release 2017.04.02

* Fix georestriction detection
* Detect live streams
+ Extract relinker metadata
* Improve ContentItem detection
+ Extract series metadata
* Fix tests

.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 *2017.03.22*. 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 Issues with outdated version will be rejected.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 **2017.03.22**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2017.04.02*. 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 Issues with outdated version will be rejected.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 **2017.04.02**
 
 ### 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 2017.03.22
+[debug] youtube-dl version 2017.04.02
 [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: {}

ChangeLog

@@ -1,3 +1,60 @@
+version 2017.04.02
+
+Core
+[YoutubeDL] Return early when extraction of url_transparent fails
+
+Extractors
+* [rai] Fix and improve extraction (#11790)
++ [vrv] Add support for series pages
+* [limelight] Improve extraction for audio only formats
+* [funimation] Fix extraction (#10696, #11773)
++ [xfileshare] Add support for vidabc.com (#12589)
++ [xfileshare] Improve extraction and extract hls formats
++ [crunchyroll] Pass geo verifcation proxy
++ [cwtv] Extract ISM formats
++ [tvplay] Bypass geo restriction
++ [vrv] Add support for vrv.co
++ [packtpub] Add support for packtpub.com (#12610)
++ [generic] Pass base_url to _parse_jwplayer_data
++ [adn] Add support for animedigitalnetwork.fr (#4866)
++ [allocine] Extract more metadata
+* [allocine] Fix extraction (#12592)
+* [openload] Fix extraction
+
+
+version 2017.03.26
+
+Core
+* Don't raise an error if JWPlayer config data is not a Javascript object
+  literal. _find_jwplayer_data now returns a dict rather than an str. (#12307)
+* Expand environment variables for options representing paths (#12556)
++ [utils] Introduce expand_path
+* [downloader/hls] Delegate downloading to ffmpeg immediately for live streams
+
+Extractors
+* [afreecatv] Fix extraction (#12179)
++ [atvat] Add support for atv.at (#5325)
++ [fox] Add metadata extraction (#12391)
++ [atresplayer] Extract DASH formats
++ [atresplayer] Extract HD manifest (#12548)
+* [atresplayer] Fix login error detection (#12548)
+* [franceculture] Fix extraction (#12547)
+* [youtube] Improve URL regular expression (#12538)
+* [generic] Do not follow redirects to the same URL
+
+
+version 2017.03.24
+
+Extractors
+- [9c9media] Remove mp4 URL extraction request
++ [bellmedia] Add support for etalk.ca and space.ca (#12447)
+* [channel9] Fix extraction (#11323)
+* [cloudy] Fix extraction (#12525)
++ [hbo] Add support for free episode URLs and new formats extraction (#12519)
+* [condenast] Fix extraction and style (#12526)
+* [viu] Relax URL regular expression (#12529)
+
+
 version 2017.03.22
 
 Extractors

README.md

@@ -181,10 +181,10 @@ Alternatively, refer to the [developer instructions](#developer-instructions) fo
     -R, --retries RETRIES            Number of retries (default is 10), or
                                      "infinite".
     --fragment-retries RETRIES       Number of retries for a fragment (default
-                                     is 10), or "infinite" (DASH and hlsnative
-                                     only)
-    --skip-unavailable-fragments     Skip unavailable fragments (DASH and
-                                     hlsnative only)
+                                     is 10), or "infinite" (DASH, hlsnative and
+                                     ISM)
+    --skip-unavailable-fragments     Skip unavailable fragments (DASH, hlsnative
+                                     and ISM)
     --abort-on-unavailable-fragment  Abort downloading when some fragment is not
                                      available
     --buffer-size SIZE               Size of download buffer (e.g. 1024 or 16K)

docs/supportedsites.md

@@ -28,6 +28,7 @@
  - **acast**
  - **acast:channel**
  - **AddAnime**
+ - **ADN**: Anime Digital Network
  - **AdobeTV**
  - **AdobeTVChannel**
  - **AdobeTVShow**
@@ -67,6 +68,7 @@
  - **arte.tv:playlist**
  - **AtresPlayer**
  - **ATTTechChannel**
+ - **ATVAt**
  - **AudiMedia**
  - **AudioBoom**
  - **audiomack**
@@ -312,8 +314,8 @@
  - **GPUTechConf**
  - **Groupon**
  - **Hark**
- - **HBO**
- - **HBOEpisode**
+ - **hbo**
+ - **hbo:episode**
  - **HearThisAt**
  - **Heise**
  - **HellPorno**
@@ -571,6 +573,8 @@
  - **orf:iptv**: iptv.ORF.at
  - **orf:oe1**: Radio Österreich 1
  - **orf:tvthek**: ORF TVthek
+ - **PacktPub**
+ - **PacktPubCourse**
  - **PandaTV**: 熊猫TV
  - **pandora.tv**: 판도라TV
  - **parliamentlive.tv**: UK parliament videos
@@ -628,7 +632,7 @@
  - **radiofrance**
  - **RadioJavan**
  - **Rai**
- - **RaiTV**
+ - **RaiPlay**
  - **RBMARadio**
  - **RDS**: RDS.ca
  - **RedBullTV**
@@ -925,6 +929,8 @@
  - **vpro**: npo.nl and ntr.nl
  - **Vrak**
  - **VRT**
+ - **vrv**
+ - **vrv:series**
  - **vube**: Vube.com
  - **VuClip**
  - **VVVVID**
@@ -952,7 +958,7 @@
  - **WSJ**: Wall Street Journal
  - **XBef**
  - **XboxClips**
- - **XFileShare**: XFileShare based sites: DaClips, FileHoot, GorillaVid, MovPod, PowerWatch, Rapidvideo.ws, TheVideoBee, Vidto, Streamin.To, XVIDSTAGE
+ - **XFileShare**: XFileShare based sites: DaClips, FileHoot, GorillaVid, MovPod, PowerWatch, Rapidvideo.ws, TheVideoBee, Vidto, Streamin.To, XVIDSTAGE, Vid ABC
  - **XHamster**
  - **XHamsterEmbed**
  - **xiami:album**: 虾米音乐 - 专辑

test/test_compat.py

@@ -27,11 +27,11 @@ from youtube_dl.compat import (
 class TestCompat(unittest.TestCase):
     def test_compat_getenv(self):
         test_str = 'тест'
-        compat_setenv('YOUTUBE-DL-TEST', test_str)
-        self.assertEqual(compat_getenv('YOUTUBE-DL-TEST'), test_str)
+        compat_setenv('YOUTUBE_DL_COMPAT_GETENV', test_str)
+        self.assertEqual(compat_getenv('YOUTUBE_DL_COMPAT_GETENV'), test_str)
 
     def test_compat_setenv(self):
-        test_var = 'YOUTUBE-DL-TEST'
+        test_var = 'YOUTUBE_DL_COMPAT_SETENV'
         test_str = 'тест'
         compat_setenv(test_var, test_str)
         compat_getenv(test_var)

test/test_download.py

@@ -71,6 +71,18 @@ class TestDownload(unittest.TestCase):
 
     maxDiff = None
 
+    def __str__(self):
+        """Identify each test with the `add_ie` attribute, if available."""
+
+        def strclass(cls):
+            """From 2.7's unittest; 2.6 had _strclass so we can't import it."""
+            return '%s.%s' % (cls.__module__, cls.__name__)
+
+        add_ie = getattr(self, self._testMethodName).add_ie
+        return '%s (%s)%s:' % (self._testMethodName,
+                               strclass(self.__class__),
+                               ' [%s]' % add_ie if add_ie else '')
+
     def setUp(self):
         self.defs = defs
 
@@ -233,6 +245,8 @@ for n, test_case in enumerate(defs):
         i += 1
     test_method = generator(test_case, tname)
     test_method.__name__ = str(tname)
+    ie_list = test_case.get('add_ie')
+    test_method.add_ie = ie_list and ','.join(ie_list)
     setattr(TestDownload, test_method.__name__, test_method)
     del test_method
 

test/test_utils.py

@@ -56,6 +56,7 @@ from youtube_dl.utils import (
     read_batch_urls,
     sanitize_filename,
     sanitize_path,
+    expand_path,
     prepend_extension,
     replace_extension,
     remove_start,
@@ -95,6 +96,8 @@ from youtube_dl.utils import (
 from youtube_dl.compat import (
     compat_chr,
     compat_etree_fromstring,
+    compat_getenv,
+    compat_setenv,
     compat_urlparse,
     compat_parse_qs,
 )
@@ -214,6 +217,18 @@ class TestUtil(unittest.TestCase):
         self.assertEqual(sanitize_path('./abc'), 'abc')
         self.assertEqual(sanitize_path('./../abc'), '..\\abc')
 
+    def test_expand_path(self):
+        def env(var):
+            return '%{0}%'.format(var) if sys.platform == 'win32' else '${0}'.format(var)
+
+        compat_setenv('YOUTUBE_DL_EXPATH_PATH', 'expanded')
+        self.assertEqual(expand_path(env('YOUTUBE_DL_EXPATH_PATH')), 'expanded')
+        self.assertEqual(expand_path(env('HOME')), compat_getenv('HOME'))
+        self.assertEqual(expand_path('~'), compat_getenv('HOME'))
+        self.assertEqual(
+            expand_path('~/%s' % env('YOUTUBE_DL_EXPATH_PATH')),
+            '%s/expanded' % compat_getenv('HOME'))
+
     def test_prepend_extension(self):
         self.assertEqual(prepend_extension('abc.ext', 'temp'), 'abc.temp.ext')
         self.assertEqual(prepend_extension('abc.ext', 'temp', 'ext'), 'abc.temp.ext')

youtube_dl/YoutubeDL.py

@@ -29,7 +29,6 @@ import random
 from .compat import (
     compat_basestring,
     compat_cookiejar,
-    compat_expanduser,
     compat_get_terminal_size,
     compat_http_client,
     compat_kwargs,
@@ -54,6 +53,7 @@ from .utils import (
     encode_compat_str,
     encodeFilename,
     error_to_compat_str,
+    expand_path,
     ExtractorError,
     format_bytes,
     formatSeconds,
@@ -672,7 +672,7 @@ class YoutubeDL(object):
                         FORMAT_RE.format(numeric_field),
                         r'%({0})s'.format(numeric_field), outtmpl)
 
-            tmpl = compat_expanduser(outtmpl)
+            tmpl = expand_path(outtmpl)
             filename = tmpl % template_dict
             # Temporary fix for #4787
             # 'Treat' all problem characters by passing filename through preferredencoding
@@ -837,6 +837,12 @@ class YoutubeDL(object):
                 ie_result['url'], ie_key=ie_result.get('ie_key'),
                 extra_info=extra_info, download=False, process=False)
 
+            # extract_info may return None when ignoreerrors is enabled and
+            # extraction failed with an error, don't crash and return early
+            # in this case
+            if not info:
+                return info
+
             force_properties = dict(
                 (k, v) for k, v in ie_result.items() if v is not None)
             for f in ('_type', 'url', 'ie_key'):
@@ -2170,7 +2176,7 @@ class YoutubeDL(object):
         if opts_cookiefile is None:
             self.cookiejar = compat_cookiejar.CookieJar()
         else:
-            opts_cookiefile = compat_expanduser(opts_cookiefile)
+            opts_cookiefile = expand_path(opts_cookiefile)
             self.cookiejar = compat_cookiejar.MozillaCookieJar(
                 opts_cookiefile)
             if os.access(opts_cookiefile, os.R_OK):

youtube_dl/__init__.py

@@ -16,7 +16,6 @@ from .options import (
     parseOpts,
 )
 from .compat import (
-    compat_expanduser,
     compat_getpass,
     compat_shlex_split,
     workaround_optparse_bug9161,
@@ -26,6 +25,7 @@ from .utils import (
     decodeOption,
     DEFAULT_OUTTMPL,
     DownloadError,
+    expand_path,
     match_filter_func,
     MaxDownloadsReached,
     preferredencoding,
@@ -88,7 +88,7 @@ def _real_main(argv=None):
                 batchfd = sys.stdin
             else:
                 batchfd = io.open(
-                    compat_expanduser(opts.batchfile),
+                    expand_path(opts.batchfile),
                     'r', encoding='utf-8', errors='ignore')
             batch_urls = read_batch_urls(batchfd)
             if opts.verbose:
@@ -238,7 +238,7 @@ def _real_main(argv=None):
 
     any_getting = opts.geturl or opts.gettitle or opts.getid or opts.getthumbnail or opts.getdescription or opts.getfilename or opts.getformat or opts.getduration or opts.dumpjson or opts.dump_single_json
     any_printing = opts.print_json
-    download_archive_fn = compat_expanduser(opts.download_archive) if opts.download_archive is not None else opts.download_archive
+    download_archive_fn = expand_path(opts.download_archive) if opts.download_archive is not None else opts.download_archive
 
     # PostProcessors
     postprocessors = []
@@ -449,7 +449,7 @@ def _real_main(argv=None):
 
         try:
             if opts.load_info_filename is not None:
-                retcode = ydl.download_with_info_file(compat_expanduser(opts.load_info_filename))
+                retcode = ydl.download_with_info_file(expand_path(opts.load_info_filename))
             else:
                 retcode = ydl.download(all_urls)
         except MaxDownloadsReached:

youtube_dl/cache.py

@@ -8,8 +8,11 @@ import re
 import shutil
 import traceback
 
-from .compat import compat_expanduser, compat_getenv
-from .utils import write_json_file
+from .compat import compat_getenv
+from .utils import (
+    expand_path,
+    write_json_file,
+)
 
 
 class Cache(object):
@@ -21,7 +24,7 @@ class Cache(object):
         if res is None:
             cache_root = compat_getenv('XDG_CACHE_HOME', '~/.cache')
             res = os.path.join(cache_root, 'youtube-dl')
-        return compat_expanduser(res)
+        return expand_path(res)
 
     def _get_cache_fn(self, section, key, dtype):
         assert re.match(r'^[a-zA-Z0-9_.-]+$', section), \

youtube_dl/downloader/__init__.py

@@ -43,6 +43,9 @@ def get_suitable_downloader(info_dict, params={}):
         if ed.can_download(info_dict):
             return ed
 
