release 2016.07.13

SOCKS tests never run on Travis CI due to unknown reasons, and
downloading them broke some tests (e.g.
https://travis-ci.org/rg3/youtube-dl/builds/144306425)

.github/ISSUE_TEMPLATE.md

@@ -6,8 +6,8 @@
 
 ---
 
-### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.06.03*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.07.13*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.06.03**
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.07.13**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2016.06.03
+[debug] youtube-dl version 2016.07.13
 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2
 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
 [debug] Proxy map: {}

.github/PULL_REQUEST_TEMPLATE.md

@@ -0,0 +1,22 @@
+## Please follow the guide below
+
+- You will be asked some questions, please read them **carefully** and answer honestly
+- Put an `x` into all the boxes [ ] relevant to your *pull request* (like that [x])
+- Use *Preview* tab to see how your *pull request* will actually look like
+
+---
+
+### Before submitting a *pull request* make sure you have:
+- [ ] At least skimmed through [adding new extractor tutorial](https://github.com/rg3/youtube-dl#adding-support-for-a-new-site) and [youtube-dl coding conventions](https://github.com/rg3/youtube-dl#youtube-dl-coding-conventions) sections
+- [ ] [Searched](https://github.com/rg3/youtube-dl/search?q=is%3Apr&type=Issues) the bugtracker for similar pull requests
+
+### What is the purpose of your *pull request*?
+- [ ] Bug fix
+- [ ] New extractor
+- [ ] New feature
+
+---
+
+### Description of your *pull request* and other information
+
+Explanation of your *pull request* in arbitrary form goes here. Please make sure the description explains the purpose and effect of your *pull request* and is worded well enough to be understood. Provide as much context and examples as possible.

.travis.yml

@@ -7,9 +7,6 @@ python:
   - "3.4"
   - "3.5"
 sudo: false
-install:
-  - bash ./devscripts/install_srelay.sh
-  - export PATH=$PATH:$(pwd)/tmp/srelay-0.4.8b6
 script: nosetests test --verbose
 notifications:
   email:

AUTHORS

@@ -173,3 +173,8 @@ Kevin Deldycke
 inondle
 Tomáš Čech
 Déstin Reed
+Roman Tsiupa
+Artur Krysiak
+Jakub Adam Wieczorek
+Aleksandar Topuzović
+Nehal Patel

CONTRIBUTING.md

@@ -97,9 +97,17 @@ If you want to add support for a new site, first of all **make sure** this site
 After you have ensured this site is distributing it's content legally, you can follow this quick list (assuming your service is called `yourextractor`):
 
 1. [Fork this repository](https://github.com/rg3/youtube-dl/fork)
-2. Check out the source code with `git clone git@github.com:YOUR_GITHUB_USERNAME/youtube-dl.git`
+2. Check out the source code with:
-3. Start a new git branch with `cd youtube-dl; git checkout -b yourextractor`
+
+        git clone git@github.com:YOUR_GITHUB_USERNAME/youtube-dl.git
+
+3. Start a new git branch with
+
+        cd youtube-dl
+        git checkout -b yourextractor
+
 4. Start with this simple template and save it to `youtube_dl/extractor/yourextractor.py`:
+
     ```python
     # coding: utf-8
     from __future__ import unicode_literals
@@ -142,17 +150,149 @@ After you have ensured this site is distributing it's content legally, you can f
     ```
 5. Add an import in [`youtube_dl/extractor/extractors.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/extractors.py).
 6. Run `python test/test_download.py TestDownload.test_YourExtractor`. This *should fail* at first, but you can continually re-run it until you're done. If you decide to add more than one test, then rename ``_TEST`` to ``_TESTS`` and make it into a list of dictionaries. The tests will then be named `TestDownload.test_YourExtractor`, `TestDownload.test_YourExtractor_1`, `TestDownload.test_YourExtractor_2`, etc.
-7. Have a look at [`youtube_dl/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 and may return](https://github.com/rg3/youtube-dl/blob/58525c94d547be1c8167d16c298bdd75506db328/youtube_dl/extractor/common.py#L68-L226). Add tests and code for as many as you want.
+7. Have a look at [`youtube_dl/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 and may return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L74-L252). Add tests and code for as many as you want.
-8. Keep in mind that the only mandatory fields in info dict for successful extraction process are `id`, `title` and either `url` or `formats`, i.e. these are the critical data the extraction does not make any sense without. This means that [any field](https://github.com/rg3/youtube-dl/blob/58525c94d547be1c8167d16c298bdd75506db328/youtube_dl/extractor/common.py#L138-L226) apart from aforementioned mandatory ones should be treated **as optional** and extraction should be **tolerate** to situations when sources for these fields can potentially be unavailable (even if they always available at the moment) and **future-proof** in order not to break the extraction of general purpose mandatory fields. For example, if you have some intermediate dict `meta` that is a source of metadata and it has a key `summary` that you want to extract and put into resulting info dict as `description`, you should be ready that this key may be missing from the `meta` dict, i.e. you should extract it as `meta.get('summary')` and not `meta['summary']`. Similarly, you should pass `fatal=False` when extracting data from a webpage with `_search_regex/_html_search_regex`.
+8. Make sure your code follows [youtube-dl coding conventions](#youtube-dl-coding-conventions) and check the code with [flake8](https://pypi.python.org/pypi/flake8). Also make sure your code works under all [Python](http://www.python.org/) versions claimed supported by youtube-dl, namely 2.6, 2.7, and 3.2+.
-9. Check the code with [flake8](https://pypi.python.org/pypi/flake8).
+9. When the tests pass, [add](http://git-scm.com/docs/git-add) the new files and [commit](http://git-scm.com/docs/git-commit) them and [push](http://git-scm.com/docs/git-push) the result, like this:
-10. When the tests pass, [add](http://git-scm.com/docs/git-add) the new files and [commit](http://git-scm.com/docs/git-commit) them and [push](http://git-scm.com/docs/git-push) the result, like this:
 
         $ git add youtube_dl/extractor/extractors.py
         $ git add youtube_dl/extractor/yourextractor.py
         $ git commit -m '[yourextractor] Add new extractor'
         $ git push origin yourextractor
 
-11. Finally, [create a pull request](https://help.github.com/articles/creating-a-pull-request). We'll then review and merge it.
+10. Finally, [create a pull request](https://help.github.com/articles/creating-a-pull-request). We'll then review and merge it.
 
 In any case, thank you very much for your contributions!
 
+## youtube-dl coding conventions
+
+This section introduces a guide lines for writing idiomatic, robust and future-proof extractor code.
+
+Extractors are very fragile by nature since they depend on the layout of the source data provided by 3rd party media hoster out of your control and this layout tend to change. As an extractor implementer your task is not only to write code that will extract media links and metadata correctly but also to minimize code dependency on source's layout changes and even to make the code foresee potential future changes and be ready for that. This is important because it will allow extractor not to break on minor layout changes thus keeping old youtube-dl versions working. Even though this breakage issue is easily fixed by emitting a new version of youtube-dl with fix incorporated all the previous version become broken in all repositories and distros' packages that may not be so prompt in fetching the update from us. Needless to say some may never receive an update at all that is possible for non rolling release distros.
+
+### Mandatory and optional metafields
+
+For extraction to work youtube-dl relies on metadata your extractor extracts and provides to youtube-dl expressed by [information dictionary](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L75-L257) or simply *info dict*. Only the following meta fields in *info dict* are considered mandatory for successful extraction process by youtube-dl:
+
+ - `id` (media identifier)
+ - `title` (media title)
+ - `url` (media download URL) or `formats`
+
+In fact only the last option is technically mandatory (i.e. if you can't figure out the download location of the media the extraction does not make any sense). But by convention youtube-dl also treats `id` and `title` to be mandatory. Thus aforementioned metafields are the critical data the extraction does not make any sense without and if any of them fail to be extracted then extractor is considered completely broken.
+
+[Any field](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L149-L257) apart from the aforementioned ones are considered **optional**. That means that extraction should be **tolerate** to situations when sources for these fields can potentially be unavailable (even if they are always available at the moment) and **future-proof** in order not to break the extraction of general purpose mandatory fields.
+
+#### Example
+
+Say you have some source dictionary `meta` that you've fetched as JSON with HTTP request and it has a key `summary`:
+
+```python
+meta = self._download_json(url, video_id)
+```
+    
+Assume at this point `meta`'s layout is:
+
+```python
+{
+    ...
+    "summary": "some fancy summary text",
+    ...
+}
+```
+
+Assume you want to extract `summary` and put into resulting info dict as `description`. Since `description` is optional metafield you should be ready that this key may be missing from the `meta` dict, so that you should extract it like:
+
+```python
+description = meta.get('summary')  # correct
+```
+
+and not like:
+
+```python
+description = meta['summary']  # incorrect
+```
+
+The latter will break extraction process with `KeyError` if `summary` disappears from `meta` at some time later but with former approach extraction will just go ahead with `description` set to `None` that is perfectly fine (remember `None` is equivalent for absence of data). 
+
+Similarly, you should pass `fatal=False` when extracting optional data from a webpage with `_search_regex`, `_html_search_regex` or similar methods, for instance:
+
+```python
+description = self._search_regex(
+    r'<span[^>]+id="title"[^>]*>([^<]+)<',
+    webpage, 'description', fatal=False)
+```
+
+With `fatal` set to `False` if `_search_regex` fails to extract `description` it will emit a warning and continue extraction.
+
+You can also pass `default=<some fallback value>`, for example:
+
+```python
+description = self._search_regex(
+    r'<span[^>]+id="title"[^>]*>([^<]+)<',
+    webpage, 'description', default=None)
+```
+
+On failure this code will silently continue the extraction with `description` set to `None`. That is useful for metafields that are known to may or may not be present.
+ 
+### Provide fallbacks
+
+When extracting metadata try to provide several scenarios for that. For example if `title` is present in several places/sources try extracting from at least some of them. This would make it more future-proof in case some of the sources became unavailable.
+
+#### Example
+
+Say `meta` from previous example has a `title` and you are about to extract it. Since `title` is mandatory meta field you should end up with something like:
+
+```python
+title = meta['title']
+```
+
+If `title` disappeares from `meta` in future due to some changes on hoster's side the extraction would fail since `title` is mandatory. That's expected.
+
+Assume that you have some another source you can extract `title` from, for example `og:title` HTML meta of a `webpage`. In this case you can provide a fallback scenario:
+
+```python
+title = meta.get('title') or self._og_search_title(webpage)
+```
+
+This code will try to extract from `meta` first and if it fails it will try extracting `og:title` from a `webpage`.
+
+### Make regular expressions flexible
+
+When using regular expressions try to write them fuzzy and flexible.
+ 
+#### Example
+
+Say you need to extract `title` from the following HTML code:
+
+```html
+<span style="position: absolute; left: 910px; width: 90px; float: right; z-index: 9999;" class="title">some fancy title</span>
+```
+
+The code for that task should look similar to:
+
+```python
+title = self._search_regex(
+    r'<span[^>]+class="title"[^>]*>([^<]+)', webpage, 'title')
+```
+
+Or even better:
+
+```python
+title = self._search_regex(
+    r'<span[^>]+class=(["\'])title\1[^>]*>(?P<title>[^<]+)',
+    webpage, 'title', group='title')
+```
+
+Note how you tolerate potential changes in `style` attribute's value or switch from using double quotes to single for `class` attribute: 
+
+The code definitely should not look like:
+
+```python
+title = self._search_regex(
+    r'<span style="position: absolute; left: 910px; width: 90px; float: right; z-index: 9999;" class="title">(.*?)</span>',
+    webpage, 'title', group='title')
+```
+
+### Use safe conversion functions
+
+Wrap all extracted numeric data into safe functions from `utils`: `int_or_none`, `float_or_none`. Use them for string to number conversions as well.
+

README.md

@@ -17,7 +17,7 @@ youtube-dl - download videos from youtube.com or other video platforms
 
 To install it right away for all UNIX users (Linux, OS X, etc.), type:
 
-    sudo curl https://yt-dl.org/latest/youtube-dl -o /usr/local/bin/youtube-dl
+    sudo curl -L https://yt-dl.org/latest/youtube-dl -o /usr/local/bin/youtube-dl
     sudo chmod a+rx /usr/local/bin/youtube-dl
 
 If you do not have curl, you can alternatively use a recent wget:
@@ -27,18 +27,24 @@ If you do not have curl, you can alternatively use a recent wget:
 
 Windows users can [download an .exe file](https://yt-dl.org/latest/youtube-dl.exe) and place it in any location on their [PATH](http://en.wikipedia.org/wiki/PATH_%28variable%29) except for `%SYSTEMROOT%\System32` (e.g. **do not** put in `C:\Windows\System32`).
 
-OS X users can install **youtube-dl** with [Homebrew](http://brew.sh/).
+You can also use pip:
+
+    sudo pip install --upgrade youtube-dl
+    
+This command will update youtube-dl if you have already installed it. See the [pypi page](https://pypi.python.org/pypi/youtube_dl) for more information.
+
+OS X users can install youtube-dl with [Homebrew](http://brew.sh/):
 
     brew install youtube-dl
 
-You can also use pip:
+Or with [MacPorts](https://www.macports.org/):
 
-    sudo pip install youtube-dl
+    sudo port install youtube-dl
 
 Alternatively, refer to the [developer instructions](#developer-instructions) for how to check out and work with the git repository. For further options, including PGP signatures, see the [youtube-dl Download Page](https://rg3.github.io/youtube-dl/download.html).
 
 # DESCRIPTION
-**youtube-dl** is a small command-line program to download videos from
+**youtube-dl** is a command-line program to download videos from
 YouTube.com and a few more sites. It requires the Python interpreter, version
 2.6, 2.7, or 3.2+, and it is not platform specific. It should work on
 your Unix box, on Windows or on Mac OS X. It is released to the public domain,
@@ -97,9 +103,9 @@ which means you can modify it, redistribute it or use it however you like.
                                      (experimental)
     -6, --force-ipv6                 Make all connections via IPv6
                                      (experimental)
-    --cn-verification-proxy URL      Use this proxy to verify the IP address for
+    --geo-verification-proxy URL     Use this proxy to verify the IP address for
-                                     some Chinese sites. The default proxy
+                                     some geo-restricted sites. The default
-                                     specified by --proxy (or none, if the
+                                     proxy specified by --proxy (or none, if the
                                      options is not present) is used for the
                                      actual downloading. (experimental)
 
@@ -249,7 +255,7 @@ which means you can modify it, redistribute it or use it however you like.
     --write-info-json                Write video metadata to a .info.json file
     --write-annotations              Write video annotations to a
                                      .annotations.xml file
-    --load-info FILE                 JSON file containing the video information
+    --load-info-json FILE            JSON file containing the video information
                                      (created with the "--write-info-json"
                                      option)
     --cookies FILE                   File to read cookies from and dump cookie
@@ -418,7 +424,7 @@ which means you can modify it, redistribute it or use it however you like.
 
 # CONFIGURATION
 
-You can configure youtube-dl by placing any supported command line option to a configuration file. On Linux and OS X, the system wide configuration file is located at `/etc/youtube-dl.conf` and the user wide configuration file at `~/.config/youtube-dl/config`. On Windows, the user wide configuration file locations are `%APPDATA%\youtube-dl\config.txt` or `C:\Users\<user name>\youtube-dl.conf`.
+You can configure youtube-dl by placing any supported command line option to a configuration file. On Linux and OS X, the system wide configuration file is located at `/etc/youtube-dl.conf` and the user wide configuration file at `~/.config/youtube-dl/config`. On Windows, the user wide configuration file locations are `%APPDATA%\youtube-dl\config.txt` or `C:\Users\<user name>\youtube-dl.conf`. Note that by default configuration file may not exist so you may need to create it yourself.
 
 For example, with the following configuration file youtube-dl will always extract the audio, not copy the mtime, use a proxy and save all videos under `Movies` directory in your home directory:
 ```
@@ -426,6 +432,7 @@ For example, with the following configuration file youtube-dl will always extrac
 --no-mtime
 --proxy 127.0.0.1:3128
 -o ~/Movies/%(title)s.%(ext)s
+# Lines starting with # are comments
 ```
 
 Note that options in configuration file are just the same options aka switches used in regular command line calls thus there **must be no whitespace** after `-` or `--`, e.g. `-o` or `--proxy` but not `- o` or `-- proxy`.
@@ -505,6 +512,9 @@ The basic usage is not to set any template arguments when downloading a single f
  - `autonumber`: Five-digit number that will be increased with each download, starting at zero
  - `playlist`: Name or id of the playlist that contains the video
  - `playlist_index`: Index of the video in the playlist padded with leading zeros according to the total length of the playlist
+ - `playlist_id`: Playlist identifier
+ - `playlist_title`: Playlist title
+
 
 Available for the video that belongs to some logical chapter or section:
  - `chapter`: Name or title of the chapter the video belongs to
@@ -544,6 +554,10 @@ The current default template is `%(title)s-%(id)s.%(ext)s`.
 
 In some cases, you don't want special characters such as 中, spaces, or &, such as when transferring the downloaded filename to a Windows system or the filename through an 8bit-unsafe channel. In these cases, add the `--restrict-filenames` flag to get a shorter title:
 
+#### Output template and Windows batch files
+
+If you are using output template inside a Windows batch file then you must escape plain percent characters (`%`) by doubling, so that `-o "%(title)s-%(id)s.%(ext)s"` should become `-o "%%(title)s-%%(id)s.%%(ext)s"`. However you should not touch `%`'s that are not plain characters, e.g. environment variables for expansion should stay intact: `-o "C:\%HOMEPATH%\Desktop\%%(title)s.%%(ext)s"`.
+
 #### Output template examples
 
 Note on Windows you may need to use double quotes instead of single.
@@ -842,6 +856,12 @@ It is *not* possible to detect whether a URL is supported or not. That's because
 
 If you want to find out whether a given URL is supported, simply call youtube-dl with it. If you get no videos back, chances are the URL is either not referring to a video or unsupported. You can find out which by examining the output (if you run youtube-dl on the console) or catching an `UnsupportedError` exception if you run it from a Python program.
 
+# Why do I need to go through that much red tape when filing bugs?
+
+Before we had the issue template, despite our extensive [bug reporting instructions](#bugs), about 80% of the issue reports we got were useless, for instance because people used ancient versions hundreds of releases old, because of simple syntactic errors (not in youtube-dl but in general shell usage), because the problem was alrady reported multiple times before, because people did not actually read an error message, even if it said "please install ffmpeg", because people did not mention the URL they were trying to download and many more simple, easy-to-avoid problems, many of whom were totally unrelated to youtube-dl.
+
+youtube-dl is an open-source project manned by too few volunteers, so we'd rather spend time fixing bugs where we are certain none of those simple problems apply, and where we can be reasonably confident to be able to reproduce the issue without asking the reporter repeatedly. As such, the output of `youtube-dl -v YOUR_URL_HERE` is really all that's required to file an issue. The issue template also guides you through some basic steps you can do, such as checking that your version of youtube-dl is current.
+
 # DEVELOPER INSTRUCTIONS
 
 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.
@@ -871,9 +891,17 @@ If you want to add support for a new site, first of all **make sure** this site
 After you have ensured this site is distributing it's content legally, you can follow this quick list (assuming your service is called `yourextractor`):
 
 1. [Fork this repository](https://github.com/rg3/youtube-dl/fork)
-2. Check out the source code with `git clone git@github.com:YOUR_GITHUB_USERNAME/youtube-dl.git`
+2. Check out the source code with:
-3. Start a new git branch with `cd youtube-dl; git checkout -b yourextractor`
+
+        git clone git@github.com:YOUR_GITHUB_USERNAME/youtube-dl.git
+
+3. Start a new git branch with
+
+        cd youtube-dl
+        git checkout -b yourextractor
+
 4. Start with this simple template and save it to `youtube_dl/extractor/yourextractor.py`:
+
     ```python
     # coding: utf-8
     from __future__ import unicode_literals
@@ -916,20 +944,152 @@ After you have ensured this site is distributing it's content legally, you can f
     ```
 5. Add an import in [`youtube_dl/extractor/extractors.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/extractors.py).
 6. Run `python test/test_download.py TestDownload.test_YourExtractor`. This *should fail* at first, but you can continually re-run it until you're done. If you decide to add more than one test, then rename ``_TEST`` to ``_TESTS`` and make it into a list of dictionaries. The tests will then be named `TestDownload.test_YourExtractor`, `TestDownload.test_YourExtractor_1`, `TestDownload.test_YourExtractor_2`, etc.
-7. Have a look at [`youtube_dl/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 and may return](https://github.com/rg3/youtube-dl/blob/58525c94d547be1c8167d16c298bdd75506db328/youtube_dl/extractor/common.py#L68-L226). Add tests and code for as many as you want.
+7. Have a look at [`youtube_dl/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 and may return](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L74-L252). Add tests and code for as many as you want.
-8. Keep in mind that the only mandatory fields in info dict for successful extraction process are `id`, `title` and either `url` or `formats`, i.e. these are the critical data the extraction does not make any sense without. This means that [any field](https://github.com/rg3/youtube-dl/blob/58525c94d547be1c8167d16c298bdd75506db328/youtube_dl/extractor/common.py#L138-L226) apart from aforementioned mandatory ones should be treated **as optional** and extraction should be **tolerate** to situations when sources for these fields can potentially be unavailable (even if they always available at the moment) and **future-proof** in order not to break the extraction of general purpose mandatory fields. For example, if you have some intermediate dict `meta` that is a source of metadata and it has a key `summary` that you want to extract and put into resulting info dict as `description`, you should be ready that this key may be missing from the `meta` dict, i.e. you should extract it as `meta.get('summary')` and not `meta['summary']`. Similarly, you should pass `fatal=False` when extracting data from a webpage with `_search_regex/_html_search_regex`.
+8. Make sure your code follows [youtube-dl coding conventions](#youtube-dl-coding-conventions) and check the code with [flake8](https://pypi.python.org/pypi/flake8). Also make sure your code works under all [Python](http://www.python.org/) versions claimed supported by youtube-dl, namely 2.6, 2.7, and 3.2+.
-9. Check the code with [flake8](https://pypi.python.org/pypi/flake8).
+9. When the tests pass, [add](http://git-scm.com/docs/git-add) the new files and [commit](http://git-scm.com/docs/git-commit) them and [push](http://git-scm.com/docs/git-push) the result, like this:
-10. When the tests pass, [add](http://git-scm.com/docs/git-add) the new files and [commit](http://git-scm.com/docs/git-commit) them and [push](http://git-scm.com/docs/git-push) the result, like this:
 
         $ git add youtube_dl/extractor/extractors.py
         $ git add youtube_dl/extractor/yourextractor.py
         $ git commit -m '[yourextractor] Add new extractor'
         $ git push origin yourextractor
 
-11. Finally, [create a pull request](https://help.github.com/articles/creating-a-pull-request). We'll then review and merge it.
+10. Finally, [create a pull request](https://help.github.com/articles/creating-a-pull-request). We'll then review and merge it.
 
 In any case, thank you very much for your contributions!
 
+## youtube-dl coding conventions
+
+This section introduces a guide lines for writing idiomatic, robust and future-proof extractor code.
+
+Extractors are very fragile by nature since they depend on the layout of the source data provided by 3rd party media hoster out of your control and this layout tend to change. As an extractor implementer your task is not only to write code that will extract media links and metadata correctly but also to minimize code dependency on source's layout changes and even to make the code foresee potential future changes and be ready for that. This is important because it will allow extractor not to break on minor layout changes thus keeping old youtube-dl versions working. Even though this breakage issue is easily fixed by emitting a new version of youtube-dl with fix incorporated all the previous version become broken in all repositories and distros' packages that may not be so prompt in fetching the update from us. Needless to say some may never receive an update at all that is possible for non rolling release distros.
+
+### Mandatory and optional metafields
+
+For extraction to work youtube-dl relies on metadata your extractor extracts and provides to youtube-dl expressed by [information dictionary](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L75-L257) or simply *info dict*. Only the following meta fields in *info dict* are considered mandatory for successful extraction process by youtube-dl:
+
+ - `id` (media identifier)
+ - `title` (media title)
+ - `url` (media download URL) or `formats`
+
+In fact only the last option is technically mandatory (i.e. if you can't figure out the download location of the media the extraction does not make any sense). But by convention youtube-dl also treats `id` and `title` to be mandatory. Thus aforementioned metafields are the critical data the extraction does not make any sense without and if any of them fail to be extracted then extractor is considered completely broken.
+
+[Any field](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/extractor/common.py#L149-L257) apart from the aforementioned ones are considered **optional**. That means that extraction should be **tolerate** to situations when sources for these fields can potentially be unavailable (even if they are always available at the moment) and **future-proof** in order not to break the extraction of general purpose mandatory fields.
+
+#### Example
+
+Say you have some source dictionary `meta` that you've fetched as JSON with HTTP request and it has a key `summary`:
+
+```python
+meta = self._download_json(url, video_id)
+```
+    
+Assume at this point `meta`'s layout is:
+
+```python
+{
+    ...
+    "summary": "some fancy summary text",
+    ...
+}
+```
+
+Assume you want to extract `summary` and put into resulting info dict as `description`. Since `description` is optional metafield you should be ready that this key may be missing from the `meta` dict, so that you should extract it like:
+
+```python
+description = meta.get('summary')  # correct
+```
+
+and not like:
+
+```python
+description = meta['summary']  # incorrect
+```
+
+The latter will break extraction process with `KeyError` if `summary` disappears from `meta` at some time later but with former approach extraction will just go ahead with `description` set to `None` that is perfectly fine (remember `None` is equivalent for absence of data). 
+
+Similarly, you should pass `fatal=False` when extracting optional data from a webpage with `_search_regex`, `_html_search_regex` or similar methods, for instance:
+
+```python
+description = self._search_regex(
+    r'<span[^>]+id="title"[^>]*>([^<]+)<',
+    webpage, 'description', fatal=False)
+```
+
+With `fatal` set to `False` if `_search_regex` fails to extract `description` it will emit a warning and continue extraction.
+
+You can also pass `default=<some fallback value>`, for example:
+
+```python
+description = self._search_regex(
+    r'<span[^>]+id="title"[^>]*>([^<]+)<',
+    webpage, 'description', default=None)
+```
+
+On failure this code will silently continue the extraction with `description` set to `None`. That is useful for metafields that are known to may or may not be present.
+ 
+### Provide fallbacks
+
+When extracting metadata try to provide several scenarios for that. For example if `title` is present in several places/sources try extracting from at least some of them. This would make it more future-proof in case some of the sources became unavailable.
+
+#### Example
+
+Say `meta` from previous example has a `title` and you are about to extract it. Since `title` is mandatory meta field you should end up with something like:
+
+```python
+title = meta['title']
+```
+
+If `title` disappeares from `meta` in future due to some changes on hoster's side the extraction would fail since `title` is mandatory. That's expected.
+
+Assume that you have some another source you can extract `title` from, for example `og:title` HTML meta of a `webpage`. In this case you can provide a fallback scenario:
+
+```python
+title = meta.get('title') or self._og_search_title(webpage)
+```
+
+This code will try to extract from `meta` first and if it fails it will try extracting `og:title` from a `webpage`.
+
+### Make regular expressions flexible
+
+When using regular expressions try to write them fuzzy and flexible.
+ 
+#### Example
+
+Say you need to extract `title` from the following HTML code:
+
+```html
+<span style="position: absolute; left: 910px; width: 90px; float: right; z-index: 9999;" class="title">some fancy title</span>
+```
+
+The code for that task should look similar to:
+
+```python
+title = self._search_regex(
+    r'<span[^>]+class="title"[^>]*>([^<]+)', webpage, 'title')
+```
+
+Or even better:
+
+```python
+title = self._search_regex(
+    r'<span[^>]+class=(["\'])title\1[^>]*>(?P<title>[^<]+)',
+    webpage, 'title', group='title')
+```
+
+Note how you tolerate potential changes in `style` attribute's value or switch from using double quotes to single for `class` attribute: 
+
+The code definitely should not look like:
+
+```python
+title = self._search_regex(
+    r'<span style="position: absolute; left: 910px; width: 90px; float: right; z-index: 9999;" class="title">(.*?)</span>',
+    webpage, 'title', group='title')
+```
+
+### Use safe conversion functions
+
+Wrap all extracted numeric data into safe functions from `utils`: `int_or_none`, `float_or_none`. Use them for string to number conversions as well.
+
 # EMBEDDING YOUTUBE-DL
 
