release 2016.05.01

* Switch to JSON source
* Add abstract IE for playlists
* Extract more track related metadata

.github/ISSUE_TEMPLATE.md

@@ -6,8 +6,8 @@
 
 ---
 
-### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.04.24*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
-- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.04.24**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.05.01*. If it's not read [this FAQ entry](https://github.com/rg3/youtube-dl/blob/master/README.md#how-do-i-update-youtube-dl) and update. Issues with outdated version will be rejected.
+- [ ] I've **verified** and **I assure** that I'm running youtube-dl **2016.05.01**
 
 ### Before submitting an *issue* make sure you have:
 - [ ] At least skimmed through [README](https://github.com/rg3/youtube-dl/blob/master/README.md) and **most notably** [FAQ](https://github.com/rg3/youtube-dl#faq) and [BUGS](https://github.com/rg3/youtube-dl#bugs) sections
@@ -35,7 +35,7 @@ $ youtube-dl -v <your command line>
 [debug] User config: []
 [debug] Command-line args: [u'-v', u'http://www.youtube.com/watch?v=BaW_jenozKcj']
 [debug] Encodings: locale cp1251, fs mbcs, out cp866, pref cp1251
-[debug] youtube-dl version 2016.04.24
+[debug] youtube-dl version 2016.05.01
 [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: {}

AUTHORS

@@ -168,3 +168,4 @@ José Joaquín Atria
 Viťas Strádal
 Kagami Hiiragi
 Philip Huppert
+blahgeek

README.md

@@ -697,6 +697,10 @@ YouTube changed their playlist format in March 2014 and later on, so you'll need
 
 If you have installed youtube-dl with a package manager, pip, setup.py or a tarball, please use that to update. Note that Ubuntu packages do not seem to get updated anymore. Since we are not affiliated with Ubuntu, there is little we can do. Feel free to [report bugs](https://bugs.launchpad.net/ubuntu/+source/youtube-dl/+filebug) to the [Ubuntu packaging guys](mailto:ubuntu-motu@lists.ubuntu.com?subject=outdated%20version%20of%20youtube-dl) - all they have to do is update the package to a somewhat recent version. See above for a way to update.
 
+### I'm getting an error when trying to use output template: `error: using output template conflicts with using title, video ID or auto number`
+
+Make sure you are not using `-o` with any of these options `-t`, `--title`, `--id`, `-A` or `--auto-number` set in command line or in a configuration file. Remove the latter if any.
+
 ### Do I always have to pass `-citw`?
 
 By default, youtube-dl intends to have the best options (incidentally, if you have a convincing case that these should be different, [please file an issue where you explain that](https://yt-dl.org/bug)). Therefore, it is unnecessary and sometimes harmful to copy long option strings from webpages. In particular, the only option out of `-citw` that is regularly useful is `-i`.
@@ -717,7 +721,7 @@ Videos or video formats streamed via RTMP protocol can only be downloaded when [
 
 ### I have downloaded a video but how can I play it?
 
-Once the video is fully downloaded, use any video player, such as [vlc](http://www.videolan.org) or [mplayer](http://www.mplayerhq.hu/).
+Once the video is fully downloaded, use any video player, such as [mpv](https://mpv.io/), [vlc](http://www.videolan.org) or [mplayer](http://www.mplayerhq.hu/).
 
 ### I extracted a video URL with `-g`, but it does not play on another machine / in my webbrowser.
 

docs/supportedsites.md

@@ -338,7 +338,6 @@
  - **mailru**: Видео@Mail.Ru
  - **MakersChannel**
  - **MakerTV**
- - **Malemotion**
  - **MatchTV**
  - **MDR**: MDR.DE and KiKA
  - **media.ccc.de**
@@ -375,8 +374,8 @@
  - **mtvservices:embedded**
  - **MuenchenTV**: münchen.tv
  - **MusicPlayOn**
- - **muzu.tv**
  - **Mwave**
+ - **MwaveMeetGreet**
  - **MySpace**
  - **MySpace:album**
  - **MySpass**
@@ -554,7 +553,6 @@
  - **SenateISVP**
  - **ServingSys**
  - **Sexu**
- - **SexyKarma**: Sexy Karma and Watch Indian Porn
  - **Shahid**
  - **Shared**: shared.sx and vivo.sx
  - **ShareSix**
@@ -567,8 +565,6 @@
  - **smotri:broadcast**: Smotri.com broadcasts
  - **smotri:community**: Smotri.com community videos
  - **smotri:user**: Smotri.com user videos
- - **SnagFilms**
- - **SnagFilmsEmbed**
  - **Snotr**
  - **Sohu**
  - **soundcloud**
@@ -610,6 +606,7 @@
  - **Syfy**
  - **SztvHu**
  - **Tagesschau**
+ - **tagesschau:player**
  - **Tapely**
  - **Tass**
  - **TDSLifeway**
@@ -725,6 +722,8 @@
  - **Vidzi**
  - **vier**
  - **vier:videos**
+ - **ViewLift**
+ - **ViewLiftEmbed**
  - **Viewster**
  - **Viidea**
  - **viki**
@@ -756,6 +755,7 @@
  - **Walla**
  - **WashingtonPost**
  - **wat.tv**
+ - **WatchIndianPorn**: Watch Indian Porn
  - **WDR**
  - **wdr:mobile**
  - **WDRMaus**: Sendung mit der Maus
@@ -775,6 +775,10 @@
  - **XFileShare**: XFileShare based sites: GorillaVid.in, daclips.in, movpod.in, fastvideo.in, realvid.net, filehoot.com and vidto.me
  - **XHamster**
  - **XHamsterEmbed**
+ - **xiami:album**: 虾米音乐 - 专辑
+ - **xiami:artist**: 虾米音乐 - 歌手
+ - **xiami:collection**: 虾米音乐 - 精选集
+ - **xiami:song**: 虾米音乐
  - **XMinus**
  - **XNXX**
  - **Xstream**

youtube_dl/extractor/arte.py

@@ -63,7 +63,7 @@ class ArteTvIE(InfoExtractor):
 
 class ArteTVPlus7IE(InfoExtractor):
     IE_NAME = 'arte.tv:+7'
-    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&+])'
+    _VALID_URL = r'https?://(?:www\.)?arte\.tv/guide/(?P<lang>fr|de|en|es)/(?:(?:sendungen|emissions|embed)/)?(?P<id>[^/]+)/(?P<name>[^/?#&]+)'
 
     @classmethod
     def _extract_url_info(cls, url):

youtube_dl/extractor/ccc.py

@@ -1,13 +1,9 @@
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
-    parse_duration,
-    qualities,
-    unified_strdate,
+    parse_iso8601,
 )
 
 
@@ -19,14 +15,14 @@ class CCCIE(InfoExtractor):
         'url': 'https://media.ccc.de/v/30C3_-_5443_-_en_-_saal_g_-_201312281830_-_introduction_to_processor_design_-_byterazor#video',
         'md5': '3a1eda8f3a29515d27f5adb967d7e740',
         'info_dict': {
-            'id': '30C3_-_5443_-_en_-_saal_g_-_201312281830_-_introduction_to_processor_design_-_byterazor',
+            'id': '1839',
             'ext': 'mp4',
             'title': 'Introduction to Processor Design',
-            'description': 'md5:80be298773966f66d56cb11260b879af',
+            'description': 'md5:df55f6d073d4ceae55aae6f2fd98a0ac',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'view_count': int,
             'upload_date': '20131228',
-            'duration': 3660,
+            'timestamp': 1388188800,
+            'duration': 3710,
         }
     }, {
         'url': 'https://media.ccc.de/v/32c3-7368-shopshifting#download',
@@ -34,79 +30,48 @@ class CCCIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        event_id = self._search_regex("data-id='(\d+)'", webpage, 'event id')
+        event_data = self._download_json('https://media.ccc.de/public/events/%s' % event_id, event_id)
 
-        if self._downloader.params.get('prefer_free_formats'):
-            preference = qualities(['mp3', 'opus', 'mp4-lq', 'webm-lq', 'h264-sd', 'mp4-sd', 'webm-sd', 'mp4', 'webm', 'mp4-hd', 'h264-hd', 'webm-hd'])
-        else:
-            preference = qualities(['opus', 'mp3', 'webm-lq', 'mp4-lq', 'webm-sd', 'h264-sd', 'mp4-sd', 'webm', 'mp4', 'webm-hd', 'mp4-hd', 'h264-hd'])
-
-        title = self._html_search_regex(
-            r'(?s)<h1>(.*?)</h1>', webpage, 'title')
-        description = self._html_search_regex(
-            r'(?s)<h3>About</h3>(.+?)<h3>',
-            webpage, 'description', fatal=False)
-        upload_date = unified_strdate(self._html_search_regex(
-            r"(?s)<span[^>]+class='[^']*fa-calendar-o'[^>]*>(.+?)</span>",
-            webpage, 'upload date', fatal=False))
-        view_count = int_or_none(self._html_search_regex(
-            r"(?s)<span class='[^']*fa-eye'></span>(.*?)</li>",
-            webpage, 'view count', fatal=False))
-        duration = parse_duration(self._html_search_regex(
-            r'(?s)<span[^>]+class=(["\']).*?fa-clock-o.*?\1[^>]*></span>(?P<duration>.+?)</li',
-            webpage, 'duration', fatal=False, group='duration'))
-
-        matches = re.finditer(r'''(?xs)
-            <(?:span|div)\s+class='label\s+filetype'>(?P<format>[^<]*)</(?:span|div)>\s*
-            <(?:span|div)\s+class='label\s+filetype'>(?P<lang>[^<]*)</(?:span|div)>\s*
-            <a\s+download\s+href='(?P<http_url>[^']+)'>\s*
-            (?:
-                .*?
-                <a\s+(?:download\s+)?href='(?P<torrent_url>[^']+\.torrent)'
-            )?''', webpage)
         formats = []
-        for m in matches:
-            format = m.group('format')
-            format_id = self._search_regex(
-                r'.*/([a-z0-9_-]+)/[^/]*$',
-                m.group('http_url'), 'format id', default=None)
-            if format_id:
-                format_id = m.group('lang') + '-' + format_id
-            vcodec = 'h264' if 'h264' in format_id else (
-                'none' if format_id in ('mp3', 'opus') else None
+        for recording in event_data.get('recordings', []):
+            recording_url = recording.get('recording_url')
+            if not recording_url:
+                continue
+            language = recording.get('language')
+            folder = recording.get('folder')
+            format_id = None
+            if language:
+                format_id = language
+            if folder:
+                if language:
+                    format_id += '-' + folder
+                else:
+                    format_id = folder
+            vcodec = 'h264' if 'h264' in folder else (
+                'none' if folder in ('mp3', 'opus') else None
             )
             formats.append({
                 'format_id': format_id,
-                'format': format,
-                'language': m.group('lang'),
-                'url': m.group('http_url'),
+                'url': recording_url,
+                'width': int_or_none(recording.get('width')),
+                'height': int_or_none(recording.get('height')),
+                'filesize': int_or_none(recording.get('size'), invscale=1024 * 1024),
+                'language': language,
                 'vcodec': vcodec,
-                'preference': preference(format_id),
             })
-
-            if m.group('torrent_url'):
-                formats.append({
-                    'format_id': 'torrent-%s' % (format if format_id is None else format_id),
-                    'format': '%s (torrent)' % format,
-                    'proto': 'torrent',
-                    'format_note': '(unsupported; will just download the .torrent file)',
-                    'vcodec': vcodec,
-                    'preference': -100 + preference(format_id),
-                    'url': m.group('torrent_url'),
-                })
         self._sort_formats(formats)
 
-        thumbnail = self._html_search_regex(
-            r"<video.*?poster='([^']+)'", webpage, 'thumbnail', fatal=False)
-
         return {
-            'id': video_id,
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'view_count': view_count,
-            'upload_date': upload_date,
-            'duration': duration,
+            'id': event_id,
+            'display_id': display_id,
+            'title': event_data['title'],
+            'description': event_data.get('description'),
+            'thumbnail': event_data.get('thumb_url'),
+            'timestamp': parse_iso8601(event_data.get('date')),
+            'duration': int_or_none(event_data.get('length')),
+            'tags': event_data.get('tags'),
             'formats': formats,
         }

youtube_dl/extractor/cloudy.py

@@ -19,7 +19,7 @@ from ..utils import (
 class CloudyIE(InfoExtractor):
     _IE_DESC = 'cloudy.ec and videoraj.ch'
     _VALID_URL = r'''(?x)
-        https?://(?:www\.)?(?P<host>cloudy\.ec|videoraj\.ch)/
+        https?://(?:www\.)?(?P<host>cloudy\.ec|videoraj\.(?:ch|to))/
         (?:v/|embed\.php\?id=)
         (?P<id>[A-Za-z0-9]+)
         '''
@@ -37,7 +37,7 @@ class CloudyIE(InfoExtractor):
             }
         },
         {
-            'url': 'http://www.videoraj.ch/v/47f399fd8bb60',
+            'url': 'http://www.videoraj.to/v/47f399fd8bb60',
             'md5': '7d0f8799d91efd4eda26587421c3c3b0',
             'info_dict': {
                 'id': '47f399fd8bb60',

youtube_dl/extractor/common.py

@@ -1006,6 +1006,13 @@ class InfoExtractor(object):
     def _parse_f4m_formats(self, manifest, manifest_url, video_id, preference=None, f4m_id=None,
                            transform_source=lambda s: fix_xml_ampersands(s).strip(),
                            fatal=True):
+        # currently youtube-dl cannot decode the playerVerificationChallenge as Akamai uses Adobe Alchemy
+        akamai_pv = manifest.find('{http://ns.adobe.com/f4m/1.0}pv-2.0')
+        if akamai_pv is not None and ';' in akamai_pv.text:
+            playerVerificationChallenge = akamai_pv.text.split(';')[0]
+            if playerVerificationChallenge.strip() != '':
+                return []
+
         formats = []
         manifest_version = '1.0'
         media_nodes = manifest.findall('{http://ns.adobe.com/f4m/1.0}media')
@@ -1054,7 +1061,7 @@ class InfoExtractor(object):
     def _extract_m3u8_formats(self, m3u8_url, video_id, ext=None,
                               entry_protocol='m3u8', preference=None,
                               m3u8_id=None, note=None, errnote=None,
-                              fatal=True):
+                              fatal=True, live=False):
 
         formats = [{
             'format_id': '-'.join(filter(None, [m3u8_id, 'meta'])),
@@ -1132,7 +1139,11 @@ class InfoExtractor(object):
                 if m3u8_id:
                     format_id.append(m3u8_id)
                 last_media_name = last_media.get('NAME') if last_media and last_media.get('TYPE') != 'SUBTITLES' else None
-                format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats)))
+                # Bandwidth of live streams may differ over time thus making
+                # format_id unpredictable. So it's better to keep provided
+                # format_id intact.
+                if not live:
+                    format_id.append(last_media_name if last_media_name else '%d' % (tbr if tbr else len(formats)))
                 f = {
                     'format_id': '-'.join(format_id),
                     'url': format_url(line.strip()),

youtube_dl/extractor/crunchyroll.py

@@ -11,7 +11,6 @@ from math import pow, sqrt, floor
 from .common import InfoExtractor
 from ..compat import (
     compat_etree_fromstring,
-    compat_urllib_parse_unquote,
     compat_urllib_parse_urlencode,
     compat_urllib_request,
     compat_urlparse,
@@ -27,6 +26,7 @@ from ..utils import (
     unified_strdate,
     urlencode_postdata,
     xpath_text,
+    extract_attributes,
 )
 from ..aes import (
     aes_cbc_decrypt,
@@ -306,28 +306,36 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             r'<a[^>]+href="/publisher/[^"]+"[^>]*>([^<]+)</a>', webpage,
             'video_uploader', fatal=False)
 