+    if protocol.startswith('m3u8') and info_dict.get('is_live'):
+        return FFmpegFD
+
     if protocol == 'm3u8' and params.get('hls_prefer_native') is True:
         return HlsFD
 

youtube_dl/extractor/adn.py

@@ -0,0 +1,136 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import base64
+import json
+import os
+
+from .common import InfoExtractor
+from ..aes import aes_cbc_decrypt
+from ..compat import compat_ord
+from ..utils import (
+    bytes_to_intlist,
+    ExtractorError,
+    float_or_none,
+    intlist_to_bytes,
+    srt_subtitles_timecode,
+    strip_or_none,
+)
+
+
+class ADNIE(InfoExtractor):
+    IE_DESC = 'Anime Digital Network'
+    _VALID_URL = r'https?://(?:www\.)?animedigitalnetwork\.fr/video/[^/]+/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://animedigitalnetwork.fr/video/blue-exorcist-kyoto-saga/7778-episode-1-debut-des-hostilites',
+        'md5': 'e497370d847fd79d9d4c74be55575c7a',
+        'info_dict': {
+            'id': '7778',
+            'ext': 'mp4',
+            'title': 'Blue Exorcist - Kyôto Saga - Épisode 1',
+            'description': 'md5:2f7b5aa76edbc1a7a92cedcda8a528d5',
+        }
+    }
+
+    def _get_subtitles(self, sub_path, video_id):
+        if not sub_path:
+            return None
+
+        enc_subtitles = self._download_webpage(
+            'http://animedigitalnetwork.fr/' + sub_path,
+            video_id, fatal=False)
+        if not enc_subtitles:
+            return None
+
+        # http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js
+        dec_subtitles = intlist_to_bytes(aes_cbc_decrypt(
+            bytes_to_intlist(base64.b64decode(enc_subtitles[24:])),
+            bytes_to_intlist(b'\xb5@\xcfq\xa3\x98"N\xe4\xf3\x12\x98}}\x16\xd8'),
+            bytes_to_intlist(base64.b64decode(enc_subtitles[:24]))
+        ))
+        subtitles_json = self._parse_json(
+            dec_subtitles[:-compat_ord(dec_subtitles[-1])],
+            None, fatal=False)
+        if not subtitles_json:
+            return None
+
+        subtitles = {}
+        for sub_lang, sub in subtitles_json.items():
+            srt = ''
+            for num, current in enumerate(sub):
+                start, end, text = (
+                    float_or_none(current.get('startTime')),
+                    float_or_none(current.get('endTime')),
+                    current.get('text'))
+                if start is None or end is None or text is None:
+                    continue
+                srt += os.linesep.join(
+                    (
+                        '%d' % num,
+                        '%s --> %s' % (
+                            srt_subtitles_timecode(start),
+                            srt_subtitles_timecode(end)),
+                        text,
+                        os.linesep,
+                    ))
+
+            if sub_lang == 'vostf':
+                sub_lang = 'fr'
+            subtitles.setdefault(sub_lang, []).extend([{
+                'ext': 'json',
+                'data': json.dumps(sub),
+            }, {
+                'ext': 'srt',
+                'data': srt,
+            }])
+        return subtitles
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        player_config = self._parse_json(self._search_regex(
+            r'playerConfig\s*=\s*({.+});', webpage, 'player config'), video_id)
+
+        video_info = {}
+        video_info_str = self._search_regex(
+            r'videoInfo\s*=\s*({.+});', webpage,
+            'video info', fatal=False)
+        if video_info_str:
+            video_info = self._parse_json(
+                video_info_str, video_id, fatal=False) or {}
+
+        options = player_config.get('options') or {}
+        metas = options.get('metas') or {}
+        title = metas.get('title') or video_info['title']
+        links = player_config.get('links') or {}
+
+        formats = []
+        for format_id, qualities in links.items():
+            for load_balancer_url in qualities.values():
+                load_balancer_data = self._download_json(
+                    load_balancer_url, video_id, fatal=False) or {}
+                m3u8_url = load_balancer_data.get('location')
+                if not m3u8_url:
+                    continue
+                m3u8_formats = self._extract_m3u8_formats(
+                    m3u8_url, video_id, 'mp4', 'm3u8_native',
+                    m3u8_id=format_id, fatal=False)
+                if format_id == 'vf':
+                    for f in m3u8_formats:
+                        f['language'] = 'fr'
+                formats.extend(m3u8_formats)
+        error = options.get('error')
+        if not formats and error:
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True)
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'description': strip_or_none(metas.get('summary') or video_info.get('resume')),
+            'thumbnail': video_info.get('image'),
+            'formats': formats,
+            'subtitles': self.extract_subtitles(player_config.get('subtitles'), video_id),
+            'episode': metas.get('subtitle') or video_info.get('videoTitle'),
+            'series': video_info.get('playlistTitle'),
+        }

youtube_dl/extractor/afreecatv.py

@@ -4,15 +4,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_urllib_parse_urlparse,
-    compat_urlparse,
-)
+from ..compat import compat_xpath
 from ..utils import (
     ExtractorError,
     int_or_none,
-    update_url_query,
-    xpath_element,
     xpath_text,
 )
 
@@ -43,7 +38,8 @@ class AfreecaTVIE(InfoExtractor):
             'uploader': 'dailyapril',
             'uploader_id': 'dailyapril',
             'upload_date': '20160503',
-        }
+        },
+        'skip': 'Video is gone',
     }, {
         'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
         'info_dict': {
@@ -71,6 +67,19 @@ class AfreecaTVIE(InfoExtractor):
                 'upload_date': '20160502',
             },
         }],
+        'skip': 'Video is gone',
+    }, {
+        'url': 'http://vod.afreecatv.com/PLAYER/STATION/18650793',
+        'info_dict': {
+            'id': '18650793',
+            'ext': 'flv',
+            'uploader': '윈아디',
+            'uploader_id': 'badkids',
+            'title': '오늘은 다르다! 쏘님의 우월한 위아래~ 댄스리액션!',
+        },
+        'params': {
+            'skip_download': True,  # requires rtmpdump
+        },
     }, {
         'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
         'only_matching': True,
@@ -90,40 +99,33 @@ class AfreecaTVIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        parsed_url = compat_urllib_parse_urlparse(url)
-        info_url = compat_urlparse.urlunparse(parsed_url._replace(
-            netloc='afbbs.afreecatv.com:8080',
-            path='/api/video/get_video_info.php'))
 
         video_xml = self._download_xml(
-            update_url_query(info_url, {'nTitleNo': video_id}), video_id)
+            'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php',
+            video_id, query={'nTitleNo': video_id})
 
-        if xpath_element(video_xml, './track/video/file') is None:
+        video_element = video_xml.findall(compat_xpath('./track/video'))[1]
+        if video_element is None or video_element.text is None:
             raise ExtractorError('Specified AfreecaTV video does not exist',
                                  expected=True)
 
-        title = xpath_text(video_xml, './track/title', 'title')
+        video_url_raw = video_element.text
+
+        app, playpath = video_url_raw.split('mp4:')
+
+        title = xpath_text(video_xml, './track/title', 'title', fatal=True)
         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
         duration = int_or_none(xpath_text(video_xml, './track/duration',
                                           'duration'))
         thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
 
-        entries = []
-        for i, video_file in enumerate(video_xml.findall('./track/video/file')):
-            video_key = self.parse_video_key(video_file.get('key', ''))
-            if not video_key:
-                continue
-            entries.append({
-                'id': '%s_%s' % (video_id, video_key.get('part', i + 1)),
-                'title': title,
-                'upload_date': video_key.get('upload_date'),
-                'duration': int_or_none(video_file.get('duration')),
-                'url': video_file.text,
-            })
-
-        info = {
+        return {
             'id': video_id,
+            'url': app,
+            'ext': 'flv',
+            'play_path': 'mp4:' + playpath,
+            'rtmp_live': True,  # downloading won't end without this
             'title': title,
             'uploader': uploader,
             'uploader_id': uploader_id,
@@ -131,20 +133,6 @@ class AfreecaTVIE(InfoExtractor):
             'thumbnail': thumbnail,
         }
 
-        if len(entries) > 1:
-            info['_type'] = 'multi_video'
-            info['entries'] = entries
-        elif len(entries) == 1:
-            info['url'] = entries[0]['url']
-            info['upload_date'] = entries[0].get('upload_date')
-        else:
-            raise ExtractorError(
-                'No files found for the specified AfreecaTV video, either'
-                ' the URL is incorrect or the video has been made private.',
-                expected=True)
-
-        return info
-
 
 class AfreecaTVGlobalIE(AfreecaTVIE):
     IE_NAME = 'afreecatv:global'

youtube_dl/extractor/allocine.py

@@ -2,9 +2,13 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
-    remove_end,
+    int_or_none,
     qualities,
+    remove_end,
+    try_get,
+    unified_timestamp,
     url_basename,
 )
 
@@ -22,6 +26,10 @@ class AllocineIE(InfoExtractor):
             'title': 'Astérix - Le Domaine des Dieux Teaser VF',
             'description': 'md5:4a754271d9c6f16c72629a8a993ee884',
             'thumbnail': r're:http://.*\.jpg',
+            'duration': 39,
+            'timestamp': 1404273600,
+            'upload_date': '20140702',
+            'view_count': int,
         },
     }, {
         'url': 'http://www.allocine.fr/video/player_gen_cmedia=19540403&cfilm=222257.html',
@@ -33,6 +41,10 @@ class AllocineIE(InfoExtractor):
             'title': 'Planes 2 Bande-annonce VF',
             'description': 'Regardez la bande annonce du film Planes 2 (Planes 2 Bande-annonce VF). Planes 2, un film de Roberts Gannaway',
             'thumbnail': r're:http://.*\.jpg',
+            'duration': 69,
+            'timestamp': 1385659800,
+            'upload_date': '20131128',
+            'view_count': int,
         },
     }, {
         'url': 'http://www.allocine.fr/video/player_gen_cmedia=19544709&cfilm=181290.html',
@@ -44,6 +56,10 @@ class AllocineIE(InfoExtractor):
             'title': 'Dragons 2 - Bande annonce finale VF',
             'description': 'md5:6cdd2d7c2687d4c6aafe80a35e17267a',
             'thumbnail': r're:http://.*\.jpg',
+            'duration': 144,
+            'timestamp': 1397589900,
+            'upload_date': '20140415',
+            'view_count': int,
         },
     }, {
         'url': 'http://www.allocine.fr/video/video-19550147/',
@@ -69,34 +85,37 @@ class AllocineIE(InfoExtractor):
             r'data-model="([^"]+)"', webpage, 'data model', default=None)
         if model:
             model_data = self._parse_json(model, display_id)
-
-            for video_url in model_data['sources'].values():
+            video = model_data['videos'][0]
+            title = video['title']
+            for video_url in video['sources'].values():
                 video_id, format_id = url_basename(video_url).split('_')[:2]
                 formats.append({
                     'format_id': format_id,
                     'quality': quality(format_id),
                     'url': video_url,
                 })
-
-            title = model_data['title']
+            duration = int_or_none(video.get('duration'))
+            view_count = int_or_none(video.get('view_count'))
+            timestamp = unified_timestamp(try_get(
+                video, lambda x: x['added_at']['date'], compat_str))
         else:
             video_id = display_id
             media_data = self._download_json(
                 'http://www.allocine.fr/ws/AcVisiondataV5.ashx?media=%s' % video_id, display_id)
+            title = remove_end(
+                self._html_search_regex(
+                    r'(?s)<title>(.+?)</title>', webpage, 'title').strip(),
+                ' - AlloCiné')
             for key, value in media_data['video'].items():
                 if not key.endswith('Path'):
                     continue
-
                 format_id = key[:-len('Path')]
                 formats.append({
                     'format_id': format_id,
                     'quality': quality(format_id),
                     'url': value,
                 })
-
-            title = remove_end(self._html_search_regex(
-                r'(?s)<title>(.+?)</title>', webpage, 'title'
-            ).strip(), ' - AlloCiné')
+            duration, view_count, timestamp = [None] * 3
 
         self._sort_formats(formats)
 
@@ -104,7 +123,10 @@ class AllocineIE(InfoExtractor):
             'id': video_id,
             'display_id': display_id,
             'title': title,
-            'thumbnail': self._og_search_thumbnail(webpage),
-            'formats': formats,
             'description': self._og_search_description(webpage),
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'duration': duration,
+            'timestamp': timestamp,
+            'view_count': view_count,
+            'formats': formats,
         }

youtube_dl/extractor/arkena.py

@@ -93,8 +93,7 @@ class ArkenaIE(InfoExtractor):
                 exts = (mimetype2ext(f.get('Type')), determine_ext(f_url, None))
                 if kind == 'm3u8' or 'm3u8' in exts:
                     formats.extend(self._extract_m3u8_formats(
-                        f_url, video_id, 'mp4',
-                        entry_protocol='m3u8' if is_live else 'm3u8_native',
+                        f_url, video_id, 'mp4', 'm3u8_native',
                         m3u8_id=kind, fatal=False, live=is_live))
                 elif kind == 'flash' or 'f4m' in exts:
                     formats.extend(self._extract_f4m_formats(

youtube_dl/extractor/atresplayer.py

@@ -90,7 +90,8 @@ class AtresPlayerIE(InfoExtractor):
             request, None, 'Logging in as %s' % username)
 
         error = self._html_search_regex(
-            r'(?s)<ul class="list_error">(.+?)</ul>', response, 'error', default=None)
+            r'(?s)<ul[^>]+class="[^"]*\blist_error\b[^"]*">(.+?)</ul>',
+            response, 'error', default=None)
         if error:
             raise ExtractorError(
                 'Unable to login: %s' % error, expected=True)
@@ -155,13 +156,17 @@ class AtresPlayerIE(InfoExtractor):
             if format_id == 'token' or not video_url.startswith('http'):
                 continue
             if 'geodeswowsmpra3player' in video_url:
-                f4m_path = video_url.split('smil:', 1)[-1].split('free_', 1)[0]
-                f4m_url = 'http://drg.antena3.com/{0}hds/es/sd.f4m'.format(f4m_path)
+                # f4m_path = video_url.split('smil:', 1)[-1].split('free_', 1)[0]
+                # f4m_url = 'http://drg.antena3.com/{0}hds/es/sd.f4m'.format(f4m_path)
                 # this videos are protected by DRM, the f4m downloader doesn't support them
                 continue
-            else:
-                f4m_url = video_url[:-9] + '/manifest.f4m'
-            formats.extend(self._extract_f4m_formats(f4m_url, video_id, f4m_id='hds', fatal=False))
+            video_url_hd = video_url.replace('free_es', 'es')
+            formats.extend(self._extract_f4m_formats(
+                video_url_hd[:-9] + '/manifest.f4m', video_id, f4m_id='hds',
+                fatal=False))
+            formats.extend(self._extract_mpd_formats(
+                video_url_hd[:-9] + '/manifest.mpd', video_id, mpd_id='dash',
+                fatal=False))
         self._sort_formats(formats)
 
         path_data = player.get('pathData')

youtube_dl/extractor/atvat.py

@@ -0,0 +1,73 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    determine_ext,
+    int_or_none,
+    unescapeHTML,
+)
+
+
+class ATVAtIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?atv\.at/(?:[^/]+/){2}(?P<id>[dv]\d+)'
+    _TESTS = [{
+        'url': 'http://atv.at/aktuell/di-210317-2005-uhr/v1698449/',
+        'md5': 'c3b6b975fb3150fc628572939df205f2',
+        'info_dict': {
+            'id': '1698447',
+            'ext': 'mp4',
+            'title': 'DI, 21.03.17 | 20:05 Uhr 1/1',
+        }
+    }, {
+        'url': 'http://atv.at/aktuell/meinrad-knapp/d8416/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        video_data = self._parse_json(unescapeHTML(self._search_regex(
+            r'class="[^"]*jsb_video/FlashPlayer[^"]*"[^>]+data-jsb="([^"]+)"',
+            webpage, 'player data')), display_id)['config']['initial_video']
+
+        video_id = video_data['id']
+        video_title = video_data['title']
+
+        parts = []
+        for part in video_data.get('parts', []):
+            part_id = part['id']
+            part_title = part['title']
+
+            formats = []
+            for source in part.get('sources', []):
+                source_url = source.get('src')
+                if not source_url:
+                    continue
+                ext = determine_ext(source_url)
+                if ext == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        source_url, part_id, 'mp4', 'm3u8_native',
+                        m3u8_id='hls', fatal=False))
+                else:
+                    formats.append({
+                        'format_id': source.get('delivery'),
+                        'url': source_url,
+                    })
+            self._sort_formats(formats)
+
+            parts.append({
+                'id': part_id,
+                'title': part_title,
+                'thumbnail': part.get('preview_image_url'),
+                'duration': int_or_none(part.get('duration')),
+                'is_live': part.get('is_livestream'),
+                'formats': formats,
+            })
+
+        return {
+            '_type': 'multi_video',
+            'id': video_id,
+            'title': video_title,
+            'entries': parts,
+        }

youtube_dl/extractor/bellmedia.py

@@ -21,10 +21,11 @@ class BellMediaIE(InfoExtractor):
                 animalplanet|
                 bravo|
                 mtv|
-                space
+                space|
+                etalk
             )\.ca|
             much\.com
-        )/.*?(?:\bvid=|-vid|~|%7E|/(?:episode)?)(?P<id>[0-9]{6,})'''
+        )/.*?(?:\bvid(?:eoid)?=|-vid|~|%7E|/(?:episode)?)(?P<id>[0-9]{6,})'''
     _TESTS = [{
         'url': 'http://www.ctv.ca/video/player?vid=706966',
         'md5': 'ff2ebbeae0aa2dcc32a830c3fd69b7b0',
@@ -58,6 +59,9 @@ class BellMediaIE(InfoExtractor):
     }, {
         'url': 'http://www.ctv.ca/DCs-Legends-of-Tomorrow/Video/S2E11-Turncoat-vid1051430',
         'only_matching': True,
+    }, {
+        'url': 'http://www.etalk.ca/video?videoid=663455',
+        'only_matching': True,
     }]
     _DOMAINS = {
         'thecomedynetwork': 'comedy',
@@ -65,6 +69,7 @@ class BellMediaIE(InfoExtractor):
         'sciencechannel': 'discsci',
         'investigationdiscovery': 'invdisc',
         'animalplanet': 'aniplan',
+        'etalk': 'ctv',
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/ceskatelevize.py