 youtube-dl makes the best effort to be a good command-line program, and thus should be callable from any programming language. If you encounter any problems parsing its output, feel free to [create a report](https://github.com/rg3/youtube-dl/issues/new).
@@ -945,7 +1105,7 @@ with youtube_dl.YoutubeDL(ydl_opts) as ydl:
     ydl.download(['http://www.youtube.com/watch?v=BaW_jenozKc'])
 ```
 
-Most likely, you'll want to use various options. For a list of what can be done, have a look at [`youtube_dl/YoutubeDL.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/YoutubeDL.py#L121-L269). For a start, if you want to intercept youtube-dl's output, set a `logger` object.
+Most likely, you'll want to use various options. For a list of options available, have a look at [`youtube_dl/YoutubeDL.py`](https://github.com/rg3/youtube-dl/blob/master/youtube_dl/YoutubeDL.py#L128-L278). For a start, if you want to intercept youtube-dl's output, set a `logger` object.
 
 Here's a more complete example of a program that outputs only errors (and a short message after the download is finished), and downloads/converts the video to an mp3 file:
 

devscripts/buildserver.py

@@ -13,6 +13,7 @@ import os.path
 
 sys.path.insert(0, os.path.dirname(os.path.dirname((os.path.abspath(__file__)))))
 from youtube_dl.compat import (
+    compat_input,
     compat_http_server,
     compat_str,
     compat_urlparse,
@@ -30,11 +31,6 @@ try:
 except ImportError:  # Python 2
     import SocketServer as compat_socketserver
 
-try:
-    compat_input = raw_input
-except NameError:  # Python 3
-    compat_input = input
-
 
 class BuildHTTPServer(compat_socketserver.ThreadingMixIn, compat_http_server.HTTPServer):
     allow_reuse_address = True

devscripts/create-github-release.py

@@ -0,0 +1,111 @@
+#!/usr/bin/env python
+from __future__ import unicode_literals
+
+import base64
+import json
+import mimetypes
+import netrc
+import optparse
+import os
+import sys
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.compat import (
+    compat_basestring,
+    compat_input,
+    compat_getpass,
+    compat_print,
+    compat_urllib_request,
+)
+from youtube_dl.utils import (
+    make_HTTPS_handler,
+    sanitized_Request,
+)
+
+
+class GitHubReleaser(object):
+    _API_URL = 'https://api.github.com/repos/rg3/youtube-dl/releases'
+    _UPLOADS_URL = 'https://uploads.github.com/repos/rg3/youtube-dl/releases/%s/assets?name=%s'
+    _NETRC_MACHINE = 'github.com'
+
+    def __init__(self, debuglevel=0):
+        self._init_github_account()
+        https_handler = make_HTTPS_handler({}, debuglevel=debuglevel)
+        self._opener = compat_urllib_request.build_opener(https_handler)
+
+    def _init_github_account(self):
+        try:
+            info = netrc.netrc().authenticators(self._NETRC_MACHINE)
+            if info is not None:
+                self._username = info[0]
+                self._password = info[2]
+                compat_print('Using GitHub credentials found in .netrc...')
+                return
+            else:
+                compat_print('No GitHub credentials found in .netrc')
+        except (IOError, netrc.NetrcParseError):
+            compat_print('Unable to parse .netrc')
+        self._username = compat_input(
+            'Type your GitHub username or email address and press [Return]: ')
+        self._password = compat_getpass(
+            'Type your GitHub password and press [Return]: ')
+
+    def _call(self, req):
+        if isinstance(req, compat_basestring):
+            req = sanitized_Request(req)
+        # Authorizing manually since GitHub does not response with 401 with
+        # WWW-Authenticate header set (see
+        # https://developer.github.com/v3/#basic-authentication)
+        b64 = base64.b64encode(
+            ('%s:%s' % (self._username, self._password)).encode('utf-8')).decode('ascii')
+        req.add_header('Authorization', 'Basic %s' % b64)
+        response = self._opener.open(req).read().decode('utf-8')
+        return json.loads(response)
+
+    def list_releases(self):
+        return self._call(self._API_URL)
+
+    def create_release(self, tag_name, name=None, body='', draft=False, prerelease=False):
+        data = {
+            'tag_name': tag_name,
+            'target_commitish': 'master',
+            'name': name,
+            'body': body,
+            'draft': draft,
+            'prerelease': prerelease,
+        }
+        req = sanitized_Request(self._API_URL, json.dumps(data).encode('utf-8'))
+        return self._call(req)
+
+    def create_asset(self, release_id, asset):
+        asset_name = os.path.basename(asset)
+        url = self._UPLOADS_URL % (release_id, asset_name)
+        # Our files are small enough to be loaded directly into memory.
+        data = open(asset, 'rb').read()
+        req = sanitized_Request(url, data)
+        mime_type, _ = mimetypes.guess_type(asset_name)
+        req.add_header('Content-Type', mime_type or 'application/octet-stream')
+        return self._call(req)
+
+
+def main():
+    parser = optparse.OptionParser(usage='%prog VERSION BUILDPATH')
+    options, args = parser.parse_args()
+    if len(args) != 2:
+        parser.error('Expected a version and a build directory')
+
+    version, build_path = args
+
+    releaser = GitHubReleaser()
+
+    new_release = releaser.create_release(version, name='youtube-dl %s' % version)
+    release_id = new_release['id']
+
+    for asset in os.listdir(build_path):
+        compat_print('Uploading %s...' % asset)
+        releaser.create_asset(release_id, os.path.join(build_path, asset))
+
+
+if __name__ == '__main__':
+    main()

devscripts/gh-pages/generate-download.py

@@ -15,13 +15,9 @@ data = urllib.request.urlopen(URL).read()
 with open('download.html.in', 'r', encoding='utf-8') as tmplf:
     template = tmplf.read()
 
-md5sum = hashlib.md5(data).hexdigest()
-sha1sum = hashlib.sha1(data).hexdigest()
 sha256sum = hashlib.sha256(data).hexdigest()
 template = template.replace('@PROGRAM_VERSION@', version)
 template = template.replace('@PROGRAM_URL@', URL)
-template = template.replace('@PROGRAM_MD5SUM@', md5sum)
-template = template.replace('@PROGRAM_SHA1SUM@', sha1sum)
 template = template.replace('@PROGRAM_SHA256SUM@', sha256sum)
 template = template.replace('@EXE_URL@', versions_info['versions'][version]['exe'][0])
 template = template.replace('@EXE_SHA256SUM@', versions_info['versions'][version]['exe'][1])

devscripts/install_srelay.sh

@@ -1,8 +0,0 @@
-#!/bin/bash
-
-mkdir -p tmp && cd tmp
-wget -N http://downloads.sourceforge.net/project/socks-relay/socks-relay/srelay-0.4.8/srelay-0.4.8b6.tar.gz
-tar zxvf srelay-0.4.8b6.tar.gz
-cd srelay-0.4.8b6
-./configure
-make

devscripts/make_lazy_extractors.py

@@ -14,15 +14,17 @@ if os.path.exists(lazy_extractors_filename):
     os.remove(lazy_extractors_filename)
 
 from youtube_dl.extractor import _ALL_CLASSES
-from youtube_dl.extractor.common import InfoExtractor
+from youtube_dl.extractor.common import InfoExtractor, SearchInfoExtractor
 
 with open('devscripts/lazy_load_template.py', 'rt') as f:
     module_template = f.read()
 
-module_contents = [module_template + '\n' + getsource(InfoExtractor.suitable)]
+module_contents = [
+    module_template + '\n' + getsource(InfoExtractor.suitable) + '\n',
+    'class LazyLoadSearchExtractor(LazyLoadExtractor):\n    pass\n']
 
 ie_template = '''
-class {name}(LazyLoadExtractor):
+class {name}({bases}):
     _VALID_URL = {valid_url!r}
     _module = '{module}'
 '''
@@ -34,10 +36,20 @@ make_valid_template = '''
 '''
 
 
+def get_base_name(base):
+    if base is InfoExtractor:
+        return 'LazyLoadExtractor'
+    elif base is SearchInfoExtractor:
+        return 'LazyLoadSearchExtractor'
+    else:
+        return base.__name__
+
+
 def build_lazy_ie(ie, name):
     valid_url = getattr(ie, '_VALID_URL', None)
     s = ie_template.format(
         name=name,
+        bases=', '.join(map(get_base_name, ie.__bases__)),
         valid_url=valid_url,
         module=ie.__module__)
     if ie.suitable.__func__ is not InfoExtractor.suitable.__func__:
@@ -47,12 +59,35 @@ def build_lazy_ie(ie, name):
         s += make_valid_template.format(valid_url=ie._make_valid_url())
     return s
 
+# find the correct sorting and add the required base classes so that sublcasses
+# can be correctly created
+classes = _ALL_CLASSES[:-1]
+ordered_cls = []
+while classes:
+    for c in classes[:]:
+        bases = set(c.__bases__) - set((object, InfoExtractor, SearchInfoExtractor))
+        stop = False
+        for b in bases:
+            if b not in classes and b not in ordered_cls:
+                if b.__name__ == 'GenericIE':
+                    exit()
+                classes.insert(0, b)
+                stop = True
+        if stop:
+            break
+        if all(b in ordered_cls for b in bases):
+            ordered_cls.append(c)
+            classes.remove(c)
+            break
+ordered_cls.append(_ALL_CLASSES[-1])
+
 names = []
-for ie in list(sorted(_ALL_CLASSES[:-1], key=lambda cls: cls.ie_key())) + _ALL_CLASSES[-1:]:
+for ie in ordered_cls:
-    name = ie.ie_key() + 'IE'
+    name = ie.__name__
     src = build_lazy_ie(ie, name)
     module_contents.append(src)
-    names.append(name)
+    if ie in _ALL_CLASSES:
+        names.append(name)
 
 module_contents.append(
     '_ALL_CLASSES = [{0}]'.format(', '.join(names)))

devscripts/release.sh

@@ -15,6 +15,7 @@
 set -e
 
 skip_tests=true
+gpg_sign_commits=""
 buildserver='localhost:8142'
 
 while true
@@ -24,6 +25,10 @@ case "$1" in
         skip_tests=false
         shift
     ;;
+    --gpg-sign-commits|-S)
+        gpg_sign_commits="-S"
+        shift
+    ;;
     --buildserver)
         buildserver="$2"
         shift 2
@@ -69,7 +74,7 @@ sed -i "s/__version__ = '.*'/__version__ = '$version'/" youtube_dl/version.py
 /bin/echo -e "\n### Committing documentation, templates and youtube_dl/version.py..."
 make README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE.md supportedsites
 git add README.md CONTRIBUTING.md .github/ISSUE_TEMPLATE.md docs/supportedsites.md youtube_dl/version.py
-git commit -m "release $version"
+git commit $gpg_sign_commits -m "release $version"
 
 /bin/echo -e "\n### Now tagging, signing and pushing..."
 git tag -s -m "Release $version" "$version"
@@ -95,15 +100,16 @@ RELEASE_FILES="youtube-dl youtube-dl.exe youtube-dl-$version.tar.gz"
 (cd build/$version/ && sha256sum $RELEASE_FILES > SHA2-256SUMS)
 (cd build/$version/ && sha512sum $RELEASE_FILES > SHA2-512SUMS)
 
-/bin/echo -e "\n### Signing and uploading the new binaries to yt-dl.org ..."
+/bin/echo -e "\n### Signing and uploading the new binaries to GitHub..."
 for f in $RELEASE_FILES; do gpg --passphrase-repeat 5 --detach-sig "build/$version/$f"; done
-scp -r "build/$version" ytdl@yt-dl.org:html/tmp/
+
-ssh ytdl@yt-dl.org "mv html/tmp/$version html/downloads/"
+ROOT=$(pwd)
+python devscripts/create-github-release.py $version "$ROOT/build/$version"
+
 ssh ytdl@yt-dl.org "sh html/update_latest.sh $version"
 
 /bin/echo -e "\n### Now switching to gh-pages..."
 git clone --branch gh-pages --single-branch . build/gh-pages
-ROOT=$(pwd)
 (
     set -e
     ORIGIN_URL=$(git config --get remote.origin.url)
@@ -115,7 +121,7 @@ ROOT=$(pwd)
     "$ROOT/devscripts/gh-pages/update-copyright.py"
     "$ROOT/devscripts/gh-pages/update-sites.py"
     git add *.html *.html.in update
-    git commit -m "release $version"
+    git commit $gpg_sign_commits -m "release $version"
     git push "$ROOT" gh-pages
     git push "$ORIGIN_URL" gh-pages
 )

devscripts/show-downloads-statistics.py

@@ -0,0 +1,41 @@
+#!/usr/bin/env python
+from __future__ import unicode_literals
+
+import json
+import os
+import re
+import sys
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+
+from youtube_dl.compat import (
+    compat_print,
+    compat_urllib_request,
+)
+from youtube_dl.utils import format_bytes
+
+
+def format_size(bytes):
+    return '%s (%d bytes)' % (format_bytes(bytes), bytes)
+
+
+total_bytes = 0
+
+releases = json.loads(compat_urllib_request.urlopen(
+    'https://api.github.com/repos/rg3/youtube-dl/releases').read().decode('utf-8'))
+
+for release in releases:
+    compat_print(release['name'])
+    for asset in release['assets']:
+        asset_name = asset['name']
+        total_bytes += asset['download_count'] * asset['size']
+        if all(not re.match(p, asset_name) for p in (
+                r'^youtube-dl$',
+                r'^youtube-dl-\d{4}\.\d{2}\.\d{2}(?:\.\d+)?\.tar\.gz$',
+                r'^youtube-dl\.exe$')):
+            continue
+        compat_print(
+            ' %s size: %s downloads: %d'
+            % (asset_name, format_size(asset['size']), asset['download_count']))
+
+compat_print('total downloads traffic: %s' % format_size(total_bytes))

docs/supportedsites.md

@@ -28,6 +28,7 @@
  - **AdobeTVVideo**
  - **AdultSwim**
  - **aenetworks**: A+E Networks: A&E, Lifetime, History.com, FYI Network
+ - **AfreecaTV**: afreecatv.com
  - **Aftonbladet**
  - **AirMozilla**
  - **AlJazeera**
@@ -43,7 +44,6 @@
  - **appletrailers:section**
  - **archive.org**: archive.org videos
  - **ARD**
- - **ARD:mediathek**: Saarländischer Rundfunk
  - **ARD:mediathek**
  - **arte.tv**
  - **arte.tv:+7**
@@ -73,6 +73,8 @@
  - **bbc**: BBC
  - **bbc.co.uk**: BBC iPlayer
  - **bbc.co.uk:article**: BBC articles
+ - **bbc.co.uk:iplayer:playlist**
+ - **bbc.co.uk:playlist**
  - **BeatportPro**
  - **Beeg**
  - **BehindKink**
@@ -103,6 +105,8 @@
  - **canalc2.tv**
  - **Canalplus**: canalplus.fr, piwiplus.fr and d8.tv
  - **Canvas**
+ - **CarambaTV**
+ - **CarambaTVPage**
  - **CBC**
  - **CBCPlayer**
  - **CBS**
@@ -123,6 +127,7 @@
  - **cliphunter**
  - **ClipRs**
  - **Clipsyndicate**
+ - **CloserToTruth**
  - **cloudtime**: CloudTime
  - **Cloudy**
  - **Clubic**
@@ -147,6 +152,8 @@
  - **CSNNE**
  - **CSpan**: C-SPAN
  - **CtsNews**: 華視新聞
+ - **CTV**
+ - **CTVNews**
  - **culturebox.francetvinfo.fr**
  - **CultureUnplugged**
  - **CWTV**
@@ -217,6 +224,7 @@
  - **Firstpost**
  - **FiveTV**
  - **Flickr**
+ - **Flipagram**
  - **Folketinget**: Folketinget (ft.dk; Danish parliament)
  - **FootyRoom**
  - **Formula1**
@@ -235,6 +243,7 @@
  - **FreeVideo**
  - **Funimation**
  - **FunnyOrDie**
+ - **Fusion**
  - **GameInformer**
  - **Gamekings**
  - **GameOne**
@@ -242,7 +251,6 @@
  - **Gamersyde**
  - **GameSpot**
  - **GameStar**
- - **Gametrailers**
  - **Gazeta**
  - **GDCVault**
  - **generic**: Generic downloader that works on some sites
@@ -253,6 +261,7 @@
  - **Globo**
  - **GloboArticle**
  - **GodTube**
+ - **GodTV**
  - **GoldenMoustache**
  - **Golem**
  - **GoogleDrive**
@@ -267,6 +276,7 @@
  - **Helsinki**: helsinki.fi
  - **HentaiStigma**
  - **HistoricFilms**
+ - **history:topic**: History.com Topic
  - **hitbox**
  - **hitbox:live**
  - **HornBunny**
@@ -274,6 +284,8 @@
  - **HotStar**
  - **Howcast**
  - **HowStuffWorks**
+ - **HRTi**
+ - **HRTiPlaylist**
  - **HuffPost**: Huffington Post
  - **Hypem**
  - **Iconosquare**
@@ -301,6 +313,7 @@
  - **jpopsuki.tv**
  - **JWPlatform**
  - **Kaltura**
+ - **Kamcord**
  - **KanalPlay**: Kanal 5/9/11 Play
  - **Kankan**
  - **Karaoketv**
@@ -320,7 +333,7 @@
  - **kuwo:mv**: 酷我音乐 - MV
  - **kuwo:singer**: 酷我音乐 - 歌手
  - **kuwo:song**: 酷我音乐
- - **la7.tv**
+ - **la7.it**
  - **Laola1Tv**
  - **Le**: 乐视网
  - **Learnr**
@@ -353,6 +366,7 @@
  - **MatchTV**
  - **MDR**: MDR.DE and KiKA
  - **media.ccc.de**
+ - **META**
  - **metacafe**
  - **Metacritic**
  - **Mgoon**
@@ -379,7 +393,7 @@
  - **MovieFap**
  - **Moviezine**
  - **MPORA**
- - **MSNBC**
+ - **MSN**
  - **MTV**
  - **mtv.de**
  - **mtviggy.com**
@@ -430,8 +444,10 @@
  - **nhl.com:videocenter**
  - **nhl.com:videocenter:category**: NHL videocenter category
  - **nick.com**
+ - **nick.de**
  - **niconico**: ニコニコ動画
  - **NiconicoPlaylist**
+ - **NineCNineMedia**
  - **njoy**: N-JOY
  - **njoy:embed**
  - **Noco**
@@ -462,6 +478,8 @@
  - **Odnoklassniki**
  - **OktoberfestTV**
  - **on.aol.com**
+ - **onet.tv**
+ - **onet.tv:channel**
  - **OnionStudios**
  - **Ooyala**
  - **OoyalaExternal**
@@ -495,8 +513,9 @@
  - **plus.google**: Google Plus
  - **pluzz.francetv.fr**
  - **podomatic**
+ - **PolskieRadio**
  - **PornHd**
- - **PornHub**
+ - **PornHub**: PornHub and Thumbzilla
  - **PornHubPlaylist**
  - **PornHubUserVideos**
  - **Pornotube**
@@ -514,6 +533,7 @@
  - **qqmusic:singer**: QQ音乐 - 歌手
  - **qqmusic:toplist**: QQ音乐 - 排行榜
  - **R7**
+ - **R7Article**
  - **radio.de**
  - **radiobremen**
  - **radiocanada**
@@ -533,6 +553,8 @@
  - **revision3:embed**
  - **RICE**
  - **RingTV**
+ - **RockstarGames**
+ - **RoosterTeeth**
  - **RottenTomatoes**
  - **Roxwel**
  - **RTBF**
@@ -546,6 +568,7 @@
  - **rtve.es:infantil**: RTVE infantil
  - **rtve.es:live**: RTVE.es live streams
  - **RTVNH**
+ - **Rudo**
  - **RUHD**
  - **RulePorn**
  - **rutube**: Rutube videos
@@ -578,8 +601,10 @@
  - **Shared**: shared.sx and vivo.sx
  - **ShareSix**
  - **Sina**
+ - **SixPlay**
+ - **skynewsarabia:article**
  - **skynewsarabia:video**
- - **skynewsarabia:video**
+ - **SkySports**
  - **Slideshare**
  - **Slutload**
  - **smotri**: Smotri.com
@@ -611,6 +636,7 @@
  - **SportBoxEmbed**
  - **SportDeutschland**
  - **Sportschau**
+ - **sr:mediathek**: Saarländischer Rundfunk
  - **SRGSSR**
  - **SRGSSRPlay**: srf.ch, rts.ch, rsi.ch, rtr.ch and swissinfo.ch play sites
  - **SSA**
@@ -645,6 +671,7 @@
  - **Telegraaf**
  - **TeleMB**
  - **TeleTask**
+ - **Telewebion**
  - **TF1**
  - **TheIntercept**
  - **ThePlatform**
@@ -696,6 +723,7 @@
  - **TVPlay**: TV3Play and related services
  - **Tweakers**
  - **twitch:chapter**
+ - **twitch:clips**
  - **twitch:past_broadcasts**
  - **twitch:profile**
  - **twitch:stream**
@@ -709,6 +737,7 @@
  - **UDNEmbed**: 聯合影音
  - **Unistra**
  - **Urort**: NRK P3 Urørt
+ - **URPlay**
  - **USAToday**
  - **ustream**
  - **ustream:channel**
@@ -726,6 +755,7 @@
  - **vh1.com**
  - **Vice**
  - **ViceShow**
+ - **Vidbit**
  - **Viddler**
  - **video.google:search**: Google Video search
  - **video.mit.edu**
@@ -738,6 +768,7 @@
  - **VideoPremium**
  - **VideoTt**: video.tt - Your True Tube (Currently broken)
  - **videoweed**: VideoWeed
+ - **Vidio**
  - **vidme**
  - **vidme:user**
  - **vidme:user:likes**
@@ -764,6 +795,7 @@
  - **vine:user**
  - **vk**: VK
  - **vk:uservideos**: VK - User's Videos
+ - **vk:wallpost**
  - **vlive**
  - **Vodlocker**
  - **VoiceRepublic**
@@ -773,7 +805,6 @@
  - **VRT**
  - **vube**: Vube.com
  - **VuClip**
- - **vulture.com**
  - **Walla**
  - **washingtonpost**
  - **washingtonpost:article**
@@ -781,10 +812,8 @@
  - **WatchIndianPorn**: Watch Indian Porn
  - **WDR**
  - **wdr:mobile**
- - **WDRMaus**: Sendung mit der Maus
  - **WebOfStories**
  - **WebOfStoriesPlaylist**
- - **Weibo**
  - **WeiqiTV**: WQTV
  - **wholecloud**: WholeCloud
  - **Wimp**
@@ -792,10 +821,11 @@
  - **WNL**
  - **WorldStarHipHop**
  - **wrzuta.pl**
+ - **wrzuta.pl:playlist**
  - **WSJ**: Wall Street Journal
  - **XBef**
  - **XboxClips**
- - **XFileShare**: XFileShare based sites: DaClips, FileHoot, GorillaVid, MovPod, PowerWatch, Rapidvideo.ws, TheVideoBee, Vidto, Streamin.To
+ - **XFileShare**: XFileShare based sites: DaClips, FileHoot, GorillaVid, MovPod, PowerWatch, Rapidvideo.ws, TheVideoBee, Vidto, Streamin.To, XVIDSTAGE
  - **XHamster**
  - **XHamsterEmbed**
  - **xiami:album**: 虾米音乐 - 专辑
@@ -820,6 +850,7 @@
  - **Ynet**
  - **YouJizz**
  - **youku**: 优酷
+ - **youku:show**
  - **YouPorn**
  - **YourUpload**
  - **youtube**: YouTube.com
@@ -833,6 +864,7 @@
  - **youtube:search**: YouTube.com searches
  - **youtube:search:date**: YouTube.com searches, newest videos first
  - **youtube:search_url**: YouTube.com search URLs
+ - **youtube:shared**
  - **youtube:show**: YouTube.com (multi-season) shows
  - **youtube:subscriptions**: YouTube.com subscriptions feed, "ytsubs" keyword (requires authentication)
  - **youtube:user**: YouTube.com user videos (URL or "ytuser" keyword)

setup.py

@@ -21,25 +21,37 @@ try:
     import py2exe
 except ImportError:
     if len(sys.argv) >= 2 and sys.argv[1] == 'py2exe':
-        print("Cannot import py2exe", file=sys.stderr)
+        print('Cannot import py2exe', file=sys.stderr)
         exit(1)
 
 py2exe_options = {
-    "bundle_files": 1,
+    'bundle_files': 1,
-    "compressed": 1,
+    'compressed': 1,
-    "optimize": 2,
+    'optimize': 2,
-    "dist_dir": '.',
+    'dist_dir': '.',
-    "dll_excludes": ['w9xpopen.exe', 'crypt32.dll'],
+    'dll_excludes': ['w9xpopen.exe', 'crypt32.dll'],
 }
 
+# Get the version from youtube_dl/version.py without importing the package
+exec(compile(open('youtube_dl/version.py').read(),
+             'youtube_dl/version.py', 'exec'))
+
+DESCRIPTION = 'YouTube video downloader'
+LONG_DESCRIPTION = 'Command-line program to download videos from YouTube.com and other video sites'
+
 py2exe_console = [{
-    "script": "./youtube_dl/__main__.py",
+    'script': './youtube_dl/__main__.py',
-    "dest_base": "youtube-dl",
+    'dest_base': 'youtube-dl',
+    'version': __version__,
+    'description': DESCRIPTION,
+    'comments': LONG_DESCRIPTION,
+    'product_name': 'youtube-dl',
+    'product_version': __version__,
 }]
 
 py2exe_params = {
     'console': py2exe_console,
-    'options': {"py2exe": py2exe_options},
+    'options': {'py2exe': py2exe_options},
     'zipfile': None
 }
 
@@ -72,7 +84,7 @@ else:
         params['scripts'] = ['bin/youtube-dl']
 
 class build_lazy_extractors(Command):
-    description = "Build the extractor lazy loading module"
+    description = 'Build the extractor lazy loading module'
     user_options = []
 
     def initialize_options(self):
@@ -87,16 +99,11 @@ class build_lazy_extractors(Command):
             dry_run=self.dry_run,
         )
 
-# Get the version from youtube_dl/version.py without importing the package
-exec(compile(open('youtube_dl/version.py').read(),
-             'youtube_dl/version.py', 'exec'))
-
 setup(
     name='youtube_dl',
     version=__version__,
-    description='YouTube video downloader',
+    description=DESCRIPTION,
-    long_description='Small command-line program to download videos from'
+    long_description=LONG_DESCRIPTION,
-    ' YouTube.com and other video sites.',
     url='https://github.com/rg3/youtube-dl',
     author='Ricardo Garcia',
     author_email='ytdl@yt-dl.org',
@@ -112,16 +119,17 @@ setup(
     # test_requires = ['nosetest'],
 
     classifiers=[
-        "Topic :: Multimedia :: Video",
+        'Topic :: Multimedia :: Video',
-        "Development Status :: 5 - Production/Stable",
+        'Development Status :: 5 - Production/Stable',
-        "Environment :: Console",
+        'Environment :: Console',
-        "License :: Public Domain",
+        'License :: Public Domain',
-        "Programming Language :: Python :: 2.6",
+        'Programming Language :: Python :: 2.6',
-        "Programming Language :: Python :: 2.7",
+        'Programming Language :: Python :: 2.7',
-        "Programming Language :: Python :: 3",
+        'Programming Language :: Python :: 3',
-        "Programming Language :: Python :: 3.2",
+        'Programming Language :: Python :: 3.2',
-        "Programming Language :: Python :: 3.3",
+        'Programming Language :: Python :: 3.3',
-        "Programming Language :: Python :: 3.4",
+        'Programming Language :: Python :: 3.4',
+        'Programming Language :: Python :: 3.5',
     ],
 
     cmdclass={'build_lazy_extractors': build_lazy_extractors},

test/test_InfoExtractor.py

@@ -11,7 +11,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 from test.helper import FakeYDL
 from youtube_dl.extractor.common import InfoExtractor
 from youtube_dl.extractor import YoutubeIE, get_info_extractor
-from youtube_dl.utils import encode_data_uri, strip_jsonp, ExtractorError
+from youtube_dl.utils import encode_data_uri, strip_jsonp, ExtractorError, RegexNotFoundError
 
 
 class TestIE(InfoExtractor):
@@ -66,6 +66,11 @@ class TestInfoExtractor(unittest.TestCase):
         self.assertEqual(ie._html_search_meta('d', html), '4')
         self.assertEqual(ie._html_search_meta('e', html), '5')
         self.assertEqual(ie._html_search_meta('f', html), '6')
+        self.assertEqual(ie._html_search_meta(('a', 'b', 'c'), html), '1')
+        self.assertEqual(ie._html_search_meta(('c', 'b', 'a'), html), '3')
+        self.assertEqual(ie._html_search_meta(('z', 'x', 'c'), html), '3')
+        self.assertRaises(RegexNotFoundError, ie._html_search_meta, 'z', html, None, fatal=True)
+        self.assertRaises(RegexNotFoundError, ie._html_search_meta, ('z', 'x'), html, None, fatal=True)
 
     def test_download_json(self):
         uri = encode_data_uri(b'{"foo": "blah"}', 'application/json')

test/test_all_urls.py

@@ -6,6 +6,7 @@ from __future__ import unicode_literals
 import os
 import sys
 import unittest
+import collections
 sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 
@@ -130,6 +131,15 @@ class TestAllURLsMatching(unittest.TestCase):
             'https://screen.yahoo.com/smartwatches-latest-wearable-gadgets-163745379-cbs.html',
             ['Yahoo'])
 
+    def test_no_duplicated_ie_names(self):
+        name_accu = collections.defaultdict(list)
+        for ie in self.ies:
+            name_accu[ie.IE_NAME.lower()].append(type(ie).__name__)
+        for (ie_name, ie_list) in name_accu.items():
+            self.assertEqual(
+                len(ie_list), 1,
+                'Multiple extractors with the same IE_NAME "%s" (%s)' % (ie_name, ', '.join(ie_list)))
+
 
 if __name__ == '__main__':
     unittest.main()

test/test_compat.py

@@ -87,6 +87,8 @@ class TestCompat(unittest.TestCase):
 
     def test_compat_shlex_split(self):
         self.assertEqual(compat_shlex_split('-option "one two"'), ['-option', 'one two'])
+        self.assertEqual(compat_shlex_split('-option "one\ntwo" \n -flag'), ['-option', 'one\ntwo', '-flag'])
+        self.assertEqual(compat_shlex_split('-val 中文'), ['-val', '中文'])
 
     def test_compat_etree_fromstring(self):
         xml = '''

test/test_http.py

@@ -138,27 +138,27 @@ class TestProxy(unittest.TestCase):
         self.proxy_thread.daemon = True
         self.proxy_thread.start()
 
-        self.cn_proxy = compat_http_server.HTTPServer(
+        self.geo_proxy = compat_http_server.HTTPServer(
-            ('localhost', 0), _build_proxy_handler('cn'))
+            ('localhost', 0), _build_proxy_handler('geo'))
-        self.cn_port = http_server_port(self.cn_proxy)
+        self.geo_port = http_server_port(self.geo_proxy)
-        self.cn_proxy_thread = threading.Thread(target=self.cn_proxy.serve_forever)
+        self.geo_proxy_thread = threading.Thread(target=self.geo_proxy.serve_forever)
-        self.cn_proxy_thread.daemon = True
+        self.geo_proxy_thread.daemon = True
-        self.cn_proxy_thread.start()
+        self.geo_proxy_thread.start()
 
     def test_proxy(self):
-        cn_proxy = 'localhost:{0}'.format(self.cn_port)
+        geo_proxy = 'localhost:{0}'.format(self.geo_port)
         ydl = YoutubeDL({
             'proxy': 'localhost:{0}'.format(self.port),
-            'cn_verification_proxy': cn_proxy,
+            'geo_verification_proxy': geo_proxy,
         })
         url = 'http://foo.com/bar'
         response = ydl.urlopen(url).read().decode('utf-8')
         self.assertEqual(response, 'normal: {0}'.format(url))
 
         req = compat_urllib_request.Request(url)
-        req.add_header('Ytdl-request-proxy', cn_proxy)
+        req.add_header('Ytdl-request-proxy', geo_proxy)
         response = ydl.urlopen(req).read().decode('utf-8')
-        self.assertEqual(response, 'cn: {0}'.format(url))
+        self.assertEqual(response, 'geo: {0}'.format(url))
 
     def test_proxy_with_idn(self):
         ydl = YoutubeDL({

test/test_utils.py

@@ -33,6 +33,7 @@ from youtube_dl.utils import (
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
+    get_element_by_class,
     InAdvancePagedList,
     intlist_to_bytes,
     is_html,
@@ -60,11 +61,13 @@ from youtube_dl.utils import (
     timeconvert,
     unescapeHTML,
     unified_strdate,
+    unified_timestamp,
     unsmuggle_url,
     uppercase_escape,
     lowercase_escape,
     url_basename,
     urlencode_postdata,
+    urshift,
     update_url_query,
     version_tuple,
     xpath_with_ns,
@@ -78,6 +81,7 @@ from youtube_dl.utils import (
     cli_option,
     cli_valueless_option,
     cli_bool_option,
+    parse_codecs,
 )
 from youtube_dl.compat import (
     compat_chr,
@@ -249,6 +253,8 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(unescapeHTML('/'), '/')
         self.assertEqual(unescapeHTML('é'), 'é')
         self.assertEqual(unescapeHTML('�'), '�')
+        # HTML5 entities
+        self.assertEqual(unescapeHTML('.''), '.\'')
 
     def test_date_from_str(self):
         self.assertEqual(date_from_str('yesterday'), date_from_str('now-1day'))
@@ -281,8 +287,28 @@ class TestUtil(unittest.TestCase):
             '20150202')
         self.assertEqual(unified_strdate('Feb 14th 2016 5:45PM'), '20160214')
         self.assertEqual(unified_strdate('25-09-2014'), '20140925')
+        self.assertEqual(unified_strdate('27.02.2016 17:30'), '20160227')
         self.assertEqual(unified_strdate('UNKNOWN DATE FORMAT'), None)
 
+    def test_unified_timestamps(self):
+        self.assertEqual(unified_timestamp('December 21, 2010'), 1292889600)
+        self.assertEqual(unified_timestamp('8/7/2009'), 1247011200)
+        self.assertEqual(unified_timestamp('Dec 14, 2012'), 1355443200)
+        self.assertEqual(unified_timestamp('2012/10/11 01:56:38 +0000'), 1349920598)
+        self.assertEqual(unified_timestamp('1968 12 10'), -33436800)
+        self.assertEqual(unified_timestamp('1968-12-10'), -33436800)
+        self.assertEqual(unified_timestamp('28/01/2014 21:00:00 +0100'), 1390939200)
+        self.assertEqual(
+            unified_timestamp('11/26/2014 11:30:00 AM PST', day_first=False),
+            1417001400)
+        self.assertEqual(
+            unified_timestamp('2/2/2015 6:47:40 PM', day_first=False),
+            1422902860)
+        self.assertEqual(unified_timestamp('Feb 14th 2016 5:45PM'), 1455471900)
+        self.assertEqual(unified_timestamp('25-09-2014'), 1411603200)
+        self.assertEqual(unified_timestamp('27.02.2016 17:30'), 1456594200)
+        self.assertEqual(unified_timestamp('UNKNOWN DATE FORMAT'), None)
+
     def test_determine_ext(self):
         self.assertEqual(determine_ext('http://example.com/foo/bar.mp4/?download'), 'mp4')
         self.assertEqual(determine_ext('http://example.com/foo/bar/?download', None), None)
@@ -381,6 +407,12 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(res_url, url)
         self.assertEqual(res_data, None)
 
+        smug_url = smuggle_url(url, {'a': 'b'})
+        smug_smug_url = smuggle_url(smug_url, {'c': 'd'})
+        res_url, res_data = unsmuggle_url(smug_smug_url)
+        self.assertEqual(res_url, url)
+        self.assertEqual(res_data, {'a': 'b', 'c': 'd'})
+
     def test_shell_quote(self):
         args = ['ffmpeg', '-i', encodeFilename('ñ€ß\'.mp4')]
         self.assertEqual(shell_quote(args), """ffmpeg -i 'ñ€ß'"'"'.mp4'""")
@@ -577,6 +609,29 @@ class TestUtil(unittest.TestCase):
             limit_length('foo bar baz asd', 12).startswith('foo bar'))
         self.assertTrue('...' in limit_length('foo bar baz asd', 12))
 
+    def test_parse_codecs(self):
+        self.assertEqual(parse_codecs(''), {})
+        self.assertEqual(parse_codecs('avc1.77.30, mp4a.40.2'), {
+            'vcodec': 'avc1.77.30',
+            'acodec': 'mp4a.40.2',
+        })
+        self.assertEqual(parse_codecs('mp4a.40.2'), {
+            'vcodec': 'none',
+            'acodec': 'mp4a.40.2',
+        })
+        self.assertEqual(parse_codecs('mp4a.40.5,avc1.42001e'), {
+            'vcodec': 'avc1.42001e',
+            'acodec': 'mp4a.40.5',
+        })
+        self.assertEqual(parse_codecs('avc3.640028'), {
+            'vcodec': 'avc3.640028',
+            'acodec': 'none',
+        })
+        self.assertEqual(parse_codecs(', h264,,newcodec,aac'), {
+            'vcodec': 'h264',
+            'acodec': 'aac',
+        })
+
     def test_escape_rfc3986(self):
         reserved = "!*'();:@&=+$,/?#[]"
         unreserved = 'ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789-_.~'
@@ -638,6 +693,9 @@ class TestUtil(unittest.TestCase):
             "1":{"src":"skipped", "type": "application/vnd.apple.mpegURL"}
         }''')
 
+        inp = '''{"foo":101}'''
+        self.assertEqual(js_to_json(inp), '''{"foo":101}''')
+
     def test_js_to_json_edgecases(self):
         on = js_to_json("{abc_def:'1\\'\\\\2\\\\\\'3\"4'}")
         self.assertEqual(json.loads(on), {"abc_def": "1'\\2\\'3\"4"})
@@ -954,5 +1012,17 @@ The first line
         self.assertRaises(ValueError, encode_base_n, 0, 70)
         self.assertRaises(ValueError, encode_base_n, 0, 60, custom_table)
 
+    def test_urshift(self):
+        self.assertEqual(urshift(3, 1), 1)
+        self.assertEqual(urshift(-3, 1), 2147483646)
+
+    def test_get_element_by_class(self):
+        html = '''
+            <span class="foo bar">nice</span>
+        '''
+
+        self.assertEqual(get_element_by_class('foo', html), 'nice')
+        self.assertEqual(get_element_by_class('no-such-class', html), None)
+
 if __name__ == '__main__':
     unittest.main()

youtube_dl/YoutubeDL.py

@@ -196,8 +196,8 @@ class YoutubeDL(object):
     prefer_insecure:   Use HTTP instead of HTTPS to retrieve information.
                        At the moment, this is only supported by YouTube.
     proxy:             URL of the proxy server to use
-    cn_verification_proxy:  URL of the proxy to use for IP address verification
+    geo_verification_proxy:  URL of the proxy to use for IP address verification
-                       on Chinese sites. (Experimental)
+                       on geo-restricted sites. (Experimental)
     socket_timeout:    Time to wait for unresponsive hosts, in seconds
     bidi_workaround:   Work around buggy terminals without bidirectional text
                        support, using fridibi
@@ -304,6 +304,11 @@ class YoutubeDL(object):
         self.params.update(params)
         self.cache = Cache(self)
 
+        if self.params.get('cn_verification_proxy') is not None:
+            self.report_warning('--cn-verification-proxy is deprecated. Use --geo-verification-proxy instead.')
+            if self.params.get('geo_verification_proxy') is None:
+                self.params['geo_verification_proxy'] = self.params['cn_verification_proxy']
+
         if params.get('bidi_workaround', False):
             try:
                 import pty
@@ -1223,6 +1228,10 @@ class YoutubeDL(object):
         if 'title' not in info_dict:
             raise ExtractorError('Missing "title" field in extractor result')
 
+        if not isinstance(info_dict['id'], compat_str):
+            self.report_warning('"id" field is not a string - forcing string conversion')
+            info_dict['id'] = compat_str(info_dict['id'])
+
         if 'playlist' not in info_dict:
             # It isn't part of a playlist
             info_dict['playlist'] = None

youtube_dl/__init__.py

@@ -18,7 +18,6 @@ from .options import (
 from .compat import (
     compat_expanduser,
     compat_getpass,
-    compat_print,
     compat_shlex_split,
     workaround_optparse_bug9161,
 )
@@ -76,7 +75,7 @@ def _real_main(argv=None):
 
     # Dump user agent
     if opts.dump_user_agent:
-        compat_print(std_headers['User-Agent'])
+        write_string(std_headers['User-Agent'] + '\n', out=sys.stdout)
         sys.exit(0)
 
     # Batch file verification
@@ -101,10 +100,10 @@ def _real_main(argv=None):
 
     if opts.list_extractors:
         for ie in list_extractors(opts.age_limit):
-            compat_print(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else ''))
+            write_string(ie.IE_NAME + (' (CURRENTLY BROKEN)' if not ie._WORKING else '') + '\n', out=sys.stdout)
             matchedUrls = [url for url in all_urls if ie.suitable(url)]
             for mu in matchedUrls:
-                compat_print('  ' + mu)
+                write_string('  ' + mu + '\n', out=sys.stdout)
         sys.exit(0)
     if opts.list_extractor_descriptions:
         for ie in list_extractors(opts.age_limit):
@@ -117,7 +116,7 @@ def _real_main(argv=None):
                 _SEARCHES = ('cute kittens', 'slithering pythons', 'falling cat', 'angry poodle', 'purple fish', 'running tortoise', 'sleeping bunny', 'burping cow')
                 _COUNTS = ('', '5', '10', 'all')
                 desc += ' (Example: "%s%s:%s" )' % (ie.SEARCH_KEY, random.choice(_COUNTS), random.choice(_SEARCHES))
-            compat_print(desc)
+            write_string(desc + '\n', out=sys.stdout)
         sys.exit(0)
 
     # Conflicting, missing and erroneous options
@@ -383,6 +382,8 @@ def _real_main(argv=None):
         'external_downloader_args': external_downloader_args,
         'postprocessor_args': postprocessor_args,
         'cn_verification_proxy': opts.cn_verification_proxy,
+        'geo_verification_proxy': opts.geo_verification_proxy,
+
     }
 
     with YoutubeDL(ydl_opts) as ydl:

youtube_dl/downloader/external.py

@@ -85,7 +85,7 @@ class ExternalFD(FileDownloader):
             cmd, stderr=subprocess.PIPE)
         _, stderr = p.communicate()
         if p.returncode != 0:
-            self.to_stderr(stderr)
+            self.to_stderr(stderr.decode('utf-8', 'replace'))
         return p.returncode
 
 
@@ -210,6 +210,7 @@ class FFmpegFD(ExternalFD):
             # args += ['-http_proxy', proxy]
             env = os.environ.copy()
             compat_setenv('HTTP_PROXY', proxy, env=env)
+            compat_setenv('http_proxy', proxy, env=env)
 
         protocol = info_dict.get('protocol')
 

youtube_dl/downloader/f4m.py

@@ -196,6 +196,11 @@ def build_fragments_list(boot_info):
     first_frag_number = fragment_run_entry_table[0]['first']
     fragments_counter = itertools.count(first_frag_number)
     for segment, fragments_count in segment_run_table['segment_run']:
+        # In some live HDS streams (for example Rai), `fragments_count` is
+        # abnormal and causing out-of-memory errors. It's OK to change the
+        # number of fragments for live streams as they are updated periodically
+        if fragments_count == 4294967295 and boot_info['live']:
+            fragments_count = 2
         for _ in range(fragments_count):
             res.append((segment, next(fragments_counter)))
 
@@ -329,7 +334,11 @@ class F4mFD(FragmentFD):
 
         base_url = compat_urlparse.urljoin(man_url, media.attrib['url'])
         bootstrap_node = doc.find(_add_ns('bootstrapInfo'))
-        boot_info, bootstrap_url = self._parse_bootstrap_node(bootstrap_node, base_url)
+        # From Adobe F4M 3.0 spec:
+        # The <baseURL> element SHALL be the base URL for all relative
+        # (HTTP-based) URLs in the manifest. If <baseURL> is not present, said
+        # URLs should be relative to the location of the containing document.
+        boot_info, bootstrap_url = self._parse_bootstrap_node(bootstrap_node, man_url)
         live = boot_info['live']
         metadata_node = media.find(_add_ns('metadata'))
         if metadata_node is not None:

youtube_dl/downloader/hls.py

@@ -2,14 +2,24 @@ from __future__ import unicode_literals
 
 import os.path
 import re
+import binascii
+try:
+    from Crypto.Cipher import AES
+    can_decrypt_frag = True
+except ImportError:
+    can_decrypt_frag = False
 
 from .fragment import FragmentFD
 from .external import FFmpegFD
 
-from ..compat import compat_urlparse
+from ..compat import (
+    compat_urlparse,
+    compat_struct_pack,
+)
 from ..utils import (
     encodeFilename,
     sanitize_open,
+    parse_m3u8_attributes,
 )
 
 
@@ -21,19 +31,27 @@ class HlsFD(FragmentFD):
     @staticmethod
     def can_download(manifest):
         UNSUPPORTED_FEATURES = (
-            r'#EXT-X-KEY:METHOD=(?!NONE)',  # encrypted streams [1]
+            r'#EXT-X-KEY:METHOD=(?!NONE|AES-128)',  # encrypted streams [1]
             r'#EXT-X-BYTERANGE',  # playlists composed of byte ranges of media files [2]
+
             # Live streams heuristic does not always work (e.g. geo restricted to Germany
             # http://hls-geo.daserste.de/i/videoportal/Film/c_620000/622873/format,716451,716457,716450,716458,716459,.mp4.csmil/index_4_av.m3u8?null=0)
             # r'#EXT-X-MEDIA-SEQUENCE:(?!0$)',  # live streams [3]
-            r'#EXT-X-PLAYLIST-TYPE:EVENT',  # media segments may be appended to the end of
+
-                                            # event media playlists [4]
+            # This heuristic also is not correct since segments may not be appended as well.
+            # Twitch vods of finished streams have EXT-X-PLAYLIST-TYPE:EVENT despite
+            # no segments will definitely be appended to the end of the playlist.
+            # r'#EXT-X-PLAYLIST-TYPE:EVENT',  # media segments may be appended to the end of
+            #                                 # event media playlists [4]
+
             # 1. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.4
             # 2. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.2.2
             # 3. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.2
             # 4. https://tools.ietf.org/html/draft-pantos-http-live-streaming-17#section-4.3.3.5
         )
-        return all(not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES)
+        check_results = [not re.search(feature, manifest) for feature in UNSUPPORTED_FEATURES]
+        check_results.append(can_decrypt_frag or '#EXT-X-KEY:METHOD=AES-128' not in manifest)
+        return all(check_results)
 
     def real_download(self, filename, info_dict):
         man_url = info_dict['url']
@@ -51,36 +69,60 @@ class HlsFD(FragmentFD):
                 fd.add_progress_hook(ph)
             return fd.real_download(filename, info_dict)
 
-        fragment_urls = []
+        total_frags = 0
         for line in s.splitlines():
             line = line.strip()
             if line and not line.startswith('#'):
-                segment_url = (
+                total_frags += 1
-                    line
-                    if re.match(r'^https?://', line)
-                    else compat_urlparse.urljoin(man_url, line))
-                fragment_urls.append(segment_url)
-                # We only download the first fragment during the test
-                if self.params.get('test', False):
-                    break
 
         ctx = {
             'filename': filename,
-            'total_frags': len(fragment_urls),
+            'total_frags': total_frags,
         }
 
         self._prepare_and_start_frag_download(ctx)
 
+        i = 0
+        media_sequence = 0
+        decrypt_info = {'METHOD': 'NONE'}
         frags_filenames = []
-        for i, frag_url in enumerate(fragment_urls):
+        for line in s.splitlines():
-            frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i)
+            line = line.strip()
-            success = ctx['dl'].download(frag_filename, {'url': frag_url})
+            if line:
-            if not success:
+                if not line.startswith('#'):
-                return False
+                    frag_url = (
-            down, frag_sanitized = sanitize_open(frag_filename, 'rb')
+                        line
-            ctx['dest_stream'].write(down.read())
+                        if re.match(r'^https?://', line)
-            down.close()
+                        else compat_urlparse.urljoin(man_url, line))
-            frags_filenames.append(frag_sanitized)
+                    frag_filename = '%s-Frag%d' % (ctx['tmpfilename'], i)
+                    success = ctx['dl'].download(frag_filename, {'url': frag_url})
+                    if not success:
+                        return False
+                    down, frag_sanitized = sanitize_open(frag_filename, 'rb')
+                    frag_content = down.read()
+                    down.close()
+                    if decrypt_info['METHOD'] == 'AES-128':
+                        iv = decrypt_info.get('IV') or compat_struct_pack('>8xq', media_sequence)
+                        frag_content = AES.new(
+                            decrypt_info['KEY'], AES.MODE_CBC, iv).decrypt(frag_content)
+                    ctx['dest_stream'].write(frag_content)
+                    frags_filenames.append(frag_sanitized)
+                    # We only download the first fragment during the test
+                    if self.params.get('test', False):
+                        break
+                    i += 1
+                    media_sequence += 1
+                elif line.startswith('#EXT-X-KEY'):
+                    decrypt_info = parse_m3u8_attributes(line[11:])
+                    if decrypt_info['METHOD'] == 'AES-128':
+                        if 'IV' in decrypt_info:
+                            decrypt_info['IV'] = binascii.unhexlify(decrypt_info['IV'][2:])
+                        if not re.match(r'^https?://', decrypt_info['URI']):
+                            decrypt_info['URI'] = compat_urlparse.urljoin(
+                                man_url, decrypt_info['URI'])
+                        decrypt_info['KEY'] = self.ydl.urlopen(decrypt_info['URI']).read()
+                elif line.startswith('#EXT-X-MEDIA-SEQUENCE'):
+                    media_sequence = int(line[22:])
 
         self._finish_frag_download(ctx)
 

youtube_dl/extractor/adobetv.py

