release 2016.09.18

- Eliminate segment_urls and initialization_url
+ Introduce manifest_url (manifest may contain unfragmented data in this case url will be used for direct media URL and manifest_url for manifest itself correspondingly)
* Rewrite dashsegments downloader to use fragments data
* Improve generic mpd extraction

.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 *2016.09.15*. 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 Issues with outdated version will be rejected.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 **2016.09.15**
+### Make sure you are using the *latest* version: run `youtube-dl --version` and ensure your version is *2016.09.18*. 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 Issues with outdated version will be rejected.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 **2016.09.18**
 
 ### 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 2016.09.15
+[debug] youtube-dl version 2016.09.18
 [debug] Python version 2.7.11 - Windows-2003Server-5.2.3790-SP2
 [debug] exe versions: ffmpeg N-75573-g1d0487f, ffprobe N-75573-g1d0487f, rtmpdump 2.4
 [debug] Proxy map: {}

ChangeLog

@@ -1,3 +1,25 @@
+version 2016.09.18
+
+Core
++ Introduce manifest_url and fragments fields in formats dictionary for
+  fragmented media
++ Provide manifest_url field for DASH segments, HLS and HDS
++ Provide fragments field for DASH segments
+* Rework DASH segments downloader to use fragments field
++ Add helper method for Wowza Streaming Engine formats extraction
+
+Extractors
++ [vyborymos] Add extractor for vybory.mos.ru (#10692)
++ [xfileshare] Add title regular expression for streamin.to (#10646)
++ [globo:article] Add support for multiple videos (#10653)
++ [thisav] Recognize HTML5 videos (#10447)
+* [jwplatform] Improve JWPlayer detection
++ [mangomolo] Add support for Mangomolo embeds
++ [toutv] Add support for authentication (#10669)
+* [franceinter] Fix upload date extraction
+* [tv4] Fix HLS and HDS formats extraction (#10659)
+
+
 version 2016.09.15
 
 Core

docs/supportedsites.md

@@ -388,6 +388,8 @@
  - **mailru**: Видео@Mail.Ru
  - **MakersChannel**
  - **MakerTV**
+ - **mangomolo:live**
+ - **mangomolo:video**
  - **MatchTV**
  - **MDR**: MDR.DE and KiKA
  - **media.ccc.de**
@@ -849,6 +851,7 @@
  - **VRT**
  - **vube**: Vube.com
  - **VuClip**
+ - **VyboryMos**
  - **Walla**
  - **washingtonpost**
  - **washingtonpost:article**

youtube_dl/downloader/dash.py

@@ -1,7 +1,6 @@
 from __future__ import unicode_literals
 
 import os
-import re
 
 from .fragment import FragmentFD
 from ..compat import compat_urllib_error
@@ -19,34 +18,32 @@ class DashSegmentsFD(FragmentFD):
     FD_NAME = 'dashsegments'
 
     def real_download(self, filename, info_dict):
-        base_url = info_dict['url']
-        segment_urls = [info_dict['segment_urls'][0]] if self.params.get('test', False) else info_dict['segment_urls']
-        initialization_url = info_dict.get('initialization_url')
+        segments = info_dict['fragments'][:1] if self.params.get(
+            'test', False) else info_dict['fragments']
 
         ctx = {
             'filename': filename,
-            'total_frags': len(segment_urls) + (1 if initialization_url else 0),
+            'total_frags': len(segments),
         }
 
         self._prepare_and_start_frag_download(ctx)
 
-        def combine_url(base_url, target_url):
-            if re.match(r'^https?://', target_url):
-                return target_url
-            return '%s%s%s' % (base_url, '' if base_url.endswith('/') else '/', target_url)
-
         segments_filenames = []
 
         fragment_retries = self.params.get('fragment_retries', 0)
         skip_unavailable_fragments = self.params.get('skip_unavailable_fragments', True)
 
-        def process_segment(segment, tmp_filename, fatal):
-            target_url, segment_name = segment
+        def process_segment(segment, tmp_filename, num):
+            segment_url = segment['url']
+            segment_name = 'Frag%d' % num
             target_filename = '%s-%s' % (tmp_filename, segment_name)
+            # In DASH, the first segment contains necessary headers to
+            # generate a valid MP4 file, so always abort for the first segment
+            fatal = num == 0 or not skip_unavailable_fragments
             count = 0
             while count <= fragment_retries:
                 try:
-                    success = ctx['dl'].download(target_filename, {'url': combine_url(base_url, target_url)})
+                    success = ctx['dl'].download(target_filename, {'url': segment_url})
                     if not success:
                         return False
                     down, target_sanitized = sanitize_open(target_filename, 'rb')
@@ -72,16 +69,8 @@ class DashSegmentsFD(FragmentFD):
                 return False
             return True
 