@@ -160,8 +160,7 @@ class CeskaTelevizeIE(InfoExtractor):
                 for format_id, stream_url in item.get('streamUrls', {}).items():
                     if 'playerType=flash' in stream_url:
                         stream_formats = self._extract_m3u8_formats(
-                            stream_url, playlist_id, 'mp4',
-                            entry_protocol='m3u8' if is_live else 'm3u8_native',
+                            stream_url, playlist_id, 'mp4', 'm3u8_native',
                             m3u8_id='hls-%s' % format_id, fatal=False)
                     else:
                         stream_formats = self._extract_mpd_formats(

youtube_dl/extractor/channel9.py

@@ -4,62 +4,62 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
+    clean_html,
     ExtractorError,
-    parse_filesize,
+    int_or_none,
+    parse_iso8601,
     qualities,
+    unescapeHTML,
 )
 
 
 class Channel9IE(InfoExtractor):
-    '''
-    Common extractor for channel9.msdn.com.
-
-    The type of provided URL (video or playlist) is determined according to
-    meta Search.PageType from web page HTML rather than URL itself, as it is
-    not always possible to do.
-    '''
     IE_DESC = 'Channel 9'
     IE_NAME = 'channel9'
-    _VALID_URL = r'https?://(?:www\.)?channel9\.msdn\.com/(?P<contentpath>.+?)(?P<rss>/RSS)?/?(?:[?#&]|$)'
+    _VALID_URL = r'https?://(?:www\.)?(?:channel9\.msdn\.com|s\.ch9\.ms)/(?P<contentpath>.+?)(?P<rss>/RSS)?/?(?:[?#&]|$)'
 
     _TESTS = [{
         'url': 'http://channel9.msdn.com/Events/TechEd/Australia/2013/KOS002',
-        'md5': 'bbd75296ba47916b754e73c3a4bbdf10',
+        'md5': '32083d4eaf1946db6d454313f44510ca',
         'info_dict': {
-            'id': 'Events/TechEd/Australia/2013/KOS002',
-            'ext': 'mp4',
+            'id': '6c413323-383a-49dc-88f9-a22800cab024',
+            'ext': 'wmv',
             'title': 'Developer Kick-Off Session: Stuff We Love',
-            'description': 'md5:c08d72240b7c87fcecafe2692f80e35f',
+            'description': 'md5:b80bf9355a503c193aff7ec6cd5a7731',
             'duration': 4576,
-            'thumbnail': r're:http://.*\.jpg',
+            'thumbnail': r're:https?://.*\.jpg',
+            'timestamp': 1377717420,
+            'upload_date': '20130828',
             'session_code': 'KOS002',
-            'session_day': 'Day 1',
             'session_room': 'Arena 1A',
-            'session_speakers': ['Ed Blankenship', 'Andrew Coates', 'Brady Gaster', 'Patrick Klug',
-                                 'Mads Kristensen'],
+            'session_speakers': ['Andrew Coates', 'Brady Gaster', 'Mads Kristensen', 'Ed Blankenship', 'Patrick Klug'],
         },
     }, {
         'url': 'http://channel9.msdn.com/posts/Self-service-BI-with-Power-BI-nuclear-testing',
-        'md5': 'b43ee4529d111bc37ba7ee4f34813e68',
+        'md5': 'dcf983ee6acd2088e7188c3cf79b46bc',
         'info_dict': {
-            'id': 'posts/Self-service-BI-with-Power-BI-nuclear-testing',
-            'ext': 'mp4',
+            'id': 'fe8e435f-bb93-4e01-8e97-a28c01887024',
+            'ext': 'wmv',
             'title': 'Self-service BI with Power BI - nuclear testing',
-            'description': 'md5:d1e6ecaafa7fb52a2cacdf9599829f5b',
+            'description': 'md5:2d17fec927fc91e9e17783b3ecc88f54',
             'duration': 1540,
-            'thumbnail': r're:http://.*\.jpg',
+            'thumbnail': r're:https?://.*\.jpg',
+            'timestamp': 1386381991,
+            'upload_date': '20131207',
             'authors': ['Mike Wilmot'],
         },
     }, {
         # low quality mp4 is best
         'url': 'https://channel9.msdn.com/Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
         'info_dict': {
-            'id': 'Events/CPP/CppCon-2015/Ranges-for-the-Standard-Library',
+            'id': '33ad69d2-6a4e-4172-83a1-a523013dec76',
             'ext': 'mp4',
             'title': 'Ranges for the Standard Library',
-            'description': 'md5:2e6b4917677af3728c5f6d63784c4c5d',
+            'description': 'md5:9895e0a9fd80822d2f01c454b8f4a372',
             'duration': 5646,
-            'thumbnail': r're:http://.*\.jpg',
+            'thumbnail': r're:https?://.*\.jpg',
+            'upload_date': '20150930',
+            'timestamp': 1443640735,
         },
         'params': {
             'skip_download': True,
@@ -70,7 +70,7 @@ class Channel9IE(InfoExtractor):
             'id': 'Niners/Splendid22/Queue/76acff796e8f411184b008028e0d492b',
             'title': 'Channel 9',
         },
-        'playlist_count': 2,
+        'playlist_mincount': 100,
     }, {
         'url': 'https://channel9.msdn.com/Events/DEVintersection/DEVintersection-2016/RSS',
         'only_matching': True,
@@ -81,189 +81,6 @@ class Channel9IE(InfoExtractor):
 
     _RSS_URL = 'http://channel9.msdn.com/%s/RSS'
 
-    def _formats_from_html(self, html):
-        FORMAT_REGEX = r'''
-            (?x)
-            <a\s+href="(?P<url>[^"]+)">(?P<quality>[^<]+)</a>\s*
-            <span\s+class="usage">\((?P<note>[^\)]+)\)</span>\s*
-            (?:<div\s+class="popup\s+rounded">\s*
-            <h3>File\s+size</h3>\s*(?P<filesize>.*?)\s*
-            </div>)?                                                # File size part may be missing
-        '''
-        quality = qualities((
-            'MP3', 'MP4',
-            'Low Quality WMV', 'Low Quality MP4',
-            'Mid Quality WMV', 'Mid Quality MP4',
-            'High Quality WMV', 'High Quality MP4'))
-        formats = [{
-            'url': x.group('url'),
-            'format_id': x.group('quality'),
-            'format_note': x.group('note'),
-            'format': '%s (%s)' % (x.group('quality'), x.group('note')),
-            'filesize_approx': parse_filesize(x.group('filesize')),
-            'quality': quality(x.group('quality')),
-            'vcodec': 'none' if x.group('note') == 'Audio only' else None,
-        } for x in list(re.finditer(FORMAT_REGEX, html))]
-
-        self._sort_formats(formats)
-
-        return formats
-
-    def _extract_title(self, html):
-        title = self._html_search_meta('title', html, 'title')
-        if title is None:
-            title = self._og_search_title(html)
-            TITLE_SUFFIX = ' (Channel 9)'
-            if title is not None and title.endswith(TITLE_SUFFIX):
-                title = title[:-len(TITLE_SUFFIX)]
-        return title
-
-    def _extract_description(self, html):
-        DESCRIPTION_REGEX = r'''(?sx)
-            <div\s+class="entry-content">\s*
-            <div\s+id="entry-body">\s*
-            (?P<description>.+?)\s*
-            </div>\s*
-            </div>
-        '''
-        m = re.search(DESCRIPTION_REGEX, html)
-        if m is not None:
-            return m.group('description')
-        return self._html_search_meta('description', html, 'description')
-
-    def _extract_duration(self, html):
-        m = re.search(r'"length": *"(?P<hours>\d{2}):(?P<minutes>\d{2}):(?P<seconds>\d{2})"', html)
-        return ((int(m.group('hours')) * 60 * 60) + (int(m.group('minutes')) * 60) + int(m.group('seconds'))) if m else None
-
-    def _extract_slides(self, html):
-        m = re.search(r'<a href="(?P<slidesurl>[^"]+)" class="slides">Slides</a>', html)
-        return m.group('slidesurl') if m is not None else None
-
-    def _extract_zip(self, html):
-        m = re.search(r'<a href="(?P<zipurl>[^"]+)" class="zip">Zip</a>', html)
-        return m.group('zipurl') if m is not None else None
-
-    def _extract_avg_rating(self, html):
-        m = re.search(r'<p class="avg-rating">Avg Rating: <span>(?P<avgrating>[^<]+)</span></p>', html)
-        return float(m.group('avgrating')) if m is not None else 0
-
-    def _extract_rating_count(self, html):
-        m = re.search(r'<div class="rating-count">\((?P<ratingcount>[^<]+)\)</div>', html)
-        return int(self._fix_count(m.group('ratingcount'))) if m is not None else 0
-
-    def _extract_view_count(self, html):
-        m = re.search(r'<li class="views">\s*<span class="count">(?P<viewcount>[^<]+)</span> Views\s*</li>', html)
-        return int(self._fix_count(m.group('viewcount'))) if m is not None else 0
-
-    def _extract_comment_count(self, html):
-        m = re.search(r'<li class="comments">\s*<a href="#comments">\s*<span class="count">(?P<commentcount>[^<]+)</span> Comments\s*</a>\s*</li>', html)
-        return int(self._fix_count(m.group('commentcount'))) if m is not None else 0
-
-    def _fix_count(self, count):
-        return int(str(count).replace(',', '')) if count is not None else None
-
-    def _extract_authors(self, html):
-        m = re.search(r'(?s)<li class="author">(.*?)</li>', html)
-        if m is None:
-            return None
-        return re.findall(r'<a href="/Niners/[^"]+">([^<]+)</a>', m.group(1))
-
-    def _extract_session_code(self, html):
-        m = re.search(r'<li class="code">\s*(?P<code>.+?)\s*</li>', html)
-        return m.group('code') if m is not None else None
-
-    def _extract_session_day(self, html):
-        m = re.search(r'<li class="day">\s*<a href="/Events/[^"]+">(?P<day>[^<]+)</a>\s*</li>', html)
-        return m.group('day').strip() if m is not None else None
-
-    def _extract_session_room(self, html):
-        m = re.search(r'<li class="room">\s*(?P<room>.+?)\s*</li>', html)
-        return m.group('room') if m is not None else None
-
-    def _extract_session_speakers(self, html):
-        return re.findall(r'<a href="/Events/Speakers/[^"]+">([^<]+)</a>', html)
-
-    def _extract_content(self, html, content_path):
-        # Look for downloadable content
-        formats = self._formats_from_html(html)
-        slides = self._extract_slides(html)
-        zip_ = self._extract_zip(html)
-
-        # Nothing to download
-        if len(formats) == 0 and slides is None and zip_ is None:
-            self._downloader.report_warning('None of recording, slides or zip are available for %s' % content_path)
-            return
-
-        # Extract meta
-        title = self._extract_title(html)
-        description = self._extract_description(html)
-        thumbnail = self._og_search_thumbnail(html)
-        duration = self._extract_duration(html)
-        avg_rating = self._extract_avg_rating(html)
-        rating_count = self._extract_rating_count(html)
-        view_count = self._extract_view_count(html)
-        comment_count = self._extract_comment_count(html)
-
-        common = {
-            '_type': 'video',
-            'id': content_path,
-            'description': description,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'avg_rating': avg_rating,
-            'rating_count': rating_count,
-            'view_count': view_count,
-            'comment_count': comment_count,
-        }
-
-        result = []
-
-        if slides is not None:
-            d = common.copy()
-            d.update({'title': title + '-Slides', 'url': slides})
-            result.append(d)
-
-        if zip_ is not None:
-            d = common.copy()
-            d.update({'title': title + '-Zip', 'url': zip_})
-            result.append(d)
-
-        if len(formats) > 0:
-            d = common.copy()
-            d.update({'title': title, 'formats': formats})
-            result.append(d)
-
-        return result
-
-    def _extract_entry_item(self, html, content_path):
-        contents = self._extract_content(html, content_path)
-        if contents is None:
-            return contents
-
-        if len(contents) > 1:
-            raise ExtractorError('Got more than one entry')
-        result = contents[0]
-        result['authors'] = self._extract_authors(html)
-
-        return result
-
-    def _extract_session(self, html, content_path):
-        contents = self._extract_content(html, content_path)
-        if contents is None:
-            return contents
-
-        session_meta = {
-            'session_code': self._extract_session_code(html),
-            'session_day': self._extract_session_day(html),
-            'session_room': self._extract_session_room(html),
-            'session_speakers': self._extract_session_speakers(html),
-        }
-
-        for content in contents:
-            content.update(session_meta)
-
-        return self.playlist_result(contents)
-
     def _extract_list(self, video_id, rss_url=None):
         if not rss_url:
             rss_url = self._RSS_URL % video_id
@@ -274,9 +91,7 @@ class Channel9IE(InfoExtractor):
         return self.playlist_result(entries, video_id, title_text)
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        content_path = mobj.group('contentpath')
-        rss = mobj.group('rss')
+        content_path, rss = re.match(self._VALID_URL, url).groups()
 
         if rss:
             return self._extract_list(content_path, url)
@@ -284,17 +99,158 @@ class Channel9IE(InfoExtractor):
         webpage = self._download_webpage(
             url, content_path, 'Downloading web page')
 
-        page_type = self._search_regex(
-            r'<meta[^>]+name=(["\'])WT\.entryid\1[^>]+content=(["\'])(?P<pagetype>[^:]+).+?\2',
-            webpage, 'page type', default=None, group='pagetype')
-        if page_type:
-            if page_type == 'Entry':      # Any 'item'-like page, may contain downloadable content
-                return self._extract_entry_item(webpage, content_path)
-            elif page_type == 'Session':  # Event session page, may contain downloadable content
-                return self._extract_session(webpage, content_path)
-            elif page_type == 'Event':
-                return self._extract_list(content_path)
+        episode_data = self._search_regex(
+            r"data-episode='([^']+)'", webpage, 'episode data', default=None)
+        if episode_data:
+            episode_data = self._parse_json(unescapeHTML(
+                episode_data), content_path)
+            content_id = episode_data['contentId']
+            is_session = '/Sessions(' in episode_data['api']
+            content_url = 'https://channel9.msdn.com/odata' + episode_data['api']
+            if is_session:
+                content_url += '?$expand=Speakers'
             else:
-                raise ExtractorError('Unexpected WT.entryid %s' % page_type, expected=True)
-        else:  # Assuming list
+                content_url += '?$expand=Authors'
+            content_data = self._download_json(content_url, content_id)
+            title = content_data['Title']
+
+            QUALITIES = (
+                'mp3',
+                'wmv', 'mp4',
+                'wmv-low', 'mp4-low',
+                'wmv-mid', 'mp4-mid',
+                'wmv-high', 'mp4-high',
+            )
+
+            quality_key = qualities(QUALITIES)
+
+            def quality(quality_id, format_url):
+                return (len(QUALITIES) if '_Source.' in format_url
+                        else quality_key(quality_id))
+
+            formats = []
+            urls = set()
+
+            SITE_QUALITIES = {
+                'MP3': 'mp3',
+                'MP4': 'mp4',
+                'Low Quality WMV': 'wmv-low',
+                'Low Quality MP4': 'mp4-low',
+                'Mid Quality WMV': 'wmv-mid',
+                'Mid Quality MP4': 'mp4-mid',
+                'High Quality WMV': 'wmv-high',
+                'High Quality MP4': 'mp4-high',
+            }
+
+            formats_select = self._search_regex(
+                r'(?s)<select[^>]+name=["\']format[^>]+>(.+?)</select', webpage,
+                'formats select', default=None)
+            if formats_select:
+                for mobj in re.finditer(
+                        r'<option\b[^>]+\bvalue=(["\'])(?P<url>(?:(?!\1).)+)\1[^>]*>\s*(?P<format>[^<]+?)\s*<',
+                        formats_select):
+                    format_url = mobj.group('url')
+                    if format_url in urls:
+                        continue
+                    urls.add(format_url)
+                    format_id = mobj.group('format')
+                    quality_id = SITE_QUALITIES.get(format_id, format_id)
+                    formats.append({
+                        'url': format_url,
+                        'format_id': quality_id,
+                        'quality': quality(quality_id, format_url),
+                        'vcodec': 'none' if quality_id == 'mp3' else None,
+                    })
+
+            API_QUALITIES = {
+                'VideoMP4Low': 'mp4-low',
+                'VideoWMV': 'wmv-mid',
+                'VideoMP4Medium': 'mp4-mid',
+                'VideoMP4High': 'mp4-high',
+                'VideoWMVHQ': 'wmv-hq',
+            }
+
+            for format_id, q in API_QUALITIES.items():
+                q_url = content_data.get(format_id)
+                if not q_url or q_url in urls:
+                    continue
+                urls.add(q_url)
+                formats.append({
+                    'url': q_url,
+                    'format_id': q,
+                    'quality': quality(q, q_url),
+                })
+
+            self._sort_formats(formats)
+
+            slides = content_data.get('Slides')
+            zip_file = content_data.get('ZipFile')
+
+            if not formats and not slides and not zip_file:
+                raise ExtractorError(
+                    'None of recording, slides or zip are available for %s' % content_path)
+
+            subtitles = {}
+            for caption in content_data.get('Captions', []):
+                caption_url = caption.get('Url')
+                if not caption_url:
+                    continue
+                subtitles.setdefault(caption.get('Language', 'en'), []).append({
+                    'url': caption_url,
+                    'ext': 'vtt',
+                })
+
+            common = {
+                'id': content_id,
+                'title': title,
+                'description': clean_html(content_data.get('Description') or content_data.get('Body')),
+                'thumbnail': content_data.get('Thumbnail') or content_data.get('VideoPlayerPreviewImage'),
+                'duration': int_or_none(content_data.get('MediaLengthInSeconds')),
+                'timestamp': parse_iso8601(content_data.get('PublishedDate')),
+                'avg_rating': int_or_none(content_data.get('Rating')),
+                'rating_count': int_or_none(content_data.get('RatingCount')),
+                'view_count': int_or_none(content_data.get('Views')),
+                'comment_count': int_or_none(content_data.get('CommentCount')),
+                'subtitles': subtitles,
+            }
+            if is_session:
+                speakers = []
+                for s in content_data.get('Speakers', []):
+                    speaker_name = s.get('FullName')
+                    if not speaker_name:
+                        continue
+                    speakers.append(speaker_name)
+
+                common.update({
+                    'session_code': content_data.get('Code'),
+                    'session_room': content_data.get('Room'),
+                    'session_speakers': speakers,
+                })
+            else:
+                authors = []
+                for a in content_data.get('Authors', []):
+                    author_name = a.get('DisplayName')
+                    if not author_name:
+                        continue
+                    authors.append(author_name)
+                common['authors'] = authors
+
+            contents = []
+
+            if slides:
+                d = common.copy()
+                d.update({'title': title + '-Slides', 'url': slides})
+                contents.append(d)
+
+            if zip_file:
+                d = common.copy()
+                d.update({'title': title + '-Zip', 'url': zip_file})
+                contents.append(d)
+
+            if formats:
+                d = common.copy()
+                d.update({'title': title, 'formats': formats})
+                contents.append(d)
+            return self.playlist_result(contents)
+        else:
             return self._extract_list(content_path)