@@ -156,7 +156,10 @@ class AdobeTVVideoIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        video_data = self._download_json(url + '?format=json', video_id)
+        webpage = self._download_webpage(url, video_id)
+
+        video_data = self._parse_json(self._search_regex(
+            r'var\s+bridge\s*=\s*([^;]+);', webpage, 'bridged data'), video_id)
 
         formats = [{
             'format_id': '%s-%s' % (determine_ext(source['src']), source.get('height')),

youtube_dl/extractor/aenetworks.py

@@ -2,23 +2,137 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
+from .theplatform import ThePlatformIE
 from ..utils import (
     smuggle_url,
     update_url_query,
     unescapeHTML,
+    extract_attributes,
+    get_element_by_attribute,
+)
+from ..compat import (
+    compat_urlparse,
 )
 
 
-class AENetworksIE(InfoExtractor):
+class AENetworksBaseIE(ThePlatformIE):
+    _THEPLATFORM_KEY = 'crazyjava'
+    _THEPLATFORM_SECRET = 's3cr3t'
+
+
+class AENetworksIE(AENetworksBaseIE):
     IE_NAME = 'aenetworks'
     IE_DESC = 'A+E Networks: A&E, Lifetime, History.com, FYI Network'
-    _VALID_URL = r'https?://(?:www\.)?(?:(?:history|aetv|mylifetime)\.com|fyi\.tv)/(?P<type>[^/]+)/(?:[^/]+/)+(?P<id>[^/]+?)(?:$|[?#])'
+    _VALID_URL = r'https?://(?:www\.)?(?P<domain>(?:history|aetv|mylifetime)\.com|fyi\.tv)/(?:shows/(?P<show_path>[^/]+(?:/[^/]+){0,2})|movies/(?P<movie_display_id>[^/]+)/full-movie)'
+    _TESTS = [{
+        'url': 'http://www.history.com/shows/mountain-men/season-1/episode-1',
+        'md5': '8ff93eb073449f151d6b90c0ae1ef0c7',
+        'info_dict': {
+            'id': '22253814',
+            'ext': 'mp4',
+            'title': 'Winter Is Coming',
+            'description': 'md5:641f424b7a19d8e24f26dea22cf59d74',
+            'timestamp': 1338306241,
+            'upload_date': '20120529',
+            'uploader': 'AENE-NEW',
+        },
+        'add_ie': ['ThePlatform'],
+    }, {
+        'url': 'http://www.history.com/shows/ancient-aliens/season-1',
+        'info_dict': {
+            'id': '71889446852',
+        },
+        'playlist_mincount': 5,
+    }, {
+        'url': 'http://www.mylifetime.com/shows/atlanta-plastic',
+        'info_dict': {
+            'id': 'SERIES4317',
+            'title': 'Atlanta Plastic',
+        },
+        'playlist_mincount': 2,
+    }, {
+        'url': 'http://www.aetv.com/shows/duck-dynasty/season-9/episode-1',
+        'only_matching': True
+    }, {
+        'url': 'http://www.fyi.tv/shows/tiny-house-nation/season-1/episode-8',
+        'only_matching': True
+    }, {
+        'url': 'http://www.mylifetime.com/shows/project-runway-junior/season-1/episode-6',
+        'only_matching': True
+    }, {
+        'url': 'http://www.mylifetime.com/movies/center-stage-on-pointe/full-movie',
+        'only_matching': True
+    }]
+    _DOMAIN_TO_REQUESTOR_ID = {
+        'history.com': 'HISTORY',
+        'aetv.com': 'AETV',
+        'mylifetime.com': 'LIFETIME',
+        'fyi.tv': 'FYI',
+    }
 
+    def _real_extract(self, url):
+        domain, show_path, movie_display_id = re.match(self._VALID_URL, url).groups()
+        display_id = show_path or movie_display_id
+        webpage = self._download_webpage(url, display_id)
+        if show_path:
+            url_parts = show_path.split('/')
+            url_parts_len = len(url_parts)
+            if url_parts_len == 1:
+                entries = []
+                for season_url_path in re.findall(r'(?s)<li[^>]+data-href="(/shows/%s/season-\d+)"' % url_parts[0], webpage):
+                    entries.append(self.url_result(
+                        compat_urlparse.urljoin(url, season_url_path), 'AENetworks'))
+                return self.playlist_result(
+                    entries, self._html_search_meta('aetn:SeriesId', webpage),
+                    self._html_search_meta('aetn:SeriesTitle', webpage))
+            elif url_parts_len == 2:
+                entries = []
+                for episode_item in re.findall(r'(?s)<div[^>]+class="[^"]*episode-item[^"]*"[^>]*>', webpage):
+                    episode_attributes = extract_attributes(episode_item)
+                    episode_url = compat_urlparse.urljoin(
+                        url, episode_attributes['data-canonical'])
+                    entries.append(self.url_result(
+                        episode_url, 'AENetworks',
+                        episode_attributes['data-videoid']))
+                return self.playlist_result(
+                    entries, self._html_search_meta('aetn:SeasonId', webpage))
+
+        query = {
+            'mbr': 'true',
+            'assetTypes': 'medium_video_s3'
+        }
+        video_id = self._html_search_meta('aetn:VideoID', webpage)
+        media_url = self._search_regex(
+            r"media_url\s*=\s*'([^']+)'", webpage, 'video url')
+        theplatform_metadata = self._download_theplatform_metadata(self._search_regex(
+            r'https?://link.theplatform.com/s/([^?]+)', media_url, 'theplatform_path'), video_id)
+        info = self._parse_theplatform_metadata(theplatform_metadata)
+        if theplatform_metadata.get('AETN$isBehindWall'):
+            requestor_id = self._DOMAIN_TO_REQUESTOR_ID[domain]
+            resource = '<rss version="2.0" xmlns:media="http://search.yahoo.com/mrss/"><channel><title>%s</title><item><title>%s</title><guid>%s</guid><media:rating scheme="urn:v-chip">%s</media:rating></item></channel></rss>' % (requestor_id, theplatform_metadata['title'], theplatform_metadata['AETN$PPL_pplProgramId'], theplatform_metadata['ratings'][0]['rating'])
+            query['auth'] = self._extract_mvpd_auth(
+                url, video_id, requestor_id, resource)
+        info.update(self._search_json_ld(webpage, video_id, fatal=False))
+        media_url = update_url_query(media_url, query)
+        media_url = self._sign_url(media_url, self._THEPLATFORM_KEY, self._THEPLATFORM_SECRET)
+        formats, subtitles = self._extract_theplatform_smil(media_url, video_id)
+        self._sort_formats(formats)
+        info.update({
+            'id': video_id,
+            'formats': formats,
+            'subtitles': subtitles,
+        })
+        return info
+
+
+class HistoryTopicIE(AENetworksBaseIE):
+    IE_NAME = 'history:topic'
+    IE_DESC = 'History.com Topic'
+    _VALID_URL = r'https?://(?:www\.)?history\.com/topics/(?:[^/]+/)?(?P<topic_id>[^/]+)(?:/[^/]+(?:/(?P<video_display_id>[^/?#]+))?)?'
     _TESTS = [{
         'url': 'http://www.history.com/topics/valentines-day/history-of-valentines-day/videos/bet-you-didnt-know-valentines-day?m=528e394da93ae&s=undefined&f=1&free=false',
         'info_dict': {
-            'id': 'g12m5Gyt3fdR',
+            'id': '40700995724',
             'ext': 'mp4',
             'title': "Bet You Didn't Know: Valentine's Day",
             'description': 'md5:7b57ea4829b391995b405fa60bd7b5f7',
@@ -31,57 +145,61 @@ class AENetworksIE(InfoExtractor):
             'skip_download': True,
         },
         'add_ie': ['ThePlatform'],
-        'expected_warnings': ['JSON-LD'],
     }, {
-        'url': 'http://www.history.com/shows/mountain-men/season-1/episode-1',
+        'url': 'http://www.history.com/topics/world-war-i/world-war-i-history/videos',
-        'md5': '8ff93eb073449f151d6b90c0ae1ef0c7',
+        'info_dict':
-        'info_dict': {
+        {
-            'id': 'eg47EERs_JsZ',
+            'id': 'world-war-i-history',
-            'ext': 'mp4',
+            'title': 'World War I History',
-            'title': 'Winter Is Coming',
-            'description': 'md5:641f424b7a19d8e24f26dea22cf59d74',
-            'timestamp': 1338306241,
-            'upload_date': '20120529',
-            'uploader': 'AENE-NEW',
         },
-        'add_ie': ['ThePlatform'],
+        'playlist_mincount': 24,
     }, {
-        'url': 'http://www.aetv.com/shows/duck-dynasty/video/inlawful-entry',
+        'url': 'http://www.history.com/topics/world-war-i-history/videos',
-        'only_matching': True
+        'only_matching': True,
     }, {
-        'url': 'http://www.fyi.tv/shows/tiny-house-nation/videos/207-sq-ft-minnesota-prairie-cottage',
+        'url': 'http://www.history.com/topics/world-war-i/world-war-i-history',
-        'only_matching': True
+        'only_matching': True,
     }, {
-        'url': 'http://www.mylifetime.com/shows/project-runway-junior/video/season-1/episode-6/superstar-clients',
+        'url': 'http://www.history.com/topics/world-war-i/world-war-i-history/speeches',
-        'only_matching': True
+        'only_matching': True,
     }]
 
-    def _real_extract(self, url):
+    def theplatform_url_result(self, theplatform_url, video_id, query):
-        page_type, video_id = re.match(self._VALID_URL, url).groups()
+        return {
-
-        webpage = self._download_webpage(url, video_id)
-
-        video_url_re = [
-            r'data-href="[^"]*/%s"[^>]+data-release-url="([^"]+)"' % video_id,
-            r"media_url\s*=\s*'([^']+)'"
-        ]
-        video_url = unescapeHTML(self._search_regex(video_url_re, webpage, 'video url'))
-        query = {'mbr': 'true'}
-        if page_type == 'shows':
-            query['assetTypes'] = 'medium_video_s3'
-        if 'switch=hds' in video_url:
-            query['switch'] = 'hls'
-
-        info = self._search_json_ld(webpage, video_id, fatal=False)
-        info.update({
             '_type': 'url_transparent',
+            'id': video_id,
             'url': smuggle_url(
-                update_url_query(video_url, query),
+                update_url_query(theplatform_url, query),
                 {
                     'sig': {
-                        'key': 'crazyjava',
+                        'key': self._THEPLATFORM_KEY,
-                        'secret': 's3cr3t'},
+                        'secret': self._THEPLATFORM_SECRET,
+                    },
                     'force_smil_url': True
                 }),
-        })
+            'ie_key': 'ThePlatform',
-        return info
+        }
+
+    def _real_extract(self, url):
+        topic_id, video_display_id = re.match(self._VALID_URL, url).groups()
+        if video_display_id:
+            webpage = self._download_webpage(url, video_display_id)
+            release_url, video_id = re.search(r"_videoPlayer.play\('([^']+)'\s*,\s*'[^']+'\s*,\s*'(\d+)'\)", webpage).groups()
+            release_url = unescapeHTML(release_url)
+
+            return self.theplatform_url_result(
+                release_url, video_id, {
+                    'mbr': 'true',
+                    'switch': 'hls'
+                })
+        else:
+            webpage = self._download_webpage(url, topic_id)
+            entries = []
+            for episode_item in re.findall(r'<a.+?data-release-url="[^"]+"[^>]*>', webpage):
+                video_attributes = extract_attributes(episode_item)
+                entries.append(self.theplatform_url_result(
+                    video_attributes['data-release-url'], video_attributes['data-id'], {
+                        'mbr': 'true',
+                        'switch': 'hls'
+                    }))
+            return self.playlist_result(entries, topic_id, get_element_by_attribute('class', 'show-title', webpage))

youtube_dl/extractor/afreecatv.py

@@ -0,0 +1,133 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse_urlparse,
+    compat_urlparse,
+)
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    xpath_element,
+    xpath_text,
+)
+
+
+class AfreecaTVIE(InfoExtractor):
+    IE_DESC = 'afreecatv.com'
+    _VALID_URL = r'''(?x)^
+        https?://(?:(live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
+        (?:
+            /app/(?:index|read_ucc_bbs)\.cgi|
+            /player/[Pp]layer\.(?:swf|html))
+        \?.*?\bnTitleNo=(?P<id>\d+)'''
+    _TESTS = [{
+        'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
+        'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
+        'info_dict': {
+            'id': '36164052',
+            'ext': 'mp4',
+            'title': '데일리 에이프릴 요정들의 시상식!',
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+            'uploader': 'dailyapril',
+            'uploader_id': 'dailyapril',
+            'upload_date': '20160503',
+        }
+    }, {
+        'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
+        'info_dict': {
+            'id': '36153164',
+            'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
+            'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
+            'uploader': 'dailyapril',
+            'uploader_id': 'dailyapril',
+        },
+        'playlist_count': 2,
+        'playlist': [{
+            'md5': 'd8b7c174568da61d774ef0203159bf97',
+            'info_dict': {
+                'id': '36153164_1',
+                'ext': 'mp4',
+                'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
+                'upload_date': '20160502',
+            },
+        }, {
+            'md5': '58f2ce7f6044e34439ab2d50612ab02b',
+            'info_dict': {
+                'id': '36153164_2',
+                'ext': 'mp4',
+                'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
+                'upload_date': '20160502',
+            },
+        }],
+    }, {
+        'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
+        'only_matching': True,
+    }]
+
+    @staticmethod
+    def parse_video_key(key):
+        video_key = {}
+        m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
+        if m:
+            video_key['upload_date'] = m.group('upload_date')
+            video_key['part'] = m.group('part')
+        return video_key
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        parsed_url = compat_urllib_parse_urlparse(url)
+        info_url = compat_urlparse.urlunparse(parsed_url._replace(
+            netloc='afbbs.afreecatv.com:8080',
+            path='/api/video/get_video_info.php'))
+        video_xml = self._download_xml(info_url, video_id)
+
+        if xpath_element(video_xml, './track/video/file') is None:
+            raise ExtractorError('Specified AfreecaTV video does not exist',
+                                 expected=True)
+
+        title = xpath_text(video_xml, './track/title', 'title')
+        uploader = xpath_text(video_xml, './track/nickname', 'uploader')
+        uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
+        duration = int_or_none(xpath_text(video_xml, './track/duration',
+                                          'duration'))
+        thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
+
+        entries = []
+        for i, video_file in enumerate(video_xml.findall('./track/video/file')):
+            video_key = self.parse_video_key(video_file.get('key', ''))
+            if not video_key:
+                continue
+            entries.append({
+                'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
+                'title': title,
+                'upload_date': video_key.get('upload_date'),
+                'duration': int_or_none(video_file.get('duration')),
+                'url': video_file.text,
+            })
+
+        info = {
+            'id': video_id,
+            'title': title,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'duration': duration,
+            'thumbnail': thumbnail,
+        }
+
+        if len(entries) > 1:
+            info['_type'] = 'multi_video'
+            info['entries'] = entries
+        elif len(entries) == 1:
+            info['url'] = entries[0]['url']
+            info['upload_date'] = entries[0].get('upload_date')
+        else:
+            raise ExtractorError(
+                'No files found for the specified AfreecaTV video, either'
+                ' the URL is incorrect or the video has been made private.',
+                expected=True)
+
+        return info

youtube_dl/extractor/aftonbladet.py

@@ -24,10 +24,10 @@ class AftonbladetIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         # find internal video meta data
-        meta_url = 'http://aftonbladet-play.drlib.aptoma.no/video/%s.json'
+        meta_url = 'http://aftonbladet-play-metadata.cdn.drvideo.aptoma.no/video/%s.json'
         player_config = self._parse_json(self._html_search_regex(
             r'data-player-config="([^"]+)"', webpage, 'player config'), video_id)
-        internal_meta_id = player_config['videoId']
+        internal_meta_id = player_config['aptomaVideoId']
         internal_meta_url = meta_url % internal_meta_id
         internal_meta_json = self._download_json(
             internal_meta_url, video_id, 'Downloading video meta data')

youtube_dl/extractor/amp.py

@@ -5,6 +5,8 @@ from .common import InfoExtractor
 from ..utils import (
     int_or_none,
     parse_iso8601,
+    mimetype2ext,
+    determine_ext,
 )
 
 
@@ -50,21 +52,25 @@ class AMPIE(InfoExtractor):
         if isinstance(media_content, dict):
             media_content = [media_content]
         for media_data in media_content:
-            media = media_data['@attributes']
+            media = media_data.get('@attributes', {})
-            media_type = media['type']
+            media_url = media.get('url')
-            if media_type in ('video/f4m', 'application/f4m+xml'):
+            if not media_url:
+                continue
+            ext = mimetype2ext(media.get('type')) or determine_ext(media_url)
+            if ext == 'f4m':
                 formats.extend(self._extract_f4m_formats(
-                    media['url'] + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
+                    media_url + '?hdcore=3.4.0&plugin=aasp-3.4.0.132.124',
                     video_id, f4m_id='hds', fatal=False))
-            elif media_type == 'application/x-mpegURL':
+            elif ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    media['url'], video_id, 'mp4', m3u8_id='hls', fatal=False))
+                    media_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
             else:
                 formats.append({
                     'format_id': media_data.get('media-category', {}).get('@attributes', {}).get('label'),
                     'url': media['url'],
                     'tbr': int_or_none(media.get('bitrate')),
                     'filesize': int_or_none(media.get('fileSize')),
+                    'ext': ext,
                 })
 
         self._sort_formats(formats)

youtube_dl/extractor/animeondemand.py

@@ -22,6 +22,7 @@ class AnimeOnDemandIE(InfoExtractor):
     _APPLY_HTML5_URL = 'https://www.anime-on-demand.de/html5apply'
     _NETRC_MACHINE = 'animeondemand'
     _TESTS = [{
+        # jap, OmU
         'url': 'https://www.anime-on-demand.de/anime/161',
         'info_dict': {
             'id': '161',
@@ -30,17 +31,21 @@ class AnimeOnDemandIE(InfoExtractor):
         },
         'playlist_mincount': 4,
     }, {
-        # Film wording is used instead of Episode
+        # Film wording is used instead of Episode, ger/jap, Dub/OmU
         'url': 'https://www.anime-on-demand.de/anime/39',
         'only_matching': True,
     }, {
-        # Episodes without titles
+        # Episodes without titles, jap, OmU
         'url': 'https://www.anime-on-demand.de/anime/162',
         'only_matching': True,
     }, {
         # ger/jap, Dub/OmU, account required
         'url': 'https://www.anime-on-demand.de/anime/169',
         'only_matching': True,
+    }, {
+        # Full length film, non-series, ger/jap, Dub/OmU, account required
+        'url': 'https://www.anime-on-demand.de/anime/185',
+        'only_matching': True,
     }]
 
     def _login(self):
@@ -110,35 +115,12 @@ class AnimeOnDemandIE(InfoExtractor):
 
         entries = []
 
-        for num, episode_html in enumerate(re.findall(
+        def extract_info(html, video_id, num=None):
-                r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', webpage), 1):
+            title, description = [None] * 2
-            episodebox_title = self._search_regex(
-                (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
-                 r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
-                episode_html, 'episodebox title', default=None, group='title')
-            if not episodebox_title:
-                continue
-
-            episode_number = int(self._search_regex(
-                r'(?:Episode|Film)\s*(\d+)',
-                episodebox_title, 'episode number', default=num))
-            episode_title = self._search_regex(
-                r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
-                episodebox_title, 'episode title', default=None)
-
-            video_id = 'episode-%d' % episode_number
-
-            common_info = {
-                'id': video_id,
-                'series': anime_title,
-                'episode': episode_title,
-                'episode_number': episode_number,
-            }
-
             formats = []
 
             for input_ in re.findall(
-                    r'<input[^>]+class=["\'].*?streamstarter_html5[^>]+>', episode_html):
+                    r'<input[^>]+class=["\'].*?streamstarter_html5[^>]+>', html):
                 attributes = extract_attributes(input_)
                 playlist_urls = []
                 for playlist_key in ('data-playlist', 'data-otherplaylist'):
@@ -161,7 +143,7 @@ class AnimeOnDemandIE(InfoExtractor):
                         format_id_list.append(lang)
                     if kind:
                         format_id_list.append(kind)
-                    if not format_id_list:
+                    if not format_id_list and num is not None:
                         format_id_list.append(compat_str(num))
                     format_id = '-'.join(format_id_list)
                     format_note = ', '.join(filter(None, (kind, lang_note)))
@@ -215,28 +197,74 @@ class AnimeOnDemandIE(InfoExtractor):
                             })
                         formats.extend(file_formats)
 
-            if formats:
+            return {
-                self._sort_formats(formats)
+                'title': title,
+                'description': description,
+                'formats': formats,
+            }
+
+        def extract_entries(html, video_id, common_info, num=None):
+            info = extract_info(html, video_id, num)
+
+            if info['formats']:
+                self._sort_formats(info['formats'])
                 f = common_info.copy()
-                f.update({
+                f.update(info)
-                    'title': title,
-                    'description': description,
-                    'formats': formats,
-                })
                 entries.append(f)
 
-            # Extract teaser only when full episode is not available
+            # Extract teaser/trailer only when full episode is not available
-            if not formats:
+            if not info['formats']:
                 m = re.search(
-                    r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>Teaser<',
+                    r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>(?P<kind>Teaser|Trailer)<',
-                    episode_html)
+                    html)
                 if m:
                     f = common_info.copy()
                     f.update({
-                        'id': '%s-teaser' % f['id'],
+                        'id': '%s-%s' % (f['id'], m.group('kind').lower()),
                         'title': m.group('title'),
                         'url': compat_urlparse.urljoin(url, m.group('href')),
                     })
                     entries.append(f)
 
+        def extract_episodes(html):
+            for num, episode_html in enumerate(re.findall(
+                    r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', html), 1):
+                episodebox_title = self._search_regex(
+                    (r'class="episodebox-title"[^>]+title=(["\'])(?P<title>.+?)\1',
+                     r'class="episodebox-title"[^>]+>(?P<title>.+?)<'),
+                    episode_html, 'episodebox title', default=None, group='title')
+                if not episodebox_title:
+                    continue
+
+                episode_number = int(self._search_regex(
+                    r'(?:Episode|Film)\s*(\d+)',
+                    episodebox_title, 'episode number', default=num))
+                episode_title = self._search_regex(
+                    r'(?:Episode|Film)\s*\d+\s*-\s*(.+)',
+                    episodebox_title, 'episode title', default=None)
+
+                video_id = 'episode-%d' % episode_number
+
+                common_info = {
+                    'id': video_id,
+                    'series': anime_title,
+                    'episode': episode_title,
+                    'episode_number': episode_number,
+                }
+
+                extract_entries(episode_html, video_id, common_info)
+
+        def extract_film(html, video_id):
+            common_info = {
+                'id': anime_id,
+                'title': anime_title,
+                'description': anime_description,
+            }
+            extract_entries(html, video_id, common_info)
+
+        extract_episodes(webpage)
+
+        if not entries:
+            extract_film(webpage, anime_id)
+
         return self.playlist_result(entries, anime_id, anime_title, anime_description)

youtube_dl/extractor/appletrailers.py

@@ -7,6 +7,8 @@ from .common import InfoExtractor
 from ..compat import compat_urlparse
 from ..utils import (
     int_or_none,
+    parse_duration,
+    unified_strdate,
 )
 
 
@@ -16,7 +18,8 @@ class AppleTrailersIE(InfoExtractor):
     _TESTS = [{
         'url': 'http://trailers.apple.com/trailers/wb/manofsteel/',
         'info_dict': {
-            'id': 'manofsteel',
+            'id': '5111',
+            'title': 'Man of Steel',
         },
         'playlist': [
             {
@@ -70,6 +73,15 @@ class AppleTrailersIE(InfoExtractor):
             'id': 'blackthorn',
         },
         'playlist_mincount': 2,
+        'expected_warnings': ['Unable to download JSON metadata'],
+    }, {
+        # json data only available from http://trailers.apple.com/trailers/feeds/data/15881.json
+        'url': 'http://trailers.apple.com/trailers/fox/kungfupanda3/',
+        'info_dict': {
+            'id': '15881',
+            'title': 'Kung Fu Panda 3',
+        },
+        'playlist_mincount': 4,
     }, {
         'url': 'http://trailers.apple.com/ca/metropole/autrui/',
         'only_matching': True,
@@ -85,6 +97,45 @@ class AppleTrailersIE(InfoExtractor):
         movie = mobj.group('movie')
         uploader_id = mobj.group('company')
 
+        webpage = self._download_webpage(url, movie)
+        film_id = self._search_regex(r"FilmId\s*=\s*'(\d+)'", webpage, 'film id')
+        film_data = self._download_json(
+            'http://trailers.apple.com/trailers/feeds/data/%s.json' % film_id,
+            film_id, fatal=False)
+
+        if film_data:
+            entries = []
+            for clip in film_data.get('clips', []):
+                clip_title = clip['title']
+
+                formats = []
+                for version, version_data in clip.get('versions', {}).items():
+                    for size, size_data in version_data.get('sizes', {}).items():
+                        src = size_data.get('src')
+                        if not src:
+                            continue
+                        formats.append({
+                            'format_id': '%s-%s' % (version, size),
+                            'url': re.sub(r'_(\d+p.mov)', r'_h\1', src),
+                            'width': int_or_none(size_data.get('width')),
+                            'height': int_or_none(size_data.get('height')),
+                            'language': version[:2],
+                        })
+                self._sort_formats(formats)
+
+                entries.append({
+                    'id': movie + '-' + re.sub(r'[^a-zA-Z0-9]', '', clip_title).lower(),
+                    'formats': formats,
+                    'title': clip_title,
+                    'thumbnail': clip.get('screen') or clip.get('thumb'),
+                    'duration': parse_duration(clip.get('runtime') or clip.get('faded')),
+                    'upload_date': unified_strdate(clip.get('posted')),
+                    'uploader_id': uploader_id,
+                })
+
+            page_data = film_data.get('page', {})
+            return self.playlist_result(entries, film_id, page_data.get('movie_title'))
+
         playlist_url = compat_urlparse.urljoin(url, 'includes/playlists/itunes.inc')
 
         def fix_html(s):

youtube_dl/extractor/ard.py

@@ -8,12 +8,12 @@ from .generic import GenericIE
 from ..utils import (
     determine_ext,
     ExtractorError,
-    get_element_by_attribute,
     qualities,
     int_or_none,
     parse_duration,
     unified_strdate,
     xpath_text,
+    update_url_query,
 )
 from ..compat import compat_etree_fromstring
 
@@ -35,6 +35,7 @@ class ARDMediathekIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         },
+        'skip': 'HTTP Error 404: Not Found',
     }, {
         'url': 'http://www.ardmediathek.de/tv/Tatort/Tatort-Scheinwelten-H%C3%B6rfassung-Video/Das-Erste/Video?documentId=29522730&bcastId=602916',
         'md5': 'f4d98b10759ac06c0072bbcd1f0b9e3e',
@@ -45,6 +46,7 @@ class ARDMediathekIE(InfoExtractor):
             'description': 'md5:196392e79876d0ac94c94e8cdb2875f1',
             'duration': 5252,
         },
+        'skip': 'HTTP Error 404: Not Found',
     }, {
         # audio
         'url': 'http://www.ardmediathek.de/tv/WDR-H%C3%B6rspiel-Speicher/Tod-eines-Fu%C3%9Fballers/WDR-3/Audio-Podcast?documentId=28488308&bcastId=23074086',
@@ -56,6 +58,7 @@ class ARDMediathekIE(InfoExtractor):
             'description': 'md5:f6e39f3461f0e1f54bfa48c8875c86ef',
             'duration': 3240,
         },
+        'skip': 'HTTP Error 404: Not Found',
     }, {
         'url': 'http://mediathek.daserste.de/sendungen_a-z/328454_anne-will/22429276_vertrauen-ist-gut-spionieren-ist-besser-geht',
         'only_matching': True,
@@ -114,11 +117,14 @@ class ARDMediathekIE(InfoExtractor):
                         continue
                     if ext == 'f4m':
                         formats.extend(self._extract_f4m_formats(
-                            stream_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124',
+                            update_url_query(stream_url, {
-                            video_id, preference=-1, f4m_id='hds', fatal=False))
+                                'hdcore': '3.1.1',
+                                'plugin': 'aasp-3.1.1.69.124'
+                            }),
+                            video_id, f4m_id='hds', fatal=False))
                     elif ext == 'm3u8':
                         formats.extend(self._extract_m3u8_formats(
-                            stream_url, video_id, 'mp4', preference=1, m3u8_id='hls', fatal=False))
+                            stream_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
                     else:
                         if server and server.startswith('rtmp'):
                             f = {
@@ -232,7 +238,8 @@ class ARDIE(InfoExtractor):
             'title': 'Die Story im Ersten: Mission unter falscher Flagge',
             'upload_date': '20140804',
             'thumbnail': 're:^https?://.*\.jpg$',
-        }
+        },
+        'skip': 'HTTP Error 404: Not Found',
     }
 
     def _real_extract(self, url):
@@ -274,41 +281,3 @@ class ARDIE(InfoExtractor):
             'upload_date': upload_date,
             'thumbnail': thumbnail,
         }
-
-
-class SportschauIE(ARDMediathekIE):
-    IE_NAME = 'Sportschau'
-    _VALID_URL = r'(?P<baseurl>https?://(?:www\.)?sportschau\.de/(?:[^/]+/)+video(?P<id>[^/#?]+))\.html'
-    _TESTS = [{
-        'url': 'http://www.sportschau.de/tourdefrance/videoseppeltkokainhatnichtsmitklassischemdopingzutun100.html',
-        'info_dict': {
-            'id': 'seppeltkokainhatnichtsmitklassischemdopingzutun100',
-            'ext': 'mp4',
-            'title': 'Seppelt: "Kokain hat nichts mit klassischem Doping zu tun"',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'description': 'Der ARD-Doping Experte Hajo Seppelt gibt seine Einschätzung zum ersten Dopingfall der diesjährigen Tour de France um den Italiener Luca Paolini ab.',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
-    }]
-
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        base_url = mobj.group('baseurl')
-
-        webpage = self._download_webpage(url, video_id)
-        title = get_element_by_attribute('class', 'headline', webpage)
-        description = self._html_search_meta('description', webpage, 'description')
-
-        info = self._extract_media_info(
-            base_url + '-mc_defaultQuality-h.json', webpage, video_id)
-
-        info.update({
-            'title': title,
-            'description': description,
-        })
-
-        return info

youtube_dl/extractor/arte.py

@@ -180,11 +180,14 @@ class ArteTVBaseIE(InfoExtractor):
 
 class ArteTVPlus7IE(ArteTVBaseIE):
     IE_NAME = 'arte.tv:+7'
-    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&]+)'
+    _VALID_URL = r'https?://(?:(?:www|sites)\.)?arte\.tv/[^/]+/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
     _TESTS = [{
         'url': 'http://www.arte.tv/guide/de/sendungen/XEN/xenius/?vid=055918-015_PLUS7-D',
         'only_matching': True,
+    }, {
+        'url': 'http://sites.arte.tv/karambolage/de/video/karambolage-22',
+        'only_matching': True,
     }]
 
     @classmethod
@@ -240,10 +243,10 @@ class ArteTVPlus7IE(ArteTVBaseIE):
             return self._extract_from_json_url(json_url, video_id, lang, title=title)
         # Different kind of embed URL (e.g.
         # http://www.arte.tv/magazine/trepalium/fr/episode-0406-replay-trepalium)
-        embed_url = self._search_regex(
+        entries = [
-            r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1',
+            self.url_result(url)
-            webpage, 'embed url', group='url')
+            for _, url in re.findall(r'<iframe[^>]+src=(["\'])(?P<url>.+?)\1', webpage)]
-        return self.url_result(embed_url)
+        return self.playlist_result(entries)
 
 
 # It also uses the arte_vp_url url from the webpage to extract the information
@@ -252,22 +255,17 @@ class ArteTVCreativeIE(ArteTVPlus7IE):
     _VALID_URL = r'https?://creative\.arte\.tv/(?P<lang>fr|de|en|es)/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
     _TESTS = [{
-        'url': 'http://creative.arte.tv/de/magazin/agentur-amateur-corporate-design',
+        'url': 'http://creative.arte.tv/fr/episode/osmosis-episode-1',
         'info_dict': {
-            'id': '72176',
+            'id': '057405-001-A',
             'ext': 'mp4',
-            'title': 'Folge 2 - Corporate Design',
+            'title': 'OSMOSIS - N\'AYEZ PLUS PEUR D\'AIMER (1)',
-            'upload_date': '20131004',
+            'upload_date': '20150716',
         },
     }, {
         'url': 'http://creative.arte.tv/fr/Monty-Python-Reunion',
-        'info_dict': {
+        'playlist_count': 11,
-            'id': '160676',
+        'add_ie': ['Youtube'],
-            'ext': 'mp4',
-            'title': 'Monty Python live (mostly)',
-            'description': 'Événement ! Quarante-cinq ans après leurs premiers succès, les légendaires Monty Python remontent sur scène.\n',
-            'upload_date': '20140805',
-        }
     }, {
         'url': 'http://creative.arte.tv/de/episode/agentur-amateur-4-der-erste-kunde',
         'only_matching': True,
@@ -349,14 +347,13 @@ class ArteTVCinemaIE(ArteTVPlus7IE):
     _VALID_URL = r'https?://cinema\.arte\.tv/(?P<lang>fr|de|en|es)/(?P<id>.+)'
 
     _TESTS = [{
-        'url': 'http://cinema.arte.tv/de/node/38291',
+        'url': 'http://cinema.arte.tv/fr/article/les-ailes-du-desir-de-julia-reck',
-        'md5': '6b275511a5107c60bacbeeda368c3aa1',
+        'md5': 'a5b9dd5575a11d93daf0e3f404f45438',
         'info_dict': {
-            'id': '055876-000_PWA12025-D',
+            'id': '062494-000-A',
             'ext': 'mp4',
-            'title': 'Tod auf dem Nil',
+            'title': 'Film lauréat du concours web - "Les ailes du désir" de Julia Reck',
-            'upload_date': '20160122',
+            'upload_date': '20150807',
-            'description': 'md5:7f749bbb77d800ef2be11d54529b96bc',
         },
     }]
 
@@ -422,6 +419,7 @@ class ArteTVPlaylistIE(ArteTVBaseIE):
         'info_dict': {
             'id': 'PL-013263',
             'title': 'Areva & Uramin',
+            'description': 'md5:a1dc0312ce357c262259139cfd48c9bf',
         },
         'playlist_mincount': 6,
     }, {

youtube_dl/extractor/audiomack.py

@@ -6,6 +6,7 @@ import time
 
 from .common import InfoExtractor
 from .soundcloud import SoundcloudIE
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     url_basename,
@@ -136,7 +137,7 @@ class AudiomackAlbumIE(InfoExtractor):
                         result[resultkey] = api_response[apikey]
                 song_id = url_basename(api_response['url']).rpartition('.')[0]
                 result['entries'].append({
-                    'id': api_response.get('id', song_id),
+                    'id': compat_str(api_response.get('id', song_id)),
                     'uploader': api_response.get('artist'),
                     'title': api_response.get('title', song_id),
                     'url': api_response['url'],

youtube_dl/extractor/azubu.py

@@ -46,6 +46,7 @@ class AzubuIE(InfoExtractor):
                 'uploader_id': 272749,
                 'view_count': int,
             },
+            'skip': 'Channel offline',
         },
     ]
 
@@ -56,22 +57,26 @@ class AzubuIE(InfoExtractor):
             'http://www.azubu.tv/api/video/%s' % video_id, video_id)['data']
 
         title = data['title'].strip()
-        description = data['description']
+        description = data.get('description')
-        thumbnail = data['thumbnail']
+        thumbnail = data.get('thumbnail')
-        view_count = data['view_count']
+        view_count = data.get('view_count')
-        uploader = data['user']['username']
+        user = data.get('user', {})
-        uploader_id = data['user']['id']
+        uploader = user.get('username')
+        uploader_id = user.get('id')
 
         stream_params = json.loads(data['stream_params'])
 
-        timestamp = float_or_none(stream_params['creationDate'], 1000)
+        timestamp = float_or_none(stream_params.get('creationDate'), 1000)
-        duration = float_or_none(stream_params['length'], 1000)
+        duration = float_or_none(stream_params.get('length'), 1000)
 
         renditions = stream_params.get('renditions') or []
         video = stream_params.get('FLVFullLength') or stream_params.get('videoFullLength')
         if video:
             renditions.append(video)
 
+        if not renditions and not user.get('channel', {}).get('is_live', True):
+            raise ExtractorError('%s said: channel is offline.' % self.IE_NAME, expected=True)
+
         formats = [{
             'url': fmt['url'],
             'width': fmt['frameWidth'],

youtube_dl/extractor/bbc.py

@@ -31,7 +31,7 @@ class BBCCoUkIE(InfoExtractor):
                             music/clips[/#]|
                             radio/player/
                         )
-                        (?P<id>%s)
+                        (?P<id>%s)(?!/(?:episodes|broadcasts|clips))
                     ''' % _ID_REGEX
 
     _MEDIASELECTOR_URLS = [
@@ -44,6 +44,8 @@ class BBCCoUkIE(InfoExtractor):
 
     _MEDIASELECTION_NS = 'http://bbc.co.uk/2008/mp/mediaselection'
     _EMP_PLAYLIST_NS = 'http://bbc.co.uk/2008/emp/playlist'
+    # Unified Streaming Platform
+    _USP_RE = r'/([^/]+)\.ism(?:\.hlsv2\.ism)?/[^/]+\.m3u8'
 
     _NAMESPACES = (
         _MEDIASELECTION_NS,
@@ -55,12 +57,11 @@ class BBCCoUkIE(InfoExtractor):
             'url': 'http://www.bbc.co.uk/programmes/b039g8p7',
             'info_dict': {
                 'id': 'b039d07m',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Leonard Cohen, Kaleidoscope - BBC Radio 4',
                 'description': 'The Canadian poet and songwriter reflects on his musical career.',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             }
         },
@@ -92,7 +93,7 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
-            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
+            'skip': 'this episode is not currently available',
         },
         {
             'url': 'http://www.bbc.co.uk/iplayer/episode/p026c7jt/tomorrows-worlds-the-unearthly-history-of-science-fiction-2-invasion',
@@ -107,7 +108,7 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
-            'skip': 'Currently BBC iPlayer TV programmes are available to play in the UK only',
+            'skip': 'this episode is not currently available',
         }, {
             'url': 'http://www.bbc.co.uk/programmes/b04v20dw',
             'info_dict': {
@@ -127,13 +128,12 @@ class BBCCoUkIE(InfoExtractor):
             'note': 'Audio',
             'info_dict': {
                 'id': 'p022h44j',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'BBC Proms Music Guides, Rachmaninov: Symphonic Dances',
                 'description': "In this Proms Music Guide, Andrew McGregor looks at Rachmaninov's Symphonic Dances.",
                 'duration': 227,
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             }
         }, {
@@ -141,13 +141,12 @@ class BBCCoUkIE(InfoExtractor):
             'note': 'Video',
             'info_dict': {
                 'id': 'p025c103',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Reading and Leeds Festival, 2014, Rae Morris - Closer (Live on BBC Three)',
                 'description': 'Rae Morris performs Closer for BBC Three at Reading 2014',
                 'duration': 226,
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             }
         }, {
@@ -163,7 +162,7 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
-            'skip': 'geolocation',
+            'skip': 'this episode is not currently available',
         }, {
             'url': 'http://www.bbc.co.uk/iplayer/episode/b05zmgwn/royal-academy-summer-exhibition',
             'info_dict': {
@@ -177,7 +176,7 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
-            'skip': 'geolocation',
+            'skip': 'this episode is not currently available',
         }, {
             # iptv-all mediaset fails with geolocation however there is no geo restriction
             # for this programme at all
@@ -192,17 +191,17 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'this episode is not currently available on BBC iPlayer Radio',
         }, {
             # compact player (https://github.com/rg3/youtube-dl/issues/8147)
             'url': 'http://www.bbc.co.uk/programmes/p028bfkf/player',
             'info_dict': {
                 'id': 'p028bfkj',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
                 'description': 'Extract from BBC documentary Look Stranger - Giant Leeks and Magic Brews',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         }, {
@@ -247,9 +246,15 @@ class BBCCoUkIE(InfoExtractor):
             elif transfer_format == 'dash':
                 pass
             elif transfer_format == 'hls':
-                formats.extend(self._extract_m3u8_formats(
+                is_unified_streaming = re.search(self._USP_RE, href)
+                if is_unified_streaming:
+                    href = re.sub(self._USP_RE, r'/\1.ism/\1.m3u8', href)
+                m3u8_formats = self._extract_m3u8_formats(
                     href, programme_id, ext='mp4', entry_protocol='m3u8_native',
-                    m3u8_id=supplier, fatal=False))
+                    m3u8_id=supplier, fatal=False)
+                if is_unified_streaming:
+                    self._check_formats(m3u8_formats, programme_id)
+                formats.extend(m3u8_formats)
             # Direct link
             else:
                 formats.append({
@@ -304,13 +309,14 @@ class BBCCoUkIE(InfoExtractor):
         for connection in self._extract_connections(media):
             conn_formats = self._extract_connection(connection, programme_id)
             for format in conn_formats:
-                format.update({
+                if format.get('protocol') != 'm3u8_native':
-                    'width': width,
+                    format.update({
-                    'height': height,
+                        'width': width,
-                    'vbr': vbr,
+                        'height': height,
-                    'vcodec': vcodec,
+                        'vbr': vbr,
-                    'filesize': file_size,
+                        'vcodec': vcodec,
-                })
+                        'filesize': file_size,
+                    })
                 if service:
                     format['format_id'] = '%s_%s' % (service, format['format_id'])
             formats.extend(conn_formats)
@@ -698,7 +704,9 @@ class BBCIE(BBCCoUkIE):
 
     @classmethod
     def suitable(cls, url):
-        return False if BBCCoUkIE.suitable(url) or BBCCoUkArticleIE.suitable(url) else super(BBCIE, cls).suitable(url)
+        EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE)
+        return (False if any(ie.suitable(url) for ie in EXCLUDE_IE)
+                else super(BBCIE, cls).suitable(url))
 
     def _extract_from_media_meta(self, media_meta, video_id):