-        playerdata_url = compat_urllib_parse_unquote(self._html_search_regex(r'"config_url":"([^"]+)', webpage, 'playerdata_url'))
-        playerdata_req = sanitized_Request(playerdata_url)
-        playerdata_req.data = urlencode_postdata({'current_page': webpage_url})
-        playerdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
-        playerdata = self._download_webpage(playerdata_req, video_id, note='Downloading media info')
-
-        stream_id = self._search_regex(r'<media_id>([^<]+)', playerdata, 'stream_id')
-        video_thumbnail = self._search_regex(r'<episode_image_url>([^<]+)', playerdata, 'thumbnail', fatal=False)
-
+        available_fmts = []
+        for a, fmt in re.findall(r'(<a[^>]+token=["\']showmedia\.([0-9]{3,4})p["\'][^>]+>)', webpage):
+            attrs = extract_attributes(a)
+            href = attrs.get('href')
+            if href and '/freetrial' in href:
+                continue
+            available_fmts.append(fmt)
+        if not available_fmts:
+            for p in (r'token=["\']showmedia\.([0-9]{3,4})p"', r'showmedia\.([0-9]{3,4})p'):
+                available_fmts = re.findall(p, webpage)
+                if available_fmts:
+                    break
+        video_encode_ids = []
         formats = []
-        for fmt in re.findall(r'showmedia\.([0-9]{3,4})p', webpage):
+        for fmt in available_fmts:
             stream_quality, stream_format = self._FORMAT_IDS[fmt]
             video_format = fmt + 'p'
             streamdata_req = sanitized_Request(
                 'http://www.crunchyroll.com/xml/?req=RpcApiVideoPlayer_GetStandardConfig&media_id=%s&video_format=%s&video_quality=%s'
-                % (stream_id, stream_format, stream_quality),
+                % (video_id, stream_format, stream_quality),
                 compat_urllib_parse_urlencode({'current_page': url}).encode('utf-8'))
             streamdata_req.add_header('Content-Type', 'application/x-www-form-urlencoded')
             streamdata = self._download_xml(
                 streamdata_req, video_id,
                 note='Downloading media info for %s' % video_format)
             stream_info = streamdata.find('./{default}preload/stream_info')
+            video_encode_id = xpath_text(stream_info, './video_encode_id')
+            if video_encode_id in video_encode_ids:
+                continue
+            video_encode_ids.append(video_encode_id)
             video_url = xpath_text(stream_info, './host')
             video_play_path = xpath_text(stream_info, './file')
             if not video_url or not video_play_path:
@@ -359,6 +367,14 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
                 'ext': 'flv',
             })
             formats.append(format_info)
+        self._sort_formats(formats)
+
+        metadata = self._download_xml(
+            'http://www.crunchyroll.com/xml', video_id,
+            note='Downloading media info', query={
+                'req': 'RpcApiVideoPlayer_GetMediaMetadata',
+                'media_id': video_id,
+            })
 
         subtitles = self.extract_subtitles(video_id, webpage)
 
@@ -366,9 +382,12 @@ Format: Layer, Start, End, Style, Name, MarginL, MarginR, MarginV, Effect, Text
             'id': video_id,
             'title': video_title,
             'description': video_description,
-            'thumbnail': video_thumbnail,
+            'thumbnail': xpath_text(metadata, 'episode_image_url'),
             'uploader': video_uploader,
             'upload_date': video_upload_date,
+            'series': xpath_text(metadata, 'series_title'),
+            'episode': xpath_text(metadata, 'episode_title'),
+            'episode_number': int_or_none(xpath_text(metadata, 'episode_number')),
             'subtitles': subtitles,
             'formats': formats,
         }

youtube_dl/extractor/cwtv.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class CWTVIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?cw(?:tv|seed)\.com/shows/(?:[^/]+/){2}\?play=(?P<id>[a-z0-9]{8}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{12})'
+    _VALID_URL = r'https?://(?:www\.)?cw(?:tv|seed)\.com/(?:shows/)?(?:[^/]+/){2}\?.*\bplay=(?P<id>[a-z0-9]{8}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{4}-[a-z0-9]{12})'
     _TESTS = [{
         'url': 'http://cwtv.com/shows/arrow/legends-of-yesterday/?play=6b15e985-9345-4f60-baf8-56e96be57c63',
         'info_dict': {
@@ -48,6 +48,9 @@ class CWTVIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
+    }, {
+        'url': 'http://cwtv.com/thecw/chroniclesofcisco/?play=8adebe35-f447-465f-ab52-e863506ff6d6',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/dfb.py

@@ -12,39 +12,46 @@ class DFBIE(InfoExtractor):
 
     _TEST = {
         'url': 'http://tv.dfb.de/video/u-19-em-stimmen-zum-spiel-gegen-russland/11633/',
-        # The md5 is different each time
+        'md5': 'ac0f98a52a330f700b4b3034ad240649',
         'info_dict': {
             'id': '11633',
             'display_id': 'u-19-em-stimmen-zum-spiel-gegen-russland',
-            'ext': 'flv',
+            'ext': 'mp4',
             'title': 'U 19-EM: Stimmen zum Spiel gegen Russland',
             'upload_date': '20150714',
         },
     }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        display_id = mobj.group('display_id')
+        display_id, video_id = re.match(self._VALID_URL, url).groups()
 
-        webpage = self._download_webpage(url, display_id)
         player_info = self._download_xml(
             'http://tv.dfb.de/server/hd_video.php?play=%s' % video_id,
             display_id)
         video_info = player_info.find('video')
+        stream_access_url = self._proto_relative_url(video_info.find('url').text.strip())
 
-        f4m_info = self._download_xml(
-            self._proto_relative_url(video_info.find('url').text.strip()), display_id)
-        token_el = f4m_info.find('token')
-        manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth'] + '&hdcore=3.2.0'
-        formats = self._extract_f4m_formats(manifest_url, display_id)
+        formats = []
+        # see http://tv.dfb.de/player/js/ajax.js for the method to extract m3u8 formats
+        for sa_url in (stream_access_url, stream_access_url + '&area=&format=iphone'):
+            stream_access_info = self._download_xml(sa_url, display_id)
+            token_el = stream_access_info.find('token')
+            manifest_url = token_el.attrib['url'] + '?' + 'hdnea=' + token_el.attrib['auth']
+            if '.f4m' in manifest_url:
+                formats.extend(self._extract_f4m_formats(
+                    manifest_url + '&hdcore=3.2.0',
+                    display_id, f4m_id='hds', fatal=False))
+            else:
+                formats.extend(self._extract_m3u8_formats(
+                    manifest_url, display_id, 'mp4',
+                    'm3u8_native', m3u8_id='hls', fatal=False))
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'display_id': display_id,
             'title': video_info.find('title').text,
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'thumbnail': 'http://tv.dfb.de/images/%s_640x360.jpg' % video_id,
             'upload_date': unified_strdate(video_info.find('time_date').text),
             'formats': formats,
         }

youtube_dl/extractor/discovery.py

@@ -33,6 +33,7 @@ class DiscoveryIE(InfoExtractor):
             'duration': 156,
             'timestamp': 1302032462,
             'upload_date': '20110405',
+            'uploader_id': '103207',
         },
         'params': {
             'skip_download': True,  # requires ffmpeg
@@ -54,7 +55,11 @@ class DiscoveryIE(InfoExtractor):
             'upload_date': '20140725',
             'timestamp': 1406246400,
             'duration': 116,
+            'uploader_id': '103207',
         },
+        'params': {
+            'skip_download': True,  # requires ffmpeg
+        }
     }]
 
     def _real_extract(self, url):
@@ -66,13 +71,19 @@ class DiscoveryIE(InfoExtractor):
         entries = []
 
         for idx, video_info in enumerate(info['playlist']):
-            formats = self._extract_m3u8_formats(
-                video_info['src'], display_id, 'mp4', 'm3u8_native', m3u8_id='hls',
-                note='Download m3u8 information for video %d' % (idx + 1))
-            self._sort_formats(formats)
+            subtitles = []
+            caption_url = video_info.get('captionsUrl')
+            if caption_url:
+                subtitles = {
+                    'en': [{
+                        'url': caption_url,
+                    }]
+                }
+
             entries.append({
+                '_type': 'url_transparent',
+                'url': 'http://players.brightcove.net/103207/default_default/index.html?videoId=ref:%s' % video_info['referenceId'],
                 'id': compat_str(video_info['id']),
-                'formats': formats,
                 'title': video_info['title'],
                 'description': video_info.get('description'),
                 'duration': parse_duration(video_info.get('video_length')),
@@ -80,6 +91,7 @@ class DiscoveryIE(InfoExtractor):
                 'thumbnail': video_info.get('thumbnailURL'),
                 'alt_title': video_info.get('secondary_title'),
                 'timestamp': parse_iso8601(video_info.get('publishedDate')),
+                'subtitles': subtitles,
             })
 
         return self.playlist_result(entries, display_id, video_title)

youtube_dl/extractor/douyutv.py

@@ -18,7 +18,7 @@ class DouyuTVIE(InfoExtractor):
             'display_id': 'iseven',
             'ext': 'flv',
             'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
-            'description': 'md5:f34981259a03e980a3c6404190a3ed61',
+            'description': 're:.*m7show@163\.com.*',
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': '7师傅',
             'uploader_id': '431925',
@@ -43,7 +43,7 @@ class DouyuTVIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
-        'skip': 'Romm not found',
+        'skip': 'Room not found',
     }, {
         'url': 'http://www.douyutv.com/17732',
         'info_dict': {
@@ -51,7 +51,7 @@ class DouyuTVIE(InfoExtractor):
             'display_id': '17732',
             'ext': 'flv',
             'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
-            'description': 'md5:f34981259a03e980a3c6404190a3ed61',
+            'description': 're:.*m7show@163\.com.*',
             'thumbnail': 're:^https?://.*\.jpg$',
             'uploader': '7师傅',
             'uploader_id': '431925',
@@ -75,13 +75,28 @@ class DouyuTVIE(InfoExtractor):
             room_id = self._html_search_regex(
                 r'"room_id"\s*:\s*(\d+),', page, 'room id')
 
-        prefix = 'room/%s?aid=android&client_sys=android&time=%d' % (
-            room_id, int(time.time()))
+        config = None
+        # Douyu API sometimes returns error "Unable to load the requested class: eticket_redis_cache"
+        # Retry with different parameters - same parameters cause same errors
+        for i in range(5):
+            prefix = 'room/%s?aid=android&client_sys=android&time=%d' % (
+                room_id, int(time.time()))
+            auth = hashlib.md5((prefix + '1231').encode('ascii')).hexdigest()
 
-        auth = hashlib.md5((prefix + '1231').encode('ascii')).hexdigest()
-        config = self._download_json(
-            'http://www.douyutv.com/api/v1/%s&auth=%s' % (prefix, auth),
-            video_id)
+            config_page = self._download_webpage(
+                'http://www.douyutv.com/api/v1/%s&auth=%s' % (prefix, auth),
+                video_id)
+            try:
+                config = self._parse_json(config_page, video_id, fatal=False)
+            except ExtractorError:
+                # Wait some time before retrying to get a different time() value
+                self._sleep(1, video_id, msg_template='%(video_id)s: Error occurs. '
+                                                      'Waiting for %(timeout)s seconds before retrying')
+                continue
+            else:
+                break
+        if config is None:
+            raise ExtractorError('Unable to fetch API result')
 
         data = config['data']
 

youtube_dl/extractor/eagleplatform.py

@@ -23,7 +23,7 @@ class EaglePlatformIE(InfoExtractor):
     _TESTS = [{
         # http://lenta.ru/news/2015/03/06/navalny/
         'url': 'http://lentaru.media.eagleplatform.com/index/player?player=new&record_id=227304&player_template_id=5201',
-        'md5': '881ee8460e1b7735a8be938e2ffb362b',
+        # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
         'info_dict': {
             'id': '227304',
             'ext': 'mp4',
@@ -109,8 +109,11 @@ class EaglePlatformIE(InfoExtractor):
             mobj = re.search('/([^/]+)/index\.m3u8', m3u8_format['url'])
             if mobj:
                 http_format = m3u8_format.copy()
+                video_url = mp4_url.replace(mp4_url_basename, mobj.group(1))
+                if not self._is_valid_url(video_url, video_id):
+                    continue
                 http_format.update({
-                    'url': mp4_url.replace(mp4_url_basename, mobj.group(1)),
+                    'url': video_url,
                     'format_id': m3u8_format['format_id'].replace('hls', 'http'),
                     'protocol': 'http',
                 })

youtube_dl/extractor/extractors.py

@@ -400,7 +400,6 @@ from .macgamestore import MacGameStoreIE
 from .mailru import MailRuIE
 from .makerschannel import MakersChannelIE
 from .makertv import MakerTVIE
-from .malemotion import MalemotionIE
 from .matchtv import MatchTVIE
 from .mdr import MDRIE
 from .metacafe import MetacafeIE
@@ -439,8 +438,7 @@ from .mtv import (
 )
 from .muenchentv import MuenchenTVIE
 from .musicplayon import MusicPlayOnIE
-from .muzu import MuzuTVIE
-from .mwave import MwaveIE
+from .mwave import MwaveIE, MwaveMeetGreetIE
 from .myspace import MySpaceIE, MySpaceAlbumIE
 from .myspass import MySpassIE
 from .myvi import MyviIE
@@ -659,7 +657,6 @@ from .screenwavemedia import ScreenwaveMediaIE, TeamFourIE
 from .senateisvp import SenateISVPIE
 from .servingsys import ServingSysIE
 from .sexu import SexuIE
-from .sexykarma import SexyKarmaIE
 from .shahid import ShahidIE
 from .shared import SharedIE
 from .sharesix import ShareSixIE
@@ -676,10 +673,6 @@ from .smotri import (
     SmotriUserIE,
     SmotriBroadcastIE,
 )
-from .snagfilms import (
-    SnagFilmsIE,
-    SnagFilmsEmbedIE,
-)
 from .snotr import SnotrIE
 from .sohu import SohuIE
 from .soundcloud import (
@@ -731,7 +724,10 @@ from .svt import (
 from .swrmediathek import SWRMediathekIE
 from .syfy import SyfyIE
 from .sztvhu import SztvHuIE
-from .tagesschau import TagesschauIE
+from .tagesschau import (
+    TagesschauPlayerIE,
+    TagesschauIE,
+)
 from .tapely import TapelyIE
 from .tass import TassIE
 from .tdslifeway import TDSLifewayIE
@@ -882,6 +878,10 @@ from .vidme import (
 )
 from .vidzi import VidziIE
 from .vier import VierIE, VierVideosIE
+from .viewlift import (
+    ViewLiftIE,
+    ViewLiftEmbedIE,
+)
 from .viewster import ViewsterIE
 from .viidea import ViideaIE
 from .vimeo import (
@@ -920,6 +920,7 @@ from .vulture import VultureIE
 from .walla import WallaIE
 from .washingtonpost import WashingtonPostIE
 from .wat import WatIE
+from .watchindianporn import WatchIndianPornIE
 from .wdr import (
     WDRIE,
     WDRMobileIE,
@@ -943,6 +944,12 @@ from .xhamster import (
     XHamsterIE,
     XHamsterEmbedIE,
 )
+from .xiami import (
+    XiamiSongIE,
+    XiamiAlbumIE,
+    XiamiArtistIE,
+    XiamiCollectionIE
+)
 from .xminus import XMinusIE
 from .xnxx import XNXXIE
 from .xstream import XstreamIE

youtube_dl/extractor/funimation.py

@@ -2,6 +2,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_HTTPError,
+    compat_urllib_parse_unquote_plus,
+)
 from ..utils import (
     clean_html,
     determine_ext,
@@ -27,6 +31,7 @@ class FunimationIE(InfoExtractor):
             'description': 'md5:1769f43cd5fc130ace8fd87232207892',
             'thumbnail': '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': {
@@ -37,6 +42,7 @@ class FunimationIE(InfoExtractor):
             'description': 'md5:b602bdc15eef4c9bbb201bb6e6a4a2dd',
             'thumbnail': 're:https?://.*\.jpg',
         },
+        'skip': 'Access without user interaction is forbidden by CloudFlare',
     }, {
         'url': 'http://www.funimation.com/shows/attack-on-titan-junior-high/videos/promotional/broadcast-dub-preview',
         'info_dict': {
@@ -47,8 +53,36 @@ class FunimationIE(InfoExtractor):
             'description': 'md5:f8ec49c0aff702a7832cd81b8a44f803',
             'thumbnail': 're:https?://.*\.(?:jpg|png)',
         },
+        'skip': 'Access without user interaction is forbidden by CloudFlare',
     }]
 
+    _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:
+            ci_session = compat_urllib_parse_unquote_plus(ci_session_cookie.value)
+            # ci_session is a string serialized by PHP function serialize()
+            # This case is simple enough to use regular expressions only
+            return self._search_regex(
+                r'"user_agent";s:\d+:"([^"]+)"', ci_session, 'user agent',
+                default=None)
+
     def _login(self):
         (username, password) = self._get_login_info()
         if username is None:
@@ -57,8 +91,11 @@ class FunimationIE(InfoExtractor):
             'email_field': username,
             'password_field': password,
         })
-        login_request = sanitized_Request('http://www.funimation.com/login', data, headers={
-            'User-Agent': 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0',
+        user_agent = self._extract_cloudflare_session_ua(self._LOGIN_URL)
+        if not user_agent:
+            user_agent = 'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:42.0) Gecko/20100101 Firefox/42.0'
+        login_request = sanitized_Request(self._LOGIN_URL, data, headers={
+            'User-Agent': user_agent,
             'Content-Type': 'application/x-www-form-urlencoded'
         })
         login_page = self._download_webpage(
@@ -103,11 +140,16 @@ class FunimationIE(InfoExtractor):
             ('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)
+                request, display_id,
+                'Downloading %s webpage' % kind if kind else 'Downloading webpage')
 
             playlist = self._parse_json(
                 self._search_regex(

youtube_dl/extractor/generic.py

@@ -51,7 +51,7 @@ from .tnaflix import TNAFlixNetworkEmbedIE
 from .vimeo import VimeoIE
 from .dailymotion import DailymotionCloudIE
 from .onionstudios import OnionStudiosIE
-from .snagfilms import SnagFilmsEmbedIE
+from .viewlift import ViewLiftEmbedIE
 from .screenwavemedia import ScreenwaveMediaIE
 from .mtv import MTVServicesEmbeddedIE
 from .pladform import PladformIE
@@ -887,6 +887,7 @@ class GenericIE(InfoExtractor):
         # Eagle.Platform embed (generic URL)
         {
             'url': 'http://lenta.ru/news/2015/03/06/navalny/',
+            # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
             'info_dict': {
                 'id': '227304',
                 'ext': 'mp4',
@@ -901,6 +902,7 @@ class GenericIE(InfoExtractor):
         # ClipYou (Eagle.Platform) embed (custom URL)
         {
             'url': 'http://muz-tv.ru/play/7129/',
+            # Not checking MD5 as sometimes the direct HTTP link results in 404 and HLS is used
             'info_dict': {
                 'id': '12820',
                 'ext': 'mp4',
@@ -1922,10 +1924,10 @@ class GenericIE(InfoExtractor):
         if onionstudios_url:
             return self.url_result(onionstudios_url)
 
