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 *2016.07.02*. 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 Issues with outdated version will be rejected.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 **2016.07.02**
+### 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 Issues with outdated version will be rejected.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 **2016.07.13**
 
 ### 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 2016.07.02
+[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

@@ -176,3 +176,5 @@ 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`
-3. Start a new git branch with `cd youtube-dl; git checkout -b yourextractor`
+2. Check out the source code with:
+
+        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
@@ -143,16 +151,148 @@ 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/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/master/youtube_dl/extractor/common.py#L148-L252) 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`.
-9. 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+.
-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:
+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. 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

@@ -103,9 +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
-                                     some Chinese sites. The default proxy
-                                     specified by --proxy (or none, if the
+    --geo-verification-proxy URL     Use this proxy to verify the IP address for
+                                     some geo-restricted sites. The default
+                                     proxy specified by --proxy (or none, if the
                                      options is not present) is used for the
                                      actual downloading. (experimental)
 
@@ -424,7 +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:
 ```
@@ -432,6 +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`.
@@ -890,9 +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`
-3. Start a new git branch with `cd youtube-dl; git checkout -b yourextractor`
+2. Check out the source code with:
+
+        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
@@ -936,19 +945,151 @@ 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/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/master/youtube_dl/extractor/common.py#L148-L252) 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`.
-9. 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+.
-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:
+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. 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).

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/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

@@ -224,6 +224,7 @@
  - **Firstpost**
  - **FiveTV**
  - **Flickr**
+ - **Flipagram**
  - **Folketinget**: Folketinget (ft.dk; Danish parliament)
  - **FootyRoom**
  - **Formula1**
@@ -283,6 +284,8 @@
  - **HotStar**
  - **Howcast**
  - **HowStuffWorks**
+ - **HRTi**
+ - **HRTiPlaylist**
  - **HuffPost**: Huffington Post
  - **Hypem**
  - **Iconosquare**
@@ -310,6 +313,7 @@
  - **jpopsuki.tv**
  - **JWPlatform**
  - **Kaltura**
+ - **Kamcord**
  - **KanalPlay**: Kanal 5/9/11 Play
  - **Kankan**
  - **Karaoketv**
@@ -329,7 +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**
@@ -474,6 +478,8 @@
  - **Odnoklassniki**
  - **OktoberfestTV**
  - **on.aol.com**
+ - **onet.tv**
+ - **onet.tv:channel**
  - **OnionStudios**
  - **Ooyala**
  - **OoyalaExternal**
@@ -548,6 +554,7 @@
  - **RICE**
  - **RingTV**
  - **RockstarGames**
+ - **RoosterTeeth**
  - **RottenTomatoes**
  - **Roxwel**
  - **RTBF**
@@ -561,6 +568,7 @@
  - **rtve.es:infantil**: RTVE infantil
  - **rtve.es:live**: RTVE.es live streams
  - **RTVNH**
+ - **Rudo**
  - **RUHD**
  - **RulePorn**
  - **rutube**: Rutube videos
@@ -787,6 +795,7 @@
  - **vine:user**
  - **vk**: VK
  - **vk:uservideos**: VK - User's Videos
+ - **vk:wallpost**
  - **vlive**
  - **Vodlocker**
  - **VoiceRepublic**
@@ -855,6 +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)

test/test_compat.py

@@ -88,6 +88,7 @@ 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(
-            ('localhost', 0), _build_proxy_handler('cn'))
-        self.cn_port = http_server_port(self.cn_proxy)
-        self.cn_proxy_thread = threading.Thread(target=self.cn_proxy.serve_forever)
-        self.cn_proxy_thread.daemon = True
-        self.cn_proxy_thread.start()
+        self.geo_proxy = compat_http_server.HTTPServer(
+            ('localhost', 0), _build_proxy_handler('geo'))
+        self.geo_port = http_server_port(self.geo_proxy)
+        self.geo_proxy_thread = threading.Thread(target=self.geo_proxy.serve_forever)
+        self.geo_proxy_thread.daemon = True
+        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,
@@ -80,6 +81,7 @@ from youtube_dl.utils import (
     cli_option,
     cli_valueless_option,
     cli_bool_option,
+    parse_codecs,
 )
 from youtube_dl.compat import (
     compat_chr,
@@ -405,6 +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'""")
@@ -601,6 +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-_.~'
@@ -985,5 +1016,13 @@ The first line
         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
-                       on Chinese sites. (Experimental)
+    geo_verification_proxy:  URL of the proxy to use for IP address verification
+                       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

youtube_dl/__init__.py

@@ -382,6 +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/compat.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 import binascii
@@ -2594,15 +2595,19 @@ except ImportError:  # Python < 3.3
             return "'" + s.replace("'", "'\"'\"'") + "'"
 
 
-if sys.version_info >= (2, 7, 3):
+try:
+    args = shlex.split('中文')
+    assert (isinstance(args, list) and
+            isinstance(args[0], compat_str) and
+            args[0] == '中文')
     compat_shlex_split = shlex.split
-else:
+except (AssertionError, UnicodeEncodeError):
     # Working around shlex issue with unicode strings on some python 2
     # versions (see http://bugs.python.org/issue1548891)
     def compat_shlex_split(s, comments=False, posix=True):
         if isinstance(s, compat_str):
             s = s.encode('utf-8')
-        return shlex.split(s, comments, posix)
+        return list(map(lambda s: s.decode('utf-8'), shlex.split(s, comments, posix)))
 
 
 def compat_ord(c):

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/extractor/aenetworks.py

@@ -2,7 +2,7 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
+from .theplatform import ThePlatformIE
 from ..utils import (
     smuggle_url,
     update_url_query,
@@ -15,28 +15,15 @@ from ..compat import (
 )
 
 
-class AENetworksBaseIE(InfoExtractor):
-    def theplatform_url_result(self, theplatform_url, video_id, query):
-        return {
-            '_type': 'url_transparent',
-            'id': video_id,
-            'url': smuggle_url(
-                update_url_query(theplatform_url, query),
-                {
-                    'sig': {
-                        'key': 'crazyjava',
-                        'secret': 's3cr3t'
-                    },
-                    'force_smil_url': True
-                }),
-            'ie_key': 'ThePlatform',
-        }
+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)/(?:shows/(?P<show_path>[^/]+(?:/[^/]+){0,2})|movies/(?P<movie_display_id>[^/]+)/full-movie)'
+    _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',
@@ -76,9 +63,15 @@ class AENetworksIE(AENetworksBaseIE):
         '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):
-        show_path, movie_display_id = re.match(self._VALID_URL, url).groups()
+        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:
@@ -103,23 +96,39 @@ class AENetworksIE(AENetworksBaseIE):
                         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')
-
-        info = self._search_json_ld(webpage, video_id, fatal=False)
-        info.update(self.theplatform_url_result(
-            media_url, video_id, {
-                'mbr': 'true',
-                'assetTypes': 'medium_video_s3'
-            }))
+        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>[^/]+)/videos(?:/(?P<video_display_id>[^/?#]+))?'
+    _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': {
@@ -147,8 +156,30 @@ class HistoryTopicIE(AENetworksBaseIE):
     }, {
         'url': 'http://www.history.com/topics/world-war-i-history/videos',
         'only_matching': True,
+    }, {
+        'url': 'http://www.history.com/topics/world-war-i/world-war-i-history',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.history.com/topics/world-war-i/world-war-i-history/speeches',
+        'only_matching': True,
     }]
 
+    def theplatform_url_result(self, theplatform_url, video_id, query):
+        return {
+            '_type': 'url_transparent',
+            'id': video_id,
+            'url': smuggle_url(
+                update_url_query(theplatform_url, query),
+                {
+                    'sig': {
+                        'key': self._THEPLATFORM_KEY,
+                        'secret': self._THEPLATFORM_SECRET,
+                    },
+                    'force_smil_url': True
+                }),
+            'ie_key': 'ThePlatform',
+        }
+
     def _real_extract(self, url):
         topic_id, video_display_id = re.match(self._VALID_URL, url).groups()
         if video_display_id:

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_type = media['type']
-            if media_type in ('video/f4m', 'application/f4m+xml'):
+            media = media_data.get('@attributes', {})
+            media_url = media.get('url')
+            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(
-                r'(?s)<h3[^>]+class="episodebox-title".+?>Episodeninhalt<', webpage), 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,
-            }
-
+        def extract_info(html, video_id, num=None):
+            title, description = [None] * 2
             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:
-                self._sort_formats(formats)
+            return {
+                '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({
-                    'title': title,
-                    'description': description,
-                    'formats': formats,
-                })
+                f.update(info)
                 entries.append(f)
 
-            # Extract teaser only when full episode is not available
-            if not formats:
+            # Extract teaser/trailer only when full episode is not available
+            if not info['formats']:
                 m = re.search(
-                    r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>Teaser<',
-                    episode_html)
+                    r'data-dialog-header=(["\'])(?P<title>.+?)\1[^>]+href=(["\'])(?P<href>.+?)\3[^>]*>(?P<kind>Teaser|Trailer)<',
+                    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/ard.py

@@ -13,6 +13,7 @@ from ..utils import (
     parse_duration,
     unified_strdate,
     xpath_text,
+    update_url_query,
 )
 from ..compat import compat_etree_fromstring
 
@@ -34,6 +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',
@@ -44,6 +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',
@@ -55,6 +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,
@@ -113,11 +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',
-                            video_id, preference=-1, f4m_id='hds', fatal=False))
+                            update_url_query(stream_url, {
+                                '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 = {
@@ -231,7 +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):

youtube_dl/extractor/bbc.py

@@ -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
@@ -192,18 +191,17 @@ class BBCCoUkIE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
-            'skip': 'Now it\'s really geo-restricted',
+            '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,
             },
         }, {
@@ -248,9 +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({
@@ -305,13 +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({
-                    'width': width,
-                    'height': height,
-                    'vbr': vbr,
-                    'vcodec': vcodec,
-                    'filesize': file_size,
-                })
+                if format.get('protocol') != 'm3u8_native':
+                    format.update({
+                        'width': width,
+                        'height': height,
+                        'vbr': vbr,
+                        'vcodec': vcodec,
+                        'filesize': file_size,
+                    })
                 if service:
                     format['format_id'] = '%s_%s' % (service, format['format_id'])
             formats.extend(conn_formats)

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/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': 'Buddhanz',
-                'title': 'Angry Ram destroys a punching bag',
+                'uploader': 'Angry Ram',
             }
         }]
     }, {
@@ -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/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/cliprs.py

@@ -1,16 +1,10 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    float_or_none,
-    int_or_none,
-    parse_iso8601,
-)
+from .onet import OnetBaseIE
 
 
-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(
-            r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
+        mvp_id = self._search_mvp_id(webpage)
 
-        response = self._download_json(
-            'http://qi.ckm.onetapi.pl/', video_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',
-            })
+        info_dict = self._extract_from_id(mvp_id, webpage)
+        info_dict['display_id'] = display_id
 
-        error = response.get('error')
-        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,
-        }
+        return info_dict

youtube_dl/extractor/common.py

@@ -44,6 +44,7 @@ from ..utils import (
     sanitized_Request,
     unescapeHTML,
     unified_strdate,
+    unified_timestamp,
     url_basename,
     xpath_element,
     xpath_text,
@@ -54,6 +55,8 @@ from ..utils import (
     update_Request,
     update_url_query,
     parse_m3u8_attributes,
+    extract_attributes,
+    parse_codecs,
 )
 
 
@@ -161,6 +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.
@@ -803,15 +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:
@@ -819,6 +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')),
@@ -837,6 +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
@@ -1186,6 +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,
                 }
@@ -1194,24 +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')
-                if codecs:
-                    vcodec, acodec = [None] * 2
-                    va_codecs = codecs.split(',')
-                    if len(va_codecs) == 1:
-                        # Audio only entries usually come with single codec and
-                        # 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]
+                # Unified Streaming Platform
+                mobj = re.search(
+                    r'audio.*?(?:%3D|=)(\d+)(?:-video.*?(?:%3D|=)(\d+))?', f['url'])
+                if mobj:
+                    abr, vbr = mobj.groups()
+                    abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000)
                     f.update({
-                        'acodec': acodec,
-                        'vcodec': vcodec,
+                        'vbr': vbr,
+                        'abr': abr,
                     })
+                f.update(parse_codecs(last_info.get('CODECS')))
                 if last_media is not None:
                     f['m3u8_media'] = last_media
                     last_media = None
@@ -1616,6 +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()
@@ -1729,6 +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/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)
-                    if type_ == 'application/x-mpegURL' or ext == 'm3u8':
+                    ext = mimetype2ext(type_) or determine_ext(media_url)
+                    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,
-            video_id, 'Downloading video formats info')
+            'http://videofarm.daum.net/controller/api/closed/v1_2/IntegratedMovieData.json',
+            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,
-            'Downloading video info')
+            'http://tvpot.daum.net/clip/ClipInfoXml.do', video_id,
+            '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$',
-            'timestamp': 1404039863.438,
+            'thumbnail': 're:https?://.*\.jpg',
+            'timestamp': 1404039863,
             'upload_date': '20140629',
             'duration': 69.544,
-            'view_count': int,
-            'categories': list,
-        }
+            'uploader_id': '1027729757001',
+        },
+        '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 = 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)
+        video_id, display_id = re.match(self._VALID_URL, url).groups()
 
         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'],
-            '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,
+            'ie_key': 'BrightcoveNew',
         }

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(
-            'http://widgets.ellentube.com/videos/%s' % video_id,
-            video_id)
+        URLS = ('http://widgets.ellentube.com/videos/%s' % video_id, url)
 
-        partner_id = self._search_regex(
-            r"var\s+partnerId\s*=\s*'([^']+)", webpage, 'partner id')
+        for num, url_ in enumerate(URLS, 1):
+            webpage = self._download_webpage(
+                url_, video_id, fatal=num == len(URLS))
 
-        kaltura_id = self._search_regex(
-            [r'id="kaltura_player_([^"]+)"',
-             r"_wb_entry_id\s*:\s*'([^']+)",
-             r'data-kaltura-entry-id="([^"]+)'],
-            webpage, 'kaltura id')
+            default = NO_DEFAULT if num == len(URLS) else None
+
+            partner_id = self._search_regex(
+                r"var\s+partnerId\s*=\s*'([^']+)", webpage, 'partner 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

@@ -139,9 +139,9 @@ 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
@@ -256,6 +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
@@ -326,6 +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
@@ -364,6 +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
@@ -577,6 +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,
@@ -670,6 +680,7 @@ 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
@@ -680,6 +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 (
@@ -977,6 +989,7 @@ from .viki import (
 from .vk import (
     VKIE,
     VKUserVideosIE,
+    VKWallPostIE,
 )
 from .vlive import VLiveIE
 from .vodlocker import VodlockerIE
@@ -1062,6 +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)
-        if m:
-            swf_params = m.group(1).replace('\\\\', '\\').replace('\\"', '"')
+        PATTERN = re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER)
+
+        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 = {}

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/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/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
@@ -66,6 +69,7 @@ from .theplatform import ThePlatformIE
 from .vessel import VesselIE
 from .kaltura import KalturaIE
 from .eagleplatform import EaglePlatformIE
+from .facebook import FacebookIE
 
 
 class GenericIE(InfoExtractor):
@@ -1260,7 +1264,87 @@ class GenericIE(InfoExtractor):
                 '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):
@@ -1617,12 +1701,16 @@ class GenericIE(InfoExtractor):
         if matches:
             return _playlist_from_matches(matches, lambda m: unescapeHTML(m))
 
-        # 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)
+        # Look for Wordpress "YouTube Video Importer" plugin
+        matches = re.findall(r'''(?x)<div[^>]+
+            class=(?P<q1>[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+
+            data-video_id=(?P<q2>[\'"])([^\'"]+)(?P=q2)''', webpage)
         if matches:
-            return _playlist_from_matches(
-                matches, lambda m: unescapeHTML(m[1]))
+            return _playlist_from_matches(matches, lambda m: m[-1])
+
+        matches = DailymotionIE._extract_urls(webpage)
+        if matches:
+            return _playlist_from_matches(matches)
 
         # Look for embedded Dailymotion playlist player (#3822)
         m = re.search(
@@ -1759,10 +1847,9 @@ class GenericIE(InfoExtractor):
             return self.url_result(mobj.group('url'))
 
         # Look for embedded Facebook player
-        mobj = re.search(
-            r'<iframe[^>]+?src=(["\'])(?P<url>https://www\.facebook\.com/video/embed.+?)\1', webpage)
-        if mobj is not None:
-            return self.url_result(mobj.group('url'), 'Facebook')
+        facebook_url = FacebookIE._extract_url(webpage)
+        if facebook_url is not None:
+            return self.url_result(facebook_url, 'Facebook')
 
         # Look for embedded VK player
         mobj = re.search(r'<iframe[^>]+?src=(["\'])(?P<url>https?://vk\.com/video_ext\.php.+?)\1', webpage)
@@ -2102,6 +2189,19 @@ class GenericIE(InfoExtractor):
         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/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/iqiyi.py

@@ -1,10 +1,8 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import binascii
 import hashlib
 import itertools
-import math
 import re
 import time
 
@@ -14,12 +12,13 @@ from ..compat import (
     compat_urllib_parse_urlencode,
 )
 from ..utils import (
+    clean_html,
     decode_packed_codes,
+    get_element_by_id,
+    get_element_by_attribute,
     ExtractorError,
-    intlist_to_bytes,
     ohdave_rsa_encrypt,
     remove_start,
-    urshift,
 )
 
 
@@ -166,7 +165,7 @@ class IqiyiIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.iqiyi.com/v_19rrojlavg.html',
-        'md5': '470a6c160618577166db1a7aac5a3606',
+        # MD5 checksum differs on my machine and Travis CI
         'info_dict': {
             'id': '9c1fb1b99d192b21c559e5a1a2cb3c73',
             'ext': 'mp4',
@@ -174,11 +173,11 @@ class IqiyiIE(InfoExtractor):
         }
     }, {
         'url': 'http://www.iqiyi.com/v_19rrhnnclk.html',
-        'md5': 'f09f0a6a59b2da66a26bf4eda669a4cc',
+        'md5': '667171934041350c5de3f5015f7f1152',
         'info_dict': {
             'id': 'e3f585b550a280af23c98b6cb2be19fb',
             'ext': 'mp4',
-            'title': '名侦探柯南 国语版',
+            'title': '名侦探柯南 国语版：第752集 迫近灰原秘密的黑影 下篇',
         },
         'skip': 'Geo-restricted to China',
     }, {
@@ -196,22 +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': [{
-            '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'],
+        'skip': 'Geo-restricted to China',
     }, {
         'url': 'http://www.iqiyi.com/a_19rrhb8ce1.html',
         'info_dict': {
@@ -224,14 +211,16 @@ class IqiyiIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    _FORMATS_MAP = [
-        ('1', 'h6'),
-        ('2', 'h5'),