         # Direct links to media in media metadata (e.g.
@@ -975,3 +983,72 @@ class BBCCoUkArticleIE(InfoExtractor):
             r'<div[^>]+typeof="Clip"[^>]+resource="([^"]+)"', webpage)]
 
         return self.playlist_result(entries, playlist_id, title, description)
+
+
+class BBCCoUkPlaylistBaseIE(InfoExtractor):
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, playlist_id)
+
+        entries = [
+            self.url_result(self._URL_TEMPLATE % video_id, BBCCoUkIE.ie_key())
+            for video_id in re.findall(
+                self._VIDEO_ID_TEMPLATE % BBCCoUkIE._ID_REGEX, webpage)]
+
+        title, description = self._extract_title_and_description(webpage)
+
+        return self.playlist_result(entries, playlist_id, title, description)
+
+
+class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE):
+    IE_NAME = 'bbc.co.uk:iplayer:playlist'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/episodes/(?P<id>%s)' % BBCCoUkIE._ID_REGEX
+    _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s'
+    _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)'
+    _TEST = {
+        'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v',
+        'info_dict': {
+            'id': 'b05rcz9v',
+            'title': 'The Disappearance',
+            'description': 'French thriller serial about a missing teenager.',
+        },
+        'playlist_mincount': 6,
+    }
+
+    def _extract_title_and_description(self, webpage):
+        title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False)
+        description = self._search_regex(
+            r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>',
+            webpage, 'description', fatal=False, group='value')
+        return title, description
+
+
+class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE):
+    IE_NAME = 'bbc.co.uk:playlist'
+    _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/programmes/(?P<id>%s)/(?:episodes|broadcasts|clips)' % BBCCoUkIE._ID_REGEX
+    _URL_TEMPLATE = 'http://www.bbc.co.uk/programmes/%s'
+    _VIDEO_ID_TEMPLATE = r'data-pid=["\'](%s)'
+    _TESTS = [{
+        'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
+        'info_dict': {
+            'id': 'b05rcz9v',
+            'title': 'The Disappearance - Clips - BBC Four',
+            'description': 'French thriller serial about a missing teenager.',
+        },
+        'playlist_mincount': 7,
+    }, {
+        'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/broadcasts/2016/06',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.bbc.co.uk/programmes/b05rcz9v/clips',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.bbc.co.uk/programmes/b055jkys/episodes/player',
+        'only_matching': True,
+    }]
+
+    def _extract_title_and_description(self, webpage):
+        title = self._og_search_title(webpage, fatal=False)
+        description = self._og_search_description(webpage)
+        return title, description

youtube_dl/extractor/bet.py

@@ -1,31 +1,27 @@
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
+from .mtv import MTVServicesInfoExtractor
-from ..compat import compat_urllib_parse_unquote
+from ..utils import unified_strdate
-from ..utils import (
+from ..compat import compat_urllib_parse_urlencode
-    xpath_text,
-    xpath_with_ns,
-    int_or_none,
-    parse_iso8601,
-)
 
 
-class BetIE(InfoExtractor):
+class BetIE(MTVServicesInfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?bet\.com/(?:[^/]+/)+(?P<id>.+?)\.html'
     _TESTS = [
         {
             'url': 'http://www.bet.com/news/politics/2014/12/08/in-bet-exclusive-obama-talks-race-and-racism.html',
             'info_dict': {
-                'id': 'news/national/2014/a-conversation-with-president-obama',
+                'id': '07e96bd3-8850-3051-b856-271b457f0ab8',
                 'display_id': 'in-bet-exclusive-obama-talks-race-and-racism',
                 'ext': 'flv',
                 'title': 'A Conversation With President Obama',
-                'description': 'md5:699d0652a350cf3e491cd15cc745b5da',
+                'description': 'President Obama urges persistence in confronting racism and bias.',
                 'duration': 1534,
-                'timestamp': 1418075340,
                 'upload_date': '20141208',
-                'uploader': 'admin',
                 'thumbnail': 're:(?i)^https?://.*\.jpg$',
+                'subtitles': {
+                    'en': 'mincount:2',
+                }
             },
             'params': {
                 # rtmp download
@@ -35,16 +31,17 @@ class BetIE(InfoExtractor):
         {
             'url': 'http://www.bet.com/video/news/national/2014/justice-for-ferguson-a-community-reacts.html',
             'info_dict': {
-                'id': 'news/national/2014/justice-for-ferguson-a-community-reacts',
+                'id': '9f516bf1-7543-39c4-8076-dd441b459ba9',
                 'display_id': 'justice-for-ferguson-a-community-reacts',
                 'ext': 'flv',
                 'title': 'Justice for Ferguson: A Community Reacts',
                 'description': 'A BET News special.',
                 'duration': 1696,
-                'timestamp': 1416942360,
                 'upload_date': '20141125',
-                'uploader': 'admin',
                 'thumbnail': 're:(?i)^https?://.*\.jpg$',
+                'subtitles': {
+                    'en': 'mincount:2',
+                }
             },
             'params': {
                 # rtmp download
@@ -53,57 +50,32 @@ class BetIE(InfoExtractor):
         }
     ]
 
+    _FEED_URL = "http://feeds.mtvnservices.com/od/feed/bet-mrss-player"
+
+    def _get_feed_query(self, uri):
+        return compat_urllib_parse_urlencode({
+            'uuid': uri,
+        })
+
+    def _extract_mgid(self, webpage):
+        return self._search_regex(r'data-uri="([^"]+)', webpage, 'mgid')
+
     def _real_extract(self, url):
         display_id = self._match_id(url)
+
         webpage = self._download_webpage(url, display_id)
+        mgid = self._extract_mgid(webpage)
+        videos_info = self._get_videos_info(mgid)
 
-        media_url = compat_urllib_parse_unquote(self._search_regex(
+        info_dict = videos_info['entries'][0]
-            [r'mediaURL\s*:\s*"([^"]+)"', r"var\s+mrssMediaUrl\s*=\s*'([^']+)'"],
-            webpage, 'media URL'))
 
-        video_id = self._search_regex(
+        upload_date = unified_strdate(self._html_search_meta('date', webpage))
-            r'/video/(.*)/_jcr_content/', media_url, 'video id')
+        description = self._html_search_meta('description', webpage)
 
-        mrss = self._download_xml(media_url, display_id)
+        info_dict.update({
-
-        item = mrss.find('./channel/item')
-
-        NS_MAP = {
-            'dc': 'http://purl.org/dc/elements/1.1/',
-            'media': 'http://search.yahoo.com/mrss/',
-            'ka': 'http://kickapps.com/karss',
-        }
-
-        title = xpath_text(item, './title', 'title')
-        description = xpath_text(
-            item, './description', 'description', fatal=False)
-
-        timestamp = parse_iso8601(xpath_text(
-            item, xpath_with_ns('./dc:date', NS_MAP),
-            'upload date', fatal=False))
-        uploader = xpath_text(
-            item, xpath_with_ns('./dc:creator', NS_MAP),
-            'uploader', fatal=False)
-
-        media_content = item.find(
-            xpath_with_ns('./media:content', NS_MAP))
-        duration = int_or_none(media_content.get('duration'))
-        smil_url = media_content.get('url')
-
-        thumbnail = media_content.find(
-            xpath_with_ns('./media:thumbnail', NS_MAP)).get('url')
-
-        formats = self._extract_smil_formats(smil_url, display_id)
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
             'display_id': display_id,
-            'title': title,
             'description': description,
-            'thumbnail': thumbnail,
+            'upload_date': upload_date,
-            'timestamp': timestamp,
+        })
-            'uploader': uploader,
+
-            'duration': duration,
+        return info_dict
-            'formats': formats,
-        }

youtube_dl/extractor/bilibili.py

@@ -102,6 +102,22 @@ class BiliBiliIE(InfoExtractor):
                 'uploader_id': '151066',
             },
         }],
+    }, {
+        # Missing upload time
+        'url': 'http://www.bilibili.com/video/av1867637/',
+        'info_dict': {
+            'id': '2880301',
+            'ext': 'flv',
+            'title': '【HDTV】【喜剧】岳父岳母真难当 （2014）【法国票房冠军】',
+            'description': '一个信奉天主教的法国旧式传统资产阶级家庭中有四个女儿。三个女儿却分别找了阿拉伯、犹太、中国丈夫，老夫老妻唯独期盼剩下未嫁的小女儿能找一个信奉天主教的法国白人，结果没想到小女儿找了一位非裔黑人……【这次应该不会跳帧了】',
+            'uploader': '黑夜为猫',
+            'uploader_id': '610729',
+        },
+        'params': {
+            # Just to test metadata extraction
+            'skip_download': True,
+        },
+        'expected_warnings': ['upload time'],
     }]
 
     # BiliBili blocks keys from time to time. The current key is extracted from
@@ -172,6 +188,7 @@ class BiliBiliIE(InfoExtractor):
         description = self._html_search_meta('description', webpage)
         datetime_str = self._html_search_regex(
             r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time', fatal=False)
+        timestamp = None
         if datetime_str:
             timestamp = calendar.timegm(datetime.datetime.strptime(datetime_str, '%Y-%m-%dT%H:%M').timetuple())
 

youtube_dl/extractor/biobiochiletv.py

@@ -2,11 +2,15 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import remove_end
+from ..utils import (
+    ExtractorError,
+    remove_end,
+)
+from .rudo import RudoIE
 
 
 class BioBioChileTVIE(InfoExtractor):
-    _VALID_URL = r'https?://tv\.biobiochile\.cl/notas/(?:[^/]+/)+(?P<id>[^/]+)\.shtml'
+    _VALID_URL = r'https?://(?:tv|www)\.biobiochile\.cl/(?:notas|noticias)/(?:[^/]+/)+(?P<id>[^/]+)\.shtml'
 
     _TESTS = [{
         'url': 'http://tv.biobiochile.cl/notas/2015/10/21/sobre-camaras-y-camarillas-parlamentarias.shtml',
@@ -18,6 +22,7 @@ class BioBioChileTVIE(InfoExtractor):
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': 'Fernando Atria',
         },
+        'skip': 'URL expired and redirected to http://www.biobiochile.cl/portada/bbtv/index.html',
     }, {
         # different uploader layout
         'url': 'http://tv.biobiochile.cl/notas/2016/03/18/natalia-valdebenito-repasa-a-diputado-hasbun-paso-a-la-categoria-de-hablar-brutalidades.shtml',
@@ -32,6 +37,16 @@ class BioBioChileTVIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+        'skip': 'URL expired and redirected to http://www.biobiochile.cl/portada/bbtv/index.html',
+    }, {
+        'url': 'http://www.biobiochile.cl/noticias/bbtv/comentarios-bio-bio/2016/07/08/edecanes-del-congreso-figuras-decorativas-que-le-cuestan-muy-caro-a-los-chilenos.shtml',
+        'info_dict': {
+            'id': 'edecanes-del-congreso-figuras-decorativas-que-le-cuestan-muy-caro-a-los-chilenos',
+            'ext': 'mp4',
+            'uploader': '(none)',
+            'upload_date': '20160708',
+            'title': 'Edecanes del Congreso: Figuras decorativas que le cuestan muy caro a los chilenos',
+        },
     }, {
         'url': 'http://tv.biobiochile.cl/notas/2015/10/22/ninos-transexuales-de-quien-es-la-decision.shtml',
         'only_matching': True,
@@ -45,42 +60,22 @@ class BioBioChileTVIE(InfoExtractor):
 
         webpage = self._download_webpage(url, video_id)
 
+        rudo_url = RudoIE._extract_url(webpage)
+        if not rudo_url:
+            raise ExtractorError('No videos found')
+
         title = remove_end(self._og_search_title(webpage), ' - BioBioChile TV')
 
-        file_url = self._search_regex(
-            r'loadFWPlayerVideo\([^,]+,\s*(["\'])(?P<url>.+?)\1',
-            webpage, 'file url', group='url')
-
-        base_url = self._search_regex(
-            r'file\s*:\s*(["\'])(?P<url>.+?)\1\s*\+\s*fileURL', webpage,
-            'base url', default='http://unlimited2-cl.digitalproserver.com/bbtv/',
-            group='url')
-
-        formats = self._extract_m3u8_formats(
-            '%s%s/playlist.m3u8' % (base_url, file_url), video_id, 'mp4',
-            entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)
-        f = {
-            'url': '%s%s' % (base_url, file_url),
-            'format_id': 'http',
-            'protocol': 'http',
-            'preference': 1,
-        }
-        if formats:
-            f_copy = formats[-1].copy()
-            f_copy.update(f)
-            f = f_copy
-        formats.append(f)
-        self._sort_formats(formats)
-
         thumbnail = self._og_search_thumbnail(webpage)
         uploader = self._html_search_regex(
-            r'<a[^>]+href=["\']https?://busca\.biobiochile\.cl/author[^>]+>(.+?)</a>',
+            r'<a[^>]+href=["\']https?://(?:busca|www)\.biobiochile\.cl/(?:lista/)?(?:author|autor)[^>]+>(.+?)</a>',
             webpage, 'uploader', fatal=False)
 
         return {
+            '_type': 'url_transparent',
+            'url': rudo_url,
             'id': video_id,
             'title': title,
             'thumbnail': thumbnail,
             'uploader': uploader,
-            'formats': formats,
         }

youtube_dl/extractor/br.py

@@ -29,7 +29,8 @@ class BRIE(InfoExtractor):
                 'duration': 180,
                 'uploader': 'Reinhard Weber',
                 'upload_date': '20150422',
-            }
+            },
+            'skip': '404 not found',
         },
         {
             'url': 'http://www.br.de/nachrichten/oberbayern/inhalt/muenchner-polizeipraesident-schreiber-gestorben-100.html',
@@ -40,7 +41,8 @@ class BRIE(InfoExtractor):
                 'title': 'Manfred Schreiber ist tot',
                 'description': 'md5:b454d867f2a9fc524ebe88c3f5092d97',
                 'duration': 26,
-            }
+            },
+            'skip': '404 not found',
         },
         {
             'url': 'https://www.br-klassik.de/audio/peeping-tom-premierenkritik-dance-festival-muenchen-100.html',
@@ -51,7 +53,8 @@ class BRIE(InfoExtractor):
                 'title': 'Kurzweilig und sehr bewegend',
                 'description': 'md5:0351996e3283d64adeb38ede91fac54e',
                 'duration': 296,
-            }
+            },
+            'skip': '404 not found',
         },
         {
             'url': 'http://www.br.de/radio/bayern1/service/team/videos/team-video-erdelt100.html',

youtube_dl/extractor/brightcove.py

@@ -26,6 +26,7 @@ from ..utils import (
     unescapeHTML,
     unsmuggle_url,
     update_url_query,
+    clean_html,
 )
 
 
@@ -90,6 +91,7 @@ class BrightcoveLegacyIE(InfoExtractor):
                 'description': 'md5:363109c02998fee92ec02211bd8000df',
                 'uploader': 'National Ballet of Canada',
             },
+            'skip': 'Video gone',
         },
         {
             # test flv videos served by akamaihd.net
@@ -108,7 +110,7 @@ class BrightcoveLegacyIE(InfoExtractor):
             },
         },
         {
-            # playlist test
+            # playlist with 'videoList'
             # from http://support.brightcove.com/en/video-cloud/docs/playlist-support-single-video-players
             'url': 'http://c.brightcove.com/services/viewer/htmlFederated?playerID=3550052898001&playerKey=AQ%7E%7E%2CAAABmA9XpXk%7E%2C-Kp7jNgisre1fG5OdqpAFUTcs0lP_ZoL',
             'info_dict': {
@@ -117,6 +119,15 @@ class BrightcoveLegacyIE(InfoExtractor):
             },
             'playlist_mincount': 7,
         },
+        {
+            # playlist with 'playlistTab' (https://github.com/rg3/youtube-dl/issues/9965)
+            'url': 'http://c.brightcove.com/services/json/experience/runtime/?command=get_programming_for_experience&playerKey=AQ%7E%7E,AAABXlLMdok%7E,NJ4EoMlZ4rZdx9eU1rkMVd8EaYPBBUlg',
+            'info_dict': {
+                'id': '1522758701001',
+                'title': 'Lesson 08',
+            },
+            'playlist_mincount': 10,
+        },
     ]
     FLV_VCODECS = {
         1: 'SORENSON',
@@ -298,13 +309,19 @@ class BrightcoveLegacyIE(InfoExtractor):
             info_url, player_key, 'Downloading playlist information')
 
         json_data = json.loads(playlist_info)
-        if 'videoList' not in json_data:
+        if 'videoList' in json_data:
+            playlist_info = json_data['videoList']
+            playlist_dto = playlist_info['mediaCollectionDTO']
+        elif 'playlistTabs' in json_data:
+            playlist_info = json_data['playlistTabs']
+            playlist_dto = playlist_info['lineupListDTO']['playlistDTOs'][0]
+        else:
             raise ExtractorError('Empty playlist')
-        playlist_info = json_data['videoList']
+
-        videos = [self._extract_video_info(video_info) for video_info in playlist_info['mediaCollectionDTO']['videoDTOs']]
+        videos = [self._extract_video_info(video_info) for video_info in playlist_dto['videoDTOs']]
 
         return self.playlist_result(videos, playlist_id='%s' % playlist_info['id'],
-                                    playlist_title=playlist_info['mediaCollectionDTO']['displayName'])
+                                    playlist_title=playlist_dto['displayName'])
 
     def _extract_video_info(self, video_info):
         video_id = compat_str(video_info['id'])
@@ -585,6 +602,13 @@ class BrightcoveNewIE(InfoExtractor):
                         'format_id': build_format_id('rtmp'),
                     })
                 formats.append(f)
+
+        errors = json_data.get('errors')
+        if not formats and errors:
+            error = errors[0]
+            raise ExtractorError(
+                error.get('message') or error.get('error_subcode') or error['error_code'], expected=True)
+
         self._sort_formats(formats)
 
         subtitles = {}
@@ -597,7 +621,7 @@ class BrightcoveNewIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
-            'description': json_data.get('description'),
+            'description': clean_html(json_data.get('description')),
             'thumbnail': json_data.get('thumbnail') or json_data.get('poster'),
             'duration': float_or_none(json_data.get('duration'), 1000),
             'timestamp': parse_iso8601(json_data.get('published_at')),

youtube_dl/extractor/buzzfeed.py

@@ -5,6 +5,7 @@ import json
 import re
 
 from .common import InfoExtractor
+from .facebook import FacebookIE
 
 
 class BuzzFeedIE(InfoExtractor):
@@ -20,11 +21,11 @@ class BuzzFeedIE(InfoExtractor):
             'info_dict': {
                 'id': 'aVCR29aE_OQ',
                 'ext': 'mp4',
+                'title': 'Angry Ram destroys a punching bag..',
+                'description': 'md5:c59533190ef23fd4458a5e8c8c872345',
                 'upload_date': '20141024',
                 'uploader_id': 'Buddhanz1',
-                'description': 'He likes to stay in shape with his heavy bag, he wont stop until its on the ground\n\nFollow Angry Ram on Facebook for regular updates -\nhttps://www.facebook.com/pages/Angry-Ram/1436897249899558?ref=hl',
+                'uploader': 'Angry Ram',
-                'uploader': 'Buddhanz',
-                'title': 'Angry Ram destroys a punching bag',
             }
         }]
     }, {
@@ -41,13 +42,30 @@ class BuzzFeedIE(InfoExtractor):
             'info_dict': {
                 'id': 'mVmBL8B-In0',
                 'ext': 'mp4',
+                'title': 're:Munchkin the Teddy Bear gets her exercise',
+                'description': 'md5:28faab95cda6e361bcff06ec12fc21d8',
                 'upload_date': '20141124',
                 'uploader_id': 'CindysMunchkin',
-                'description': 're:© 2014 Munchkin the',
                 'uploader': 're:^Munchkin the',
-                'title': 're:Munchkin the Teddy Bear gets her exercise',
             },
         }]
+    }, {
+        'url': 'http://www.buzzfeed.com/craigsilverman/the-most-adorable-crash-landing-ever#.eq7pX0BAmK',
+        'info_dict': {
+            'id': 'the-most-adorable-crash-landing-ever',
+            'title': 'Watch This Baby Goose Make The Most Adorable Crash Landing',
+            'description': 'This gosling knows how to stick a landing.',
+        },
+        'playlist': [{
+            'md5': '763ca415512f91ca62e4621086900a23',
+            'info_dict': {
+                'id': '971793786185728',
+                'ext': 'mp4',
+                'title': 'We set up crash pads so that the goslings on our roof would have a safe landi...',
+                'uploader': 'Calgary Outdoor Centre-University of Calgary',
+            },
+        }],
+        'add_ie': ['Facebook'],
     }]
 
     def _real_extract(self, url):
@@ -66,6 +84,10 @@ class BuzzFeedIE(InfoExtractor):
                 continue
             entries.append(self.url_result(video['url']))
 
+        facebook_url = FacebookIE._extract_url(webpage)
+        if facebook_url:
+            entries.append(self.url_result(facebook_url))
+
         return {
             '_type': 'playlist',
             'id': playlist_id,

youtube_dl/extractor/canalplus.py

@@ -4,11 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urllib_parse_urlparse
 from ..utils import (
     ExtractorError,
     HEADRequest,
     unified_strdate,
-    url_basename,
     qualities,
     int_or_none,
 )
@@ -16,24 +16,38 @@ from ..utils import (
 
 class CanalplusIE(InfoExtractor):
     IE_DESC = 'canalplus.fr, piwiplus.fr and d8.tv'
-    _VALID_URL = r'https?://(?:www\.(?P<site>canalplus\.fr|piwiplus\.fr|d8\.tv|itele\.fr)/.*?/(?P<path>.*)|player\.canalplus\.fr/#/(?P<id>[0-9]+))'
+    _VALID_URL = r'''(?x)
+                        https?://
+                            (?:
+                                (?:
+                                    (?:(?:www|m)\.)?canalplus\.fr|
+                                    (?:www\.)?piwiplus\.fr|
+                                    (?:www\.)?d8\.tv|
+                                    (?:www\.)?d17\.tv|
+                                    (?:www\.)?itele\.fr
+                                )/(?:(?:[^/]+/)*(?P<display_id>[^/?#&]+))?(?:\?.*\bvid=(?P<vid>\d+))?|
+                                player\.canalplus\.fr/#/(?P<id>\d+)
+                            )
+
+                    '''
     _VIDEO_INFO_TEMPLATE = 'http://service.canal-plus.com/video/rest/getVideosLiees/%s/%s?format=json'
     _SITE_ID_MAP = {
-        'canalplus.fr': 'cplus',
+        'canalplus': 'cplus',
-        'piwiplus.fr': 'teletoon',
+        'piwiplus': 'teletoon',
-        'd8.tv': 'd8',
+        'd8': 'd8',
-        'itele.fr': 'itele',
+        'd17': 'd17',
+        'itele': 'itele',
     }
 
     _TESTS = [{
-        'url': 'http://www.canalplus.fr/c-emissions/pid1830-c-zapping.html?vid=1263092',
+        'url': 'http://www.canalplus.fr/c-emissions/pid1830-c-zapping.html?vid=1192814',
-        'md5': '12164a6f14ff6df8bd628e8ba9b10b78',
+        'md5': '41f438a4904f7664b91b4ed0dec969dc',
         'info_dict': {
-            'id': '1263092',
+            'id': '1192814',
             'ext': 'mp4',
-            'title': 'Le Zapping - 13/05/15',
+            'title': "L'Année du Zapping 2014 - L'Année du Zapping 2014",
-            'description': 'md5:09738c0d06be4b5d06a0940edb0da73f',
+            'description': "Toute l'année 2014 dans un Zapping exceptionnel !",
-            'upload_date': '20150513',
+            'upload_date': '20150105',
         },
     }, {
         'url': 'http://www.piwiplus.fr/videos-piwi/pid1405-le-labyrinthe-boing-super-ranger.html?vid=1108190',
@@ -46,35 +60,45 @@ class CanalplusIE(InfoExtractor):
         },
         'skip': 'Only works from France',
     }, {
-        'url': 'http://www.d8.tv/d8-docs-mags/pid6589-d8-campagne-intime.html',
+        'url': 'http://www.d8.tv/d8-docs-mags/pid5198-d8-en-quete-d-actualite.html?vid=1390231',
         'info_dict': {
-            'id': '966289',
+            'id': '1390231',
-            'ext': 'flv',
-            'title': 'Campagne intime - Documentaire exceptionnel',
-            'description': 'md5:d2643b799fb190846ae09c61e59a859f',
-            'upload_date': '20131108',
-        },
-        'skip': 'videos get deleted after a while',
-    }, {
-        'url': 'http://www.itele.fr/france/video/aubervilliers-un-lycee-en-colere-111559',
-        'md5': '38b8f7934def74f0d6f3ba6c036a5f82',
-        'info_dict': {
-            'id': '1213714',
             'ext': 'mp4',
-            'title': 'Aubervilliers : un lycée en colère - Le 11/02/2015 à 06h45',
+            'title': "Vacances pas chères : prix discount ou grosses dépenses ? - En quête d'actualité",
-            'description': 'md5:8216206ec53426ea6321321f3b3c16db',
+            'description': 'md5:edb6cf1cb4a1e807b5dd089e1ac8bfc6',
-            'upload_date': '20150211',
+            'upload_date': '20160512',
         },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.itele.fr/chroniques/invite-bruce-toussaint/thierry-solere-nicolas-sarkozy-officialisera-sa-candidature-a-la-primaire-quand-il-le-voudra-167224',
+        'info_dict': {
+            'id': '1398334',
+            'ext': 'mp4',
+            'title': "L'invité de Bruce Toussaint du 07/06/2016 - ",
+            'description': 'md5:40ac7c9ad0feaeb6f605bad986f61324',
+            'upload_date': '20160607',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://m.canalplus.fr/?vid=1398231',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.d17.tv/emissions/pid8303-lolywood.html?vid=1397061',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.groupdict().get('id')
+        video_id = mobj.groupdict().get('id') or mobj.groupdict().get('vid')
 
-        site_id = self._SITE_ID_MAP[mobj.group('site') or 'canal']
+        site_id = self._SITE_ID_MAP[compat_urllib_parse_urlparse(url).netloc.rsplit('.', 2)[-2]]
 
         # Beware, some subclasses do not define an id group
-        display_id = url_basename(mobj.group('path'))
+        display_id = mobj.group('display_id') or video_id
 
         if video_id is None:
             webpage = self._download_webpage(url, display_id)

youtube_dl/extractor/carambatv.py

@@ -0,0 +1,88 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    float_or_none,
+    int_or_none,
+    try_get,
+)
+
+
+class CarambaTVIE(InfoExtractor):
+    _VALID_URL = r'(?:carambatv:|https?://video1\.carambatv\.ru/v/)(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://video1.carambatv.ru/v/191910501',
+        'md5': '2f4a81b7cfd5ab866ee2d7270cb34a2a',
+        'info_dict': {
+            'id': '191910501',
+            'ext': 'mp4',
+            'title': '[BadComedian] - Разборка в Маниле (Абсолютный обзор)',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'duration': 2678.31,
+        },
+    }, {
+        'url': 'carambatv:191910501',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        video = self._download_json(
+            'http://video1.carambatv.ru/v/%s/videoinfo.js' % video_id,
+            video_id)
+
+        title = video['title']
+
+        base_url = video.get('video') or 'http://video1.carambatv.ru/v/%s/' % video_id
+
+        formats = [{
+            'url': base_url + f['fn'],
+            'height': int_or_none(f.get('height')),
+            'format_id': '%sp' % f['height'] if f.get('height') else None,
+        } for f in video['qualities'] if f.get('fn')]
+        self._sort_formats(formats)
+
+        thumbnail = video.get('splash')
+        duration = float_or_none(try_get(
+            video, lambda x: x['annotations'][0]['end_time'], compat_str))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }
+
+
+class CarambaTVPageIE(InfoExtractor):
+    _VALID_URL = r'https?://carambatv\.ru/(?:[^/]+/)+(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'http://carambatv.ru/movie/bad-comedian/razborka-v-manile/',
+        'md5': '',
+        'info_dict': {
+            'id': '191910501',
+            'ext': 'mp4',
+            'title': '[BadComedian] - Разборка в Маниле (Абсолютный обзор)',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 2678.31,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        video_url = self._og_search_property('video:iframe', webpage, default=None)
+
+        if not video_url:
+            video_id = self._search_regex(
+                r'(?:video_id|crmb_vuid)\s*[:=]\s*["\']?(\d+)',
+                webpage, 'video id')
+            video_url = 'carambatv:%s' % video_id
+
+        return self.url_result(video_url, CarambaTVIE.ie_key())

youtube_dl/extractor/cbs.py

@@ -1,17 +1,13 @@
 from __future__ import unicode_literals
 
-import re
+from .theplatform import ThePlatformFeedIE
-
-from .theplatform import ThePlatformIE
 from ..utils import (
-    xpath_text,
-    xpath_element,
     int_or_none,
     find_xpath_attr,
 )
 
 
-class CBSBaseIE(ThePlatformIE):
+class CBSBaseIE(ThePlatformFeedIE):
     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
         closed_caption_e = find_xpath_attr(smil, self._xpath_ns('.//param', namespace), 'name', 'ClosedCaptionURL')
         return {
@@ -21,9 +17,22 @@ class CBSBaseIE(ThePlatformIE):
             }]
         } if closed_caption_e is not None and closed_caption_e.attrib.get('value') else []
 
+    def _extract_video_info(self, filter_query, video_id):
+        return self._extract_feed_info(
+            'dJ5BDC', 'VxxJg8Ymh8sE', filter_query, video_id, lambda entry: {
+                'series': entry.get('cbs$SeriesTitle'),
+                'season_number': int_or_none(entry.get('cbs$SeasonNumber')),
+                'episode': entry.get('cbs$EpisodeTitle'),
+                'episode_number': int_or_none(entry.get('cbs$EpisodeNumber')),
+            }, {
+                'StreamPack': {
+                    'manifest': 'm3u',
+                }
+            })
+
 
 class CBSIE(CBSBaseIE):
-    _VALID_URL = r'(?:cbs:(?P<content_id>\w+)|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/(?:video|artist)|colbertlateshow\.com/(?:video|podcasts))/[^/]+/(?P<display_id>[^/]+))'
+    _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
 
     _TESTS = [{
         'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
@@ -38,25 +47,7 @@ class CBSIE(CBSBaseIE):
             'upload_date': '20131127',
             'uploader': 'CBSI-NEW',
         },
-        'params': {
+        'expected_warnings': ['Failed to download m3u8 information'],
-            # rtmp download
-            'skip_download': True,
-        },
-        '_skip': 'Blocked outside the US',
-    }, {
-        'url': 'http://www.cbs.com/shows/liveonletterman/artist/221752/st-vincent/',
-        'info_dict': {
-            'id': 'WWF_5KqY3PK1',
-            'display_id': 'st-vincent',
-            'ext': 'flv',
-            'title': 'Live on Letterman - St. Vincent',
-            'description': 'Live On Letterman: St. Vincent in concert from New York\'s Ed Sullivan Theater on Tuesday, July 16, 2014.',
-            'duration': 3221,
-        },
-        'params': {
-            # rtmp download
-            'skip_download': True,
-        },
         '_skip': 'Blocked outside the US',
     }, {
         'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
@@ -68,44 +59,5 @@ class CBSIE(CBSBaseIE):
     TP_RELEASE_URL_TEMPLATE = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true'
 
     def _real_extract(self, url):
-        content_id, display_id = re.match(self._VALID_URL, url).groups()
+        content_id = self._match_id(url)
-        if not content_id:
+        return self._extract_video_info('byGuid=%s' % content_id, content_id)
-            webpage = self._download_webpage(url, display_id)
-            content_id = self._search_regex(
-                [r"video\.settings\.content_id\s*=\s*'([^']+)';", r"cbsplayer\.contentId\s*=\s*'([^']+)';"],
-                webpage, 'content id')
-        items_data = self._download_xml(
-            'http://can.cbs.com/thunder/player/videoPlayerService.php',
-            content_id, query={'partner': 'cbs', 'contentId': content_id})
-        video_data = xpath_element(items_data, './/item')
-        title = xpath_text(video_data, 'videoTitle', 'title', True)
-
-        subtitles = {}
-        formats = []
-        for item in items_data.findall('.//item'):
-            pid = xpath_text(item, 'pid')
-            if not pid:
-                continue
-            tp_release_url = self.TP_RELEASE_URL_TEMPLATE % pid
-            if '.m3u8' in xpath_text(item, 'contentUrl', default=''):
-                tp_release_url += '&manifest=m3u'
-            tp_formats, tp_subtitles = self._extract_theplatform_smil(
-                tp_release_url, content_id, 'Downloading %s SMIL data' % pid)
-            formats.extend(tp_formats)
-            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
-        self._sort_formats(formats)
-
-        info = self.get_metadata('dJ5BDC/media/guid/2198311517/%s' % content_id, content_id)
-        info.update({
-            'id': content_id,
-            'display_id': display_id,
-            'title': title,
-            'series': xpath_text(video_data, 'seriesTitle'),
-            'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
-            'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
-            'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
-            'thumbnail': xpath_text(video_data, 'previewImageURL'),
-            'formats': formats,
-            'subtitles': subtitles,
-        })
-        return info

youtube_dl/extractor/cbsinteractive.py

@@ -80,9 +80,6 @@ class CBSInteractiveIE(ThePlatformIE):
 
         media_guid_path = 'media/guid/%d/%s' % (self.MPX_ACCOUNTS[site], vdata['mpxRefId'])
         formats, subtitles = [], {}
-        if site == 'cnet':
-            formats, subtitles = self._extract_theplatform_smil(
-                self.TP_RELEASE_URL_TEMPLATE % media_guid_path, video_id)
         for (fkey, vid) in vdata['files'].items():
             if fkey == 'hls_phone' and 'hls_tablet' in vdata['files']:
                 continue
@@ -94,7 +91,7 @@ class CBSInteractiveIE(ThePlatformIE):
             subtitles = self._merge_subtitles(subtitles, tp_subtitles)
         self._sort_formats(formats)
 
-        info = self.get_metadata('kYEXFC/%s' % media_guid_path, video_id)
+        info = self._extract_theplatform_metadata('kYEXFC/%s' % media_guid_path, video_id)
         info.update({
             'id': video_id,
             'display_id': display_id,

youtube_dl/extractor/cbsnews.py

@@ -30,9 +30,12 @@ class CBSNewsIE(CBSBaseIE):
         {
             'url': 'http://www.cbsnews.com/videos/fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack/',
             'info_dict': {
-                'id': 'fort-hood-shooting-army-downplays-mental-illness-as-cause-of-attack',
+                'id': 'SNJBOYzXiWBOvaLsdzwH8fmtP1SCd91Y',
                 'ext': 'mp4',
                 'title': 'Fort Hood shooting: Army downplays mental illness as cause of attack',
+                'description': 'md5:4a6983e480542d8b333a947bfc64ddc7',
+                'upload_date': '19700101',
+                'uploader': 'CBSI-NEW',
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'duration': 205,
                 'subtitles': {
@@ -58,30 +61,8 @@ class CBSNewsIE(CBSBaseIE):
             webpage, 'video JSON info'), video_id)
 
         item = video_info['item'] if 'item' in video_info else video_info
-        title = item.get('articleTitle') or item.get('hed')
+        guid = item['mpxRefId']
-        duration = item.get('duration')
+        return self._extract_video_info('byGuid=%s' % guid, guid)
-        thumbnail = item.get('mediaImage') or item.get('thumbnail')
-
-        subtitles = {}
-        formats = []
-        for format_id in ['RtmpMobileLow', 'RtmpMobileHigh', 'Hls', 'RtmpDesktop']:
-            pid = item.get('media' + format_id)
-            if not pid:
-                continue
-            release_url = 'http://link.theplatform.com/s/dJ5BDC/%s?mbr=true' % pid
-            tp_formats, tp_subtitles = self._extract_theplatform_smil(release_url, video_id, 'Downloading %s SMIL data' % pid)
-            formats.extend(tp_formats)
-            subtitles = self._merge_subtitles(subtitles, tp_subtitles)
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'title': title,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats,
-            'subtitles': subtitles,
-        }
 
 
 class CBSNewsLiveVideoIE(InfoExtractor):

youtube_dl/extractor/cbssports.py

@@ -1,30 +1,28 @@
 from __future__ import unicode_literals
 
-import re
+from .cbs import CBSBaseIE
-
-from .common import InfoExtractor
 
 
-class CBSSportsIE(InfoExtractor):
+class CBSSportsIE(CBSBaseIE):
-    _VALID_URL = r'https?://www\.cbssports\.com/video/player/(?P<section>[^/]+)/(?P<id>[^/]+)'
+    _VALID_URL = r'https?://www\.cbssports\.com/video/player/[^/]+/(?P<id>\d+)'
 
-    _TEST = {
+    _TESTS = [{
-        'url': 'http://www.cbssports.com/video/player/tennis/318462531970/0/us-open-flashbacks-1990s',
+        'url': 'http://www.cbssports.com/video/player/videos/708337219968/0/ben-simmons-the-next-lebron?-not-so-fast',
         'info_dict': {
-            'id': '_d5_GbO8p1sT',
+            'id': '708337219968',
-            'ext': 'flv',
+            'ext': 'mp4',
-            'title': 'US Open flashbacks: 1990s',
+            'title': 'Ben Simmons the next LeBron? Not so fast',
-            'description': 'Bill Macatee relives the best moments in US Open history from the 1990s.',
+            'description': 'md5:854294f627921baba1f4b9a990d87197',
+            'timestamp': 1466293740,
+            'upload_date': '20160618',
+            'uploader': 'CBSI-NEW',
         },
-    }
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        }
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        section = mobj.group('section')
+        return self._extract_video_info('byId=%s' % video_id, video_id)
-        video_id = mobj.group('id')
-        all_videos = self._download_json(
-            'http://www.cbssports.com/data/video/player/getVideos/%s?as=json' % section,
-            video_id)
-        # The json file contains the info of all the videos in the section
-        video_info = next(v for v in all_videos if v['pcid'] == video_id)
-        return self.url_result('theplatform:%s' % video_info['pid'], 'ThePlatform')

youtube_dl/extractor/cda.py

@@ -58,7 +58,8 @@ class CDAIE(InfoExtractor):
         def extract_format(page, version):
             unpacked = decode_packed_codes(page)
             format_url = self._search_regex(
-                r"url:\\'(.+?)\\'", unpacked, '%s url' % version, fatal=False)
+                r"(?:file|url)\s*:\s*(\\?[\"'])(?P<url>http.+?)\1", unpacked,
+                '%s url' % version, fatal=False, group='url')
             if not format_url:
                 return
             f = {
@@ -75,7 +76,8 @@ class CDAIE(InfoExtractor):
             info_dict['formats'].append(f)
             if not info_dict['duration']:
                 info_dict['duration'] = parse_duration(self._search_regex(
-                    r"duration:\\'(.+?)\\'", unpacked, 'duration', fatal=False))
+                    r"duration\s*:\s*(\\?[\"'])(?P<duration>.+?)\1",
+                    unpacked, 'duration', fatal=False, group='duration'))
 
         extract_format(webpage, 'default')
 

youtube_dl/extractor/channel9.py

@@ -20,54 +20,64 @@ class Channel9IE(InfoExtractor):
     '''
     IE_DESC = 'Channel 9'
     IE_NAME = 'channel9'
-    _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+)/?'
+    _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+?)(?P<rss>/RSS)?/?(?:[?#&]|$)'
 
-    _TESTS = [
+    _TESTS = [{
-        {
+        'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
-            'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
+        'md5': 'bbd75296ba47916b754e73c3a4bbdf10',
-            'md5': 'bbd75296ba47916b754e73c3a4bbdf10',
+        'info_dict': {
-            'info_dict': {
+            'id': 'Events/TechEd/Australia/2013/KOS002',
-                'id': 'Events/TechEd/Australia/2013/KOS002',
+            'ext': 'mp4',
-                'ext': 'mp4',
+            'title': 'Developer Kick-Off Session: Stuff We Love',
-                'title': 'Developer Kick-Off Session: Stuff We Love',
+            'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
-                'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
+            'duration': 4576,
-                'duration': 4576,
+            'thumbnail': 're:http://.*\.jpg',
-                'thumbnail': 're:http://.*\.jpg',
+            'session_code': 'KOS002',
-                'session_code': 'KOS002',
+            'session_day': 'Day 1',
-                'session_day': 'Day 1',
+            'session_room': 'Arena 1A',
-                'session_room': 'Arena 1A',
+            'session_speakers': ['Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug',
-                'session_speakers': ['Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug', 'Mads Kristensen'],
+                                 'Mads Kristensen'],
-            },
         },
-        {
+    }, {
-            'url': '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',
-            'md5': 'b43ee4529d111bc37ba7ee4f34813e68',
+        'md5': 'b43ee4529d111bc37ba7ee4f34813e68',
-            'info_dict': {
+        'info_dict': {
-                'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing',
+            'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing',
-                'ext': 'mp4',
+            'ext': 'mp4',
-                'title': 'Self-service BI with Power BI - nuclear testing',
+            'title': 'Self-service BI with Power BI - nuclear testing',
-                'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
+            'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
-                'duration': 1540,
+            'duration': 1540,
-                'thumbnail': 're:http://.*\.jpg',
+            'thumbnail': 're:http://.*\.jpg',
-                'authors': ['Mike Wilmot'],
+            'authors': ['Mike Wilmot'],
-            },
         },
-        {
+    }, {
-            # low quality mp4 is best
+        # low quality mp4 is best
-            'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+        'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
-            'info_dict': {
+        'info_dict': {
-                'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+            'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
-                'ext': 'mp4',
+            'ext': 'mp4',
-                'title': 'Ranges for the Standard Library',
+            'title': 'Ranges for the Standard Library',
-                'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d',
+            'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d',
-                'duration': 5646,
+            'duration': 5646,
-                'thumbnail': 're:http://.*\.jpg',
+            'thumbnail': 're:http://.*\.jpg',
-            },
+        },
-            'params': {
+        'params': {
-                'skip_download': True,
+            'skip_download': True,
-            },
+        },
-        }
+    }, {
-    ]
+        'url': 'https://channel9.msdn.com/Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b/RSS',
+        'info_dict': {
+            'id': 'Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b',
+            'title': 'Channel 9',
+        },
+        'playlist_count': 2,
+    }, {
+        'url': 'https://channel9.msdn.com/Events/DEVintersection/DEVintersection-2016/RSS',
+        'only_matching': True,
+    }, {
+        'url': 'https://channel9.msdn.com/Events/Speakers/scott-hanselman/RSS?UrlSafeName=scott-hanselman',
+        'only_matching': True,
+    }]
 
     _RSS_URL = 'http://channel9.msdn.com/%s/RSS'
 
@@ -254,22 +264,30 @@ class Channel9IE(InfoExtractor):
 
         return self.playlist_result(contents)
 
-    def _extract_list(self, content_path):
+    def _extract_list(self, video_id, rss_url=None):
-        rss = self._download_xml(self._RSS_URL % content_path, content_path, 'Downloading RSS')
+        if not rss_url:
+            rss_url = self._RSS_URL % video_id
+        rss = self._download_xml(rss_url, video_id, '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
-        return self.playlist_result(entries, content_path, title_text)
+        return self.playlist_result(entries, video_id, title_text)
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         content_path = mobj.group('contentpath')
+        rss = mobj.group('rss')
 
-        webpage = self._download_webpage(url, content_path, 'Downloading web page')
+        if rss:
+            return self._extract_list(content_path, url)
 
-        page_type_m = re.search(r'<meta name="WT.entryid" content="(?P<pagetype>[^:]+)[^"]+"/>', webpage)
+        webpage = self._download_webpage(
-        if page_type_m is not None:
+            url, content_path, 'Downloading web page')
-            page_type = page_type_m.group('pagetype')
+
+        page_type = self._search_regex(
+            r'<meta[^>]+name=(["\'])WT\.entryid\1[^>]+content=(["\'])(?P<pagetype>[^:]+).+?\2',
+            webpage, 'page type', default=None, group='pagetype')
+        if page_type:
             if page_type == 'Entry':      # Any 'item'-like page, may contain downloadable content
                 return self._extract_entry_item(webpage, content_path)
             elif page_type == 'Session':  # Event session page, may contain downloadable content
