': v.get('videotitle'),
+                    'duration': int_or_none(v.get('videoduration') or v.get('episodeduration')),
+                    'upload_date': unified_strdate(v.get('clipreleasetime')),
+                    'uploader': v.get('channel'),
+                }
+
+        result['formats'] = formats
+
+        return result

youtube_dl/extractor/rtlnow.py

@@ -146,7 +146,7 @@ class RTLnowIE(InfoExtractor):
                 mobj = re.search(r'.*/(?P<hoster>[^/]+)/videos/(?P<play_path>.+)\.f4m', filename.text)
                 if mobj:
                     fmt = {
-                        'url': 'rtmpe://fmspay-fra2.rtl.de/' + mobj.group('hoster'),
+                        'url': 'rtmpe://fms.rtl.de/' + mobj.group('hoster'),
                         'play_path': 'mp4:' + mobj.group('play_path'),
                         'page_url': url,
                         'player_url': video_page_url + 'includes/vodplayer.swf',

youtube_dl/extractor/rtve.py

@@ -8,8 +8,9 @@ import time
 from .common import InfoExtractor
 from ..compat import compat_urlparse
 from ..utils import (
-    struct_unpack,
+    float_or_none,
     remove_end,
+    struct_unpack,
 )
 
 
@@ -67,6 +68,7 @@ class RTVEALaCartaIE(InfoExtractor):
             'id': '2491869',
             'ext': 'mp4',
             'title': 'Balonmano - Swiss Cup masculina. Final: España-Suecia',
+            'duration': 5024.566,
         },
     }, {
         'note': 'Live stream',
@@ -113,6 +115,7 @@ class RTVEALaCartaIE(InfoExtractor):
             'thumbnail': info.get('image'),
             'page_url': url,
             'subtitles': subtitles,
+            'duration': float_or_none(info.get('duration'), scale=1000),
         }
 
     def _get_subtitles(self, video_id, sub_file):

youtube_dl/extractor/svtplay.py

@@ -1,6 +1,8 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..utils import (
     determine_ext,
@@ -8,23 +10,40 @@ from ..utils import (
 
 
 class SVTPlayIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?svtplay\.se/video/(?P<id>[0-9]+)'
-    _TEST = {
+    IE_DESC = 'SVT Play and Öppet arkiv'
+    _VALID_URL = r'https?://(?:www\.)?(?P<host>svtplay|oppetarkiv)\.se/video/(?P<id>[0-9]+)'
+    _TESTS = [{
         'url': 'http://www.svtplay.se/video/2609989/sm-veckan/sm-veckan-rally-final-sasong-1-sm-veckan-rally-final',
-        'md5': 'f4a184968bc9c802a9b41316657aaa80',
+        'md5': 'ade3def0643fa1c40587a422f98edfd9',
         'info_dict': {
             'id': '2609989',
-            'ext': 'mp4',
+            'ext': 'flv',
             'title': 'SM veckan vinter, Örebro - Rally, final',
             'duration': 4500,
             'thumbnail': 're:^https?://.*[\.-]jpg$',
+            'age_limit': 0,
         },
-    }
+    }, {
+        'url': 'http://www.oppetarkiv.se/video/1058509/rederiet-sasong-1-avsnitt-1-av-318',
+        'md5': 'c3101a17ce9634f4c1f9800f0746c187',
+        'info_dict': {
+            'id': '1058509',
+            'ext': 'flv',
+            'title': 'Farlig kryssning',
+            'duration': 2566,
+            'thumbnail': 're:^https?://.*[\.-]jpg$',
+            'age_limit': 0,
+        },
+        'skip': 'Only works from Sweden',
+    }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        host = mobj.group('host')
+
         info = self._download_json(
-            'http://www.svtplay.se/video/%s?output=json' % video_id, video_id)
+            'http://www.%s.se/video/%s?output=json' % (host, video_id), video_id)
 
