release 2015.02.28

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

README.md

@@ -408,6 +408,8 @@ which means you can modify it, redistribute it or use it however you like.
                                      downloading, similar to find's -exec
                                      syntax. Example: --exec 'adb push {}
                                      /sdcard/Music/ && rm {}'
+    --convert-subtitles FORMAT       Convert the subtitles to other format
+                                     (currently supported: srt|ass|vtt)
 
 # CONFIGURATION
 

docs/supportedsites.md

@@ -210,6 +210,7 @@
  - **Jove**
  - **jpopsuki.tv**
  - **Jukebox**
+ - **Kaltura**
  - **Kankan**
  - **Karaoketv**
  - **keek**
@@ -308,6 +309,7 @@
  - **Nuvid**
  - **NYTimes**
  - **ocw.mit.edu**
+ - **Odnoklassniki**
  - **OktoberfestTV**
  - **on.aol.com**
  - **Ooyala**
@@ -334,6 +336,7 @@
  - **PornoXO**
  - **PromptFile**
  - **prosiebensat1**: ProSiebenSat.1 Digital
+ - **Puls4**
  - **Pyvideo**
  - **QuickVid**
  - **R7**
@@ -412,7 +415,7 @@
  - **StreamCZ**
  - **StreetVoice**
  - **SunPorno**
- - **SVTPlay**
+ - **SVTPlay**: SVT Play and Öppet arkiv
  - **SWRMediathek**
  - **Syfy**
  - **SztvHu**

youtube_dl/YoutubeDL.py

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

youtube_dl/__init__.py

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

youtube_dl/extractor/__init__.py

@@ -227,6 +227,7 @@ from .jeuxvideo import JeuxVideoIE
 from .jove import JoveIE
 from .jukebox import JukeboxIE
 from .jpopsukitv import JpopsukiIE
+from .kaltura import KalturaIE
 from .kankan import KankanIE
 from .karaoketv import KaraoketvIE
 from .keezmovies import KeezMoviesIE
@@ -345,6 +346,7 @@ from .ntvde import NTVDeIE
 from .ntvru import NTVRuIE
 from .nytimes import NYTimesIE
 from .nuvid import NuvidIE
+from .odnoklassniki import OdnoklassnikiIE
 from .oktoberfesttv import OktoberfestTVIE
 from .ooyala import OoyalaIE
 from .openfilm import OpenFilmIE
@@ -372,6 +374,7 @@ from .pornotube import PornotubeIE
 from .pornoxo import PornoXOIE
 from .promptfile import PromptFileIE
 from .prosiebensat1 import ProSiebenSat1IE
+from .puls4 import Puls4IE
 from .pyvideo import PyvideoIE
 from .quickvid import QuickVidIE
 from .r7 import R7IE

youtube_dl/extractor/comedycentral.py

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

youtube_dl/extractor/escapist.py

@@ -8,6 +8,7 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     js_to_json,
+    parse_duration,
 )
 
 
@@ -25,6 +26,7 @@ class EscapistIE(InfoExtractor):
             'uploader': 'The Escapist Presents',
             'title': "Breaking Down Baldur's Gate",
             'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 264,
         }
     }
 
@@ -41,6 +43,7 @@ class EscapistIE(InfoExtractor):
             r"<h1\s+class='headline'>(.*?)</a>",
             webpage, 'uploader', fatal=False)
         description = self._html_search_meta('description', webpage)
+        duration = parse_duration(self._html_search_meta('duration', webpage))
 
         raw_title = self._html_search_meta('title', webpage, fatal=True)
         title = raw_title.partition(' : ')[2]
@@ -105,6 +108,7 @@ class EscapistIE(InfoExtractor):
             'title': title,
             'thumbnail': self._og_search_thumbnail(webpage),
             'description': description,
+            'duration': duration,
         }
 
         if self._downloader.params.get('include_ads') and ad_formats:

youtube_dl/extractor/gamestar.py

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

youtube_dl/extractor/generic.py

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

youtube_dl/extractor/kaltura.py

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

youtube_dl/extractor/laola1tv.py

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

youtube_dl/extractor/lynda.py

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

youtube_dl/extractor/mpora.py

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

youtube_dl/extractor/musicvault.py

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

youtube_dl/extractor/odnoklassniki.py

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

youtube_dl/extractor/puls4.py

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

youtube_dl/extractor/rtve.py

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

youtube_dl/extractor/svtplay.py

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

youtube_dl/extractor/twitch.py

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

youtube_dl/options.py

@@ -751,6 +751,10 @@ def parseOpts(overrideArguments=None):
         '--exec',
         metavar='CMD', dest='exec_cmd',
         help='Execute a command on the file after downloading, similar to find\'s -exec syntax. Example: --exec \'adb push {} /sdcard/Music/ && rm {}\'')
+    postproc.add_option(
+        '--convert-subtitles', '--convert-subs',
+        metavar='FORMAT', dest='convertsubtitles', default=None,
+        help='Convert the subtitles to other format (currently supported: srt|ass|vtt)')
 
     parser.add_option_group(general)
     parser.add_option_group(network)

youtube_dl/postprocessor/__init__.py

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

youtube_dl/postprocessor/ffmpeg.py

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

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2015.02.26.2'
+__version__ = '2015.02.28'