-        ('3', 'h4'),
-        ('4', 'h3'),
-        ('5', 'h2'),
-        ('10', 'h1'),
-    ]
+    _FORMATS_MAP = {
+        '96': 1,    # 216p, 240p
+        '1': 2,     # 336p, 360p
+        '2': 3,     # 480p, 504p
+        '21': 4,    # 504p
+        '4': 5,     # 720p
+        '17': 5,    # 720p
+        '5': 6,     # 1072p, 1080p
+        '18': 7,    # 1080p
+    }
 
     def _real_initialize(self):
         self._login()
@@ -291,101 +280,23 @@ class IqiyiIE(InfoExtractor):
 
         return True
 
-    @staticmethod
-    def _gen_sc(tvid, timestamp):
-        M = [1732584193, -271733879]
-        M.extend([~M[0], ~M[1]])
-        I_table = [7, 12, 17, 22, 5, 9, 14, 20, 4, 11, 16, 23, 6, 10, 15, 21]
-        C_base = [0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 8388608, 432]
-
-        def L(n, t):
-            if t is None:
-                t = 0
-            return trunc(((n >> 1) + (t >> 1) << 1) + (n & 1) + (t & 1))
-
-        def trunc(n):
-            n = n % 0x100000000
-            if n > 0x7fffffff:
-                n -= 0x100000000
-            return n
-
-        def transform(string, mod):
-            num = int(string, 16)
-            return (num >> 8 * (i % 4) & 255 ^ i % mod) << ((a & 3) << 3)
-
-        C = list(C_base)
-        o = list(M)
-        k = str(timestamp - 7)
-        for i in range(13):
-            a = i
-            C[a >> 2] |= ord(k[a]) << 8 * (a % 4)
-
-        for i in range(16):
-            a = i + 13
-            start = (i >> 2) * 8
-            r = '03967743b643f66763d623d637e30733'
-            C[a >> 2] |= transform(''.join(reversed(r[start:start + 8])), 7)
-
-        for i in range(16):
-            a = i + 29
-            start = (i >> 2) * 8
-            r = '7038766939776a32776a32706b337139'
-            C[a >> 2] |= transform(r[start:start + 8], 1)
-
-        for i in range(9):
-            a = i + 45
-            if i < len(tvid):
-                C[a >> 2] |= ord(tvid[i]) << 8 * (a % 4)
-
-        for a in range(64):
-            i = a
-            I = i >> 4
-            C_index = [i, 5 * i + 1, 3 * i + 5, 7 * i][I] % 16 + urshift(a, 6)
-            m = L(L(o[0], [
-                trunc(o[1] & o[2]) | trunc(~o[1] & o[3]),
-                trunc(o[3] & o[1]) | trunc(~o[3] & o[2]),
-                o[1] ^ o[2] ^ o[3],
-                o[2] ^ trunc(o[1] | ~o[3])
-            ][I]), L(
-                trunc(int(abs(math.sin(i + 1)) * 4294967296)),
-                C[C_index] if C_index < len(C) else None))
-            I = I_table[4 * I + i % 4]
-            o = [o[3],
-                 L(o[1], trunc(trunc(m << I) | urshift(m, 32 - I))),
-                 o[1],
-                 o[2]]
-
-        new_M = [L(o[0], M[0]), L(o[1], M[1]), L(o[2], M[2]), L(o[3], M[3])]
-        s = [new_M[a >> 3] >> (1 ^ a & 7) * 4 & 15 for a in range(32)]
-        return binascii.hexlify(intlist_to_bytes(s))[1::2].decode('ascii')
-
     def get_raw_data(self, tvid, video_id):
         tm = int(time.time() * 1000)
 
-        sc = self._gen_sc(tvid, tm)
+        key = 'd5fb4bd9d50c4be6948c97edd7254b0e'
+        sc = md5_text(compat_str(tm) + key + tvid)
         params = {
-            'platForm': 'h5',
-            'rate': 1,
             'tvid': tvid,
             'vid': video_id,
-            'cupid': 'qc_100001_100186',
-            'type': 'mp4',
-            'nolimit': 0,
-            'agenttype': 13,
-            'src': 'd846d0c32d664d32b6b54ea48997a589',
+            'src': '76f90cbd92f94a2e925d83e8ccd22cb7',
             'sc': sc,
-            't': tm - 7,
-            '__jsT': None,
+            't': tm,
         }
 
-        headers = {}
-        cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
-        if cn_verification_proxy:
-            headers['Ytdl-request-proxy'] = cn_verification_proxy
         return self._download_json(
             'http://cache.m.iqiyi.com/jp/tmts/%s/%s/' % (tvid, video_id),
             video_id, transform_source=lambda s: remove_start(s, 'var tvInfoJs='),
-            query=params, headers=headers)
+            query=params, headers=self.geo_verification_headers())
 
     def _extract_playlist(self, webpage):
         PAGE_SIZE = 50
@@ -435,6 +346,7 @@ class IqiyiIE(InfoExtractor):
         video_id = self._search_regex(
             r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
 
+        formats = []
         for _ in range(5):
             raw_data = self.get_raw_data(tvid, video_id)
 
@@ -445,16 +357,29 @@ class IqiyiIE(InfoExtractor):
 
             data = raw_data['data']
 
-            # iQiYi sometimes returns Ads
-            if not isinstance(data['playInfo'], dict):
-                self._sleep(5, video_id)
-                continue
+            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['playInfo']['an']
-            break
+            if formats:
+                break
+
+            self._sleep(5, video_id)
+
+        self._sort_formats(formats)
+        title = (get_element_by_id('widget-videotitle', webpage) or
+                 clean_html(get_element_by_attribute('class', 'mod-play-tit', webpage)))
 
         return {
             'id': video_id,
             'title': title,
-            'url': data['m3u'],
+            'formats': formats,
         }

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',
@@ -88,18 +89,26 @@ class KalturaIE(InfoExtractor):
                     (?P<q3>["\'])(?P<id>.+?)(?P=q3)
                 ''', webpage))
         if mobj:
-            return 'kaltura:%(partner_id)s:%(id)s' % mobj.groupdict()
+            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, *args, **kwargs):
+    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)
-        url = self._API_BASE + query
-        data = self._download_json(url, video_id, *args, **kwargs)
+        data = self._download_json(
+            (service_url or self._SERVICE_URL) + self._SERVICE_BASE,
+            video_id, query=params, *args, **kwargs)
 
         status = data if len(actions) == 1 else data[0]
         if status.get('objectType') == 'KalturaAPIException':
@@ -108,7 +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,
@@ -118,10 +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):
-        signature = self._get_kaltura_signature(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, service_url)
         actions = [
             {
                 'action': 'null',
@@ -144,7 +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, {})
@@ -153,7 +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:
@@ -201,12 +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'),
@@ -219,9 +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'))
-        formats.extend(self._extract_m3u8_formats(
-            m3u8_url, entry_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+        if '/playManifest/' in data_url:
+            m3u8_url = sign_url(data_url.replace(
+                '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:

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'
-    _VALID_URL = r'''(?x)
-        https?://(?:www\.)?la7\.tv/
-        (?:
-            richplayer/\?assetid=|
-            \?contentId=
-        )
-        (?P<id>[0-9]+)'''
+    IE_NAME = 'la7.it'
+    _VALID_URL = r'''(?x)(https?://)?(?:
+        (?:www\.)?la7\.it/([^/]+)/(?:rivedila7|video)/|
+        tg\.la7\.it/repliche-tgla7\?id=
+    )(?P<id>.+)'''
 
-    _TEST = {
-        'url': 'http://www.la7.tv/richplayer/?assetid=50355319',
-        'md5': 'ec7d1f0224d20ba293ab56cf2259651f',
+    _TESTS = [{
+        # 'src' is a plain URL
+        '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',
-            'description': 'Un film di Paolo Sorrentino con Toni Servillo, Anna Bonaiuto, Giulio Bosetti  e Flavio Bucci',
-            'duration': 6254,
+            'title': 'Inc.Cool8',
+            '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',
+            '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
-        description = doc.find('description').text
-        duration = parse_duration(doc.find('duration').text)
-        thumbnail = doc.find('img').text
-        view_count = int(doc.find('views').text)
+        webpage = self._download_webpage(url, video_id)
 
-        prefix = doc.find('.//fqdn').text.strip().replace('auto:', 'http:')
-
-        formats = [{
-            '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)
+        player_data = self._parse_json(
+            self._search_regex(r'videoLa7\(({[^;]+})\);', webpage, 'player data'),
+            video_id, transform_source=js_to_json)
 
         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,
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats,
-            'view_count': view_count,
+            'title': player_data['title'],
+            'description': self._og_search_description(webpage, default=None),
+            'thumbnail': player_data.get('poster'),
+            'ie_key': 'Kaltura',
         }

youtube_dl/extractor/leeco.py

@@ -20,10 +20,10 @@ from ..utils import (
     int_or_none,
     orderedSet,
     parse_iso8601,
-    sanitized_Request,
     str_or_none,
     url_basename,
     urshift,
+    update_url_query,
 )
 
 
@@ -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):
-        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')
-
+    def _check_errors(self, play_json):
         # 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']
-        dispatch = playurl['dispatch']
+        play_json_h5 = self._download_json(
+            '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 = []
-        for format_id in formats:
-            if format_id in dispatch:
-                media_url = playurl['domain'][0] + dispatch[format_id][0]
-                media_url += '&' + compat_urllib_parse_urlencode({
-                    'm3v': 1,
+        play_json_flash = self._download_json(
+            'http://api.le.com/mms/out/video/playJson',
+            media_id, 'Downloading flash playJson data', query={
+                'id': media_id,
+                'platid': 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(
-                    media_url, media_id,
-                    'Download JSON metadata for format %s' % format_id)
+            return {
+                'http': update_url_query(location, {'tss': 'no'}),
+                'hls': update_url_query(location, {'tss': 'ios'}),
+            }
 
-                req = self._request_webpage(
-                    nodes_data['nodelist'][0]['location'], media_id,
-                    note='Downloading m3u8 information for format %s' % format_id)
+        def get_flash_urls(media_url, format_id):
+            media_url += '&' + compat_urllib_parse_urlencode({
+                '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 = {
-                    'url': encode_data_uri(m3u8_data, 'application/vnd.apple.mpegurl'),
-                    'ext': determine_ext(dispatch[format_id][1]),
-                    'format_id': format_id,
-                    'protocol': 'm3u8',
-                }
+            req = self._request_webpage(
+                nodes_data['nodelist'][0]['location'], media_id,
+                note='Downloading m3u8 information for format %s' % format_id)
 
-                if format_id[-1:] == 'p':
-                    url_info_dict['height'] = int_or_none(format_id[:-1])
+            m3u8_data = self.decrypt_m3u8(req.read())
 
-                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,

youtube_dl/extractor/lynda.py

@@ -100,7 +100,7 @@ class LyndaIE(LyndaBaseIE):
 
     _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',

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',
-                'description': 'md5:38c711dd98f5bb87acf973d573442e67',
+                'uploader': 'AnyClip',
+                '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)
-        if mobj is None:
-            raise ExtractorError('Invalid URL: %s' % url)
-
-        video_id = mobj.group(1)
+        video_id, display_id = re.match(self._VALID_URL, url).groups()
 
         # 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
-        req = sanitized_Request('http://www.metacafe.com/watch/%s/' % video_id)
+        # self._confirm_age()
 
         # AnyClip videos require the flashversion cookie so that we get the link
         # to the mp4 file
-        mobj_an = re.match(r'^an-(.*?)$', video_id)
-        if mobj_an:
-            req.headers['Cookie'] = 'flashVersion=0;'
-        webpage = self._download_webpage(req, video_id)
+        headers = {}
+        if video_id.startswith('an-'):
+            headers['Cookie'] = 'flashVersion=0;'
+
+        # 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(
-            r'(?im)<title>(.*) - Video</title>', webpage, 'title')
-        description = self._og_search_description(webpage)
-        thumbnail = self._og_search_thumbnail(webpage)
+        description = self._html_search_meta(
+            ['og:description', 'twitter:description', 'description'],
+            webpage, 'title', fatal=False)
+        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):
-        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}
-
+    def _extract_mioplayer(self, webpage, video_id, title, http_headers):
         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

@@ -12,12 +12,69 @@ from ..utils import (
     get_element_by_attribute,
     int_or_none,
     remove_start,
+    extract_attributes,
+    determine_ext,
 )
 
 
-class MiTeleIE(InfoExtractor):
+class MiTeleBaseIE(InfoExtractor):
+    def _get_player_info(self, url, webpage):
+        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']
+
+        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/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'
+    _VALID_URL = r'https?://www\.mitele\.es/(?:[^/]+/){3}(?P<id>[^/]+)/'
 
     _TESTS = [{
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
@@ -25,7 +82,7 @@ class MiTeleIE(InfoExtractor):
         'info_dict': {
             'id': '0NF1jJnxS1Wu3pHrmvFyw2',
             'display_id': 'programa-144',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Tor, la web invisible',
             'description': 'md5:3b6fce7eaa41b2d97358726378d9369f',
             'series': 'Diario de',
@@ -40,7 +97,7 @@ class MiTeleIE(InfoExtractor):
         'info_dict': {
             'id': 'eLZSwoEd1S3pVyUm8lc6F',
             'display_id': 'programa-226',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Cuarto Milenio - Temporada 6 - Programa 226',
             'description': 'md5:50daf9fadefa4e62d9fc866d0c015701',
             'series': 'Cuarto Milenio',
@@ -59,40 +116,7 @@ class MiTeleIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
-        config_url = self._search_regex(
-            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)
+        info = self._get_player_info(url, webpage)
 
         title = self._search_regex(
             r'class="Destacado-text"[^>]*>\s*<strong>([^<]+)</strong>',
@@ -112,21 +136,12 @@ class MiTeleIE(InfoExtractor):
                 title = remove_start(self._search_regex(
                     r'<title>([^<]+)</title>', webpage, 'title'), 'Ver online ')
 
-        video_id = self._search_regex(
-            r'data-media-id\s*=\s*"([^"]+)"', webpage,
-            'data media id', default=None) or display_id
-        thumbnail = config.get('poster', {}).get('imageUrl')
-        duration = int_or_none(mmc.get('duration'))
-
-        return {
-            'id': video_id,
+        info.update({
             'display_id': display_id,
             'title': title,
             'description': get_element_by_attribute('class', 'text', webpage),
             'series': series,
             'season': season,
             'episode': episode,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'formats': formats,
-        }
+        })
+        return info

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,
         }

youtube_dl/extractor/nick.py

@@ -8,7 +8,7 @@ from ..utils import update_url_query
 
 class NickIE(MTVServicesInfoExtractor):
     IE_NAME = 'nick.com'
-    _VALID_URL = r'https?://(?:www\.)?nick\.com/videos/clip/(?P<id>[^/?#.]+)'
+    _VALID_URL = r'https?://(?:www\.)?nick(?:jr)?\.com/(?:videos/clip|[^/]+/videos)/(?P<id>[^/?#.]+)'
     _FEED_URL = 'http://udat.mtvnservices.com/service1/dispatch.htm'
     _TESTS = [{
         'url': 'http://www.nick.com/videos/clip/alvinnn-and-the-chipmunks-112-full-episode.html',
@@ -52,6 +52,9 @@ class NickIE(MTVServicesInfoExtractor):
                 }
             },
         ],
+    }, {
+        'url': 'http://www.nickjr.com/paw-patrol/videos/pups-save-a-goldrush-s3-ep302-full-episode/',
+        'only_matching': True,
     }]
 
     def _get_feed_query(self, uri):

youtube_dl/extractor/onet.py

@@ -0,0 +1,172 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    ExtractorError,
+    float_or_none,
+    get_element_by_class,
+    int_or_none,
+    js_to_json,
+    parse_iso8601,
+    remove_start,
+    strip_or_none,
+    url_basename,
+)
+
+
+class OnetBaseIE(InfoExtractor):
+    def _search_mvp_id(self, webpage):
+        return self._search_regex(
+            r'id=(["\'])mvp:(?P<id>.+?)\1', webpage, 'mvp id', group='id')
+
+    def _extract_from_id(self, video_id, webpage):
+        response = self._download_json(
+            'http://qi.ckm.onetapi.pl/', video_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')
+        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:
+                    video_url = f.get('url')
+                    if not video_url:
+                        continue
+                    ext = determine_ext(video_url)
+                    if format_id == 'ism':
+                        # TODO: Support Microsoft Smooth Streaming
+                        continue
+                    elif ext == 'mpd':
+                        # TODO: Current DASH formats are broken - $Time$ pattern in
+                        # <SegmentTemplate> not implemented yet
+                        # formats.extend(self._extract_mpd_formats(
+                        #    video_url, video_id, mpd_id='dash', fatal=False))
+                        continue
+                    else:
+                        formats.append({
+                            'url': video_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,
+        }
+
+
+class OnetIE(OnetBaseIE):
+    _VALID_URL = 'https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/(?P<display_id>[0-9a-z-]+)/(?P<id>[0-9a-z]+)'
+    IE_NAME = 'onet.tv'
+
+    _TEST = {
+        'url': 'http://onet.tv/k/openerfestival/open-er-festival-2016-najdziwniejsze-wymagania-gwiazd/qbpyqc',
+        'md5': 'e3ffbf47590032ac3f27249204173d50',
+        'info_dict': {
+            'id': 'qbpyqc',
+            'display_id': 'open-er-festival-2016-najdziwniejsze-wymagania-gwiazd',
+            'ext': 'mp4',
+            'title': 'Open\'er Festival 2016: najdziwniejsze wymagania gwiazd',
+            'description': 'Trzy samochody, których nigdy nie użyto, prywatne spa, hotel dekorowany czarnym suknem czy nielegalne używki. Organizatorzy koncertów i festiwali muszą stawać przed nie lada wyzwaniem zapraszając gwia...',
+            'upload_date': '20160705',
+            'timestamp': 1467721580,
+        },
+    }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        display_id, video_id = mobj.group('display_id', 'id')
+
+        webpage = self._download_webpage(url, display_id)
+
+        mvp_id = self._search_mvp_id(webpage)
+
+        info_dict = self._extract_from_id(mvp_id, webpage)
+        info_dict.update({
+            'id': video_id,
+            'display_id': display_id,
+        })
+
+        return info_dict
+
+
+class OnetChannelIE(OnetBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?onet\.tv/[a-z]/(?P<id>[a-z]+)(?:[?#]|$)'
+    IE_NAME = 'onet.tv:channel'
+
+    _TEST = {
+        'url': 'http://onet.tv/k/openerfestival',
+        'info_dict': {
+            'id': 'openerfestival',
+            'title': 'Open\'er Festival Live',
+            'description': 'Dziękujemy, że oglądaliście transmisje. Zobaczcie nasze relacje i wywiady z artystami.',
+        },
+        'playlist_mincount': 46,
+    }
+
+    def _real_extract(self, url):
+        channel_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, channel_id)
+
+        current_clip_info = self._parse_json(self._search_regex(
+            r'var\s+currentClip\s*=\s*({[^}]+})', webpage, 'video info'), channel_id,
+            transform_source=lambda s: js_to_json(re.sub(r'\'\s*\+\s*\'', '', s)))
+        video_id = remove_start(current_clip_info['ckmId'], 'mvp:')
+        video_name = url_basename(current_clip_info['url'])
+
+        if self._downloader.params.get('noplaylist'):
+            self.to_screen(
+                'Downloading just video %s because of --no-playlist' % video_name)
+            return self._extract_from_id(video_id, webpage)
+
+        self.to_screen(
+            'Downloading channel %s - add --no-playlist to just download video %s' % (
+                channel_id, video_name))
+        matches = re.findall(
+            r'<a[^>]+href=[\'"](https?://(?:www\.)?onet\.tv/[a-z]/[a-z]+/[0-9a-z-]+/[0-9a-z]+)',
+            webpage)
+        entries = [
+            self.url_result(video_link, OnetIE.ie_key())
+            for video_link in matches]
+
+        channel_title = strip_or_none(get_element_by_class('o_channelName', webpage))
+        channel_description = strip_or_none(get_element_by_class('o_channelDesc', webpage))
+        return self.playlist_result(entries, channel_id, channel_title, channel_description)

youtube_dl/extractor/onionstudios.py

@@ -7,6 +7,8 @@ from .common import InfoExtractor
 from ..utils import (
     determine_ext,
     int_or_none,
+    float_or_none,
+    mimetype2ext,
 )
 
 
@@ -15,15 +17,14 @@ class OnionStudiosIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://www.onionstudios.com/videos/hannibal-charges-forward-stops-for-a-cocktail-2937',
-        'md5': 'd4851405d31adfadf71cd7a487b765bb',
+        'md5': 'e49f947c105b8a78a675a0ee1bddedfe',
         'info_dict': {
             'id': '2937',
             'ext': 'mp4',
             'title': 'Hannibal charges forward, stops for a cocktail',
-            'description': 'md5:e786add7f280b7f0fe237b64cc73df76',
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': 'The A.V. Club',
-            'uploader_id': 'TheAVClub',
+            'uploader_id': 'the-av-club',
         },
     }, {
         'url': 'http://www.onionstudios.com/embed?id=2855&autoplay=true',
@@ -40,50 +41,38 @@ class OnionStudiosIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(
-            'http://www.onionstudios.com/embed?id=%s' % video_id, video_id)
+        video_data = self._download_json(
+            'http://www.onionstudios.com/video/%s.json' % video_id, video_id)
+
+        title = video_data['title']
 
         formats = []
-        for src in re.findall(r'<source[^>]+src="([^"]+)"', webpage):
-            ext = determine_ext(src)
+        for source in video_data.get('sources', []):
+            source_url = source.get('url')
+            if not source_url:
+                continue
+            ext = mimetype2ext(source.get('content_type')) or determine_ext(source_url)
             if ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    src, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+                    source_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