youtube_dl/extractor/cloudy.py

@@ -1,97 +1,56 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
-from ..compat import (
-    compat_parse_qs,
-    compat_HTTPError,
-)
 from ..utils import (
-    ExtractorError,
-    HEADRequest,
-    remove_end,
+    str_to_int,
+    unified_strdate,
 )
 
 
 class CloudyIE(InfoExtractor):
     _IE_DESC = 'cloudy.ec'
-    _VALID_URL = r'''(?x)
-        https?://(?:www\.)?cloudy\.ec/
-        (?:v/|embed\.php\?id=)
-        (?P<id>[A-Za-z0-9]+)
-        '''
-    _EMBED_URL = 'http://www.cloudy.ec/embed.php?id=%s'
-    _API_URL = 'http://www.cloudy.ec/api/player.api.php'
-    _MAX_TRIES = 2
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?cloudy\.ec/(?:v/|embed\.php\?.*?\bid=)(?P<id>[A-Za-z0-9]+)'
+    _TESTS = [{
         'url': 'https://www.cloudy.ec/v/af511e2527aac',
-        'md5': '5cb253ace826a42f35b4740539bedf07',
+        'md5': '29832b05028ead1b58be86bf319397ca',
         'info_dict': {
             'id': 'af511e2527aac',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'Funny Cats and Animals Compilation june 2013',
+            'upload_date': '20130913',
+            'view_count': int,
         }
-    }
-
-    def _extract_video(self, video_id, file_key, error_url=None, try_num=0):
-
-        if try_num > self._MAX_TRIES - 1:
-            raise ExtractorError('Unable to extract video URL', expected=True)
-
-        form = {
-            'file': video_id,
-            'key': file_key,
-        }
-
-        if error_url:
-            form.update({
-                'numOfErrors': try_num,
-                'errorCode': '404',
-                'errorUrl': error_url,
-            })
-
-        player_data = self._download_webpage(
-            self._API_URL, video_id, 'Downloading player data', query=form)
-        data = compat_parse_qs(player_data)
-
-        try_num += 1
-
-        if 'error' in data:
-            raise ExtractorError(
-                '%s error: %s' % (self.IE_NAME, ' '.join(data['error_msg'])),
-                expected=True)
-
-        title = data.get('title', [None])[0]
-        if title:
-            title = remove_end(title, '&asdasdas').strip()
-
-        video_url = data.get('url', [None])[0]
-
-        if video_url:
-            try:
-                self._request_webpage(HEADRequest(video_url), video_id, 'Checking video URL')
-            except ExtractorError as e:
-                if isinstance(e.cause, compat_HTTPError) and e.cause.code in [404, 410]:
-                    self.report_warning('Invalid video URL, requesting another', video_id)
-                    return self._extract_video(video_id, file_key, video_url, try_num)
-
-        return {
-            'id': video_id,
-            'url': video_url,
-            'title': title,
-        }
+    }, {
+        'url': 'http://www.cloudy.ec/embed.php?autoplay=1&id=af511e2527aac',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
+        video_id = self._match_id(url)
 
-        url = self._EMBED_URL % video_id
-        webpage = self._download_webpage(url, video_id)
+        webpage = self._download_webpage(
+            'http://www.cloudy.ec/embed.php?id=%s' % video_id, video_id)
 
-        file_key = self._search_regex(
-            [r'key\s*:\s*"([^"]+)"', r'filekey\s*=\s*"([^"]+)"'],
-            webpage, 'file_key')
+        info = self._parse_html5_media_entries(url, webpage, video_id)[0]
 
-        return self._extract_video(video_id, file_key)
+        webpage = self._download_webpage(
+            'https://www.cloudy.ec/v/%s' % video_id, video_id, fatal=False)
+
+        if webpage:
+            info.update({
+                'title': self._search_regex(
+                    r'<h\d[^>]*>([^<]+)<', webpage, 'title'),
+                'upload_date': unified_strdate(self._search_regex(
+                    r'>Published at (\d{4}-\d{1,2}-\d{1,2})', webpage,
+                    'upload date', fatal=False)),
+                'view_count': str_to_int(self._search_regex(
+                    r'([\d,.]+) views<', webpage, 'view count', fatal=False)),
+            })
+
+        if not info.get('title'):
+            info['title'] = video_id
+
+        info['id'] = video_id
+
+        return info

youtube_dl/extractor/common.py

@@ -2169,18 +2169,24 @@ class InfoExtractor(object):
                     })
         return formats
 
-    @staticmethod
-    def _find_jwplayer_data(webpage):
+    def _find_jwplayer_data(self, webpage, video_id=None, transform_source=js_to_json):
         mobj = re.search(
             r'jwplayer\((?P<quote>[\'"])[^\'" ]+(?P=quote)\)\.setup\s*\((?P<options>[^)]+)\)',
             webpage)
         if mobj:
-            return mobj.group('options')
+            try:
+                jwplayer_data = self._parse_json(mobj.group('options'),
+                                                 video_id=video_id,
+                                                 transform_source=transform_source)
+            except ExtractorError:
+                pass
+            else:
+                if isinstance(jwplayer_data, dict):
+                    return jwplayer_data
 
     def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs):
-        jwplayer_data = self._parse_json(
-            self._find_jwplayer_data(webpage), video_id,
-            transform_source=js_to_json)
+        jwplayer_data = self._find_jwplayer_data(
+            webpage, video_id, transform_source=js_to_json)
         return self._parse_jwplayer_data(
             jwplayer_data, video_id, *args, **kwargs)
 

youtube_dl/extractor/condenast.py

@@ -9,13 +9,14 @@ from ..compat import (
     compat_urlparse,
 )
 from ..utils import (
-    orderedSet,
-    remove_end,
-    extract_attributes,
-    mimetype2ext,
     determine_ext,
+    extract_attributes,
     int_or_none,
+    js_to_json,
+    mimetype2ext,
+    orderedSet,
     parse_iso8601,
+    remove_end,
 )
 
 
@@ -66,6 +67,16 @@ class CondeNastIE(InfoExtractor):
             'upload_date': '20130314',
             'timestamp': 1363219200,
         }
+    }, {
+        'url': 'http://video.gq.com/watch/the-closer-with-keith-olbermann-the-only-true-surprise-trump-s-an-idiot?c=series',
+        'info_dict': {
+            'id': '58d1865bfd2e6126e2000015',
+            'ext': 'mp4',
+            'title': 'The Only True Surprise? Trump’s an Idiot',
+            'uploader': 'gq',
+            'upload_date': '20170321',
+            'timestamp': 1490126427,
+        },
     }, {
         # JS embed
         'url': 'http://player.cnevids.com/embedjs/55f9cf8b61646d1acf00000c/5511d76261646d5566020000.js',
@@ -114,26 +125,33 @@ class CondeNastIE(InfoExtractor):
             })
         video_id = query['videoId']
         video_info = None
-        info_page = self._download_webpage(
+        info_page = self._download_json(
             'http://player.cnevids.com/player/video.js',
-            video_id, 'Downloading video info', query=query, fatal=False)
+            video_id, 'Downloading video info', fatal=False, query=query)
         if info_page:
-            video_info = self._parse_json(self._search_regex(
-                r'loadCallback\(({.+})\)', info_page, 'video info'), video_id)['video']
-        else:
+            video_info = info_page.get('video')
+        if not video_info:
             info_page = self._download_webpage(
                 'http://player.cnevids.com/player/loader.js',
                 video_id, 'Downloading loader info', query=query)
-            video_info = self._parse_json(self._search_regex(
-                r'var\s+video\s*=\s*({.+?});', info_page, 'video info'), video_id)
+            video_info = self._parse_json(
+                self._search_regex(
+                    r'(?s)var\s+config\s*=\s*({.+?});', info_page, 'config'),
+                video_id, transform_source=js_to_json)['video']
+
         title = video_info['title']
 
         formats = []
-        for fdata in video_info.get('sources', [{}])[0]:
+        for fdata in video_info['sources']:
             src = fdata.get('src')
             if not src:
                 continue
             ext = mimetype2ext(fdata.get('type')) or determine_ext(src)
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    src, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id='hls', fatal=False))
+                continue
             quality = fdata.get('quality')
             formats.append({
                 'format_id': ext + ('-%s' % quality if quality else ''),
@@ -169,7 +187,6 @@ class CondeNastIE(InfoExtractor):
                 path=remove_end(parsed_url.path, '.js').replace('/embedjs/', '/embed/')))
             url_type = 'embed'
 
-        self.to_screen('Extracting from %s with the Condé Nast extractor' % self._SITES[site])
         webpage = self._download_webpage(url, item_id)
 
         if url_type == 'series':

youtube_dl/extractor/crunchyroll.py

@@ -390,7 +390,9 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
         else:
             webpage_url = 'http://www.' + mobj.group('url')
 
-        webpage = self._download_webpage(self._add_skip_wall(webpage_url), video_id, 'Downloading webpage')
+        webpage = self._download_webpage(
+            self._add_skip_wall(webpage_url), video_id,
+            headers=self.geo_verification_headers())
         note_m = self._html_search_regex(
             r'<div class="showmedia-trailer-notice">(.+?)</div>',
             webpage, 'trailer-notice', default='')
@@ -565,7 +567,9 @@ class CrunchyrollShowPlaylistIE(CrunchyrollBaseIE):
     def _real_extract(self, url):
         show_id = self._match_id(url)
 
-        webpage = self._download_webpage(self._add_skip_wall(url), show_id)
+        webpage = self._download_webpage(
+            self._add_skip_wall(url), show_id,
+            headers=self.geo_verification_headers())
         title = self._html_search_regex(
             r'(?s)<h1[^>]*>\s*<span itemprop="name">(.*?)</span>',
             webpage, 'title')

youtube_dl/extractor/cwtv.py

@@ -82,6 +82,11 @@ class CWTVIE(InfoExtractor):
                             'url': quality_url,
                             'tbr': tbr,
                         })
+        video_metadata = video_data['assetFields']
+        ism_url = video_metadata.get('smoothStreamingUrl')
+        if ism_url:
+            formats.extend(self._extract_ism_formats(
+                ism_url, video_id, ism_id='mss', fatal=False))
         self._sort_formats(formats)
 
         thumbnails = [{
@@ -90,8 +95,6 @@ class CWTVIE(InfoExtractor):
             'height': image.get('height'),
         } for image_id, image in video_data['images'].items() if image.get('uri')] if video_data.get('images') else None
 
-        video_metadata = video_data['assetFields']
-
         subtitles = {
             'en': [{
                 'url': video_metadata['UnicornCcUrl'],

youtube_dl/extractor/extractors.py

@@ -19,6 +19,7 @@ from .acast import (
     ACastChannelIE,
 )
 from .addanime import AddAnimeIE
+from .adn import ADNIE
 from .adobetv import (
     AdobeTVIE,
     AdobeTVShowIE,
@@ -71,6 +72,7 @@ from .arte import (
 )
 from .atresplayer import AtresPlayerIE
 from .atttechchannel import ATTTechChannelIE
+from .atvat import ATVAtIE
 from .audimedia import AudiMediaIE
 from .audioboom import AudioBoomIE
 from .audiomack import AudiomackIE, AudiomackAlbumIE
@@ -727,6 +729,10 @@ from .orf import (
     ORFFM4IE,
     ORFIPTVIE,
 )
+from .packtpub import (
+    PacktPubIE,
+    PacktPubCourseIE,
+)
 from .pandatv import PandaTVIE
 from .pandoratv import PandoraTVIE
 from .parliamentliveuk import ParliamentLiveUKIE
@@ -796,7 +802,7 @@ from .radiojavan import RadioJavanIE
 from .radiobremen import RadioBremenIE
 from .radiofrance import RadioFranceIE
 from .rai import (
-    RaiTVIE,
+    RaiPlayIE,
     RaiIE,
 )
 from .rbmaradio import RBMARadioIE
@@ -1176,6 +1182,10 @@ from .voxmedia import VoxMediaIE
 from .vporn import VpornIE
 from .vrt import VRTIE
 from .vrak import VrakIE
+from .vrv import (
+    VRVIE,
+    VRVSeriesIE,
+)
 from .medialaan import MedialaanIE
 from .vube import VubeIE
 from .vuclip import VuClipIE

youtube_dl/extractor/eyedotv.py

@@ -54,7 +54,7 @@ class EyedoTVIE(InfoExtractor):
             'id': video_id,
             'title': title,
             'formats': self._extract_m3u8_formats(
-                m3u8_url, video_id, 'mp4', 'm3u8' if is_live else 'm3u8_native'),
+                m3u8_url, video_id, 'mp4', 'm3u8_native'),
             'description': xpath_text(video_data, _add_ns('Description')),
             'duration': parse_duration(xpath_text(video_data, _add_ns('Duration'))),
             'uploader': xpath_text(video_data, _add_ns('Createur')),

youtube_dl/extractor/fox.py

@@ -47,9 +47,12 @@ class FOXIE(AdobePassIE):
             resource = self._get_mvpd_resource('fbc-fox', None, ap_p['videoGUID'], rating)
             query['auth'] = self._extract_mvpd_auth(url, video_id, 'fbc-fox', resource)
 
-        return {
+        info = self._search_json_ld(webpage, video_id, fatal=False)
+        info.update({
             '_type': 'url_transparent',
             'ie_key': 'ThePlatform',
             'url': smuggle_url(update_url_query(release_url, query), {'force_smil_url': True}),
             'id': video_id,
-        }
+        })
+
+        return info

youtube_dl/extractor/franceculture.py

@@ -4,7 +4,8 @@ from __future__ import unicode_literals
 from .common import InfoExtractor
 from ..utils import (
     determine_ext,
-    unified_strdate,
+    extract_attributes,
+    int_or_none,
 )
 