@@ -278,6 +296,5 @@ class Channel9IE(InfoExtractor):
                 return self._extract_list(content_path)
             else:
                 raise ExtractorError('Unexpected WT.entryid %s' % page_type, expected=True)
-
         else:  # Assuming list
             return self._extract_list(content_path)

youtube_dl/extractor/cliprs.py

@@ -1,16 +1,10 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
+from .onet import OnetBaseIE
-from ..utils import (
-    ExtractorError,
-    float_or_none,
-    int_or_none,
-    parse_iso8601,
-)
 
 
-class ClipRsIE(InfoExtractor):
+class ClipRsIE(OnetBaseIE):
     _VALID_URL = r'https?://(?:www\.)?clip\.rs/(?P<id>[^/]+)/\d+'
     _TEST = {
         'url': 'http://www.clip.rs/premijera-frajle-predstavljaju-novi-spot-za-pesmu-moli-me-moli/3732',
@@ -27,64 +21,13 @@ class ClipRsIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        display_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(url, display_id)
 
-        video_id = self._search_regex(
+        mvp_id = self._search_mvp_id(webpage)
-            r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
 
-        response = self._download_json(
+        info_dict = self._extract_from_id(mvp_id, webpage)
-            'http://qi.ckm.onetapi.pl/', video_id,
+        info_dict['display_id'] = display_id
-            query={
-                'body[id]': video_id,
-                'body[jsonrpc]': '2.0',
-                'body[method]': 'get_asset_detail',
-                'body[params][ID_Publikacji]': video_id,
-                'body[params][Service]': 'www.onet.pl',
-                'content-type': 'application/jsonp',
-                'x-onet-app': 'player.front.onetapi.pl',
-            })
 
-        error = response.get('error')
+        return info_dict
-        if error:
-            raise ExtractorError(
-                '%s said: %s' % (self.IE_NAME, error['message']), expected=True)
-
-        video = response['result'].get('0')
-
-        formats = []
-        for _, formats_dict in video['formats'].items():
-            if not isinstance(formats_dict, dict):
-                continue
-            for format_id, format_list in formats_dict.items():
-                if not isinstance(format_list, list):
-                    continue
-                for f in format_list:
-                    if not f.get('url'):
-                        continue
-                    formats.append({
-                        'url': f['url'],
-                        'format_id': format_id,
-                        'height': int_or_none(f.get('vertical_resolution')),
-                        'width': int_or_none(f.get('horizontal_resolution')),
-                        'abr': float_or_none(f.get('audio_bitrate')),
-                        'vbr': float_or_none(f.get('video_bitrate')),
-                    })
-        self._sort_formats(formats)
-
-        meta = video.get('meta', {})
-
-        title = self._og_search_title(webpage, default=None) or meta['title']
-        description = self._og_search_description(webpage, default=None) or meta.get('description')
-        duration = meta.get('length') or meta.get('lenght')
-        timestamp = parse_iso8601(meta.get('addDate'), ' ')
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'duration': duration,
-            'timestamp': timestamp,
-            'formats': formats,
-        }

youtube_dl/extractor/closertotruth.py

@@ -0,0 +1,92 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+
+
+class CloserToTruthIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?closertotruth\.com/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'http://closertotruth.com/series/solutions-the-mind-body-problem#video-3688',
+        'info_dict': {
+            'id': '0_zof1ktre',
+            'display_id': 'solutions-the-mind-body-problem',
+            'ext': 'mov',
+            'title': 'Solutions to the Mind-Body Problem?',
+            'upload_date': '20140221',
+            'timestamp': 1392956007,
+            'uploader_id': 'CTTXML'
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://closertotruth.com/episodes/how-do-brains-work',
+        'info_dict': {
+            'id': '0_iuxai6g6',
+            'display_id': 'how-do-brains-work',
+            'ext': 'mov',
+            'title': 'How do Brains Work?',
+            'upload_date': '20140221',
+            'timestamp': 1392956024,
+            'uploader_id': 'CTTXML'
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://closertotruth.com/interviews/1725',
+        'info_dict': {
+            'id': '1725',
+            'title': 'AyaFr-002',
+        },
+        'playlist_mincount': 2,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        partner_id = self._search_regex(
+            r'<script[^>]+src=["\'].*?\b(?:partner_id|p)/(\d+)',
+            webpage, 'kaltura partner_id')
+
+        title = self._search_regex(
+            r'<title>(.+?)\s*\|\s*.+?</title>', webpage, 'video title')
+
+        select = self._search_regex(
+            r'(?s)<select[^>]+id="select-version"[^>]*>(.+?)</select>',
+            webpage, 'select version', default=None)
+        if select:
+            entry_ids = set()
+            entries = []
+            for mobj in re.finditer(
+                    r'<option[^>]+value=(["\'])(?P<id>[0-9a-z_]+)(?:#.+?)?\1[^>]*>(?P<title>[^<]+)',
+                    webpage):
+                entry_id = mobj.group('id')
+                if entry_id in entry_ids:
+                    continue
+                entry_ids.add(entry_id)
+                entries.append({
+                    '_type': 'url_transparent',
+                    'url': 'kaltura:%s:%s' % (partner_id, entry_id),
+                    'ie_key': 'Kaltura',
+                    'title': mobj.group('title'),
+                })
+            if entries:
+                return self.playlist_result(entries, display_id, title)
+
+        entry_id = self._search_regex(
+            r'<a[^>]+id=(["\'])embed-kaltura\1[^>]+data-kaltura=(["\'])(?P<id>[0-9a-z_]+)\2',
+            webpage, 'kaltura entry_id', group='id')
+
+        return {
+            '_type': 'url_transparent',
+            'display_id': display_id,
+            'url': 'kaltura:%s:%s' % (partner_id, entry_id),
+            'ie_key': 'Kaltura',
+            'title': title
+        }

youtube_dl/extractor/common.py

@@ -44,7 +44,9 @@ from ..utils import (
     sanitized_Request,
     unescapeHTML,
     unified_strdate,
+    unified_timestamp,
     url_basename,
+    xpath_element,
     xpath_text,
     xpath_with_ns,
     determine_protocol,
@@ -52,6 +54,9 @@ from ..utils import (
     mimetype2ext,
     update_Request,
     update_url_query,
+    parse_m3u8_attributes,
+    extract_attributes,
+    parse_codecs,
 )
 
 
@@ -159,6 +164,7 @@ class InfoExtractor(object):
                         * "height" (optional, int)
                         * "resolution" (optional, string "{width}x{height"},
                                         deprecated)
+                        * "filesize" (optional, int)
     thumbnail:      Full URL to a video thumbnail image.
     description:    Full video description.
     uploader:       Full name of the video uploader.
@@ -747,10 +753,12 @@ class InfoExtractor(object):
         return self._og_search_property('url', html, **kargs)
 
     def _html_search_meta(self, name, html, display_name=None, fatal=False, **kwargs):
+        if not isinstance(name, (list, tuple)):
+            name = [name]
         if display_name is None:
-            display_name = name
+            display_name = name[0]
         return self._html_search_regex(
-            self._meta_regex(name),
+            [self._meta_regex(n) for n in name],
             html, display_name, fatal=fatal, group='content', **kwargs)
 
     def _dc_search_uploader(self, html):
@@ -799,15 +807,17 @@ class InfoExtractor(object):
         return self._html_search_meta('twitter:player', html,
                                       'twitter card player')
 
-    def _search_json_ld(self, html, video_id, **kwargs):
+    def _search_json_ld(self, html, video_id, expected_type=None, **kwargs):
         json_ld = self._search_regex(
             r'(?s)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>',
             html, 'JSON-LD', group='json_ld', **kwargs)
         if not json_ld:
             return {}
-        return self._json_ld(json_ld, video_id, fatal=kwargs.get('fatal', True))
+        return self._json_ld(
+            json_ld, video_id, fatal=kwargs.get('fatal', True),
+            expected_type=expected_type)
 
-    def _json_ld(self, json_ld, video_id, fatal=True):
+    def _json_ld(self, json_ld, video_id, fatal=True, expected_type=None):
         if isinstance(json_ld, compat_str):
             json_ld = self._parse_json(json_ld, video_id, fatal=fatal)
         if not json_ld:
@@ -815,6 +825,8 @@ class InfoExtractor(object):
         info = {}
         if json_ld.get('@context') == 'http://schema.org':
             item_type = json_ld.get('@type')
+            if expected_type is not None and expected_type != item_type:
+                return info
             if item_type == 'TVEpisode':
                 info.update({
                     'episode': unescapeHTML(json_ld.get('name')),
@@ -833,6 +845,19 @@ class InfoExtractor(object):
                     'title': unescapeHTML(json_ld.get('headline')),
                     'description': unescapeHTML(json_ld.get('articleBody')),
                 })
+            elif item_type == 'VideoObject':
+                info.update({
+                    'url': json_ld.get('contentUrl'),
+                    'title': unescapeHTML(json_ld.get('name')),
+                    'description': unescapeHTML(json_ld.get('description')),
+                    'thumbnail': json_ld.get('thumbnailUrl'),
+                    'duration': parse_duration(json_ld.get('duration')),
+                    'timestamp': unified_timestamp(json_ld.get('uploadDate')),
+                    'filesize': float_or_none(json_ld.get('contentSize')),
+                    'tbr': int_or_none(json_ld.get('bitrate')),
+                    'width': int_or_none(json_ld.get('width')),
+                    'height': int_or_none(json_ld.get('height')),
+                })
         return dict((k, v) for k, v in info.items() if v is not None)
 
     @staticmethod
@@ -874,7 +899,11 @@ class InfoExtractor(object):
                 f['ext'] = determine_ext(f['url'])
 
             if isinstance(field_preference, (list, tuple)):
-                return tuple(f.get(field) if f.get(field) is not None else -1 for field in field_preference)
+                return tuple(
+                    f.get(field)
+                    if f.get(field) is not None
+                    else ('' if field == 'format_id' else -1)
+                    for field in field_preference)
 
             preference = f.get('preference')
             if preference is None:
@@ -1030,7 +1059,7 @@ class InfoExtractor(object):
         if base_url:
             base_url = base_url.strip()
 
-        bootstrap_info = xpath_text(
+        bootstrap_info = xpath_element(
             manifest, ['{http://ns.adobe.com/f4m/1.0}bootstrapInfo', '{http://ns.adobe.com/f4m/2.0}bootstrapInfo'],
             'bootstrap info', default=None)
 
@@ -1085,7 +1114,7 @@ class InfoExtractor(object):
             formats.append({
                 'format_id': format_id,
                 'url': manifest_url,
-                'ext': 'flv' if bootstrap_info else None,
+                'ext': 'flv' if bootstrap_info is not None else None,
                 'tbr': tbr,
                 'width': width,
                 'height': height,
@@ -1149,23 +1178,11 @@ class InfoExtractor(object):
             }]
         last_info = None
         last_media = None
-        kv_rex = re.compile(
-            r'(?P<key>[a-zA-Z_-]+)=(?P<val>"[^"]+"|[^",]+)(?:,|$)')
         for line in m3u8_doc.splitlines():
             if line.startswith('#EXT-X-STREAM-INF:'):
-                last_info = {}
+                last_info = parse_m3u8_attributes(line)
-                for m in kv_rex.finditer(line):
-                    v = m.group('val')
-                    if v.startswith('"'):
-                        v = v[1:-1]
-                    last_info[m.group('key')] = v
             elif line.startswith('#EXT-X-MEDIA:'):
-                last_media = {}
+                last_media = parse_m3u8_attributes(line)
-                for m in kv_rex.finditer(line):
-                    v = m.group('val')
-                    if v.startswith('"'):
-                        v = v[1:-1]
-                    last_media[m.group('key')] = v
             elif line.startswith('#') or not line.strip():
                 continue
             else:
@@ -1190,6 +1207,7 @@ class InfoExtractor(object):
                     'url': format_url(line.strip()),
                     'tbr': tbr,
                     'ext': ext,
+                    'fps': float_or_none(last_info.get('FRAME-RATE')),
                     'protocol': entry_protocol,
                     'preference': preference,
                 }
@@ -1198,24 +1216,17 @@ class InfoExtractor(object):
                     width_str, height_str = resolution.split('x')
                     f['width'] = int(width_str)
                     f['height'] = int(height_str)
-                codecs = last_info.get('CODECS')
+                # Unified Streaming Platform
-                if codecs:
+                mobj = re.search(
-                    vcodec, acodec = [None] * 2
+                    r'audio.*?(?:%3D|=)(\d+)(?:-video.*?(?:%3D|=)(\d+))?', f['url'])
-                    va_codecs = codecs.split(',')
+                if mobj:
-                    if len(va_codecs) == 1:
+                    abr, vbr = mobj.groups()
-                        # Audio only entries usually come with single codec and
+                    abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000)
-                        # no resolution. For more robustness we also check it to
-                        # be mp4 audio.
-                        if not resolution and va_codecs[0].startswith('mp4a'):
-                            vcodec, acodec = 'none', va_codecs[0]
-                        else:
-                            vcodec = va_codecs[0]
-                    else:
-                        vcodec, acodec = va_codecs[:2]
                     f.update({
-                        'acodec': acodec,
+                        'vbr': vbr,
-                        'vcodec': vcodec,
+                        'abr': abr,
                     })
+                f.update(parse_codecs(last_info.get('CODECS')))
                 if last_media is not None:
                     f['m3u8_media'] = last_media
                     last_media = None
@@ -1620,6 +1631,62 @@ class InfoExtractor(object):
                         self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type)
         return formats
 
+    def _parse_html5_media_entries(self, base_url, webpage):
+        def absolute_url(video_url):
+            return compat_urlparse.urljoin(base_url, video_url)
+
+        def parse_content_type(content_type):
+            if not content_type:
+                return {}
+            ctr = re.search(r'(?P<mimetype>[^/]+/[^;]+)(?:;\s*codecs="?(?P<codecs>[^"]+))?', content_type)
+            if ctr:
+                mimetype, codecs = ctr.groups()
+                f = parse_codecs(codecs)
+                f['ext'] = mimetype2ext(mimetype)
+                return f
+            return {}
+
+        entries = []
+        for media_tag, media_type, media_content in re.findall(r'(?s)(<(?P<tag>video|audio)[^>]*>)(.*?)</(?P=tag)>', webpage):
+            media_info = {
+                'formats': [],
+                'subtitles': {},
+            }
+            media_attributes = extract_attributes(media_tag)
+            src = media_attributes.get('src')
+            if src:
+                media_info['formats'].append({
+                    'url': absolute_url(src),
+                    'vcodec': 'none' if media_type == 'audio' else None,
+                })
+            media_info['thumbnail'] = media_attributes.get('poster')
+            if media_content:
+                for source_tag in re.findall(r'<source[^>]+>', media_content):
+                    source_attributes = extract_attributes(source_tag)
+                    src = source_attributes.get('src')
+                    if not src:
+                        continue
+                    f = parse_content_type(source_attributes.get('type'))
+                    f.update({
+                        'url': absolute_url(src),
+                        'vcodec': 'none' if media_type == 'audio' else None,
+                    })
+                    media_info['formats'].append(f)
+                for track_tag in re.findall(r'<track[^>]+>', media_content):
+                    track_attributes = extract_attributes(track_tag)
+                    kind = track_attributes.get('kind')
+                    if not kind or kind == 'subtitles':
+                        src = track_attributes.get('src')
+                        if not src:
+                            continue
+                        lang = track_attributes.get('srclang') or track_attributes.get('lang') or track_attributes.get('label')
+                        media_info['subtitles'].setdefault(lang, []).append({
+                            'url': absolute_url(src),
+                        })
+            if media_info['formats']:
+                entries.append(media_info)
+        return entries
+
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()
@@ -1733,6 +1800,13 @@ class InfoExtractor(object):
     def _mark_watched(self, *args, **kwargs):
         raise NotImplementedError('This method must be implemented by subclasses')
 
+    def geo_verification_headers(self):
+        headers = {}
+        geo_verification_proxy = self._downloader.params.get('geo_verification_proxy')
+        if geo_verification_proxy:
+            headers['Ytdl-request-proxy'] = geo_verification_proxy
+        return headers
+
 
 class SearchInfoExtractor(InfoExtractor):
     """

youtube_dl/extractor/ctv.py

@@ -0,0 +1,30 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class CTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?ctv\.ca/video/player\?vid=(?P<id>[0-9.]+)'
+    _TESTS = [{
+        'url': 'http://www.ctv.ca/video/player?vid=706966',
+        'md5': 'ff2ebbeae0aa2dcc32a830c3fd69b7b0',
+        'info_dict': {
+            'id': '706966',
+            'ext': 'mp4',
+            'title': 'Larry Day and Richard Jutras on the TIFF red carpet of \'Stonewall\'',
+            'description': 'etalk catches up with Larry Day and Richard Jutras on the TIFF red carpet of "Stonewall”.',
+            'upload_date': '20150919',
+            'timestamp': 1442624700,
+        },
+        'expected_warnings': ['HTTP Error 404'],
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        return {
+            '_type': 'url_transparent',
+            'id': video_id,
+            'url': '9c9media:ctv_web:%s' % video_id,
+            'ie_key': 'NineCNineMedia',
+        }

youtube_dl/extractor/ctvnews.py

@@ -0,0 +1,65 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import orderedSet
+
+
+class CTVNewsIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?ctvnews\.ca/(?:video\?(?:clip|playlist|bin)Id=|.*?)(?P<id>[0-9.]+)'
+    _TESTS = [{
+        'url': 'http://www.ctvnews.ca/video?clipId=901995',
+        'md5': '10deb320dc0ccb8d01d34d12fc2ea672',
+        'info_dict': {
+            'id': '901995',
+            'ext': 'mp4',
+            'title': 'Extended: \'That person cannot be me\' Johnson says',
+            'description': 'md5:958dd3b4f5bbbf0ed4d045c790d89285',
+            'timestamp': 1467286284,
+            'upload_date': '20160630',
+        }
+    }, {
+        'url': 'http://www.ctvnews.ca/video?playlistId=1.2966224',
+        'info_dict':
+        {
+            'id': '1.2966224',
+        },
+        'playlist_mincount': 19,
+    }, {
+        'url': 'http://www.ctvnews.ca/video?binId=1.2876780',
+        'info_dict':
+        {
+            'id': '1.2876780',
+        },
+        'playlist_mincount': 100,
+    }, {
+        'url': 'http://www.ctvnews.ca/1.810401',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.ctvnews.ca/canadiens-send-p-k-subban-to-nashville-in-blockbuster-trade-1.2967231',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        page_id = self._match_id(url)
+
+        def ninecninemedia_url_result(clip_id):
+            return {
+                '_type': 'url_transparent',
+                'id': clip_id,
+                'url': '9c9media:ctvnews_web:%s' % clip_id,
+                'ie_key': 'NineCNineMedia',
+            }
+
+        if page_id.isdigit():
+            return ninecninemedia_url_result(page_id)
+        else:
+            webpage = self._download_webpage('http://www.ctvnews.ca/%s' % page_id, page_id, query={
+                'ot': 'example.AjaxPageLayout.ot',
+                'maxItemsPerPage': 1000000,
+            })
+            entries = [ninecninemedia_url_result(clip_id) for clip_id in orderedSet(
+                re.findall(r'clip\.id\s*=\s*(\d+);', webpage))]
+            return self.playlist_result(entries, page_id)

youtube_dl/extractor/dailymotion.py

@@ -16,6 +16,7 @@ from ..utils import (
     sanitized_Request,
     str_to_int,
     unescapeHTML,
+    mimetype2ext,
 )
 
 
@@ -111,6 +112,13 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
         }
     ]
 
+    @staticmethod
+    def _extract_urls(webpage):
+        # Look for embedded Dailymotion player
+        matches = re.findall(
+            r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
+        return list(map(lambda m: unescapeHTML(m[1]), matches))
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
@@ -153,18 +161,19 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                     type_ = media.get('type')
                     if type_ == 'application/vnd.lumberjack.manifest':
                         continue
-                    ext = determine_ext(media_url)
+                    ext = mimetype2ext(type_) or determine_ext(media_url)
-                    if type_ == 'application/x-mpegURL' or ext == 'm3u8':
+                    if ext == 'm3u8':
                         formats.extend(self._extract_m3u8_formats(
                             media_url, video_id, 'mp4', preference=-1,
                             m3u8_id='hls', fatal=False))
-                    elif type_ == 'application/f4m' or ext == 'f4m':
+                    elif ext == 'f4m':
                         formats.extend(self._extract_f4m_formats(
                             media_url, video_id, preference=-1, f4m_id='hds', fatal=False))
                     else:
                         f = {
                             'url': media_url,
                             'format_id': 'http-%s' % quality,
+                            'ext': ext,
                         }
                         m = re.search(r'H264-(?P<width>\d+)x(?P<height>\d+)', media_url)
                         if m:

youtube_dl/extractor/daum.py

@@ -66,22 +66,32 @@ class DaumIE(InfoExtractor):
             'view_count': int,
             'comment_count': int,
         },
+    }, {
+        # Requires dte_type=WEB (#9972)
+        'url': 'http://tvpot.daum.net/v/s3794Uf1NZeZ1qMpGpeqeRU',
+        'md5': 'a8917742069a4dd442516b86e7d66529',
+        'info_dict': {
+            'id': 's3794Uf1NZeZ1qMpGpeqeRU',
+            'ext': 'mp4',
+            'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny) [쇼! 음악중심] 508회 20160611',
+            'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\n\n[쇼! 음악중심] 20160611, 507회',
+            'upload_date': '20160611',
+        },
     }]
 
     def _real_extract(self, url):
         video_id = compat_urllib_parse_unquote(self._match_id(url))
-        query = compat_urllib_parse_urlencode({'vid': video_id})
         movie_data = self._download_json(
-            'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json?' + query,
+            'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json',
-            video_id, 'Downloading video formats info')
+            video_id, 'Downloading video formats info', query={'vid': video_id, 'dte_type': 'WEB'})
 
         # For urls like http://m.tvpot.daum.net/v/65139429, where the video_id is really a clipid
         if not movie_data.get('output_list', {}).get('output_list') and re.match(r'^\d+$', video_id):
             return self.url_result('http://tvpot.daum.net/clip/ClipView.do?clipid=%s' % video_id)
 
         info = self._download_xml(
-            'http://tvpot.daum.net/clip/ClipInfoXml.do?' + query, video_id,
+            'http://tvpot.daum.net/clip/ClipInfoXml.do', video_id,
-            'Downloading video info')
+            'Downloading video info', query={'vid': video_id})
 
         formats = []
         for format_el in movie_data['output_list']['output_list']:

youtube_dl/extractor/dbtv.py

@@ -4,78 +4,47 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_str
-from ..utils import (
-    float_or_none,
-    int_or_none,
-    clean_html,
-)
 
 
 class DBTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dbtv\.no/(?:(?:lazyplayer|player)/)?(?P<id>[0-9]+)(?:#(?P<display_id>.+))?'
+    _VALID_URL = r'https?://(?:www\.)?dbtv\.no/(?:[^/]+/)?(?P<id>[0-9]+)(?:#(?P<display_id>.+))?'
     _TESTS = [{
         'url': 'http://dbtv.no/3649835190001#Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
-        'md5': 'b89953ed25dacb6edb3ef6c6f430f8bc',
+        'md5': '2e24f67936517b143a234b4cadf792ec',
         'info_dict': {
-            'id': '33100',
+            'id': '3649835190001',
             'display_id': 'Skulle_teste_ut_fornøyelsespark,_men_kollegaen_var_bare_opptatt_av_bikinikroppen',
             'ext': 'mp4',
             'title': 'Skulle teste ut fornøyelsespark, men kollegaen var bare opptatt av bikinikroppen',
             'description': 'md5:1504a54606c4dde3e4e61fc97aa857e0',
-            'thumbnail': 're:https?://.*\.jpg$',
+            'thumbnail': 're:https?://.*\.jpg',
-            'timestamp': 1404039863.438,
+            'timestamp': 1404039863,
             'upload_date': '20140629',
             'duration': 69.544,
-            'view_count': int,
+            'uploader_id': '1027729757001',
-            'categories': list,
+        },
-        }
+        'add_ie': ['BrightcoveNew']
     }, {
         'url': 'http://dbtv.no/3649835190001',
         'only_matching': True,
     }, {
         'url': 'http://www.dbtv.no/lazyplayer/4631135248001',
         'only_matching': True,
+    }, {
+        'url': 'http://dbtv.no/vice/5000634109001',
+        'only_matching': True,
+    }, {
+        'url': 'http://dbtv.no/filmtrailer/3359293614001',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id, display_id = re.match(self._VALID_URL, url).groups()
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id') or video_id
-
-        data = self._download_json(
-            'http://api.dbtv.no/discovery/%s' % video_id, display_id)
-
-        video = data['playlist'][0]
-
-        formats = [{
-            'url': f['URL'],
-            'vcodec': f.get('container'),
-            'width': int_or_none(f.get('width')),
-            'height': int_or_none(f.get('height')),
-            'vbr': float_or_none(f.get('rate'), 1000),
-            'filesize': int_or_none(f.get('size')),
-        } for f in video['renditions'] if 'URL' in f]
-
-        if not formats:
-            for url_key, format_id in [('URL', 'mp4'), ('HLSURL', 'hls')]:
-                if url_key in video:
-                    formats.append({
-                        'url': video[url_key],
-                        'format_id': format_id,
-                    })
-
-        self._sort_formats(formats)
 
         return {
-            'id': compat_str(video['id']),
+            '_type': 'url_transparent',
+            'url': 'http://players.brightcove.net/1027729757001/default_default/index.html?videoId=%s' % video_id,
+            'id': video_id,
             'display_id': display_id,
-            'title': video['title'],
+            'ie_key': 'BrightcoveNew',
-            'description': clean_html(video['desc']),
-            'thumbnail': video.get('splash') or video.get('thumb'),
-            'timestamp': float_or_none(video.get('publishedAt'), 1000),
-            'duration': float_or_none(video.get('length'), 1000),
-            'view_count': int_or_none(video.get('views')),
-            'categories': video.get('tags'),
-            'formats': formats,
         }

youtube_dl/extractor/dcn.py

@@ -20,7 +20,7 @@ from ..utils import (
 
 
 class DCNIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?show/(?P<show_id>\d+)/[^/]+(?:/(?P<video_id>\d+)/(?P<season_id>\d+))?'
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?show/(?P<show_id>\d+)/[^/]+(?:/(?P<video_id>\d+)/(?P<season_id>\d+))?'
 
     def _real_extract(self, url):
         show_id, video_id, season_id = re.match(self._VALID_URL, url).groups()
@@ -55,30 +55,32 @@ class DCNBaseIE(InfoExtractor):
             'is_live': is_live,
         }
 
-    def _extract_video_formats(self, webpage, video_id, entry_protocol):
+    def _extract_video_formats(self, webpage, video_id, m3u8_entry_protocol):
         formats = []
-        m3u8_url = self._html_search_regex(
+        format_url_base = 'http' + self._html_search_regex(
-            r'file\s*:\s*"([^"]+)', webpage, 'm3u8 url', fatal=False)
+            [
-        if m3u8_url:
+                r'file\s*:\s*"https?(://[^"]+)/playlist.m3u8',
-            formats.extend(self._extract_m3u8_formats(
+                r'<a[^>]+href="rtsp(://[^"]+)"'
-                m3u8_url, video_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=None))
+            ], webpage, 'format url')
-
+        # TODO: Current DASH formats are broken - $Time$ pattern in
-        rtsp_url = self._search_regex(
+        # <SegmentTemplate> not implemented yet
-            r'<a[^>]+href="(rtsp://[^"]+)"', webpage, 'rtsp url', fatal=False)
+        # formats.extend(self._extract_mpd_formats(
-        if rtsp_url:
+        #     format_url_base + '/manifest.mpd',
-            formats.append({
+        #     video_id, mpd_id='dash', fatal=False))
-                'url': rtsp_url,
+        formats.extend(self._extract_m3u8_formats(
-                'format_id': 'rtsp',
+            format_url_base + '/playlist.m3u8', video_id, 'mp4',
-            })
+            m3u8_entry_protocol, m3u8_id='hls', fatal=False))
-
+        formats.extend(self._extract_f4m_formats(
+            format_url_base + '/manifest.f4m',
+            video_id, f4m_id='hds', fatal=False))
         self._sort_formats(formats)
         return formats
 
 
 class DCNVideoIE(DCNBaseIE):
     IE_NAME = 'dcn:video'
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?(?:video/[^/]+|media|catchup/[^/]+/[^/]+)/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?(?:video(?:/[^/]+)?|media|catchup/[^/]+/[^/]+)/(?P<id>\d+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.dcndigital.ae/#/video/%D8%B1%D8%AD%D9%84%D8%A9-%D8%A7%D9%84%D8%B9%D9%85%D8%B1-%D8%A7%D9%84%D8%AD%D9%84%D9%82%D8%A9-1/17375',
         'info_dict':
         {
@@ -94,7 +96,10 @@ class DCNVideoIE(DCNBaseIE):
             # m3u8 download
             'skip_download': True,
         },
-    }
+    }, {
+        'url': 'http://awaan.ae/video/26723981/%D8%AF%D8%A7%D8%B1-%D8%A7%D9%84%D8%B3%D9%84%D8%A7%D9%85:-%D8%AE%D9%8A%D8%B1-%D8%AF%D9%88%D8%B1-%D8%A7%D9%84%D8%A3%D9%86%D8%B5%D8%A7%D8%B1',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -120,7 +125,7 @@ class DCNVideoIE(DCNBaseIE):
 
 class DCNLiveIE(DCNBaseIE):
     IE_NAME = 'dcn:live'
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?live/(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?live/(?P<id>\d+)'
 
     def _real_extract(self, url):
         channel_id = self._match_id(url)
@@ -147,7 +152,7 @@ class DCNLiveIE(DCNBaseIE):
 
 class DCNSeasonIE(InfoExtractor):
     IE_NAME = 'dcn:season'
-    _VALID_URL = r'https?://(?:www\.)?dcndigital\.ae/(?:#/)?program/(?:(?P<show_id>\d+)|season/(?P<season_id>\d+))'
+    _VALID_URL = r'https?://(?:www\.)?(?:awaan|dcndigital)\.ae/(?:#/)?program/(?:(?P<show_id>\d+)|season/(?P<season_id>\d+))'
     _TEST = {
         'url': 'http://dcndigital.ae/#/program/205024/%D9%85%D8%AD%D8%A7%D8%B6%D8%B1%D8%A7%D8%AA-%D8%A7%D9%84%D8%B4%D9%8A%D8%AE-%D8%A7%D9%84%D8%B4%D8%B9%D8%B1%D8%A7%D9%88%D9%8A',
         'info_dict':

youtube_dl/extractor/dw.py

@@ -35,6 +35,7 @@ class DWIE(InfoExtractor):
             'upload_date': '20160311',
         }
     }, {
+        # DW documentaries, only last for one or two weeks
         'url': 'http://www.dw.com/en/documentaries-welcome-to-the-90s-2016-05-21/e-19220158-9798',
         'md5': '56b6214ef463bfb9a3b71aeb886f3cf1',
         'info_dict': {
@@ -44,6 +45,7 @@ class DWIE(InfoExtractor):
             'description': 'Welcome to the 90s - The Golden Decade of Hip Hop',
             'upload_date': '20160521',
         },
+        'skip': 'Video removed',
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/eagleplatform.py

@@ -50,6 +50,14 @@ class EaglePlatformIE(InfoExtractor):
         'skip': 'Georestricted',
     }]
 
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(
+            r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//.+?\.media\.eagleplatform\.com/index/player\?.+?)\1',
+            webpage)
+        if mobj is not None:
+            return mobj.group('url')
+
     @staticmethod
     def _handle_error(response):
         status = int_or_none(response.get('status', 200))

youtube_dl/extractor/ellentv.py

@@ -6,12 +6,13 @@ import json
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
+    NO_DEFAULT,
 )
 
 
 class EllenTVIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?(?:ellentv|ellentube)\.com/videos/(?P<id>[a-z0-9_-]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.ellentv.com/videos/0-ipq1gsai/',
         'md5': '4294cf98bc165f218aaa0b89e0fd8042',
         'info_dict': {
@@ -22,24 +23,47 @@ class EllenTVIE(InfoExtractor):
             'timestamp': 1428035648,
             'upload_date': '20150403',
             'uploader_id': 'batchUser',
-        }
+        },
-    }
+    }, {
+        # not available via http://widgets.ellentube.com/
+        'url': 'http://www.ellentv.com/videos/1-szkgu2m2/',
+        'info_dict': {
+            'id': '1_szkgu2m2',
+            'ext': 'flv',
+            'title': "Ellen's Amazingly Talented Audience",
+            'description': 'md5:86ff1e376ff0d717d7171590e273f0a5',
+            'timestamp': 1255140900,
+            'upload_date': '20091010',
+            'uploader_id': 'ellenkaltura@gmail.com',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(
+        URLS = ('http://widgets.ellentube.com/videos/%s' % video_id, url)
-            'http://widgets.ellentube.com/videos/%s' % video_id,
-            video_id)
 
-        partner_id = self._search_regex(
+        for num, url_ in enumerate(URLS, 1):
-            r"var\s+partnerId\s*=\s*'([^']+)", webpage, 'partner id')
+            webpage = self._download_webpage(
+                url_, video_id, fatal=num == len(URLS))
 
-        kaltura_id = self._search_regex(
+            default = NO_DEFAULT if num == len(URLS) else None
-            [r'id="kaltura_player_([^"]+)"',
+
-             r"_wb_entry_id\s*:\s*'([^']+)",
+            partner_id = self._search_regex(
-             r'data-kaltura-entry-id="([^"]+)'],
+                r"var\s+partnerId\s*=\s*'([^']+)", webpage, 'partner id',
-            webpage, 'kaltura id')
+                default=default)
+
+            kaltura_id = self._search_regex(
+                [r'id="kaltura_player_([^"]+)"',
+                 r"_wb_entry_id\s*:\s*'([^']+)",
+                 r'data-kaltura-entry-id="([^"]+)'],
+                webpage, 'kaltura id', default=default)
+
+            if partner_id and kaltura_id:
+                break
 
         return self.url_result('kaltura:%s:%s' % (partner_id, kaltura_id), 'Kaltura')
 

youtube_dl/extractor/extractors.py

@@ -20,7 +20,11 @@ from .adobetv import (
     AdobeTVVideoIE,
 )
 from .adultswim import AdultSwimIE
-from .aenetworks import AENetworksIE
+from .aenetworks import (
+    AENetworksIE,
+    HistoryTopicIE,
+)
+from .afreecatv import AfreecaTVIE
 from .aftonbladet import AftonbladetIE
 from .airmozilla import AirMozillaIE
 from .aljazeera import AlJazeeraIE
@@ -43,7 +47,6 @@ from .archiveorg import ArchiveOrgIE
 from .ard import (
     ARDIE,
     ARDMediathekIE,
-    SportschauIE,
 )
 from .arte import (
     ArteTvIE,
@@ -70,6 +73,8 @@ from .bandcamp import BandcampIE, BandcampAlbumIE
 from .bbc import (
     BBCCoUkIE,
     BBCCoUkArticleIE,
+    BBCCoUkIPlayerPlaylistIE,
+    BBCCoUkPlaylistIE,
     BBCIE,
 )
 from .beeg import BeegIE
@@ -107,6 +112,10 @@ from .camwithher import CamWithHerIE
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
 from .canvas import CanvasIE
+from .carambatv import (
+    CarambaTVIE,
+    CarambaTVPageIE,
+)
 from .cbc import (
     CBCIE,
     CBCPlayerIE,
@@ -130,10 +139,11 @@ from .chirbit import (
     ChirbitProfileIE,
 )
 from .cinchcast import CinchcastIE
-from .cliprs import ClipRsIE
 from .clipfish import ClipfishIE
 from .cliphunter import CliphunterIE
+from .cliprs import ClipRsIE
 from .clipsyndicate import ClipsyndicateIE
+from .closertotruth import CloserToTruthIE
 from .cloudy import CloudyIE
 from .clubic import ClubicIE
 from .clyp import ClypIE
@@ -161,6 +171,8 @@ from .crunchyroll import (
 )
 from .cspan import CSpanIE
 from .ctsnews import CtsNewsIE
+from .ctv import CTVIE
+from .ctvnews import CTVNewsIE
 from .cultureunplugged import CultureUnpluggedIE
 from .cwtv import CWTVIE
 from .dailymail import DailyMailIE
@@ -244,6 +256,7 @@ from .fivemin import FiveMinIE
 from .fivetv import FiveTVIE
 from .fktv import FKTVIE
 from .flickr import FlickrIE
+from .flipagram import FlipagramIE
 from .folketinget import FolketingetIE
 from .footyroom import FootyRoomIE
 from .formula1 import Formula1IE
@@ -269,6 +282,7 @@ from .freespeech import FreespeechIE
 from .freevideo import FreeVideoIE
 from .funimation import FunimationIE
 from .funnyordie import FunnyOrDieIE
+from .fusion import FusionIE
 from .gameinformer import GameInformerIE
 from .gamekings import GamekingsIE
 from .gameone import (
@@ -278,7 +292,6 @@ from .gameone import (
 from .gamersyde import GamersydeIE
 from .gamespot import GameSpotIE
 from .gamestar import GameStarIE
-from .gametrailers import GametrailersIE
 from .gazeta import GazetaIE
 from .gdcvault import GDCVaultIE
 from .generic import GenericIE
@@ -291,6 +304,7 @@ from .globo import (
     GloboArticleIE,
 )
 from .godtube import GodTubeIE
+from .godtv import GodTVIE
 from .goldenmoustache import GoldenMoustacheIE
 from .golem import GolemIE
 from .googledrive import GoogleDriveIE
@@ -313,6 +327,10 @@ from .hotnewhiphop import HotNewHipHopIE
 from .hotstar import HotStarIE
 from .howcast import HowcastIE
 from .howstuffworks import HowStuffWorksIE
+from .hrti import (
+    HRTiIE,
+    HRTiPlaylistIE,
+)
 from .huffpost import HuffPostIE
 from .hypem import HypemIE
 from .iconosquare import IconosquareIE
@@ -351,6 +369,7 @@ from .jove import JoveIE
 from .jwplatform import JWPlatformIE
 from .jpopsukitv import JpopsukiIE
 from .kaltura import KalturaIE
+from .kamcord import KamcordIE
 from .kanalplay import KanalPlayIE
 from .kankan import KankanIE
 from .karaoketv import KaraoketvIE
@@ -415,6 +434,7 @@ from .makerschannel import MakersChannelIE
 from .makertv import MakerTVIE
 from .matchtv import MatchTVIE
 from .mdr import MDRIE
+from .meta import METAIE
 from .metacafe import MetacafeIE
 from .metacritic import MetacriticIE
 from .mgoon import MgoonIE
@@ -447,6 +467,7 @@ from .motherless import MotherlessIE
 from .motorsport import MotorsportIE
 from .movieclips import MovieClipsIE
 from .moviezine import MoviezineIE
+from .msn import MSNIE
 from .mtv import (
     MTVIE,
     MTVServicesEmbeddedIE,
@@ -473,7 +494,6 @@ from .nbc import (
     NBCNewsIE,
     NBCSportsIE,
     NBCSportsVPlayerIE,
-    MSNBCIE,
 )
 from .ndr import (
     NDRIE,
@@ -510,8 +530,12 @@ from .nhl import (
     NHLVideocenterCategoryIE,
     NHLIE,
 )
-from .nick import NickIE
+from .nick import (
+    NickIE,
+    NickDeIE,
+)
 from .niconico import NiconicoIE, NiconicoPlaylistIE
+from .ninecninemedia import NineCNineMediaIE
 from .ninegag import NineGagIE
 from .noco import NocoIE
 from .normalboots import NormalbootsIE
@@ -559,6 +583,10 @@ from .nytimes import (
 from .nuvid import NuvidIE
 from .odnoklassniki import OdnoklassnikiIE
 from .oktoberfesttv import OktoberfestTVIE
+from .onet import (
+    OnetIE,
+    OnetChannelIE,
+)
 from .onionstudios import OnionStudiosIE
 from .ooyala import (
     OoyalaIE,
@@ -597,6 +625,7 @@ from .pluralsight import (
     PluralsightCourseIE,
 )
 from .podomatic import PodomaticIE
+from .polskieradio import PolskieRadioIE
 from .porn91 import Porn91IE
 from .pornhd import PornHdIE
 from .pornhub import (
@@ -620,7 +649,10 @@ from .qqmusic import (
     QQMusicToplistIE,
     QQMusicPlaylistIE,
 )
-from .r7 import R7IE
+from .r7 import (
+    R7IE,
+    R7ArticleIE,
+)
 from .radiocanada import (
     RadioCanadaIE,
     RadioCanadaAudioVideoIE,
@@ -647,6 +679,8 @@ from .revision3 import (
 from .rice import RICEIE
 from .ringtv import RingTVIE
 from .ro220 import Ro220IE
+from .rockstargames import RockstarGamesIE
+from .roosterteeth import RoosterTeethIE
 from .rottentomatoes import RottenTomatoesIE
 from .roxwel import RoxwelIE
 from .rtbf import RTBFIE
@@ -657,6 +691,7 @@ from .rtp import RTPIE
 from .rts import RTSIE
 from .rtve import RTVEALaCartaIE, RTVELiveIE, RTVEInfantilIE
 from .rtvnh import RTVNHIE
+from .rudo import RudoIE
 from .ruhd import RUHDIE
 from .ruleporn import RulePornIE
 from .rutube import (
@@ -691,10 +726,12 @@ from .shahid import ShahidIE
 from .shared import SharedIE
 from .sharesix import ShareSixIE
 from .sina import SinaIE
+from .sixplay import SixPlayIE
 from .skynewsarabia import (
     SkyNewsArabiaIE,
     SkyNewsArabiaArticleIE,
 )
+from .skysports import SkySportsIE
 from .slideshare import SlideshareIE
 from .slutload import SlutloadIE
 from .smotri import (
@@ -735,6 +772,7 @@ from .sportbox import (
     SportBoxEmbedIE,
 )
 from .sportdeutschland import SportDeutschlandIE
+from .sportschau import SportschauIE
 from .srgssr import (
     SRGSSRIE,
     SRGSSRPlayIE,
@@ -775,6 +813,7 @@ from .telecinco import TelecincoIE
 from .telegraaf import TelegraafIE
 from .telemb import TeleMBIE
 from .teletask import TeleTaskIE
+from .telewebion import TelewebionIE
 from .testurl import TestURLIE
 from .tf1 import TF1IE
 from .theintercept import TheInterceptIE
@@ -859,6 +898,7 @@ from .twitch import (
     TwitchProfileIE,
     TwitchPastBroadcastsIE,
     TwitchStreamIE,
+    TwitchClipsIE,
 )
 from .twitter import (
     TwitterCardIE,
@@ -873,6 +913,7 @@ from .udn import UDNEmbedIE
 from .digiteka import DigitekaIE
 from .unistra import UnistraIE
 from .urort import UrortIE
+from .urplay import URPlayIE
 from .usatoday import USATodayIE
 from .ustream import UstreamIE, UstreamChannelIE
 from .ustudio import (
@@ -899,6 +940,7 @@ from .vice import (
     ViceIE,
     ViceShowIE,
 )
+from .vidbit import VidbitIE
 from .viddler import ViddlerIE
 from .videodetective import VideoDetectiveIE
 from .videofyme import VideofyMeIE
@@ -910,6 +952,7 @@ from .videomore import (
 )
 from .videopremium import VideoPremiumIE
 from .videott import VideoTtIE
+from .vidio import VidioIE
 from .vidme import (
     VidmeIE,
     VidmeUserIE,
@@ -946,6 +989,7 @@ from .viki import (
 from .vk import (
     VKIE,
     VKUserVideosIE,
+    VKWallPostIE,
 )
 from .vlive import VLiveIE
 from .vodlocker import VodlockerIE
@@ -955,7 +999,6 @@ from .vporn import VpornIE
 from .vrt import VRTIE
 from .vube import VubeIE
 from .vuclip import VuClipIE
-from .vulture import VultureIE
 from .walla import WallaIE
 from .washingtonpost import (
     WashingtonPostIE,
@@ -966,18 +1009,19 @@ from .watchindianporn import WatchIndianPornIE
 from .wdr import (
     WDRIE,
     WDRMobileIE,
-    WDRMausIE,
 )
 from .webofstories import (
     WebOfStoriesIE,
     WebOfStoriesPlaylistIE,
 )
-from .weibo import WeiboIE
 from .weiqitv import WeiqiTVIE
 from .wimp import WimpIE
 from .wistia import WistiaIE
 from .worldstarhiphop import WorldStarHipHopIE
-from .wrzuta import WrzutaIE
+from .wrzuta import (
+    WrzutaIE,
+    WrzutaPlaylistIE,
+)
 from .wsj import WSJIE
 from .xbef import XBefIE
 from .xboxclips import XboxClipsIE
@@ -1013,7 +1057,10 @@ from .yesjapan import YesJapanIE
 from .yinyuetai import YinYueTaiIE
 from .ynet import YnetIE
 from .youjizz import YouJizzIE
-from .youku import YoukuIE
+from .youku import (
+    YoukuIE,
+    YoukuShowIE,
+)
 from .youporn import YouPornIE
 from .yourupload import YourUploadIE
 from .youtube import (
@@ -1028,6 +1075,7 @@ from .youtube import (
     YoutubeSearchDateIE,
     YoutubeSearchIE,
     YoutubeSearchURLIE,
+    YoutubeSharedVideoIE,
     YoutubeShowIE,
     YoutubeSubscriptionsIE,
     YoutubeTruncatedIDIE,

youtube_dl/extractor/facebook.py

@@ -129,6 +129,21 @@ class FacebookIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = re.search(