             else:
-                height = int_or_none(self._search_regex(
-                    r'/(\d+)\.%s' % ext, src, 'height', default=None))
+                tbr = int_or_none(source.get('bitrate'))
                 formats.append({
-                    'format_id': ext + ('-%sp' % height if height else ''),
-                    'url': src,
-                    'height': height,
+                    'format_id': ext + ('-%d' % tbr if tbr else ''),
+                    'url': source_url,
+                    'width': int_or_none(source.get('width')),
+                    'tbr': tbr,
                     'ext': ext,
-                    'preference': 1,
                 })
         self._sort_formats(formats)
 
-        title = self._search_regex(
-            r'share_title\s*=\s*(["\'])(?P<title>[^\1]+?)\1',
-            webpage, 'title', group='title')
-        description = self._search_regex(
-            r'share_description\s*=\s*(["\'])(?P<description>[^\'"]+?)\1',
-            webpage, 'description', default=None, group='description')
-        thumbnail = self._search_regex(
-            r'poster\s*=\s*(["\'])(?P<thumbnail>[^\1]+?)\1',
-            webpage, 'thumbnail', default=False, group='thumbnail')
-
-        uploader_id = self._search_regex(
-            r'twitter_handle\s*=\s*(["\'])(?P<uploader_id>[^\1]+?)\1',
-            webpage, 'uploader id', fatal=False, group='uploader_id')
-        uploader = self._search_regex(
-            r'window\.channelName\s*=\s*(["\'])Embedded:(?P<uploader>[^\1]+?)\1',
-            webpage, 'uploader', default=False, group='uploader')
-
         return {
             'id': video_id,
             'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'uploader': uploader,
-            'uploader_id': uploader_id,
+            'thumbnail': video_data.get('poster_url'),
+            'uploader': video_data.get('channel_name'),
+            'uploader_id': video_data.get('channel_slug'),
+            'duration': float_or_none(video_data.get('duration', 1000)),
+            'tags': video_data.get('tags'),
             'formats': formats,
         }

youtube_dl/extractor/playvid.py

@@ -15,7 +15,7 @@ from ..utils import (
 
 class PlayvidIE(InfoExtractor):
     _VALID_URL = r'https?://www\.playvid\.com/watch(\?v=|/)(?P<id>.+?)(?:#|$)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.playvid.com/watch/RnmBNgtrrJu',
         'md5': 'ffa2f6b2119af359f544388d8c01eb6c',
         'info_dict': {
@@ -24,8 +24,19 @@ class PlayvidIE(InfoExtractor):
             'title': 'md5:9256d01c6317e3f703848b5906880dc8',
             'duration': 82,
             'age_limit': 18,
-        }
-    }
+        },
+        'skip': 'Video removed due to ToS',
+    }, {
+        'url': 'http://www.playvid.com/watch/hwb0GpNkzgH',
+        'md5': '39d49df503ad7b8f23a4432cbf046477',
+        'info_dict': {
+            'id': 'hwb0GpNkzgH',
+            'ext': 'mp4',
+            'title': 'Ellen Euro Cutie Blond Takes a Sexy Survey Get Facial in The Park',
+            'age_limit': 18,
+            'thumbnail': 're:^https?://.*\.jpg$',
+        },
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

youtube_dl/extractor/polskieradio.py

@@ -33,6 +33,7 @@ class PolskieRadioIE(InfoExtractor):
                 'timestamp': 1456594200,
                 'upload_date': '20160227',
                 'duration': 2364,
+                'thumbnail': 're:^https?://static\.prsa\.pl/images/.*\.jpg$'
             },
         }],
     }, {
@@ -68,6 +69,8 @@ class PolskieRadioIE(InfoExtractor):
             r'(?s)<span[^>]+id="datetime2"[^>]*>(.+?)</span>',
             webpage, 'timestamp', fatal=False))
 
+        thumbnail_url = self._og_search_thumbnail(webpage)
+
         entries = []
 
         media_urls = set()
@@ -87,6 +90,7 @@ class PolskieRadioIE(InfoExtractor):
                 'duration': int_or_none(media.get('length')),
                 'vcodec': 'none' if media.get('provider') == 'audio' else None,
                 'timestamp': timestamp,
+                'thumbnail': thumbnail_url
             })
 
         title = self._og_search_title(webpage).strip()

youtube_dl/extractor/pornhub.py

@@ -82,6 +82,10 @@ class PornHubIE(InfoExtractor):
         # removed by uploader
         'url': 'http://www.pornhub.com/view_video.php?viewkey=ph572716d15a111',
         'only_matching': True,
+    }, {
+        # private video
+        'url': 'http://www.pornhub.com/view_video.php?viewkey=ph56fd731fce6b7',
+        'only_matching': True,
     }, {
         'url': 'https://www.thumbzilla.com/video/ph56c6114abd99a/horny-girlfriend-sex',
         'only_matching': True,
@@ -107,7 +111,7 @@ class PornHubIE(InfoExtractor):
         webpage = self._download_webpage(req, video_id)
 
         error_msg = self._html_search_regex(
-            r'(?s)<div[^>]+class=(["\']).*?\bremoved\b.*?\1[^>]*>(?P<error>.+?)</div>',
+            r'(?s)<div[^>]+class=(["\']).*?\b(?:removed|userMessageSection)\b.*?\1[^>]*>(?P<error>.+?)</div>',
             webpage, 'error message', default=None, group='error')
         if error_msg:
             error_msg = re.sub(r'\s+', ' ', error_msg)

youtube_dl/extractor/prosiebensat1.py

@@ -5,7 +5,7 @@ import re
 
 from hashlib import sha1
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_urlencode
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     determine_ext,
@@ -71,6 +71,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'This video is unavailable',
         },
         {
             'url': 'http://www.sixx.de/stars-style/video/sexy-laufen-in-ugg-boots-clip',
@@ -86,6 +87,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'This video is unavailable',
         },
         {
             'url': 'http://www.sat1.de/film/der-ruecktritt/video/im-interview-kai-wiesinger-clip',
@@ -101,6 +103,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'This video is unavailable',
         },
         {
             'url': 'http://www.kabeleins.de/tv/rosins-restaurants/videos/jagd-auf-fertigkost-im-elsthal-teil-2-ganze-folge',
@@ -116,6 +119,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'This video is unavailable',
         },
         {
             'url': 'http://www.ran.de/fussball/bundesliga/video/schalke-toennies-moechte-raul-zurueck-ganze-folge',
@@ -131,6 +135,7 @@ class ProSiebenSat1IE(InfoExtractor):
                 # rtmp download
                 'skip_download': True,
             },
+            'skip': 'This video is unavailable',
         },
         {
             'url': 'http://www.the-voice-of-germany.de/video/31-andreas-kuemmert-rocket-man-clip',
@@ -227,70 +232,42 @@ class ProSiebenSat1IE(InfoExtractor):
     ]
 
     def _extract_clip(self, url, webpage):
-        clip_id = self._html_search_regex(self._CLIPID_REGEXES, webpage, 'clip id')
+        clip_id = self._html_search_regex(
+            self._CLIPID_REGEXES, webpage, 'clip id')
 
         access_token = 'prosieben'
         client_name = 'kolibri-2.0.19-splec4'
         client_location = url
 
-        videos_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos?%s' % compat_urllib_parse_urlencode({
-            'access_token': access_token,
-            'client_location': client_location,
-            'client_name': client_name,
-            'ids': clip_id,
-        })
-
-        video = self._download_json(videos_api_url, clip_id, 'Downloading videos JSON')[0]
+        video = self._download_json(
+            'http://vas.sim-technik.de/vas/live/v2/videos',
+            clip_id, 'Downloading videos JSON', query={
+                'access_token': access_token,
+                'client_location': client_location,
+                'client_name': client_name,
+                'ids': clip_id,
+            })[0]
 
         if video.get('is_protected') is True:
             raise ExtractorError('This video is DRM protected.', expected=True)
 
         duration = float_or_none(video.get('duration'))
-        source_ids = [source['id'] for source in video['sources']]
-        source_ids_str = ','.join(map(str, source_ids))
+        source_ids = [compat_str(source['id']) for source in video['sources']]
 
         g = '01!8d8F_)r9]4s[qeuXfP%'
+        client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name]).encode('utf-8')).hexdigest()
 
-        client_id = g[:2] + sha1(''.join([clip_id, g, access_token, client_location, g, client_name])
-                                 .encode('utf-8')).hexdigest()
-
-        sources_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources?%s' % (clip_id, compat_urllib_parse_urlencode({
-            'access_token': access_token,
-            'client_id': client_id,
-            'client_location': client_location,
-            'client_name': client_name,
-        }))
-
-        sources = self._download_json(sources_api_url, clip_id, 'Downloading sources JSON')
+        sources = self._download_json(
+            'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources' % clip_id,
+            clip_id, 'Downloading sources JSON', query={
+                'access_token': access_token,
+                'client_id': client_id,
+                'client_location': client_location,
+                'client_name': client_name,
+            })
         server_id = sources['server_id']
 
-        client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id,
-                                          client_location, source_ids_str, g, client_name])
-                                 .encode('utf-8')).hexdigest()
-
-        url_api_url = 'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url?%s' % (clip_id, compat_urllib_parse_urlencode({
-            'access_token': access_token,
-            'client_id': client_id,
-            'client_location': client_location,
-            'client_name': client_name,
-            'server_id': server_id,
-            'source_ids': source_ids_str,
-        }))
-
-        urls = self._download_json(url_api_url, clip_id, 'Downloading urls JSON')
-
         title = self._html_search_regex(self._TITLE_REGEXES, webpage, 'title')
-        description = self._html_search_regex(self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
-        thumbnail = self._og_search_thumbnail(webpage)
-
-        upload_date = unified_strdate(self._html_search_regex(
-            self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
-
-        formats = []
-
-        urls_sources = urls['sources']
-        if isinstance(urls_sources, dict):
-            urls_sources = urls_sources.values()
 
         def fix_bitrate(bitrate):
             bitrate = int_or_none(bitrate)
@@ -298,37 +275,73 @@ class ProSiebenSat1IE(InfoExtractor):
                 return None
             return (bitrate // 1000) if bitrate % 1000 == 0 else bitrate
 
-        for source in urls_sources:
-            protocol = source['protocol']
-            source_url = source['url']
-            if protocol == 'rtmp' or protocol == 'rtmpe':
-                mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
-                if not mobj:
+        formats = []
+        for source_id in source_ids:
+            client_id = g[:2] + sha1(''.join([g, clip_id, access_token, server_id, client_location, source_id, g, client_name]).encode('utf-8')).hexdigest()
+            urls = self._download_json(
+                'http://vas.sim-technik.de/vas/live/v2/videos/%s/sources/url' % clip_id,
+                clip_id, 'Downloading urls JSON', fatal=False, query={
+                    'access_token': access_token,
+                    'client_id': client_id,
+                    'client_location': client_location,
+                    'client_name': client_name,
+                    'server_id': server_id,
+                    'source_ids': source_id,
+                })
+            if not urls:
+                continue
+            if urls.get('status_code') != 0:
+                raise ExtractorError('This video is unavailable', expected=True)
+            urls_sources = urls['sources']
+            if isinstance(urls_sources, dict):
+                urls_sources = urls_sources.values()
+            for source in urls_sources:
+                source_url = source.get('url')
+                if not source_url:
                     continue
-                path = mobj.group('path')
-                mp4colon_index = path.rfind('mp4:')
-                app = path[:mp4colon_index]
-                play_path = path[mp4colon_index:]
-                formats.append({
-                    'url': '%s/%s' % (mobj.group('url'), app),
-                    'app': app,
-                    'play_path': play_path,
-                    'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
-                    'page_url': 'http://www.prosieben.de',
-                    'vbr': fix_bitrate(source['bitrate']),
-                    'ext': 'mp4',
-                    'format_id': '%s_%s' % (source['cdn'], source['bitrate']),
-                })
-            elif 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
-                formats.extend(self._extract_f4m_formats(source_url, clip_id))
-            else:
-                formats.append({
-                    'url': source_url,
-                    'vbr': fix_bitrate(source['bitrate']),
-                })
-
+                protocol = source.get('protocol')
+                mimetype = source.get('mimetype')
+                if mimetype == 'application/f4m+xml' or 'f4mgenerator' in source_url or determine_ext(source_url) == 'f4m':
+                    formats.extend(self._extract_f4m_formats(
+                        source_url, clip_id, f4m_id='hds', fatal=False))
+                elif mimetype == 'application/x-mpegURL':
+                    formats.extend(self._extract_m3u8_formats(
+                        source_url, clip_id, 'mp4', 'm3u8_native',
+                        m3u8_id='hls', fatal=False))
+                else:
+                    tbr = fix_bitrate(source['bitrate'])
+                    if protocol in ('rtmp', 'rtmpe'):
+                        mobj = re.search(r'^(?P<url>rtmpe?://[^/]+)/(?P<path>.+)$', source_url)
+                        if not mobj:
+                            continue
+                        path = mobj.group('path')
+                        mp4colon_index = path.rfind('mp4:')
+                        app = path[:mp4colon_index]
+                        play_path = path[mp4colon_index:]
+                        formats.append({
+                            'url': '%s/%s' % (mobj.group('url'), app),
+                            'app': app,
+                            'play_path': play_path,
+                            'player_url': 'http://livepassdl.conviva.com/hf/ver/2.79.0.17083/LivePassModuleMain.swf',
+                            'page_url': 'http://www.prosieben.de',
+                            'tbr': tbr,
+                            'ext': 'flv',
+                            'format_id': 'rtmp%s' % ('-%d' % tbr if tbr else ''),
+                        })
+                    else:
+                        formats.append({
+                            'url': source_url,
+                            'tbr': tbr,
+                            'format_id': 'http%s' % ('-%d' % tbr if tbr else ''),
+                        })
         self._sort_formats(formats)
 
+        description = self._html_search_regex(
+            self._DESCRIPTION_REGEXES, webpage, 'description', fatal=False)
+        thumbnail = self._og_search_thumbnail(webpage)
+        upload_date = unified_strdate(self._html_search_regex(
+            self._UPLOAD_DATE_REGEXES, webpage, 'upload date', default=None))
+
         return {
             'id': clip_id,
             'title': title,

youtube_dl/extractor/radiocanada.py

@@ -12,6 +12,7 @@ from ..utils import (
     unified_strdate,
     xpath_element,
     ExtractorError,
+    determine_protocol,
 )
 
 
@@ -22,13 +23,13 @@ class RadioCanadaIE(InfoExtractor):
         'url': 'http://ici.radio-canada.ca/widgets/mediaconsole/medianet/7184272',
         'info_dict': {
             'id': '7184272',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Le parcours du tireur capté sur vidéo',
             'description': 'Images des caméras de surveillance fournies par la GRC montrant le parcours du tireur d\'Ottawa',
             'upload_date': '20141023',
         },
         'params': {
-            # rtmp download
+            # m3u8 download
             'skip_download': True,
         },
     }
@@ -36,11 +37,14 @@ class RadioCanadaIE(InfoExtractor):
     def _real_extract(self, url):
         app_code, video_id = re.match(self._VALID_URL, url).groups()
 
+        device_types = ['ipad', 'android']
+        if app_code != 'toutv':
+            device_types.append('flash')
+
         formats = []
-        # TODO: extract m3u8 and f4m formats
-        # m3u8 formats can be extracted using ipad device_type return 403 error code when ffmpeg try to download segements
+        # TODO: extract f4m formats
         # f4m formats can be extracted using flashhd device_type but they produce unplayable file
-        for device_type in ('flash',):
+        for device_type in device_types:
             v_data = self._download_xml(
                 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx',
                 video_id, note='Downloading %s XML' % device_type, query={
@@ -52,7 +56,7 @@ class RadioCanadaIE(InfoExtractor):
                     # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
                     'paysJ391wsHjbOJwvCs26toz': 'CA',
                     'bypasslock': 'NZt5K62gRqfc',
-                })
+                }, fatal=False)
             v_url = xpath_text(v_data, 'url')
             if not v_url:
                 continue
@@ -64,7 +68,8 @@ class RadioCanadaIE(InfoExtractor):
                 formats.extend(self._extract_m3u8_formats(
                     v_url, video_id, 'mp4', m3u8_id='hls', fatal=False))
             elif ext == 'f4m':
-                formats.extend(self._extract_f4m_formats(v_url, video_id, f4m_id='hds', fatal=False))
+                formats.extend(self._extract_f4m_formats(
+                    v_url, video_id, f4m_id='hds', fatal=False))
             else:
                 ext = determine_ext(v_url)
                 bitrates = xpath_element(v_data, 'bitrates')
@@ -72,15 +77,28 @@ class RadioCanadaIE(InfoExtractor):
                     tbr = int_or_none(url_e.get('bitrate'))
                     if not tbr:
                         continue
+                    f_url = re.sub(r'\d+\.%s' % ext, '%d.%s' % (tbr, ext), v_url)
+                    protocol = determine_protocol({'url': f_url})
                     formats.append({
-                        'format_id': 'rtmp-%d' % tbr,
-                        'url': re.sub(r'\d+\.%s' % ext, '%d.%s' % (tbr, ext), v_url),
-                        'ext': 'flv',
-                        'protocol': 'rtmp',
+                        'format_id': '%s-%d' % (protocol, tbr),
+                        'url': f_url,
+                        'ext': 'flv' if protocol == 'rtmp' else ext,
+                        'protocol': protocol,
                         'width': int_or_none(url_e.get('width')),
                         'height': int_or_none(url_e.get('height')),
                         'tbr': tbr,
                     })
+                    if protocol == 'rtsp':
+                        base_url = self._search_regex(
+                            r'rtsp://([^?]+)', f_url, 'base url', default=None)
+                        if base_url:
+                            base_url = 'http://' + base_url
+                            formats.extend(self._extract_m3u8_formats(
+                                base_url + '/playlist.m3u8', video_id, 'mp4',
+                                'm3u8_native', m3u8_id='hls', fatal=False))
+                            formats.extend(self._extract_f4m_formats(
+                                base_url + '/manifest.f4m', video_id,
+                                f4m_id='hds', fatal=False))
         self._sort_formats(formats)
 
         metadata = self._download_xml(
@@ -115,13 +133,13 @@ class RadioCanadaAudioVideoIE(InfoExtractor):
         'url': 'http://ici.radio-canada.ca/audio-video/media-7527184/barack-obama-au-vietnam',
         'info_dict': {
             'id': '7527184',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Barack Obama au Vietnam',
             'description': 'Les États-Unis lèvent l\'embargo sur la vente d\'armes qui datait de la guerre du Vietnam',
             'upload_date': '20160523',
         },
         'params': {
-            # rtmp download
+            # m3u8 download
             'skip_download': True,
         },
     }

youtube_dl/extractor/rai.py

@@ -1,47 +1,141 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse,
-    compat_urlparse,
-)
+from ..compat import compat_urlparse
 from ..utils import (
-    ExtractorError,
     determine_ext,
+    ExtractorError,
+    find_xpath_attr,
+    fix_xml_ampersands,
+    int_or_none,
     parse_duration,
     unified_strdate,
-    int_or_none,
+    update_url_query,
     xpath_text,
 )
 
 
-class RaiTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/(?:[^/]+/)+media/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html'
+class RaiBaseIE(InfoExtractor):
+    def _extract_relinker_formats(self, relinker_url, video_id):
+        formats = []
+
+        for platform in ('mon', 'flash', 'native'):
+            relinker = self._download_xml(
+                relinker_url, video_id,
+                note='Downloading XML metadata for platform %s' % platform,
+                transform_source=fix_xml_ampersands,
+                query={'output': 45, 'pl': platform},
+                headers=self.geo_verification_headers())
+
+            media_url = find_xpath_attr(relinker, './url', 'type', 'content').text
+            if media_url == 'http://download.rai.it/video_no_available.mp4':
+                self.raise_geo_restricted()
+
+            ext = determine_ext(media_url)
+            if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'):
+                continue
+
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    media_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
+            elif ext == 'f4m':
+                manifest_url = update_url_query(
+                    media_url.replace('manifest#live_hds.f4m', 'manifest.f4m'),
+                    {'hdcore': '3.7.0', 'plugin': 'aasp-3.7.0.39.44'})
+                formats.extend(self._extract_f4m_formats(
+                    manifest_url, video_id, f4m_id='hds', fatal=False))
+            else:
+                bitrate = int_or_none(xpath_text(relinker, 'bitrate'))
+                formats.append({
+                    'url': media_url,
+                    'tbr': bitrate if bitrate > 0 else None,
+                    'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http',
+                })
+
+        return formats
+
+    def _extract_from_content_id(self, content_id, base_url):
+        media = self._download_json(
+            'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
+            content_id, 'Downloading video JSON')
+
+        thumbnails = []
+        for image_type in ('image', 'image_medium', 'image_300'):
+            thumbnail_url = media.get(image_type)
+            if thumbnail_url:
+                thumbnails.append({