 
@@ -19,6 +20,7 @@ class FranceCultureIE(InfoExtractor):
             'title': 'Rendez-vous au pays des geeks',
             'thumbnail': r're:^https?://.*\.jpg$',
             'upload_date': '20140301',
+            'timestamp': 1393642916,
             'vcodec': 'none',
         }
     }
@@ -28,30 +30,34 @@ class FranceCultureIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
-        video_url = self._search_regex(
-            r'(?s)<div[^>]+class="[^"]*?title-zone-diffusion[^"]*?"[^>]*>.*?<button[^>]+data-asset-source="([^"]+)"',
-            webpage, 'video path')
+        video_data = extract_attributes(self._search_regex(
+            r'(?s)<div[^>]+class="[^"]*?(?:title-zone-diffusion|heading-zone-(?:wrapper|player-button))[^"]*?"[^>]*>.*?(<button[^>]+data-asset-source="[^"]+"[^>]+>)',
+            webpage, 'video data'))
 
-        title = self._og_search_title(webpage)
+        video_url = video_data['data-asset-source']
+        title = video_data.get('data-asset-title') or self._og_search_title(webpage)
 
-        upload_date = unified_strdate(self._search_regex(
-            '(?s)<div[^>]+class="date"[^>]*>.*?<span[^>]+class="inner"[^>]*>([^<]+)<',
-            webpage, 'upload date', fatal=False))
+        description = self._html_search_regex(
+            r'(?s)<div[^>]+class="intro"[^>]*>.*?<h2>(.+?)</h2>',
+            webpage, 'description', default=None)
         thumbnail = self._search_regex(
-            r'(?s)<figure[^>]+itemtype="https://schema.org/ImageObject"[^>]*>.*?<img[^>]+data-dejavu-src="([^"]+)"',
+            r'(?s)<figure[^>]+itemtype="https://schema.org/ImageObject"[^>]*>.*?<img[^>]+(?:data-dejavu-)?src="([^"]+)"',
             webpage, 'thumbnail', fatal=False)
         uploader = self._html_search_regex(
-            r'(?s)<div id="emission".*?<span class="author">(.*?)</span>',
+            r'(?s)<span class="author">(.*?)</span>',
             webpage, 'uploader', default=None)
-        vcodec = 'none' if determine_ext(video_url.lower()) == 'mp3' else None
+        ext = determine_ext(video_url.lower())
 
         return {
             'id': display_id,
             'display_id': display_id,
             'url': video_url,
             'title': title,
+            'description': description,
             'thumbnail': thumbnail,
-            'vcodec': vcodec,
+            'ext': ext,
+            'vcodec': 'none' if ext == 'mp3' else None,
             'uploader': uploader,
-            'upload_date': upload_date,
+            'timestamp': int_or_none(video_data.get('data-asset-created-date')),
+            'duration': int_or_none(video_data.get('data-duration')),
         }

youtube_dl/extractor/freshlive.py

@@ -56,9 +56,8 @@ class FreshLiveIE(InfoExtractor):
         is_live = info.get('liveStreamUrl') is not None
 
         formats = self._extract_m3u8_formats(
-            stream_url, video_id, ext='mp4',
-            entry_protocol='m3u8' if is_live else 'm3u8_native',
-            m3u8_id='hls')
+            stream_url, video_id, 'mp4',
+            'm3u8_native', m3u8_id='hls')
 
         if is_live:
             title = self._live_title(title)

youtube_dl/extractor/funimation.py

@@ -7,9 +7,9 @@ from ..compat import (
     compat_urllib_parse_unquote_plus,
 )
 from ..utils import (
-    clean_html,
     determine_ext,
     int_or_none,
+    js_to_json,
     sanitized_Request,
     ExtractorError,
     urlencode_postdata
@@ -17,34 +17,26 @@ from ..utils import (
 
 
 class FunimationIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?funimation\.com/shows/[^/]+/videos/(?:official|promotional)/(?P<id>[^/?#&]+)'
+    _VALID_URL = r'https?://(?:www\.)?funimation(?:\.com|now\.uk)/shows/[^/]+/(?P<id>[^/?#&]+)'
 
     _NETRC_MACHINE = 'funimation'
 
     _TESTS = [{
-        'url': 'http://www.funimation.com/shows/air/videos/official/breeze',
+        'url': 'https://www.funimation.com/shows/hacksign/role-play/',
         'info_dict': {
-            'id': '658',
-            'display_id': 'breeze',
-            'ext': 'mp4',
-            'title': 'Air - 1 - Breeze',
-            'description': 'md5:1769f43cd5fc130ace8fd87232207892',
-            'thumbnail': r're:https?://.*\.jpg',
-        },
-        'skip': 'Access without user interaction is forbidden by CloudFlare, and video removed',
-    }, {
-        'url': 'http://www.funimation.com/shows/hacksign/videos/official/role-play',
-        'info_dict': {
-            'id': '31128',
+            'id': '91144',
             'display_id': 'role-play',
             'ext': 'mp4',
-            'title': '.hack//SIGN - 1 - Role Play',
+            'title': '.hack//SIGN - Role Play',
             'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
             'thumbnail': r're:https?://.*\.jpg',
         },
-        'skip': 'Access without user interaction is forbidden by CloudFlare',
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
     }, {
-        'url': 'http://www.funimation.com/shows/attack-on-titan-junior-high/videos/promotional/broadcast-dub-preview',
+        'url': 'https://www.funimation.com/shows/attack-on-titan-junior-high/broadcast-dub-preview/',
         'info_dict': {
             'id': '9635',
             'display_id': 'broadcast-dub-preview',
@@ -54,25 +46,13 @@ class FunimationIE(InfoExtractor):
             'thumbnail': r're:https?://.*\.(?:jpg|png)',
         },
         'skip': 'Access without user interaction is forbidden by CloudFlare',
+    }, {
+        'url': 'https://www.funimationnow.uk/shows/puzzle-dragons-x/drop-impact/simulcast/',
+        'only_matching': True,
     }]
 
     _LOGIN_URL = 'http://www.funimation.com/login'
 
-    def _download_webpage(self, *args, **kwargs):
-        try:
-            return super(FunimationIE, self)._download_webpage(*args, **kwargs)
-        except ExtractorError as ee:
-            if isinstance(ee.cause, compat_HTTPError) and ee.cause.code == 403:
-                response = ee.cause.read()
-                if b'>Please complete the security check to access<' in response:
-                    raise ExtractorError(
-                        'Access to funimation.com is blocked by CloudFlare. '
-                        'Please browse to http://www.funimation.com/, solve '
-                        'the reCAPTCHA, export browser cookies to a text file,'
-                        ' and then try again with --cookies YOUR_COOKIE_FILE.',
-                        expected=True)
-            raise
-
     def _extract_cloudflare_session_ua(self, url):
         ci_session_cookie = self._get_cookies(url).get('ci_session')
         if ci_session_cookie:
@@ -114,119 +94,74 @@ class FunimationIE(InfoExtractor):
 
     def _real_extract(self, url):
         display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        def _search_kane(name):
+            return self._search_regex(
+                r"KANE_customdimensions\.%s\s*=\s*'([^']+)';" % name,
+                webpage, name, default=None)
+
+        title_data = self._parse_json(self._search_regex(
+            r'TITLE_DATA\s*=\s*({[^}]+})',
+            webpage, 'title data', default=''),
+            display_id, js_to_json, fatal=False) or {}
+
+        video_id = title_data.get('id') or self._search_regex([
+            r"KANE_customdimensions.videoID\s*=\s*'(\d+)';",
+            r'<iframe[^>]+src="/player/(\d+)"',
+        ], webpage, 'video_id', default=None)
+        if not video_id:
+            player_url = self._html_search_meta([
+                'al:web:url',
+                'og:video:url',
+                'og:video:secure_url',
+            ], webpage, fatal=True)
+            video_id = self._search_regex(r'/player/(\d+)', player_url, 'video id')
+
+        title = episode = title_data.get('title') or _search_kane('videoTitle') or self._og_search_title(webpage)
+        series = _search_kane('showName')
+        if series:
+            title = '%s - %s' % (series, title)
+        description = self._html_search_meta(['description', 'og:description'], webpage, fatal=True)
+
+        try:
+            sources = self._download_json(
+                'https://prod-api-funimationnow.dadcdigital.com/api/source/catalog/video/%s/signed/' % video_id,
+                video_id)['items']
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                error = self._parse_json(e.cause.read(), video_id)['errors'][0]
+                raise ExtractorError('%s said: %s' % (
+                    self.IE_NAME, error.get('detail') or error.get('title')), expected=True)
+            raise
 
-        errors = []
         formats = []
-
-        ERRORS_MAP = {
-            'ERROR_MATURE_CONTENT_LOGGED_IN': 'matureContentLoggedIn',
-            'ERROR_MATURE_CONTENT_LOGGED_OUT': 'matureContentLoggedOut',
-            'ERROR_SUBSCRIPTION_LOGGED_OUT': 'subscriptionLoggedOut',
-            'ERROR_VIDEO_EXPIRED': 'videoExpired',
-            'ERROR_TERRITORY_UNAVAILABLE': 'territoryUnavailable',
-            'SVODBASIC_SUBSCRIPTION_IN_PLAYER': 'basicSubscription',
-            'SVODNON_SUBSCRIPTION_IN_PLAYER': 'nonSubscription',
-            'ERROR_PLAYER_NOT_RESPONDING': 'playerNotResponding',
-            'ERROR_UNABLE_TO_CONNECT_TO_CDN': 'unableToConnectToCDN',
-            'ERROR_STREAM_NOT_FOUND': 'streamNotFound',
-        }
-
-        USER_AGENTS = (
-            # PC UA is served with m3u8 that provides some bonus lower quality formats
-            ('pc', 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0'),
-            # Mobile UA allows to extract direct links and also does not fail when
-            # PC UA fails with hulu error (e.g.
-            # http://www.funimation.com/shows/hacksign/videos/official/role-play)
-            ('mobile', 'Mozilla/5.0 (Linux; Android 4.4.2; Nexus 4 Build/KOT49H) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/34.0.1847.114 Mobile Safari/537.36'),
-        )
-
-        user_agent = self._extract_cloudflare_session_ua(url)
-        if user_agent:
-            USER_AGENTS = ((None, user_agent),)
-
-        for kind, user_agent in USER_AGENTS:
-            request = sanitized_Request(url)
-            request.add_header('User-Agent', user_agent)
-            webpage = self._download_webpage(
-                request, display_id,
-                'Downloading %s webpage' % kind if kind else 'Downloading webpage')
-
-            playlist = self._parse_json(
-                self._search_regex(
-                    r'var\s+playersData\s*=\s*(\[.+?\]);\n',
-                    webpage, 'players data'),
-                display_id)[0]['playlist']
-
-            items = next(item['items'] for item in playlist if item.get('items'))
-            item = next(item for item in items if item.get('itemAK') == display_id)
-
-            error_messages = {}
-            video_error_messages = self._search_regex(
-                r'var\s+videoErrorMessages\s*=\s*({.+?});\n',
-                webpage, 'error messages', default=None)
-            if video_error_messages:
-                error_messages_json = self._parse_json(video_error_messages, display_id, fatal=False)
-                if error_messages_json:
-                    for _, error in error_messages_json.items():
-                        type_ = error.get('type')
-                        description = error.get('description')
-                        content = error.get('content')
-                        if type_ == 'text' and description and content:
-                            error_message = ERRORS_MAP.get(description)
-                            if error_message:
-                                error_messages[error_message] = content
-
-            for video in item.get('videoSet', []):
-                auth_token = video.get('authToken')
-                if not auth_token:
-                    continue
-                funimation_id = video.get('FUNImationID') or video.get('videoId')
-                preference = 1 if video.get('languageMode') == 'dub' else 0
-                if not auth_token.startswith('?'):
-                    auth_token = '?%s' % auth_token
-                for quality, height in (('sd', 480), ('hd', 720), ('hd1080', 1080)):
-                    format_url = video.get('%sUrl' % quality)
-                    if not format_url:
-                        continue
-                    if not format_url.startswith(('http', '//')):
-                        errors.append(format_url)
-                        continue
-                    if determine_ext(format_url) == 'm3u8':
-                        formats.extend(self._extract_m3u8_formats(
-                            format_url + auth_token, display_id, 'mp4', entry_protocol='m3u8_native',
-                            preference=preference, m3u8_id='%s-hls' % funimation_id, fatal=False))
-                    else:
-                        tbr = int_or_none(self._search_regex(
-                            r'-(\d+)[Kk]', format_url, 'tbr', default=None))
-                        formats.append({
-                            'url': format_url + auth_token,
-                            'format_id': '%s-http-%dp' % (funimation_id, height),
-                            'height': height,
-                            'tbr': tbr,
-                            'preference': preference,
-                        })
-
-        if not formats and errors:
-            raise ExtractorError(
-                '%s returned error: %s'
-                % (self.IE_NAME, clean_html(error_messages.get(errors[0], errors[0]))),
-                expected=True)
-
+        for source in sources:
+            source_url = source.get('src')
+            if not source_url:
+                continue
+            source_type = source.get('videoType') or determine_ext(source_url)
+            if source_type == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    source_url, video_id, 'mp4',
+                    m3u8_id='hls', fatal=False))
+            else:
+                formats.append({
+                    'format_id': source_type,
+                    'url': source_url,
+                })
         self._sort_formats(formats)
 
-        title = item['title']
-        artist = item.get('artist')
-        if artist:
-            title = '%s - %s' % (artist, title)
-        description = self._og_search_description(webpage) or item.get('description')
-        thumbnail = self._og_search_thumbnail(webpage) or item.get('posterUrl')
-        video_id = item.get('itemId') or display_id
-
         return {
             'id': video_id,
             'display_id': display_id,
             'title': title,
             'description': description,
-            'thumbnail': thumbnail,
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'series': series,
+            'season_number': int_or_none(title_data.get('seasonNum') or _search_kane('season')),
+            'episode_number': int_or_none(title_data.get('episodeNum')),
+            'episode': episode,
+            'season_id': title_data.get('seriesId'),
             'formats': formats,
         }

youtube_dl/extractor/generic.py

@@ -902,12 +902,13 @@ class GenericIE(InfoExtractor):
         },
         # LazyYT
         {
-            'url': 'http://discourse.ubuntu.com/t/unity-8-desktop-mode-windows-on-mir/1986',
+            'url': 'https://skiplagged.com/',
             'info_dict': {
-                'id': '1986',
-                'title': 'Unity 8 desktop-mode windows on Mir! - Ubuntu Discourse',
+                'id': 'skiplagged',
+                'title': 'Skiplagged: The smart way to find cheap flights',
             },
-            'playlist_mincount': 2,
+            'playlist_mincount': 1,
+            'add_ie': ['Youtube'],
         },
         # Cinchcast embed
         {
@@ -990,6 +991,20 @@ class GenericIE(InfoExtractor):
                 'thumbnail': r're:^https?://.*\.jpg$',
             },
         },
+        {
+            # JWPlayer config passed as variable
+            'url': 'http://www.txxx.com/videos/3326530/ariele/',
+            'info_dict': {
+                'id': '3326530_hq',
+                'ext': 'mp4',
+                'title': 'ARIELE | Tube Cup',
+                'uploader': 'www.txxx.com',
+                'age_limit': 18,
+            },
+            'params': {
+                'skip_download': True,
+            }
+        },
         # rtl.nl embed
         {
             'url': 'http://www.rtlnieuws.nl/nieuws/buitenland/aanslagen-kopenhagen',
@@ -2549,18 +2564,14 @@ class GenericIE(InfoExtractor):
                 self._sort_formats(entry['formats'])
             return self.playlist_result(entries)
 
-        jwplayer_data_str = self._find_jwplayer_data(webpage)
-        if jwplayer_data_str:
-            try:
-                jwplayer_data = self._parse_json(
-                    jwplayer_data_str, video_id, transform_source=js_to_json)
-                info = self._parse_jwplayer_data(
-                    jwplayer_data, video_id, require_title=False)
-                if not info.get('title'):
-                    info['title'] = video_title
-                return info
-            except ExtractorError:
-                pass
+        jwplayer_data = self._find_jwplayer_data(
+            webpage, video_id, transform_source=js_to_json)
+        if jwplayer_data:
+            info = self._parse_jwplayer_data(
+                jwplayer_data, video_id, require_title=False, base_url=url)
+            if not info.get('title'):
+                info['title'] = video_title
+            return info
 