+            r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
+        if mobj is not None:
+            return mobj.group('url')
+
+        # Facebook API embed
+        # see https://developers.facebook.com/docs/plugins/embedded-video-player
+        mobj = re.search(r'''(?x)<div[^>]+
+                class=(?P<q1>[\'"])[^\'"]*\bfb-(?:video|post)\b[^\'"]*(?P=q1)[^>]+
+                data-href=(?P<q2>[\'"])(?P<url>(?:https?:)?//(?:www\.)?facebook.com/.+?)(?P=q2)''', webpage)
+        if mobj is not None:
+            return mobj.group('url')
+
     def _login(self):
         (useremail, password) = self._get_login_info()
         if useremail is None:
@@ -204,12 +219,25 @@ class FacebookIE(InfoExtractor):
 
         BEFORE = '{swf.addParam(param[0], param[1]);});'
         AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});'
-        m = re.search(re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER), webpage)
+        PATTERN = re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER)
-        if m:
+
-            swf_params = m.group(1).replace('\\\\', '\\').replace('\\"', '"')
+        for m in re.findall(PATTERN, webpage):
+            swf_params = m.replace('\\\\', '\\').replace('\\"', '"')
             data = dict(json.loads(swf_params))
             params_raw = compat_urllib_parse_unquote(data['params'])
-            video_data = json.loads(params_raw)['video_data']
+            video_data_candidate = json.loads(params_raw)['video_data']
+            for _, f in video_data_candidate.items():
+                if not f:
+                    continue
+                if isinstance(f, dict):
+                    f = [f]
+                if not isinstance(f, list):
+                    continue
+                if f[0].get('video_id') == video_id:
+                    video_data = video_data_candidate
+                    break
+            if video_data:
+                break
 
         def video_data_list2dict(video_data):
             ret = {}
@@ -239,6 +267,8 @@ class FacebookIE(InfoExtractor):
 
         formats = []
         for format_id, f in video_data.items():
+            if f and isinstance(f, dict):
+                f = [f]
             if not f or not isinstance(f, list):
                 continue
             for quality in ('sd', 'hd'):

youtube_dl/extractor/flipagram.py

@@ -0,0 +1,115 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    float_or_none,
+    try_get,
+    unified_timestamp,
+)
+
+
+class FlipagramIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?flipagram\.com/f/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'https://flipagram.com/f/nyvTSJMKId',
+        'md5': '888dcf08b7ea671381f00fab74692755',
+        'info_dict': {
+            'id': 'nyvTSJMKId',
+            'ext': 'mp4',
+            'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction',
+            'description': 'md5:d55e32edc55261cae96a41fa85ff630e',
+            'duration': 35.571,
+            'timestamp': 1461244995,
+            'upload_date': '20160421',
+            'uploader': 'kitty juria',
+            'uploader_id': 'sjuria101',
+            'creator': 'kitty juria',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'comment_count': int,
+            'comments': list,
+            'formats': 'mincount:2',
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        video_data = self._parse_json(
+            self._search_regex(
+                r'window\.reactH2O\s*=\s*({.+});', webpage, 'video data'),
+            video_id)
+
+        flipagram = video_data['flipagram']
+        video = flipagram['video']
+
+        json_ld = self._search_json_ld(webpage, video_id, default=False)
+        title = json_ld.get('title') or flipagram['captionText']
+        description = json_ld.get('description') or flipagram.get('captionText')
+
+        formats = [{
+            'url': video['url'],
+            'width': int_or_none(video.get('width')),
+            'height': int_or_none(video.get('height')),
+            'filesize': int_or_none(video_data.get('size')),
+        }]
+
+        preview_url = try_get(
+            flipagram, lambda x: x['music']['track']['previewUrl'], compat_str)
+        if preview_url:
+            formats.append({
+                'url': preview_url,
+                'ext': 'm4a',
+                'vcodec': 'none',
+            })
+
+        self._sort_formats(formats)
+
+        counts = flipagram.get('counts', {})
+        user = flipagram.get('user', {})
+        video_data = flipagram.get('video', {})
+
+        thumbnails = [{
+            'url': self._proto_relative_url(cover['url']),
+            'width': int_or_none(cover.get('width')),
+            'height': int_or_none(cover.get('height')),
+            'filesize': int_or_none(cover.get('size')),
+        } for cover in flipagram.get('covers', []) if cover.get('url')]
+
+        # Note that this only retrieves comments that are initally loaded.
+        # For videos with large amounts of comments, most won't be retrieved.
+        comments = []
+        for comment in video_data.get('comments', {}).get(video_id, {}).get('items', []):
+            text = comment.get('comment')
+            if not text or not isinstance(text, list):
+                continue
+            comments.append({
+                'author': comment.get('user', {}).get('name'),
+                'author_id': comment.get('user', {}).get('username'),
+                'id': comment.get('id'),
+                'text': text[0],
+                'timestamp': unified_timestamp(comment.get('created')),
+            })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': description,
+            'duration': float_or_none(flipagram.get('duration'), 1000),
+            'thumbnails': thumbnails,
+            'timestamp': unified_timestamp(flipagram.get('iso8601Created')),
+            'uploader': user.get('name'),
+            'uploader_id': user.get('username'),
+            'creator': user.get('name'),
+            'view_count': int_or_none(counts.get('plays')),
+            'like_count': int_or_none(counts.get('likes')),
+            'repost_count': int_or_none(counts.get('reflips')),
+            'comment_count': int_or_none(counts.get('comments')),
+            'comments': comments,
+            'formats': formats,
+        }

youtube_dl/extractor/foxsports.py

@@ -1,7 +1,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import smuggle_url
+from ..utils import (
+    smuggle_url,
+    update_url_query,
+)
 
 
 class FoxSportsIE(InfoExtractor):
@@ -9,11 +12,15 @@ class FoxSportsIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://www.foxsports.com/video?vid=432609859715',
+        'md5': 'b49050e955bebe32c301972e4012ac17',
         'info_dict': {
-            'id': 'gA0bHB3Ladz3',
+            'id': 'i0qKWsk3qJaM',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Courtney Lee on going up 2-0 in series vs. Blazers',
             'description': 'Courtney Lee talks about Memphis being focused.',
+            'upload_date': '20150423',
+            'timestamp': 1429761109,
+            'uploader': 'NEWA-FNG-FOXSPORTS',
         },
         'add_ie': ['ThePlatform'],
     }
@@ -28,5 +35,8 @@ class FoxSportsIE(InfoExtractor):
                 r"data-player-config='([^']+)'", webpage, 'data player config'),
             video_id)
 
-        return self.url_result(smuggle_url(
+        return self.url_result(smuggle_url(update_url_query(
-            config['releaseURL'] + '&manifest=f4m', {'force_smil_url': True}))
+            config['releaseURL'], {
+                'mbr': 'true',
+                'switch': 'http',
+            }), {'force_smil_url': True}))

youtube_dl/extractor/francetv.py

@@ -14,7 +14,10 @@ from ..utils import (
     parse_duration,
     determine_ext,
 )
-from .dailymotion import DailymotionCloudIE
+from .dailymotion import (
+    DailymotionIE,
+    DailymotionCloudIE,
+)
 
 
 class FranceTVBaseInfoExtractor(InfoExtractor):
@@ -188,6 +191,21 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
         'params': {
             'skip_download': True,
         },
+    }, {
+        # Dailymotion embed
+        'url': 'http://www.francetvinfo.fr/politique/notre-dame-des-landes/video-sur-france-inter-cecile-duflot-denonce-le-regard-meprisant-de-patrick-cohen_1520091.html',
+        'md5': 'ee7f1828f25a648addc90cb2687b1f12',
+        'info_dict': {
+            'id': 'x4iiko0',
+            'ext': 'mp4',
+            'title': 'NDDL, référendum, Brexit : Cécile Duflot répond à Patrick Cohen',
+            'description': 'Au lendemain de la victoire du "oui" au référendum sur l\'aéroport de Notre-Dame-des-Landes, l\'ancienne ministre écologiste est l\'invitée de Patrick Cohen. Plus d\'info : https://www.franceinter.fr/emissions/le-7-9/le-7-9-27-juin-2016',
+            'timestamp': 1467011958,
+            'upload_date': '20160627',
+            'uploader': 'France Inter',
+            'uploader_id': 'x2q2ez',
+        },
+        'add_ie': ['Dailymotion'],
     }]
 
     def _real_extract(self, url):
@@ -197,7 +215,13 @@ class FranceTvInfoIE(FranceTVBaseInfoExtractor):
 
         dmcloud_url = DailymotionCloudIE._extract_dmcloud_url(webpage)
         if dmcloud_url:
-            return self.url_result(dmcloud_url, 'DailymotionCloud')
+            return self.url_result(dmcloud_url, DailymotionCloudIE.ie_key())
+
+        dailymotion_urls = DailymotionIE._extract_urls(webpage)
+        if dailymotion_urls:
+            return self.playlist_result([
+                self.url_result(dailymotion_url, DailymotionIE.ie_key())
+                for dailymotion_url in dailymotion_urls])
 
         video_id, catalogue = self._search_regex(
             (r'id-video=([^@]+@[^"]+)',

youtube_dl/extractor/fusion.py

@@ -0,0 +1,35 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from .ooyala import OoyalaIE
+
+
+class FusionIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?fusion\.net/video/(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://fusion.net/video/201781/u-s-and-panamanian-forces-work-together-to-stop-a-vessel-smuggling-drugs/',
+        'info_dict': {
+            'id': 'ZpcWNoMTE6x6uVIIWYpHh0qQDjxBuq5P',
+            'ext': 'mp4',
+            'title': 'U.S. and Panamanian forces work together to stop a vessel smuggling drugs',
+            'description': 'md5:0cc84a9943c064c0f46b128b41b1b0d7',
+            'duration': 140.0,
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'add_ie': ['Ooyala'],
+    }, {
+        'url': 'http://fusion.net/video/201781',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        ooyala_code = self._search_regex(
+            r'data-video-id=(["\'])(?P<code>.+?)\1',
+            webpage, 'ooyala code', group='code')
+
+        return OoyalaIE._build_url_result(ooyala_code)

youtube_dl/extractor/gamespot.py

@@ -1,19 +1,19 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
-from .common import InfoExtractor
+from .once import OnceIE
 from ..compat import (
     compat_urllib_parse_unquote,
-    compat_urlparse,
 )
 from ..utils import (
     unescapeHTML,
+    url_basename,
+    dict_get,
 )
 
 
-class GameSpotIE(InfoExtractor):
+class GameSpotIE(OnceIE):
     _VALID_URL = r'https?://(?:www\.)?gamespot\.com/.*-(?P<id>\d+)/?'
     _TESTS = [{
         'url': 'http://www.gamespot.com/videos/arma-3-community-guide-sitrep-i/2300-6410818/',
@@ -39,29 +39,73 @@ class GameSpotIE(InfoExtractor):
         webpage = self._download_webpage(url, page_id)
         data_video_json = self._search_regex(
             r'data-video=["\'](.*?)["\']', webpage, 'data video')
-        data_video = json.loads(unescapeHTML(data_video_json))
+        data_video = self._parse_json(unescapeHTML(data_video_json), page_id)
         streams = data_video['videoStreams']
 
+        manifest_url = None
         formats = []
         f4m_url = streams.get('f4m_stream')
-        if f4m_url is not None:
+        if f4m_url:
-            # Transform the manifest url to a link to the mp4 files
+            manifest_url = f4m_url
-            # they are used in mobile devices.
+            formats.extend(self._extract_f4m_formats(
-            f4m_path = compat_urlparse.urlparse(f4m_url).path
+                f4m_url + '?hdcore=3.7.0', page_id, f4m_id='hds', fatal=False))
-            QUALITIES_RE = r'((,\d+)+,?)'
+        m3u8_url = streams.get('m3u8_stream')
-            qualities = self._search_regex(QUALITIES_RE, f4m_path, 'qualities').strip(',').split(',')
+        if m3u8_url:
-            http_path = f4m_path[1:].split('/', 1)[1]
+            manifest_url = m3u8_url
-            http_template = re.sub(QUALITIES_RE, r'%s', http_path)
+            m3u8_formats = self._extract_m3u8_formats(
-            http_template = http_template.replace('.csmil/manifest.f4m', '')
+                m3u8_url, page_id, 'mp4', 'm3u8_native',
-            http_template = compat_urlparse.urljoin(
+                m3u8_id='hls', fatal=False)
-                'http://video.gamespotcdn.com/', http_template)
+            formats.extend(m3u8_formats)
-            for q in qualities:
+        progressive_url = dict_get(
-                formats.append({
+            streams, ('progressive_hd', 'progressive_high', 'progressive_low'))
-                    'url': http_template % q,
+        if progressive_url and manifest_url:
-                    'ext': 'mp4',
+            qualities_basename = self._search_regex(
-                    'format_id': q,
+                '/([^/]+)\.csmil/',
-                })
+                manifest_url, 'qualities basename', default=None)
-        else:
+            if qualities_basename:
+                QUALITIES_RE = r'((,\d+)+,?)'
+                qualities = self._search_regex(
+                    QUALITIES_RE, qualities_basename,
+                    'qualities', default=None)
+                if qualities:
+                    qualities = list(map(lambda q: int(q), qualities.strip(',').split(',')))
+                    qualities.sort()
+                    http_template = re.sub(QUALITIES_RE, r'%d', qualities_basename)
+                    http_url_basename = url_basename(progressive_url)
+                    if m3u8_formats:
+                        self._sort_formats(m3u8_formats)
+                        m3u8_formats = list(filter(
+                            lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
+                            m3u8_formats))
+                    if len(qualities) == len(m3u8_formats):
+                        for q, m3u8_format in zip(qualities, m3u8_formats):
+                            f = m3u8_format.copy()
+                            f.update({
+                                'url': progressive_url.replace(
+                                    http_url_basename, http_template % q),
+                                'format_id': f['format_id'].replace('hls', 'http'),
+                                'protocol': 'http',
+                            })
+                            formats.append(f)
+                    else:
+                        for q in qualities:
+                            formats.append({
+                                'url': progressive_url.replace(
+                                    http_url_basename, http_template % q),
+                                'ext': 'mp4',
+                                'format_id': 'http-%d' % q,
+                                'tbr': q,
+                            })
+
+        onceux_json = self._search_regex(
+            r'data-onceux-options=["\'](.*?)["\']', webpage, 'data video', default=None)
+        if onceux_json:
+            onceux_url = self._parse_json(unescapeHTML(onceux_json), page_id).get('metadataUri')
+            if onceux_url:
+                formats.extend(self._extract_once_formats(re.sub(
+                    r'https?://[^/]+', 'http://once.unicornmedia.com', onceux_url).replace('ads/vmap/', '')))
+
+        if not formats:
             for quality in ['sd', 'hd']:
                 # It's actually a link to a flv file
                 flv_url = streams.get('f4m_{0}'.format(quality))
@@ -71,6 +115,7 @@ class GameSpotIE(InfoExtractor):
                         'ext': 'flv',
                         'format_id': quality,
                     })
+        self._sort_formats(formats)
 
         return {
             'id': data_video['guid'],

youtube_dl/extractor/gametrailers.py

@@ -1,62 +0,0 @@
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-from ..utils import (
-    int_or_none,
-    parse_age_limit,
-    url_basename,
-)
-
-
-class GametrailersIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.gametrailers\.com/videos/view/[^/]+/(?P<id>.+)'
-
-    _TEST = {
-        'url': 'http://www.gametrailers.com/videos/view/gametrailers-com/116437-Just-Cause-3-Review',
-        'md5': 'f28c4efa0bdfaf9b760f6507955b6a6a',
-        'info_dict': {
-            'id': '2983958',
-            'ext': 'mp4',
-            'display_id': '116437-Just-Cause-3-Review',
-            'title': 'Just Cause 3 - Review',
-            'description': 'It\'s a lot of fun to shoot at things and then watch them explode in Just Cause 3, but should there be more to the experience than that?',
-        },
-    }
-
-    def _real_extract(self, url):
-        display_id = self._match_id(url)
-        webpage = self._download_webpage(url, display_id)
-        title = self._html_search_regex(
-            r'<title>(.+?)\|', webpage, 'title').strip()
-        embed_url = self._proto_relative_url(
-            self._search_regex(
-                r'src=\'(//embed.gametrailers.com/embed/[^\']+)\'', webpage,
-                'embed url'),
-            scheme='http:')
-        video_id = url_basename(embed_url)
-        embed_page = self._download_webpage(embed_url, video_id)
-        embed_vars_json = self._search_regex(
-            r'(?s)var embedVars = (\{.*?\})\s*</script>', embed_page,
-            'embed vars')
-        info = self._parse_json(embed_vars_json, video_id)
-
-        formats = []
-        for media in info['media']:
-            if media['mediaPurpose'] == 'play':
-                formats.append({
-                    'url': media['uri'],
-                    'height': media['height'],
-                    'width:': media['width'],
-                })
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'display_id': display_id,
-            'title': title,
-            'formats': formats,
-            'thumbnail': info.get('thumbUri'),
-            'description': self._og_search_description(webpage),
-            'duration': int_or_none(info.get('videoLengthInSeconds')),
-            'age_limit': parse_age_limit(info.get('audienceRating')),
-        }

youtube_dl/extractor/generic.py

@@ -49,7 +49,10 @@ from .pornhub import PornHubIE
 from .xhamster import XHamsterEmbedIE
 from .tnaflix import TNAFlixNetworkEmbedIE
 from .vimeo import VimeoIE
-from .dailymotion import DailymotionCloudIE
+from .dailymotion import (
+    DailymotionIE,
+    DailymotionCloudIE,
+)
 from .onionstudios import OnionStudiosIE
 from .viewlift import ViewLiftEmbedIE
 from .screenwavemedia import ScreenwaveMediaIE
@@ -63,6 +66,10 @@ from .instagram import InstagramIE
 from .liveleak import LiveLeakIE
 from .threeqsdn import ThreeQSDNIE
 from .theplatform import ThePlatformIE
+from .vessel import VesselIE
+from .kaltura import KalturaIE
+from .eagleplatform import EaglePlatformIE
+from .facebook import FacebookIE
 
 
 class GenericIE(InfoExtractor):
@@ -626,13 +633,13 @@ class GenericIE(InfoExtractor):
         },
         # MTVSercices embed
         {
-            'url': 'http://www.gametrailers.com/news-post/76093/north-america-europe-is-getting-that-mario-kart-8-mercedes-dlc-too',
+            'url': 'http://www.vulture.com/2016/06/new-key-peele-sketches-released.html',
-            'md5': '35727f82f58c76d996fc188f9755b0d5',
+            'md5': 'ca1aef97695ef2c1d6973256a57e5252',
             'info_dict': {
-                'id': '0306a69b-8adf-4fb5-aace-75f8e8cbfca9',
+                'id': '769f7ec0-0692-4d62-9b45-0d88074bffc1',
                 'ext': 'mp4',
-                'title': 'Review',
+                'title': 'Key and Peele|October 10, 2012|2|203|Liam Neesons - Uncensored',
-                'description': 'Mario\'s life in the fast lane has never looked so good.',
+                'description': 'Two valets share their love for movie star Liam Neesons.',
             },
         },
         # YouTube embed via <data-embed-url="">
@@ -919,6 +926,24 @@ class GenericIE(InfoExtractor):
             },
             'add_ie': ['Kaltura'],
         },
+        {
+            # Kaltura embedded via quoted entry_id
+            'url': 'https://www.oreilly.com/ideas/my-cloud-makes-pretty-pictures',
+            'info_dict': {
+                'id': '0_utuok90b',
+                'ext': 'mp4',
+                'title': '06_matthew_brender_raj_dutt',
+                'timestamp': 1466638791,
+                'upload_date': '20160622',
+            },
+            'add_ie': ['Kaltura'],
+            'expected_warnings': [
+                'Could not send HEAD request'
+            ],
+            'params': {
+                'skip_download': True,
+            }
+        },
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
@@ -1031,6 +1056,17 @@ class GenericIE(InfoExtractor):
                 'timestamp': 1389118457,
             },
         },
+        # NBC News embed
+        {
+            'url': 'http://www.vulture.com/2016/06/letterman-couldnt-care-less-about-late-night.html',
+            'md5': '1aa589c675898ae6d37a17913cf68d66',
+            'info_dict': {
+                'id': '701714499682',
+                'ext': 'mp4',
+                'title': 'PREVIEW: On Assignment: David Letterman',
+                'description': 'A preview of Tom Brokaw\'s interview with David Letterman as part of the On Assignment series powered by Dateline. Airs Sunday June 12 at 7/6c.',
+            },
+        },
         # UDN embed
         {
             'url': 'https://video.udn.com/news/300346',
@@ -1061,20 +1097,6 @@ class GenericIE(InfoExtractor):
                 'skip_download': True,
             }
         },
-        # Contains a SMIL manifest
-        {
-            'url': 'http://www.telewebion.com/fa/1263668/%D9%82%D8%B1%D8%B9%D9%87%E2%80%8C%DA%A9%D8%B4%DB%8C-%D9%84%DB%8C%DA%AF-%D9%82%D9%87%D8%B1%D9%85%D8%A7%D9%86%D8%A7%D9%86-%D8%A7%D8%B1%D9%88%D9%BE%D8%A7/%2B-%D9%81%D9%88%D8%AA%D8%A8%D8%A7%D9%84.html',
-            'info_dict': {
-                'id': 'file',
-                'ext': 'flv',
-                'title': '+ Football: Lottery Champions League Europe',
-                'uploader': 'www.telewebion.com',
-            },
-            'params': {
-                # rtmpe downloads
-                'skip_download': True,
-            }
-        },
         # Brightcove URL in single quotes
         {
             'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
@@ -1093,12 +1115,17 @@ class GenericIE(InfoExtractor):
         # Dailymotion Cloud video
         {
             'url': 'http://replay.publicsenat.fr/vod/le-debat/florent-kolandjian,dominique-cena,axel-decourtye,laurence-abeille,bruno-parmentier/175910',
-            'md5': '49444254273501a64675a7e68c502681',
+            'md5': 'dcaf23ad0c67a256f4278bce6e0bae38',
             'info_dict': {
-                'id': '5585de919473990de4bee11b',
+                'id': 'x2uy8t3',
                 'ext': 'mp4',
-                'title': 'Le débat',
+                'title': 'Sauvons les abeilles ! - Le débat',
+                'description': 'md5:d9082128b1c5277987825d684939ca26',
                 'thumbnail': 're:^https?://.*\.jpe?g$',
+                'timestamp': 1434970506,
+                'upload_date': '20150622',
+                'uploader': 'Public Sénat',
+                'uploader_id': 'xa9gza',
             }
         },
         # OnionStudios embed
@@ -1222,6 +1249,102 @@ class GenericIE(InfoExtractor):
                 'uploader': 'www.hudl.com',
             },
         },
+        # twitter:player embed
+        {
+            'url': 'http://www.theatlantic.com/video/index/484130/what-do-black-holes-sound-like/',
+            'md5': 'a3e0df96369831de324f0778e126653c',
+            'info_dict': {
+                'id': '4909620399001',
+                'ext': 'mp4',
+                'title': 'What Do Black Holes Sound Like?',
+                'description': 'what do black holes sound like',
+                'upload_date': '20160524',
+                'uploader_id': '29913724001',
+                'timestamp': 1464107587,
+                'uploader': 'TheAtlantic',
+            },
+            'add_ie': ['BrightcoveLegacy'],
+        },
+        # Facebook <iframe> embed
+        {
+            'url': 'https://www.hostblogger.de/blog/archives/6181-Auto-jagt-Betonmischer.html',
+            'md5': 'fbcde74f534176ecb015849146dd3aee',
+            'info_dict': {
+                'id': '599637780109885',
+                'ext': 'mp4',
+                'title': 'Facebook video #599637780109885',
+            },
+        },
+        # Facebook API embed
+        {
+            'url': 'http://www.lothype.com/blue-stars-2016-preview-standstill-full-show/',
+            'md5': 'a47372ee61b39a7b90287094d447d94e',
+            'info_dict': {
+                'id': '10153467542406923',
+                'ext': 'mp4',
+                'title': 'Facebook video #10153467542406923',
+            },
+        },
+        # Wordpress "YouTube Video Importer" plugin
+        {
+            'url': 'http://www.lothype.com/blue-devils-drumline-stanford-lot-2016/',
+            'md5': 'd16797741b560b485194eddda8121b48',
+            'info_dict': {
+                'id': 'HNTXWDXV9Is',
+                'ext': 'mp4',
+                'title': 'Blue Devils Drumline Stanford lot 2016',
+                'upload_date': '20160627',
+                'uploader_id': 'GENOCIDE8GENERAL10',
+                'uploader': 'cylus cyrus',
+            },
+        },
+        {
+            # video stored on custom kaltura server
+            'url': 'http://www.expansion.com/multimedia/videos.html?media=EQcM30NHIPv',
+            'md5': '537617d06e64dfed891fa1593c4b30cc',
+            'info_dict': {
+                'id': '0_1iotm5bh',
+                'ext': 'mp4',
+                'title': 'Elecciones británicas: 5 lecciones para Rajoy',
+                'description': 'md5:435a89d68b9760b92ce67ed227055f16',
+                'uploader_id': 'videos.expansion@el-mundo.net',
+                'upload_date': '20150429',
+                'timestamp': 1430303472,
+            },
+            'add_ie': ['Kaltura'],
+        },
+        {
+            # Non-standard Vimeo embed
+            'url': 'https://openclassrooms.com/courses/understanding-the-web',
+            'md5': '64d86f1c7d369afd9a78b38cbb88d80a',
+            'info_dict': {
+                'id': '148867247',
+                'ext': 'mp4',
+                'title': 'Understanding the web - Teaser',
+                'description': 'This is "Understanding the web - Teaser" by openclassrooms on Vimeo, the home for high quality videos and the people who love them.',
+                'upload_date': '20151214',
+                'uploader': 'OpenClassrooms',
+                'uploader_id': 'openclassrooms',
+            },
+            'add_ie': ['Vimeo'],
+        },
+        # {
+        #     # TODO: find another test
+        #     # http://schema.org/VideoObject
+        #     'url': 'https://flipagram.com/f/nyvTSJMKId',
+        #     'md5': '888dcf08b7ea671381f00fab74692755',
+        #     'info_dict': {
+        #         'id': 'nyvTSJMKId',
+        #         'ext': 'mp4',
+        #         'title': 'Flipagram by sjuria101 featuring Midnight Memories by One Direction',
+        #         'description': '#love for cats.',
+        #         'timestamp': 1461244995,
+        #         'upload_date': '20160421',
+        #     },
+        #     'params': {
+        #         'force_generic_extractor': True,
+        #     },
+        # }
     ]
 
     def report_following_redirect(self, new_url):
@@ -1533,6 +1656,11 @@ class GenericIE(InfoExtractor):
         if tp_urls:
             return _playlist_from_matches(tp_urls, ie='ThePlatform')
 
+        # Look for Vessel embeds
+        vessel_urls = VesselIE._extract_urls(webpage)
+        if vessel_urls:
+            return _playlist_from_matches(vessel_urls, ie=VesselIE.ie_key())
+
         # Look for embedded rtl.nl player
         matches = re.findall(
             r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
@@ -1573,12 +1701,16 @@ class GenericIE(InfoExtractor):
         if matches:
             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
 
-        # Look for embedded Dailymotion player
+        # Look for Wordpress "YouTube Video Importer" plugin
-        matches = re.findall(
+        matches = re.findall(r'''(?x)<div[^>]+
-            r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1', webpage)
+            class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
+            data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
         if matches:
-            return _playlist_from_matches(
+            return _playlist_from_matches(matches, lambda m: m[-1])
-                matches, lambda m: unescapeHTML(m[1]))
+
+        matches = DailymotionIE._extract_urls(webpage)
+        if matches:
+            return _playlist_from_matches(matches)
 
         # Look for embedded Dailymotion playlist player (#3822)
         m = re.search(
@@ -1715,10 +1847,9 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'))
 
         # Look for embedded Facebook player
-        mobj = re.search(
+        facebook_url = FacebookIE._extract_url(webpage)
-            r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
+        if facebook_url is not None:
-        if mobj is not None:
+            return self.url_result(facebook_url, 'Facebook')
-            return self.url_result(mobj.group('url'), 'Facebook')
 
         # Look for embedded VK player
         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
@@ -1840,14 +1971,6 @@ class GenericIE(InfoExtractor):
             url = unescapeHTML(mobj.group('url'))
             return self.url_result(url)
 
-        # Look for embedded vulture.com player
-        mobj = re.search(
-            r'<iframe src="(?P<url>https?://video\.vulture\.com/[^"]+)"',
-            webpage)
-        if mobj is not None:
-            url = unescapeHTML(mobj.group('url'))
-            return self.url_result(url, ie='Vulture')
-
         # Look for embedded mtvservices player
         mtvservices_url = MTVServicesEmbeddedIE._extract_url(webpage)
         if mtvservices_url:
@@ -1908,18 +2031,14 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'), 'Zapiks')
 
         # Look for Kaltura embeds
-        mobj = (re.search(r"(?s)kWidget\.(?:thumb)?[Ee]mbed\(\{.*?(?P<q1>['\"])wid(?P=q1)\s*:\s*(?P<q2>['\"])_?(?P<partner_id>[^'\"]+)(?P=q2),.*?(?P<q3>['\"])entry_?[Ii]d(?P=q3)\s*:\s*(?P<q4>['\"])(?P<id>[^'\"]+)(?P=q4),", webpage) or
+        kaltura_url = KalturaIE._extract_url(webpage)
-                re.search(r'(?s)(?P<q1>["\'])(?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?(?P=q1).*?entry_?[Ii]d\s*:\s*(?P<q2>["\'])(?P<id>.+?)(?P=q2)', webpage))
+        if kaltura_url:
-        if mobj is not None:
+            return self.url_result(smuggle_url(kaltura_url, {'source_url': url}), KalturaIE.ie_key())
-            return self.url_result(smuggle_url(
-                'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict(),
-                {'source_url': url}), 'Kaltura')
 
         # Look for Eagle.Platform embeds
-        mobj = re.search(
+        eagleplatform_url = EaglePlatformIE._extract_url(webpage)
-            r'<iframe[^>]+src="(?P<url>https?://.+?\.media\.eagleplatform\.com/index/player\?.+?)"', webpage)
+        if eagleplatform_url:
-        if mobj is not None:
+            return self.url_result(eagleplatform_url, EaglePlatformIE.ie_key())
-            return self.url_result(mobj.group('url'), 'EaglePlatform')
 
         # Look for ClipYou (uses Eagle.Platform) embeds
         mobj = re.search(
@@ -1960,6 +2079,12 @@ class GenericIE(InfoExtractor):
         if nbc_sports_url:
             return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
 
+        # Look for NBC News embeds
+        nbc_news_embed_url = re.search(
+            r'<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//www\.nbcnews\.com/widget/video-embed/[^"\']+)\1', webpage)
+        if nbc_news_embed_url:
+            return self.url_result(nbc_news_embed_url.group('url'), 'NBCNews')
+
         # Look for Google Drive embeds
         google_drive_url = GoogleDriveIE._extract_url(webpage)
         if google_drive_url:
@@ -2059,6 +2184,24 @@ class GenericIE(InfoExtractor):
                 'uploader': video_uploader,
             }
 
+        # https://dev.twitter.com/cards/types/player#On_twitter.com_via_desktop_browser
+        embed_url = self._html_search_meta('twitter:player', webpage, default=None)
+        if embed_url:
+            return self.url_result(embed_url)
+
+        # Looking for http://schema.org/VideoObject
+        json_ld = self._search_json_ld(
+            webpage, video_id, default=None, expected_type='VideoObject')
+        if json_ld and json_ld.get('url'):
+            info_dict.update({
+                'title': video_title or info_dict['title'],
+                'description': video_description,
+                'thumbnail': video_thumbnail,
+                'age_limit': age_limit
+            })
+            info_dict.update(json_ld)
+            return info_dict
+
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
                 return True

youtube_dl/extractor/godtv.py

@@ -0,0 +1,66 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from .ooyala import OoyalaIE
+from ..utils import js_to_json
+
+
+class GodTVIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?god\.tv(?:/[^/]+)*/(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'http://god.tv/jesus-image/video/jesus-conference-2016/randy-needham',
+        'info_dict': {
+            'id': 'lpd3g2MzE6D1g8zFAKz8AGpxWcpu6o_3',
+            'ext': 'mp4',
+            'title': 'Randy Needham',
+            'duration': 3615.08,
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        'url': 'http://god.tv/playlist/bible-study',
+        'info_dict': {
+            'id': 'bible-study',
+        },
+        'playlist_mincount': 37,
+    }, {
+        'url': 'http://god.tv/node/15097',
+        'only_matching': True,
+    }, {
+        'url': 'http://god.tv/live/africa',
+        'only_matching': True,
+    }, {
+        'url': 'http://god.tv/liveevents',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        settings = self._parse_json(
+            self._search_regex(
+                r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
+                webpage, 'settings', default='{}'),
+            display_id, transform_source=js_to_json, fatal=False)
+
+        ooyala_id = None
+
+        if settings:
+            playlist = settings.get('playlist')
+            if playlist and isinstance(playlist, list):
+                entries = [
+                    OoyalaIE._build_url_result(video['content_id'])
+                    for video in playlist if video.get('content_id')]
+                if entries:
+                    return self.playlist_result(entries, display_id)
+            ooyala_id = settings.get('ooyala', {}).get('content_id')
+
+        if not ooyala_id:
+            ooyala_id = self._search_regex(
+                r'["\']content_id["\']\s*:\s*(["\'])(?P<id>[\w-]+)\1',
+                webpage, 'ooyala id', group='id')
+
+        return OoyalaIE._build_url_result(ooyala_id)

youtube_dl/extractor/hrti.py

@@ -0,0 +1,202 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_HTTPError
+from ..utils import (
+    clean_html,
+    ExtractorError,
+    int_or_none,
+    parse_age_limit,
+    sanitized_Request,
+    try_get,
+)
+
+
+class HRTiBaseIE(InfoExtractor):
+    """
+        Base Information Extractor for Croatian Radiotelevision
+        video on demand site https://hrti.hrt.hr
+        Reverse engineered from the JavaScript app in app.min.js
+    """
+    _NETRC_MACHINE = 'hrti'
+
+    _APP_LANGUAGE = 'hr'
+    _APP_VERSION = '1.1'
+    _APP_PUBLICATION_ID = 'all_in_one'
+    _API_URL = 'http://clientapi.hrt.hr/client_api.php/config/identify/format/json'
+
+    def _initialize_api(self):
+        init_data = {
+            'application_publication_id': self._APP_PUBLICATION_ID
+        }
+
+        uuid = self._download_json(
+            self._API_URL, None, note='Downloading uuid',
+            errnote='Unable to download uuid',
+            data=json.dumps(init_data).encode('utf-8'))['uuid']
+
+        app_data = {
+            'uuid': uuid,
+            'application_publication_id': self._APP_PUBLICATION_ID,
+            'application_version': self._APP_VERSION
+        }
+
+        req = sanitized_Request(self._API_URL, data=json.dumps(app_data).encode('utf-8'))
+        req.get_method = lambda: 'PUT'
+
+        resources = self._download_json(
+            req, None, note='Downloading session information',
+            errnote='Unable to download session information')
+
+        self._session_id = resources['session_id']
+
+        modules = resources['modules']
+
+        self._search_url = modules['vod_catalog']['resources']['search']['uri'].format(
+            language=self._APP_LANGUAGE,
+            application_id=self._APP_PUBLICATION_ID)
+
+        self._login_url = (modules['user']['resources']['login']['uri'] +
+                           '/format/json').format(session_id=self._session_id)
+
+        self._logout_url = modules['user']['resources']['logout']['uri']
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        # TODO: figure out authentication with cookies
+        if username is None or password is None:
+            self.raise_login_required()
+
+        auth_data = {
+            'username': username,
+            'password': password,
+        }
+
+        try:
+            auth_info = self._download_json(
+                self._login_url, None, note='Logging in', errnote='Unable to log in',
+                data=json.dumps(auth_data).encode('utf-8'))
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 406:
+                auth_info = self._parse_json(e.cause.read().encode('utf-8'), None)
+            else:
+                raise
+
+        error_message = auth_info.get('error', {}).get('message')
+        if error_message:
+            raise ExtractorError(
+                '%s said: %s' % (self.IE_NAME, error_message),
+                expected=True)
+
+        self._token = auth_info['secure_streaming_token']
+
+    def _real_initialize(self):
+        self._initialize_api()
+        self._login()
+
+
+class HRTiIE(HRTiBaseIE):
+    _VALID_URL = r'''(?x)
+                        (?:
+                            hrti:(?P<short_id>[0-9]+)|
+                            https?://
+                                hrti\.hrt\.hr/\#/video/show/(?P<id>[0-9]+)/(?P<display_id>[^/]+)?
+                        )
+                    '''
+    _TESTS = [{
+        'url': 'https://hrti.hrt.hr/#/video/show/2181385/republika-dokumentarna-serija-16-hd',
+        'info_dict': {
+            'id': '2181385',
+            'display_id': 'republika-dokumentarna-serija-16-hd',
+            'ext': 'mp4',
+            'title': 'REPUBLIKA, dokumentarna serija (1/6) (HD)',
+            'description': 'md5:48af85f620e8e0e1df4096270568544f',
+            'duration': 2922,
+            'view_count': int,
+            'average_rating': int,
+            'episode_number': int,
+            'season_number': int,
+            'age_limit': 12,
+        },
+        'skip': 'Requires account credentials',
+    }, {
+        'url': 'https://hrti.hrt.hr/#/video/show/2181385/',
+        'only_matching': True,
+    }, {
+        'url': 'hrti:2181385',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('short_id') or mobj.group('id')
+        display_id = mobj.group('display_id') or video_id
+
+        video = self._download_json(
+            '%s/video_id/%s/format/json' % (self._search_url, video_id),
+            display_id, 'Downloading video metadata JSON')['video'][0]
+
+        title_info = video['title']
+        title = title_info['title_long']
+
+        movie = video['video_assets']['movie'][0]
+        m3u8_url = movie['url'].format(TOKEN=self._token)
+        formats = self._extract_m3u8_formats(
+            m3u8_url, display_id, 'mp4', entry_protocol='m3u8_native',
+            m3u8_id='hls')
+        self._sort_formats(formats)
+
+        description = clean_html(title_info.get('summary_long'))
+        age_limit = parse_age_limit(video.get('parental_control', {}).get('rating'))
+        view_count = int_or_none(video.get('views'))
+        average_rating = int_or_none(video.get('user_rating'))
+        duration = int_or_none(movie.get('duration'))
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'duration': duration,
+            'view_count': view_count,
+            'average_rating': average_rating,
+            'age_limit': age_limit,
+            'formats': formats,
+        }
+
+
+class HRTiPlaylistIE(HRTiBaseIE):
+    _VALID_URL = r'https?://hrti.hrt.hr/#/video/list/category/(?P<id>[0-9]+)/(?P<display_id>[^/]+)?'
+    _TESTS = [{
+        'url': 'https://hrti.hrt.hr/#/video/list/category/212/ekumena',
+        'info_dict': {
+            'id': '212',
+            'title': 'ekumena',
+        },
+        'playlist_mincount': 8,
+        'skip': 'Requires account credentials',
+    }, {
+        'url': 'https://hrti.hrt.hr/#/video/list/category/212/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        category_id = mobj.group('id')
+        display_id = mobj.group('display_id') or category_id
+
+        response = self._download_json(
+            '%s/category_id/%s/format/json' % (self._search_url, category_id),
+            display_id, 'Downloading video metadata JSON')
+
+        video_ids = try_get(
+            response, lambda x: x['video_listings'][0]['alternatives'][0]['list'],
+            list) or [video['id'] for video in response.get('videos', []) if video.get('id')]
+
+        entries = [self.url_result('hrti:%s' % video_id) for video_id in video_ids]
+
+        return self.playlist_result(entries, category_id, display_id)

youtube_dl/extractor/imdb.py

@@ -12,7 +12,7 @@ from ..utils import (
 class ImdbIE(InfoExtractor):
     IE_NAME = 'imdb'
     IE_DESC = 'Internet Movie Database trailers'
-    _VALID_URL = r'https?://(?:www|m)\.imdb\.com/video/[^/]+/vi(?P<id>\d+)'
+    _VALID_URL = r'https?://(?:www|m)\.imdb\.com/(?:video/[^/]+/|title/tt\d+.*?#lb-)vi(?P<id>\d+)'
 
     _TESTS = [{
         'url': 'http://www.imdb.com/video/imdb/vi2524815897',
@@ -25,6 +25,12 @@ class ImdbIE(InfoExtractor):
     }, {
         'url': 'http://www.imdb.com/video/_/vi2524815897',
         'only_matching': True,
+    }, {
+        'url': 'http://www.imdb.com/title/tt1667889/?ref_=ext_shr_eml_vi#lb-vi2524815897',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.imdb.com/title/tt1667889/#lb-vi2524815897',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/indavideo.py

@@ -60,7 +60,8 @@ class IndavideoEmbedIE(InfoExtractor):
 
         formats = [{
             'url': video_url,
-            'height': self._search_regex(r'\.(\d{3,4})\.mp4$', video_url, 'height', default=None),
+            'height': int_or_none(self._search_regex(
+                r'\.(\d{3,4})\.mp4(?:\?|$)', video_url, 'height', default=None)),
         } for video_url in video_urls]
         self._sort_formats(formats)
 

youtube_dl/extractor/instagram.py

@@ -8,6 +8,7 @@ from ..utils import (
     int_or_none,
     limit_length,
     lowercase_escape,
+    try_get,
 )
 
 
@@ -19,10 +20,16 @@ class InstagramIE(InfoExtractor):
         'info_dict': {
             'id': 'aye83DjauH',
             'ext': 'mp4',
-            'uploader_id': 'naomipq',
             'title': 'Video by naomipq',
             'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8',
-        }
+            'thumbnail': 're:^https?://.*\.jpg',
+            'timestamp': 1371748545,
+            'upload_date': '20130620',
+            'uploader_id': 'naomipq',
+            'uploader': 'Naomi Leonor Phan-Quang',
+            'like_count': int,
+            'comment_count': int,
+        },
     }, {
         # missing description
         'url': 'https://www.instagram.com/p/BA-pQFBG8HZ/?taken-by=britneyspears',
@@ -31,6 +38,13 @@ class InstagramIE(InfoExtractor):
             'ext': 'mp4',
             'uploader_id': 'britneyspears',
             'title': 'Video by britneyspears',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'timestamp': 1453760977,
+            'upload_date': '20160125',
+            'uploader_id': 'britneyspears',
+            'uploader': 'Britney Spears',
+            'like_count': int,
+            'comment_count': int,
         },
         'params': {
             'skip_download': True,
@@ -67,21 +81,57 @@ class InstagramIE(InfoExtractor):
         url = mobj.group('url')
 
         webpage = self._download_webpage(url, video_id)
-        uploader_id = self._search_regex(r'"owner":{"username":"(.+?)"',
+
-                                         webpage, 'uploader id', fatal=False)