+                    'url': compat_urlparse.urljoin(base_url, thumbnail_url),
+                })
+
+        formats = []
+        media_type = media['type']
+        if 'Audio' in media_type:
+            formats.append({
+                'format_id': media.get('formatoAudio'),
+                'url': media['audioUrl'],
+                'ext': media.get('formatoAudio'),
+            })
+        elif 'Video' in media_type:
+            formats.extend(self._extract_relinker_formats(media['mediaUri'], content_id))
+            self._sort_formats(formats)
+        else:
+            raise ExtractorError('not a media file')
+
+        subtitles = {}
+        captions = media.get('subtitlesUrl')
+        if captions:
+            STL_EXT = '.stl'
+            SRT_EXT = '.srt'
+            if captions.endswith(STL_EXT):
+                captions = captions[:-len(STL_EXT)] + SRT_EXT
+            subtitles['it'] = [{
+                'ext': 'srt',
+                'url': captions,
+            }]
+
+        return {
+            'id': content_id,
+            'title': media['name'],
+            'description': media.get('desc'),
+            'thumbnails': thumbnails,
+            'uploader': media.get('author'),
+            'upload_date': unified_strdate(media.get('date')),
+            'duration': parse_duration(media.get('length')),
+            'formats': formats,
+            'subtitles': subtitles,
+        }
+
+
+class RaiTVIE(RaiBaseIE):
+    _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/(?:[^/]+/)+(?:media|ondemand)/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html'
     _TESTS = [
         {
             'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
-            'md5': '96382709b61dd64a6b88e0f791e6df4c',
+            'md5': '8970abf8caf8aef4696e7b1f2adfc696',
             'info_dict': {
                 'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Report del 07/04/2014',
                 'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
                 'upload_date': '20140407',
                 'duration': 6160,
+                'thumbnail': 're:^https?://.*\.jpg$',
             }
         },
         {
+            # no m3u8 stream
             'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
-            'md5': 'd9751b78eac9710d62c2447b224dea39',
+            # HDS download, MD5 is unstable
             'info_dict': {
                 'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
                 'ext': 'flv',
                 'title': 'TG PRIMO TEMPO',
                 'upload_date': '20140612',
                 'duration': 1758,
+                'thumbnail': 're:^https?://.*\.jpg$',
             },
+            'skip': 'Geo-restricted to Italy',
         },
         {
             'url': 'http://www.rainews.it/dl/rainews/media/state-of-the-net-Antonella-La-Carpia-regole-virali-7aafdea9-0e5d-49d5-88a6-7e65da67ae13.html',
@@ -67,127 +161,70 @@ class RaiTVIE(InfoExtractor):
         },
         {
             'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html',
-            'md5': '496ab63e420574447f70d02578333437',
+            'md5': 'e57493e1cb8bc7c564663f363b171847',
             'info_dict': {
                 'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Il Candidato - Primo episodio: "Le Primarie"',
                 'description': 'md5:364b604f7db50594678f483353164fb8',
                 'upload_date': '20140923',
                 'duration': 386,
+                'thumbnail': 're:^https?://.*\.jpg$',
             }
         },
     ]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        media = self._download_json(
-            'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % video_id,
-            video_id, 'Downloading video JSON')
 
-        thumbnails = []
-        for image_type in ('image', 'image_medium', 'image_300'):
-            thumbnail_url = media.get(image_type)
-            if thumbnail_url:
-                thumbnails.append({
-                    'url': thumbnail_url,
-                })
-
-        subtitles = []
-        formats = []
-        media_type = media['type']
-        if 'Audio' in media_type:
-            formats.append({
-                'format_id': media.get('formatoAudio'),
-                'url': media['audioUrl'],
-                'ext': media.get('formatoAudio'),
-            })
-        elif 'Video' in media_type:
-            def fix_xml(xml):
-                return xml.replace(' tag elementi', '').replace('>/', '</')
-
-            relinker = self._download_xml(
-                media['mediaUri'] + '&output=43',
-                video_id, transform_source=fix_xml)
-
-            has_subtitle = False
-
-            for element in relinker.findall('element'):
-                media_url = xpath_text(element, 'url')
-                ext = determine_ext(media_url)
-                content_type = xpath_text(element, 'content-type')
-                if ext == 'm3u8':
-                    formats.extend(self._extract_m3u8_formats(
-                        media_url, video_id, 'mp4', 'm3u8_native',
-                        m3u8_id='hls', fatal=False))
-                elif ext == 'f4m':
-                    formats.extend(self._extract_f4m_formats(
-                        media_url + '?hdcore=3.7.0&plugin=aasp-3.7.0.39.44',
-                        video_id, f4m_id='hds', fatal=False))
-                elif ext == 'stl':
-                    has_subtitle = True
-                elif content_type.startswith('video/'):
-                    bitrate = int_or_none(xpath_text(element, 'bitrate'))
-                    formats.append({
-                        'url': media_url,
-                        'tbr': bitrate if bitrate > 0 else None,
-                        'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http',
-                    })
-                elif content_type.startswith('image/'):
-                    thumbnails.append({
-                        'url': media_url,
-                    })
-
-            self._sort_formats(formats)
-
-            if has_subtitle:
-                webpage = self._download_webpage(url, video_id)
-                subtitles = self._get_subtitles(video_id, webpage)
-        else:
-            raise ExtractorError('not a media file')
-
-        return {
-            'id': video_id,
-            'title': media['name'],
-            'description': media.get('desc'),
-            'thumbnails': thumbnails,
-            'uploader': media.get('author'),
-            'upload_date': unified_strdate(media.get('date')),
-            'duration': parse_duration(media.get('length')),
-            'formats': formats,
-            'subtitles': subtitles,
-        }
-
-    def _get_subtitles(self, video_id, webpage):
-        subtitles = {}
-        m = re.search(r'<meta name="closedcaption" content="(?P<captions>[^"]+)"', webpage)
-        if m:
-            captions = m.group('captions')
-            STL_EXT = '.stl'
-            SRT_EXT = '.srt'
-            if captions.endswith(STL_EXT):
-                captions = captions[:-len(STL_EXT)] + SRT_EXT
-            subtitles['it'] = [{
-                'ext': 'srt',
-                'url': 'http://www.rai.tv%s' % compat_urllib_parse.quote(captions),
-            }]
-        return subtitles
+        return self._extract_from_content_id(video_id, url)
 
 
-class RaiIE(InfoExtractor):
+class RaiIE(RaiBaseIE):
     _VALID_URL = r'https?://(?:.+?\.)?(?:rai\.it|rai\.tv|rainews\.it)/dl/.+?-(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})(?:-.+?)?\.html'
     _TESTS = [
         {
             'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
-            'md5': 'e0e7a8a131e249d1aa0ebf270d1d8db7',
+            'md5': '2dd727e61114e1ee9c47f0da6914e178',
             'info_dict': {
                 'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Il pacco',
                 'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
                 'upload_date': '20141221',
             },
-        }
+        },
+        {
+            # Direct relinker URL
+            'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
+            # HDS live stream, MD5 is unstable
+            'info_dict': {
+                'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
+                'ext': 'flv',
+                'title': 'EuroNews',
+            },
+            'skip': 'Geo-restricted to Italy',
+        },
+        {
+            # Embedded content item ID
+            'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
+            'md5': '84c1135ce960e8822ae63cec34441d63',
+            'info_dict': {
+                'id': '0960e765-62c8-474a-ac4b-7eb3e2be39c8',
+                'ext': 'mp4',
+                'title': 'TG1 ore 20:00 del 02/07/2016',
+                'upload_date': '20160702',
+            },
+        },
+        {
+            'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
+            # HDS live stream, MD5 is unstable
+            'info_dict': {
+                'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
+                'ext': 'flv',
+                'title': 'La diretta di Rainews24',
+            },
+        },
     ]
 
     @classmethod
@@ -201,7 +238,30 @@ class RaiIE(InfoExtractor):
         iframe_url = self._search_regex(
             [r'<iframe[^>]+src="([^"]*/dl/[^"]+\?iframe\b[^"]*)"',
              r'drawMediaRaiTV\(["\'](.+?)["\']'],
-            webpage, 'iframe')
-        if not iframe_url.startswith('http'):
-            iframe_url = compat_urlparse.urljoin(url, iframe_url)
-        return self.url_result(iframe_url)
+            webpage, 'iframe', default=None)
+        if iframe_url:
+            if not iframe_url.startswith('http'):
+                iframe_url = compat_urlparse.urljoin(url, iframe_url)
+            return self.url_result(iframe_url)
+
+        content_item_id = self._search_regex(
+            r'initEdizione\((?P<q1>[\'"])ContentItem-(?P<content_id>[^\'"]+)(?P=q1)',
+            webpage, 'content item ID', group='content_id', default=None)
+        if content_item_id:
+            return self._extract_from_content_id(content_item_id, url)
+
+        relinker_url = compat_urlparse.urljoin(url, self._search_regex(
+            r'(?:var\s+videoURL|mediaInfo\.mediaUri)\s*=\s*(?P<q1>[\'"])(?P<url>(https?:)?//mediapolis\.rai\.it/relinker/relinkerServlet\.htm\?cont=\d+)(?P=q1)',
+            webpage, 'relinker URL', group='url'))
+        formats = self._extract_relinker_formats(relinker_url, video_id)
+        self._sort_formats(formats)
+
+        title = self._search_regex(
+            r'var\s+videoTitolo\s*=\s*([\'"])(?P<title>[^\'"]+)\1',
+            webpage, 'title', group='title', default=None) or self._og_search_title(webpage)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+        }

youtube_dl/extractor/roosterteeth.py

@@ -0,0 +1,148 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    strip_or_none,
+    unescapeHTML,
+    urlencode_postdata,
+)
+
+
+class RoosterTeethIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:.+?\.)?roosterteeth\.com/episode/(?P<id>[^/?#&]+)'
+    _LOGIN_URL = 'https://roosterteeth.com/login'
+    _NETRC_MACHINE = 'roosterteeth'
+    _TESTS = [{
+        'url': 'http://roosterteeth.com/episode/million-dollars-but-season-2-million-dollars-but-the-game-announcement',
+        'md5': 'e2bd7764732d785ef797700a2489f212',
+        'info_dict': {
+            'id': '26576',
+            'display_id': 'million-dollars-but-season-2-million-dollars-but-the-game-announcement',
+            'ext': 'mp4',
+            'title': 'Million Dollars, But...: Million Dollars, But... The Game Announcement',
+            'description': 'md5:0cc3b21986d54ed815f5faeccd9a9ca5',
+            'thumbnail': 're:^https?://.*\.png$',
+            'series': 'Million Dollars, But...',
+            'episode': 'Million Dollars, But... The Game Announcement',
+            'comment_count': int,
+        },
+    }, {
+        'url': 'http://achievementhunter.roosterteeth.com/episode/off-topic-the-achievement-hunter-podcast-2016-i-didn-t-think-it-would-pass-31',
+        'only_matching': True,
+    }, {
+        'url': 'http://funhaus.roosterteeth.com/episode/funhaus-shorts-2016-austin-sucks-funhaus-shorts',
+        'only_matching': True,
+    }, {
+        'url': 'http://screwattack.roosterteeth.com/episode/death-battle-season-3-mewtwo-vs-shadow',
+        'only_matching': True,
+    }, {
+        'url': 'http://theknow.roosterteeth.com/episode/the-know-game-news-season-1-boring-steam-sales-are-better',
+        'only_matching': True,
+    }, {
+        # only available for FIRST members
+        'url': 'http://roosterteeth.com/episode/rt-docs-the-world-s-greatest-head-massage-the-world-s-greatest-head-massage-an-asmr-journey-part-one',
+        'only_matching': True,
+    }]
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+
+        login_page = self._download_webpage(
+            self._LOGIN_URL, None,
+            note='Downloading login page',
+            errnote='Unable to download login page')
+
+        login_form = self._hidden_inputs(login_page)
+
+        login_form.update({
+            'username': username,
+            'password': password,
+        })
+
+        login_request = self._download_webpage(
+            self._LOGIN_URL, None,
+            note='Logging in as %s' % username,
+            data=urlencode_postdata(login_form),
+            headers={
+                'Referer': self._LOGIN_URL,
+            })
+
+        if not any(re.search(p, login_request) for p in (
+                r'href=["\']https?://(?:www\.)?roosterteeth\.com/logout"',
+                r'>Sign Out<')):
+            error = self._html_search_regex(
+                r'(?s)<div[^>]+class=(["\']).*?\balert-danger\b.*?\1[^>]*>(?:\s*<button[^>]*>.*?</button>)?(?P<error>.+?)</div>',
+                login_request, 'alert', default=None, group='error')
+            if error:
+                raise ExtractorError('Unable to login: %s' % error, expected=True)
+            raise ExtractorError('Unable to log in')
+
+    def _real_initialize(self):
+        self._login()
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        episode = strip_or_none(unescapeHTML(self._search_regex(
+            (r'videoTitle\s*=\s*(["\'])(?P<title>(?:(?!\1).)+)\1',
+             r'<title>(?P<title>[^<]+)</title>'), webpage, 'title',
+            default=None, group='title')))
+
+        title = strip_or_none(self._og_search_title(
+            webpage, default=None)) or episode
+
+        m3u8_url = self._search_regex(
+            r'file\s*:\s*(["\'])(?P<url>http.+?\.m3u8.*?)\1',
+            webpage, 'm3u8 url', default=None, group='url')
+
+        if not m3u8_url:
+            if re.search(r'<div[^>]+class=["\']non-sponsor', webpage):
+                self.raise_login_required(
+                    '%s is only available for FIRST members' % display_id)
+
+            if re.search(r'<div[^>]+class=["\']golive-gate', webpage):
+                self.raise_login_required('%s is not available yet' % display_id)
+
+            raise ExtractorError('Unable to extract m3u8 URL')
+
+        formats = self._extract_m3u8_formats(
+            m3u8_url, display_id, ext='mp4',
+            entry_protocol='m3u8_native', m3u8_id='hls')
+        self._sort_formats(formats)
+
+        description = strip_or_none(self._og_search_description(webpage))
+        thumbnail = self._proto_relative_url(self._og_search_thumbnail(webpage))
+
+        series = self._search_regex(
+            (r'<h2>More ([^<]+)</h2>', r'<a[^>]+>See All ([^<]+) Videos<'),
+            webpage, 'series', fatal=False)
+
+        comment_count = int_or_none(self._search_regex(
+            r'>Comments \((\d+)\)<', webpage,
+            'comment count', fatal=False))
+
+        video_id = self._search_regex(
+            (r'containerId\s*=\s*["\']episode-(\d+)\1',
+             r'<div[^<]+id=["\']episode-(\d+)'), webpage,
+            'video id', default=display_id)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': description,
+            'thumbnail': thumbnail,
+            'series': series,
+            'episode': episode,
+            'comment_count': comment_count,
+            'formats': formats,
+        }

youtube_dl/extractor/rtvnh.py

@@ -9,7 +9,7 @@ class RTVNHIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?rtvnh\.nl/video/(?P<id>[0-9]+)'
     _TEST = {
         'url': 'http://www.rtvnh.nl/video/131946',
-        'md5': '6e1d0ab079e2a00b6161442d3ceacfc1',
+        'md5': 'cdbec9f44550763c8afc96050fa747dc',
         'info_dict': {
             'id': '131946',
             'ext': 'mp4',
@@ -29,15 +29,29 @@ class RTVNHIE(InfoExtractor):
             raise ExtractorError(
                 '%s returned error code %d' % (self.IE_NAME, status), expected=True)
 
-        formats = self._extract_smil_formats(
-            'http://www.rtvnh.nl/video/smil?m=' + video_id, video_id, fatal=False)
+        formats = []
+        rtmp_formats = self._extract_smil_formats(
+            'http://www.rtvnh.nl/video/smil?m=' + video_id, video_id)
+        formats.extend(rtmp_formats)
 
-        for item in meta['source']['fb']:
-            if item.get('type') == 'hls':
-                formats.extend(self._extract_m3u8_formats(
-                    item['file'], video_id, ext='mp4', entry_protocol='m3u8_native'))
-            elif item.get('type') == '':
-                formats.append({'url': item['file']})
+        for rtmp_format in rtmp_formats:
+            rtmp_url = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path'])
+            rtsp_format = rtmp_format.copy()
+            del rtsp_format['play_path']
+            del rtsp_format['ext']
+            rtsp_format.update({
+                'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'),
+                'url': rtmp_url.replace('rtmp://', 'rtsp://'),
+                'protocol': 'rtsp',
+            })
+            formats.append(rtsp_format)
+            http_base_url = rtmp_url.replace('rtmp://', 'http://')
+            formats.extend(self._extract_m3u8_formats(
+                http_base_url + '/playlist.m3u8', video_id, 'mp4',
+                'm3u8_native', m3u8_id='hls', fatal=False))
+            formats.extend(self._extract_f4m_formats(
+                http_base_url + '/manifest.f4m',
+                video_id, f4m_id='hds', fatal=False))
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/rudo.py

@@ -0,0 +1,53 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .jwplatform import JWPlatformBaseIE
+from ..utils import (
+    js_to_json,
+    get_element_by_class,
+    unified_strdate,
+)
+
+
+class RudoIE(JWPlatformBaseIE):
+    _VALID_URL = r'https?://rudo\.video/vod/(?P<id>[0-9a-zA-Z]+)'
+
+    _TEST = {
+        'url': 'http://rudo.video/vod/oTzw0MGnyG',
+        'md5': '2a03a5b32dd90a04c83b6d391cf7b415',
+        'info_dict': {
+            'id': 'oTzw0MGnyG',
+            'ext': 'mp4',
+            'title': 'Comentario Tomás Mosciatti',
+            'upload_date': '20160617',
+        },
+    }
+
+    @classmethod
+    def _extract_url(self, webpage):
+        mobj = re.search(
+            '<iframe[^>]+src=(?P<q1>[\'"])(?P<url>(?:https?:)?//rudo\.video/vod/[0-9a-zA-Z]+)(?P=q1)',
+            webpage)
+        if mobj:
+            return mobj.group('url')
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id, encoding='iso-8859-1')
+
+        jwplayer_data = self._parse_json(self._search_regex(
+            r'(?s)playerInstance\.setup\(({.+?})\)', webpage, 'jwplayer data'), video_id,
+            transform_source=lambda s: js_to_json(re.sub(r'encodeURI\([^)]+\)', '""', s)))
+
+        info_dict = self._parse_jwplayer_data(
+            jwplayer_data, video_id, require_title=False, m3u8_id='hls')
+
+        info_dict.update({
+            'title': self._og_search_title(webpage),
+            'upload_date': unified_strdate(get_element_by_class('date', webpage)),
+        })
+
+        return info_dict

youtube_dl/extractor/sandia.py

@@ -1,18 +1,12 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import itertools
 import json
-import re
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
 from ..utils import (
     int_or_none,
-    js_to_json,
     mimetype2ext,
-    sanitized_Request,
-    unified_strdate,
 )
 
 
@@ -27,7 +21,8 @@ class SandiaIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Xyce Software Training - Section 1',
             'description': 're:(?s)SAND Number: SAND 2013-7800.{200,}',
-            'upload_date': '20120904',
+            'upload_date': '20120409',
+            'timestamp': 1333983600,
             'duration': 7794,
         }
     }
@@ -35,81 +30,36 @@ class SandiaIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        req = sanitized_Request(url)
-        req.add_header('Cookie', 'MediasitePlayerCaps=ClientPlugins=4')
-        webpage = self._download_webpage(req, video_id)
+        presentation_data = self._download_json(
+            'http://digitalops.sandia.gov/Mediasite/PlayerService/PlayerService.svc/json/GetPlayerOptions',
+            video_id, data=json.dumps({
+                'getPlayerOptionsRequest': {
+                    'ResourceId': video_id,
+                    'QueryString': '',
+                }
+            }), headers={
+                'Content-Type': 'application/json; charset=utf-8',
+            })['d']['Presentation']
 
-        js_path = self._search_regex(
-            r'<script type="text/javascript" src="(/Mediasite/FileServer/Presentation/[^"]+)"',
-            webpage, 'JS code URL')
-        js_url = compat_urlparse.urljoin(url, js_path)
-
-        js_code = self._download_webpage(
-            js_url, video_id, note='Downloading player')
-
-        def extract_str(key, **args):
-            return self._search_regex(
-                r'Mediasite\.PlaybackManifest\.%s\s*=\s*(.+);\s*?\n' % re.escape(key),
-                js_code, key, **args)
-
-        def extract_data(key, **args):
-            data_json = extract_str(key, **args)
-            if data_json is None:
-                return data_json
-            return self._parse_json(
-                data_json, video_id, transform_source=js_to_json)
+        title = presentation_data['Title']
 
         formats = []
-        for i in itertools.count():
-            fd = extract_data('VideoUrls[%d]' % i, default=None)
-            if fd is None:
-                break
-            formats.append({
-                'format_id': '%s' % i,
-                'format_note': fd['MimeType'].partition('/')[2],
-                'ext': mimetype2ext(fd['MimeType']),
-                'url': fd['Location'],
-                'protocol': 'f4m' if fd['MimeType'] == 'video/x-mp4-fragmented' else None,
-            })
+        for stream in presentation_data.get('Streams', []):
+            for fd in stream.get('VideoUrls', []):
+                formats.append({
+                    'format_id': fd['MediaType'],
+                    'format_note': fd['MimeType'].partition('/')[2],
+                    'ext': mimetype2ext(fd['MimeType']),
+                    'url': fd['Location'],
+                    'protocol': 'f4m' if fd['MimeType'] == 'video/x-mp4-fragmented' else None,
+                })
         self._sort_formats(formats)
 
-        slide_baseurl = compat_urlparse.urljoin(
-            url, extract_data('SlideBaseUrl'))