         title = info['context']['title']
         thumbnail = info['context'].get('thumbnailImage')
@@ -33,11 +52,16 @@ class SVTPlayIE(InfoExtractor):
         formats = []
         for vr in video_info['videoReferences']:
             vurl = vr['url']
-            if determine_ext(vurl) == 'm3u8':
+            ext = determine_ext(vurl)
+            if ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
                     vurl, video_id,
                     ext='mp4', entry_protocol='m3u8_native',
                     m3u8_id=vr.get('playerType')))
+            elif ext == 'f4m':
+                formats.extend(self._extract_f4m_formats(
+                    vurl + '?hdcore=3.3.0', video_id,
+                    f4m_id=vr.get('playerType')))
             else:
                 formats.append({
                     'format_id': vr.get('playerType'),
@@ -46,6 +70,7 @@ class SVTPlayIE(InfoExtractor):
         self._sort_formats(formats)
 
         duration = video_info.get('materialLength')
+        age_limit = 18 if video_info.get('inappropriateForChildren') else 0
 
         return {
             'id': video_id,
@@ -53,4 +78,5 @@ class SVTPlayIE(InfoExtractor):
             'formats': formats,
             'thumbnail': thumbnail,
             'duration': duration,
+            'age_limit': age_limit,
         }

youtube_dl/extractor/telecinco.py

@@ -6,9 +6,9 @@ from .mitele import MiTeleIE
 
 class TelecincoIE(MiTeleIE):
     IE_NAME = 'telecinco.es'
-    _VALID_URL = r'https?://www\.telecinco\.es/[^/]+/[^/]+/[^/]+/(?P<id>.*?)\.html'
+    _VALID_URL = r'https?://www\.telecinco\.es/[^/]+/[^/]+/(?:[^/]+/)?(?P<id>.*?)\.html'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
         'info_dict': {
             'id': 'MDSVID20141015_0058',
@@ -16,4 +16,7 @@ class TelecincoIE(MiTeleIE):
             'title': 'Con Martín Berasategui, hacer un bacalao al ...',
             'duration': 662,
         },
-    }
+    }, {
+        'url': 'http://www.telecinco.es/informativos/nacional/Pablo_Iglesias-Informativos_Telecinco-entrevista-Pedro_Piqueras_2_1945155182.html',
+        'only_matching': True,
+    }]

youtube_dl/extractor/twitch.py

@@ -34,7 +34,15 @@ class TwitchBaseIE(InfoExtractor):
                 expected=True)
 
     def _download_json(self, url, video_id, note='Downloading JSON metadata'):
-        response = super(TwitchBaseIE, self)._download_json(url, video_id, note)
+        headers = {
+            'Referer': 'http://api.twitch.tv/crossdomain/receiver.html?v=2',
+            'X-Requested-With': 'XMLHttpRequest',
+        }
+        for cookie in self._downloader.cookiejar:
+            if cookie.name == 'api_token':
+                headers['Twitch-Api-Token'] = cookie.value
+        request = compat_urllib_request.Request(url, headers=headers)
+        response = super(TwitchBaseIE, self)._download_json(request, video_id, note)
         self._handle_error(response)
         return response
 