+        (video_url, description, thumbnail, timestamp, uploader,
-        desc = self._search_regex(
+         uploader_id, like_count, comment_count) = [None] * 8
-            r'"caption":"(.+?)"', webpage, 'description', default=None)
+
-        if desc is not None:
+        shared_data = self._parse_json(
-            desc = lowercase_escape(desc)
+            self._search_regex(
+                r'window\._sharedData\s*=\s*({.+?});',
+                webpage, 'shared data', default='{}'),
+            video_id, fatal=False)
+        if shared_data:
+            media = try_get(
+                shared_data, lambda x: x['entry_data']['PostPage'][0]['media'], dict)
+            if media:
+                video_url = media.get('video_url')
+                description = media.get('caption')
+                thumbnail = media.get('display_src')
+                timestamp = int_or_none(media.get('date'))
+                uploader = media.get('owner', {}).get('full_name')
+                uploader_id = media.get('owner', {}).get('username')
+                like_count = int_or_none(media.get('likes', {}).get('count'))
+                comment_count = int_or_none(media.get('comments', {}).get('count'))
+
+        if not video_url:
+            video_url = self._og_search_video_url(webpage, secure=False)
+
+        if not uploader_id:
+            uploader_id = self._search_regex(
+                r'"owner"\s*:\s*{\s*"username"\s*:\s*"(.+?)"',
+                webpage, 'uploader id', fatal=False)
+
+        if not description:
+            description = self._search_regex(
+                r'"caption"\s*:\s*"(.+?)"', webpage, 'description', default=None)
+            if description is not None:
+                description = lowercase_escape(description)
+
+        if not thumbnail:
+            thumbnail = self._og_search_thumbnail(webpage)
 
         return {
             'id': video_id,
-            'url': self._og_search_video_url(webpage, secure=False),
+            'url': video_url,
             'ext': 'mp4',
             'title': 'Video by %s' % uploader_id,
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'description': description,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
             'uploader_id': uploader_id,
-            'description': desc,
+            'uploader': uploader,
+            'like_count': like_count,
+            'comment_count': comment_count,
         }
 
 

youtube_dl/extractor/iqiyi.py

@@ -3,28 +3,22 @@ from __future__ import unicode_literals
 
 import hashlib
 import itertools
-import math
-import os
-import random
 import re
 import time
-import uuid
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_parse_qs,
     compat_str,
     compat_urllib_parse_urlencode,
-    compat_urllib_parse_urlparse,
 )
 from ..utils import (
+    clean_html,
     decode_packed_codes,
+    get_element_by_id,
+    get_element_by_attribute,
     ExtractorError,
     ohdave_rsa_encrypt,
     remove_start,
-    sanitized_Request,
-    urlencode_postdata,
-    url_basename,
 )
 
 
@@ -171,70 +165,21 @@ class IqiyiIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
-        'md5': '2cb594dc2781e6c941a110d8f358118b',
+        # MD5 checksum differs on my machine and Travis CI
         'info_dict': {
             'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
+            'ext': 'mp4',
             'title': '美国德州空中惊现奇异云团 酷似UFO',
-            'ext': 'f4v',
         }
     }, {
         'url': 'http://www.iqiyi.com/v_19rrhnnclk.html',
+        'md5': '667171934041350c5de3f5015f7f1152',
         'info_dict': {
             'id': 'e3f585b550a280af23c98b6cb2be19fb',
-            'title': '名侦探柯南第752集',
+            'ext': 'mp4',
-        },
+            'title': '名侦探柯南 国语版：第752集 迫近灰原秘密的黑影 下篇',
-        'playlist': [{
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part1',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part2',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part3',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part4',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part5',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part6',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part7',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }, {
-            'info_dict': {
-                'id': 'e3f585b550a280af23c98b6cb2be19fb_part8',
-                'ext': 'f4v',
-                'title': '名侦探柯南第752集',
-            },
-        }],
-        'params': {
-            'skip_download': True,
         },
+        'skip': 'Geo-restricted to China',
     }, {
         'url': 'http://www.iqiyi.com/w_19rt6o8t9p.html',
         'only_matching': True,
@@ -250,22 +195,10 @@ class IqiyiIE(InfoExtractor):
         'url': 'http://www.iqiyi.com/v_19rrny4w8w.html',
         'info_dict': {
             'id': 'f3cf468b39dddb30d676f89a91200dc1',
+            'ext': 'mp4',
             'title': '泰坦尼克号',
         },
-        'playlist': [{
+        'skip': 'Geo-restricted to China',
-            'info_dict': {
-                'id': 'f3cf468b39dddb30d676f89a91200dc1_part1',
-                'ext': 'f4v',
-                'title': '泰坦尼克号',
-            },
-        }, {
-            'info_dict': {
-                'id': 'f3cf468b39dddb30d676f89a91200dc1_part2',
-                'ext': 'f4v',
-                'title': '泰坦尼克号',
-            },
-        }],
-        'expected_warnings': ['Needs a VIP account for full video'],
     }, {
         'url': 'http://www.iqiyi.com/a_19rrhb8ce1.html',
         'info_dict': {
@@ -278,20 +211,15 @@ class IqiyiIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    _FORMATS_MAP = [
+    _FORMATS_MAP = {
-        ('1', 'h6'),
+        '96': 1,    # 216p, 240p
-        ('2', 'h5'),
+        '1': 2,     # 336p, 360p
-        ('3', 'h4'),
+        '2': 3,     # 480p, 504p
-        ('4', 'h3'),
+        '21': 4,    # 504p
-        ('5', 'h2'),
+        '4': 5,     # 720p
-        ('10', 'h1'),
+        '17': 5,    # 720p
-    ]
+        '5': 6,     # 1072p, 1080p
-
+        '18': 7,    # 1080p
-    AUTH_API_ERRORS = {
-        # No preview available (不允许试看鉴权失败)
-        'Q00505': 'This video requires a VIP account',
-        # End of preview time (试看结束鉴权失败)
-        'Q00506': 'Needs a VIP account for full video',
     }
 
     def _real_initialize(self):
@@ -352,177 +280,23 @@ class IqiyiIE(InfoExtractor):
 
         return True
 
-    def _authenticate_vip_video(self, api_video_url, video_id, tvid, _uuid, do_report_warning):
+    def get_raw_data(self, tvid, video_id):
-        auth_params = {
+        tm = int(time.time() * 1000)
-            # version and platform hard-coded in com/qiyi/player/core/model/remote/AuthenticationRemote.as
+
-            'version': '2.0',
+        key = 'd5fb4bd9d50c4be6948c97edd7254b0e'
-            'platform': 'b6c13e26323c537d',
+        sc = md5_text(compat_str(tm) + key + tvid)
-            'aid': tvid,
+        params = {
             'tvid': tvid,
-            'uid': '',
-            'deviceId': _uuid,
-            'playType': 'main',  # XXX: always main?
-            'filename': os.path.splitext(url_basename(api_video_url))[0],
-        }
-
-        qd_items = compat_parse_qs(compat_urllib_parse_urlparse(api_video_url).query)
-        for key, val in qd_items.items():
-            auth_params[key] = val[0]
-
-        auth_req = sanitized_Request(
-            'http://api.vip.iqiyi.com/services/ckn.action',
-            urlencode_postdata(auth_params))
-        # iQiyi server throws HTTP 405 error without the following header
-        auth_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        auth_result = self._download_json(
-            auth_req, video_id,
-            note='Downloading video authentication JSON',
-            errnote='Unable to download video authentication JSON')
-
-        code = auth_result.get('code')
-        msg = self.AUTH_API_ERRORS.get(code) or auth_result.get('msg') or code
-        if code == 'Q00506':
-            if do_report_warning:
-                self.report_warning(msg)
-            return False
-        if 'data' not in auth_result:
-            if msg is not None:
-                raise ExtractorError('%s said: %s' % (self.IE_NAME, msg), expected=True)
-            raise ExtractorError('Unexpected error from Iqiyi auth API')
-
-        return auth_result['data']
-
-    def construct_video_urls(self, data, video_id, _uuid, tvid):
-        def do_xor(x, y):
-            a = y % 3
-            if a == 1:
-                return x ^ 121
-            if a == 2:
-                return x ^ 72
-            return x ^ 103
-
-        def get_encode_code(l):
-            a = 0
-            b = l.split('-')
-            c = len(b)
-            s = ''
-            for i in range(c - 1, -1, -1):
-                a = do_xor(int(b[c - i - 1], 16), i)
-                s += chr(a)
-            return s[::-1]
-
-        def get_path_key(x, format_id, segment_index):
-            mg = ')(*&^flash@#$%a'
-            tm = self._download_json(
-                'http://data.video.qiyi.com/t?tn=' + str(random.random()), video_id,
-                note='Download path key of segment %d for format %s' % (segment_index + 1, format_id)
-            )['t']
-            t = str(int(math.floor(int(tm) / (600.0))))
-            return md5_text(t + mg + x)
-
-        video_urls_dict = {}
-        need_vip_warning_report = True
-        for format_item in data['vp']['tkl'][0]['vs']:
-            if 0 < int(format_item['bid']) <= 10:
-                format_id = self.get_format(format_item['bid'])
-            else:
-                continue
-
-            video_urls = []
-
-            video_urls_info = format_item['fs']
-            if not format_item['fs'][0]['l'].startswith('/'):
-                t = get_encode_code(format_item['fs'][0]['l'])
-                if t.endswith('mp4'):
-                    video_urls_info = format_item['flvs']
-
-            for segment_index, segment in enumerate(video_urls_info):
-                vl = segment['l']
-                if not vl.startswith('/'):
-                    vl = get_encode_code(vl)
-                is_vip_video = '/vip/' in vl
-                filesize = segment['b']
-                base_url = data['vp']['du'].split('/')
-                if not is_vip_video:
-                    key = get_path_key(
-                        vl.split('/')[-1].split('.')[0], format_id, segment_index)
-                    base_url.insert(-1, key)
-                base_url = '/'.join(base_url)
-                param = {
-                    'su': _uuid,
-                    'qyid': uuid.uuid4().hex,
-                    'client': '',
-                    'z': '',
-                    'bt': '',
-                    'ct': '',
-                    'tn': str(int(time.time()))
-                }
-                api_video_url = base_url + vl
-                if is_vip_video:
-                    api_video_url = api_video_url.replace('.f4v', '.hml')
-                    auth_result = self._authenticate_vip_video(
-                        api_video_url, video_id, tvid, _uuid, need_vip_warning_report)
-                    if auth_result is False:
-                        need_vip_warning_report = False
-                        break
-                    param.update({
-                        't': auth_result['t'],
-                        # cid is hard-coded in com/qiyi/player/core/player/RuntimeData.as
-                        'cid': 'afbe8fd3d73448c9',
-                        'vid': video_id,
-                        'QY00001': auth_result['u'],
-                    })
-                api_video_url += '?' if '?' not in api_video_url else '&'
-                api_video_url += compat_urllib_parse_urlencode(param)
-                js = self._download_json(
-                    api_video_url, video_id,
-                    note='Download video info of segment %d for format %s' % (segment_index + 1, format_id))
-                video_url = js['l']
-                video_urls.append(
-                    (video_url, filesize))
-
-            video_urls_dict[format_id] = video_urls
-        return video_urls_dict
-
-    def get_format(self, bid):
-        matched_format_ids = [_format_id for _bid, _format_id in self._FORMATS_MAP if _bid == str(bid)]
-        return matched_format_ids[0] if len(matched_format_ids) else None
-
-    def get_bid(self, format_id):
-        matched_bids = [_bid for _bid, _format_id in self._FORMATS_MAP if _format_id == format_id]
-        return matched_bids[0] if len(matched_bids) else None
-
-    def get_raw_data(self, tvid, video_id, enc_key, _uuid):
-        tm = str(int(time.time()))
-        tail = tm + tvid
-        param = {
-            'key': 'fvip',
-            'src': md5_text('youtube-dl'),
-            'tvId': tvid,
             'vid': video_id,
-            'vinfo': 1,
+            'src': '76f90cbd92f94a2e925d83e8ccd22cb7',
-            'tm': tm,
+            'sc': sc,
-            'enc': md5_text(enc_key + tail),
+            't': tm,
-            'qyid': _uuid,
-            'tn': random.random(),
-            # In iQiyi's flash player, um is set to 1 if there's a logged user
-            # Some 1080P formats are only available with a logged user.
-            # Here force um=1 to trick the iQiyi server
-            'um': 1,
-            'authkey': md5_text(md5_text('') + tail),
-            'k_tag': 1,
         }
 
-        api_url = 'http://cache.video.qiyi.com/vms' + '?' + \
+        return self._download_json(
-            compat_urllib_parse_urlencode(param)
+            'http://cache.m.iqiyi.com/jp/tmts/%s/%s/' % (tvid, video_id),
-        raw_data = self._download_json(api_url, video_id)
+            video_id, transform_source=lambda s: remove_start(s, 'var tvInfoJs='),
-        return raw_data
+            query=params, headers=self.geo_verification_headers())
-
-    def get_enc_key(self, video_id):
-        # TODO: automatic key extraction
-        # last update at 2016-01-22 for Zombie::bite
-        enc_key = '4a1caba4b4465345366f28da7c117d20'
-        return enc_key
 
     def _extract_playlist(self, webpage):
         PAGE_SIZE = 50
@@ -571,58 +345,41 @@ class IqiyiIE(InfoExtractor):
             r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid')
         video_id = self._search_regex(
             r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
-        _uuid = uuid.uuid4().hex
 
-        enc_key = self.get_enc_key(video_id)
+        formats = []
+        for _ in range(5):
+            raw_data = self.get_raw_data(tvid, video_id)
 
-        raw_data = self.get_raw_data(tvid, video_id, enc_key, _uuid)
+            if raw_data['code'] != 'A00000':
+                if raw_data['code'] == 'A00111':
+                    self.raise_geo_restricted()
+                raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
 
-        if raw_data['code'] != 'A000000':
+            data = raw_data['data']
-            raise ExtractorError('Unable to load data. Error code: ' + raw_data['code'])
 
-        data = raw_data['data']
+            for stream in data['vidl']:
+                if 'm3utx' not in stream:
+                    continue
+                vd = compat_str(stream['vd'])
+                formats.append({
+                    'url': stream['m3utx'],
+                    'format_id': vd,
+                    'ext': 'mp4',
+                    'preference': self._FORMATS_MAP.get(vd, -1),
+                    'protocol': 'm3u8_native',
+                })
 
-        title = data['vi']['vn']
+            if formats:
+                break
 
-        # generate video_urls_dict
+            self._sleep(5, video_id)
-        video_urls_dict = self.construct_video_urls(
-            data, video_id, _uuid, tvid)
 
-        # construct info
+        self._sort_formats(formats)
-        entries = []
+        title = (get_element_by_id('widget-videotitle', webpage) or
-        for format_id in video_urls_dict:
+                 clean_html(get_element_by_attribute('class', 'mod-play-tit', webpage)))
-            video_urls = video_urls_dict[format_id]
-            for i, video_url_info in enumerate(video_urls):
-                if len(entries) < i + 1:
-                    entries.append({'formats': []})
-                entries[i]['formats'].append(
-                    {
-                        'url': video_url_info[0],
-                        'filesize': video_url_info[-1],
-                        'format_id': format_id,
-                        'preference': int(self.get_bid(format_id))
-                    }
-                )
 
-        for i in range(len(entries)):
+        return {
-            self._sort_formats(entries[i]['formats'])
+            'id': video_id,
-            entries[i].update(
+            'title': title,
-                {
+            'formats': formats,
-                    'id': '%s_part%d' % (video_id, i + 1),
+        }
-                    'title': title,
-                }
-            )
-
-        if len(entries) > 1:
-            info = {
-                '_type': 'multi_video',
-                'id': video_id,
-                'title': title,
-                'entries': entries,
-            }
-        else:
-            info = entries[0]
-            info['id'] = video_id
-            info['title'] = title
-
-        return info

youtube_dl/extractor/jwplatform.py

@@ -12,9 +12,35 @@ from ..utils import (
 
 
 class JWPlatformBaseIE(InfoExtractor):
+    @staticmethod
+    def _find_jwplayer_data(webpage):
+        # TODO: Merge this with JWPlayer-related codes in generic.py
+
+        mobj = re.search(
+            'jwplayer\((?P<quote>[\'"])[^\'" ]+(?P=quote)\)\.setup\((?P<options>[^)]+)\)',
+            webpage)
+        if mobj:
+            return mobj.group('options')
+
+    def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs):
+        jwplayer_data = self._parse_json(
+            self._find_jwplayer_data(webpage), video_id)
+        return self._parse_jwplayer_data(
+            jwplayer_data, video_id, *args, **kwargs)
+
     def _parse_jwplayer_data(self, jwplayer_data, video_id, require_title=True, m3u8_id=None, rtmp_params=None):
+        # JWPlayer backward compatibility: flattened playlists
+        # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/api/config.js#L81-L96
+        if 'playlist' not in jwplayer_data:
+            jwplayer_data = {'playlist': [jwplayer_data]}
+
         video_data = jwplayer_data['playlist'][0]
 
+        # JWPlayer backward compatibility: flattened sources
+        # https://github.com/jwplayer/jwplayer/blob/v7.4.3/src/js/playlist/item.js#L29-L35
+        if 'sources' not in video_data:
+            video_data['sources'] = [video_data]
+
         formats = []
         for source in video_data['sources']:
             source_url = self._proto_relative_url(source['file'])

youtube_dl/extractor/kaltura.py

@@ -6,7 +6,6 @@ import base64
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_urllib_parse_urlencode,
     compat_urlparse,
     compat_parse_qs,
 )
@@ -15,6 +14,7 @@ from ..utils import (
     ExtractorError,
     int_or_none,
     unsmuggle_url,
+    smuggle_url,
 )
 
 
@@ -34,7 +34,8 @@ class KalturaIE(InfoExtractor):
                         )(?:/(?P<path>[^?]+))?(?:\?(?P<query>.*))?
                 )
                 '''
-    _API_BASE = 'http://cdnapi.kaltura.com/api_v3/index.php?'
+    _SERVICE_URL = 'http://cdnapi.kaltura.com'
+    _SERVICE_BASE = '/api_v3/index.php'
     _TESTS = [
         {
             'url': 'kaltura:269692:1_1jc2y3e4',
@@ -64,16 +65,50 @@ class KalturaIE(InfoExtractor):
         }
     ]
 
-    def _kaltura_api_call(self, video_id, actions, *args, **kwargs):
+    @staticmethod
+    def _extract_url(webpage):
+        mobj = (
+            re.search(
+                r"""(?xs)
+                    kWidget\.(?:thumb)?[Ee]mbed\(
+                    \{.*?
+                        (?P<q1>['\"])wid(?P=q1)\s*:\s*
+                        (?P<q2>['\"])_?(?P<partner_id>[^'\"]+)(?P=q2),.*?
+                        (?P<q3>['\"])entry_?[Ii]d(?P=q3)\s*:\s*
+                        (?P<q4>['\"])(?P<id>[^'\"]+)(?P=q4),
+                """, webpage) or
+            re.search(
+                r'''(?xs)
+                    (?P<q1>["\'])
+                        (?:https?:)?//cdnapi(?:sec)?\.kaltura\.com/.*?(?:p|partner_id)/(?P<partner_id>\d+).*?
+                    (?P=q1).*?
+                    (?:
+                        entry_?[Ii]d|
+                        (?P<q2>["\'])entry_?[Ii]d(?P=q2)
+                    )\s*:\s*
+                    (?P<q3>["\'])(?P<id>.+?)(?P=q3)
+                ''', webpage))
+        if mobj:
+            embed_info = mobj.groupdict()
+            url = 'kaltura:%(partner_id)s:%(id)s' % embed_info
+            escaped_pid = re.escape(embed_info['partner_id'])
+            service_url = re.search(
+                r'<script[^>]+src=["\']((?:https?:)?//.+?)/p/%s/sp/%s00/embedIframeJs' % (escaped_pid, escaped_pid),
+                webpage)
+            if service_url:
+                url = smuggle_url(url, {'service_url': service_url.group(1)})
+            return url
+
+    def _kaltura_api_call(self, video_id, actions, service_url=None, *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)
+        data = self._download_json(
-        url = self._API_BASE + query
+            (service_url or self._SERVICE_URL) + self._SERVICE_BASE,
-        data = self._download_json(url, video_id, *args, **kwargs)
+            video_id, query=params, *args, **kwargs)
 
         status = data if len(actions) == 1 else data[0]
         if status.get('objectType') == 'KalturaAPIException':
@@ -82,7 +117,7 @@ class KalturaIE(InfoExtractor):
 
         return data
 
-    def _get_kaltura_signature(self, video_id, partner_id):
+    def _get_kaltura_signature(self, video_id, partner_id, service_url=None):
         actions = [{
             'apiVersion': '3.1',
             'expiry': 86400,
@@ -92,10 +127,10 @@ class KalturaIE(InfoExtractor):
             'widgetId': '_%s' % partner_id,
         }]
         return self._kaltura_api_call(
-            video_id, actions, note='Downloading Kaltura signature')['ks']
+            video_id, actions, service_url, note='Downloading Kaltura signature')['ks']
 
-    def _get_video_info(self, video_id, partner_id):
+    def _get_video_info(self, video_id, partner_id, service_url=None):
-        signature = self._get_kaltura_signature(video_id, partner_id)
+        signature = self._get_kaltura_signature(video_id, partner_id, service_url)
         actions = [
             {
                 'action': 'null',
@@ -118,7 +153,7 @@ class KalturaIE(InfoExtractor):
             },
         ]
         return self._kaltura_api_call(
-            video_id, actions, note='Downloading video info JSON')
+            video_id, actions, service_url, note='Downloading video info JSON')
 
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
@@ -127,7 +162,7 @@ class KalturaIE(InfoExtractor):
         partner_id, entry_id = mobj.group('partner_id', 'id')
         ks = None
         if partner_id and entry_id:
-            info, flavor_assets = self._get_video_info(entry_id, partner_id)
+            info, flavor_assets = self._get_video_info(entry_id, partner_id, smuggled_data.get('service_url'))
         else:
             path, query = mobj.group('path', 'query')
             if not path and not query:
@@ -175,12 +210,17 @@ class KalturaIE(InfoExtractor):
                 unsigned_url += '?referrer=%s' % referrer
             return unsigned_url
 
+        data_url = info['dataUrl']
+        if '/flvclipper/' in data_url:
+            data_url = re.sub(r'/flvclipper/.*', '/serveFlavor', data_url)
+
         formats = []
         for f in flavor_assets:
             # Continue if asset is not ready
             if f['status'] != 2:
                 continue
-            video_url = sign_url('%s/flavorId/%s' % (info['dataUrl'], f['id']))
+            video_url = sign_url(
+                '%s/flavorId/%s' % (data_url, f['id']))
             formats.append({
                 'format_id': '%(fileExt)s-%(bitrate)s' % f,
                 'ext': f.get('fileExt'),
@@ -193,9 +233,12 @@ class KalturaIE(InfoExtractor):
                 'width': int_or_none(f.get('width')),
                 'url': video_url,
             })
-        m3u8_url = sign_url(info['dataUrl'].replace('format/url', 'format/applehttp'))
+        if '/playManifest/' in data_url:
-        formats.extend(self._extract_m3u8_formats(
+            m3u8_url = sign_url(data_url.replace(
-            m3u8_url, entry_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+                'format/url', 'format/applehttp'))
+            formats.extend(self._extract_m3u8_formats(
+                m3u8_url, entry_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False))
 
         self._check_formats(formats, entry_id)
         self._sort_formats(formats)

youtube_dl/extractor/kamcord.py

@@ -0,0 +1,71 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    int_or_none,
+    qualities,
+)
+
+
+class KamcordIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?kamcord\.com/v/(?P<id>[^/?#&]+)'
+    _TEST = {
+        'url': 'https://www.kamcord.com/v/hNYRduDgWb4',
+        'md5': 'c3180e8a9cfac2e86e1b88cb8751b54c',
+        'info_dict': {
+            'id': 'hNYRduDgWb4',
+            'ext': 'mp4',
+            'title': 'Drinking Madness',
+            'uploader': 'jacksfilms',
+            'uploader_id': '3044562',
+            'view_count': int,
+            'like_count': int,
+            'comment_count': int,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        video = self._parse_json(
+            self._search_regex(
+                r'window\.__props\s*=\s*({.+?});?(?:\n|\s*</script)',
+                webpage, 'video'),
+            video_id)['video']
+
+        title = video['title']
+
+        formats = self._extract_m3u8_formats(
+            video['play']['hls'], video_id, 'mp4', entry_protocol='m3u8_native')
+        self._sort_formats(formats)
+
+        uploader = video.get('user', {}).get('username')
+        uploader_id = video.get('user', {}).get('id')
+
+        view_count = int_or_none(video.get('viewCount'))
+        like_count = int_or_none(video.get('heartCount'))
+        comment_count = int_or_none(video.get('messageCount'))
+
+        preference_key = qualities(('small', 'medium', 'large'))
+
+        thumbnails = [{
+            'url': thumbnail_url,
+            'id': thumbnail_id,
+            'preference': preference_key(thumbnail_id),
+        } for thumbnail_id, thumbnail_url in (video.get('thumbnail') or {}).items()
+            if isinstance(thumbnail_id, compat_str) and isinstance(thumbnail_url, compat_str)]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'uploader': uploader,
+            'uploader_id': uploader_id,
+            'view_count': view_count,
+            'like_count': like_count,
+            'comment_count': comment_count,
+            'thumbnails': thumbnails,
+            'formats': formats,
+        }

youtube_dl/extractor/kuwo.py

@@ -26,11 +26,6 @@ class KuwoBaseIE(InfoExtractor):
     def _get_formats(self, song_id, tolerate_ip_deny=False):
         formats = []
         for file_format in self._FORMATS:
-            headers = {}
-            cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
-            if cn_verification_proxy:
-                headers['Ytdl-request-proxy'] = cn_verification_proxy
-
             query = {
                 'format': file_format['ext'],
                 'br': file_format.get('br', ''),
@@ -42,7 +37,7 @@ class KuwoBaseIE(InfoExtractor):
             song_url = self._download_webpage(
                 'http://antiserver.kuwo.cn/anti.s',
                 song_id, note='Download %s url info' % file_format['format'],
-                query=query, headers=headers,
+                query=query, headers=self.geo_verification_headers(),
             )
 
             if song_url == 'IPDeny' and not tolerate_ip_deny:
@@ -148,8 +143,8 @@ class KuwoAlbumIE(InfoExtractor):
         'url': 'http://www.kuwo.cn/album/502294/',
         'info_dict': {
             'id': '502294',
-            'title': 'M',
+            'title': 'Made\xa0Series\xa0《M》',
-            'description': 'md5:6a7235a84cc6400ec3b38a7bdaf1d60c',
+            'description': 'md5:d463f0d8a0ff3c3ea3d6ed7452a9483f',
         },
         'playlist_count': 2,
     }
@@ -209,7 +204,7 @@ class KuwoSingerIE(InfoExtractor):
         'url': 'http://www.kuwo.cn/mingxing/bruno+mars/',
         'info_dict': {
             'id': 'bruno+mars',
-            'title': 'Bruno Mars',
+            'title': 'Bruno\xa0Mars',
         },
         'playlist_mincount': 329,
     }, {
@@ -306,7 +301,7 @@ class KuwoMvIE(KuwoBaseIE):
             'id': '6480076',
             'ext': 'mp4',
             'title': 'My HouseMV',
-            'creator': '2PM',
+            'creator': 'PM02:00',
         },
         # In this video, music URLs (anti.s) are blocked outside China and
         # USA, while the MV URL (mvurl) is available globally, so force the MV

youtube_dl/extractor/la7.py

@@ -1,60 +1,65 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..utils import (
-    parse_duration,
+    js_to_json,
+    smuggle_url,
 )
 
 
 class LA7IE(InfoExtractor):
-    IE_NAME = 'la7.tv'
+    IE_NAME = 'la7.it'
-    _VALID_URL = r'''(?x)
+    _VALID_URL = r'''(?x)(https?://)?(?:
-        https?://(?:www\.)?la7\.tv/
+        (?:www\.)?la7\.it/([^/]+)/(?:rivedila7|video)/|
-        (?:
+        tg\.la7\.it/repliche-tgla7\?id=
-            richplayer/\?assetid=|
+    )(?P<id>.+)'''
-            \?contentId=
-        )
-        (?P<id>[0-9]+)'''
 
-    _TEST = {
+    _TESTS = [{
-        'url': 'http://www.la7.tv/richplayer/?assetid=50355319',
+        # 'src' is a plain URL
-        'md5': 'ec7d1f0224d20ba293ab56cf2259651f',
+        'url': 'http://www.la7.it/crozza/video/inccool8-02-10-2015-163722',
+        'md5': '8b613ffc0c4bf9b9e377169fc19c214c',
         'info_dict': {
-            'id': '50355319',
+            'id': 'inccool8-02-10-2015-163722',
             'ext': 'mp4',
-            'title': 'IL DIVO',
+            'title': 'Inc.Cool8',
-            'description': 'Un film di Paolo Sorrentino con Toni Servillo, Anna Bonaiuto, Giulio Bosetti  e Flavio Bucci',
+            'description': 'Benvenuti nell\'incredibile mondo della INC. COOL. 8. dove “INC.” sta per “Incorporated” “COOL” sta per “fashion” ed Eight sta per il gesto  atletico',
-            'duration': 6254,
+            'thumbnail': 're:^https?://.*',
+            'uploader_id': 'kdla7pillole@iltrovatore.it',
+            'timestamp': 1443814869,
+            'upload_date': '20151002',
         },
-        'skip': 'Blocked in the US',
+    }, {
-    }
+        # 'src' is a dictionary
+        'url': 'http://tg.la7.it/repliche-tgla7?id=189080',
+        'md5': '6b0d8888d286e39870208dfeceaf456b',
+        'info_dict': {
+            'id': '189080',
+            'ext': 'mp4',
+            'title': 'TG LA7',
+        },
+    }, {
+        'url': 'http://www.la7.it/omnibus/rivedila7/omnibus-news-02-07-2016-189077',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        xml_url = 'http://www.la7.tv/repliche/content/index.php?contentId=%s' % video_id
-        doc = self._download_xml(xml_url, video_id)
 
-        video_title = doc.find('title').text
+        webpage = self._download_webpage(url, video_id)
-        description = doc.find('description').text
-        duration = parse_duration(doc.find('duration').text)
-        thumbnail = doc.find('img').text
-        view_count = int(doc.find('views').text)
 
-        prefix = doc.find('.//fqdn').text.strip().replace('auto:', 'http:')
+        player_data = self._parse_json(
-
+            self._search_regex(r'videoLa7\(({[^;]+})\);', webpage, 'player data'),
-        formats = [{
+            video_id, transform_source=js_to_json)
-            'format': vnode.find('quality').text,
-            'tbr': int(vnode.find('quality').text),
-            'url': vnode.find('fms').text.strip().replace('mp4:', prefix),
-        } for vnode in doc.findall('.//videos/video')]
-        self._sort_formats(formats)
 
         return {
+            '_type': 'url_transparent',
+            'url': smuggle_url('kaltura:103:%s' % player_data['vid'], {
+                'service_url': 'http://kdam.iltrovatore.it',
+            }),
             'id': video_id,
-            'title': video_title,
+            'title': player_data['title'],
-            'description': description,
+            'description': self._og_search_description(webpage, default=None),
-            'thumbnail': thumbnail,
+            'thumbnail': player_data.get('poster'),
-            'duration': duration,
+            'ie_key': 'Kaltura',
-            'formats': formats,
-            'view_count': view_count,
         }

youtube_dl/extractor/leeco.py

@@ -20,15 +20,16 @@ from ..utils import (
     int_or_none,
     orderedSet,
     parse_iso8601,
-    sanitized_Request,
     str_or_none,
     url_basename,
+    urshift,
+    update_url_query,
 )
 
 
 class LeIE(InfoExtractor):
     IE_DESC = '乐视网'
-    _VALID_URL = r'https?://www\.le\.com/ptv/vplay/(?P<id>\d+)\.html'
+    _VALID_URL = r'https?://(?:www\.le\.com/ptv/vplay|sports\.le\.com/video)/(?P<id>\d+)\.html'
 
     _URL_TEMPLATE = 'http://www.le.com/ptv/vplay/%s.html'
 
@@ -69,17 +70,16 @@ class LeIE(InfoExtractor):
             'hls_prefer_native': True,
         },
         'skip': 'Only available in China',
+    }, {
+        'url': 'http://sports.le.com/video/25737697.html',
+        'only_matching': True,
     }]
 
-    @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)
+            param1 = urshift(param1, 1) + ((param1 & 1) << 31)
             _loc3_ += 1
         return param1
 
@@ -90,6 +90,10 @@ class LeIE(InfoExtractor):
         _loc3_ = self.ror(_loc3_, _loc2_ % 17)
         return _loc3_
 
+    # reversed from http://jstatic.letvcdn.com/sdk/player.js
+    def get_mms_key(self, time):
+        return self.ror(time, 8) ^ 185025305
+
     # see M3U8Encryption class in KLetvPlayer.swf
     @staticmethod
     def decrypt_m3u8(encrypted_data):
@@ -110,28 +114,7 @@ class LeIE(InfoExtractor):
 
         return bytes(_loc7_)
 
-    def _real_extract(self, url):
+    def _check_errors(self, play_json):
-        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.le.com'
-        }
-        play_json_req = sanitized_Request(
-            'http://api.le.com/mms/out/video/playJson?' + compat_urllib_parse_urlencode(params)
-        )
-        cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
-        if cn_verification_proxy:
-            play_json_req.add_header('Ytdl-request-proxy', cn_verification_proxy)
-
-        play_json = self._download_json(
-            play_json_req,
-            media_id, 'Downloading playJson data')
-
         # Check for errors
         playstatus = play_json['playstatus']
         if playstatus['status'] == 0:
@@ -142,43 +125,99 @@ class LeIE(InfoExtractor):
                 msg = 'Generic error. flag = %d' % flag
             raise ExtractorError(msg, expected=True)
 
-        playurl = play_json['playurl']
+    def _real_extract(self, url):
+        media_id = self._match_id(url)
+        page = self._download_webpage(url, media_id)
 
-        formats = ['350', '1000', '1300', '720p', '1080p']
+        play_json_h5 = self._download_json(
-        dispatch = playurl['dispatch']
+            'http://api.le.com/mms/out/video/playJsonH5',
+            media_id, 'Downloading html5 playJson data', query={
+                'id': media_id,
+                'platid': 3,
+                'splatid': 304,
+                'format': 1,
+                'tkey': self.get_mms_key(int(time.time())),
+                'domain': 'www.le.com',
+                'tss': 'no',
+            },
+            headers=self.geo_verification_headers())
+        self._check_errors(play_json_h5)
 
-        urls = []
+        play_json_flash = self._download_json(
-        for format_id in formats:
+            'http://api.le.com/mms/out/video/playJson',
-            if format_id in dispatch:
+            media_id, 'Downloading flash playJson data', query={
-                media_url = playurl['domain'][0] + dispatch[format_id][0]
+                'id': media_id,
-                media_url += '&' + compat_urllib_parse_urlencode({
+                'platid': 1,
-                    'm3v': 1,
+                'splatid': 101,
+                'format': 1,
+                'tkey': self.calc_time_key(int(time.time())),
+                'domain': 'www.le.com',
+            },
+            headers=self.geo_verification_headers())
+        self._check_errors(play_json_flash)
+
+        def get_h5_urls(media_url, format_id):
+            location = self._download_json(
+                media_url, media_id,
+                'Download JSON metadata for format %s' % format_id, query={
                     'format': 1,
                     'expect': 3,
-                    'rateid': format_id,
+                    'tss': 'no',
-                })
+                })['location']
 
-                nodes_data = self._download_json(
+            return {
-                    media_url, media_id,
+                'http': update_url_query(location, {'tss': 'no'}),
-                    'Download JSON metadata for format %s' % format_id)
+                'hls': update_url_query(location, {'tss': 'ios'}),
+            }
 
-                req = self._request_webpage(
+        def get_flash_urls(media_url, format_id):
-                    nodes_data['nodelist'][0]['location'], media_id,
+            media_url += '&' + compat_urllib_parse_urlencode({
-                    note='Downloading m3u8 information for format %s' % format_id)
+                'm3v': 1,
+                'format': 1,
+                'expect': 3,
+                'rateid': format_id,
+            })
 
-                m3u8_data = self.decrypt_m3u8(req.read())
+            nodes_data = self._download_json(
+                media_url, media_id,
+                'Download JSON metadata for format %s' % format_id)
 
-                url_info_dict = {
+            req = self._request_webpage(
-                    'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
+                nodes_data['nodelist'][0]['location'], media_id,
-                    'ext': determine_ext(dispatch[format_id][1]),
+                note='Downloading m3u8 information for format %s' % format_id)
-                    'format_id': format_id,
-                    'protocol': 'm3u8',
-                }
 
-                if format_id[-1:] == 'p':
+            m3u8_data = self.decrypt_m3u8(req.read())
-                    url_info_dict['height'] = int_or_none(format_id[:-1])
 
-                urls.append(url_info_dict)
+            return {
+                'hls': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
+            }
+
+        extracted_formats = []
+        formats = []
+        for play_json, get_urls in ((play_json_h5, get_h5_urls), (play_json_flash, get_flash_urls)):
+            playurl = play_json['playurl']
+            play_domain = playurl['domain'][0]
+
+            for format_id, format_data in playurl.get('dispatch', []).items():
+                if format_id in extracted_formats:
+                    continue
+                extracted_formats.append(format_id)
+
+                media_url = play_domain + format_data[0]
+                for protocol, format_url in get_urls(media_url, format_id).items():
+                    f = {
+                        'url': format_url,
+                        'ext': determine_ext(format_data[1]),
+                        'format_id': '%s-%s' % (protocol, format_id),
+                        'protocol': 'm3u8_native' if protocol == 'hls' else 'http',
+                        'quality': int_or_none(format_id),
+                    }
+
+                    if format_id[-1:] == 'p':
+                        f['height'] = int_or_none(format_id[:-1])
+
+                    formats.append(f)
+        self._sort_formats(formats, ('height', 'quality', 'format_id'))
 
         publish_time = parse_iso8601(self._html_search_regex(
             r'发布时间&nbsp;([^<>]+) ', page, 'publish time', default=None),
@@ -187,7 +226,7 @@ class LeIE(InfoExtractor):
 
         return {
             'id': media_id,
-            'formats': urls,
+            'formats': formats,
             'title': playurl['title'],
             'thumbnail': playurl['pic'],
             'description': description,
@@ -196,7 +235,7 @@ class LeIE(InfoExtractor):
 
 
 class LePlaylistIE(InfoExtractor):
-    _VALID_URL = r'https?://[a-z]+\.le\.com/[a-z]+/(?P<id>[a-z0-9_]+)'
+    _VALID_URL = r'https?://[a-z]+\.le\.com/(?!video)[a-z]+/(?P<id>[a-z0-9_]+)'
 
     _TESTS = [{
         'url': 'http://www.le.com/tv/46177.html',

youtube_dl/extractor/libraryofcongress.py

@@ -1,20 +1,24 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 
 from ..utils import (
     determine_ext,
     float_or_none,
     int_or_none,
+    parse_filesize,
 )
 
 
 class LibraryOfCongressIE(InfoExtractor):
     IE_NAME = 'loc'
     IE_DESC = 'Library of Congress'
-    _VALID_URL = r'https?://(?:www\.)?loc\.gov/item/(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://(?:www\.)?loc\.gov/(?:item/|today/cyberlc/feature_wdesc\.php\?.*\brec=)(?P<id>[0-9]+)'
-    _TEST = {
+    _TESTS = [{
+        # embedded via <div class="media-player"
         'url': 'http://loc.gov/item/90716351/',
         'md5': '353917ff7f0255aa6d4b80a034833de8',
         'info_dict': {
@@ -25,7 +29,35 @@ class LibraryOfCongressIE(InfoExtractor):
             'duration': 0,
             'view_count': int,
         },
-    }
+    }, {
+        # webcast embedded via mediaObjectId
+        'url': 'https://www.loc.gov/today/cyberlc/feature_wdesc.php?rec=5578',
+        'info_dict': {
+            'id': '5578',
+            'ext': 'mp4',
+            'title': 'Help! Preservation Training Needs Here, There & Everywhere',
+            'duration': 3765,
+            'view_count': int,
+            'subtitles': 'mincount:1',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # with direct download links
+        'url': 'https://www.loc.gov/item/78710669/',
+        'info_dict': {
+            'id': '78710669',
+            'ext': 'mp4',
+            'title': 'La vie et la passion de Jesus-Christ',
+            'duration': 0,
+            'view_count': int,
+            'formats': 'mincount:4',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -34,18 +66,20 @@ class LibraryOfCongressIE(InfoExtractor):
         media_id = self._search_regex(
             (r'id=(["\'])media-player-(?P<id>.+?)\1',
              r'<video[^>]+id=(["\'])uuid-(?P<id>.+?)\1',
-             r'<video[^>]+data-uuid=(["\'])(?P<id>.+?)\1'),
+             r'<video[^>]+data-uuid=(["\'])(?P<id>.+?)\1',
+             r'mediaObjectId\s*:\s*(["\'])(?P<id>.+?)\1'),
             webpage, 'media id', group='id')
 
-        data = self._parse_json(
+        data = self._download_json(
-            self._download_webpage(
+            'https://media.loc.gov/services/v1/media?id=%s&context=json' % media_id,
-                'https://media.loc.gov/services/v1/media?id=%s&context=json' % media_id,
-                video_id),
             video_id)['mediaObject']
 
         derivative = data['derivatives'][0]
         media_url = derivative['derivativeUrl']
 
+        title = derivative.get('shortName') or data.get('shortName') or self._og_search_title(
+            webpage)
+
         # Following algorithm was extracted from setAVSource js function
         # found in webpage
         media_url = media_url.replace('rtmp', 'https')
@@ -61,6 +95,7 @@ class LibraryOfCongressIE(InfoExtractor):
                 'format_id': 'hls',
                 'ext': 'mp4',
                 'protocol': 'm3u8_native',
+                'quality': 1,
             }]
         elif 'vod/mp3:' in media_url:
             formats = [{
@@ -68,17 +103,41 @@ class LibraryOfCongressIE(InfoExtractor):
                 'vcodec': 'none',
             }]
 
+        download_urls = set()
+        for m in re.finditer(
+                r'<option[^>]+value=(["\'])(?P<url>.+?)\1[^>]+data-file-download=[^>]+>\s*(?P<id>.+?)(?:(?:&nbsp;|\s+)\((?P<size>.+?)\))?\s*<', webpage):
+            format_id = m.group('id').lower()
+            if format_id == 'gif':
+                continue
+            download_url = m.group('url')
+            if download_url in download_urls:
+                continue
+            download_urls.add(download_url)
+            formats.append({
+                'url': download_url,
+                'format_id': format_id,
+                'filesize_approx': parse_filesize(m.group('size')),
+            })
+
         self._sort_formats(formats)
 
-        title = derivative.get('shortName') or data.get('shortName') or self._og_search_title(webpage)
         duration = float_or_none(data.get('duration'))
         view_count = int_or_none(data.get('viewCount'))
 
+        subtitles = {}
+        cc_url = data.get('ccUrl')
+        if cc_url:
+            subtitles.setdefault('en', []).append({
+                'url': cc_url,
+                'ext': 'ttml',
+            })
+
         return {
             'id': video_id,
             'title': title,