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):
@@ -2635,11 +2646,14 @@ class GenericIE(InfoExtractor):
                     found = re.search(REDIRECT_REGEX, refresh_header)
             if found:
                 new_url = compat_urlparse.urljoin(url, unescapeHTML(found.group(1)))
-                self.report_following_redirect(new_url)
-                return {
-                    '_type': 'url',
-                    'url': new_url,
-                }
+                if new_url != url:
+                    self.report_following_redirect(new_url)
+                    return {
+                        '_type': 'url',
+                        'url': new_url,
+                    }
+                else:
+                    found = None
 
         if not found:
             # twitter:player is a https URL to iframe player that may or may not

youtube_dl/extractor/hbo.py

@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     xpath_text,
     xpath_element,
@@ -14,14 +15,26 @@ from ..utils import (
 
 class HBOBaseIE(InfoExtractor):
     _FORMATS_INFO = {
+        'pro7': {
+            'width': 1280,
+            'height': 720,
+        },
         '1920': {
             'width': 1280,
             'height': 720,
         },
+        'pro6': {
+            'width': 768,
+            'height': 432,
+        },
         '640': {
             'width': 768,
             'height': 432,
         },
+        'pro5': {
+            'width': 640,
+            'height': 360,
+        },
         'highwifi': {
             'width': 640,
             'height': 360,
@@ -78,6 +91,17 @@ class HBOBaseIE(InfoExtractor):
                     formats.extend(self._extract_m3u8_formats(
                         video_url.replace('.tar', '/base_index_w8.m3u8'),
                         video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+                elif source.tag == 'hls':
+                    # #EXT-X-BYTERANGE is not supported by native hls downloader
+                    # and ffmpeg (#10955)
+                    # formats.extend(self._extract_m3u8_formats(
+                    #     video_url.replace('.tar', '/base_index.m3u8'),
+                    #     video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+                    continue
+                elif source.tag == 'dash':
+                    formats.extend(self._extract_mpd_formats(
+                        video_url.replace('.tar', '/manifest.mpd'),
+                        video_id, mpd_id='dash', fatal=False))
                 else:
                     format_info = self._FORMATS_INFO.get(source.tag, {})
                     formats.append({
@@ -112,10 +136,11 @@ class HBOBaseIE(InfoExtractor):
 
 
 class HBOIE(HBOBaseIE):
+    IE_NAME = 'hbo'
     _VALID_URL = r'https?://(?:www\.)?hbo\.com/video/video\.html\?.*vid=(?P<id>[0-9]+)'
     _TEST = {
         'url': 'http://www.hbo.com/video/video.html?autoplay=true&g=u&vid=1437839',
-        'md5': '1c33253f0c7782142c993c0ba62a8753',
+        'md5': '2c6a6bc1222c7e91cb3334dad1746e5a',
         'info_dict': {
             'id': '1437839',
             'ext': 'mp4',
@@ -131,11 +156,12 @@ class HBOIE(HBOBaseIE):
 
 
 class HBOEpisodeIE(HBOBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?!video)([^/]+/)+video/(?P<id>[0-9a-z-]+)\.html'
+    IE_NAME = 'hbo:episode'
+    _VALID_URL = r'https?://(?:www\.)?hbo\.com/(?P<path>(?!video)(?:(?:[^/]+/)+video|watch-free-episodes)/(?P<id>[0-9a-z-]+))(?:\.html)?'
 
     _TESTS = [{
         'url': 'http://www.hbo.com/girls/episodes/5/52-i-love-you-baby/video/ep-52-inside-the-episode.html?autoplay=true',
-        'md5': '689132b253cc0ab7434237fc3a293210',
+        'md5': '61ead79b9c0dfa8d3d4b07ef4ac556fb',
         'info_dict': {
             'id': '1439518',
             'display_id': 'ep-52-inside-the-episode',
@@ -147,16 +173,19 @@ class HBOEpisodeIE(HBOBaseIE):
     }, {
         'url': 'http://www.hbo.com/game-of-thrones/about/video/season-5-invitation-to-the-set.html?autoplay=true',
         'only_matching': True,
+    }, {
+        'url': 'http://www.hbo.com/watch-free-episodes/last-week-tonight-with-john-oliver',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
+        path, display_id = re.match(self._VALID_URL, url).groups()
 
-        webpage = self._download_webpage(url, display_id)
+        content = self._download_json(
+            'http://www.hbo.com/api/content/' + path, display_id)['content']
 
-        video_id = self._search_regex(
-            r'(?P<q1>[\'"])videoId(?P=q1)\s*:\s*(?P<q2>[\'"])(?P<video_id>\d+)(?P=q2)',
-            webpage, 'video ID', group='video_id')
+        video_id = compat_str((content.get('parsed', {}).get(
+            'common:FullBleedVideo', {}) or content['selectedEpisode'])['videoId'])
 
         info_dict = self._extract_from_id(video_id)
         info_dict['display_id'] = display_id

youtube_dl/extractor/limelight.py

@@ -62,13 +62,21 @@ class LimelightBaseIE(InfoExtractor):
                 fmt = {
                     'url': stream_url,
                     'abr': float_or_none(stream.get('audioBitRate')),
-                    'vbr': float_or_none(stream.get('videoBitRate')),
                     'fps': float_or_none(stream.get('videoFrameRate')),
-                    'width': int_or_none(stream.get('videoWidthInPixels')),
-                    'height': int_or_none(stream.get('videoHeightInPixels')),
                     'ext': ext,
                 }
-                rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp4:.+)$', stream_url)
+                width = int_or_none(stream.get('videoWidthInPixels'))
+                height = int_or_none(stream.get('videoHeightInPixels'))
+                vbr = float_or_none(stream.get('videoBitRate'))
+                if width or height or vbr:
+                    fmt.update({
+                        'width': width,
+                        'height': height,
+                        'vbr': vbr,
+                    })
+                else:
+                    fmt['vcodec'] = 'none'
+                rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', stream_url)
                 if rtmp:
                     format_id = 'rtmp'
                     if stream.get('videoBitRate'):

youtube_dl/extractor/livestream.py

@@ -119,7 +119,8 @@ class LivestreamIE(InfoExtractor):
         m3u8_url = video_data.get('m3u8_url')
         if m3u8_url:
             formats.extend(self._extract_m3u8_formats(
-                m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+                m3u8_url, video_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False))
 
         f4m_url = video_data.get('f4m_url')
         if f4m_url:
@@ -158,11 +159,11 @@ class LivestreamIE(InfoExtractor):
         if smil_url:
             formats.extend(self._extract_smil_formats(smil_url, broadcast_id))
 
-        entry_protocol = 'm3u8' if is_live else 'm3u8_native'
         m3u8_url = stream_info.get('m3u8_url')
         if m3u8_url:
             formats.extend(self._extract_m3u8_formats(
-                m3u8_url, broadcast_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=False))
+                m3u8_url, broadcast_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False))
 
         rtsp_url = stream_info.get('rtsp_url')
         if rtsp_url:
@@ -276,7 +277,7 @@ class LivestreamOriginalIE(InfoExtractor):
             'view_count': view_count,
         }
 
-    def _extract_video_formats(self, video_data, video_id, entry_protocol):
+    def _extract_video_formats(self, video_data, video_id):
         formats = []
 
         progressive_url = video_data.get('progressiveUrl')
@@ -289,7 +290,8 @@ class LivestreamOriginalIE(InfoExtractor):
         m3u8_url = video_data.get('httpUrl')
         if m3u8_url:
             formats.extend(self._extract_m3u8_formats(
-                m3u8_url, video_id, 'mp4', entry_protocol, m3u8_id='hls', fatal=False))
+                m3u8_url, video_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False))
 
         rtsp_url = video_data.get('rtspUrl')
         if rtsp_url:
@@ -340,11 +342,10 @@ class LivestreamOriginalIE(InfoExtractor):
                 }
             video_data = self._download_json(stream_url, content_id)
             is_live = video_data.get('isLive')
-            entry_protocol = 'm3u8' if is_live else 'm3u8_native'
             info.update({
                 'id': content_id,
                 'title': self._live_title(info['title']) if is_live else info['title'],
-                'formats': self._extract_video_formats(video_data, content_id, entry_protocol),
+                'formats': self._extract_video_formats(video_data, content_id),
                 'is_live': is_live,
             })
             return info

youtube_dl/extractor/ninecninemedia.py

@@ -34,12 +34,6 @@ class NineCNineMediaStackIE(NineCNineMediaBaseIE):
         formats.extend(self._extract_f4m_formats(
             stack_base_url + 'f4m', stack_id,
             f4m_id='hds', fatal=False))
-        mp4_url = self._download_webpage(stack_base_url + 'pd', stack_id, fatal=False)
-        if mp4_url:
-            formats.append({
-                'url': mp4_url,
-                'format_id': 'mp4',
-            })
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/openload.py

@@ -75,51 +75,38 @@ class OpenloadIE(InfoExtractor):
             '<span[^>]+id="[^"]+"[^>]*>([0-9A-Za-z]+)</span>',
             webpage, 'openload ID')
 
-        video_url_chars = []
-
-        first_char = ord(ol_id[0])
-        key = first_char - 55
-        maxKey = max(2, key)
-        key = min(maxKey, len(ol_id) - 38)
-        t = ol_id[key:key + 36]
-
-        hashMap = {}
-        v = ol_id.replace(t, '')
-        h = 0
-
-        while h < len(t):
-            f = t[h:h + 3]
-            i = int(f, 8)
-            hashMap[h / 3] = i
-            h += 3
-
-        h = 0
-        H = 0
-        while h < len(v):
-            B = ''
-            C = ''
-            if len(v) >= h + 2:
-                B = v[h:h + 2]
-            if len(v) >= h + 3:
-                C = v[h:h + 3]
-            i = int(B, 16)
-            h += 2
-            if H % 3 == 0:
-                i = int(C, 8)
-                h += 1
-            elif H % 2 == 0 and H != 0 and ord(v[H - 1]) < 60:
-                i = int(C, 10)
-                h += 1
-            index = H % 7
-
-            A = hashMap[index]
-            i ^= 213
-            i ^= A
-            video_url_chars.append(compat_chr(i))
-            H += 1
+        decoded = ''
+        a = ol_id[0:24]
+        b = []
+        for i in range(0, len(a), 8):
+            b.append(int(a[i:i + 8] or '0', 16))
+        ol_id = ol_id[24:]
+        j = 0
+        k = 0
+        while j < len(ol_id):
+            c = 128
+            d = 0
+            e = 0
+            f = 0
+            _more = True
+            while _more:
+                if j + 1 >= len(ol_id):
+                    c = 143
+                f = int(ol_id[j:j + 2] or '0', 16)
+                j += 2
+                d += (f & 127) << e
+                e += 7
+                _more = f >= c
+            g = d ^ b[k % 3]
+            for i in range(4):
+                char_dec = (g >> 8 * i) & (c + 127)
+                char = compat_chr(char_dec)
+                if char != '#':
+                    decoded += char
+            k += 1
 
         video_url = 'https://openload.co/stream/%s?mime=true'
-        video_url = video_url % (''.join(video_url_chars))
+        video_url = video_url % decoded
 
         title = self._og_search_title(webpage, default=None) or self._search_regex(
             r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,

youtube_dl/extractor/packtpub.py

@@ -0,0 +1,138 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_str
+from ..utils import (
+    clean_html,
+    ExtractorError,
+    remove_end,
+    strip_or_none,
+    unified_timestamp,
+    urljoin,
+)
+
+
+class PacktPubBaseIE(InfoExtractor):
+    _PACKT_BASE = 'https://www.packtpub.com'
+    _MAPT_REST = '%s/mapt-rest' % _PACKT_BASE
+
+
+class PacktPubIE(PacktPubBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?packtpub\.com/mapt/video/[^/]+/(?P<course_id>\d+)/(?P<chapter_id>\d+)/(?P<id>\d+)'
+
+    _TEST = {
+        'url': 'https://www.packtpub.com/mapt/video/web-development/9781787122215/20528/20530/Project+Intro',
+        'md5': '1e74bd6cfd45d7d07666f4684ef58f70',
+        'info_dict': {
+            'id': '20530',
+            'ext': 'mp4',
+            'title': 'Project Intro',
+            'thumbnail': r're:(?i)^https?://.*\.jpg',
+            'timestamp': 1490918400,
+            'upload_date': '20170331',
+        },
+    }
+
+    def _handle_error(self, response):
+        if response.get('status') != 'success':
+            raise ExtractorError(
+                '% said: %s' % (self.IE_NAME, response['message']),
+                expected=True)
+
+    def _download_json(self, *args, **kwargs):
+        response = super(PacktPubIE, self)._download_json(*args, **kwargs)
+        self._handle_error(response)
+        return response
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        course_id, chapter_id, video_id = mobj.group(
+            'course_id', 'chapter_id', 'id')
+
+        video = self._download_json(
+            '%s/users/me/products/%s/chapters/%s/sections/%s'
+            % (self._MAPT_REST, course_id, chapter_id, video_id), video_id,
+            'Downloading JSON video')['data']
+
+        content = video.get('content')
+        if not content:
+            raise ExtractorError('This video is locked', expected=True)
+
+        video_url = content['file']
+
+        metadata = self._download_json(
+            '%s/products/%s/chapters/%s/sections/%s/metadata'
+            % (self._MAPT_REST, course_id, chapter_id, video_id),
+            video_id)['data']
+
+        title = metadata['pageTitle']
+        course_title = metadata.get('title')
+        if course_title:
+            title = remove_end(title, ' - %s' % course_title)
+        timestamp = unified_timestamp(metadata.get('publicationDate'))
+        thumbnail = urljoin(self._PACKT_BASE, metadata.get('filepath'))
+
+        return {
+            'id': video_id,
+            'url': video_url,
+            'title': title,
+            'thumbnail': thumbnail,
+            'timestamp': timestamp,
+        }
+
+
+class PacktPubCourseIE(PacktPubBaseIE):
+    _VALID_URL = r'(?P<url>https?://(?:www\.)?packtpub\.com/mapt/video/[^/]+/(?P<id>\d+))'
+    _TEST = {
+        'url': 'https://www.packtpub.com/mapt/video/web-development/9781787122215',
+        'info_dict': {
+            'id': '9781787122215',
+            'title': 'Learn Nodejs by building 12 projects [Video]',
+        },
+        'playlist_count': 90,
+    }
+
+    @classmethod
+    def suitable(cls, url):
+        return False if PacktPubIE.suitable(url) else super(
+            PacktPubCourseIE, cls).suitable(url)
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        url, course_id = mobj.group('url', 'id')
+
+        course = self._download_json(
+            '%s/products/%s/metadata' % (self._MAPT_REST, course_id),
+            course_id)['data']
+
+        entries = []
+        for chapter_num, chapter in enumerate(course['tableOfContents'], 1):
+            if chapter.get('type') != 'chapter':
+                continue
+            children = chapter.get('children')
+            if not isinstance(children, list):
+                continue
+            chapter_info = {
+                'chapter': chapter.get('title'),
+                'chapter_number': chapter_num,
+                'chapter_id': chapter.get('id'),
+            }
+            for section in children:
+                if section.get('type') != 'section':
+                    continue
+                section_url = section.get('seoUrl')
+                if not isinstance(section_url, compat_str):
+                    continue
+                entry = {
+                    '_type': 'url_transparent',
+                    'url': urljoin(url + '/', section_url),
+                    'title': strip_or_none(section.get('title')),
+                    'description': clean_html(section.get('summary')),
+                    'ie_key': PacktPubIE.ie_key(),
+                }
+                entry.update(chapter_info)
+                entries.append(entry)
+
+        return self.playlist_result(entries, course_id, course.get('title'))

youtube_dl/extractor/pluralsight.py

@@ -169,11 +169,10 @@ class PluralsightIE(PluralsightBaseIE):
 
         collection = course['modules']
 
-        module, clip = None, None
+        clip = None
 
         for module_ in collection:
             if name in (module_.get('moduleName'), module_.get('name')):
-                module = module_
                 for clip_ in module_.get('clips', []):
                     clip_index = clip_.get('clipIndex')
                     if clip_index is None:

youtube_dl/extractor/rai.py

@@ -1,23 +1,40 @@
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..compat import (
+    compat_urlparse,
+    compat_str,
+)
 from ..utils import (
-    determine_ext,
     ExtractorError,
+    determine_ext,
     find_xpath_attr,
     fix_xml_ampersands,
+    GeoRestrictedError,
     int_or_none,
     parse_duration,
+    strip_or_none,
+    try_get,
     unified_strdate,
+    unified_timestamp,
     update_url_query,
+    urljoin,
     xpath_text,
 )
 
 
 class RaiBaseIE(InfoExtractor):
-    def _extract_relinker_formats(self, relinker_url, video_id):
+    _UUID_RE = r'[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12}'
+    _GEO_COUNTRIES = ['IT']
+    _GEO_BYPASS = False
+
+    def _extract_relinker_info(self, relinker_url, video_id):
         formats = []
+        geoprotection = None
+        is_live = None
+        duration = None
 
         for platform in ('mon', 'flash', 'native'):
             relinker = self._download_xml(
@@ -27,9 +44,27 @@ class RaiBaseIE(InfoExtractor):
                 query={'output': 45, 'pl': platform},
                 headers=self.geo_verification_headers())
 