youtube_dl/extractor/wdr.py

@@ -28,6 +28,7 @@ class WDRIE(InfoExtractor):
                 'title': 'Servicezeit',
                 'description': 'md5:c8f43e5e815eeb54d0b96df2fba906cb',
                 'upload_date': '20140310',
+                'is_live': False
             },
             'params': {
                 'skip_download': True,
@@ -41,6 +42,7 @@ class WDRIE(InfoExtractor):
                 'title': 'Marga Spiegel ist tot',
                 'description': 'md5:2309992a6716c347891c045be50992e4',
                 'upload_date': '20140311',
+                'is_live': False
             },
             'params': {
                 'skip_download': True,
@@ -55,6 +57,7 @@ class WDRIE(InfoExtractor):
                 'title': 'Erlebte Geschichten: Marga Spiegel (29.11.2009)',
                 'description': 'md5:2309992a6716c347891c045be50992e4',
                 'upload_date': '20091129',
+                'is_live': False
             },
         },
         {
@@ -66,6 +69,7 @@ class WDRIE(InfoExtractor):
                 'title': 'Flavia Coelho: Amar é Amar',
                 'description': 'md5:7b29e97e10dfb6e265238b32fa35b23a',
                 'upload_date': '20140717',
+                'is_live': False
             },
         },
         {
@@ -74,6 +78,20 @@ class WDRIE(InfoExtractor):
             'info_dict': {
                 'id': 'mediathek/video/sendungen/quarks_und_co/filterseite-quarks-und-co100',
             }
+        },
+        {
+            'url': 'http://www1.wdr.de/mediathek/video/livestream/index.html',
+            'info_dict': {
+                'id': 'mdb-103364',
+                'title': 're:^WDR Fernsehen [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+                'description': 'md5:ae2ff888510623bf8d4b115f95a9b7c9',
+                'ext': 'flv',
+                'upload_date': '20150212',
+                'is_live': True
+            },
+            'params': {
+                'skip_download': True,
+            },
         }
     ]
 
@@ -119,6 +137,10 @@ class WDRIE(InfoExtractor):
         video_url = flashvars['dslSrc'][0]
         title = flashvars['trackerClipTitle'][0]
         thumbnail = flashvars['startPicture'][0] if 'startPicture' in flashvars else None
+        is_live = flashvars.get('isLive', ['0'])[0] == '1'
+
+        if is_live:
+            title = self._live_title(title)
 
         if 'trackerClipAirTime' in flashvars:
             upload_date = flashvars['trackerClipAirTime'][0]
@@ -131,6 +153,13 @@ class WDRIE(InfoExtractor):
         if video_url.endswith('.f4m'):
             video_url += '?hdcore=3.2.0&plugin=aasp-3.2.0.77.18'
             ext = 'flv'
+        elif video_url.endswith('.smil'):
+            fmt = self._extract_smil_formats(video_url, page_id)[0]
+            video_url = fmt['url']
+            sep = '&' if '?' in video_url else '?'
+            video_url += sep
+            video_url += 'hdcore=3.3.0&plugin=aasp-3.3.0.99.43'
+            ext = fmt['ext']
         else:
             ext = determine_ext(video_url)
 
@@ -144,6 +173,7 @@ class WDRIE(InfoExtractor):
             'description': description,
             'thumbnail': thumbnail,
             'upload_date': upload_date,
+            'is_live': is_live
         }
 
 

youtube_dl/options.py

@@ -272,6 +272,10 @@ def parseOpts(overrideArguments=None):
         '--no-playlist',
         action='store_true', dest='noplaylist', default=False,
         help='If the URL refers to a video and a playlist, download only the video.')
+    selection.add_option(
+        '--yes-playlist',
+        action='store_false', dest='noplaylist', default=False,
+        help='If the URL refers to a video and a playlist, download the playlist.')
     selection.add_option(
         '--age-limit',
         metavar='YEARS', dest='age_limit', default=None, type=int,
@@ -747,6 +751,10 @@ def parseOpts(overrideArguments=None):
         '--exec',
         metavar='CMD', dest='exec_cmd',
         help='Execute a command on the file after downloading, similar to find\'s -exec syntax. Example: --exec \'adb push {} /sdcard/Music/ && rm {}\'')
+    postproc.add_option(
+        '--convert-subtitles', '--convert-subs',
+        metavar='FORMAT', dest='convertsubtitles', default=None,
+        help='Convert the subtitles to other format (currently supported: srt|ass|vtt)')
 
     parser.add_option_group(general)
     parser.add_option_group(network)