-        segments_to_download = [(initialization_url, 'Init')] if initialization_url else []
-        segments_to_download.extend([
-            (segment_url, 'Seg%d' % i)
-            for i, segment_url in enumerate(segment_urls)])
-
-        for i, segment in enumerate(segments_to_download):
-            # In DASH, the first segment contains necessary headers to
-            # generate a valid MP4 file, so always abort for the first segment
-            fatal = i == 0 or not skip_unavailable_fragments
-            if not process_segment(segment, ctx['tmpfilename'], fatal):
+        for i, segment in enumerate(segments):
+            if not process_segment(segment, ctx['tmpfilename'], i):
                 return False
 
         self._finish_frag_download(ctx)

youtube_dl/extractor/awaan.py

@@ -50,25 +50,6 @@ class AWAANBaseIE(InfoExtractor):
             'is_live': is_live,
         }
 
-    def _extract_video_formats(self, webpage, video_id, m3u8_entry_protocol):
-        formats = []
-        format_url_base = 'http' + self._html_search_regex(
-            [
-                r'file\s*:\s*"https?(://[^"]+)/playlist.m3u8',
-                r'<a[^>]+href="rtsp(://[^"]+)"'
-            ], webpage, 'format url')
-        formats.extend(self._extract_mpd_formats(
-            format_url_base + '/manifest.mpd',
-            video_id, mpd_id='dash', fatal=False))
-        formats.extend(self._extract_m3u8_formats(
-            format_url_base + '/playlist.m3u8', video_id, 'mp4',
-            m3u8_entry_protocol, m3u8_id='hls', fatal=False))
-        formats.extend(self._extract_f4m_formats(
-            format_url_base + '/manifest.f4m',
-            video_id, f4m_id='hds', fatal=False))
-        self._sort_formats(formats)
-        return formats
-
 
 class AWAANVideoIE(AWAANBaseIE):
     IE_NAME = 'awaan:video'
@@ -99,16 +80,18 @@ class AWAANVideoIE(AWAANBaseIE):
             video_id, headers={'Origin': 'http://awaan.ae'})
         info = self._parse_video_data(video_data, video_id, False)
 
-        webpage = self._download_webpage(
-            'http://admin.mangomolo.com/analytics/index.php/customers/embed/video?' +
-            compat_urllib_parse_urlencode({
-                'id': video_data['id'],
-                'user_id': video_data['user_id'],
-                'signature': video_data['signature'],
-                'countries': 'Q0M=',
-                'filter': 'DENY',
-            }), video_id)
-        info['formats'] = self._extract_video_formats(webpage, video_id, 'm3u8_native')
+        embed_url = 'http://admin.mangomolo.com/analytics/index.php/customers/embed/video?' + compat_urllib_parse_urlencode({
+            'id': video_data['id'],
+            'user_id': video_data['user_id'],
+            'signature': video_data['signature'],
+            'countries': 'Q0M=',
+            'filter': 'DENY',
+        })
+        info.update({
+            '_type': 'url_transparent',
+            'url': embed_url,
+            'ie_key': 'MangomoloVideo',
+        })
         return info
 
 
@@ -138,16 +121,18 @@ class AWAANLiveIE(AWAANBaseIE):
             channel_id, headers={'Origin': 'http://awaan.ae'})
         info = self._parse_video_data(channel_data, channel_id, True)
 
-        webpage = self._download_webpage(
-            'http://admin.mangomolo.com/analytics/index.php/customers/embed/index?' +
-            compat_urllib_parse_urlencode({
-                'id': base64.b64encode(channel_data['user_id'].encode()).decode(),
-                'channelid': base64.b64encode(channel_data['id'].encode()).decode(),
-                'signature': channel_data['signature'],
-                'countries': 'Q0M=',
-                'filter': 'DENY',
-            }), channel_id)
-        info['formats'] = self._extract_video_formats(webpage, channel_id, 'm3u8')
+        embed_url = 'http://admin.mangomolo.com/analytics/index.php/customers/embed/index?' + compat_urllib_parse_urlencode({
+            'id': base64.b64encode(channel_data['user_id'].encode()).decode(),
+            'channelid': base64.b64encode(channel_data['id'].encode()).decode(),
+            'signature': channel_data['signature'],
+            'countries': 'Q0M=',
+            'filter': 'DENY',
+        })
+        info.update({
+            '_type': 'url_transparent',
+            'url': embed_url,
+            'ie_key': 'MangomoloLive',
+        })
         return info
 