-        # Look for SnagFilms embeds
-        snagfilms_url = SnagFilmsEmbedIE._extract_url(webpage)
-        if snagfilms_url:
-            return self.url_result(snagfilms_url)
+        # Look for ViewLift embeds
+        viewlift_url = ViewLiftEmbedIE._extract_url(webpage)
+        if viewlift_url:
+            return self.url_result(viewlift_url)
 
         # Look for JWPlatform embeds
         jwplatform_url = JWPlatformIE._extract_url(webpage)

youtube_dl/extractor/malemotion.py

@@ -1,46 +0,0 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
-
-
-class MalemotionIE(InfoExtractor):
-    _VALID_URL = r'https?://malemotion\.com/video/(.+?)\.(?P<id>.+?)(#|$)'
-    _TEST = {
-        'url': 'http://malemotion.com/video/bete-de-concours.ltc',
-        'md5': '3013e53a0afbde2878bc39998c33e8a5',
-        'info_dict': {
-            'id': 'ltc',
-            'ext': 'mp4',
-            'title': 'Bête de Concours',
-            'age_limit': 18,
-        },
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-
-        video_url = compat_urllib_parse_unquote(self._search_regex(
-            r'<source type="video/mp4" src="(.+?)"', webpage, 'video URL'))
-        video_title = self._html_search_regex(
-            r'<title>(.*?)</title', webpage, 'title')
-        video_thumbnail = self._search_regex(
-            r'<video .+?poster="(.+?)"', webpage, 'thumbnail', fatal=False)
-
-        formats = [{
-            'url': video_url,
-            'ext': 'mp4',
-            'format_id': 'mp4',
-            'preference': 1,
-        }]
-        self._sort_formats(formats)
-
-        return {
-            'id': video_id,
-            'formats': formats,
-            'title': video_title,
-            'thumbnail': video_thumbnail,
-            'age_limit': 18,
-        }

youtube_dl/extractor/mitele.py

@@ -15,9 +15,9 @@ class MiTeleIE(InfoExtractor):
     IE_DESC = 'mitele.es'
     _VALID_URL = r'https?://www\.mitele\.es/[^/]+/[^/]+/[^/]+/(?P<id>[^/]+)/'
 
-    _TESTS = [{
+    _TEST = {
         'url': 'http://www.mitele.es/programas-tv/diario-de/la-redaccion/programa-144/',
-        'md5': '0ff1a13aebb35d9bc14081ff633dd324',
+        # MD5 is unstable
         'info_dict': {
             'id': '0NF1jJnxS1Wu3pHrmvFyw2',
             'display_id': 'programa-144',
@@ -27,7 +27,7 @@ class MiTeleIE(InfoExtractor):
             'thumbnail': 're:(?i)^https?://.*\.jpg$',
             'duration': 2913,
         },
-    }]
+    }
 
     def _real_extract(self, url):
         display_id = self._match_id(url)

youtube_dl/extractor/muzu.py

@@ -1,63 +0,0 @@
-from __future__ import unicode_literals
-
-from .common import InfoExtractor
-from ..compat import compat_urllib_parse_urlencode
-
-
-class MuzuTVIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.muzu\.tv/(.+?)/(.+?)/(?P<id>\d+)'
-    IE_NAME = 'muzu.tv'
-
-    _TEST = {
-        'url': 'http://www.muzu.tv/defected/marcashken-featuring-sos-cat-walk-original-mix-music-video/1981454/',
-        'md5': '98f8b2c7bc50578d6a0364fff2bfb000',
-        'info_dict': {
-            'id': '1981454',
-            'ext': 'mp4',
-            'title': 'Cat Walk (Original Mix)',
-            'description': 'md5:90e868994de201b2570e4e5854e19420',
-            'uploader': 'MarcAshken featuring SOS',
-        },
-    }
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        info_data = compat_urllib_parse_urlencode({
-            'format': 'json',
-            'url': url,
-        })
-        info = self._download_json(
-            'http://www.muzu.tv/api/oembed/?%s' % info_data,
-            video_id, 'Downloading video info')
-
-        player_info = self._download_json(
-            'http://player.muzu.tv/player/playerInit?ai=%s' % video_id,
-            video_id, 'Downloading player info')
-        video_info = player_info['videos'][0]
-        for quality in ['1080', '720', '480', '360']:
-            if video_info.get('v%s' % quality):
-                break
-
-        data = compat_urllib_parse_urlencode({
-            'ai': video_id,
-            # Even if each time you watch a video the hash changes,
-            # it seems to work for different videos, and it will work
-            # even if you use any non empty string as a hash
-            'viewhash': 'VBNff6djeV4HV5TRPW5kOHub2k',
-            'device': 'web',
-            'qv': quality,
-        })
-        video_url_info = self._download_json(
-            'http://player.muzu.tv/player/requestVideo?%s' % data,
-            video_id, 'Downloading video url')
-        video_url = video_url_info['url']
-
-        return {
-            'id': video_id,
-            'title': info['title'],
-            'url': video_url,
-            'thumbnail': info['thumbnail_url'],
-            'description': info['description'],
-            'uploader': info['author_name'],
-        }

youtube_dl/extractor/mwave.py

@@ -10,6 +10,7 @@ from ..utils import (
 
 class MwaveIE(InfoExtractor):
     _VALID_URL = r'https?://mwave\.interest\.me/mnettv/videodetail\.m\?searchVideoDetailVO\.clip_id=(?P<id>[0-9]+)'
+    _URL_TEMPLATE = 'http://mwave.interest.me/mnettv/videodetail.m?searchVideoDetailVO.clip_id=%s'
     _TEST = {
         'url': 'http://mwave.interest.me/mnettv/videodetail.m?searchVideoDetailVO.clip_id=168859',
         # md5 is unstable
@@ -56,3 +57,28 @@ class MwaveIE(InfoExtractor):
             'view_count': int_or_none(vod_info.get('hit')),
             'formats': formats,
         }
+
+
+class MwaveMeetGreetIE(InfoExtractor):
+    _VALID_URL = r'https?://mwave\.interest\.me/meetgreet/view/(?P<id>\d+)'
+    _TEST = {
+        'url': 'http://mwave.interest.me/meetgreet/view/256',
+        'info_dict': {
+            'id': '173294',
+            'ext': 'flv',
+            'title': '[MEET&GREET] Park BoRam',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'uploader': 'Mwave',
+            'duration': 3634,
+            'view_count': int,
+        }
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        clip_id = self._html_search_regex(
+            r'<iframe[^>]+src="/mnettv/ifr_clip\.m\?searchVideoDetailVO\.clip_id=(\d+)',
+            webpage, 'clip ID')
+        clip_url = MwaveIE._URL_TEMPLATE % clip_id
+        return self.url_result(clip_url, 'Mwave', clip_id)

youtube_dl/extractor/normalboots.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from .screenwavemedia import ScreenwaveMediaIE
 
 from ..utils import (
     unified_strdate,
@@ -12,7 +13,6 @@ class NormalbootsIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?normalboots\.com/video/(?P<id>[0-9a-z-]*)/?$'
     _TEST = {
         'url': 'http://normalboots.com/video/home-alone-games-jontron/',
-        'md5': '8bf6de238915dd501105b44ef5f1e0f6',
         'info_dict': {
             'id': 'home-alone-games-jontron',
             'ext': 'mp4',
@@ -22,9 +22,10 @@ class NormalbootsIE(InfoExtractor):
             'upload_date': '20140125',
         },
         'params': {
-            # rtmp download
+            # m3u8 download
             'skip_download': True,
         },
+        'add_ie': ['ScreenwaveMedia'],
     }
 
     def _real_extract(self, url):
@@ -38,16 +39,15 @@ class NormalbootsIE(InfoExtractor):
             r'<span style="text-transform:uppercase; font-size:inherit;">[A-Za-z]+, (?P<date>.*)</span>',
             webpage, 'date', fatal=False))
 
-        player_url = self._html_search_regex(
-            r'<iframe\swidth="[0-9]+"\sheight="[0-9]+"\ssrc="(?P<url>[\S]+)"',
-            webpage, 'player url')
-        player_page = self._download_webpage(player_url, video_id)
-        video_url = self._html_search_regex(
-            r"file:\s'(?P<file>[^']+\.mp4)'", player_page, 'file')
+        screenwavemedia_url = self._html_search_regex(
+            ScreenwaveMediaIE.EMBED_PATTERN, webpage, 'screenwave URL',
+            group='url')
 
         return {
+            '_type': 'url_transparent',
             'id': video_id,
-            'url': video_url,
+            'url': screenwavemedia_url,
+            'ie_key': ScreenwaveMediaIE.ie_key(),
             'title': self._og_search_title(webpage),
             'description': self._og_search_description(webpage),
             'thumbnail': self._og_search_thumbnail(webpage),

youtube_dl/extractor/nrk.py

@@ -23,7 +23,7 @@ class NRKIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://www.nrk.no/video/PS*150533',
-            'md5': 'bccd850baebefe23b56d708a113229c2',
+            # MD5 is unstable
             'info_dict': {
                 'id': '150533',
                 'ext': 'flv',
@@ -34,7 +34,7 @@ class NRKIE(InfoExtractor):
         },
         {
             'url': 'http://www.nrk.no/video/PS*154915',
-            'md5': '0b1493ba1aae7d9579a5ad5531bc395a',
+            # MD5 is unstable
             'info_dict': {
                 'id': '154915',
                 'ext': 'flv',

youtube_dl/extractor/nuvid.py

@@ -5,8 +5,6 @@ import re
 from .common import InfoExtractor
 from ..utils import (
     parse_duration,
-    sanitized_Request,
-    unified_strdate,
 )
 
 
@@ -20,7 +18,6 @@ class NuvidIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Horny babes show their awesome bodeis and',
             'duration': 129,
-            'upload_date': '20140508',
             'age_limit': 18,
         }
     }
@@ -28,28 +25,31 @@ class NuvidIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        formats = []
+        page_url = 'http://m.nuvid.com/video/%s' % video_id
+        webpage = self._download_webpage(
+            page_url, video_id, 'Downloading video page')
+        # When dwnld_speed exists and has a value larger than the MP4 file's
+        # bitrate, Nuvid returns the MP4 URL
+        # It's unit is 100bytes/millisecond, see mobile-nuvid-min.js for the algorithm
+        self._set_cookie('nuvid.com', 'dwnld_speed', '10.0')
+        mp4_webpage = self._download_webpage(
+            page_url, video_id, 'Downloading video page for MP4 format')
 
-        for dwnld_speed, format_id in [(0, '3gp'), (5, 'mp4')]:
-            request = sanitized_Request(
-                'http://m.nuvid.com/play/%s' % video_id)
-            request.add_header('Cookie', 'skip_download_page=1; dwnld_speed=%d; adv_show=1' % dwnld_speed)
-            webpage = self._download_webpage(
-                request, video_id, 'Downloading %s page' % format_id)
-            video_url = self._html_search_regex(
-                r'<a\s+href="([^"]+)"\s+class="b_link">', webpage, '%s video URL' % format_id, fatal=False)
-            if not video_url:
-                continue
+        html5_video_re = r'(?s)<(?:video|audio)[^<]*(?:>.*?<source[^>]*)?\s+src=["\'](.*?)["\']',
+        video_url = self._html_search_regex(html5_video_re, webpage, video_id)
+        mp4_video_url = self._html_search_regex(html5_video_re, mp4_webpage, video_id)
+        formats = [{
+            'url': video_url,
+        }]
+        if mp4_video_url != video_url:
             formats.append({
-                'url': video_url,
-                'format_id': format_id,
+                'url': mp4_video_url,
             })
 
-        webpage = self._download_webpage(
-            'http://m.nuvid.com/video/%s' % video_id, video_id, 'Downloading video page')
         title = self._html_search_regex(
             [r'<span title="([^"]+)">',
-             r'<div class="thumb-holder video">\s*<h5[^>]*>([^<]+)</h5>'], webpage, 'title').strip()
+             r'<div class="thumb-holder video">\s*<h5[^>]*>([^<]+)</h5>',
+             r'<span[^>]+class="title_thumb">([^<]+)</span>'], webpage, 'title').strip()
         thumbnails = [
             {
                 'url': thumb_url,
@@ -57,9 +57,8 @@ class NuvidIE(InfoExtractor):
         ]
         thumbnail = thumbnails[0]['url'] if thumbnails else None
         duration = parse_duration(self._html_search_regex(
-            r'<i class="fa fa-clock-o"></i>\s*(\d{2}:\d{2})', webpage, 'duration', fatal=False))
-        upload_date = unified_strdate(self._html_search_regex(
-            r'<i class="fa fa-user"></i>\s*(\d{4}-\d{2}-\d{2})', webpage, 'upload date', fatal=False))
+            [r'<i class="fa fa-clock-o"></i>\s*(\d{2}:\d{2})',
+             r'<span[^>]+class="view_time">([^<]+)</span>'], webpage, 'duration', fatal=False))
 
         return {
             'id': video_id,
@@ -67,7 +66,6 @@ class NuvidIE(InfoExtractor):
             'thumbnails': thumbnails,
             'thumbnail': thumbnail,
             'duration': duration,
-            'upload_date': upload_date,
             'age_limit': 18,
             'formats': formats,
         }

youtube_dl/extractor/odnoklassniki.py

@@ -2,7 +2,11 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_urllib_parse_unquote
+from ..compat import (
+    compat_parse_qs,
+    compat_urllib_parse_unquote,
+    compat_urllib_parse_urlparse,
+)
 from ..utils import (
     ExtractorError,
     unified_strdate,
@@ -32,7 +36,7 @@ class OdnoklassnikiIE(InfoExtractor):
         'skip': 'Video has been blocked',
     }, {
         # metadataUrl
-        'url': 'http://ok.ru/video/63567059965189-0',
+        'url': 'http://ok.ru/video/63567059965189-0?fromTime=5',
         'md5': '9676cf86eff5391d35dea675d224e131',
         'info_dict': {
             'id': '63567059965189-0',
@@ -44,6 +48,7 @@ class OdnoklassnikiIE(InfoExtractor):
             'uploader': '☭ Андрей Мещанинов ☭',
             'like_count': int,
             'age_limit': 0,
+            'start_time': 5,
         },
     }, {
         # YouTube embed (metadataUrl, provider == USER_YOUTUBE)
@@ -60,6 +65,22 @@ class OdnoklassnikiIE(InfoExtractor):
             'uploader': 'Алина П',
             'age_limit': 0,
         },
+    }, {
+        # YouTube embed (metadata, provider == USER_YOUTUBE, no metadata.movie.title field)
+        'url': 'http://ok.ru/video/62036049272859-0',
+        'info_dict': {
+            'id': '62036049272859-0',
+            'ext': 'mp4',
+            'title': 'МУЗЫКА     ДОЖДЯ .',
+            'description': 'md5:6f1867132bd96e33bf53eda1091e8ed0',
+            'upload_date': '20120106',
+            'uploader_id': '473534735899',
+            'uploader': 'МARINA D',
+            'age_limit': 0,
+        },
+        'params': {
+            'skip_download': True,
+        },
     }, {
         'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452',
         'only_matching': True,
@@ -78,6 +99,9 @@ class OdnoklassnikiIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
+        start_time = int_or_none(compat_parse_qs(
+            compat_urllib_parse_urlparse(url).query).get('fromTime', [None])[0])
+
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(
@@ -106,7 +130,14 @@ class OdnoklassnikiIE(InfoExtractor):
                 video_id, 'Downloading metadata JSON')
 
         movie = metadata['movie']
-        title = movie['title']
+
+        # Some embedded videos may not contain title in movie dict (e.g.
+        # http://ok.ru/video/62036049272859-0) thus we allow missing title
+        # here and it's going to be extracted later by an extractor that
+        # will process the actual embed.
+        provider = metadata.get('provider')
+        title = movie['title'] if provider == 'UPLOADED_ODKL' else movie.get('title')
+
         thumbnail = movie.get('poster')
         duration = int_or_none(movie.get('duration'))
 