-        slide_template = slide_baseurl + re.sub(
-            r'\{0:D?([0-9+])\}', r'%0\1d', extract_data('SlideImageFileNameTemplate'))
-        slides = []
-        last_slide_time = 0
-        for i in itertools.count(1):
-            sd = extract_str('Slides[%d]' % i, default=None)
-            if sd is None:
-                break
-            timestamp = int_or_none(self._search_regex(
-                r'^Mediasite\.PlaybackManifest\.CreateSlide\("[^"]*"\s*,\s*([0-9]+),',
-                sd, 'slide %s timestamp' % i, fatal=False))
-            slides.append({
-                'url': slide_template % i,
-                'duration': timestamp - last_slide_time,
-            })
-            last_slide_time = timestamp
-        formats.append({
-            'format_id': 'slides',
-            'protocol': 'slideshow',
-            'url': json.dumps(slides),
-            'preference': -10000,  # Downloader not yet written
-        })
-        self._sort_formats(formats)
-
-        title = extract_data('Title')
-        description = extract_data('Description', fatal=False)
-        duration = int_or_none(extract_data(
-            'Duration', fatal=False), scale=1000)
-        upload_date = unified_strdate(extract_data('AirDate', fatal=False))
-
         return {
             'id': video_id,
             'title': title,
-            'description': description,
+            'description': presentation_data.get('Description'),
             'formats': formats,
-            'upload_date': upload_date,
-            'duration': duration,
+            'timestamp': int_or_none(presentation_data.get('UnixTime'), 1000),
+            'duration': int_or_none(presentation_data.get('Duration'), 1000),
         }

youtube_dl/extractor/shahid.py

@@ -2,11 +2,11 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_urlencode
 from ..utils import (
     ExtractorError,
     int_or_none,
     parse_iso8601,
+    str_or_none,
 )
 
 
@@ -33,45 +33,27 @@ class ShahidIE(InfoExtractor):
         'only_matching': True
     }]
 
-    def _handle_error(self, response):
-        if not isinstance(response, dict):
-            return
-        error = response.get('error')
+    def _call_api(self, path, video_id, note):
+        data = self._download_json(
+            'http://api.shahid.net/api/v1_1/' + path, video_id, note, query={
+                'apiKey': 'sh@hid0nlin3',
+                'hash': 'b2wMCTHpSmyxGqQjJFOycRmLSex+BpTK/ooxy6vHaqs=',
+            }).get('data', {})
+
+        error = data.get('error')
         if error:
             raise ExtractorError(
                 '%s returned error: %s' % (self.IE_NAME, '\n'.join(error.values())),
                 expected=True)
 
-    def _download_json(self, url, video_id, note='Downloading JSON metadata'):
-        response = super(ShahidIE, self)._download_json(url, video_id, note)['data']
-        self._handle_error(response)
-        return response
+        return data
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
-
-        api_vars = {
-            'id': video_id,
-            'type': 'player',
-            'url': 'http://api.shahid.net/api/v1_1',
-            'playerType': 'episode',
-        }
-
-        flashvars = self._search_regex(
-            r'var\s+flashvars\s*=\s*({[^}]+})', webpage, 'flashvars', default=None)
-        if flashvars:
-            for key in api_vars.keys():
-                value = self._search_regex(
-                    r'\b%s\s*:\s*(?P<q>["\'])(?P<value>.+?)(?P=q)' % key,
-                    flashvars, 'type', default=None, group='value')
-                if value:
-                    api_vars[key] = value
-
-        player = self._download_json(
-            'https://shahid.mbc.net/arContent/getPlayerContent-param-.id-%s.type-%s.html'
-            % (video_id, api_vars['type']), video_id, 'Downloading player JSON')
+        player = self._call_api(
+            'Content/Episode/%s' % video_id,
+            video_id, 'Downloading player JSON')
 
         if player.get('drm'):
             raise ExtractorError('This video is DRM protected.', expected=True)
@@ -79,22 +61,11 @@ class ShahidIE(InfoExtractor):
         formats = self._extract_m3u8_formats(player['url'], video_id, 'mp4')
         self._sort_formats(formats)
 
-        video = self._download_json(
-            '%s/%s/%s?%s' % (
-                api_vars['url'], api_vars['playerType'], api_vars['id'],
-                compat_urllib_parse_urlencode({
-                    'apiKey': 'sh@hid0nlin3',
-                    'hash': 'b2wMCTHpSmyxGqQjJFOycRmLSex+BpTK/ooxy6vHaqs=',
-                })),
-            video_id, 'Downloading video JSON')
-
-        video = video[api_vars['playerType']]
+        video = self._call_api(
+            'episode/%s' % video_id, video_id,
+            'Downloading video JSON')['episode']
 
         title = video['title']
-        description = video.get('description')
-        thumbnail = video.get('thumbnailUrl')
-        duration = int_or_none(video.get('duration'))
-        timestamp = parse_iso8601(video.get('referenceDate'))
         categories = [
             category['name']
             for category in video.get('genres', []) if 'name' in category]
@@ -102,10 +73,16 @@ class ShahidIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'timestamp': timestamp,
+            'description': video.get('description'),
+            'thumbnail': video.get('thumbnailUrl'),
+            'duration': int_or_none(video.get('duration')),
+            'timestamp': parse_iso8601(video.get('referenceDate')),
             'categories': categories,
+            'series': video.get('showTitle') or video.get('showName'),
+            'season': video.get('seasonTitle'),
+            'season_number': int_or_none(video.get('seasonNumber')),
+            'season_id': str_or_none(video.get('seasonId')),
+            'episode_number': int_or_none(video.get('number')),
+            'episode_id': video_id,
             'formats': formats,
         }

youtube_dl/extractor/sixplay.py

@@ -5,6 +5,8 @@ from .common import InfoExtractor
 from ..utils import (
     qualities,
     int_or_none,
+    mimetype2ext,
+    determine_ext,
 )
 
 
@@ -34,19 +36,21 @@ class SixPlayIE(InfoExtractor):
             source_type, source_url = source.get('type'), source.get('src')
             if not source_url or source_type == 'hls/primetime':
                 continue
-            if source_type == 'application/vnd.apple.mpegURL':
+            ext = mimetype2ext(source_type) or determine_ext(source_url)
+            if ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
                     source_url, video_id, 'mp4', 'm3u8_native',
                     m3u8_id='hls', fatal=False))
                 formats.extend(self._extract_f4m_formats(
                     source_url.replace('.m3u8', '.f4m'),
                     video_id, f4m_id='hds', fatal=False))
-            elif source_type == 'video/mp4':
+            elif ext == 'mp4':
                 quality = source.get('quality')
                 formats.append({
                     'url': source_url,
                     'format_id': quality,
                     'quality': quality_key(quality),
+                    'ext': ext,
                 })
         self._sort_formats(formats)
 

youtube_dl/extractor/slideshare.py

@@ -9,6 +9,7 @@ from ..compat import (
 )
 from ..utils import (
     ExtractorError,
+    get_element_by_id,
 )
 
 
@@ -40,7 +41,7 @@ class SlideshareIE(InfoExtractor):
         bucket = info['jsplayer']['video_bucket']
         ext = info['jsplayer']['video_extension']
         video_url = compat_urlparse.urljoin(bucket, doc + '-SD.' + ext)
-        description = self._html_search_regex(
+        description = get_element_by_id('slideshow-description-paragraph', webpage) or self._html_search_regex(
             r'(?s)<p[^>]+itemprop="description"[^>]*>(.+?)</p>', webpage,
             'description', fatal=False)
 
@@ -51,5 +52,5 @@ class SlideshareIE(InfoExtractor):
             'ext': ext,
             'url': video_url,
             'thumbnail': info['slideshow']['pin_image_url'],
-            'description': description,
+            'description': description.strip() if description else None,
         }

youtube_dl/extractor/sohu.py

@@ -8,10 +8,7 @@ from ..compat import (
     compat_str,
     compat_urllib_parse_urlencode,
 )
-from ..utils import (
-    ExtractorError,
-    sanitized_Request,
-)
+from ..utils import ExtractorError
 
 
 class SohuIE(InfoExtractor):
@@ -96,15 +93,10 @@ class SohuIE(InfoExtractor):
             else:
                 base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
 
-            req = sanitized_Request(base_data_url + vid_id)
-
-            cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
-            if cn_verification_proxy:
-                req.add_header('Ytdl-request-proxy', cn_verification_proxy)
-
             return self._download_json(
-                req, video_id,
-                'Downloading JSON data for %s' % vid_id)
+                base_data_url + vid_id, video_id,
+                'Downloading JSON data for %s' % vid_id,
+                headers=self.geo_verification_headers())
 
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')

youtube_dl/extractor/spiegel.py

@@ -4,8 +4,13 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
 from .spiegeltv import SpiegeltvIE
+from ..compat import compat_urlparse
+from ..utils import (
+    extract_attributes,
+    unified_strdate,
+    get_element_by_attribute,
+)
 
 
 class SpiegelIE(InfoExtractor):
@@ -19,6 +24,7 @@ class SpiegelIE(InfoExtractor):
             'title': 'Vulkanausbruch in Ecuador: Der "Feuerschlund" ist wieder aktiv',
             'description': 'md5:8029d8310232196eb235d27575a8b9f4',
             'duration': 49,
+            'upload_date': '20130311',
         },
     }, {
         'url': 'http://www.spiegel.de/video/schach-wm-videoanalyse-des-fuenften-spiels-video-1309159.html',
@@ -29,6 +35,7 @@ class SpiegelIE(InfoExtractor):
             'title': 'Schach-WM in der Videoanalyse: Carlsen nutzt die Fehlgriffe des Titelverteidigers',
             'description': 'md5:c2322b65e58f385a820c10fa03b2d088',
             'duration': 983,
+            'upload_date': '20131115',
         },
     }, {
         'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-embed.html',
@@ -38,6 +45,7 @@ class SpiegelIE(InfoExtractor):
             'ext': 'mp4',
             'description': 'SPIEGEL ONLINE-Nutzer durften den deutschen Astronauten Alexander Gerst über sein Leben auf der ISS-Station befragen. Hier kommen seine Antworten auf die besten sechs Fragen.',
             'title': 'Fragen an Astronaut Alexander Gerst: "Bekommen Sie die Tageszeiten mit?"',
+            'upload_date': '20140904',
         }
     }, {
         'url': 'http://www.spiegel.de/video/astronaut-alexander-gerst-von-der-iss-station-beantwortet-fragen-video-1519126-iframe.html',
@@ -52,10 +60,10 @@ class SpiegelIE(InfoExtractor):
         if SpiegeltvIE.suitable(handle.geturl()):
             return self.url_result(handle.geturl(), 'Spiegeltv')
 
-        title = re.sub(r'\s+', ' ', self._html_search_regex(
-            r'(?s)<(?:h1|div) class="module-title"[^>]*>(.*?)</(?:h1|div)>',
-            webpage, 'title'))
-        description = self._html_search_meta('description', webpage, 'description')
+        video_data = extract_attributes(self._search_regex(r'(<div[^>]+id="spVideoElements"[^>]+>)', webpage, 'video element', default=''))
+
+        title = video_data.get('data-video-title') or get_element_by_attribute('class', 'module-title', webpage)
+        description = video_data.get('data-video-teaser') or self._html_search_meta('description', webpage, 'description')
 
         base_url = self._search_regex(
             [r'server\s*:\s*(["\'])(?P<url>.+?)\1', r'var\s+server\s*=\s*"(?P<url>[^"]+)\"'],
@@ -87,8 +95,9 @@ class SpiegelIE(InfoExtractor):
         return {
             'id': video_id,
             'title': title,
-            'description': description,
+            'description': description.strip() if description else None,
             'duration': duration,
+            'upload_date': unified_strdate(video_data.get('data-video-date')),
             'formats': formats,
         }
 
@@ -104,6 +113,7 @@ class SpiegelArticleIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Faszination Badminton: Nennt es bloß nicht Federball',
             'description': 're:^Patrick Kämnitz gehört.{100,}',
+            'upload_date': '20140825',
         },
     }, {
         'url': 'http://www.spiegel.de/wissenschaft/weltall/astronaut-alexander-gerst-antwortet-spiegel-online-lesern-a-989876.html',

youtube_dl/extractor/srmediathek.py

@@ -11,7 +11,7 @@ from ..utils import (
 class SRMediathekIE(ARDMediathekIE):
     IE_NAME = 'sr:mediathek'
     IE_DESC = 'Saarländischer Rundfunk'
-    _VALID_URL = r'https?://sr-mediathek\.sr-online\.de/index\.php\?.*?&id=(?P<id>[0-9]+)'
+    _VALID_URL = r'https?://sr-mediathek(?:\.sr-online)?\.de/index\.php\?.*?&id=(?P<id>[0-9]+)'
 
     _TESTS = [{
         'url': 'http://sr-mediathek.sr-online.de/index.php?seite=7&id=28455',
@@ -35,7 +35,9 @@ class SRMediathekIE(ARDMediathekIE):
             # m3u8 download
             'skip_download': True,
         },
-        'expected_warnings': ['Unable to download f4m manifest']
+    }, {
+        'url': 'http://sr-mediathek.de/index.php?seite=7&id=7480',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/stitcher.py

@@ -56,7 +56,7 @@ class StitcherIE(InfoExtractor):
 
         episode = self._parse_json(
             js_to_json(self._search_regex(
-                r'(?s)var\s+stitcher\s*=\s*({.+?});\n', webpage, 'episode config')),
+                r'(?s)var\s+stitcher(?:Config)?\s*=\s*({.+?});\n', webpage, 'episode config')),
             display_id)['config']['episode']
 
         title = unescapeHTML(episode['title'])

youtube_dl/extractor/telecinco.py

@@ -1,50 +1,41 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import json
-
-from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse_unquote,
-    compat_urllib_parse_urlencode,
-    compat_urlparse,
-)
-from ..utils import (
-    get_element_by_attribute,
-    parse_duration,
-    strip_jsonp,
-)
+from .mitele import MiTeleBaseIE
 
 
-class TelecincoIE(InfoExtractor):
+class TelecincoIE(MiTeleBaseIE):
     IE_DESC = 'telecinco.es, cuatro.com and mediaset.es'
     _VALID_URL = r'https?://www\.(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html'
 
     _TESTS = [{
         'url': 'http://www.telecinco.es/robinfood/temporada-01/t01xp14/Bacalao-cocochas-pil-pil_0_1876350223.html',
-        'md5': '5cbef3ad5ef17bf0d21570332d140729',
+        'md5': '8d7b2d5f699ee2709d992a63d5cd1712',
         'info_dict': {
-            'id': 'MDSVID20141015_0058',
+            'id': 'JEA5ijCnF6p5W08A1rNKn7',
             'ext': 'mp4',
-            'title': 'Con Martín Berasategui, hacer un bacalao al ...',
+            'title': 'Bacalao con kokotxas al pil-pil',
+            'description': 'md5:1382dacd32dd4592d478cbdca458e5bb',
             'duration': 662,
         },
     }, {
         'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
-        'md5': '0a5b9f3cc8b074f50a0578f823a12694',
+        'md5': '284393e5387b3b947b77c613ef04749a',
         'info_dict': {
-            'id': 'MDSVID20150916_0128',
+            'id': 'jn24Od1zGLG4XUZcnUnZB6',
             'ext': 'mp4',
-            'title': '¿Quién es este ex futbolista con el que hablan ...',
+            'title': '¿Quién es este ex futbolista con el que hablan Leo Messi y Luis Suárez?',
+            'description': 'md5:a62ecb5f1934fc787107d7b9a2262805',
             'duration': 79,
         },
     }, {
         'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html',
-        'md5': 'ad1bfaaba922dd4a295724b05b68f86a',
+        'md5': '749afab6ea5a136a8806855166ae46a2',
         'info_dict': {
-            'id': 'MDSVID20150513_0220',
+            'id': 'aywerkD2Sv1vGNqq9b85Q2',
             'ext': 'mp4',
             'title': '#DOYLACARA. Con la trata no hay trato',
+            'description': 'md5:2771356ff7bfad9179c5f5cd954f1477',
             'duration': 50,
         },
     }, {
@@ -56,40 +47,16 @@ class TelecincoIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        episode = self._match_id(url)
-        webpage = self._download_webpage(url, episode)
-        embed_data_json = self._search_regex(
-            r'(?s)MSV\.embedData\[.*?\]\s*=\s*({.*?});', webpage, 'embed data',
-        ).replace('\'', '"')
-        embed_data = json.loads(embed_data_json)
-
-        domain = embed_data['mediaUrl']
-        if not domain.startswith('http'):
-            # only happens in telecinco.es videos
-            domain = 'http://' + domain
-        info_url = compat_urlparse.urljoin(
-            domain,
-            compat_urllib_parse_unquote(embed_data['flashvars']['host'])
-        )
-        info_el = self._download_xml(info_url, episode).find('./video/info')
-
-        video_link = info_el.find('videoUrl/link').text
-        token_query = compat_urllib_parse_urlencode({'id': video_link})
-        token_info = self._download_json(
-            embed_data['flashvars']['ov_tk'] + '?' + token_query,
-            episode,
-            transform_source=strip_jsonp
-        )
-        formats = self._extract_m3u8_formats(
-            token_info['tokenizedUrl'], episode, ext='mp4', entry_protocol='m3u8_native')
-        self._sort_formats(formats)
-
-        return {
-            'id': embed_data['videoId'],
-            'display_id': episode,
-            'title': info_el.find('title').text,
-            'formats': formats,
-            'description': get_element_by_attribute('class', 'text', webpage),
-            'thumbnail': info_el.find('thumb').text,
-            'duration': parse_duration(info_el.find('duration').text),
-        }
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        title = self._html_search_meta(
+            ['og:title', 'twitter:title'], webpage, 'title')
+        info = self._get_player_info(url, webpage)
+        info.update({
+            'display_id': display_id,
+            'title': title,
+            'description': self._html_search_meta(
+                ['og:description', 'twitter:description'],
+                webpage, 'title', fatal=False),
+        })
+        return info

youtube_dl/extractor/theplatform.py

@@ -6,6 +6,7 @@ import time
 import hmac
 import binascii
 import hashlib
+import netrc
 
 
 from .once import OnceIE
@@ -24,6 +25,9 @@ from ..utils import (
     xpath_with_ns,
     mimetype2ext,
     find_xpath_attr,
+    unescapeHTML,
+    urlencode_postdata,
+    unified_timestamp,
 )
 
 default_ns = 'http://www.w3.org/2005/SMIL21/Language'
@@ -62,10 +66,11 @@ class ThePlatformBaseIE(OnceIE):
 
         return formats, subtitles
 
-    def get_metadata(self, path, video_id):
+    def _download_theplatform_metadata(self, path, video_id):
         info_url = 'http://link.theplatform.com/s/%s?format=preview' % path
-        info = self._download_json(info_url, video_id)
+        return self._download_json(info_url, video_id)
 
+    def _parse_theplatform_metadata(self, info):
         subtitles = {}
         captions = info.get('captions')
         if isinstance(captions, list):
@@ -86,6 +91,10 @@ class ThePlatformBaseIE(OnceIE):
             'uploader': info.get('billingCode'),
         }
 
+    def _extract_theplatform_metadata(self, path, video_id):
+        info = self._download_theplatform_metadata(path, video_id)
+        return self._parse_theplatform_metadata(info)
+
 
 class ThePlatformIE(ThePlatformBaseIE):
     _VALID_URL = r'''(?x)
@@ -158,6 +167,7 @@ class ThePlatformIE(ThePlatformBaseIE):
         'url': 'http://player.theplatform.com/p/NnzsPC/onsite_universal/select/media/guid/2410887629/2928790?fwsitesection=nbc_the_blacklist_video_library&autoPlay=true&carouselID=137781',
         'only_matching': True,
     }]
+    _SERVICE_PROVIDER_TEMPLATE = 'https://sp.auth.adobe.com/adobe-services/%s'
 
     @classmethod
     def _extract_urls(cls, webpage):
@@ -192,6 +202,96 @@ class ThePlatformIE(ThePlatformBaseIE):
         sig = flags + expiration_date + checksum + str_to_hex(sig_secret)
         return '%s&sig=%s' % (url, sig)
 
+    def _extract_mvpd_auth(self, url, video_id, requestor_id, resource):
+        def xml_text(xml_str, tag):
+            return self._search_regex(
+                '<%s>(.+?)</%s>' % (tag, tag), xml_str, tag)
+
+        mvpd_headers = {
+            'ap_42': 'anonymous',
+            'ap_11': 'Linux i686',
+            'ap_z': 'Mozilla/5.0 (X11; Linux i686; rv:47.0) Gecko/20100101 Firefox/47.0',
+            'User-Agent': 'Mozilla/5.0 (X11; Linux i686; rv:47.0) Gecko/20100101 Firefox/47.0',
+        }
+
+        guid = xml_text(resource, 'guid')
+        requestor_info = self._downloader.cache.load('mvpd', requestor_id) or {}
+        authn_token = requestor_info.get('authn_token')
+        if authn_token:
+            token_expires = unified_timestamp(xml_text(authn_token, 'simpleTokenExpires').replace('_GMT', ''))
+            if token_expires and token_expires >= time.time():
+                authn_token = None
+        if not authn_token:
+            # TODO add support for other TV Providers
+            mso_id = 'DTV'
+            login_info = netrc.netrc().authenticators(mso_id)
+            if not login_info:
+                return None
+
+            def post_form(form_page, note, data={}):
+                post_url = self._html_search_regex(r'<form[^>]+action=(["\'])(?P<url>.+?)\1', form_page, 'post url', group='url')
+                return self._download_webpage(
+                    post_url, video_id, note, data=urlencode_postdata(data or self._hidden_inputs(form_page)), headers={
+                        'Content-Type': 'application/x-www-form-urlencoded',
+                    })
+
+            provider_redirect_page = self._download_webpage(
+                self._SERVICE_PROVIDER_TEMPLATE % 'authenticate/saml', video_id,
+                'Downloading Provider Redirect Page', query={
+                    'noflash': 'true',
+                    'mso_id': mso_id,
+                    'requestor_id': requestor_id,
+                    'no_iframe': 'false',
+                    'domain_name': 'adobe.com',
+                    'redirect_url': url,
+                })
+            provider_login_page = post_form(
+                provider_redirect_page, 'Downloading Provider Login Page')
+            mvpd_confirm_page = post_form(provider_login_page, 'Logging in', {
+                'username': login_info[0],
+                'password': login_info[2],
+            })
+            post_form(mvpd_confirm_page, 'Confirming Login')
+
+            session = self._download_webpage(
+                self._SERVICE_PROVIDER_TEMPLATE % 'session', video_id,
+                'Retrieving Session', data=urlencode_postdata({
+                    '_method': 'GET',
+                    'requestor_id': requestor_id,
+                }), headers=mvpd_headers)
+            authn_token = unescapeHTML(xml_text(session, 'authnToken'))
+            requestor_info['authn_token'] = authn_token
+            self._downloader.cache.store('mvpd', requestor_id, requestor_info)
+
+        authz_token = requestor_info.get(guid)
+        if not authz_token:
+            authorize = self._download_webpage(