 

youtube_dl/extractor/common.py

@@ -87,6 +87,9 @@ class InfoExtractor(object):
 
                     Potential fields:
                     * url        Mandatory. The URL of the video file
+                    * manifest_url
+                                 The URL of the manifest file in case of
+                                 fragmented media (DASH, hls, hds)
                     * ext        Will be calculated from URL if missing
                     * format     A human-readable description of the format
                                  ("mp4 container with h264/opus").
@@ -115,6 +118,11 @@ class InfoExtractor(object):
                                  download, lower-case.
                                  "http", "https", "rtsp", "rtmp", "rtmpe",
                                  "m3u8", "m3u8_native" or "http_dash_segments".
+                    * fragments  A list of fragments of the fragmented media,
+                                 with the following entries:
+                                 * "url" (mandatory) - fragment's URL
+                                 * "duration" (optional, int or float)
+                                 * "filesize" (optional, int)
                     * preference Order number of this format. If this field is
                                  present and not None, the formats get sorted
                                  by this field, regardless of all other values.
@@ -1142,6 +1150,7 @@ class InfoExtractor(object):
             formats.append({
                 'format_id': format_id,
                 'url': manifest_url,
+                'manifest_url': manifest_url,
                 'ext': 'flv' if bootstrap_info is not None else None,
                 'tbr': tbr,
                 'width': width,
@@ -1247,9 +1256,11 @@ class InfoExtractor(object):
                 # format_id intact.
                 if not live:
                     format_id.append(stream_name if stream_name else '%d' % (tbr if tbr else len(formats)))
+                manifest_url = format_url(line.strip())
                 f = {
                     'format_id': '-'.join(format_id),
-                    'url': format_url(line.strip()),
+                    'url': manifest_url,
+                    'manifest_url': manifest_url,
                     'tbr': tbr,
                     'ext': ext,
                     'fps': float_or_none(last_info.get('FRAME-RATE')),
@@ -1521,9 +1532,10 @@ class InfoExtractor(object):
         mpd_base_url = re.match(r'https?://.+/', urlh.geturl()).group()
 
         return self._parse_mpd_formats(
-            compat_etree_fromstring(mpd.encode('utf-8')), mpd_id, mpd_base_url, formats_dict=formats_dict)
+            compat_etree_fromstring(mpd.encode('utf-8')), mpd_id, mpd_base_url,
+            formats_dict=formats_dict, mpd_url=mpd_url)
 
-    def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}):
+    def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}, mpd_url=None):
         """
         Parse formats from MPD manifest.
         References:
@@ -1544,42 +1556,52 @@ class InfoExtractor(object):
 