-            media_url = find_xpath_attr(relinker, './url', 'type', 'content').text
+            if not geoprotection:
+                geoprotection = xpath_text(
+                    relinker, './geoprotection', default=None) == 'Y'
+
+            if not is_live:
+                is_live = xpath_text(
+                    relinker, './is_live', default=None) == 'Y'
+            if not duration:
+                duration = parse_duration(xpath_text(
+                    relinker, './duration', default=None))
+
+            url_elem = find_xpath_attr(relinker, './url', 'type', 'content')
+            if url_elem is None:
+                continue
+
+            media_url = url_elem.text
+
+            # This does not imply geo restriction (e.g.
+            # http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html)
             if media_url == 'http://download.rai.it/video_no_available.mp4':
-                self.raise_geo_restricted()
+                continue
 
             ext = determine_ext(media_url)
             if (ext == 'm3u8' and platform != 'mon') or (ext == 'f4m' and platform != 'flash'):
@@ -53,35 +88,225 @@ class RaiBaseIE(InfoExtractor):
                     'format_id': 'http-%d' % bitrate if bitrate > 0 else 'http',
                 })
 
-        return formats
+        if not formats and geoprotection is True:
+            self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
 
-    def _extract_from_content_id(self, content_id, base_url):
+        return dict((k, v) for k, v in {
+            'is_live': is_live,
+            'duration': duration,
+            'formats': formats,
+        }.items() if v is not None)
+
+
+class RaiPlayIE(RaiBaseIE):
+    _VALID_URL = r'(?P<url>https?://(?:www\.)?raiplay\.it/.+?-(?P<id>%s)\.html)' % RaiBaseIE._UUID_RE
+    _TESTS = [{
+        'url': 'http://www.raiplay.it/video/2016/10/La-Casa-Bianca-e06118bb-59a9-4636-b914-498e4cfd2c66.html?source=twitter',
+        'md5': '340aa3b7afb54bfd14a8c11786450d76',
+        'info_dict': {
+            'id': 'e06118bb-59a9-4636-b914-498e4cfd2c66',
+            'ext': 'mp4',
+            'title': 'La Casa Bianca',
+            'alt_title': 'S2016 - Puntata del 23/10/2016',
+            'description': 'md5:a09d45890850458077d1f68bb036e0a5',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'Rai 3',
+            'creator': 'Rai 3',
+            'duration': 3278,
+            'timestamp': 1477764300,
+            'upload_date': '20161029',
+            'series': 'La Casa Bianca',
+            'season': '2016',
+        },
+    }, {
+        'url': 'http://www.raiplay.it/video/2014/04/Report-del-07042014-cb27157f-9dd0-4aee-b788-b1f67643a391.html',
+        'md5': '8970abf8caf8aef4696e7b1f2adfc696',
+        'info_dict': {
+            'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
+            'ext': 'mp4',
+            'title': 'Report del 07/04/2014',
+            'alt_title': 'S2013/14 - Puntata del 07/04/2014',
+            'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'Rai 5',
+            'creator': 'Rai 5',
+            'duration': 6160,
+            'series': 'Report',
+            'season_number': 5,
+            'season': '2013/14',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.raiplay.it/video/2016/11/gazebotraindesi-efebe701-969c-4593-92f3-285f0d1ce750.html?',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        url, video_id = mobj.group('url', 'id')
+
+        media = self._download_json(
+            '%s?json' % url, video_id, 'Downloading video JSON')
+
+        title = media['name']
+
+        video = media['video']
+
+        relinker_info = self._extract_relinker_info(video['contentUrl'], video_id)
+        self._sort_formats(relinker_info['formats'])
+
+        thumbnails = []
+        if 'images' in media:
+            for _, value in media.get('images').items():
+                if value:
+                    thumbnails.append({
+                        'url': value.replace('[RESOLUTION]', '600x400')
+                    })
+
+        timestamp = unified_timestamp(try_get(
+            media, lambda x: x['availabilities'][0]['start'], compat_str))
+
+        info = {
+            'id': video_id,
+            'title': title,
+            'alt_title': media.get('subtitle'),
+            'description': media.get('description'),
+            'uploader': media.get('channel'),
+            'creator': media.get('editor'),
+            'duration': parse_duration(video.get('duration')),
+            'timestamp': timestamp,
+            'thumbnails': thumbnails,
+            'series': try_get(
+                media, lambda x: x['isPartOf']['name'], compat_str),
+            'season_number': int_or_none(try_get(
+                media, lambda x: x['isPartOf']['numeroStagioni'])),
+            'season': media.get('stagione') or None,
+        }
+
+        info.update(relinker_info)
+
+        return info
+
+
+class RaiIE(RaiBaseIE):
+    _VALID_URL = r'https?://[^/]+\.(?:rai\.(?:it|tv)|rainews\.it)/dl/.+?-(?P<id>%s)(?:-.+?)?\.html' % RaiBaseIE._UUID_RE
+    _TESTS = [{
+        # var uniquename = "ContentItem-..."
+        # data-id="ContentItem-..."
+        'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
+        'info_dict': {
+            'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
+            'ext': 'mp4',
+            'title': 'TG PRIMO TEMPO',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 1758,
+            'upload_date': '20140612',
+        }
+    }, {
+        # with ContentItem in many metas
+        'url': 'http://www.rainews.it/dl/rainews/media/Weekend-al-cinema-da-Hollywood-arriva-il-thriller-di-Tate-Taylor-La-ragazza-del-treno-1632c009-c843-4836-bb65-80c33084a64b.html',
+        'info_dict': {
+            'id': '1632c009-c843-4836-bb65-80c33084a64b',
+            'ext': 'mp4',
+            'title': 'Weekend al cinema, da Hollywood arriva il thriller di Tate Taylor "La ragazza del treno"',
+            'description': 'I film in uscita questa settimana.',
+            'thumbnail': r're:^https?://.*\.png$',
+            'duration': 833,
+            'upload_date': '20161103',
+        }
+    }, {
+        # with ContentItem in og:url
+        'url': 'http://www.rai.it/dl/RaiTV/programmi/media/ContentItem-efb17665-691c-45d5-a60c-5301333cbb0c.html',
+        'md5': '11959b4e44fa74de47011b5799490adf',
+        'info_dict': {
+            'id': 'efb17665-691c-45d5-a60c-5301333cbb0c',
+            'ext': 'mp4',
+            'title': 'TG1 ore 20:00 del 03/11/2016',
+            'description': 'TG1 edizione integrale ore 20:00 del giorno 03/11/2016',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2214,
+            'upload_date': '20161103',
+        }
+    }, {
+        # drawMediaRaiTV(...)
+        'url': 'http://www.report.rai.it/dl/Report/puntata/ContentItem-0c7a664b-d0f4-4b2c-8835-3f82e46f433e.html',
+        'md5': '2dd727e61114e1ee9c47f0da6914e178',
+        'info_dict': {
+            'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
+            'ext': 'mp4',
+            'title': 'Il pacco',
+            'description': 'md5:4b1afae1364115ce5d78ed83cd2e5b3a',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'upload_date': '20141221',
+        },
+    }, {
+        # initEdizione('ContentItem-...'
+        'url': 'http://www.tg1.rai.it/dl/tg1/2010/edizioni/ContentSet-9b6e0cba-4bef-4aef-8cf0-9f7f665b7dfb-tg1.html?item=undefined',
+        'info_dict': {
+            'id': 'c2187016-8484-4e3a-8ac8-35e475b07303',
+            'ext': 'mp4',
+            'title': r're:TG1 ore \d{2}:\d{2} del \d{2}/\d{2}/\d{4}',
+            'duration': 2274,
+            'upload_date': '20170401',
+        },
+        'skip': 'Changes daily',
+    }, {
+        # HDS live stream with only relinker URL
+        'url': 'http://www.rai.tv/dl/RaiTV/dirette/PublishingBlock-1912dbbf-3f96-44c3-b4cf-523681fbacbc.html?channel=EuroNews',
+        'info_dict': {
+            'id': '1912dbbf-3f96-44c3-b4cf-523681fbacbc',
+            'ext': 'flv',
+            'title': 'EuroNews',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        # HLS live stream with ContentItem in og:url
+        'url': 'http://www.rainews.it/dl/rainews/live/ContentItem-3156f2f2-dc70-4953-8e2f-70d7489d4ce9.html',
+        'info_dict': {
+            'id': '3156f2f2-dc70-4953-8e2f-70d7489d4ce9',
+            'ext': 'mp4',
+            'title': 'La diretta di Rainews24',
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }]
+
+    def _extract_from_content_id(self, content_id, url):
         media = self._download_json(
             'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-%s.html?json' % content_id,
             content_id, 'Downloading video JSON')
 
+        title = media['name'].strip()
+
+        media_type = media['type']
+        if 'Audio' in media_type:
+            relinker_info = {
+                'formats': {
+                    'format_id': media.get('formatoAudio'),
+                    'url': media['audioUrl'],
+                    'ext': media.get('formatoAudio'),
+                }
+            }
+        elif 'Video' in media_type:
+            relinker_info = self._extract_relinker_info(media['mediaUri'], content_id)
+        else:
+            raise ExtractorError('not a media file')
+
+        self._sort_formats(relinker_info['formats'])
+
         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),
+                    'url': compat_urlparse.urljoin(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:
@@ -94,174 +319,90 @@ class RaiBaseIE(InfoExtractor):
                 'url': captions,
             }]
 
-        return {
+        info = {
             'id': content_id,
-            'title': media['name'],
-            'description': media.get('desc'),
+            'title': title,
+            'description': strip_or_none(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,
         }
 
+        info.update(relinker_info)
 
-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': '8970abf8caf8aef4696e7b1f2adfc696',
-            'info_dict': {
-                'id': 'cb27157f-9dd0-4aee-b788-b1f67643a391',
-                'ext': 'mp4',
-                'title': 'Report del 07/04/2014',
-                'description': 'md5:f27c544694cacb46a078db84ec35d2d9',
-                'upload_date': '20140407',
-                'duration': 6160,
-                'thumbnail': r're:^https?://.*\.jpg$',
-            }
-        },
-        {
-            # no m3u8 stream
-            'url': 'http://www.raisport.rai.it/dl/raiSport/media/rassegna-stampa-04a9f4bd-b563-40cf-82a6-aad3529cb4a9.html',
-            # HDS download, MD5 is unstable
-            'info_dict': {
-                'id': '04a9f4bd-b563-40cf-82a6-aad3529cb4a9',
-                'ext': 'flv',
-                'title': 'TG PRIMO TEMPO',
-                'upload_date': '20140612',
-                'duration': 1758,
-                'thumbnail': r're:^https?://.*\.jpg$',
-            },
-            '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',
-            'md5': '35cf7c229f22eeef43e48b5cf923bef0',
-            'info_dict': {
-                'id': '7aafdea9-0e5d-49d5-88a6-7e65da67ae13',
-                'ext': 'mp4',
-                'title': 'State of the Net, Antonella La Carpia: regole virali',
-                'description': 'md5:b0ba04a324126903e3da7763272ae63c',
-                'upload_date': '20140613',
-            },
-            'skip': 'Error 404',
-        },
-        {
-            'url': 'http://www.rai.tv/dl/RaiTV/programmi/media/ContentItem-b4a49761-e0cc-4b14-8736-2729f6f73132-tg2.html',
-            'info_dict': {
-                'id': 'b4a49761-e0cc-4b14-8736-2729f6f73132',
-                'ext': 'mp4',
-                'title': 'Alluvione in Sardegna e dissesto idrogeologico',
-                'description': 'Edizione delle ore 20:30 ',
-            },
-            'skip': 'invalid urls',
-        },
-        {
-            'url': 'http://www.ilcandidato.rai.it/dl/ray/media/Il-Candidato---Primo-episodio-Le-Primarie-28e5525a-b495-45e8-a7c3-bc48ba45d2b6.html',
-            'md5': 'e57493e1cb8bc7c564663f363b171847',
-            'info_dict': {
-                'id': '28e5525a-b495-45e8-a7c3-bc48ba45d2b6',
-                'ext': 'mp4',
-                'title': 'Il Candidato - Primo episodio: "Le Primarie"',
-                'description': 'md5:364b604f7db50594678f483353164fb8',
-                'upload_date': '20140923',
-                'duration': 386,
-                'thumbnail': r're:^https?://.*\.jpg$',
-            }
-        },
-    ]
+        return info
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        return self._extract_from_content_id(video_id, url)
-
-
-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': '2dd727e61114e1ee9c47f0da6914e178',
-            'info_dict': {
-                'id': '59d69d28-6bb6-409d-a4b5-ed44096560af',
-                '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
-    def suitable(cls, url):
-        return False if RaiTVIE.suitable(url) else super(RaiIE, cls).suitable(url)
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        iframe_url = self._search_regex(
-            [r'<iframe[^>]+src="([^"]*/dl/[^"]+\?iframe\b[^"]*)"',
-             r'drawMediaRaiTV\(["\'](.+?)["\']'],
-            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 = None
 
-        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)
+        content_item_url = self._html_search_meta(
+            ('og:url', 'og:video', 'og:video:secure_url', 'twitter:url',
+             'twitter:player', 'jsonlink'), webpage, default=None)
+        if content_item_url:
+            content_item_id = self._search_regex(
+                r'ContentItem-(%s)' % self._UUID_RE, content_item_url,
+                'content item id', default=None)
+
+        if not content_item_id:
+            content_item_id = self._search_regex(
+                r'''(?x)
+                    (?:
+                        (?:initEdizione|drawMediaRaiTV)\(|
+                        <(?:[^>]+\bdata-id|var\s+uniquename)=
+                    )
+                    (["\'])
+                    (?:(?!\1).)*\bContentItem-(?P<id>%s)
+                ''' % self._UUID_RE,
+                webpage, 'content item id', default=None, group='id')
+
+        content_item_ids = set()
         if content_item_id:
-            return self._extract_from_content_id(content_item_id, url)
+            content_item_ids.add(content_item_id)
+        if video_id not in content_item_ids:
+            content_item_ids.add(video_id)
 
-        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)
+        for content_item_id in content_item_ids:
+            try:
+                return self._extract_from_content_id(content_item_id, url)
+            except GeoRestrictedError:
+                raise
+            except ExtractorError:
+                pass
+
+        relinker_url = self._search_regex(
+            r'''(?x)
+                (?:
+                    var\s+videoURL|
+                    mediaInfo\.mediaUri
+                )\s*=\s*
+                ([\'"])
+                (?P<url>
+                    (?:https?:)?
+                    //mediapolis(?:vod)?\.rai\.it/relinker/relinkerServlet\.htm\?
+                    (?:(?!\1).)*\bcont=(?:(?!\1).)+)\1
+            ''',
+            webpage, 'relinker URL', group='url')
+
+        relinker_info = self._extract_relinker_info(
+            urljoin(url, relinker_url), video_id)
+        self._sort_formats(relinker_info['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)
+            webpage, 'title', group='title',
+            default=None) or self._og_search_title(webpage)
 