+                self._SERVICE_PROVIDER_TEMPLATE % 'authorize', video_id,
+                'Retrieving Authorization Token', data=urlencode_postdata({
+                    'resource_id': resource,
+                    'requestor_id': requestor_id,
+                    'authentication_token': authn_token,
+                    'mso_id': xml_text(authn_token, 'simpleTokenMsoID'),
+                    'userMeta': '1',
+                }), headers=mvpd_headers)
+            authz_token = unescapeHTML(xml_text(authorize, 'authzToken'))
+            requestor_info[guid] = authz_token
+            self._downloader.cache.store('mvpd', requestor_id, requestor_info)
+
+        mvpd_headers.update({
+            'ap_19': xml_text(authn_token, 'simpleSamlNameID'),
+            'ap_23': xml_text(authn_token, 'simpleSamlSessionIndex'),
+        })
+
+        return self._download_webpage(
+            self._SERVICE_PROVIDER_TEMPLATE % 'shortAuthorize',
+            video_id, 'Retrieving Media Token', data=urlencode_postdata({
+                'authz_token': authz_token,
+                'requestor_id': requestor_id,
+                'session_guid': xml_text(authn_token, 'simpleTokenAuthenticationGuid'),
+                'hashed_guid': 'false',
+            }), headers=mvpd_headers)
+
     def _real_extract(self, url):
         url, smuggled_data = unsmuggle_url(url, {})
 
@@ -265,7 +365,7 @@ class ThePlatformIE(ThePlatformBaseIE):
         formats, subtitles = self._extract_theplatform_smil(smil_url, video_id)
         self._sort_formats(formats)
 
-        ret = self.get_metadata(path, video_id)
+        ret = self._extract_theplatform_metadata(path, video_id)
         combined_subtitles = self._merge_subtitles(ret.get('subtitles', {}), subtitles)
         ret.update({
             'id': video_id,
@@ -339,7 +439,7 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
         timestamp = int_or_none(entry.get('media$availableDate'), scale=1000)
         categories = [item['media$name'] for item in entry.get('media$categories', [])]
 
-        ret = self.get_metadata('%s/%s' % (provider_id, first_video_id), video_id)
+        ret = self._extract_theplatform_metadata('%s/%s' % (provider_id, first_video_id), video_id)
         subtitles = self._merge_subtitles(subtitles, ret['subtitles'])
         ret.update({
             'id': video_id,

youtube_dl/extractor/threeqsdn.py

@@ -92,12 +92,11 @@ class ThreeQSDNIE(InfoExtractor):
             if not item_url or item_url in urls:
                 return
             urls.add(item_url)
-            type_ = item.get('type')
-            ext = determine_ext(item_url, default_ext=None)
-            if type_ == 'application/dash+xml' or ext == 'mpd':
+            ext = mimetype2ext(item.get('type')) or determine_ext(item_url, default_ext=None)
+            if ext == 'mpd':
                 formats.extend(self._extract_mpd_formats(
                     item_url, video_id, mpd_id='mpd', fatal=False))
-            elif type_ in ('application/vnd.apple.mpegURL', 'application/x-mpegurl') or ext == 'm3u8':
+            elif ext == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
                     item_url, video_id, 'mp4',
                     entry_protocol='m3u8' if live else 'm3u8_native',
@@ -111,7 +110,7 @@ class ThreeQSDNIE(InfoExtractor):
                 formats.append({
                     'url': item_url,
                     'format_id': item.get('quality'),
-                    'ext': 'mp4' if item_url.startswith('rtsp') else mimetype2ext(type_) or ext,
+                    'ext': 'mp4' if item_url.startswith('rtsp') else ext,
                     'vcodec': 'none' if stream_type == 'audio' else None,
                 })
 

youtube_dl/extractor/tmz.py

@@ -5,31 +5,27 @@ from .common import InfoExtractor
 
 
 class TMZIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?tmz\.com/videos/(?P<id>[^/]+)/?'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?tmz\.com/videos/(?P<id>[^/?#]+)'
+    _TESTS = [{
         'url': 'http://www.tmz.com/videos/0_okj015ty/',
-        'md5': '791204e3bf790b1426cb2db0706184c0',
+        'md5': '4d22a51ef205b6c06395d8394f72d560',
         'info_dict': {
             'id': '0_okj015ty',
-            'url': 'http://tmz.vo.llnwd.net/o28/2014-03/13/0_okj015ty_0_rt8ro3si_2.mp4',
             'ext': 'mp4',
             'title': 'Kim Kardashian\'s Boobs Unlock a Mystery!',
             'description': 'Did Kim Kardasain try to one-up Khloe by one-upping Kylie???  Or is she just showing off her amazing boobs?',
-            'thumbnail': r're:http://cdnbakmi\.kaltura\.com/.*thumbnail.*',
+            'timestamp': 1394747163,
+            'uploader_id': 'batchUser',
+            'upload_date': '20140313',
         }
-    }
+    }, {
+        'url': 'http://www.tmz.com/videos/0-cegprt2p/',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        return {
-            'id': video_id,
-            'url': self._html_search_meta('VideoURL', webpage, fatal=True),
-            'title': self._og_search_title(webpage),
-            'description': self._og_search_description(webpage),
-            'thumbnail': self._html_search_meta('ThumbURL', webpage),
-        }
+        video_id = self._match_id(url).replace('-', '_')
+        return self.url_result('kaltura:591531:%s' % video_id, 'Kaltura', video_id)
 
 
 class TMZArticleIE(InfoExtractor):

youtube_dl/extractor/toutv.py

@@ -1,74 +1,41 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..utils import (
-    ExtractorError,
-    unified_strdate,
-)
+from ..utils import int_or_none
 
 
 class TouTvIE(InfoExtractor):
     IE_NAME = 'tou.tv'
-    _VALID_URL = r'https?://www\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+(?:/(?P<episode>S[0-9]+E[0-9]+)))'
+    _VALID_URL = r'https?://ici\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+/S[0-9]+E[0-9]+)'
 
     _TEST = {
-        'url': 'http://www.tou.tv/30-vies/S04E41',
+        'url': 'http://ici.tou.tv/garfield-tout-court/S2015E17',
         'info_dict': {
-            'id': '30-vies_S04E41',
+            'id': '122017',
             'ext': 'mp4',
-            'title': '30 vies Saison 4 / Épisode 41',
-            'description': 'md5:da363002db82ccbe4dafeb9cab039b09',
-            'age_limit': 8,
-            'uploader': 'Groupe des Nouveaux Médias',
-            'duration': 1296,
-            'upload_date': '20131118',
-            'thumbnail': 'http://static.tou.tv/medias/images/2013-11-18_19_00_00_30VIES_0341_01_L.jpeg',
+            'title': 'Saison 2015 Épisode 17',
+            'description': 'La photo de famille 2',
+            'upload_date': '20100717',
         },
         'params': {
-            'skip_download': True,  # Requires rtmpdump
+            # m3u8 download
+            'skip_download': True,
         },
-        'skip': 'Only available in Canada'
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-
-        mediaId = self._search_regex(
-            r'"idMedia":\s*"([^"]+)"', webpage, 'media ID')
-
-        streams_url = 'http://release.theplatform.com/content.select?pid=' + mediaId
-        streams_doc = self._download_xml(
-            streams_url, video_id, note='Downloading stream list')
-
-        video_url = next(n.text
-                         for n in streams_doc.findall('.//choice/url')
-                         if '//ad.doubleclick' not in n.text)
-        if video_url.endswith('/Unavailable.flv'):
-            raise ExtractorError(
-                'Access to this video is blocked from outside of Canada',
-                expected=True)
-
-        duration_str = self._html_search_meta(
-            'video:duration', webpage, 'duration')
-        duration = int(duration_str) if duration_str else None
-        upload_date_str = self._html_search_meta(
-            'video:release_date', webpage, 'upload date')
-        upload_date = unified_strdate(upload_date_str) if upload_date_str else None
+        path = self._match_id(url)
+        metadata = self._download_json('http://ici.tou.tv/presentation/%s' % path, path)
+        video_id = metadata['IdMedia']
+        details = metadata['Details']
+        title = details['OriginalTitle']
 
         return {
+            '_type': 'url_transparent',
+            'url': 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id),
             'id': video_id,
-            'title': self._og_search_title(webpage),
-            'url': video_url,
-            'description': self._og_search_description(webpage),
-            'uploader': self._dc_search_uploader(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'age_limit': self._media_rating_search(webpage),
-            'duration': duration,
-            'upload_date': upload_date,
-            'ext': 'mp4',
+            'title': title,
+            'thumbnail': details.get('ImageUrl'),
+            'duration': int_or_none(details.get('LengthInSeconds')),
         }

youtube_dl/extractor/tvplay.py

@@ -8,43 +8,36 @@ from ..compat import compat_str
 from ..utils import (
     parse_iso8601,
     qualities,
+    determine_ext,
+    update_url_query,
+    int_or_none,
 )
 
 
 class TVPlayIE(InfoExtractor):
     IE_DESC = 'TV3Play and related services'
     _VALID_URL = r'''(?x)https?://(?:www\.)?
-        (?:tvplay\.lv/parraides|
-           tv3play\.lt/programos|
-           play\.tv3\.lt/programos|
-           tv3play\.ee/sisu|
-           tv3play\.se/program|
-           tv6play\.se/program|
-           tv8play\.se/program|
-           tv10play\.se/program|
-           tv3play\.no/programmer|
-           viasat4play\.no/programmer|
-           tv6play\.no/programmer|
-           tv3play\.dk/programmer|
+        (?:tvplay(?:\.skaties)?\.lv/parraides|
+           (?:tv3play|play\.tv3)\.lt/programos|
+           tv3play(?:\.tv3)?\.ee/sisu|
+           tv(?:3|6|8|10)play\.se/program|
+           (?:(?:tv3play|viasat4play|tv6play)\.no|tv3play\.dk)/programmer|
            play\.novatv\.bg/programi
         )/[^/]+/(?P<id>\d+)
         '''
     _TESTS = [
         {
             'url': 'http://www.tvplay.lv/parraides/vinas-melo-labak/418113?autostart=true',
+            'md5': 'a1612fe0849455423ad8718fe049be21',
             'info_dict': {
                 'id': '418113',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Kādi ir īri? - Viņas melo labāk',
                 'description': 'Baiba apsmej īrus, kādi tie ir un ko viņi dara.',
                 'duration': 25,
                 'timestamp': 1406097056,
                 'upload_date': '20140723',
             },
-            'params': {
-                # rtmp download
-                'skip_download': True,
-            },
         },
         {
             'url': 'http://play.tv3.lt/programos/moterys-meluoja-geriau/409229?autostart=true',
@@ -82,7 +75,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.tv3play.se/program/husraddarna/395385?autostart=true',
             'info_dict': {
                 'id': '395385',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Husräddarna S02E07',
                 'description': 'md5:f210c6c89f42d4fc39faa551be813777',
                 'duration': 2574,
@@ -90,7 +83,6 @@ class TVPlayIE(InfoExtractor):
                 'upload_date': '20140520',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -98,7 +90,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.tv6play.se/program/den-sista-dokusapan/266636?autostart=true',
             'info_dict': {
                 'id': '266636',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Den sista dokusåpan S01E08',
                 'description': 'md5:295be39c872520221b933830f660b110',
                 'duration': 1492,
@@ -107,7 +99,6 @@ class TVPlayIE(InfoExtractor):
                 'age_limit': 18,
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -115,7 +106,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.tv8play.se/program/antikjakten/282756?autostart=true',
             'info_dict': {
                 'id': '282756',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Antikjakten S01E10',
                 'description': 'md5:1b201169beabd97e20c5ad0ad67b13b8',
                 'duration': 2646,
@@ -123,7 +114,6 @@ class TVPlayIE(InfoExtractor):
                 'upload_date': '20120925',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -131,7 +121,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.tv3play.no/programmer/anna-anka-soker-assistent/230898?autostart=true',
             'info_dict': {
                 'id': '230898',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Anna Anka søker assistent - Ep. 8',
                 'description': 'md5:f80916bf5bbe1c5f760d127f8dd71474',
                 'duration': 2656,
@@ -139,7 +129,6 @@ class TVPlayIE(InfoExtractor):
                 'upload_date': '20100628',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -147,7 +136,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.viasat4play.no/programmer/budbringerne/21873?autostart=true',
             'info_dict': {
                 'id': '21873',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Budbringerne program 10',
                 'description': 'md5:4db78dc4ec8a85bb04fd322a3ee5092d',
                 'duration': 1297,
@@ -155,7 +144,6 @@ class TVPlayIE(InfoExtractor):
                 'upload_date': '20090929',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -163,7 +151,7 @@ class TVPlayIE(InfoExtractor):
             'url': 'http://www.tv6play.no/programmer/hotelinspektor-alex-polizzi/361883?autostart=true',
             'info_dict': {
                 'id': '361883',
-                'ext': 'flv',
+                'ext': 'mp4',
                 'title': 'Hotelinspektør Alex Polizzi - Ep. 10',
                 'description': 'md5:3ecf808db9ec96c862c8ecb3a7fdaf81',
                 'duration': 2594,
@@ -171,7 +159,6 @@ class TVPlayIE(InfoExtractor):
                 'upload_date': '20140224',
             },
             'params': {
-                # rtmp download
                 'skip_download': True,
             },
         },
@@ -191,6 +178,14 @@ class TVPlayIE(InfoExtractor):
                 'skip_download': True,
             },
         },
+        {
+            'url': 'http://tvplay.skaties.lv/parraides/vinas-melo-labak/418113?autostart=true',
+            'only_matching': True,
+        },
+        {
+            'url': 'http://tv3play.tv3.ee/sisu/kodu-keset-linna/238551?autostart=true',
+            'only_matching': True,
+        }
     ]
 
     def _real_extract(self, url):
@@ -199,7 +194,9 @@ class TVPlayIE(InfoExtractor):
         video = self._download_json(
             'http://playapi.mtgx.tv/v1/videos/%s' % video_id, video_id, 'Downloading video JSON')
 
-        if video['is_geo_blocked']:
+        title = video['title']
+
+        if video.get('is_geo_blocked'):
             self.report_warning(
                 'This content might not be available in your country due to copyright reasons')
 
@@ -208,42 +205,50 @@ class TVPlayIE(InfoExtractor):
 
         quality = qualities(['hls', 'medium', 'high'])
         formats = []
-        for format_id, video_url in streams['streams'].items():
+        for format_id, video_url in streams.get('streams', {}).items():
             if not video_url or not isinstance(video_url, compat_str):
                 continue
-            fmt = {
-                'format_id': format_id,
-                'preference': quality(format_id),
-            }
-            if video_url.startswith('rtmp'):
-                m = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', video_url)
-                if not m:
-                    continue
-                fmt.update({
-                    'ext': 'flv',
-                    'url': m.group('url'),
-                    'app': m.group('app'),
-                    'play_path': m.group('playpath'),
-                })
-            elif video_url.endswith('.f4m'):
+            ext = determine_ext(video_url)
+            if ext == 'f4m':
                 formats.extend(self._extract_f4m_formats(
-                    video_url + '?hdcore=3.5.0&plugin=aasp-3.5.0.151.81', video_id))
-                continue
+                    update_url_query(video_url, {
+                        'hdcore': '3.5.0',
+                        'plugin': 'aasp-3.5.0.151.81'
+                    }), video_id, f4m_id='hds', fatal=False))
+            elif ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    video_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False))
             else:
-                fmt.update({
-                    'url': video_url,
-                })
-            formats.append(fmt)
-
+                fmt = {
+                    'format_id': format_id,
+                    'quality': quality(format_id),
+                    'ext': ext,
+                }
+                if video_url.startswith('rtmp'):
+                    m = re.search(r'^(?P<url>rtmp://[^/]+/(?P<app>[^/]+))/(?P<playpath>.+)$', video_url)
+                    if not m:
+                        continue
+                    fmt.update({
+                        'ext': 'flv',
+                        'url': m.group('url'),
+                        'app': m.group('app'),
+                        'play_path': m.group('playpath'),
+                    })
+                else:
+                    fmt.update({
+                        'url': video_url,
+                    })
+                formats.append(fmt)
         self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'title': video['title'],
-            'description': video['description'],
-            'duration': video['duration'],
-            'timestamp': parse_iso8601(video['created_at']),
-            'view_count': video['views']['total'],
-            'age_limit': video.get('age_limit', 0),
+            'title': title,
+            'description': video.get('description'),
+            'duration': int_or_none(video.get('duration')),
+            'timestamp': parse_iso8601(video.get('created_at')),
+            'view_count': int_or_none(video.get('views', {}).get('total')),
+            'age_limit': int_or_none(video.get('age_limit', 0)),
             'formats': formats,
         }

youtube_dl/extractor/tweakers.py

@@ -1,25 +1,62 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    determine_ext,
+    mimetype2ext,
+)
 
 
 class TweakersIE(InfoExtractor):
     _VALID_URL = r'https?://tweakers\.net/video/(?P<id>\d+)'
     _TEST = {
         'url': 'https://tweakers.net/video/9926/new-nintendo-3ds-xl-op-alle-fronten-beter.html',
-        'md5': '3147e4ddad366f97476a93863e4557c8',
+        'md5': 'fe73e417c093a788e0160c4025f88b15',
         'info_dict': {
             'id': '9926',
             'ext': 'mp4',
             'title': 'New Nintendo 3DS XL - Op alle fronten beter',
-            'description': 'md5:f97324cc71e86e11c853f0763820e3ba',
+            'description': 'md5:3789b21fed9c0219e9bcaacd43fab280',
             'thumbnail': 're:^https?://.*\.jpe?g$',
             'duration': 386,
+            'uploader_id': 's7JeEm',
         }
     }
 
     def _real_extract(self, url):
-        playlist_id = self._match_id(url)
-        entries = self._extract_xspf_playlist(
-            'https://tweakers.net/video/s1playlist/%s/playlist.xspf' % playlist_id, playlist_id)
-        return self.playlist_result(entries, playlist_id)
+        video_id = self._match_id(url)
+        video_data = self._download_json(
+            'https://tweakers.net/video/s1playlist/%s/1920/1080/playlist.json' % video_id,
+            video_id)['items'][0]
+
+        title = video_data['title']
+
+        formats = []
+        for location in video_data.get('locations', {}).get('progressive', []):
+            format_id = location.get('label')
+            width = int_or_none(location.get('width'))
+            height = int_or_none(location.get('height'))
+            for source in location.get('sources', []):
+                source_url = source.get('src')
+                if not source_url:
+                    continue
+                ext = mimetype2ext(source.get('type')) or determine_ext(source_url)
+                formats.append({
+                    'format_id': format_id,
+                    'url': source_url,
+                    'width': width,
+                    'height': height,
+                    'ext': ext,
+                })
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': video_data.get('description'),
+            'thumbnail': video_data.get('poster'),
+            'duration': int_or_none(video_data.get('duration')),
+            'uploader_id': video_data.get('account'),
+            'formats': formats,
+        }

youtube_dl/extractor/vidzi.py

@@ -9,8 +9,8 @@ from ..utils import (
 
 
 class VidziIE(JWPlatformBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?vidzi\.tv/(?P<id>\w+)'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?vidzi\.tv/(?:embed-)?(?P<id>[0-9a-zA-Z]+)'
+    _TESTS = [{
         'url': 'http://vidzi.tv/cghql9yq6emu.html',
         'md5': '4f16c71ca0c8c8635ab6932b5f3f1660',
         'info_dict': {
@@ -22,12 +22,16 @@ class VidziIE(JWPlatformBaseIE):
             # m3u8 download
             'skip_download': True,
         },
-    }
+    }, {
+        'url': 'http://vidzi.tv/embed-4z2yb0rzphe9-600x338.html',
+        'skip_download': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(
+            'http://vidzi.tv/%s' % video_id, video_id)
         title = self._html_search_regex(
             r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title')
 

youtube_dl/extractor/vimeo.py

@@ -364,6 +364,11 @@ class VimeoIE(VimeoBaseInfoExtractor):
             r'<embed[^>]+?src="((?:https?:)?//(?:www\.)?vimeo\.com/moogaloop\.swf.+?)"', webpage)
         if mobj:
             return mobj.group(1)
+        # Look more for non-standard embedded Vimeo player
+        mobj = re.search(
+            r'<video[^>]+src=(?P<q1>[\'"])(?P<url>(?:https?:)?//(?:www\.)?vimeo\.com/[0-9]+)(?P=q1)', webpage)
+        if mobj:
+            return mobj.group('url')
 
     def _verify_player_video_password(self, url, video_id):
         password = self._downloader.params.get('videopassword')

youtube_dl/extractor/vk.py

@@ -6,11 +6,18 @@ import json
 import sys
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_str,
+    compat_urlparse,
+)
 from ..utils import (
+    clean_html,
     ExtractorError,
+    get_element_by_class,
     int_or_none,
     orderedSet,
+    parse_duration,
+    remove_start,
     str_to_int,
     unescapeHTML,
     unified_strdate,
@@ -20,26 +27,72 @@ from .vimeo import VimeoIE
 from .pladform import PladformIE
 
 
-class VKIE(InfoExtractor):
+class VKBaseIE(InfoExtractor):
+    _NETRC_MACHINE = 'vk'
+
+    def _login(self):
+        (username, password) = self._get_login_info()
+        if username is None:
+            return
+
+        login_page, url_handle = self._download_webpage_handle(
+            'https://vk.com', None, 'Downloading login page')
+
+        login_form = self._hidden_inputs(login_page)
+
+        login_form.update({
+            'email': username.encode('cp1251'),
+            'pass': password.encode('cp1251'),
+        })
+
+        # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
+        # and expects the first one to be set rather than second (see
+        # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
+        # As of RFC6265 the newer one cookie should be set into cookie store
+        # what actually happens.