         def extract_multisegment_info(element, ms_parent_info):
             ms_info = ms_parent_info.copy()
+
+            # As per [1, 5.3.9.2.2] SegmentList and SegmentTemplate share some
+            # common attributes and elements.  We will only extract relevant
+            # for us.
+            def extract_common(source):
+                segment_timeline = source.find(_add_ns('SegmentTimeline'))
+                if segment_timeline is not None:
+                    s_e = segment_timeline.findall(_add_ns('S'))
+                    if s_e:
+                        ms_info['total_number'] = 0
+                        ms_info['s'] = []
+                        for s in s_e:
+                            r = int(s.get('r', 0))
+                            ms_info['total_number'] += 1 + r
+                            ms_info['s'].append({
+                                't': int(s.get('t', 0)),
+                                # @d is mandatory (see [1, 5.3.9.6.2, Table 17, page 60])
+                                'd': int(s.attrib['d']),
+                                'r': r,
+                            })
+                start_number = source.get('startNumber')
+                if start_number:
+                    ms_info['start_number'] = int(start_number)
+                timescale = source.get('timescale')
+                if timescale:
+                    ms_info['timescale'] = int(timescale)
+                segment_duration = source.get('duration')
+                if segment_duration:
+                    ms_info['segment_duration'] = int(segment_duration)
+
+            def extract_Initialization(source):
+                initialization = source.find(_add_ns('Initialization'))
+                if initialization is not None:
+                    ms_info['initialization_url'] = initialization.attrib['sourceURL']
+
             segment_list = element.find(_add_ns('SegmentList'))
             if segment_list is not None:
+                extract_common(segment_list)
+                extract_Initialization(segment_list)
                 segment_urls_e = segment_list.findall(_add_ns('SegmentURL'))
                 if segment_urls_e:
                     ms_info['segment_urls'] = [segment.attrib['media'] for segment in segment_urls_e]
-                initialization = segment_list.find(_add_ns('Initialization'))
-                if initialization is not None:
-                    ms_info['initialization_url'] = initialization.attrib['sourceURL']
             else:
                 segment_template = element.find(_add_ns('SegmentTemplate'))
                 if segment_template is not None:
-                    start_number = segment_template.get('startNumber')
-                    if start_number:
-                        ms_info['start_number'] = int(start_number)
-                    segment_timeline = segment_template.find(_add_ns('SegmentTimeline'))
-                    if segment_timeline is not None:
-                        s_e = segment_timeline.findall(_add_ns('S'))
-                        if s_e:
-                            ms_info['total_number'] = 0
-                            ms_info['s'] = []
-                            for s in s_e:
-                                r = int(s.get('r', 0))
-                                ms_info['total_number'] += 1 + r
-                                ms_info['s'].append({
-                                    't': int(s.get('t', 0)),
-                                    # @d is mandatory (see [1, 5.3.9.6.2, Table 17, page 60])
-                                    'd': int(s.attrib['d']),
-                                    'r': r,
-                                })
-                    else:
-                        timescale = segment_template.get('timescale')
-                        if timescale:
-                            ms_info['timescale'] = int(timescale)
-                        segment_duration = segment_template.get('duration')
-                        if segment_duration:
-                            ms_info['segment_duration'] = int(segment_duration)
+                    extract_common(segment_template)
                     media_template = segment_template.get('media')
                     if media_template:
                         ms_info['media_template'] = media_template
@@ -1587,11 +1609,14 @@ class InfoExtractor(object):
                     if initialization:
                         ms_info['initialization_url'] = initialization
                     else:
-                        initialization = segment_template.find(_add_ns('Initialization'))
-                        if initialization is not None:
-                            ms_info['initialization_url'] = initialization.attrib['sourceURL']
+                        extract_Initialization(segment_template)
             return ms_info
 
+        def combine_url(base_url, target_url):
+            if re.match(r'^https?://', target_url):
+                return target_url
+            return '%s%s%s' % (base_url, '' if base_url.endswith('/') else '/', target_url)
+
         mpd_duration = parse_duration(mpd_doc.get('mediaPresentationDuration'))
         formats = []
         for period in mpd_doc.findall(_add_ns('Period')):
@@ -1634,6 +1659,7 @@ class InfoExtractor(object):
                         f = {
                             'format_id': '%s-%s' % (mpd_id, representation_id) if mpd_id else representation_id,
                             'url': base_url,
+                            'manifest_url': mpd_url,
                             'ext': mimetype2ext(mime_type),
                             'width': int_or_none(representation_attrib.get('width')),
                             'height': int_or_none(representation_attrib.get('height')),
@@ -1648,9 +1674,7 @@ class InfoExtractor(object):
                         }
                         representation_ms_info = extract_multisegment_info(representation, adaption_set_ms_info)
                         if 'segment_urls' not in representation_ms_info and 'media_template' in representation_ms_info:
-                            if 'total_number' not in representation_ms_info and 'segment_duration':
-                                segment_duration = float(representation_ms_info['segment_duration']) / float(representation_ms_info['timescale'])
-                                representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration))
+
                             media_template = representation_ms_info['media_template']
                             media_template = media_template.replace('$RepresentationID$', representation_id)
                             media_template = re.sub(r'\$(Number|Bandwidth|Time)\$', r'%(\1)d', media_template)
@@ -1659,46 +1683,79 @@ class InfoExtractor(object):
 