@@ -135,9 +166,10 @@ class OdnoklassnikiIE(InfoExtractor):
             'uploader_id': uploader_id,
             'like_count': like_count,
             'age_limit': age_limit,
+            'start_time': start_time,
         }
 
-        if metadata.get('provider') == 'USER_YOUTUBE':
+        if provider == 'USER_YOUTUBE':
             info.update({
                 '_type': 'url_transparent',
                 'url': movie['contentId'],

youtube_dl/extractor/ooyala.py

@@ -96,6 +96,8 @@ class OoyalaIE(OoyalaBaseIE):
                 'description': 'How badly damaged does a drive have to be to defeat Russell and his crew? Apparently, smashed to bits.',
                 'duration': 853.386,
             },
+            # The video in the original webpage now uses PlayWire
+            'skip': 'Ooyala said: movie expired',
         }, {
             # Only available for ipad
             'url': 'http://player.ooyala.com/player.js?embedCode=x1b3lqZDq9y_7kMyC2Op5qo-p077tXD0',

youtube_dl/extractor/openload.py

@@ -6,8 +6,10 @@ import re
 from .common import InfoExtractor
 from ..compat import compat_chr
 from ..utils import (
+    determine_ext,
     encode_base_n,
     ExtractorError,
+    mimetype2ext,
 )
 
 
@@ -29,6 +31,11 @@ class OpenloadIE(InfoExtractor):
     }, {
         'url': 'https://openload.io/f/ZAn6oz-VZGE/',
         'only_matching': True,
+    }, {
+        # unavailable via https://openload.co/f/Sxz5sADo82g/, different layout
+        # for title and ext
+        'url': 'https://openload.co/embed/Sxz5sADo82g/',
+        'only_matching': True,
     }]
 
     @staticmethod
@@ -96,12 +103,25 @@ class OpenloadIE(InfoExtractor):
             r'<video[^>]+>\s*<script[^>]+>([^<]+)</script>',
             webpage, 'JS code')
 
+        decoded = self.openload_decode(code)
+
         video_url = self._search_regex(
-            r'return\s+"(https?://[^"]+)"', self.openload_decode(code), 'video URL')
+            r'return\s+"(https?://[^"]+)"', decoded, 'video URL')
+
+        title = self._og_search_title(webpage, default=None) or self._search_regex(
+            r'<span[^>]+class=["\']title["\'][^>]*>([^<]+)', webpage,
+            'title', default=None) or self._html_search_meta(
+            'description', webpage, 'title', fatal=True)
+
+        ext = mimetype2ext(self._search_regex(
+            r'window\.vt\s*=\s*(["\'])(?P<mimetype>.+?)\1', decoded,
+            'mimetype', default=None, group='mimetype')) or determine_ext(
+            video_url, 'mp4')
 
         return {
             'id': video_id,
-            'title': self._og_search_title(webpage),
-            'thumbnail': self._og_search_thumbnail(webpage),
+            'title': title,
+            'ext': ext,
+            'thumbnail': self._og_search_thumbnail(webpage, default=None),
             'url': video_url,
         }

youtube_dl/extractor/orf.py

@@ -185,6 +185,7 @@ class ORFFM4IE(InfoExtractor):
             'timestamp': 1452456073,
             'upload_date': '20160110',
         },
+        'skip': 'Live streams on FM4 got deleted soon',
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/pbs.py

@@ -196,7 +196,7 @@ class PBSIE(InfoExtractor):
     _TESTS = [
         {
             'url': 'http://www.pbs.org/tpt/constitution-usa-peter-sagal/watch/a-more-perfect-union/',
-            'md5': 'ce1888486f0908d555a8093cac9a7362',
+            'md5': '173dc391afd361fa72eab5d3d918968d',
             'info_dict': {
                 'id': '2365006249',
                 'ext': 'mp4',
@@ -204,13 +204,10 @@ class PBSIE(InfoExtractor):
                 'description': 'md5:36f341ae62e251b8f5bd2b754b95a071',
                 'duration': 3190,
             },
-            'params': {
-                'skip_download': True,  # requires ffmpeg
-            },
         },
         {
             'url': 'http://www.pbs.org/wgbh/pages/frontline/losing-iraq/',
-            'md5': '143c98aa54a346738a3d78f54c925321',
+            'md5': '6f722cb3c3982186d34b0f13374499c7',
             'info_dict': {
                 'id': '2365297690',
                 'ext': 'mp4',
@@ -218,9 +215,6 @@ class PBSIE(InfoExtractor):
                 'description': 'md5:4d3eaa01f94e61b3e73704735f1196d9',
                 'duration': 5050,
             },
-            'params': {
-                'skip_download': True,  # requires ffmpeg
-            }
         },
         {
             'url': 'http://www.pbs.org/newshour/bb/education-jan-june12-cyberschools_02-23/',
@@ -244,9 +238,6 @@ class PBSIE(InfoExtractor):
                 'duration': 6559,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
-            'params': {
-                'skip_download': True,  # requires ffmpeg
-            },
         },
         {
             'url': 'http://www.pbs.org/wgbh/nova/earth/killer-typhoon.html',
@@ -262,9 +253,6 @@ class PBSIE(InfoExtractor):
                 'upload_date': '20140122',
                 'age_limit': 10,
             },
-            'params': {
-                'skip_download': True,  # requires ffmpeg
-            },
         },
         {
             'url': 'http://www.pbs.org/wgbh/pages/frontline/united-states-of-secrets/',
@@ -290,6 +278,7 @@ class PBSIE(InfoExtractor):
         },
         {
             'url': 'http://www.pbs.org/video/2365245528/',
+            'md5': '115223d41bd55cda8ae5cd5ed4e11497',
             'info_dict': {
                 'id': '2365245528',
                 'display_id': '2365245528',
@@ -299,15 +288,13 @@ class PBSIE(InfoExtractor):
                 'duration': 6851,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
-            'params': {
-                'skip_download': True,  # requires ffmpeg
-            },
         },
         {
             # Video embedded in iframe containing angle brackets as attribute's value (e.g.
             # "<iframe style='position: absolute;<br />\ntop: 0; left: 0;' ...", see
             # https://github.com/rg3/youtube-dl/issues/7059)
             'url': 'http://www.pbs.org/food/features/a-chefs-life-season-3-episode-5-prickly-business/',
+            'md5': '84ced42850d78f1d4650297356e95e6f',
             'info_dict': {
                 'id': '2365546844',
                 'display_id': 'a-chefs-life-season-3-episode-5-prickly-business',
@@ -317,9 +304,6 @@ class PBSIE(InfoExtractor):
                 'duration': 1480,
                 'thumbnail': 're:^https?://.*\.jpg$',
             },
-            'params': {
-                'skip_download': True,  # requires ffmpeg
-            },
         },
         {
             # Frontline video embedded via flp2012.js
@@ -340,6 +324,7 @@ class PBSIE(InfoExtractor):
         {
             # Serves hd only via wigget/partnerplayer page
             'url': 'http://www.pbs.org/video/2365641075/',
+            'md5': 'acfd4c400b48149a44861cb16dd305cf',
             'info_dict': {
                 'id': '2365641075',
                 'ext': 'mp4',
@@ -348,9 +333,6 @@ class PBSIE(InfoExtractor):
                 'thumbnail': 're:^https?://.*\.jpg$',
                 'formats': 'mincount:8',
             },
-            'params': {
-                'skip_download': True,  # requires ffmpeg
-            },
         },
         {
             'url': 'http://player.pbs.org/widget/partnerplayer/2365297708/?start=0&end=0&chapterbar=false&endscreen=false&topbar=true',
@@ -494,6 +476,7 @@ class PBSIE(InfoExtractor):
                         info = video_info
 
         formats = []
+        http_url = None
         for num, redirect in enumerate(redirects):
             redirect_id = redirect.get('eeid')
 
@@ -514,13 +497,32 @@ class PBSIE(InfoExtractor):
 
             if determine_ext(format_url) == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
-                    format_url, display_id, 'mp4', preference=1, m3u8_id='hls'))
+                    format_url, display_id, 'mp4', m3u8_id='hls', fatal=False))
             else:
                 formats.append({
                     'url': format_url,
                     'format_id': redirect_id,
                 })
+                if re.search(r'^https?://.*(?:\d+k|baseline)', format_url):
+                    http_url = format_url
         self._remove_duplicate_formats(formats)
+        m3u8_formats = list(filter(
+            lambda f: f.get('protocol') == 'm3u8' and f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
+            formats))
+        if http_url:
+            for m3u8_format in m3u8_formats:
+                bitrate = self._search_regex(r'(\d+k)', m3u8_format['url'], 'bitrate', default=None)
+                # extract only the formats that we know that they will be available as http format.
+                # https://projects.pbs.org/confluence/display/coveapi/COVE+Video+Specifications
+                if not bitrate or bitrate not in ('400k', '800k', '1200k', '2500k'):
+                    continue
+                f = m3u8_format.copy()
+                f.update({
+                    'url': re.sub(r'\d+k|baseline', bitrate, http_url),
+                    'format_id': m3u8_format['format_id'].replace('hls', 'http'),
+                    'protocol': 'http',
+                })
+                formats.append(f)
         self._sort_formats(formats)
 
         rating_str = info.get('rating')
@@ -535,6 +537,19 @@ class PBSIE(InfoExtractor):
                 'ext': 'ttml',
                 'url': closed_captions_url,
             }]
+            mobj = re.search(r'/(\d+)_Encoded\.dfxp', closed_captions_url)
+            if mobj:
+                ttml_caption_suffix, ttml_caption_id = mobj.group(0, 1)
+                ttml_caption_id = int(ttml_caption_id)
+                subtitles['en'].extend([{
+                    'url': closed_captions_url.replace(
+                        ttml_caption_suffix, '/%d_Encoded.srt' % (ttml_caption_id + 1)),
+                    'ext': 'srt',
+                }, {
+                    'url': closed_captions_url.replace(
+                        ttml_caption_suffix, '/%d_Encoded.vtt' % (ttml_caption_id + 2)),
+                    'ext': 'vtt',
+                }])
 
         # info['title'] is often incomplete (e.g. 'Full Episode', 'Episode 5', etc)
         # Try turning it to 'program - title' naming scheme if possible

youtube_dl/extractor/rtlnl.py

@@ -20,18 +20,19 @@ class RtlNlIE(InfoExtractor):
         (?P<id>[0-9a-f-]+)'''
 
     _TESTS = [{
-        'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/6e4203a6-0a5e-3596-8424-c599a59e0677',
-        'md5': 'cc16baa36a6c169391f0764fa6b16654',
+        'url': 'http://www.rtlxl.nl/#!/rtl-nieuws-132237/82b1aad1-4a14-3d7b-b554-b0aed1b2c416',
+        'md5': '473d1946c1fdd050b2c0161a4b13c373',
         'info_dict': {
-            'id': '6e4203a6-0a5e-3596-8424-c599a59e0677',
+            'id': '82b1aad1-4a14-3d7b-b554-b0aed1b2c416',
             'ext': 'mp4',
-            'title': 'RTL Nieuws - Laat',
-            'description': 'md5:6b61f66510c8889923b11f2778c72dc5',
-            'timestamp': 1408051800,
-            'upload_date': '20140814',
-            'duration': 576.880,
+            'title': 'RTL Nieuws',
+            'description': 'md5:d41d8cd98f00b204e9800998ecf8427e',
+            'timestamp': 1461951000,
+            'upload_date': '20160429',
+            'duration': 1167.96,
         },
     }, {
+        # best format avaialble a3t
         'url': 'http://www.rtl.nl/system/videoplayer/derden/rtlnieuws/video_embed.html#uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed/autoplay=false',
         'md5': 'dea7474214af1271d91ef332fb8be7ea',
         'info_dict': {
@@ -39,18 +40,19 @@ class RtlNlIE(InfoExtractor):
             'ext': 'mp4',
             'timestamp': 1424039400,
             'title': 'RTL Nieuws - Nieuwe beelden Kopenhagen: chaos direct na aanslag',
-            'thumbnail': 're:^https?://screenshots\.rtl\.nl/system/thumb/sz=[0-9]+x[0-9]+/uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed$',
+            'thumbnail': 're:^https?://screenshots\.rtl\.nl/(?:[^/]+/)*sz=[0-9]+x[0-9]+/uuid=84ae5571-ac25-4225-ae0c-ef8d9efb2aed$',
             'upload_date': '20150215',
             'description': 'Er zijn nieuwe beelden vrijgegeven die vlak na de aanslag in Kopenhagen zijn gemaakt. Op de video is goed te zien hoe omstanders zich bekommeren om één van de slachtoffers, terwijl de eerste agenten ter plaatse komen.',
         }
     }, {
         # empty synopsis and missing episodes (see https://github.com/rg3/youtube-dl/issues/6275)
+        # best format available nettv
         'url': 'http://www.rtl.nl/system/videoplayer/derden/rtlnieuws/video_embed.html#uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a/autoplay=false',
         'info_dict': {
             'id': 'f536aac0-1dc3-4314-920e-3bd1c5b3811a',
             'ext': 'mp4',
             'title': 'RTL Nieuws - Meer beelden van overval juwelier',
-            'thumbnail': 're:^https?://screenshots\.rtl\.nl/system/thumb/sz=[0-9]+x[0-9]+/uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a$',
+            'thumbnail': 're:^https?://screenshots\.rtl\.nl/(?:[^/]+/)*sz=[0-9]+x[0-9]+/uuid=f536aac0-1dc3-4314-920e-3bd1c5b3811a$',
             'timestamp': 1437233400,
             'upload_date': '20150718',
             'duration': 30.474,
@@ -94,22 +96,46 @@ class RtlNlIE(InfoExtractor):
         videopath = material['videopath']
         m3u8_url = meta.get('videohost', 'http://manifest.us.rtl.nl') + videopath
 
-        formats = self._extract_m3u8_formats(m3u8_url, uuid, ext='mp4')
+        formats = self._extract_m3u8_formats(
+            m3u8_url, uuid, 'mp4', m3u8_id='hls', fatal=False)
 
         video_urlpart = videopath.split('/adaptive/')[1][:-5]
         PG_URL_TEMPLATE = 'http://pg.us.rtl.nl/rtlxl/network/%s/progressive/%s.mp4'
 