-        return {
+        info = {
             'id': video_id,
             'title': title,
-            'formats': formats,
         }
+
+        info.update(relinker_info)
+
+        return info

youtube_dl/extractor/tvnoe.py

@@ -31,9 +31,8 @@ class TVNoeIE(InfoExtractor):
             r'<iframe[^>]+src="([^"]+)"', webpage, 'iframe URL')
 
         ifs_page = self._download_webpage(iframe_url, video_id)
-        jwplayer_data = self._parse_json(
-            self._find_jwplayer_data(ifs_page),
-            video_id, transform_source=js_to_json)
+        jwplayer_data = self._find_jwplayer_data(
+            ifs_page, video_id, transform_source=js_to_json)
         info_dict = self._parse_jwplayer_data(
             jwplayer_data, video_id, require_title=False, base_url=iframe_url)
 

youtube_dl/extractor/tvplay.py

@@ -225,7 +225,11 @@ class TVPlayIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-
+        geo_country = self._search_regex(
+            r'https?://[^/]+\.([a-z]{2})', url,
+            'geo country', default=None)
+        if geo_country:
+            self._initialize_geo_bypass([geo_country.upper()])
         video = self._download_json(
             'http://playapi.mtgx.tv/v3/videos/%s' % video_id, video_id, 'Downloading video JSON')
 

youtube_dl/extractor/viu.py

@@ -44,7 +44,7 @@ class ViuBaseIE(InfoExtractor):
 
 
 class ViuIE(ViuBaseIE):
-    _VALID_URL = r'(?:viu:|https?://www\.viu\.com/[a-z]{2}/media/)(?P<id>\d+)'
+    _VALID_URL = r'(?:viu:|https?://[^/]+\.viu\.com/[a-z]{2}/media/)(?P<id>\d+)'
     _TESTS = [{
         'url': 'https://www.viu.com/en/media/1116705532?containerId=playlist-22168059',
         'info_dict': {
@@ -69,6 +69,9 @@ class ViuIE(ViuBaseIE):
             'skip_download': 'm3u8 download',
         },
         'skip': 'Geo-restricted to Indonesia',
+    }, {
+        'url': 'https://india.viu.com/en/media/1126286865',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/vk.py

@@ -432,8 +432,7 @@ class VKIE(VKBaseIE):
                 })
             elif format_id == 'hls':
                 formats.extend(self._extract_m3u8_formats(
-                    format_url, video_id, 'mp4',
-                    entry_protocol='m3u8' if is_live else 'm3u8_native',
+                    format_url, video_id, 'mp4', 'm3u8_native',
                     m3u8_id=format_id, fatal=False, live=is_live))
             elif format_id == 'rtmp':
                 formats.append({

youtube_dl/extractor/vrv.py

@@ -0,0 +1,191 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import base64
+import json
+import hashlib
+import hmac
+import random
+import string
+import time
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse_urlencode,
+    compat_urlparse,
+)
+from ..utils import (
+    float_or_none,
+    int_or_none,
+)
+
+
+class VRVBaseIE(InfoExtractor):
+    _API_DOMAIN = None
+    _API_PARAMS = {}
+    _CMS_SIGNING = {}
+
+    def _call_api(self, path, video_id, note, data=None):
+        base_url = self._API_DOMAIN + '/core/' + path
+        encoded_query = compat_urllib_parse_urlencode({
+            'oauth_consumer_key': self._API_PARAMS['oAuthKey'],
+            'oauth_nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
+            'oauth_signature_method': 'HMAC-SHA1',
+            'oauth_timestamp': int(time.time()),
+            'oauth_version': '1.0',
+        })
+        headers = self.geo_verification_headers()
+        if data:
+            data = json.dumps(data).encode()
+            headers['Content-Type'] = 'application/json'
+        method = 'POST' if data else 'GET'
+        base_string = '&'.join([method, compat_urlparse.quote(base_url, ''), compat_urlparse.quote(encoded_query, '')])
+        oauth_signature = base64.b64encode(hmac.new(
+            (self._API_PARAMS['oAuthSecret'] + '&').encode('ascii'),
+            base_string.encode(), hashlib.sha1).digest()).decode()
+        encoded_query += '&oauth_signature=' + compat_urlparse.quote(oauth_signature, '')
+        return self._download_json(
+            '?'.join([base_url, encoded_query]), video_id,
+            note='Downloading %s JSON metadata' % note, headers=headers, data=data)
+
+    def _call_cms(self, path, video_id, note):
+        if not self._CMS_SIGNING:
+            self._CMS_SIGNING = self._call_api('index', video_id, 'CMS Signing')['cms_signing']
+        return self._download_json(
+            self._API_DOMAIN + path, video_id, query=self._CMS_SIGNING,
+            note='Downloading %s JSON metadata' % note, headers=self.geo_verification_headers())
+
+    def _set_api_params(self, webpage, video_id):
+        if not self._API_PARAMS:
+            self._API_PARAMS = self._parse_json(self._search_regex(
+                r'window\.__APP_CONFIG__\s*=\s*({.+?})</script>',
+                webpage, 'api config'), video_id)['cxApiParams']
+            self._API_DOMAIN = self._API_PARAMS.get('apiDomain', 'https://api.vrv.co')
+
+    def _get_cms_resource(self, resource_key, video_id):
+        return self._call_api(
+            'cms_resource', video_id, 'resource path', data={
+                'resource_key': resource_key,
+            })['__links__']['cms_resource']['href']
+
+
+class VRVIE(VRVBaseIE):
+    IE_NAME = 'vrv'
+    _VALID_URL = r'https?://(?:www\.)?vrv\.co/watch/(?P<id>[A-Z0-9]+)'
+    _TEST = {
+        'url': 'https://vrv.co/watch/GR9PNZ396/Hidden-America-with-Jonah-Ray:BOSTON-WHERE-THE-PAST-IS-THE-PRESENT',
+        'info_dict': {
+            'id': 'GR9PNZ396',
+            'ext': 'mp4',
+            'title': 'BOSTON: WHERE THE PAST IS THE PRESENT',
+            'description': 'md5:4ec8844ac262ca2df9e67c0983c6b83f',
+            'uploader_id': 'seeso',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(
+            url, video_id,
+            headers=self.geo_verification_headers())
+        media_resource = self._parse_json(self._search_regex(
+            r'window\.__INITIAL_STATE__\s*=\s*({.+?})</script>',
+            webpage, 'inital state'), video_id).get('watch', {}).get('mediaResource') or {}
+
+        video_data = media_resource.get('json')
+        if not video_data:
+            self._set_api_params(webpage, video_id)
+            episode_path = self._get_cms_resource(
+                'cms:/episodes/' + video_id, video_id)
+            video_data = self._call_cms(episode_path, video_id, 'video')
+        title = video_data['title']
+
+        streams_json = media_resource.get('streams', {}).get('json', {})
+        if not streams_json:
+            self._set_api_params(webpage, video_id)
+            streams_path = video_data['__links__']['streams']['href']
+            streams_json = self._call_cms(streams_path, video_id, 'streams')
+
+        audio_locale = streams_json.get('audio_locale')
+        formats = []
+        for stream_id, stream in streams_json.get('streams', {}).get('adaptive_hls', {}).items():
+            stream_url = stream.get('url')
+            if not stream_url:
+                continue
+            stream_id = stream_id or audio_locale
+            m3u8_formats = self._extract_m3u8_formats(
+                stream_url, video_id, 'mp4', m3u8_id=stream_id,
+                note='Downloading %s m3u8 information' % stream_id,
+                fatal=False)
+            if audio_locale:
+                for f in m3u8_formats:
+                    f['language'] = audio_locale
+            formats.extend(m3u8_formats)
+        self._sort_formats(formats)
+
+        thumbnails = []
+        for thumbnail in video_data.get('images', {}).get('thumbnails', []):
+            thumbnail_url = thumbnail.get('source')
+            if not thumbnail_url:
+                continue
+            thumbnails.append({
+                'url': thumbnail_url,
+                'width': int_or_none(thumbnail.get('width')),
+                'height': int_or_none(thumbnail.get('height')),
+            })
+
+        return {
+            'id': video_id,
+            'title': title,
+            'formats': formats,
+            'thumbnails': thumbnails,
+            'description': video_data.get('description'),
+            'duration': float_or_none(video_data.get('duration_ms'), 1000),
+            'uploader_id': video_data.get('channel_id'),
+            'series': video_data.get('series_title'),
+            'season': video_data.get('season_title'),
+            'season_number': int_or_none(video_data.get('season_number')),
+            'season_id': video_data.get('season_id'),
+            'episode': title,
+            'episode_number': int_or_none(video_data.get('episode_number')),
+            'episode_id': video_data.get('production_episode_id'),
+        }
+
+
+class VRVSeriesIE(VRVBaseIE):
+    IE_NAME = 'vrv:series'
+    _VALID_URL = r'https?://(?:www\.)?vrv\.co/series/(?P<id>[A-Z0-9]+)'
+    _TEST = {
+        'url': 'https://vrv.co/series/G68VXG3G6/The-Perfect-Insider',
+        'info_dict': {
+            'id': 'G68VXG3G6',
+        },
+        'playlist_mincount': 11,
+    }
+
+    def _real_extract(self, url):
+        series_id = self._match_id(url)
+        webpage = self._download_webpage(
+            url, series_id,
+            headers=self.geo_verification_headers())
+
+        self._set_api_params(webpage, series_id)
+        seasons_path = self._get_cms_resource(
+            'cms:/seasons?series_id=' + series_id, series_id)
+        seasons_data = self._call_cms(seasons_path, series_id, 'seasons')
+
+        entries = []
+        for season in seasons_data.get('items', []):
+            episodes_path = season['__links__']['season/episodes']['href']
+            episodes = self._call_cms(episodes_path, series_id, 'episodes')
+            for episode in episodes.get('items', []):
+                episode_id = episode['id']
+                entries.append(self.url_result(
+                    'https://vrv.co/watch/' + episode_id,
+                    'VRV', episode_id, episode.get('title')))
+
+        return self.playlist_result(entries, series_id)

youtube_dl/extractor/xfileshare.py

@@ -6,6 +6,7 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     decode_packed_codes,
+    determine_ext,
     ExtractorError,
     int_or_none,
     NO_DEFAULT,
@@ -26,6 +27,7 @@ class XFileShareIE(InfoExtractor):
         ('vidto.me', 'Vidto'),
         ('streamin.to', 'Streamin.To'),
         ('xvidstage.com', 'XVIDSTAGE'),
+        ('vidabc.com', 'Vid ABC'),
     )
 
     IE_DESC = 'XFileShare based sites: %s' % ', '.join(list(zip(*_SITES))[1])
@@ -95,6 +97,16 @@ class XFileShareIE(InfoExtractor):
         # removed by administrator
         'url': 'http://xvidstage.com/amfy7atlkx25',
         'only_matching': True,
+    }, {
+        'url': 'http://vidabc.com/i8ybqscrphfv',
+        'info_dict': {
+            'id': 'i8ybqscrphfv',
+            'ext': 'mp4',
+            'title': 're:Beauty and the Beast 2017',
+        },
+        'params': {
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
@@ -133,31 +145,45 @@ class XFileShareIE(InfoExtractor):
             webpage, 'title', default=None) or self._og_search_title(
             webpage, default=None) or video_id).strip()
 
-        def extract_video_url(default=NO_DEFAULT):
-            return self._search_regex(
-                (r'file\s*:\s*(["\'])(?P<url>http.+?)\1,',
-                 r'file_link\s*=\s*(["\'])(?P<url>http.+?)\1',
-                 r'addVariable\((\\?["\'])file\1\s*,\s*(\\?["\'])(?P<url>http.+?)\2\)',
-                 r'<embed[^>]+src=(["\'])(?P<url>http.+?)\1'),
-                webpage, 'file url', default=default, group='url')
+        def extract_formats(default=NO_DEFAULT):
+            urls = []
+            for regex in (
+                    r'file\s*:\s*(["\'])(?P<url>http(?:(?!\1).)+\.(?:m3u8|mp4|flv)(?:(?!\1).)*)\1',
+                    r'file_link\s*=\s*(["\'])(?P<url>http(?:(?!\1).)+)\1',
+                    r'addVariable\((\\?["\'])file\1\s*,\s*(\\?["\'])(?P<url>http(?:(?!\2).)+)\2\)',
+                    r'<embed[^>]+src=(["\'])(?P<url>http(?:(?!\1).)+\.(?:m3u8|mp4|flv)(?:(?!\1).)*)\1'):
+                for mobj in re.finditer(regex, webpage):
+                    video_url = mobj.group('url')
+                    if video_url not in urls:
+                        urls.append(video_url)
+            formats = []
+            for video_url in urls:
+                if determine_ext(video_url) == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        video_url, video_id, 'mp4',
+                        entry_protocol='m3u8_native', m3u8_id='hls',
+                        fatal=False))
+                else:
+                    formats.append({
+                        'url': video_url,
+                        'format_id': 'sd',
+                    })
+            if not formats and default is not NO_DEFAULT:
+                return default
+            self._sort_formats(formats)
+            return formats
 
-        video_url = extract_video_url(default=None)
+        formats = extract_formats(default=None)
 
-        if not video_url:
+        if not formats:
             webpage = decode_packed_codes(self._search_regex(
                 r"(}\('(.+)',(\d+),(\d+),'[^']*\b(?:file|embed)\b[^']*'\.split\('\|'\))",
                 webpage, 'packed code'))
-            video_url = extract_video_url()
+            formats = extract_formats()
 
         thumbnail = self._search_regex(
             r'image\s*:\s*["\'](http[^"\']+)["\'],', webpage, 'thumbnail', default=None)
 
-        formats = [{
-            'format_id': 'sd',
-            'url': video_url,
-            'quality': 1,
-        }]
-
         return {
             'id': video_id,
             'title': title,

youtube_dl/extractor/youtube.py

@@ -59,6 +59,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
     # If True it will raise an error if no login info is provided
     _LOGIN_REQUIRED = False
 
+    _PLAYLIST_ID_RE = r'(?:PL|LL|EC|UU|FL|RD|UL|TL)[0-9A-Za-z-_]{10,}'
+
     def _set_language(self):
         self._set_cookie(
             '.youtube.com', 'PREF', 'f1=50000000&hl=en',
@@ -265,9 +267,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                          )
                      )?                                                       # all until now is optional -> you can pass the naked ID
                      ([0-9A-Za-z_-]{11})                                      # here is it! the YouTube video ID
-                     (?!.*?\blist=)                                            # combined list/video URLs are handled by the playlist IE
+                     (?!.*?\blist=
+                        (?:
+                            %(playlist_id)s|                                  # combined list/video URLs are handled by the playlist IE
+                            WL                                                # WL are handled by the watch later IE
+                        )
+                     )
                      (?(1).+)?                                                # if we found the ID, everything can follow
-                     $"""
+                     $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
     _NEXT_URL_RE = r'[\?&]next_url=([^&]+)'
     _formats = {
         '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'},
@@ -924,6 +931,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             'url': 'sJL6WA-aGkQ',
             'only_matching': True,
         },
+        {
+            'url': 'https://www.youtube.com/watch?v=MuAGGZNfUkU&list=RDMM',
+            'only_matching': True,
+        },
     ]
 
     def __init__(self, *args, **kwargs):
@@ -1864,8 +1875,8 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
                         )
                         .*
                      |
-                        ((?:PL|LL|EC|UU|FL|RD|UL|TL)[0-9A-Za-z-_]{10,})
-                     )"""
+                        (%(playlist_id)s)
+                     )""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
     _TEMPLATE_URL = 'https://www.youtube.com/playlist?list=%s&disable_polymer=true'
     _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})&amp;[^"]*?index=(?P<index>\d+)(?:[^>]+>(?P<title>[^<]+))?'
     IE_NAME = 'youtube:playlist'

youtube_dl/options.py

@@ -459,11 +459,11 @@ def parseOpts(overrideArguments=None):
     downloader.add_option(
         '--fragment-retries',
         dest='fragment_retries', metavar='RETRIES', default=10,
-        help='Number of retries for a fragment (default is %default), or "infinite" (DASH and hlsnative only)')
+        help='Number of retries for a fragment (default is %default), or "infinite" (DASH, hlsnative and ISM)')
     downloader.add_option(
         '--skip-unavailable-fragments',
         action='store_true', dest='skip_unavailable_fragments', default=True,
-        help='Skip unavailable fragments (DASH and hlsnative only)')
+        help='Skip unavailable fragments (DASH, hlsnative and ISM)')
     downloader.add_option(
         '--abort-on-unavailable-fragment',
         action='store_false', dest='skip_unavailable_fragments',

youtube_dl/utils.py

@@ -39,6 +39,7 @@ from .compat import (
     compat_basestring,
     compat_chr,
     compat_etree_fromstring,
+    compat_expanduser,
     compat_html_entities,
     compat_html_entities_html5,
     compat_http_client,
@@ -539,6 +540,11 @@ def sanitized_Request(url, *args, **kwargs):
     return compat_urllib_request.Request(sanitize_url(url), *args, **kwargs)
 
 
+def expand_path(s):
+    """Expand shell variables and ~"""
+    return os.path.expandvars(compat_expanduser(s))
+
+
 def orderedSet(iterable):
     """ Remove all duplicates from the input iterable """
     res = []

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2017.03.22'
+__version__ = '2017.04.02'