                             # As per [1, 5.3.9.4.4, Table 16, page 55] $Number$ and $Time$
                             # can't be used at the same time
-                            if '%(Number' in media_template:
-                                representation_ms_info['segment_urls'] = [
-                                    media_template % {
+                            if '%(Number' in media_template and 's' not in representation_ms_info:
+                                segment_duration = None
+                                if 'total_number' not in representation_ms_info and 'segment_duration':
+                                    segment_duration = float_or_none(representation_ms_info['segment_duration'], representation_ms_info['timescale'])
+                                    representation_ms_info['total_number'] = int(math.ceil(float(period_duration) / segment_duration))
+                                representation_ms_info['fragments'] = [{
+                                    'url': media_template % {
                                         'Number': segment_number,
                                         'Bandwidth': representation_attrib.get('bandwidth'),
-                                    }
-                                    for segment_number in range(
-                                        representation_ms_info['start_number'],
-                                        representation_ms_info['total_number'] + representation_ms_info['start_number'])]
+                                    },
+                                    'duration': segment_duration,
+                                } for segment_number in range(
+                                    representation_ms_info['start_number'],
+                                    representation_ms_info['total_number'] + representation_ms_info['start_number'])]
                             else:
-                                representation_ms_info['segment_urls'] = []
+                                # $Number*$ or $Time$ in media template with S list available
+                                # Example $Number*$: http://www.svtplay.se/klipp/9023742/stopptid-om-bjorn-borg
+                                # Example $Time$: https://play.arkena.com/embed/avp/v2/player/media/b41dda37-d8e7-4d3f-b1b5-9a9db578bdfe/1/129411
+                                representation_ms_info['fragments'] = []
                                 segment_time = 0
+                                segment_d = None
+                                segment_number = representation_ms_info['start_number']
 
                                 def add_segment_url():
-                                    representation_ms_info['segment_urls'].append(
-                                        media_template % {
-                                            'Time': segment_time,
-                                            'Bandwidth': representation_attrib.get('bandwidth'),
-                                        }
-                                    )
+                                    segment_url = media_template % {
+                                        'Time': segment_time,
+                                        'Bandwidth': representation_attrib.get('bandwidth'),
+                                        'Number': segment_number,
+                                    }
+                                    representation_ms_info['fragments'].append({
+                                        'url': segment_url,
+                                        'duration': float_or_none(segment_d, representation_ms_info['timescale']),
+                                    })
 
                                 for num, s in enumerate(representation_ms_info['s']):
                                     segment_time = s.get('t') or segment_time
+                                    segment_d = s['d']
                                     add_segment_url()
+                                    segment_number += 1
                                     for r in range(s.get('r', 0)):
-                                        segment_time += s['d']
+                                        segment_time += segment_d
                                         add_segment_url()
-                                    segment_time += s['d']
-                        if 'segment_urls' in representation_ms_info:
+                                        segment_number += 1
+                                    segment_time += segment_d
+                        elif 'segment_urls' in representation_ms_info and 's' in representation_ms_info:
+                            # No media template
+                            # Example: https://www.youtube.com/watch?v=iXZV5uAYMJI
+                            # or any YouTube dashsegments video
+                            fragments = []
+                            s_num = 0
+                            for segment_url in representation_ms_info['segment_urls']:
+                                s = representation_ms_info['s'][s_num]
+                                for r in range(s.get('r', 0) + 1):
+                                    fragments.append({
+                                        'url': segment_url,
+                                        'duration': float_or_none(s['d'], representation_ms_info['timescale']),
+                                    })
+                            representation_ms_info['fragments'] = fragments
+                        # NB: MPD manifest may contain direct URLs to unfragmented media.
+                        # No fragments key is present in this case.
+                        if 'fragments' in representation_ms_info:
                             f.update({
-                                'segment_urls': representation_ms_info['segment_urls'],
+                                'fragments': [],
                                 'protocol': 'http_dash_segments',
                             })
                             if 'initialization_url' in representation_ms_info:
                                 initialization_url = representation_ms_info['initialization_url'].replace('$RepresentationID$', representation_id)
-                                f.update({
-                                    'initialization_url': initialization_url,
-                                })
                                 if not f.get('url'):
                                     f['url'] = initialization_url
+                                f['fragments'].append({'url': initialization_url})
+                            f['fragments'].extend(representation_ms_info['fragments'])
+                            for fragment in f['fragments']:
+                                fragment['url'] = combine_url(base_url, fragment['url'])
                         try:
                             existing_format = next(
                                 fo for fo in formats
@@ -1795,6 +1852,49 @@ class InfoExtractor(object):
             m3u8_id='hls', fatal=False))
         return formats
 