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage, default=None),
             'duration': duration,
             'view_count': view_count,
             'formats': formats,
+            'subtitles': subtitles,
         }

youtube_dl/extractor/limelight.py

@@ -98,13 +98,19 @@ class LimelightBaseIE(InfoExtractor):
         } for thumbnail in properties.get('thumbnails', []) if thumbnail.get('url')]
 
         subtitles = {}
-        for caption in properties.get('captions', {}):
+        for caption in properties.get('captions', []):
             lang = caption.get('language_code')
             subtitles_url = caption.get('url')
             if lang and subtitles_url:
-                subtitles[lang] = [{
+                subtitles.setdefault(lang, []).append({
                     'url': subtitles_url,
-                }]
+                })
+        closed_captions_url = properties.get('closed_captions_url')
+        if closed_captions_url:
+            subtitles.setdefault('en', []).append({
+                'url': closed_captions_url,
+                'ext': 'ttml',
+            })
 
         return {
             'id': video_id,
@@ -123,7 +129,18 @@ class LimelightBaseIE(InfoExtractor):
 
 class LimelightMediaIE(LimelightBaseIE):
     IE_NAME = 'limelight'
-    _VALID_URL = r'(?:limelight:media:|https?://link\.videoplatform\.limelight\.com/media/\??\bmediaId=)(?P<id>[a-z0-9]{32})'
+    _VALID_URL = r'''(?x)
+                        (?:
+                            limelight:media:|
+                            https?://
+                                (?:
+                                    link\.videoplatform\.limelight\.com/media/|
+                                    assets\.delvenetworks\.com/player/loader\.swf
+                                )
+                                \?.*?\bmediaId=
+                        )
+                        (?P<id>[a-z0-9]{32})
+                    '''
     _TESTS = [{
         'url': 'http://link.videoplatform.limelight.com/media/?mediaId=3ffd040b522b4485b6d84effc750cd86',
         'info_dict': {
@@ -158,6 +175,9 @@ class LimelightMediaIE(LimelightBaseIE):
             # rtmp download
             'skip_download': True,
         },
+    }, {
+        'url': 'https://assets.delvenetworks.com/player/loader.swf?mediaId=8018a574f08d416e95ceaccae4ba0452',
+        'only_matching': True,
     }]
     _PLAYLIST_SERVICE_PATH = 'media'
     _API_PATH = 'media'
@@ -176,15 +196,29 @@ class LimelightMediaIE(LimelightBaseIE):
 
 class LimelightChannelIE(LimelightBaseIE):
     IE_NAME = 'limelight:channel'
-    _VALID_URL = r'(?:limelight:channel:|https?://link\.videoplatform\.limelight\.com/media/\??\bchannelId=)(?P<id>[a-z0-9]{32})'
+    _VALID_URL = r'''(?x)
-    _TEST = {
+                        (?:
+                            limelight:channel:|
+                            https?://
+                                (?:
+                                    link\.videoplatform\.limelight\.com/media/|
+                                    assets\.delvenetworks\.com/player/loader\.swf
+                                )
+                                \?.*?\bchannelId=
+                        )
+                        (?P<id>[a-z0-9]{32})
+                    '''
+    _TESTS = [{
         'url': 'http://link.videoplatform.limelight.com/media/?channelId=ab6a524c379342f9b23642917020c082',
         'info_dict': {
             'id': 'ab6a524c379342f9b23642917020c082',
             'title': 'Javascript Sample Code',
         },
         'playlist_mincount': 3,
-    }
+    }, {
+        'url': 'http://assets.delvenetworks.com/player/loader.swf?channelId=ab6a524c379342f9b23642917020c082',
+        'only_matching': True,
+    }]
     _PLAYLIST_SERVICE_PATH = 'channel'
     _API_PATH = 'channels'
 
@@ -207,15 +241,29 @@ class LimelightChannelIE(LimelightBaseIE):
 
 class LimelightChannelListIE(LimelightBaseIE):
     IE_NAME = 'limelight:channel_list'
-    _VALID_URL = r'(?:limelight:channel_list:|https?://link\.videoplatform\.limelight\.com/media/\?.*?\bchannelListId=)(?P<id>[a-z0-9]{32})'
+    _VALID_URL = r'''(?x)
-    _TEST = {
+                        (?:
+                            limelight:channel_list:|
+                            https?://
+                                (?:
+                                    link\.videoplatform\.limelight\.com/media/|
+                                    assets\.delvenetworks\.com/player/loader\.swf
+                                )
+                                \?.*?\bchannelListId=
+                        )
+                        (?P<id>[a-z0-9]{32})
+                    '''
+    _TESTS = [{
         'url': 'http://link.videoplatform.limelight.com/media/?channelListId=301b117890c4465c8179ede21fd92e2b',
         'info_dict': {
             'id': '301b117890c4465c8179ede21fd92e2b',
             'title': 'Website - Hero Player',
         },
         'playlist_mincount': 2,
-    }
+    }, {
+        'url': 'https://assets.delvenetworks.com/player/loader.swf?channelListId=301b117890c4465c8179ede21fd92e2b',
+        'only_matching': True,
+    }]
     _PLAYLIST_SERVICE_PATH = 'channel_list'
 
     def _real_extract(self, url):

youtube_dl/extractor/livestream.py

@@ -203,9 +203,10 @@ class LivestreamIE(InfoExtractor):
             if not videos_info:
                 break
             for v in videos_info:
+                v_id = compat_str(v['id'])
                 entries.append(self.url_result(
-                    'http://livestream.com/accounts/%s/events/%s/videos/%s' % (account_id, event_id, v['id']),
+                    'http://livestream.com/accounts/%s/events/%s/videos/%s' % (account_id, event_id, v_id),
-                    'Livestream', v['id'], v['caption']))
+                    'Livestream', v_id, v.get('caption')))
             last_video = videos_info[-1]['id']
         return self.playlist_result(entries, event_id, event_data['full_name'])
 

youtube_dl/extractor/lynda.py

@@ -1,106 +1,106 @@
 from __future__ import unicode_literals
 
 import re
-import json
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_HTTPError,
+    compat_str,
+    compat_urlparse,
+)
 from ..utils import (
     ExtractorError,
-    clean_html,
     int_or_none,
-    sanitized_Request,
     urlencode_postdata,
 )
 
 
 class LyndaBaseIE(InfoExtractor):
-    _LOGIN_URL = 'https://www.lynda.com/login/login.aspx'
+    _SIGNIN_URL = 'https://www.lynda.com/signin'
+    _PASSWORD_URL = 'https://www.lynda.com/signin/password'
+    _USER_URL = 'https://www.lynda.com/signin/user'
     _ACCOUNT_CREDENTIALS_HINT = 'Use --username and --password options to provide lynda.com account credentials.'
     _NETRC_MACHINE = 'lynda'
 
     def _real_initialize(self):
         self._login()
 
+    @staticmethod
+    def _check_error(json_string, key_or_keys):
+        keys = [key_or_keys] if isinstance(key_or_keys, compat_str) else key_or_keys
+        for key in keys:
+            error = json_string.get(key)
+            if error:
+                raise ExtractorError('Unable to login: %s' % error, expected=True)
+
+    def _login_step(self, form_html, fallback_action_url, extra_form_data, note, referrer_url):
+        action_url = self._search_regex(
+            r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_html,
+            'post url', default=fallback_action_url, group='url')
+
+        if not action_url.startswith('http'):
+            action_url = compat_urlparse.urljoin(self._SIGNIN_URL, action_url)
+
+        form_data = self._hidden_inputs(form_html)
+        form_data.update(extra_form_data)
+
+        try:
+            response = self._download_json(
+                action_url, None, note,
+                data=urlencode_postdata(form_data),
+                headers={
+                    'Referer': referrer_url,
+                    'X-Requested-With': 'XMLHttpRequest',
+                })
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 500:
+                response = self._parse_json(e.cause.read().decode('utf-8'), None)
+                self._check_error(response, ('email', 'password'))
+            raise
+
+        self._check_error(response, 'ErrorMessage')
+
+        return response, action_url
+
     def _login(self):
         username, password = self._get_login_info()
         if username is None:
             return
 
-        login_form = {
+        # Step 1: download signin page
-            'username': username,
+        signin_page = self._download_webpage(
-            'password': password,
+            self._SIGNIN_URL, None, 'Downloading signin page')
-            'remember': 'false',
-            'stayPut': 'false'
-        }
-        request = sanitized_Request(
-            self._LOGIN_URL, urlencode_postdata(login_form))
-        login_page = self._download_webpage(
-            request, None, 'Logging in as %s' % username)
 
-        # Not (yet) logged in
+        # Already logged in
-        m = re.search(r'loginResultJson\s*=\s*\'(?P<json>[^\']+)\';', login_page)
+        if any(re.search(p, signin_page) for p in (
-        if m is not None:
+                'isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')):
-            response = m.group('json')
-            response_json = json.loads(response)
-            state = response_json['state']
-
-            if state == 'notlogged':
-                raise ExtractorError(
-                    'Unable to login, incorrect username and/or password',
-                    expected=True)
-
-            # This is when we get popup:
-            # > You're already logged in to lynda.com on two devices.
-            # > If you log in here, we'll log you out of another device.
-            # So, we need to confirm this.
-            if state == 'conflicted':
-                confirm_form = {
-                    'username': '',
-                    'password': '',
-                    'resolve': 'true',
-                    'remember': 'false',
-                    'stayPut': 'false',
-                }
-                request = sanitized_Request(
-                    self._LOGIN_URL, urlencode_postdata(confirm_form))
-                login_page = self._download_webpage(
-                    request, None,
-                    'Confirming log in and log out from another device')
-
-        if all(not re.search(p, login_page) for p in ('isLoggedIn\s*:\s*true', r'logout\.aspx', r'>Log out<')):
-            if 'login error' in login_page:
-                mobj = re.search(
-                    r'(?s)<h1[^>]+class="topmost">(?P<title>[^<]+)</h1>\s*<div>(?P<description>.+?)</div>',
-                    login_page)
-                if mobj:
-                    raise ExtractorError(
-                        'lynda returned error: %s - %s'
-                        % (mobj.group('title'), clean_html(mobj.group('description'))),
-                        expected=True)
-            raise ExtractorError('Unable to log in')
-
-    def _logout(self):
-        username, _ = self._get_login_info()
-        if username is None:
             return
 
-        self._download_webpage(
+        # Step 2: submit email
-            'http://www.lynda.com/ajax/logout.aspx', None,
+        signin_form = self._search_regex(
-            'Logging out', 'Unable to log out', fatal=False)
+            r'(?s)(<form[^>]+data-form-name=["\']signin["\'][^>]*>.+?</form>)',
+            signin_page, 'signin form')
+        signin_page, signin_url = self._login_step(
+            signin_form, self._PASSWORD_URL, {'email': username},
+            'Submitting email', self._SIGNIN_URL)
+
+        # Step 3: submit password
+        password_form = signin_page['body']
+        self._login_step(
+            password_form, self._USER_URL, {'email': username, 'password': password},
+            'Submitting password', signin_url)
 
 
 class LyndaIE(LyndaBaseIE):
     IE_NAME = 'lynda'
     IE_DESC = 'lynda.com videos'
     _VALID_URL = r'https?://www\.lynda\.com/(?:[^/]+/[^/]+/\d+|player/embed)/(?P<id>\d+)'
-    _NETRC_MACHINE = 'lynda'
 
     _TIMECODE_REGEX = r'\[(?P<timecode>\d+:\d+:\d+[\.,]\d+)\]'
 
     _TESTS = [{
         'url': 'http://www.lynda.com/Bootstrap-tutorials/Using-exercise-files/110885/114408-4.html',
-        'md5': 'ecfc6862da89489161fb9cd5f5a6fac1',
+        # md5 is unstable
         'info_dict': {
             'id': '114408',
             'ext': 'mp4',
@@ -212,8 +212,6 @@ class LyndaCourseIE(LyndaBaseIE):
             'http://www.lynda.com/ajax/player?courseId=%s&type=course' % course_id,
             course_id, 'Downloading course JSON')
 
-        self._logout()
-
         if course.get('Status') == 'NotFound':
             raise ExtractorError(
                 'Course %s does not exist' % course_id, expected=True)
@@ -246,5 +244,6 @@ class LyndaCourseIE(LyndaBaseIE):
                 % unaccessible_videos + self._ACCOUNT_CREDENTIALS_HINT)
 
         course_title = course.get('Title')
+        course_description = course.get('Description')
 
-        return self.playlist_result(entries, course_id, course_title)
+        return self.playlist_result(entries, course_id, course_title, course_description)

youtube_dl/extractor/m6.py

@@ -1,8 +1,6 @@
 # encoding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 
 
@@ -23,34 +21,5 @@ class M6IE(InfoExtractor):
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        video_id = self._match_id(url)
-        video_id = mobj.group('id')
+        return self.url_result('6play:%s' % video_id, 'SixPlay', video_id)
-
-        rss = self._download_xml('http://ws.m6.fr/v1/video/info/m6/bonus/%s' % video_id, video_id,
-                                 'Downloading video RSS')
-
-        title = rss.find('./channel/item/title').text
-        description = rss.find('./channel/item/description').text
-        thumbnail = rss.find('./channel/item/visuel_clip_big').text
-        duration = int(rss.find('./channel/item/duration').text)
-        view_count = int(rss.find('./channel/item/nombre_vues').text)
-
-        formats = []
-        for format_id in ['lq', 'sd', 'hq', 'hd']:
-            video_url = rss.find('./channel/item/url_video_%s' % format_id)
-            if video_url is None:
-                continue
-            formats.append({
-                'url': video_url.text,
-                'format_id': format_id,
-            })
-
-        return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'view_count': view_count,
-            'formats': formats,
-        }

youtube_dl/extractor/matchtv.py

@@ -4,16 +4,12 @@ from __future__ import unicode_literals
 import random
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_urlencode
+from ..utils import xpath_text
-from ..utils import (
-    sanitized_Request,
-    xpath_text,
-)
 
 
 class MatchTVIE(InfoExtractor):
-    _VALID_URL = r'https?://matchtv\.ru/?#live-player'
+    _VALID_URL = r'https?://matchtv\.ru(?:/on-air|/?#live-player)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://matchtv.ru/#live-player',
         'info_dict': {
             'id': 'matchtv-live',
@@ -24,12 +20,16 @@ class MatchTVIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
-    }
+    }, {
+        'url': 'http://matchtv.ru/on-air/',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = 'matchtv-live'
-        request = sanitized_Request(
+        video_url = self._download_json(
-            'http://player.matchtv.ntvplus.tv/player/smil?%s' % compat_urllib_parse_urlencode({
+            'http://player.matchtv.ntvplus.tv/player/smil', video_id,
+            query={
                 'ts': '',
                 'quality': 'SD',
                 'contentId': '561d2c0df7159b37178b4567',
@@ -40,11 +40,10 @@ class MatchTVIE(InfoExtractor):
                 'contentType': 'channel',
                 'timeShift': '0',
                 'platform': 'portal',
-            }),
+            },
             headers={
                 'Referer': 'http://player.matchtv.ntvplus.tv/embed-player/NTVEmbedPlayer.swf',
-            })
+            })['data']['videoUrl']
-        video_url = self._download_json(request, video_id)['data']['videoUrl']
         f4m_url = xpath_text(self._download_xml(video_url, video_id), './to')
         formats = self._extract_f4m_formats(f4m_url, video_id)
         self._sort_formats(formats)

youtube_dl/extractor/meta.py

@@ -0,0 +1,73 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from .pladform import PladformIE
+from ..utils import (
+    unescapeHTML,
+    int_or_none,
+    ExtractorError,
+)
+
+
+class METAIE(InfoExtractor):
+    _VALID_URL = r'https?://video\.meta\.ua/(?:iframe/)?(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://video.meta.ua/5502115.video',
+        'md5': '71b6f3ee274bef16f1ab410f7f56b476',
+        'info_dict': {
+            'id': '5502115',
+            'ext': 'mp4',
+            'title': 'Sony Xperia Z camera test [HQ]',
+            'description': 'Xperia Z shoots video in FullHD HDR.',
+            'uploader_id': 'nomobile',
+            'uploader': 'CHЁZA.TV',
+            'upload_date': '20130211',
+        },
+        'add_ie': ['Youtube'],
+    }, {
+        'url': 'http://video.meta.ua/iframe/5502115',
+        'only_matching': True,
+    }, {
+        # pladform embed
+        'url': 'http://video.meta.ua/7121015.video',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        st_html5 = self._search_regex(
+            r"st_html5\s*=\s*'#([^']+)'", webpage, 'uppod html5 st', default=None)
+
+        if st_html5:
+            # uppod st decryption algorithm is reverse engineered from function un(s) at uppod.js
+            json_str = ''
+            for i in range(0, len(st_html5), 3):
+                json_str += '&#x0%s;' % st_html5[i:i + 3]
+            uppod_data = self._parse_json(unescapeHTML(json_str), video_id)
+            error = uppod_data.get('customnotfound')
+            if error:
+                raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
+
+            video_url = uppod_data['file']
+            info = {
+                'id': video_id,
+                'url': video_url,
+                'title': uppod_data.get('comment') or self._og_search_title(webpage),
+                'description': self._og_search_description(webpage, default=None),
+                'thumbnail': uppod_data.get('poster') or self._og_search_thumbnail(webpage),
+                'duration': int_or_none(self._og_search_property(
+                    'video:duration', webpage, default=None)),
+            }
+            if 'youtube.com/' in video_url:
+                info.update({
+                    '_type': 'url_transparent',
+                    'ie_key': 'Youtube',
+                })
+            return info
+
+        pladform_url = PladformIE._extract_url(webpage)
+        if pladform_url:
+            return self.url_result(pladform_url)

youtube_dl/extractor/metacafe.py

@@ -11,13 +11,14 @@ from ..utils import (
     determine_ext,
     ExtractorError,
     int_or_none,
-    sanitized_Request,
     urlencode_postdata,
+    get_element_by_attribute,
+    mimetype2ext,
 )
 
 
 class MetacafeIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?metacafe\.com/watch/([^/]+)/([^/]+)/.*'
+    _VALID_URL = r'https?://(?:www\.)?metacafe\.com/watch/(?P<video_id>[^/]+)/(?P<display_id>[^/?#]+)'
     _DISCLAIMER = 'http://www.metacafe.com/family_filter/'
     _FILTER_POST = 'http://www.metacafe.com/f/index.php?inputType=filter&controllerGroup=user'
     IE_NAME = 'metacafe'
@@ -47,6 +48,7 @@ class MetacafeIE(InfoExtractor):
                 'uploader': 'ign',
                 'description': 'Sony released a massive FAQ on the PlayStation Blog detailing the PS4\'s capabilities and limitations.',
             },
+            'skip': 'Page is temporarily unavailable.',
         },
         # AnyClip video
         {
@@ -55,8 +57,8 @@ class MetacafeIE(InfoExtractor):
                 'id': 'an-dVVXnuY7Jh77J',
                 'ext': 'mp4',
                 'title': 'The Andromeda Strain (1971): Stop the Bomb Part 3',
-                'uploader': 'anyclip',
+                'uploader': 'AnyClip',
-                'description': 'md5:38c711dd98f5bb87acf973d573442e67',
+                'description': 'md5:cbef0460d31e3807f6feb4e7a5952e5b',
             },
         },
         # age-restricted video
@@ -110,28 +112,25 @@ class MetacafeIE(InfoExtractor):
     def report_disclaimer(self):
         self.to_screen('Retrieving disclaimer')
 
-    def _real_initialize(self):
+    def _confirm_age(self):
         # Retrieve disclaimer
         self.report_disclaimer()
         self._download_webpage(self._DISCLAIMER, None, False, 'Unable to retrieve disclaimer')
 
         # Confirm age
-        disclaimer_form = {
-            'filters': '0',
-            'submit': "Continue - I'm over 18",
-        }
-        request = sanitized_Request(self._FILTER_POST, urlencode_postdata(disclaimer_form))
-        request.add_header('Content-Type', 'application/x-www-form-urlencoded')
         self.report_age_confirmation()
-        self._download_webpage(request, None, False, 'Unable to confirm age')
+        self._download_webpage(
+            self._FILTER_POST, None, False, 'Unable to confirm age',
+            data=urlencode_postdata({
+                'filters': '0',
+                'submit': "Continue - I'm over 18",
+            }), headers={
+                'Content-Type': 'application/x-www-form-urlencoded',
+            })
 
     def _real_extract(self, url):
         # Extract id and simplified title from URL
-        mobj = re.match(self._VALID_URL, url)
+        video_id, display_id = re.match(self._VALID_URL, url).groups()
-        if mobj is None:
-            raise ExtractorError('Invalid URL: %s' % url)
-
-        video_id = mobj.group(1)
 
         # the video may come from an external site
         m_external = re.match('^(\w{2})-(.*)$', video_id)
@@ -144,15 +143,24 @@ class MetacafeIE(InfoExtractor):
             if prefix == 'cb':
                 return self.url_result('theplatform:%s' % ext_id, 'ThePlatform')
 
-        # Retrieve video webpage to extract further information
+        # self._confirm_age()
-        req = sanitized_Request('http://www.metacafe.com/watch/%s/' % video_id)
 
         # AnyClip videos require the flashversion cookie so that we get the link
         # to the mp4 file
-        mobj_an = re.match(r'^an-(.*?)$', video_id)
+        headers = {}
-        if mobj_an:
+        if video_id.startswith('an-'):
-            req.headers['Cookie'] = 'flashVersion=0;'
+            headers['Cookie'] = 'flashVersion=0;'
-        webpage = self._download_webpage(req, video_id)
+
+        # Retrieve video webpage to extract further information
+        webpage = self._download_webpage(url, video_id, headers=headers)
+
+        error = get_element_by_attribute(
+            'class', 'notfound-page-title', webpage)
+        if error:
+            raise ExtractorError(error, expected=True)
+
+        video_title = self._html_search_meta(
+            ['og:title', 'twitter:title'], webpage, 'title', default=None) or self._search_regex(r'<h1>(.*?)</h1>', webpage, 'title')
 
         # Extract URL, uploader and title from webpage
         self.report_extraction(video_id)
@@ -216,20 +224,40 @@ class MetacafeIE(InfoExtractor):
                         'player_url': player_url,
                         'ext': play_path.partition(':')[0],
                     })
+        if video_url is None:
+            flashvars = self._parse_json(self._search_regex(
+                r'flashvars\s*=\s*({.*});', webpage, 'flashvars',
+                default=None), video_id, fatal=False)
+            if flashvars:
+                video_url = []
+                for source in flashvars.get('sources'):
+                    source_url = source.get('src')
+                    if not source_url:
+                        continue
+                    ext = mimetype2ext(source.get('type')) or determine_ext(source_url)
+                    if ext == 'm3u8':
+                        video_url.extend(self._extract_m3u8_formats(
+                            source_url, video_id, 'mp4',
+                            'm3u8_native', m3u8_id='hls', fatal=False))
+                    else:
+                        video_url.append({
+                            'url': source_url,
+                            'ext': ext,
+                        })
 
         if video_url is None:
             raise ExtractorError('Unsupported video type')
 
-        video_title = self._html_search_regex(
+        description = self._html_search_meta(
-            r'(?im)<title>(.*) - Video</title>', webpage, 'title')
+            ['og:description', 'twitter:description', 'description'],
-        description = self._og_search_description(webpage)
+            webpage, 'title', fatal=False)
-        thumbnail = self._og_search_thumbnail(webpage)
+        thumbnail = self._html_search_meta(
+            ['og:image', 'twitter:image'], webpage, 'title', fatal=False)
         video_uploader = self._html_search_regex(
             r'submitter=(.*?);|googletag\.pubads\(\)\.setTargeting\("(?:channel|submiter)","([^"]+)"\);',
             webpage, 'uploader nickname', fatal=False)
         duration = int_or_none(
-            self._html_search_meta('video:duration', webpage))
+            self._html_search_meta('video:duration', webpage, default=None))
-
         age_limit = (
             18
             if re.search(r'(?:"contentRating":|"rating",)"restricted"', webpage)
@@ -242,10 +270,11 @@ class MetacafeIE(InfoExtractor):
                 'url': video_url,
                 'ext': video_ext,
             }]
-
         self._sort_formats(formats)
+
         return {
             'id': video_id,
+            'display_id': display_id,
             'description': description,
             'uploader': video_uploader,
             'title': video_title,

youtube_dl/extractor/mgtv.py

@@ -26,7 +26,8 @@ class MGTVIE(InfoExtractor):
         video_id = self._match_id(url)
         api_data = self._download_json(
             'http://v.api.mgtv.com/player/video', video_id,
-            query={'video_id': video_id})['data']
+            query={'video_id': video_id},
+            headers=self.geo_verification_headers())['data']
         info = api_data['info']
 
         formats = []

youtube_dl/extractor/miomio.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import random
 
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 from ..utils import (
     xpath_text,
     int_or_none,
@@ -18,13 +19,16 @@ class MioMioIE(InfoExtractor):
     _TESTS = [{
         # "type=video" in flashvars
         'url': 'http://www.miomio.tv/watch/cc88912/',
-        'md5': '317a5f7f6b544ce8419b784ca8edae65',
         'info_dict': {
             'id': '88912',
             'ext': 'flv',
             'title': '【SKY】字幕 铠武昭和VS平成 假面骑士大战FEAT战队 魔星字幕组 字幕',
             'duration': 5923,
         },
+        'params': {
+            # The server provides broken file
+            'skip_download': True,
+        }
     }, {
         'url': 'http://www.miomio.tv/watch/cc184024/',
         'info_dict': {
@@ -32,7 +36,7 @@ class MioMioIE(InfoExtractor):
             'title': '《动漫同人插画绘制》',
         },
         'playlist_mincount': 86,
-        'skip': 'This video takes time too long for retrieving the URL',
+        'skip': 'Unable to load videos',
     }, {
         'url': 'http://www.miomio.tv/watch/cc173113/',
         'info_dict': {
@@ -40,20 +44,23 @@ class MioMioIE(InfoExtractor):
             'title': 'The New Macbook 2015 上手试玩与简评'
         },
         'playlist_mincount': 2,
+        'skip': 'Unable to load videos',
+    }, {
+        # new 'h5' player
+        'url': 'http://www.miomio.tv/watch/cc273295/',
+        'md5': '',
+        'info_dict': {
+            'id': '273295',
+            'ext': 'mp4',
+            'title': 'アウト×デラックス 20160526',
+        },
+        'params': {
+            # intermittent HTTP 500
+            'skip_download': True,
+        },
     }]
 
-    def _real_extract(self, url):
+    def _extract_mioplayer(self, webpage, video_id, title, http_headers):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        title = self._html_search_meta(
-            'description', webpage, 'title', fatal=True)
-
-        mioplayer_path = self._search_regex(
-            r'src="(/mioplayer/[^"]+)"', webpage, 'ref_path')
-
-        http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path}
-
         xml_config = self._search_regex(
             r'flashvars="type=(?:sina|video)&(.+?)&',
             webpage, 'xml config')
@@ -92,10 +99,34 @@ class MioMioIE(InfoExtractor):
                 'http_headers': http_headers,
             })
 
+        return entries
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._html_search_meta(
+            'description', webpage, 'title', fatal=True)
+
+        mioplayer_path = self._search_regex(
+            r'src="(/mioplayer(?:_h5)?/[^"]+)"', webpage, 'ref_path')
+
+        if '_h5' in mioplayer_path:
+            player_url = compat_urlparse.urljoin(url, mioplayer_path)
+            player_webpage = self._download_webpage(
+                player_url, video_id,
+                note='Downloading player webpage', headers={'Referer': url})
+            entries = self._parse_html5_media_entries(player_url, player_webpage)
+            http_headers = {'Referer': player_url}
+        else:
+            http_headers = {'Referer': 'http://www.miomio.tv%s' % mioplayer_path}
+            entries = self._extract_mioplayer(webpage, video_id, title, http_headers)
+
         if len(entries) == 1:
             segment = entries[0]
             segment['id'] = video_id
             segment['title'] = title
+            segment['http_headers'] = http_headers
             return segment
 
         return {

youtube_dl/extractor/mitele.py

@@ -1,5 +1,8 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse_urlencode,
@@ -8,82 +11,137 @@ from ..compat import (
 from ..utils import (
     get_element_by_attribute,
     int_or_none,
+    remove_start,
+    extract_attributes,
+    determine_ext,
 )
 
 
-class MiTeleIE(InfoExtractor):
+class MiTeleBaseIE(InfoExtractor):
-    IE_DESC = 'mitele.es'
+    def _get_player_info(self, url, webpage):
-    _VALID_URL = r'https?://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'
+        player_data = extract_attributes(self._search_regex(
+            r'(?s)(<ms-video-player.+?</ms-video-player>)',
+            webpage, 'ms video player'))
+        video_id = player_data['data-media-id']
+        config_url = compat_urlparse.urljoin(url, player_data['data-config'])
+        config = self._download_json(
+            config_url, video_id, 'Downloading config JSON')
+        mmc_url = config['services']['mmc']
 
-    _TEST = {
+        duration = None
+        formats = []
+        for m_url in (mmc_url, mmc_url.replace('/flash.json', '/html5.json')):
+            mmc = self._download_json(
+                m_url, video_id, 'Downloading mmc JSON')
+            if not duration:
+                duration = int_or_none(mmc.get('duration'))
+            for location in mmc['locations']:
+                gat = self._proto_relative_url(location.get('gat'), 'http:')
+                bas = location.get('bas')
+                loc = location.get('loc')
+                ogn = location.get('ogn')
+                if None in (gat, bas, loc, ogn):
+                    continue
+                token_data = {
+                    'bas': bas,
+                    'icd': loc,
+                    'ogn': ogn,
+                    'sta': '0',
+                }
+                media = self._download_json(
+                    '%s/?%s' % (gat, compat_urllib_parse_urlencode(token_data)),
+                    video_id, 'Downloading %s JSON' % location['loc'])
+                file_ = media.get('file')
+                if not file_:
+                    continue
+                ext = determine_ext(file_)
+                if ext == 'f4m':
+                    formats.extend(self._extract_f4m_formats(
+                        file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
+                        video_id, f4m_id='hds', fatal=False))
+                elif ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        file_, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'formats': formats,
+            'thumbnail': player_data.get('data-poster') or config.get('poster', {}).get('imageUrl'),
+            'duration': duration,
+        }
+
+
+class MiTeleIE(MiTeleBaseIE):
+    IE_DESC = 'mitele.es'
+    _VALID_URL = r'https?://www\.mitele\.es/(?:[^/]+/){3}(?P<id>[^/]+)/'
+
+    _TESTS = [{
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
         # MD5 is unstable
         'info_dict': {
             'id': '0NF1jJnxS1Wu3pHrmvFyw2',
             'display_id': 'programa-144',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Tor, la web invisible',
             'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
+            'series': 'Diario de',
+            'season': 'La redacción',
+            'episode': 'Programa 144',
             'thumbnail': 're:(?i)^https?://.*\.jpg$',
             'duration': 2913,
         },
-    }
+    }, {
+        # no explicit title
+        'url': 'http://www.mitele.es/programas-tv/cuarto-milenio/temporada-6/programa-226/',
+        'info_dict': {
+            'id': 'eLZSwoEd1S3pVyUm8lc6F',
+            'display_id': 'programa-226',
+            'ext': 'mp4',
+            'title': 'Cuarto Milenio - Temporada 6 - Programa 226',
+            'description': 'md5:50daf9fadefa4e62d9fc866d0c015701',
+            'series': 'Cuarto Milenio',
+            'season': 'Temporada 6',
+            'episode': 'Programa 226',
+            'thumbnail': 're:(?i)^https?://.*\.jpg$',
+            'duration': 7312,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
 
         webpage = self._download_webpage(url, display_id)
 
-        config_url = self._search_regex(
+        info = self._get_player_info(url, webpage)
-            r'data-config\s*=\s*"([^"]+)"', webpage, 'data config url')
-        config_url = compat_urlparse.urljoin(url, config_url)
-
-        config = self._download_json(
-            config_url, display_id, 'Downloading config JSON')
-
-        mmc = self._download_json(
-            config['services']['mmc'], display_id, 'Downloading mmc JSON')
-
-        formats = []
-        for location in mmc['locations']:
-            gat = self._proto_relative_url(location.get('gat'), 'http:')
-            bas = location.get('bas')
-            loc = location.get('loc')
-            ogn = location.get('ogn')
-            if None in (gat, bas, loc, ogn):
-                continue
-            token_data = {
-                'bas': bas,
-                'icd': loc,
-                'ogn': ogn,
-                'sta': '0',
-            }
-            media = self._download_json(
-                '%s/?%s' % (gat, compat_urllib_parse_urlencode(token_data)),
-                display_id, 'Downloading %s JSON' % location['loc'])
-            file_ = media.get('file')
-            if not file_:
-                continue
-            formats.extend(self._extract_f4m_formats(
-                file_ + '&hdcore=3.2.0&plugin=aasp-3.2.0.77.18',
-                display_id, f4m_id=loc))
-        self._sort_formats(formats)
 
         title = self._search_regex(
-            r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>', webpage, 'title')
+            r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>',
+            webpage, 'title', default=None)
 
-        video_id = self._search_regex(
+        mobj = re.search(r'''(?sx)
-            r'data-media-id\s*=\s*"([^"]+)"', webpage,
+                            class="Destacado-text"[^>]*>.*?<h1>\s*
-            'data media id', default=None) or display_id
+                            <span>(?P<series>[^<]+)</span>\s*
-        thumbnail = config.get('poster', {}).get('imageUrl')
+                            <span>(?P<season>[^<]+)</span>\s*
-        duration = int_or_none(mmc.get('duration'))
+                            <span>(?P<episode>[^<]+)</span>''', webpage)
+        series, season, episode = mobj.groups() if mobj else [None] * 3
 
-        return {
+        if not title:
-            'id': video_id,
+            if mobj:
+                title = '%s - %s - %s' % (series, season, episode)
+            else:
+                title = remove_start(self._search_regex(
+                    r'<title>([^<]+)</title>', webpage, 'title'), 'Ver online ')
+
+        info.update({
             'display_id': display_id,
             'title': title,
             'description': get_element_by_attribute('class', 'text', webpage),
-            'thumbnail': thumbnail,
+            'series': series,
-            'duration': duration,
+            'season': season,
-            'formats': formats,
+            'episode': episode,
-        }
+        })
+        return info

youtube_dl/extractor/mixcloud.py

@@ -102,11 +102,11 @@ class MixcloudIE(InfoExtractor):
         description = self._og_search_description(webpage)
         like_count = parse_count(self._search_regex(
             r'\bbutton-favorite[^>]+>.*?<span[^>]+class=["\']toggle-number[^>]+>\s*([^<]+)',
-            webpage, 'like count', fatal=False))
+            webpage, 'like count', default=None))
         view_count = str_to_int(self._search_regex(
             [r'<meta itemprop="interactionCount" content="UserPlays:([0-9]+)"',
              r'/listeners/?">([0-9,.]+)</a>'],
-            webpage, 'play count', fatal=False))
+            webpage, 'play count', default=None))
 
         return {
             'id': track_id,

youtube_dl/extractor/msn.py

@@ -0,0 +1,122 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    determine_ext,
+    ExtractorError,
+    int_or_none,
+    unescapeHTML,
+)
+
+
+class MSNIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?msn\.com/(?:[^/]+/)+(?P<display_id>[^/]+)/[a-z]{2}-(?P<id>[\da-zA-Z]+)'
+    _TESTS = [{
+        'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/criminal-minds-shemar-moore-shares-a-touching-goodbye-message/vp-BBqQYNE',
+        'md5': '8442f66c116cbab1ff7098f986983458',
+        'info_dict': {
+            'id': 'BBqQYNE',
+            'display_id': 'criminal-minds-shemar-moore-shares-a-touching-goodbye-message',
+            'ext': 'mp4',
+            'title': 'Criminal Minds - Shemar Moore Shares A Touching Goodbye Message',
+            'description': 'md5:e8e89b897b222eb33a6b5067a8f1bc25',
+            'duration': 104,
+            'uploader': 'CBS Entertainment',
+            'uploader_id': 'IT0X5aoJ6bJgYerJXSDCgFmYPB1__54v',
+        },
+    }, {
+        'url': 'http://www.msn.com/en-ae/news/offbeat/meet-the-nine-year-old-self-made-millionaire/ar-BBt6ZKf',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.msn.com/en-ae/video/watch/obama-a-lot-of-people-will-be-disappointed/vi-AAhxUMH',
+        'only_matching': True,
+    }, {
+        # geo restricted
+        'url': 'http://www.msn.com/en-ae/foodanddrink/joinourtable/the-first-fart-makes-you-laugh-the-last-fart-makes-you-cry/vp-AAhzIBU',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.msn.com/en-ae/entertainment/bollywood/watch-how-salman-khan-reacted-when-asked-if-he-would-apologize-for-his-‘raped-woman’-comment/vi-AAhvzW6',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id, display_id = mobj.group('id', 'display_id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        video = self._parse_json(
+            self._search_regex(
+                r'data-metadata\s*=\s*(["\'])(?P<data>.+?)\1',
+                webpage, 'video data', default='{}', group='data'),
+            display_id, transform_source=unescapeHTML)
+
+        if not video:
+            error = unescapeHTML(self._search_regex(
+                r'data-error=(["\'])(?P<error>.+?)\1',
+                webpage, 'error', group='error'))
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
+
+        title = video['title']
+
+        formats = []
+        for file_ in video.get('videoFiles', []):
+            format_url = file_.get('url')
+            if not format_url:
+                continue
+            ext = determine_ext(format_url)
+            # .ism is not yet supported (see
+            # https://github.com/rg3/youtube-dl/issues/8118)
+            if ext == 'ism':
+                continue
+            if 'm3u8' in format_url:
+                # m3u8_native should not be used here until
+                # https://github.com/rg3/youtube-dl/issues/9913 is fixed
+                m3u8_formats = self._extract_m3u8_formats(
+                    format_url, display_id, 'mp4',
+                    m3u8_id='hls', fatal=False)
+                # Despite metadata in m3u8 all video+audio formats are
+                # actually video-only (no audio)
+                for f in m3u8_formats:
+                    if f.get('acodec') != 'none' and f.get('vcodec') != 'none':
+                        f['acodec'] = 'none'
+                formats.extend(m3u8_formats)
+            else:
+                formats.append({
+                    'url': format_url,
+                    'ext': 'mp4',
+                    'format_id': 'http',
+                    'width': int_or_none(file_.get('width')),
+                    'height': int_or_none(file_.get('height')),
+                })
+        self._sort_formats(formats)
+
+        subtitles = {}
+        for file_ in video.get('files', []):
+            format_url = file_.get('url')
+            format_code = file_.get('formatCode')
+            if not format_url or not format_code:
+                continue
+            if compat_str(format_code) == '3100':
+                subtitles.setdefault(file_.get('culture', 'en'), []).append({
+                    'ext': determine_ext(format_url, 'ttml'),
+                    'url': format_url,
+                })
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': video.get('description'),
+            'thumbnail': video.get('headlineImage', {}).get('url'),
+            'duration': int_or_none(video.get('durationSecs')),
+            'uploader': video.get('sourceFriendly'),
+            'uploader_id': video.get('providerId'),
+            'creator': video.get('creator'),
+            'subtitles': subtitles,
+            'formats': formats,
+        }

youtube_dl/extractor/mtv.py

@@ -6,6 +6,7 @@ from .common import InfoExtractor
 from ..compat import (
     compat_urllib_parse_urlencode,
     compat_str,
+    compat_xpath,
 )
 from ..utils import (
     ExtractorError,
@@ -84,9 +85,10 @@ class MTVServicesInfoExtractor(InfoExtractor):
                 rtmp_video_url = rendition.find('./src').text
                 if rtmp_video_url.endswith('siteunavail.png'):
                     continue
+                new_url = self._transform_rtmp_url(rtmp_video_url)
                 formats.append({
-                    'ext': ext,
+                    'ext': 'flv' if new_url.startswith('rtmp') else ext,
-                    'url': self._transform_rtmp_url(rtmp_video_url),
+                    'url': new_url,
                     'format_id': rendition.get('bitrate'),
                     'width': int(rendition.get('width')),
                     'height': int(rendition.get('height')),
@@ -139,9 +141,9 @@ class MTVServicesInfoExtractor(InfoExtractor):
                 itemdoc, './/{http://search.yahoo.com/mrss/}category',
                 'scheme', 'urn:mtvn:video_title')
         if title_el is None:
-            title_el = itemdoc.find('.//{http://search.yahoo.com/mrss/}title')
+            title_el = itemdoc.find(compat_xpath('.//{http://search.yahoo.com/mrss/}title'))
         if title_el is None:
-            title_el = itemdoc.find('.//title') or itemdoc.find('./title')
+            title_el = itemdoc.find(compat_xpath('.//title'))
             if title_el.text is None:
                 title_el = None
 

youtube_dl/extractor/nationalgeographic.py

@@ -1,6 +1,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from .theplatform import ThePlatformIE
 from ..utils import (
     smuggle_url,
     url_basename,
@@ -61,7 +62,7 @@ class NationalGeographicIE(InfoExtractor):
         }
 
 
-class NationalGeographicChannelIE(InfoExtractor):
+class NationalGeographicChannelIE(ThePlatformIE):
     IE_NAME = 'natgeo:channel'
     _VALID_URL = r'https?://channel\.nationalgeographic\.com/(?:wild/)?[^/]+/videos/(?P<id>[^/?]+)'
 
@@ -102,12 +103,22 @@ class NationalGeographicChannelIE(InfoExtractor):
         release_url = self._search_regex(
             r'video_auth_playlist_url\s*=\s*"([^"]+)"',
             webpage, 'release url')
+        query = {
+            'mbr': 'true',
+            'switch': 'http',
+        }
+        is_auth = self._search_regex(r'video_is_auth\s*=\s*"([^"]+)"', webpage, 'is auth', fatal=False)
+        if is_auth == 'auth':
+            auth_resource_id = self._search_regex(
+                r"video_auth_resourceId\s*=\s*'([^']+)'",
+                webpage, 'auth resource id')
+            query['auth'] = self._extract_mvpd_auth(url, display_id, 'natgeo', auth_resource_id) or ''
 
         return {
             '_type': 'url_transparent',
             'ie_key': 'ThePlatform',
             'url': smuggle_url(
-                update_url_query(release_url, {'mbr': 'true', 'switch': 'http'}),
+                update_url_query(release_url, query),
                 {'force_smil_url': True}),
             'display_id': display_id,
         }