+        # We will workaround this VK issue by resetting the remixlhk cookie to
+        # the first one manually.
+        cookies = url_handle.headers.get('Set-Cookie')
+        if cookies:
+            if sys.version_info[0] >= 3:
+                cookies = cookies.encode('iso-8859-1')
+            cookies = cookies.decode('utf-8')
+            remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
+            if remixlhk:
+                value, domain = remixlhk.groups()
+                self._set_cookie(domain, 'remixlhk', value)
+
+        login_page = self._download_webpage(
+            'https://login.vk.com/?act=login', None,
+            note='Logging in as %s' % username,
+            data=urlencode_postdata(login_form))
+
+        if re.search(r'onLoginFailed', login_page):
+            raise ExtractorError(
+                'Unable to login, incorrect username and/or password', expected=True)
+
+    def _real_initialize(self):
+        self._login()
+
+
+class VKIE(VKBaseIE):
     IE_NAME = 'vk'
     IE_DESC = 'VK'
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
                             (?:
-                                (?:m\.)?vk\.com/video_|
+                                (?:(?:m|new)\.)?vk\.com/video_|
                                 (?:www\.)?daxab.com/
                             )
                             ext\.php\?(?P<embed_query>.*?\boid=(?P<oid>-?\d+).*?\bid=(?P<id>\d+).*)|
                             (?:
-                                (?:m\.)?vk\.com/(?:.+?\?.*?z=)?video|
+                                (?:(?:m|new)\.)?vk\.com/(?:.+?\?.*?z=)?video|
                                 (?:www\.)?daxab.com/embed/
                             )
                             (?P<videoid>-?\d+_\d+)(?:.*\blist=(?P<list_id>[\da-f]+))?
                         )
                     '''
-    _NETRC_MACHINE = 'vk'
-
     _TESTS = [
         {
             'url': 'http://vk.com/videos-77521?z=video-77521_162222515%2Fclub77521',
@@ -182,52 +235,13 @@ class VKIE(InfoExtractor):
             # pladform embed
             'url': 'https://vk.com/video-76116461_171554880',
             'only_matching': True,
+        },
+        {
+            'url': 'http://new.vk.com/video205387401_165548505',
+            'only_matching': True,
         }
     ]
 
-    def _login(self):
-        (username, password) = self._get_login_info()
-        if username is None:
-            return
-
-        login_page, url_handle = self._download_webpage_handle(
-            'https://vk.com', None, 'Downloading login page')
-
-        login_form = self._hidden_inputs(login_page)
-
-        login_form.update({
-            'email': username.encode('cp1251'),
-            'pass': password.encode('cp1251'),
-        })
-
-        # https://new.vk.com/ serves two same remixlhk cookies in Set-Cookie header
-        # and expects the first one to be set rather than second (see
-        # https://github.com/rg3/youtube-dl/issues/9841#issuecomment-227871201).
-        # As of RFC6265 the newer one cookie should be set into cookie store
-        # what actually happens.
-        # We will workaround this VK issue by resetting the remixlhk cookie to
-        # the first one manually.
-        cookies = url_handle.headers.get('Set-Cookie')
-        if sys.version_info[0] >= 3:
-            cookies = cookies.encode('iso-8859-1')
-        cookies = cookies.decode('utf-8')
-        remixlhk = re.search(r'remixlhk=(.+?);.*?\bdomain=(.+?)(?:[,;]|$)', cookies)
-        if remixlhk:
-            value, domain = remixlhk.groups()
-            self._set_cookie(domain, 'remixlhk', value)
-
-        login_page = self._download_webpage(
-            'https://login.vk.com/?act=login', None,
-            note='Logging in as %s' % username,
-            data=urlencode_postdata(login_form))
-
-        if re.search(r'onLoginFailed', login_page):
-            raise ExtractorError(
-                'Unable to login, incorrect username and/or password', expected=True)
-
-    def _real_initialize(self):
-        self._login()
-
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('videoid')
@@ -351,10 +365,10 @@ class VKIE(InfoExtractor):
         }
 
 
-class VKUserVideosIE(InfoExtractor):
+class VKUserVideosIE(VKBaseIE):
     IE_NAME = 'vk:uservideos'
     IE_DESC = "VK - User's Videos"
-    _VALID_URL = r'https?://vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
+    _VALID_URL = r'https?://(?:(?:m|new)\.)?vk\.com/videos(?P<id>-?[0-9]+)(?!\?.*\bz=video)(?:[/?#&]|$)'
     _TEMPLATE_URL = 'https://vk.com/videos'
     _TESTS = [{
         'url': 'http://vk.com/videos205387401',
@@ -369,6 +383,12 @@ class VKUserVideosIE(InfoExtractor):
     }, {
         'url': 'http://vk.com/videos-97664626?section=all',
         'only_matching': True,
+    }, {
+        'url': 'http://m.vk.com/videos205387401',
+        'only_matching': True,
+    }, {
+        'url': 'http://new.vk.com/videos205387401',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
@@ -386,3 +406,121 @@ class VKUserVideosIE(InfoExtractor):
             webpage, 'title', default=page_id))
 
         return self.playlist_result(entries, page_id, title)
+
+
+class VKWallPostIE(VKBaseIE):
+    IE_NAME = 'vk:wallpost'
+    _VALID_URL = r'https?://(?:(?:(?:(?:m|new)\.)?vk\.com/(?:[^?]+\?.*\bw=)?wall(?P<id>-?\d+_\d+)))'
+    _TESTS = [{
+        # public page URL, audio playlist
+        'url': 'https://vk.com/bs.official?w=wall-23538238_35',
+        'info_dict': {
+            'id': '23538238_35',
+            'title': 'Black Shadow - Wall post 23538238_35',
+            'description': 'md5:3f84b9c4f9ef499731cf1ced9998cc0c',
+        },
+        'playlist': [{
+            'md5': '5ba93864ec5b85f7ce19a9af4af080f6',
+            'info_dict': {
+                'id': '135220665_111806521',
+                'ext': 'mp3',
+                'title': 'Black Shadow - Слепое Верование',
+                'duration': 370,
+                'uploader': 'Black Shadow',
+                'artist': 'Black Shadow',
+                'track': 'Слепое Верование',
+            },
+        }, {
+            'md5': '4cc7e804579122b17ea95af7834c9233',
+            'info_dict': {
+                'id': '135220665_111802303',
+                'ext': 'mp3',
+                'title': 'Black Shadow - Война - Негасимое Бездны Пламя!',
+                'duration': 423,
+                'uploader': 'Black Shadow',
+                'artist': 'Black Shadow',
+                'track': 'Война - Негасимое Бездны Пламя!',
+            },
+            'params': {
+                'skip_download': True,
+            },
+        }],
+        'skip': 'Requires vk account credentials',
+    }, {
+        # single YouTube embed, no leading -
+        'url': 'https://vk.com/wall85155021_6319',
+        'info_dict': {
+            'id': '85155021_6319',
+            'title': 'Sergey Gorbunov - Wall post 85155021_6319',
+        },
+        'playlist_count': 1,
+        'skip': 'Requires vk account credentials',
+    }, {
+        # wall page URL
+        'url': 'https://vk.com/wall-23538238_35',
+        'only_matching': True,
+    }, {
+        # mobile wall page URL
+        'url': 'https://m.vk.com/wall-23538238_35',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        post_id = self._match_id(url)
+
+        wall_url = 'https://vk.com/wall%s' % post_id
+
+        post_id = remove_start(post_id, '-')
+
+        webpage = self._download_webpage(wall_url, post_id)
+
+        error = self._html_search_regex(
+            r'>Error</div>\s*<div[^>]+class=["\']body["\'][^>]*>([^<]+)',
+            webpage, 'error', default=None)
+        if error:
+            raise ExtractorError('VK said: %s' % error, expected=True)
+
+        description = clean_html(get_element_by_class('wall_post_text', webpage))
+        uploader = clean_html(get_element_by_class(
+            'fw_post_author', webpage)) or self._og_search_description(webpage)
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        entries = []
+
+        for audio in re.finditer(r'''(?sx)
+                            <input[^>]+
+                                id=(?P<q1>["\'])audio_info(?P<id>\d+_\d+).*?(?P=q1)[^>]+
+                                value=(?P<q2>["\'])(?P<url>http.+?)(?P=q2)
+                                .+?
+                            </table>''', webpage):
+            audio_html = audio.group(0)
+            audio_id = audio.group('id')
+            duration = parse_duration(get_element_by_class('duration', audio_html))
+            track = self._html_search_regex(
+                r'<span[^>]+id=["\']title%s[^>]*>([^<]+)' % audio_id,
+                audio_html, 'title', default=None)
+            artist = self._html_search_regex(
+                r'>([^<]+)</a></b>\s*&ndash', audio_html,
+                'artist', default=None)
+            entries.append({
+                'id': audio_id,
+                'url': audio.group('url'),
+                'title': '%s - %s' % (artist, track) if artist and track else audio_id,
+                'thumbnail': thumbnail,
+                'duration': duration,
+                'uploader': uploader,
+                'artist': artist,
+                'track': track,
+            })
+
+        for video in re.finditer(
+                r'<a[^>]+href=(["\'])(?P<url>/video(?:-?[\d_]+).*?)\1', webpage):
+            entries.append(self.url_result(
+                compat_urlparse.urljoin(url, video.group('url')), VKIE.ie_key()))
+
+        title = 'Wall post %s' % post_id
+
+        return self.playlist_result(
+            orderedSet(entries), post_id,
+            '%s - %s' % (uploader, title) if uploader else title,
+            description)

youtube_dl/extractor/vuclip.py

@@ -9,7 +9,7 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     parse_duration,
-    qualities,
+    remove_end,
 )
 
 
@@ -22,7 +22,7 @@ class VuClipIE(InfoExtractor):
             'id': '922692425',
             'ext': '3gp',
             'title': 'The Toy Soldiers - Hollywood Movie Trailer',
-            'duration': 180,
+            'duration': 177,
         }
     }
 
@@ -46,34 +46,21 @@ class VuClipIE(InfoExtractor):
                 '%s said: %s' % (self.IE_NAME, error_msg), expected=True)
 
         # These clowns alternate between two page types
-        links_code = self._search_regex(
-            r'''(?xs)
-                (?:
-                    <img\s+src="[^"]*/play.gif".*?>|
-                    <!--\ player\ end\ -->\s*</div><!--\ thumb\ end-->
-                )
-                (.*?)