+    def _extract_wowza_formats(self, url, video_id, m3u8_entry_protocol='m3u8_native', skip_protocols=[]):
+        url = re.sub(r'/(?:manifest|playlist|jwplayer)\.(?:m3u8|f4m|mpd|smil)', '', url)
+        url_base = self._search_regex(r'(?:https?|rtmp|rtsp)(://[^?]+)', url, 'format url')
+        http_base_url = 'http' + url_base
+        formats = []
+        if 'm3u8' not in skip_protocols:
+            formats.extend(self._extract_m3u8_formats(
+                http_base_url + '/playlist.m3u8', video_id, 'mp4',
+                m3u8_entry_protocol, m3u8_id='hls', fatal=False))
+        if 'f4m' not in skip_protocols:
+            formats.extend(self._extract_f4m_formats(
+                http_base_url + '/manifest.f4m',
+                video_id, f4m_id='hds', fatal=False))
+        if re.search(r'(?:/smil:|\.smil)', url_base):
+            if 'dash' not in skip_protocols:
+                formats.extend(self._extract_mpd_formats(
+                    http_base_url + '/manifest.mpd',
+                    video_id, mpd_id='dash', fatal=False))
+            if 'smil' not in skip_protocols:
+                rtmp_formats = self._extract_smil_formats(
+                    http_base_url + '/jwplayer.smil',
+                    video_id, fatal=False)
+                for rtmp_format in rtmp_formats:
+                    rtsp_format = rtmp_format.copy()
+                    rtsp_format['url'] = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path'])
+                    del rtsp_format['play_path']
+                    del rtsp_format['ext']
+                    rtsp_format.update({
+                        'url': rtsp_format['url'].replace('rtmp://', 'rtsp://'),
+                        'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'),
+                        'protocol': 'rtsp',
+                    })
+                    formats.extend([rtmp_format, rtsp_format])
+        else:
+            for protocol in ('rtmp', 'rtsp'):
+                if protocol not in skip_protocols:
+                    formats.append({
+                        'url': protocol + url_base,
+                        'format_id': protocol,
+                        'protocol': protocol,
+                    })
+        return formats
+
     def _live_title(self, name):
         """ Generate the title for a live video """
         now = datetime.datetime.now()

youtube_dl/extractor/extractors.py

@@ -472,6 +472,10 @@ from .macgamestore import MacGameStoreIE
 from .mailru import MailRuIE
 from .makerschannel import MakersChannelIE
 from .makertv import MakerTVIE
+from .mangomolo import (
+    MangomoloVideoIE,
+    MangomoloLiveIE,
+)
 from .matchtv import MatchTVIE
 from .mdr import MDRIE
 from .meta import METAIE
@@ -1065,6 +1069,7 @@ from .vporn import VpornIE
 from .vrt import VRTIE
 from .vube import VubeIE
 from .vuclip import VuClipIE
+from .vyborymos import VyboryMosIE
 from .walla import WallaIE
 from .washingtonpost import (
     WashingtonPostIE,

youtube_dl/extractor/franceinter.py

@@ -2,7 +2,6 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..compat import compat_str
 from ..utils import month_by_name
 
 
@@ -10,14 +9,14 @@ class FranceInterIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?franceinter\.fr/emissions/(?P<id>[^?#]+)'
 
     _TEST = {
-        'url': 'https://www.franceinter.fr/emissions/la-marche-de-l-histoire/la-marche-de-l-histoire-18-decembre-2013',
-        'md5': '4764932e466e6f6c79c317d2e74f6884',
+        'url': 'https://www.franceinter.fr/emissions/affaires-sensibles/affaires-sensibles-07-septembre-2016',
+        'md5': '9e54d7bdb6fdc02a841007f8a975c094',
         'info_dict': {
-            'id': 'la-marche-de-l-histoire/la-marche-de-l-histoire-18-decembre-2013',
+            'id': 'affaires-sensibles/affaires-sensibles-07-septembre-2016',
             'ext': 'mp3',
-            'title': 'L’Histoire dans les jeux vidéo du 18 décembre 2013 - France Inter',
-            'description': 'md5:7f2ce449894d1e585932273080fb410d',
-            'upload_date': '20131218',
+            'title': 'Affaire Cahuzac : le contentieux du compte en Suisse',
+            'description': 'md5:401969c5d318c061f86bda1fa359292b',
+            'upload_date': '20160907',
         },
     }
 