youtube_dl/postprocessor/__init__.py

@@ -11,6 +11,7 @@ from .ffmpeg import (
     FFmpegMergerPP,
     FFmpegMetadataPP,
     FFmpegVideoConvertorPP,
+    FFmpegSubtitlesConvertorPP,
 )
 from .xattrpp import XAttrMetadataPP
 from .execafterdownload import ExecAfterDownloadPP
@@ -31,6 +32,7 @@ __all__ = [
     'FFmpegMergerPP',
     'FFmpegMetadataPP',
     'FFmpegPostProcessor',
+    'FFmpegSubtitlesConvertorPP',
     'FFmpegVideoConvertorPP',
     'XAttrMetadataPP',
 ]

youtube_dl/postprocessor/ffmpeg.py

@@ -1,5 +1,6 @@
 from __future__ import unicode_literals
 
+import io
 import os
 import subprocess
 import sys
@@ -635,3 +636,40 @@ class FFmpegFixupM4aPP(FFmpegPostProcessor):
         os.rename(encodeFilename(temp_filename), encodeFilename(filename))
 
         return True, info
+
+
+class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor):
+    def __init__(self, downloader=None, format=None):
+        super(FFmpegSubtitlesConvertorPP, self).__init__(downloader)
+        self.format = format
+
+    def run(self, info):
+        subs = info.get('requested_subtitles')
+        filename = info['filepath']
+        new_ext = self.format
+        new_format = new_ext
+        if new_format == 'vtt':
+            new_format = 'webvtt'
+        if subs is None:
+            self._downloader.to_screen('[ffmpeg] There aren\'t any subtitles to convert')
+            return True, info
+        self._downloader.to_screen('[ffmpeg] Converting subtitles')
+        for lang, sub in subs.items():
+            ext = sub['ext']
+            if ext == new_ext:
+                self._downloader.to_screen(
+                    '[ffmpeg] Subtitle file for %s is already in the requested'
+                    'format' % new_ext)
+                continue
+            new_file = subtitles_filename(filename, lang, new_ext)
+            self.run_ffmpeg(
+                subtitles_filename(filename, lang, ext),
+                new_file, ['-f', new_format])
+
+            with io.open(new_file, 'rt', encoding='utf-8') as f:
+                subs[lang] = {
+                    'ext': ext,
+                    'data': f.read(),
+                }
+
+        return True, info

youtube_dl/utils.py

@@ -54,7 +54,7 @@ from .compat import (
 compiled_regex_type = type(re.compile(''))
 
 std_headers = {
-    'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20100101 Firefox/10.0 (Chrome)',
+    'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/20.0 (Chrome)',
     'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
     'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
     'Accept-Encoding': 'gzip, deflate',
@@ -1290,6 +1290,7 @@ def parse_duration(s):
             (?P<only_mins>[0-9.]+)\s*(?:mins?|minutes?)\s*|
             (?P<only_hours>[0-9.]+)\s*(?:hours?)|
 
+            \s*(?P<hours_reversed>[0-9]+)\s*(?:[:h]|hours?)\s*(?P<mins_reversed>[0-9]+)\s*(?:[:m]|mins?|minutes?)\s*|
             (?:
                 (?:
                     (?:(?P<days>[0-9]+)\s*(?:[:d]|days?)\s*)?
@@ -1308,10 +1309,14 @@ def parse_duration(s):
         return float_or_none(m.group('only_hours'), invscale=60 * 60)
     if m.group('secs'):
         res += int(m.group('secs'))
+    if m.group('mins_reversed'):
+        res += int(m.group('mins_reversed')) * 60
     if m.group('mins'):
         res += int(m.group('mins')) * 60
     if m.group('hours'):
         res += int(m.group('hours')) * 60 * 60
+    if m.group('hours_reversed'):
+        res += int(m.group('hours_reversed')) * 60 * 60
     if m.group('days'):
         res += int(m.group('days')) * 24 * 60 * 60
     if m.group('ms'):

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.02.24.2'
+__version__ = '2015.02.28'