-        formats.extend([
-            {
-                'url': PG_URL_TEMPLATE % ('a2m', video_urlpart),
-                'format_id': 'pg-sd',
-            },
-            {
-                'url': PG_URL_TEMPLATE % ('a3m', video_urlpart),
-                'format_id': 'pg-hd',
-                'quality': 0,
+        PG_FORMATS = (
+            ('a2t', 512, 288),
+            ('a3t', 704, 400),
+            ('nettv', 1280, 720),
+        )
+
+        def pg_format(format_id, width, height):
+            return {
+                'url': PG_URL_TEMPLATE % (format_id, video_urlpart),
+                'format_id': 'pg-%s' % format_id,
+                'protocol': 'http',
+                'width': width,
+                'height': height,
             }
-        ])
+
+        if not formats:
+            formats = [pg_format(*pg_tuple) for pg_tuple in PG_FORMATS]
+        else:
+            pg_formats = []
+            for format_id, width, height in PG_FORMATS:
+                try:
+                    # Find hls format with the same width and height corresponding
+                    # to progressive format and copy metadata from it.
+                    f = next(f for f in formats if f.get('height') == height)
+                    # hls formats may have invalid width
+                    f['width'] = width
+                    f_copy = f.copy()
+                    f_copy.update(pg_format(format_id, width, height))
+                    pg_formats.append(f_copy)
+                except StopIteration:
+                    # Missing hls format does mean that no progressive format with
+                    # such width and height exists either.
+                    pass
+            formats.extend(pg_formats)
         self._sort_formats(formats)
 
         thumbnails = []

youtube_dl/extractor/scivee.py

@@ -18,6 +18,7 @@ class SciVeeIE(InfoExtractor):
             'title': 'Adam Arkin at the 2014 DOE JGI Genomics of Energy & Environment Meeting',
             'description': 'md5:81f1710638e11a481358fab1b11059d7',
         },
+        'skip': 'Not accessible from Travis CI server',
     }
 
     def _real_extract(self, url):

youtube_dl/extractor/screenwavemedia.py

@@ -12,7 +12,7 @@ from ..utils import (
 
 
 class ScreenwaveMediaIE(InfoExtractor):
-    _VALID_URL = r'https?://player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?.*\bid=(?P<id>[A-Za-z0-9-]+)'
+    _VALID_URL = r'(?:https?:)?//player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?.*\bid=(?P<id>[A-Za-z0-9-]+)'
     EMBED_PATTERN = r'src=(["\'])(?P<url>(?:https?:)?//player\d?\.screenwavemedia\.com/(?:play/)?[a-zA-Z]+\.php\?.*\bid=.+?)\1'
     _TESTS = [{
         'url': 'http://player.screenwavemedia.com/play/play.php?playerdiv=videoarea&companiondiv=squareAd&id=Cinemassacre-19911',

youtube_dl/extractor/tagesschau.py

@@ -4,42 +4,178 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
-from ..utils import parse_filesize
+from ..utils import (
+    determine_ext,
+    js_to_json,
+    parse_iso8601,
+    parse_filesize,
+)
+
+
+class TagesschauPlayerIE(InfoExtractor):
+    IE_NAME = 'tagesschau:player'
+    _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?P<kind>audio|video)/(?P=kind)-(?P<id>\d+)~player(?:_[^/?#&]+)?\.html'
+
+    _TESTS = [{
+        'url': 'http://www.tagesschau.de/multimedia/video/video-179517~player.html',
+        'md5': '8d09548d5c15debad38bee3a4d15ca21',
+        'info_dict': {
+            'id': '179517',
+            'ext': 'mp4',
+            'title': 'Marie Kristin Boese, ARD Berlin, über den zukünftigen Kurs der AfD',
+            'thumbnail': 're:^https?:.*\.jpg$',
+            'formats': 'mincount:6',
+        },
+    }, {
+        'url': 'https://www.tagesschau.de/multimedia/audio/audio-29417~player.html',
+        'md5': '76e6eec6ebd40740671cf0a2c88617e5',
+        'info_dict': {
+            'id': '29417',
+            'ext': 'mp3',
+            'title': 'Trabi - Bye, bye Rennpappe',
+            'thumbnail': 're:^https?:.*\.jpg$',
+            'formats': 'mincount:2',
+        },
+    }, {
+        'url': 'http://www.tagesschau.de/multimedia/audio/audio-29417~player_autoplay-true.html',
+        'only_matching': True,
+    }]
+
+    _FORMATS = {
+        'xs': {'quality': 0},
+        's': {'width': 320, 'height': 180, 'quality': 1},
+        'm': {'width': 512, 'height': 288, 'quality': 2},
+        'l': {'width': 960, 'height': 540, 'quality': 3},
+        'xl': {'width': 1280, 'height': 720, 'quality': 4},
+        'xxl': {'quality': 5},
+    }
+
+    def _extract_via_api(self, kind, video_id):
+        info = self._download_json(
+            'https://www.tagesschau.de/api/multimedia/{0}/{0}-{1}.json'.format(kind, video_id),
+            video_id)
+        title = info['headline']
+        formats = []
+        for media in info['mediadata']:
+            for format_id, format_url in media.items():
+                if determine_ext(format_url) == 'm3u8':
+                    formats.extend(self._extract_m3u8_formats(
+                        format_url, video_id, 'mp4',
+                        entry_protocol='m3u8_native', m3u8_id='hls'))
+                else:
+                    formats.append({
+                        'url': format_url,
+                        'format_id': format_id,
+                        'vcodec': 'none' if kind == 'audio' else None,
+                    })
+        self._sort_formats(formats)
+        timestamp = parse_iso8601(info.get('date'))
+        return {
+            'id': video_id,
+            'title': title,
+            'timestamp': timestamp,
+            'formats': formats,
+        }
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+
+        # kind = mobj.group('kind').lower()
+        # if kind == 'video':
+        #     return self._extract_via_api(kind, video_id)
+
+        # JSON api does not provide some audio formats (e.g. ogg) thus
+        # extractiong audio via webpage
+
+        webpage = self._download_webpage(url, video_id)
+
+        title = self._og_search_title(webpage).strip()
+        formats = []
+
+        for media_json in re.findall(r'({src\s*:\s*["\']http[^}]+type\s*:[^}]+})', webpage):
+            media = self._parse_json(js_to_json(media_json), video_id, fatal=False)
+            if not media:
+                continue
+            src = media.get('src')
+            if not src:
+                return
+            quality = media.get('quality')
+            kind = media.get('type', '').split('/')[0]
+            ext = determine_ext(src)
+            f = {
+                'url': src,
+                'format_id': '%s_%s' % (quality, ext) if quality else ext,
+                'ext': ext,
+                'vcodec': 'none' if kind == 'audio' else None,
+            }
+            f.update(self._FORMATS.get(quality, {}))
+            formats.append(f)
+
+        self._sort_formats(formats)
+
+        thumbnail = self._og_search_thumbnail(webpage)
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'formats': formats,
+        }
 
 
 class TagesschauIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/multimedia/(?:[^/]+/)*?[^/#?]+?(?P<id>-?[0-9]+)(?:~_[^/#?]+?)?\.html'
+    _VALID_URL = r'https?://(?:www\.)?tagesschau\.de/(?P<path>[^/]+/(?:[^/]+/)*?(?P<id>[^/#?]+?(?:-?[0-9]+)?))(?:~_?[^/#?]+?)?\.html'
 
     _TESTS = [{
         'url': 'http://www.tagesschau.de/multimedia/video/video-102143.html',
-        'md5': '917a228bc7df7850783bc47979673a09',
+        'md5': 'f7c27a0eff3bfe8c7727e65f8fe1b1e6',
         'info_dict': {
-            'id': '102143',
+            'id': 'video-102143',
             'ext': 'mp4',
             'title': 'Regierungsumbildung in Athen: Neue Minister in Griechenland vereidigt',
-            'description': 'md5:171feccd9d9b3dd54d05d501568f6359',
+            'description': '18.07.2015 20:10 Uhr',
             'thumbnail': 're:^https?:.*\.jpg$',
         },
     }, {
         'url': 'http://www.tagesschau.de/multimedia/sendung/ts-5727.html',
         'md5': '3c54c1f6243d279b706bde660ceec633',
         'info_dict': {
-            'id': '5727',
+            'id': 'ts-5727',
             'ext': 'mp4',
-            'description': 'md5:695c01bfd98b7e313c501386327aea59',
             'title': 'Sendung: tagesschau \t04.12.2014 20:00 Uhr',
+            'description': 'md5:695c01bfd98b7e313c501386327aea59',
             'thumbnail': 're:^https?:.*\.jpg$',
         },
     }, {
-        'url': 'http://www.tagesschau.de/multimedia/politikimradio/audio-18407.html',
-        'md5': 'aef45de271c4bf0a5db834aa40bf774c',
+        # exclusive audio
+        'url': 'http://www.tagesschau.de/multimedia/audio/audio-29417.html',
+        'md5': '76e6eec6ebd40740671cf0a2c88617e5',
         'info_dict': {
-            'id': '18407',
+            'id': 'audio-29417',
             'ext': 'mp3',
-            'title': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich',
-            'description': 'Flüchtlingsdebatte: Hitzig, aber wenig hilfreich',
+            'title': 'Trabi - Bye, bye Rennpappe',
+            'description': 'md5:8687dda862cbbe2cfb2df09b56341317',
             'thumbnail': 're:^https?:.*\.jpg$',
         },
+    }, {
+        # audio in article
+        'url': 'http://www.tagesschau.de/inland/bnd-303.html',
+        'md5': 'e0916c623e85fc1d2b26b78f299d3958',
+        'info_dict': {
+            'id': 'bnd-303',
+            'ext': 'mp3',
+            'title': 'Viele Baustellen für neuen BND-Chef',
+            'description': 'md5:1e69a54be3e1255b2b07cdbce5bcd8b4',
+            'thumbnail': 're:^https?:.*\.jpg$',
+        },
+    }, {
+        'url': 'http://www.tagesschau.de/inland/afd-parteitag-135.html',
+        'info_dict': {
+            'id': 'afd-parteitag-135',
+            'title': 'Möchtegern-Underdog mit Machtanspruch',
+        },
+        'playlist_count': 2,
     }, {
         'url': 'http://www.tagesschau.de/multimedia/sendung/tsg-3771.html',
         'only_matching': True,
@@ -61,88 +197,108 @@ class TagesschauIE(InfoExtractor):
     }, {
         'url': 'http://www.tagesschau.de/multimedia/video/video-102303~_bab-sendung-211.html',
         'only_matching': True,
+    }, {
+        'url': 'http://www.tagesschau.de/100sekunden/index.html',
+        'only_matching': True,
+    }, {
+        # playlist article with collapsing sections
+        'url': 'http://www.tagesschau.de/wirtschaft/faq-freihandelszone-eu-usa-101.html',
+        'only_matching': True,
     }]
 
-    _FORMATS = {
-        's': {'width': 256, 'height': 144, 'quality': 1},
-        'm': {'width': 512, 'height': 288, 'quality': 2},
-        'l': {'width': 960, 'height': 544, 'quality': 3},
-    }
+    @classmethod
+    def suitable(cls, url):
+        return False if TagesschauPlayerIE.suitable(url) else super(TagesschauIE, cls).suitable(url)
+
+    def _extract_formats(self, download_text, media_kind):
+        links = re.finditer(
+            r'<div class="button" title="(?P<title>[^"]*)"><a href="(?P<url>[^"]+)">(?P<name>.+?)</a></div>',
+            download_text)
+        formats = []
+        for l in links:
+            link_url = l.group('url')
+            if not link_url:
+                continue
+            format_id = self._search_regex(
+                r'.*/[^/.]+\.([^/]+)\.[^/.]+$', link_url, 'format ID',
+                default=determine_ext(link_url))
+            format = {
+                'format_id': format_id,
+                'url': l.group('url'),
+                'format_name': l.group('name'),
+            }
+            title = l.group('title')
+            if title:
+                if media_kind.lower() == 'video':
+                    m = re.match(
+                        r'''(?x)
+                            Video:\s*(?P<vcodec>[a-zA-Z0-9/._-]+)\s*&\#10;
+                            (?P<width>[0-9]+)x(?P<height>[0-9]+)px&\#10;
+                            (?P<vbr>[0-9]+)kbps&\#10;
+                            Audio:\s*(?P<abr>[0-9]+)kbps,\s*(?P<audio_desc>[A-Za-z\.0-9]+)&\#10;
+                            Gr&ouml;&szlig;e:\s*(?P<filesize_approx>[0-9.,]+\s+[a-zA-Z]*B)''',
+                        title)
+                    if m:
+                        format.update({
+                            'format_note': m.group('audio_desc'),
+                            'vcodec': m.group('vcodec'),
+                            'width': int(m.group('width')),
+                            'height': int(m.group('height')),
+                            'abr': int(m.group('abr')),
+                            'vbr': int(m.group('vbr')),
+                            'filesize_approx': parse_filesize(m.group('filesize_approx')),
+                        })
+                else:
+                    m = re.match(
+                        r'(?P<format>.+?)-Format\s*:\s*(?P<abr>\d+)kbps\s*,\s*(?P<note>.+)',
+                        title)
+                    if m:
+                        format.update({
+                            'format_note': '%s, %s' % (m.group('format'), m.group('note')),
+                            'vcodec': 'none',
+                            'abr': int(m.group('abr')),
+                        })
+            formats.append(format)
+        self._sort_formats(formats)
+        return formats
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id') or mobj.group('path')
         display_id = video_id.lstrip('-')
+
         webpage = self._download_webpage(url, display_id)
 
-        player_url = self._html_search_meta(
-            'twitter:player', webpage, 'player URL', default=None)
-        if player_url:
-            playerpage = self._download_webpage(
-                player_url, display_id, 'Downloading player page')
+        title = self._html_search_regex(
+            r'<span[^>]*class="headline"[^>]*>(.+?)</span>',
+            webpage, 'title', default=None) or self._og_search_title(webpage)
 
-            formats = []
-            for media in re.finditer(
-                    r'''(?x)
-                        (?P<q_url>["\'])(?P<url>http://media.+?)(?P=q_url)
-                        ,\s*type:(?P<q_type>["\'])(?P<type>video|audio)/(?P<ext>.+?)(?P=q_type)
-                        (?:,\s*quality:(?P<q_quality>["\'])(?P<quality>.+?)(?P=q_quality))?
-                    ''', playerpage):
-                url = media.group('url')
-                type_ = media.group('type')
-                ext = media.group('ext')
-                res = media.group('quality')
-                f = {
-                    'format_id': '%s_%s' % (res, ext) if res else ext,
-                    'url': url,
-                    'ext': ext,
-                    'vcodec': 'none' if type_ == 'audio' else None,
-                }
-                f.update(self._FORMATS.get(res, {}))
-                formats.append(f)
-            thumbnail = self._og_search_thumbnail(playerpage)
-            title = self._og_search_title(webpage).strip()
-            description = self._og_search_description(webpage).strip()
-        else:
+        DOWNLOAD_REGEX = r'(?s)<p>Wir bieten dieses (?P<kind>Video|Audio) in folgenden Formaten zum Download an:</p>\s*<div class="controls">(?P<links>.*?)</div>\s*<p>'
+
+        webpage_type = self._og_search_property('type', webpage, default=None)
+        if webpage_type == 'website':  # Article
+            entries = []
+            for num, (entry_title, media_kind, download_text) in enumerate(re.findall(
+                    r'(?s)<p[^>]+class="infotext"[^>]*>\s*(?:<a[^>]+>)?\s*<strong>(.+?)</strong>.*?</p>.*?%s' % DOWNLOAD_REGEX,
+                    webpage), 1):
+                entries.append({
+                    'id': '%s-%d' % (display_id, num),
+                    'title': '%s' % entry_title,
+                    'formats': self._extract_formats(download_text, media_kind),
+                })
+            if len(entries) > 1:
+                return self.playlist_result(entries, display_id, title)
+            formats = entries[0]['formats']
+        else:  # Assume single video
             download_text = self._search_regex(
-                r'(?s)<p>Wir bieten dieses Video in folgenden Formaten zum Download an:</p>\s*<div class="controls">(.*?)</div>\s*<p>',
-                webpage, 'download links')
-            links = re.finditer(
-                r'<div class="button" title="(?P<title>[^"]*)"><a href="(?P<url>[^"]+)">(?P<name>.+?)</a></div>',
-                download_text)
-            formats = []
-            for l in links:
-                format_id = self._search_regex(
-                    r'.*/[^/.]+\.([^/]+)\.[^/.]+', l.group('url'), 'format ID')
-                format = {
-                    'format_id': format_id,
-                    'url': l.group('url'),
-                    'format_name': l.group('name'),
-                }
-                m = re.match(
-                    r'''(?x)
-                        Video:\s*(?P<vcodec>[a-zA-Z0-9/._-]+)\s*&\#10;
-                        (?P<width>[0-9]+)x(?P<height>[0-9]+)px&\#10;
-                        (?P<vbr>[0-9]+)kbps&\#10;
-                        Audio:\s*(?P<abr>[0-9]+)kbps,\s*(?P<audio_desc>[A-Za-z\.0-9]+)&\#10;
-                        Gr&ouml;&szlig;e:\s*(?P<filesize_approx>[0-9.,]+\s+[a-zA-Z]*B)''',
-                    l.group('title'))
-                if m:
-                    format.update({
-                        'format_note': m.group('audio_desc'),
-                        'vcodec': m.group('vcodec'),
-                        'width': int(m.group('width')),
-                        'height': int(m.group('height')),
-                        'abr': int(m.group('abr')),
-                        'vbr': int(m.group('vbr')),
-                        'filesize_approx': parse_filesize(m.group('filesize_approx')),
-                    })
-                formats.append(format)
-            thumbnail = self._og_search_thumbnail(webpage)
-            description = self._html_search_regex(
-                r'(?s)<p class="teasertext">(.*?)</p>',
-                webpage, 'description', default=None)
-            title = self._html_search_regex(
-                r'<span class="headline".*?>(.*?)</span>', webpage, 'title')
+                DOWNLOAD_REGEX, webpage, 'download links', group='links')
+            media_kind = self._search_regex(
+                DOWNLOAD_REGEX, webpage, 'media kind', default='Video', group='kind')
+            formats = self._extract_formats(download_text, media_kind)
+        thumbnail = self._og_search_thumbnail(webpage)
+        description = self._html_search_regex(
+            r'(?s)<p class="teasertext">(.*?)</p>',
+            webpage, 'description', default=None)
 
         self._sort_formats(formats)
 

youtube_dl/extractor/ted.py

@@ -27,7 +27,7 @@ class TEDIE(InfoExtractor):
         '''
     _TESTS = [{
         'url': 'http://www.ted.com/talks/dan_dennett_on_our_consciousness.html',
-        'md5': 'fc94ac279feebbce69f21c0c6ee82810',
+        'md5': '0de43ac406aa3e4ea74b66c9c7789b13',
         'info_dict': {
             'id': '102',
             'ext': 'mp4',
@@ -37,21 +37,26 @@ class TEDIE(InfoExtractor):
                             'consciousness, but that half the time our brains are '
                             'actively fooling us.'),
             'uploader': 'Dan Dennett',
-            'width': 854,
+            'width': 853,
             'duration': 1308,
         }
     }, {
         'url': 'http://www.ted.com/watch/ted-institute/ted-bcg/vishal-sikka-the-beauty-and-power-of-algorithms',
-        'md5': '226f4fb9c62380d11b7995efa4c87994',
+        'md5': 'b899ac15e345fb39534d913f7606082b',
         'info_dict': {
-            'id': 'vishal-sikka-the-beauty-and-power-of-algorithms',
+            'id': 'tSVI8ta_P4w',
             'ext': 'mp4',
             'title': 'Vishal Sikka: The beauty and power of algorithms',
             'thumbnail': 're:^https?://.+\.jpg',
-            'description': 'Adaptive, intelligent, and consistent, algorithms are emerging as the ultimate app for everything from matching consumers to products to assessing medical diagnoses. Vishal Sikka shares his appreciation for the algorithm, charting both its inherent beauty and its growing power.',
-        }
+            'description': 'md5:6261fdfe3e02f4f579cbbfc00aff73f4',
+            'upload_date': '20140122',
+            'uploader_id': 'TEDInstitute',
+            'uploader': 'TED Institute',
+        },
+        'add_ie': ['Youtube'],
     }, {
         'url': 'http://www.ted.com/talks/gabby_giffords_and_mark_kelly_be_passionate_be_courageous_be_your_best',
+        'md5': '71b3ab2f4233012dce09d515c9c39ce2',
         'info_dict': {
             'id': '1972',
             'ext': 'mp4',
@@ -102,9 +107,9 @@ class TEDIE(InfoExtractor):
     }]
 