@@ -39,7 +38,8 @@ class FranceInterIE(InfoExtractor):
         if upload_date_str:
             upload_date_list = upload_date_str.split()
             upload_date_list.reverse()
-            upload_date_list[1] = compat_str(month_by_name(upload_date_list[1], lang='fr'))
+            upload_date_list[1] = '%02d' % (month_by_name(upload_date_list[1], lang='fr') or 0)
+            upload_date_list[2] = '%02d' % int(upload_date_list[2])
             upload_date = ''.join(upload_date_list)
         else:
             upload_date = None

youtube_dl/extractor/generic.py

@@ -1657,7 +1657,9 @@ class GenericIE(InfoExtractor):
                 return self.playlist_result(self._parse_xspf(doc, video_id), video_id)
             elif re.match(r'(?i)^(?:{[^}]+})?MPD$', doc.tag):
                 info_dict['formats'] = self._parse_mpd_formats(
-                    doc, video_id, mpd_base_url=url.rpartition('/')[0])
+                    doc, video_id,
+                    mpd_base_url=full_response.geturl().rpartition('/')[0],
+                    mpd_url=url)
                 self._sort_formats(info_dict['formats'])
                 return info_dict
             elif re.match(r'^{http://ns\.adobe\.com/f4m/[12]\.0}manifest$', doc.tag):
@@ -2254,6 +2256,35 @@ class GenericIE(InfoExtractor):
             return self.url_result(
                 self._proto_relative_url(unescapeHTML(mobj.group('url'))), 'VODPlatform')
 
+        # Look for Mangomolo embeds
+        mobj = re.search(
+            r'''(?x)<iframe[^>]+src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?admin\.mangomolo\.com/analytics/index\.php/customers/embed/
+                (?:
+                    video\?.*?\bid=(?P<video_id>\d+)|
+                    index\?.*?\bchannelid=(?P<channel_id>(?:[A-Za-z0-9+/=]|%2B|%2F|%3D)+)
+                ).+?)\1''', webpage)
+        if mobj is not None:
+            info = {
+                '_type': 'url_transparent',
+                'url': self._proto_relative_url(unescapeHTML(mobj.group('url'))),
+                'title': video_title,
+                'description': video_description,
+                'thumbnail': video_thumbnail,
+                'uploader': video_uploader,
+            }
+            video_id = mobj.group('video_id')
+            if video_id:
+                info.update({
+                    'ie_key': 'MangomoloVideo',
+                    'id': video_id,
+                })
+            else:
+                info.update({
+                    'ie_key': 'MangomoloLive',
+                    'id': mobj.group('channel_id'),
+                })
+            return info
+
         # Look for Instagram embeds
         instagram_embed_url = InstagramIE._extract_embed_url(webpage)
         if instagram_embed_url is not None:

youtube_dl/extractor/globo.py

@@ -2,6 +2,7 @@
 from __future__ import unicode_literals
 
 import random
+import re
 import math
 
 from .common import InfoExtractor
@@ -14,6 +15,7 @@ from ..utils import (
     ExtractorError,
     float_or_none,
     int_or_none,
+    orderedSet,
     str_or_none,
 )
 
@@ -63,6 +65,9 @@ class GloboIE(InfoExtractor):
     }, {
         'url': 'http://canaloff.globo.com/programas/desejar-profundo/videos/4518560.html',
         'only_matching': True,
+    }, {
+        'url': 'globo:3607726',
+        'only_matching': True,
     }]
 
     class MD5(object):
@@ -396,7 +401,7 @@ class GloboIE(InfoExtractor):
 
 
 class GloboArticleIE(InfoExtractor):
-    _VALID_URL = r'https?://.+?\.globo\.com/(?:[^/]+/)*(?P<id>[^/]+)(?:\.html)?'
+    _VALID_URL = r'https?://.+?\.globo\.com/(?:[^/]+/)*(?P<id>[^/.]+)(?:\.html)?'
 
     _VIDEOID_REGEXES = [
         r'\bdata-video-id=["\'](\d{7,})',
@@ -408,15 +413,20 @@ class GloboArticleIE(InfoExtractor):
 