-                (?:
-                    <a\s+href="fblike|<div\s+class="social">
-                )
-            ''', webpage, 'links')
-        title = self._html_search_regex(
-            r'<title>(.*?)-\s*Vuclip</title>', webpage, 'title').strip()
+        video_url = self._search_regex(
+            r'<a[^>]+href="([^"]+)"[^>]*><img[^>]+src="[^"]*/play\.gif',
+            webpage, 'video URL', default=None)
+        if video_url:
+            formats = [{
+                'url': video_url,
+            }]
+        else:
+            formats = self._parse_html5_media_entries(url, webpage)[0]['formats']
 
-        quality_order = qualities(['Reg', 'Hi'])
-        formats = []
-        for url, q in re.findall(
-                r'<a\s+href="(?P<url>[^"]+)".*?>(?:<button[^>]*>)?(?P<q>[^<]+)(?:</button>)?</a>', links_code):
-            format_id = compat_urllib_parse_urlparse(url).scheme + '-' + q
-            formats.append({
-                'format_id': format_id,
-                'url': url,
-                'quality': quality_order(q),
-            })
-        self._sort_formats(formats)
+        title = remove_end(self._html_search_regex(
+            r'<title>(.*?)-\s*Vuclip</title>', webpage, 'title').strip(), ' - Video')
 
-        duration = parse_duration(self._search_regex(
-            r'\(([0-9:]+)\)</span>', webpage, 'duration', fatal=False))
+        duration = parse_duration(self._html_search_regex(
+            r'[(>]([0-9]+:[0-9]+)(?:<span|\))', webpage, 'duration', fatal=False))
 
         return {
             'id': video_id,

youtube_dl/extractor/wat.py

@@ -9,7 +9,6 @@ from ..utils import (
     ExtractorError,
     unified_strdate,
     HEADRequest,
-    float_or_none,
 )
 
 
@@ -95,16 +94,7 @@ class WatIE(InfoExtractor):
             m3u8_url.replace('ios.', 'web.').replace('.m3u8', '.f4m'),
             video_id, f4m_id='hds', fatal=False))
         for m3u8_format in m3u8_formats:
-            mobj = re.search(
-                r'audio.*?%3D(\d+)(?:-video.*?%3D(\d+))?', m3u8_format['url'])
-            if not mobj:
-                continue
-            abr, vbr = mobj.groups()
-            abr, vbr = float_or_none(abr, 1000), float_or_none(vbr, 1000)
-            m3u8_format.update({
-                'vbr': vbr,
-                'abr': abr,
-            })
+            vbr, abr = m3u8_format.get('vbr'), m3u8_format.get('abr')
             if not vbr or not abr:
                 continue
             f = m3u8_format.copy()

youtube_dl/extractor/xtube.py

@@ -4,17 +4,23 @@ import itertools
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
 from ..utils import (
     int_or_none,
     orderedSet,
+    parse_duration,
     sanitized_Request,
     str_to_int,
 )
 
 
 class XTubeIE(InfoExtractor):
-    _VALID_URL = r'(?:xtube:|https?://(?:www\.)?xtube\.com/(?:watch\.php\?.*\bv=|video-watch/(?P<display_id>[^/]+)-))(?P<id>[^/?&#]+)'
+    _VALID_URL = r'''(?x)
+                        (?:
+                            xtube:|
+                            https?://(?:www\.)?xtube\.com/(?:watch\.php\?.*\bv=|video-watch/(?P<display_id>[^/]+)-)
+                        )
+                        (?P<id>[^/?&#]+)
+                    '''
 
     _TESTS = [{
         # old URL schema
@@ -27,6 +33,8 @@ class XTubeIE(InfoExtractor):
             'description': 'contains:an ET kind of thing',
             'uploader': 'greenshowers',
             'duration': 450,
+            'view_count': int,
+            'comment_count': int,
             'age_limit': 18,
         }
     }, {
@@ -51,21 +59,30 @@ class XTubeIE(InfoExtractor):
         req.add_header('Cookie', 'age_verified=1; cookiesAccepted=1')
         webpage = self._download_webpage(req, display_id)
 
-        flashvars = self._parse_json(
-            self._search_regex(
-                r'xt\.playerOps\s*=\s*({.+?});', webpage, 'player ops'),
-            video_id)['flashvars']
+        sources = self._parse_json(self._search_regex(
+            r'sources\s*:\s*({.+?}),', webpage, 'sources'), video_id)
 
-        title = flashvars.get('title') or self._search_regex(
-            r'<h1>([^<]+)</h1>', webpage, 'title')
-        video_url = compat_urllib_parse_unquote(flashvars['video_url'])
-        duration = int_or_none(flashvars.get('video_duration'))
+        formats = []
+        for format_id, format_url in sources.items():
+            formats.append({
+                'url': format_url,
+                'format_id': format_id,
+                'height': int_or_none(format_id),
+            })
+        self._sort_formats(formats)
 
-        uploader = self._search_regex(
-            r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"',
-            webpage, 'uploader', fatal=False)
+        title = self._search_regex(
+            (r'<h1>(?P<title>[^<]+)</h1>', r'videoTitle\s*:\s*(["\'])(?P<title>.+?)\1'),
+            webpage, 'title', group='title')
         description = self._search_regex(
             r'</h1>\s*<p>([^<]+)', webpage, 'description', fatal=False)
+        uploader = self._search_regex(
+            (r'<input[^>]+name="contentOwnerId"[^>]+value="([^"]+)"',
+             r'<span[^>]+class="nickname"[^>]*>([^<]+)'),
+            webpage, 'uploader', fatal=False)
+        duration = parse_duration(self._search_regex(
+            r'<dt>Runtime:</dt>\s*<dd>([^<]+)</dd>',
+            webpage, 'duration', fatal=False))
         view_count = str_to_int(self._search_regex(
             r'<dt>Views:</dt>\s*<dd>([\d,\.]+)</dd>',
             webpage, 'view count', fatal=False))
@@ -76,7 +93,6 @@ class XTubeIE(InfoExtractor):
         return {
             'id': video_id,
             'display_id': display_id,
-            'url': video_url,
             'title': title,
             'description': description,
             'uploader': uploader,
@@ -84,6 +100,7 @@ class XTubeIE(InfoExtractor):
             'view_count': view_count,
             'comment_count': comment_count,
             'age_limit': 18,
+            'formats': formats,
         }
 
 

youtube_dl/extractor/xuite.py

@@ -67,6 +67,20 @@ class XuiteIE(InfoExtractor):
             'categories': ['電玩動漫'],
         },
         'skip': 'Video removed',
+    }, {
+        # Video with encoded media id
+        # from http://forgetfulbc.blogspot.com/2016/06/date.html
+        'url': 'http://vlog.xuite.net/embed/cE1xbENoLTI3NDQ3MzM2LmZsdg==?ar=0&as=0',
+        'info_dict': {
+            'id': 'cE1xbENoLTI3NDQ3MzM2LmZsdg==',
+            'ext': 'mp4',
+            'title': '男女平權只是口號？專家解釋約會時男生是否該幫女生付錢 (中字)',
+            'description': 'md5:f0abdcb69df300f522a5442ef3146f2a',
+            'timestamp': 1466160960,
+            'upload_date': '20160617',
+            'uploader': 'B.C. & Lowy',
+            'uploader_id': '232279340',
+        },
     }, {
         'url': 'http://vlog.xuite.net/play/S1dDUjdyLTMyOTc3NjcuZmx2/%E5%AD%AB%E7%87%95%E5%A7%BF-%E7%9C%BC%E6%B7%9A%E6%88%90%E8%A9%A9',
         'only_matching': True,
@@ -80,10 +94,9 @@ class XuiteIE(InfoExtractor):
     def base64_encode_utf8(data):
         return base64.b64encode(data.encode('utf-8')).decode('utf-8')
 
-    def _extract_flv_config(self, media_id):
-        base64_media_id = self.base64_encode_utf8(media_id)
+    def _extract_flv_config(self, encoded_media_id):
         flv_config = self._download_xml(
-            'http://vlog.xuite.net/flash/player?media=%s' % base64_media_id,
+            'http://vlog.xuite.net/flash/player?media=%s' % encoded_media_id,
             'flv config')
         prop_dict = {}
         for prop in flv_config.findall('./property'):
@@ -108,9 +121,14 @@ class XuiteIE(InfoExtractor):
                 '%s returned error: %s' % (self.IE_NAME, error_msg),
                 expected=True)
 
-        video_id = self._html_search_regex(
-            r'data-mediaid="(\d+)"', webpage, 'media id')
-        flv_config = self._extract_flv_config(video_id)
+        encoded_media_id = self._search_regex(
+            r'attributes\.name\s*=\s*"([^"]+)"', webpage,
+            'encoded media id', default=None)
+        if encoded_media_id is None:
+            video_id = self._html_search_regex(
+                r'data-mediaid="(\d+)"', webpage, 'media id')
+            encoded_media_id = self.base64_encode_utf8(video_id)
+        flv_config = self._extract_flv_config(encoded_media_id)
 
         FORMATS = {
             'audio': 'mp3',

youtube_dl/extractor/yahoo.py

@@ -19,6 +19,7 @@ from ..utils import (
     mimetype2ext,
 )
 
+from .brightcove import BrightcoveNewIE
 from .nbc import NBCSportsVPlayerIE
 
 
@@ -227,7 +228,12 @@ class YahooIE(InfoExtractor):
         # Look for NBCSports iframes
         nbc_sports_url = NBCSportsVPlayerIE._extract_url(webpage)
         if nbc_sports_url:
-            return self.url_result(nbc_sports_url, 'NBCSportsVPlayer')
+            return self.url_result(nbc_sports_url, NBCSportsVPlayerIE.ie_key())
+
+        # Look for Brightcove New Studio embeds
+        bc_url = BrightcoveNewIE._extract_url(webpage)
+        if bc_url:
+            return self.url_result(bc_url, BrightcoveNewIE.ie_key())
 
         # Query result is often embedded in webpage as JSON. Sometimes explicit requests
         # to video API results in a failure with geo restriction reason therefore using

youtube_dl/extractor/youku.py

@@ -16,7 +16,6 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     get_element_by_attribute,
-    sanitized_Request,
 )
 
 
@@ -218,14 +217,10 @@ class YoukuIE(InfoExtractor):
             headers = {
                 'Referer': req_url,
             }
+            headers.update(self.geo_verification_headers())
             self._set_cookie('youku.com', 'xreferrer', 'http://www.youku.com')
-            req = sanitized_Request(req_url, headers=headers)
 
-            cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
-            if cn_verification_proxy:
-                req.add_header('Ytdl-request-proxy', cn_verification_proxy)
-
-            raw_data = self._download_json(req, video_id, note=note)
+            raw_data = self._download_json(req_url, video_id, note=note, headers=headers)
 
             return raw_data['data']
 

youtube_dl/extractor/youtube.py

@@ -137,7 +137,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
         # Two-Factor
         # TODO add SMS and phone call support - these require making a request and then prompting the user
 
-        if re.search(r'(?i)<form[^>]* id="challenge"', login_results) is not None:
+        if re.search(r'(?i)<form[^>]+id="challenge"', login_results) is not None:
             tfa_code = self._get_tfa_info('2-step verification code')
 
             if not tfa_code:
@@ -165,17 +165,17 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             if tfa_results is False:
                 return False
 
-            if re.search(r'(?i)<form[^>]* id="challenge"', tfa_results) is not None:
+            if re.search(r'(?i)<form[^>]+id="challenge"', tfa_results) is not None:
                 self._downloader.report_warning('Two-factor code expired or invalid. Please try again, or use a one-use backup code instead.')
                 return False
-            if re.search(r'(?i)<form[^>]* id="gaia_loginform"', tfa_results) is not None:
+            if re.search(r'(?i)<form[^>]+id="gaia_loginform"', tfa_results) is not None:
                 self._downloader.report_warning('unable to log in - did the page structure change?')
                 return False
             if re.search(r'smsauth-interstitial-reviewsettings', tfa_results) is not None:
                 self._downloader.report_warning('Your Google account has a security notice. Please log in on your web browser, resolve the notice, and try again.')
                 return False
 
-        if re.search(r'(?i)<form[^>]* id="gaia_loginform"', login_results) is not None:
+        if re.search(r'(?i)<form[^>]+id="gaia_loginform"', login_results) is not None:
             self._downloader.report_warning('unable to log in: bad username or password')
             return False
         return True
@@ -858,6 +858,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         {
             'url': 'https://www.youtube.com/watch?feature=player_embedded&amp;amp;v=V36LpHqtcDY',
             'only_matching': True,
+        },
+        {
+            # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
+            'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
+            'only_matching': True,
         }
     ]
 
@@ -1730,6 +1735,39 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         }
 
 
+class YoutubeSharedVideoIE(InfoExtractor):
+    _VALID_URL = r'(?:https?:)?//(?:www\.)?youtube\.com/shared\?ci=(?P<id>[0-9A-Za-z_-]{11})'
+    IE_NAME = 'youtube:shared'
+
+    _TEST = {
+        'url': 'https://www.youtube.com/shared?ci=1nEzmT-M4fU',
+        'info_dict': {
+            'id': 'uPDB5I9wfp8',
+            'ext': 'webm',
+            'title': 'Pocoyo: 90 minutos de episódios completos Português para crianças - PARTE 3',
+            'description': 'md5:d9e4d9346a2dfff4c7dc4c8cec0f546d',
+            'upload_date': '20160219',
+            'uploader': 'Pocoyo - Português (BR)',
+            'uploader_id': 'PocoyoBrazil',
+        },
+        'add_ie': ['Youtube'],
+        'params': {
+            # There are already too many Youtube downloads
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        real_video_id = self._html_search_meta(
+            'videoId', webpage, 'YouTube video id', fatal=True)
+
+        return self.url_result(real_video_id, YoutubeIE.ie_key())
+
+
 class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
     IE_DESC = 'YouTube.com playlists'
     _VALID_URL = r"""(?x)(?:
@@ -1945,10 +1983,13 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
         return (False if YoutubePlaylistsIE.suitable(url) or YoutubeLiveIE.suitable(url)
                 else super(YoutubeChannelIE, cls).suitable(url))
 
+    def _build_template_url(self, url, channel_id):
+        return self._TEMPLATE_URL % channel_id
+
     def _real_extract(self, url):
         channel_id = self._match_id(url)
 
-        url = self._TEMPLATE_URL % channel_id
+        url = self._build_template_url(url, channel_id)
 
         # Channel by page listing is restricted to 35 pages of 30 items, i.e. 1050 videos total (see #5778)
         # Workaround by extracting as a playlist if managed to obtain channel playlist URL
@@ -1962,9 +2003,13 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
             channel_playlist_id = self._html_search_meta(
                 'channelId', channel_page, 'channel id', default=None)
             if not channel_playlist_id:
-                channel_playlist_id = self._search_regex(
-                    r'data-(?:channel-external-|yt)id="([^"]+)"',
-                    channel_page, 'channel id', default=None)
+                channel_url = self._html_search_meta(
+                    ('al:ios:url', 'twitter:app:url:iphone', 'twitter:app:url:ipad'),
+                    channel_page, 'channel url', default=None)
+                if channel_url:
+                    channel_playlist_id = self._search_regex(
+                        r'vnd\.youtube://user/([0-9A-Za-z_-]+)',
+                        channel_url, 'channel id', default=None)
         if channel_playlist_id and channel_playlist_id.startswith('UC'):
             playlist_id = 'UU' + channel_playlist_id[2:]
             return self.url_result(
@@ -1987,20 +2032,39 @@ class YoutubeChannelIE(YoutubePlaylistBaseInfoExtractor):
                 for video_id, video_title in self.extract_videos_from_page(channel_page)]
             return self.playlist_result(entries, channel_id)
 
+        try:
+            next(self._entries(channel_page, channel_id))
+        except StopIteration:
+            alert_message = self._html_search_regex(
+                r'(?s)<div[^>]+class=(["\']).*?\byt-alert-message\b.*?\1[^>]*>(?P<alert>[^<]+)</div>',
+                channel_page, 'alert', default=None, group='alert')
+            if alert_message:
+                raise ExtractorError('Youtube said: %s' % alert_message, expected=True)
+
         return self.playlist_result(self._entries(channel_page, channel_id), channel_id)
 
 
 class YoutubeUserIE(YoutubeChannelIE):
     IE_DESC = 'YouTube.com user videos (URL or "ytuser" keyword)'
-    _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:user/|c/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
-    _TEMPLATE_URL = 'https://www.youtube.com/user/%s/videos'
+    _VALID_URL = r'(?:(?:https?://(?:\w+\.)?youtube\.com/(?:(?P<user>user|c)/)?(?!(?:attribution_link|watch|results)(?:$|[^a-z_A-Z0-9-])))|ytuser:)(?!feed/)(?P<id>[A-Za-z0-9_-]+)'
+    _TEMPLATE_URL = 'https://www.youtube.com/%s/%s/videos'
     IE_NAME = 'youtube:user'
 
     _TESTS = [{
         'url': 'https://www.youtube.com/user/TheLinuxFoundation',
         'playlist_mincount': 320,
         'info_dict': {
-            'title': 'TheLinuxFoundation',
+            'id': 'UUfX55Sx5hEFjoC3cNs6mCUQ',
+            'title': 'Uploads from The Linux Foundation',
+        }
+    }, {
+        # Only available via https://www.youtube.com/c/12minuteathlete/videos
+        # but not https://www.youtube.com/user/12minuteathlete/videos
+        'url': 'https://www.youtube.com/c/12minuteathlete/videos',
+        'playlist_mincount': 249,
+        'info_dict': {
+            'id': 'UUVjM-zV6_opMDx7WYxnjZiQ',
+            'title': 'Uploads from 12 Minute Athlete',
         }
     }, {
         'url': 'ytuser:phihag',
@@ -2008,6 +2072,13 @@ class YoutubeUserIE(YoutubeChannelIE):
     }, {
         'url': 'https://www.youtube.com/c/gametrailers',
         'only_matching': True,
+    }, {
+        'url': 'https://www.youtube.com/gametrailers',
+        'only_matching': True,
+    }, {
+        # This channel is not available.
+        'url': 'https://www.youtube.com/user/kananishinoSMEJ/videos',
+        'only_matching': True,
     }]
 
     @classmethod
@@ -2020,6 +2091,10 @@ class YoutubeUserIE(YoutubeChannelIE):
         else:
             return super(YoutubeUserIE, cls).suitable(url)
 
+    def _build_template_url(self, url, channel_id):
+        mobj = re.match(self._VALID_URL, url)
+        return self._TEMPLATE_URL % (mobj.group('user') or 'user', mobj.group('id'))
+
 
 class YoutubeLiveIE(YoutubeBaseInfoExtractor):
     IE_DESC = 'YouTube.com live streams'

youtube_dl/options.py

@@ -26,7 +26,11 @@ def parseOpts(overrideArguments=None):
         except IOError:
             return default  # silently skip if file is not present
         try:
-            res = compat_shlex_split(optionf.read(), comments=True)
+            # FIXME: https://github.com/rg3/youtube-dl/commit/dfe5fa49aed02cf36ba9f743b11b0903554b5e56
+            contents = optionf.read()
+            if sys.version_info < (3,):
+                contents = contents.decode(preferredencoding())
+            res = compat_shlex_split(contents, comments=True)
         finally:
             optionf.close()
         return res
@@ -209,11 +213,16 @@ def parseOpts(overrideArguments=None):
         action='store_const', const='::', dest='source_address',
         help='Make all connections via IPv6 (experimental)',
     )
+    network.add_option(
+        '--geo-verification-proxy',
+        dest='geo_verification_proxy', default=None, metavar='URL',
+        help='Use this proxy to verify the IP address for some geo-restricted sites. '
+        'The default proxy specified by --proxy (or none, if the options is not present) is used for the actual downloading. (experimental)'
+    )
     network.add_option(
         '--cn-verification-proxy',
         dest='cn_verification_proxy', default=None, metavar='URL',
-        help='Use this proxy to verify the IP address for some Chinese sites. '
-        'The default proxy specified by --proxy (or none, if the options is not present) is used for the actual downloading. (experimental)'
+        help=optparse.SUPPRESS_HELP,
     )
 
     selection = optparse.OptionGroup(parser, 'Video Selection')
@@ -807,11 +816,11 @@ def parseOpts(overrideArguments=None):
             system_conf = []
             user_conf = []
         else:
-            system_conf = compat_conf(_readOptions('/etc/youtube-dl.conf'))
+            system_conf = _readOptions('/etc/youtube-dl.conf')
             if '--ignore-config' in system_conf:
                 user_conf = []
             else:
-                user_conf = compat_conf(_readUserConf())
+                user_conf = _readUserConf()
         argv = system_conf + user_conf + command_line_conf
 
         opts, args = parser.parse_args(argv)

youtube_dl/postprocessor/ffmpeg.py

@@ -363,8 +363,10 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
         input_files = [filename] + sub_filenames
 
         opts = [
-            '-map', '0',
-            '-c', 'copy',
+            '-map', '0:v',
+            '-c:v', 'copy',
+            '-map', '0:a',
+            '-c:a', 'copy',
             # Don't copy the existing subtitles, we may be running the
             # postprocessor a second time
             '-map', '-0:s',

youtube_dl/utils.py

@@ -310,9 +310,17 @@ def get_element_by_id(id, html):
     return get_element_by_attribute('id', id, html)
 
 
-def get_element_by_attribute(attribute, value, html):
+def get_element_by_class(class_name, html):
+    return get_element_by_attribute(
+        'class', r'[^\'"]*\b%s\b[^\'"]*' % re.escape(class_name),
+        html, escape_value=False)
+
+
+def get_element_by_attribute(attribute, value, html, escape_value=True):
     """Return the content of the tag with the specified attribute in the passed HTML document"""
 
+    value = re.escape(value) if escape_value else value
+
     m = re.search(r'''(?xs)
         <([a-zA-Z0-9:._-]+)
          (?:\s+[a-zA-Z0-9:._-]+(?:=[a-zA-Z0-9:._-]*|="[^"]*"|='[^']*'))*?
@@ -321,7 +329,7 @@ def get_element_by_attribute(attribute, value, html):
         \s*>
         (?P<content>.*?)
         </\1>
-    ''' % (re.escape(attribute), re.escape(value)), html)
+    ''' % (re.escape(attribute), value), html)
 
     if not m:
         return None
@@ -1444,6 +1452,8 @@ def shell_quote(args):
 def smuggle_url(url, data):
     """ Pass additional data in a URL for internal use. """
 
+    url, idata = unsmuggle_url(url, {})
+    data.update(idata)
     sdata = compat_urllib_parse_urlencode(
         {'__youtubedl_smuggle': json.dumps(data)})
     return url + '#' + sdata
@@ -1625,6 +1635,11 @@ class HEADRequest(compat_urllib_request.Request):
         return 'HEAD'
 
 
+class PUTRequest(compat_urllib_request.Request):
+    def get_method(self):
+        return 'PUT'
+
+
 def int_or_none(v, scale=1, default=None, get_attr=None, invscale=1):
     if get_attr:
         if v is not None:
@@ -1920,7 +1935,13 @@ def update_Request(req, url=None, data=None, headers={}, query={}):
     req_headers.update(headers)
     req_data = data or req.data
     req_url = update_url_query(url or req.get_full_url(), query)
-    req_type = HEADRequest if req.get_method() == 'HEAD' else compat_urllib_request.Request
+    req_get_method = req.get_method()
+    if req_get_method == 'HEAD':
+        req_type = HEADRequest
+    elif req_get_method == 'PUT':
+        req_type = PUTRequest
+    else:
+        req_type = compat_urllib_request.Request
     new_req = req_type(
         req_url, data=req_data, headers=req_headers,
         origin_req_host=req.origin_req_host, unverifiable=req.unverifiable)
@@ -2084,6 +2105,7 @@ def mimetype2ext(mt):
         return ext
 
     _, _, res = mt.rpartition('/')
+    res = res.lower()
 
     return {
         '3gpp': '3gp',
@@ -2095,9 +2117,51 @@ def mimetype2ext(mt):
         'x-flv': 'flv',
         'x-mp4-fragmented': 'mp4',
         'x-ms-wmv': 'wmv',
+        'mpegurl': 'm3u8',
+        'x-mpegurl': 'm3u8',
+        'vnd.apple.mpegurl': 'm3u8',
+        'dash+xml': 'mpd',
+        'f4m': 'f4m',
+        'f4m+xml': 'f4m',
     }.get(res, res)
 
 
+def parse_codecs(codecs_str):
+    # http://tools.ietf.org/html/rfc6381
+    if not codecs_str:
+        return {}
+    splited_codecs = list(filter(None, map(
+        lambda str: str.strip(), codecs_str.strip().strip(',').split(','))))
+    vcodec, acodec = None, None
+    for full_codec in splited_codecs:
+        codec = full_codec.split('.')[0]
+        if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v'):
+            if not vcodec:
+                vcodec = full_codec
+        elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac'):
+            if not acodec:
+                acodec = full_codec
+        else:
+            write_string('WARNING: Unknown codec %s' % full_codec, sys.stderr)
+    if not vcodec and not acodec:
+        if len(splited_codecs) == 2:
+            return {
+                'vcodec': vcodec,
+                'acodec': acodec,
+            }
+        elif len(splited_codecs) == 1:
+            return {
+                'vcodec': 'none',
+                'acodec': vcodec,
+            }
+    else:
+        return {
+            'vcodec': vcodec or 'none',
+            'acodec': acodec or 'none',
+        }
+    return {}
+
+
 def urlhandle_detect_ext(url_handle):
     getheader = url_handle.headers.get
 

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.07.02'
+__version__ = '2016.07.13'