     _NATIVE_FORMATS = {
-        'low': {'preference': 1, 'width': 320, 'height': 180},
-        'medium': {'preference': 2, 'width': 512, 'height': 288},
-        'high': {'preference': 3, 'width': 854, 'height': 480},
+        'low': {'width': 320, 'height': 180},
+        'medium': {'width': 512, 'height': 288},
+        'high': {'width': 854, 'height': 480},
     }
 
     def _extract_info(self, webpage):
@@ -171,15 +176,21 @@ class TEDIE(InfoExtractor):
                 if finfo:
                     f.update(finfo)
 
+        http_url = None
         for format_id, resources in talk_info['resources'].items():
             if format_id == 'h264':
                 for resource in resources:
+                    h264_url = resource.get('file')
+                    if not h264_url:
+                        continue
                     bitrate = int_or_none(resource.get('bitrate'))
                     formats.append({
-                        'url': resource['file'],
+                        'url': h264_url,
                         'format_id': '%s-%sk' % (format_id, bitrate),
                         'tbr': bitrate,
                     })
+                    if re.search('\d+k', h264_url):
+                        http_url = h264_url
             elif format_id == 'rtmp':
                 streamer = talk_info.get('streamer')
                 if not streamer:
@@ -195,16 +206,24 @@ class TEDIE(InfoExtractor):
                         'tbr': int_or_none(resource.get('bitrate')),
                     })
             elif format_id == 'hls':
-                hls_formats = self._extract_m3u8_formats(
-                    resources.get('stream'), video_name, 'mp4', m3u8_id=format_id)
-                for f in hls_formats:
-                    if f.get('format_id') == 'hls-meta':
-                        continue
-                    if not f.get('height'):
-                        f['vcodec'] = 'none'
-                    else:
-                        f['acodec'] = 'none'
-                formats.extend(hls_formats)
+                formats.extend(self._extract_m3u8_formats(
+                    resources.get('stream'), video_name, 'mp4', m3u8_id=format_id, fatal=False))
+
+        m3u8_formats = list(filter(
+            lambda f: f.get('protocol') == 'm3u8' and f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
+            formats))
+        if http_url:
+            for m3u8_format in m3u8_formats:
+                bitrate = self._search_regex(r'(\d+k)', m3u8_format['url'], 'bitrate', default=None)
+                if not bitrate:
+                    continue
+                f = m3u8_format.copy()
+                f.update({
+                    'url': re.sub(r'\d+k', bitrate, http_url),
+                    'format_id': m3u8_format['format_id'].replace('hls', 'http'),
+                    'protocol': 'http',
+                })
+                formats.append(f)
 
         audio_download = talk_info.get('audioDownload')
         if audio_download:
@@ -212,7 +231,6 @@ class TEDIE(InfoExtractor):
                 'url': audio_download,
                 'format_id': 'audio',
                 'vcodec': 'none',
-                'preference': -0.5,
             })
 
         self._sort_formats(formats)
@@ -254,7 +272,11 @@ class TEDIE(InfoExtractor):
 
         config_json = self._html_search_regex(
             r'"pages\.jwplayer"\s*,\s*({.+?})\s*\)\s*</script>',
-            webpage, 'config')
+            webpage, 'config', default=None)
+        if not config_json:
+            embed_url = self._search_regex(
+                r"<iframe[^>]+class='pages-video-embed__video__object'[^>]+src='([^']+)'", webpage, 'embed url')
+            return self.url_result(self._proto_relative_url(embed_url))
         config = json.loads(config_json)['config']
         video_url = config['video']['url']
         thumbnail = config.get('image', {}).get('url')

youtube_dl/extractor/theplatform.py

@@ -159,11 +159,11 @@ class ThePlatformIE(ThePlatformBaseIE):
         def str_to_hex(str):
             return binascii.b2a_hex(str.encode('ascii')).decode('ascii')
 
-        def hex_to_str(hex):
-            return binascii.a2b_hex(hex)
+        def hex_to_bytes(hex):
+            return binascii.a2b_hex(hex.encode('ascii'))
 
         relative_path = re.match(r'https?://link.theplatform.com/s/([^?]+)', url).group(1)
-        clear_text = hex_to_str(flags + expiration_date + str_to_hex(relative_path))
+        clear_text = hex_to_bytes(flags + expiration_date + str_to_hex(relative_path))
         checksum = hmac.new(sig_key.encode('ascii'), clear_text, hashlib.sha1).hexdigest()
         sig = flags + expiration_date + checksum + str_to_hex(sig_secret)
         return '%s&sig=%s' % (url, sig)

youtube_dl/extractor/twentymin.py

@@ -32,7 +32,22 @@ class TwentyMinutenIE(InfoExtractor):
             'title': '«Wir müssen mutig nach vorne schauen»',
             'description': 'Kein Land sei innovativer als die Schweiz, sagte Johann Schneider-Ammann in seiner Neujahrsansprache. Das Land müsse aber seine Hausaufgaben machen.',
             'thumbnail': 'http://www.20min.ch/images/content/2/2/0/22050469/10/teaserbreit.jpg'
-        }
+        },
+        'skip': '"This video is no longer available" is shown both on the web page and in the downloaded file.',
+    }, {
+        # YouTube embed
+        'url': 'http://www.20min.ch/ro/sports/football/story/Il-marque-une-bicyclette-de-plus-de-30-metres--21115184',
+        'md5': 'cec64d59aa01c0ed9dbba9cf639dd82f',
+        'info_dict': {
+            'id': 'ivM7A7SpDOs',
+            'ext': 'mp4',
+            'title': 'GOLAZO DE CHILENA DE JAVI GÓMEZ, FINALISTA AL BALÓN DE CLM 2016',
+            'description': 'md5:903c92fbf2b2f66c09de514bc25e9f5a',
+            'upload_date': '20160424',
+            'uploader': 'RTVCM Castilla-La Mancha',
+            'uploader_id': 'RTVCM',
+        },
+        'add_ie': ['Youtube'],
     }, {
         'url': 'http://www.20min.ch/videotv/?cid=44&vid=468738',
         'only_matching': True,
@@ -48,6 +63,12 @@ class TwentyMinutenIE(InfoExtractor):
 
         webpage = self._download_webpage(url, display_id)
 
+        youtube_url = self._html_search_regex(
+            r'<iframe[^>]+src="((?:https?:)?//www\.youtube\.com/embed/[^"]+)"',
+            webpage, 'YouTube embed URL', default=None)
+        if youtube_url is not None:
+            return self.url_result(youtube_url, 'Youtube')
+
         title = self._html_search_regex(
             r'<h1>.*?<span>(.+?)</span></h1>',
             webpage, 'title', default=None)

youtube_dl/extractor/unistra.py

@@ -49,6 +49,7 @@ class UnistraIE(InfoExtractor):
                 'format_id': format_id,
                 'quality': quality(format_id)
             })
+        self._sort_formats(formats)
 
         title = self._html_search_regex(
             r'<title>UTV - (.*?)</', webpage, 'title')

youtube_dl/extractor/viewlift.py

@@ -13,8 +13,12 @@ from ..utils import (
 )
 
 
-class SnagFilmsEmbedIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:(?:www|embed)\.)?snagfilms\.com/embed/player\?.*\bfilmId=(?P<id>[\da-f-]{36})'
+class ViewLiftBaseIE(InfoExtractor):
+    _DOMAINS_REGEX = '(?:snagfilms|snagxtreme|funnyforfree|kiddovid|winnersview|monumentalsportsnetwork|vayafilm)\.com|kesari\.tv'
+
+
+class ViewLiftEmbedIE(ViewLiftBaseIE):
+    _VALID_URL = r'https?://(?:(?:www|embed)\.)?(?:%s)/embed/player\?.*\bfilmId=(?P<id>[\da-f-]{36})' % ViewLiftBaseIE._DOMAINS_REGEX
     _TESTS = [{
         'url': 'http://embed.snagfilms.com/embed/player?filmId=74849a00-85a9-11e1-9660-123139220831&w=500',
         'md5': '2924e9215c6eff7a55ed35b72276bd93',
@@ -40,7 +44,7 @@ class SnagFilmsEmbedIE(InfoExtractor):
     @staticmethod
     def _extract_url(webpage):
         mobj = re.search(
-            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:embed\.)?snagfilms\.com/embed/player.+?)\1',
+            r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:embed\.)?(?:%s)/embed/player.+?)\1' % ViewLiftBaseIE._DOMAINS_REGEX,
             webpage)
         if mobj:
             return mobj.group('url')
@@ -55,6 +59,7 @@ class SnagFilmsEmbedIE(InfoExtractor):
                 'Film %s is not playable in your area.' % video_id, expected=True)
 
         formats = []
+        has_bitrate = False
         for source in self._parse_json(js_to_json(self._search_regex(
                 r'(?s)sources:\s*(\[.+?\]),', webpage, 'json')), video_id):
             file_ = source.get('file')
@@ -63,22 +68,25 @@ class SnagFilmsEmbedIE(InfoExtractor):
             type_ = source.get('type')
             ext = determine_ext(file_)
             format_id = source.get('label') or ext
-            if all(v == 'm3u8' for v in (type_, ext)):
+            if all(v == 'm3u8' or v == 'hls' for v in (type_, ext)):
                 formats.extend(self._extract_m3u8_formats(
                     file_, video_id, 'mp4', m3u8_id='hls'))
             else:
                 bitrate = int_or_none(self._search_regex(
                     [r'(\d+)kbps', r'_\d{1,2}x\d{1,2}_(\d{3,})\.%s' % ext],
                     file_, 'bitrate', default=None))
+                if not has_bitrate and bitrate:
+                    has_bitrate = True
                 height = int_or_none(self._search_regex(
                     r'^(\d+)[pP]$', format_id, 'height', default=None))
                 formats.append({
                     'url': file_,
-                    'format_id': format_id,
+                    'format_id': 'http-%s%s' % (format_id, ('-%dk' % bitrate if bitrate else '')),
                     'tbr': bitrate,
                     'height': height,
                 })
-        self._sort_formats(formats)
+        field_preference = None if has_bitrate else ('height', 'tbr', 'format_id')
+        self._sort_formats(formats, field_preference)
 
         title = self._search_regex(
             [r"title\s*:\s*'([^']+)'", r'<title>([^<]+)</title>'],
@@ -91,8 +99,8 @@ class SnagFilmsEmbedIE(InfoExtractor):
         }
 
 
-class SnagFilmsIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?snagfilms\.com/(?:films/title|show)/(?P<id>[^?#]+)'
+class ViewLiftIE(ViewLiftBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?(?P<domain>%s)/(?:films/title|show|(?:news/)?videos?)/(?P<id>[^?#]+)' % ViewLiftBaseIE._DOMAINS_REGEX
     _TESTS = [{
         'url': 'http://www.snagfilms.com/films/title/lost_for_life',
         'md5': '19844f897b35af219773fd63bdec2942',
@@ -127,10 +135,16 @@ class SnagFilmsIE(InfoExtractor):
         # Film is not available.
         'url': 'http://www.snagfilms.com/show/augie_alone/flirting',
         'only_matching': True,
+    }, {
+        'url': 'http://www.winnersview.com/videos/the-good-son',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.kesari.tv/news/video/1461919076414',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        display_id = self._match_id(url)
+        domain, display_id = re.match(self._VALID_URL, url).groups()
 
         webpage = self._download_webpage(url, display_id)
 
@@ -170,7 +184,7 @@ class SnagFilmsIE(InfoExtractor):
 
         return {
             '_type': 'url_transparent',
-            'url': 'http://embed.snagfilms.com/embed/player?filmId=%s' % film_id,
+            'url': 'http://%s/embed/player?filmId=%s' % (domain, film_id),
             'id': film_id,
             'display_id': display_id,
             'title': title,
@@ -178,4 +192,5 @@ class SnagFilmsIE(InfoExtractor):
             'thumbnail': thumbnail,
             'duration': duration,
             'categories': categories,
+            'ie_key': 'ViewLiftEmbed',
         }

youtube_dl/extractor/viewster.py

@@ -6,7 +6,6 @@ import re
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,
-    compat_urllib_parse,
     compat_urllib_parse_unquote,
 )
 from ..utils import (
@@ -78,11 +77,11 @@ class ViewsterIE(InfoExtractor):
 
     _ACCEPT_HEADER = 'application/json, text/javascript, */*; q=0.01'
 
-    def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True):
+    def _download_json(self, url, video_id, note='Downloading JSON metadata', fatal=True, query={}):
         request = sanitized_Request(url)
         request.add_header('Accept', self._ACCEPT_HEADER)
         request.add_header('Auth-token', self._AUTH_TOKEN)
-        return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal)
+        return super(ViewsterIE, self)._download_json(request, video_id, note, fatal=fatal, query=query)
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -117,57 +116,85 @@ class ViewsterIE(InfoExtractor):
             return self.playlist_result(entries, video_id, title, description)
 
         formats = []
-        manifest_url = None
-        for media_type in ('application/f4m+xml', 'application/x-mpegURL', 'video/mp4'):
-            media = self._download_json(
-                'https://public-api.viewster.com/movies/%s/video?mediaType=%s'
-                % (entry_id, compat_urllib_parse.quote(media_type)),
-                video_id, 'Downloading %s JSON' % media_type, fatal=False)
-            if not media:
-                continue
-            video_url = media.get('Uri')
-            if not video_url:
-                continue
-            ext = determine_ext(video_url)
-            if ext == 'f4m':
-                manifest_url = video_url
-                video_url += '&' if '?' in video_url else '?'
-                video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1'
-                formats.extend(self._extract_f4m_formats(
-                    video_url, video_id, f4m_id='hds'))
-            elif ext == 'm3u8':
-                manifest_url = video_url
-                m3u8_formats = self._extract_m3u8_formats(
-                    video_url, video_id, 'mp4', m3u8_id='hls',
-                    fatal=False)  # m3u8 sometimes fail
-                if m3u8_formats:
-                    formats.extend(m3u8_formats)
-            else:
-                qualities_basename = self._search_regex(
-                    '/([^/]+)\.csmil/',
-                    manifest_url, 'qualities basename', default=None)
-                if not qualities_basename:
-                    continue
-                QUALITIES_RE = r'((,\d+k)+,?)'
-                qualities = self._search_regex(
-                    QUALITIES_RE, qualities_basename,
-                    'qualities', default=None)
-                if not qualities:
-                    continue
-                qualities = qualities.strip(',').split(',')
-                http_template = re.sub(QUALITIES_RE, r'%s', qualities_basename)
-                http_url_basename = url_basename(video_url)
-                for q in qualities:
-                    tbr = int_or_none(self._search_regex(
-                        r'(\d+)k', q, 'bitrate', default=None))
-                    formats.append({
-                        'url': video_url.replace(http_url_basename, http_template % q),
-                        'ext': 'mp4',
-                        'format_id': 'http' + ('-%d' % tbr if tbr else ''),
-                        'tbr': tbr,
-                    })
+        for language_set in info.get('LanguageSets', []):
+            manifest_url = None
+            m3u8_formats = []
+            audio = language_set.get('Audio') or ''
+            subtitle = language_set.get('Subtitle') or ''
+            base_format_id = audio
+            if subtitle:
+                base_format_id += '-%s' % subtitle
 
-        if not formats and not info.get('LanguageSets') and not info.get('VODSettings'):
+            def concat(suffix, sep='-'):
+                return (base_format_id + '%s%s' % (sep, suffix)) if base_format_id else suffix
+
+            for media_type in ('application/f4m+xml', 'application/x-mpegURL', 'video/mp4'):
+                media = self._download_json(
+                    'https://public-api.viewster.com/movies/%s/video' % entry_id,
+                    video_id, 'Downloading %s JSON' % concat(media_type, ' '), fatal=False, query={
+                        'mediaType': media_type,
+                        'language': audio,
+                        'subtitle': subtitle,
+                    })
+                if not media:
+                    continue
+                video_url = media.get('Uri')
+                if not video_url:
+                    continue
+                ext = determine_ext(video_url)
+                if ext == 'f4m':
+                    manifest_url = video_url
+                    video_url += '&' if '?' in video_url else '?'
+                    video_url += 'hdcore=3.2.0&plugin=flowplayer-3.2.0.1'
+                    formats.extend(self._extract_f4m_formats(
+                        video_url, video_id, f4m_id=concat('hds')))
+                elif ext == 'm3u8':
+                    manifest_url = video_url
+                    m3u8_formats = self._extract_m3u8_formats(
+                        video_url, video_id, 'mp4', m3u8_id=concat('hls'),
+                        fatal=False)  # m3u8 sometimes fail
+                    if m3u8_formats:
+                        formats.extend(m3u8_formats)
+                else:
+                    qualities_basename = self._search_regex(
+                        '/([^/]+)\.csmil/',
+                        manifest_url, 'qualities basename', default=None)
+                    if not qualities_basename:
+                        continue
+                    QUALITIES_RE = r'((,\d+k)+,?)'
+                    qualities = self._search_regex(
+                        QUALITIES_RE, qualities_basename,
+                        'qualities', default=None)
+                    if not qualities:
+                        continue
+                    qualities = list(map(lambda q: int(q[:-1]), qualities.strip(',').split(',')))
+                    qualities.sort()
+                    http_template = re.sub(QUALITIES_RE, r'%dk', qualities_basename)
+                    http_url_basename = url_basename(video_url)
+                    if m3u8_formats:
+                        self._sort_formats(m3u8_formats)
+                        m3u8_formats = list(filter(
+                            lambda f: f.get('vcodec') != 'none' and f.get('resolution') != 'multiple',
+                            m3u8_formats))
+                    if len(qualities) == len(m3u8_formats):
+                        for q, m3u8_format in zip(qualities, m3u8_formats):
+                            f = m3u8_format.copy()
+                            f.update({
+                                'url': video_url.replace(http_url_basename, http_template % q),
+                                'format_id': f['format_id'].replace('hls', 'http'),
+                                'protocol': 'http',
+                            })
+                            formats.append(f)
+                    else:
+                        for q in qualities:
+                            formats.append({
+                                'url': video_url.replace(http_url_basename, http_template % q),
+                                'ext': 'mp4',
+                                'format_id': 'http-%d' % q,
+                                'tbr': q,
+                            })
+
+        if not formats and not info.get('VODSettings'):
             self.raise_geo_restricted()
 
         self._sort_formats(formats)

youtube_dl/extractor/vlive.py

@@ -1,9 +1,13 @@
 # coding: utf-8
-from __future__ import unicode_literals
+from __future__ import division, unicode_literals
+
+import re
+import time
 
 from .common import InfoExtractor
 from ..utils import (
     dict_get,
+    ExtractorError,
     float_or_none,
     int_or_none,
 )
@@ -31,16 +35,76 @@ class VLiveIE(InfoExtractor):
         webpage = self._download_webpage(
             'http://www.vlive.tv/video/%s' % video_id, video_id)
 
-        long_video_id = self._search_regex(
-            r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"([^"]+)"',
-            webpage, 'long video id')
+        # UTC+x - UTC+9 (KST)
+        tz = time.altzone if time.localtime().tm_isdst == 1 else time.timezone
+        tz_offset = -tz // 60 - 9 * 60
+        self._set_cookie('vlive.tv', 'timezoneOffset', '%d' % tz_offset)
 
-        key = self._search_regex(
-            r'vlive\.tv\.video\.ajax\.request\.handler\.init\(\s*"[0-9]+"\s*,\s*"[^"]*"\s*,\s*"[^"]+"\s*,\s*"([^"]+)"',
-            webpage, 'key')
+        status_params = self._download_json(
+            'http://www.vlive.tv/video/status?videoSeq=%s' % video_id,
+            video_id, 'Downloading JSON status',
+            headers={'Referer': url.encode('utf-8')})
+        status = status_params.get('status')
+        air_start = status_params.get('onAirStartAt', '')
+        is_live = status_params.get('isLive')
 
+        video_params = self._search_regex(
+            r'vlive\.tv\.video\.ajax\.request\.handler\.init\((.+)\)',
+            webpage, 'video params')
+        live_params, long_video_id, key = re.split(
+            r'"\s*,\s*"', video_params)[1:4]
+
+        if status == 'LIVE_ON_AIR' or status == 'BIG_EVENT_ON_AIR':
+            live_params = self._parse_json('"%s"' % live_params, video_id)
+            live_params = self._parse_json(live_params, video_id)
+            return self._live(video_id, webpage, live_params)
+        elif status == 'VOD_ON_AIR' or status == 'BIG_EVENT_INTRO':
+            if long_video_id and key:
+                return self._replay(video_id, webpage, long_video_id, key)
+            elif is_live:
+                status = 'LIVE_END'
+            else:
+                status = 'COMING_SOON'
+
+        if status == 'LIVE_END':
+            raise ExtractorError('Uploading for replay. Please wait...',
+                                 expected=True)
+        elif status == 'COMING_SOON':
+            raise ExtractorError('Coming soon! %s' % air_start, expected=True)
+        elif status == 'CANCELED':
+            raise ExtractorError('We are sorry, '
+                                 'but the live broadcast has been canceled.',
+                                 expected=True)
+        else:
+            raise ExtractorError('Unknown status %s' % status)
+
+    def _get_common_fields(self, webpage):
         title = self._og_search_title(webpage)
+        creator = self._html_search_regex(
+            r'<div[^>]+class="info_area"[^>]*>\s*<a\s+[^>]*>([^<]+)',
+            webpage, 'creator', fatal=False)
+        thumbnail = self._og_search_thumbnail(webpage)
+        return {
+            'title': title,
+            'creator': creator,
+            'thumbnail': thumbnail,
+        }
 
+    def _live(self, video_id, webpage, live_params):
+        formats = []
+        for vid in live_params.get('resolutions', []):
+            formats.extend(self._extract_m3u8_formats(
+                vid['cdnUrl'], video_id, 'mp4',
+                m3u8_id=vid.get('name'),
+                fatal=False, live=True))
+        self._sort_formats(formats)
+
+        return dict(self._get_common_fields(webpage),
+                    id=video_id,
+                    formats=formats,
+                    is_live=True)
+
+    def _replay(self, video_id, webpage, long_video_id, key):
         playinfo = self._download_json(
             'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?%s'
             % compat_urllib_parse_urlencode({
@@ -62,11 +126,6 @@ class VLiveIE(InfoExtractor):
         } for vid in playinfo.get('videos', {}).get('list', []) if vid.get('source')]
         self._sort_formats(formats)
 
-        thumbnail = self._og_search_thumbnail(webpage)
-        creator = self._html_search_regex(
-            r'<div[^>]+class="info_area"[^>]*>\s*<a\s+[^>]*>([^<]+)',
-            webpage, 'creator', fatal=False)
-
         view_count = int_or_none(playinfo.get('meta', {}).get('count'))
 
         subtitles = {}
@@ -77,12 +136,8 @@ class VLiveIE(InfoExtractor):
                     'ext': 'vtt',
                     'url': caption['source']}]
 
-        return {
-            'id': video_id,
-            'title': title,
-            'creator': creator,
-            'thumbnail': thumbnail,
-            'view_count': view_count,
-            'formats': formats,
-            'subtitles': subtitles,
-        }
+        return dict(self._get_common_fields(webpage),
+                    id=video_id,
+                    formats=formats,
+                    view_count=view_count,
+                    subtitles=subtitles)

youtube_dl/extractor/watchindianporn.py

@@ -11,61 +11,27 @@ from ..utils import (
 )
 
 
-class SexyKarmaIE(InfoExtractor):
-    IE_DESC = 'Sexy Karma and Watch Indian Porn'
-    _VALID_URL = r'https?://(?:www\.)?(?:sexykarma\.com|watchindianporn\.net)/(?:[^/]+/)*video/(?P<display_id>[^/]+)-(?P<id>[a-zA-Z0-9]+)\.html'
-    _TESTS = [{
-        'url': 'http://www.sexykarma.com/gonewild/video/taking-a-quick-pee-yHI70cOyIHt.html',
-        'md5': 'b9798e7d1ef1765116a8f516c8091dbd',
+class WatchIndianPornIE(InfoExtractor):
+    IE_DESC = 'Watch Indian Porn'
+    _VALID_URL = r'https?://(?:www\.)?watchindianporn\.net/(?:[^/]+/)*video/(?P<display_id>[^/]+)-(?P<id>[a-zA-Z0-9]+)\.html'
+    _TEST = {
+        'url': 'http://www.watchindianporn.net/video/hot-milf-from-kerala-shows-off-her-gorgeous-large-breasts-on-camera-RZa2avywNPa.html',
+        'md5': '249589a164dde236ec65832bfce17440',
         'info_dict': {
-            'id': 'yHI70cOyIHt',
-            'display_id': 'taking-a-quick-pee',
+            'id': 'RZa2avywNPa',
+            'display_id': 'hot-milf-from-kerala-shows-off-her-gorgeous-large-breasts-on-camera',
             'ext': 'mp4',
-            'title': 'Taking a quick pee.',
+            'title': 'Hot milf from kerala shows off her gorgeous large breasts on camera',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'uploader': 'wildginger7',
-            'upload_date': '20141008',
-            'duration': 22,
+            'uploader': 'LoveJay',
+            'upload_date': '20160428',
+            'duration': 226,
             'view_count': int,
             'comment_count': int,
             'categories': list,
             'age_limit': 18,
         }
-    }, {
-        'url': 'http://www.sexykarma.com/gonewild/video/pot-pixie-tribute-8Id6EZPbuHf.html',
-        'md5': 'dd216c68d29b49b12842b9babe762a5d',
-        'info_dict': {
-            'id': '8Id6EZPbuHf',
-            'display_id': 'pot-pixie-tribute',
-            'ext': 'mp4',
-            'title': 'pot_pixie tribute',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'uploader': 'banffite',
-            'upload_date': '20141013',
-            'duration': 16,
-            'view_count': int,
-            'comment_count': int,
-            'categories': list,
-            'age_limit': 18,
-        }
-    }, {
-        'url': 'http://www.watchindianporn.net/video/desi-dancer-namrata-stripping-completely-nude-and-dancing-on-a-hot-number-dW2mtctxJfs.html',
-        'md5': '9afb80675550406ed9a63ac2819ef69d',
-        'info_dict': {
-            'id': 'dW2mtctxJfs',
-            'display_id': 'desi-dancer-namrata-stripping-completely-nude-and-dancing-on-a-hot-number',
-            'ext': 'mp4',
-            'title': 'Desi dancer namrata stripping completely nude and dancing on a hot number',
-            'thumbnail': 're:^https?://.*\.jpg$',
-            'uploader': 'Don',
-            'upload_date': '20140213',
-            'duration': 83,
-            'view_count': int,
-            'comment_count': int,
-            'categories': list,
-            'age_limit': 18,
-        }
-    }]
+    }
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -109,6 +75,9 @@ class SexyKarmaIE(InfoExtractor):
             'id': video_id,
             'display_id': display_id,
             'url': video_url,
+            'http_headers': {
+                'Referer': url,
+            },
             'title': title,
             'thumbnail': thumbnail,
             'uploader': uploader,

youtube_dl/extractor/wsj.py

@@ -4,16 +4,22 @@ from __future__ import unicode_literals
 from .common import InfoExtractor
 from ..utils import (
     int_or_none,
+    float_or_none,
     unified_strdate,
 )
 
 
 class WSJIE(InfoExtractor):
-    _VALID_URL = r'https?://video-api\.wsj\.com/api-video/player/iframe\.html\?guid=(?P<id>[a-zA-Z0-9-]+)'
+    _VALID_URL = r'''(?x)https?://
+        (?:
+            video-api\.wsj\.com/api-video/player/iframe\.html\?guid=|
+            (?:www\.)?wsj\.com/video/[^/]+/
+        )
+        (?P<id>[a-zA-Z0-9-]+)'''
     IE_DESC = 'Wall Street Journal'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://video-api.wsj.com/api-video/player/iframe.html?guid=1BD01A4C-BFE8-40A5-A42F-8A8AF9898B1A',
-        'md5': '9747d7a6ebc2f4df64b981e1dde9efa9',
+        'md5': 'e230a5bb249075e40793b655a54a02e4',
         'info_dict': {
             'id': '1BD01A4C-BFE8-40A5-A42F-8A8AF9898B1A',
             'ext': 'mp4',
@@ -24,65 +30,60 @@ class WSJIE(InfoExtractor):
             'duration': 90,
             'title': 'Bills Coach Rex Ryan Updates His Old Jets Tattoo',
         },
-    }
+    }, {
+        'url': 'http://www.wsj.com/video/can-alphabet-build-a-smarter-city/359DDAA8-9AC1-489C-82E6-0429C1E430E0.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        bitrates = [128, 174, 264, 320, 464, 664, 1264]
         api_url = (
             'http://video-api.wsj.com/api-video/find_all_videos.asp?'
-            'type=guid&count=1&query=%s&'
-            'fields=hls,adZone,thumbnailList,guid,state,secondsUntilStartTime,'
-            'author,description,name,linkURL,videoStillURL,duration,videoURL,'
-            'adCategory,catastrophic,linkShortURL,doctypeID,youtubeID,'
-            'titletag,rssURL,wsj-section,wsj-subsection,allthingsd-section,'
-            'allthingsd-subsection,sm-section,sm-subsection,provider,'
-            'formattedCreationDate,keywords,keywordsOmniture,column,editor,'
-            'emailURL,emailPartnerID,showName,omnitureProgramName,'
-            'omnitureVideoFormat,linkRelativeURL,touchCastID,'
-            'omniturePublishDate,%s') % (
-                video_id, ','.join('video%dkMP4Url' % br for br in bitrates))
+            'type=guid&count=1&query=%s&fields=type,hls,videoMP4List,'
+            'thumbnailList,author,description,name,duration,videoURL,'
+            'titletag,formattedCreationDate,keywords,editor' % video_id)
         info = self._download_json(api_url, video_id)['items'][0]
-
-        # Thumbnails are conveniently in the correct format already
-        thumbnails = info.get('thumbnailList')
-        creator = info.get('author')
-        uploader_id = info.get('editor')
-        categories = info.get('keywords')
-        duration = int_or_none(info.get('duration'))
-        upload_date = unified_strdate(
-            info.get('formattedCreationDate'), day_first=False)
         title = info.get('name', info.get('titletag'))
 
-        formats = [{
-            'format_id': 'f4m',
-            'format_note': 'f4m (meta URL)',
-            'url': info['videoURL'],
-        }]
-        if info.get('hls'):
+        formats = []
+
+        f4m_url = info.get('videoURL')
+        if f4m_url:
+            formats.extend(self._extract_f4m_formats(
+                f4m_url, video_id, f4m_id='hds', fatal=False))
+
+        m3u8_url = info.get('hls')
+        if m3u8_url:
             formats.extend(self._extract_m3u8_formats(
                 info['hls'], video_id, ext='mp4',
-                preference=0, entry_protocol='m3u8_native'))
-        for br in bitrates:
-            field = 'video%dkMP4Url' % br
-            if info.get(field):
-                formats.append({
-                    'format_id': 'mp4-%d' % br,
-                    'container': 'mp4',
-                    'tbr': br,
-                    'url': info[field],
-                })
+                entry_protocol='m3u8_native', m3u8_id='hls', fatal=False))
+
+        for v in info.get('videoMP4List', []):
+            mp4_url = v.get('url')
+            if not mp4_url:
+                continue
+            tbr = int_or_none(v.get('bitrate'))
+            formats.append({
+                'url': mp4_url,
+                'format_id': 'http' + ('-%d' % tbr if tbr else ''),
+                'tbr': tbr,
+                'width': int_or_none(v.get('width')),
+                'height': int_or_none(v.get('height')),
+                'fps': float_or_none(v.get('fps')),
+            })
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'formats': formats,
-            'thumbnails': thumbnails,
-            'creator': creator,
-            'uploader_id': uploader_id,
-            'duration': duration,
-            'upload_date': upload_date,
+            # Thumbnails are conveniently in the correct format already
+            'thumbnails': info.get('thumbnailList'),
+            'creator': info.get('author'),
+            'uploader_id': info.get('editor'),
+            'duration': int_or_none(info.get('duration')),
+            'upload_date': unified_strdate(info.get(
+                'formattedCreationDate'), day_first=False),
             'title': title,
-            'categories': categories,
+            'categories': info.get('keywords'),
         }

youtube_dl/extractor/xiami.py

@@ -0,0 +1,158 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
+from ..utils import int_or_none
+
+
+class XiamiBaseIE(InfoExtractor):
+    _API_BASE_URL = 'http://www.xiami.com/song/playlist/cat/json/id'
+
+    def _extract_track(self, track, track_id=None):
+        title = track['title']
+        track_url = self._decrypt(track['location'])
+
+        subtitles = {}
+        lyrics_url = track.get('lyric_url') or track.get('lyric')
+        if lyrics_url and lyrics_url.startswith('http'):
+            subtitles['origin'] = [{'url': lyrics_url}]
+
+        return {
+            'id': track.get('song_id') or track_id,
+            'url': track_url,
+            'title': title,
+            'thumbnail': track.get('pic') or track.get('album_pic'),
+            'duration': int_or_none(track.get('length')),
+            'creator': track.get('artist', '').split(';')[0],
+            'track': title,
+            'album': track.get('album_name'),
+            'artist': track.get('artist'),
+            'subtitles': subtitles,
+        }
+
+    def _extract_tracks(self, item_id, typ=None):
+        playlist = self._download_json(
+            '%s/%s%s' % (self._API_BASE_URL, item_id, '/type/%s' % typ if typ else ''), item_id)
+        return [
+            self._extract_track(track, item_id)
+            for track in playlist['data']['trackList']]
+
+    @staticmethod
+    def _decrypt(origin):
+        n = int(origin[0])
+        origin = origin[1:]
+        short_lenth = len(origin) // n
+        long_num = len(origin) - short_lenth * n
+        l = tuple()
+        for i in range(0, n):
+            length = short_lenth
+            if i < long_num:
+                length += 1
+            l += (origin[0:length], )
+            origin = origin[length:]
+        ans = ''
+        for i in range(0, short_lenth + 1):
+            for j in range(0, n):
+                if len(l[j]) > i:
+                    ans += l[j][i]
+        return compat_urllib_parse_unquote(ans).replace('^', '0')
+
+
+class XiamiSongIE(XiamiBaseIE):
+    IE_NAME = 'xiami:song'
+    IE_DESC = '虾米音乐'
+    _VALID_URL = r'https?://(?:www\.)?xiami\.com/song/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'http://www.xiami.com/song/1775610518',
+        'md5': '521dd6bea40fd5c9c69f913c232cb57e',
+        'info_dict': {
+            'id': '1775610518',
+            'ext': 'mp3',
+            'title': 'Woman',
+            'thumbnail': r're:http://img\.xiami\.net/images/album/.*\.jpg',
+            'duration': 265,
+            'creator': 'HONNE',
+            'track': 'Woman',
+            'album': 'Woman',
+            'artist': 'HONNE',
+            'subtitles': {
+                'origin': [{
+                    'ext': 'lrc',
+                }],
+            },
+        }
+    }, {
+        'url': 'http://www.xiami.com/song/1775256504',
+        'md5': '932a3abd45c6aa2b1fdbe028fcb4c4fc',
+        'info_dict': {
+            'id': '1775256504',
+            'ext': 'mp3',
+            'title': '悟空',
+            'thumbnail': r're:http://img\.xiami\.net/images/album/.*\.jpg',
+            'duration': 200,
+            'creator': '戴荃',
+            'track': '悟空',
+            'album': '悟空',
+            'artist': '戴荃',
+            'subtitles': {
+                'origin': [{
+                    'ext': 'lrc',
+                }],
+            },
+        }
+    }]
+
+    def _real_extract(self, url):
+        return self._extract_tracks(self._match_id(url))[0]
+
+
+class XiamiPlaylistBaseIE(XiamiBaseIE):
+    def _real_extract(self, url):
+        item_id = self._match_id(url)
+        return self.playlist_result(self._extract_tracks(item_id, self._TYPE), item_id)
+
+
+class XiamiAlbumIE(XiamiPlaylistBaseIE):
+    IE_NAME = 'xiami:album'
+    IE_DESC = '虾米音乐 - 专辑'
+    _VALID_URL = r'https?://(?:www\.)?xiami\.com/album/(?P<id>[0-9]+)'
+    _TYPE = '1'
+    _TESTS = [{
+        'url': 'http://www.xiami.com/album/2100300444',
+        'info_dict': {
+            'id': '2100300444',
+        },
+        'playlist_count': 10,
+    }, {
+        'url': 'http://www.xiami.com/album/512288?spm=a1z1s.6843761.1110925389.6.hhE9p9',
+        'only_matching': True,
+    }]
+
+
+class XiamiArtistIE(XiamiPlaylistBaseIE):
+    IE_NAME = 'xiami:artist'
+    IE_DESC = '虾米音乐 - 歌手'
+    _VALID_URL = r'https?://(?:www\.)?xiami\.com/artist/(?P<id>[0-9]+)'
+    _TYPE = '2'
+    _TEST = {
+        'url': 'http://www.xiami.com/artist/2132?spm=0.0.0.0.dKaScp',
+        'info_dict': {
+            'id': '2132',
+        },
+        'playlist_count': 20,
+    }
+
+
+class XiamiCollectionIE(XiamiPlaylistBaseIE):
+    IE_NAME = 'xiami:collection'
+    IE_DESC = '虾米音乐 - 精选集'
+    _VALID_URL = r'https?://(?:www\.)?xiami\.com/collect/(?P<id>[0-9]+)'
+    _TYPE = '3'
+    _TEST = {
+        'url': 'http://www.xiami.com/collect/156527391?spm=a1z1s.2943601.6856193.12.4jpBnr',
+        'info_dict': {
+            'id': '156527391',
+        },
+        'playlist_mincount': 29,
+    }

youtube_dl/extractor/xminus.py

@@ -2,15 +2,15 @@
 from __future__ import unicode_literals
 
 import re
+import time
 
 from .common import InfoExtractor
 from ..compat import (
-    compat_chr,
     compat_ord,
 )
 from ..utils import (
     int_or_none,
-    parse_filesize,
+    parse_duration,
 )
 
 
@@ -22,7 +22,7 @@ class XMinusIE(InfoExtractor):
         'info_dict': {
             'id': '4542',
             'ext': 'mp3',
-            'title': 'Леонид Агутин-Песенка шофера',
+            'title': 'Леонид Агутин-Песенка шофёра',
             'duration': 156,
             'tbr': 320,
             'filesize_approx': 5900000,
@@ -36,38 +36,41 @@ class XMinusIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
 
         artist = self._html_search_regex(
-            r'minus_track\.artist="(.+?)"', webpage, 'artist')
+            r'<a[^>]+href="/artist/\d+">([^<]+)</a>', webpage, 'artist')
         title = artist + '-' + self._html_search_regex(
-            r'minus_track\.title="(.+?)"', webpage, 'title')
-        duration = int_or_none(self._html_search_regex(
-            r'minus_track\.dur_sec=\'([0-9]*?)\'',
+            r'<span[^>]+class="minustrack-full-title(?:\s+[^"]+)?"[^>]*>([^<]+)', webpage, 'title')
+        duration = parse_duration(self._html_search_regex(
+            r'<span[^>]+class="player-duration(?:\s+[^"]+)?"[^>]*>([^<]+)',
             webpage, 'duration', fatal=False))
-        filesize_approx = parse_filesize(self._html_search_regex(
-            r'<div id="finfo"[^>]*>\s*↓\s*([0-9.]+\s*[a-zA-Z][bB])',
-            webpage, 'approximate filesize', fatal=False))
-        tbr = int_or_none(self._html_search_regex(
-            r'<div class="quality[^"]*"></div>\s*([0-9]+)\s*kbps',
-            webpage, 'bitrate', fatal=False))
+        mobj = re.search(
+            r'<div[^>]+class="dw-info(?:\s+[^"]+)?"[^>]*>(?P<tbr>\d+)\s*кбит/c\s+(?P<filesize>[0-9.]+)\s*мб</div>',
+            webpage)
+        tbr = filesize_approx = None
+        if mobj:
+            filesize_approx = float(mobj.group('filesize')) * 1000000
+            tbr = float(mobj.group('tbr'))
         view_count = int_or_none(self._html_search_regex(
-            r'<div class="quality.*?► ([0-9]+)',
+            r'<span><[^>]+class="icon-chart-bar".*?>(\d+)</span>',
             webpage, 'view count', fatal=False))
         description = self._html_search_regex(
-            r'(?s)<div id="song_texts">(.*?)</div><br',
+            r'(?s)<pre[^>]+id="lyrics-original"[^>]*>(.*?)</pre>',
             webpage, 'song lyrics', fatal=False)
         if description:
             description = re.sub(' *\r *', '\n', description)
 
-        enc_token = self._html_search_regex(
-            r'minus_track\.s?tkn="(.+?)"', webpage, 'enc_token')
-        token = ''.join(
-            c if pos == 3 else compat_chr(compat_ord(c) - 1)
-            for pos, c in enumerate(reversed(enc_token)))
-        video_url = 'http://x-minus.org/dwlf/%s/%s.mp3' % (video_id, token)
+        k = self._search_regex(
+            r'<div[^>]+id="player-bottom"[^>]+data-k="([^"]+)">', webpage,
+            'encoded data')
+        h = time.time() / 3600
+        a = sum(map(int, [compat_ord(c) for c in k])) + int(video_id) + h
+        video_url = 'http://x-minus.me/dl/minus?id=%s&tkn2=%df%d' % (video_id, a, h)
 
         return {
             'id': video_id,
             'title': title,
             'url': video_url,
+            # The extension is unknown until actual downloading
+            'ext': 'mp3',
             'duration': duration,
             'filesize_approx': filesize_approx,
             'tbr': tbr,

youtube_dl/extractor/yandexmusic.py

@@ -18,9 +18,23 @@ from ..utils import (
 class YandexMusicBaseIE(InfoExtractor):
     @staticmethod
     def _handle_error(response):
-        error = response.get('error')
-        if error:
-            raise ExtractorError(error, expected=True)
+        if isinstance(response, dict):
+            error = response.get('error')
+            if error:
+                raise ExtractorError(error, expected=True)
+
+    def _download_webpage(self, *args, **kwargs):
+        webpage = super(YandexMusicBaseIE, self)._download_webpage(*args, **kwargs)
+        if 'Нам очень жаль, но запросы, поступившие с вашего IP-адреса, похожи на автоматические.' in webpage:
+            raise ExtractorError(
+                'YandexMusic has considered youtube-dl requests automated and '
+                'asks you to solve a CAPTCHA. You can either wait for some '
+                'time until unblocked and optionally use --sleep-interval '
+                'in future or alternatively you can go to https://music.yandex.ru/ '
+                'solve CAPTCHA, then export cookies and pass cookie file to '
+                'youtube-dl with --cookies',
+                expected=True)
+        return webpage
 
     def _download_json(self, *args, **kwargs):
         response = super(YandexMusicBaseIE, self)._download_json(*args, **kwargs)
@@ -47,7 +61,8 @@ class YandexMusicTrackIE(YandexMusicBaseIE):
             'album_artist': 'Carlo Ambrosio',
             'artist': 'Carlo Ambrosio & Fabio Di Bari, Carlo Ambrosio',
             'release_year': '2009',
-        }
+        },
+        'skip': 'Travis CI servers blocked by YandexMusic',
     }
 
     def _get_track_url(self, storage_dir, track_id):
@@ -139,6 +154,7 @@ class YandexMusicAlbumIE(YandexMusicPlaylistBaseIE):
             'title': 'Carlo Ambrosio - Gypsy Soul (2009)',
         },
         'playlist_count': 50,
+        'skip': 'Travis CI servers blocked by YandexMusic',
     }
 
     def _real_extract(self, url):
@@ -171,6 +187,7 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
             'description': 'md5:3b9f27b0efbe53f2ee1e844d07155cc9',
         },
         'playlist_count': 6,
+        'skip': 'Travis CI servers blocked by YandexMusic',
     }, {
         # playlist exceeding the limit of 150 tracks shipped with webpage (see
         # https://github.com/rg3/youtube-dl/issues/6666)
@@ -180,6 +197,7 @@ class YandexMusicPlaylistIE(YandexMusicPlaylistBaseIE):
             'title': 'Музыка 90-х',
         },
         'playlist_count': 310,
+        'skip': 'Travis CI servers blocked by YandexMusic',
     }]
 
     def _real_extract(self, url):

youtube_dl/extractor/youtube.py

@@ -2139,10 +2139,11 @@ class YoutubeSearchDateIE(YoutubeSearchIE):
     _EXTRA_QUERY_ARGS = {'search_sort': 'video_date_uploaded'}
 
 
-class YoutubeSearchURLIE(InfoExtractor):
+class YoutubeSearchURLIE(YoutubePlaylistBaseInfoExtractor):
     IE_DESC = 'YouTube.com search URLs'
     IE_NAME = 'youtube:search_url'
     _VALID_URL = r'https?://(?:www\.)?youtube\.com/results\?(.*?&)?(?:search_query|q)=(?P<query>[^&]+)(?:[&]|$)'
+    _VIDEO_RE = r'href="\s*/watch\?v=(?P<id>[0-9A-Za-z_-]{11})(?:[^"]*"[^>]+\btitle="(?P<title>[^"]+))?'
     _TESTS = [{
         'url': 'https://www.youtube.com/results?baz=bar&search_query=youtube-dl+test+video&filters=video&lclk=video',
         'playlist_mincount': 5,
@@ -2157,32 +2158,8 @@ class YoutubeSearchURLIE(InfoExtractor):
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
         query = compat_urllib_parse_unquote_plus(mobj.group('query'))
-
         webpage = self._download_webpage(url, query)
-        result_code = self._search_regex(
-            r'(?s)<ol[^>]+class="item-section"(.*?)</ol>', webpage, 'result HTML')
-
-        part_codes = re.findall(
-            r'(?s)<h3[^>]+class="[^"]*yt-lockup-title[^"]*"[^>]*>(.*?)</h3>', result_code)
-        entries = []
-        for part_code in part_codes:
-            part_title = self._html_search_regex(
-                [r'(?s)title="([^"]+)"', r'>([^<]+)</a>'], part_code, 'item title', fatal=False)
-            part_url_snippet = self._html_search_regex(
-                r'(?s)href="([^"]+)"', part_code, 'item URL')
-            part_url = compat_urlparse.urljoin(
-                'https://www.youtube.com/', part_url_snippet)
-            entries.append({
-                '_type': 'url',
-                'url': part_url,
-                'title': part_title,
-            })
-
-        return {
-            '_type': 'playlist',
-            'entries': entries,
-            'title': query,
-        }
+        return self.playlist_result(self._process_page(webpage), playlist_title=query)
 
 
 class YoutubeShowIE(YoutubePlaylistsBaseInfoExtractor):

youtube_dl/postprocessor/ffmpeg.py

@@ -389,23 +389,30 @@ class FFmpegEmbedSubtitlePP(FFmpegPostProcessor):
 class FFmpegMetadataPP(FFmpegPostProcessor):
     def run(self, info):
         metadata = {}
-        if info.get('title') is not None:
-            metadata['title'] = info['title']
-        if info.get('upload_date') is not None:
-            metadata['date'] = info['upload_date']
-        if info.get('artist') is not None:
-            metadata['artist'] = info['artist']
-        elif info.get('uploader') is not None:
-            metadata['artist'] = info['uploader']
-        elif info.get('uploader_id') is not None:
-            metadata['artist'] = info['uploader_id']
-        if info.get('description') is not None:
-            metadata['description'] = info['description']
-            metadata['comment'] = info['description']
-        if info.get('webpage_url') is not None:
-            metadata['purl'] = info['webpage_url']
-        if info.get('album') is not None:
-            metadata['album'] = info['album']
+
+        def add(meta_list, info_list=None):
+            if not info_list:
+                info_list = meta_list
+            if not isinstance(meta_list, (list, tuple)):
+                meta_list = (meta_list,)
+            if not isinstance(info_list, (list, tuple)):
+                info_list = (info_list,)
+            for info_f in info_list:
+                if info.get(info_f) is not None:
+                    for meta_f in meta_list:
+                        metadata[meta_f] = info[info_f]
+                    break
+
+        add('title', ('track', 'title'))
+        add('date', 'upload_date')
+        add(('description', 'comment'), 'description')
+        add('purl', 'webpage_url')
+        add('track', 'track_number')
+        add('artist', ('artist', 'creator', 'uploader', 'uploader_id'))
+        add('genre')
+        add('album')
+        add('album_artist')
+        add('disc', 'disc_number')
 
         if not metadata:
             self._downloader.to_screen('[ffmpeg] There isn\'t any metadata to add')

youtube_dl/utils.py

@@ -1935,6 +1935,9 @@ def error_to_compat_str(err):
 
 
 def mimetype2ext(mt):
+    if mt is None:
+        return None
+
     ext = {
         'audio/mp4': 'm4a',
     }.get(mt)

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.04.24'
+__version__ = '2016.05.01'