     _TESTS = [{
         'url': 'http://g1.globo.com/jornal-nacional/noticia/2014/09/novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes.html',
-        'md5': '307fdeae4390ccfe6ba1aa198cf6e72b',
         'info_dict': {
-            'id': '3652183',
-            'ext': 'mp4',
-            'title': 'Receita Federal explica como vai fiscalizar bagagens de quem retorna ao Brasil de avião',
-            'duration': 110.711,
-            'uploader': 'Rede Globo',
-            'uploader_id': '196',
-        }
+            'id': 'novidade-na-fiscalizacao-de-bagagem-pela-receita-provoca-discussoes',
+            'title': 'Novidade na fiscalização de bagagem pela Receita provoca discussões',
+            'description': 'md5:c3c4b4d4c30c32fce460040b1ac46b12',
+        },
+        'playlist_count': 1,
+    }, {
+        'url': 'http://g1.globo.com/pr/parana/noticia/2016/09/mpf-denuncia-lula-marisa-e-mais-seis-na-operacao-lava-jato.html',
+        'info_dict': {
+            'id': 'mpf-denuncia-lula-marisa-e-mais-seis-na-operacao-lava-jato',
+            'title': "Lula era o 'comandante máximo' do esquema da Lava Jato, diz MPF",
+            'description': 'md5:8aa7cc8beda4dc71cc8553e00b77c54c',
+        },
+        'playlist_count': 6,
     }, {
         'url': 'http://gq.globo.com/Prazeres/Poder/noticia/2015/10/all-o-desafio-assista-ao-segundo-capitulo-da-serie.html',
         'only_matching': True,
@@ -435,5 +445,12 @@ class GloboArticleIE(InfoExtractor):
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
-        video_id = self._search_regex(self._VIDEOID_REGEXES, webpage, 'video id')
-        return self.url_result('globo:%s' % video_id, 'Globo')
+        video_ids = []
+        for video_regex in self._VIDEOID_REGEXES:
+            video_ids.extend(re.findall(video_regex, webpage))
+        entries = [
+            self.url_result('globo:%s' % video_id, GloboIE.ie_key())
+            for video_id in orderedSet(video_ids)]
+        title = self._og_search_title(webpage, fatal=False)
+        description = self._html_search_meta('description', webpage)
+        return self.playlist_result(entries, display_id, title, description)

youtube_dl/extractor/jwplatform.py

@@ -9,6 +9,7 @@ from ..utils import (
     determine_ext,
     float_or_none,
     int_or_none,
+    js_to_json,
     mimetype2ext,
 )
 
@@ -19,14 +20,15 @@ class JWPlatformBaseIE(InfoExtractor):
         # TODO: Merge this with JWPlayer-related codes in generic.py
 
         mobj = re.search(
-            'jwplayer\((?P<quote>[\'"])[^\'" ]+(?P=quote)\)\.setup\((?P<options>[^)]+)\)',
+            r'jwplayer\((?P<quote>[\'"])[^\'" ]+(?P=quote)\)\.setup\s*\((?P<options>[^)]+)\)',
             webpage)
         if mobj:
             return mobj.group('options')
 
     def _extract_jwplayer_data(self, webpage, video_id, *args, **kwargs):
         jwplayer_data = self._parse_json(
-            self._find_jwplayer_data(webpage), video_id)
+            self._find_jwplayer_data(webpage), video_id,
+            transform_source=js_to_json)
         return self._parse_jwplayer_data(
             jwplayer_data, video_id, *args, **kwargs)
 

youtube_dl/extractor/mangomolo.py

@@ -0,0 +1,54 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import base64
+
+from .common import InfoExtractor
+from ..compat import compat_urllib_parse_unquote
+from ..utils import (
+    int_or_none,
+)
+
+
+class MangomoloBaseIE(InfoExtractor):
+    def _get_real_id(self, page_id):
+        return page_id
+
+    def _real_extract(self, url):
+        page_id = self._get_real_id(self._match_id(url))
+        webpage = self._download_webpage(url, page_id)
+        hidden_inputs = self._hidden_inputs(webpage)
+        m3u8_entry_protocol = 'm3u8' if self._IS_LIVE else 'm3u8_native'
+
+        format_url = self._html_search_regex(
+            [
+                r'file\s*:\s*"(https?://[^"]+?/playlist.m3u8)',
+                r'<a[^>]+href="(rtsp://[^"]+)"'
+            ], webpage, 'format url')
+        formats = self._extract_wowza_formats(
+            format_url, page_id, m3u8_entry_protocol, ['smil'])
+        self._sort_formats(formats)
+
+        return {
+            'id': page_id,
+            'title': self._live_title(page_id) if self._IS_LIVE else page_id,
+            'uploader_id': hidden_inputs.get('userid'),
+            'duration': int_or_none(hidden_inputs.get('duration')),
+            'is_live': self._IS_LIVE,
+            'formats': formats,
+        }
+
+
+class MangomoloVideoIE(MangomoloBaseIE):
+    IE_NAME = 'mangomolo:video'
+    _VALID_URL = r'https?://admin\.mangomolo\.com/analytics/index\.php/customers/embed/video\?.*?\bid=(?P<id>\d+)'
+    _IS_LIVE = False
+
+
+class MangomoloLiveIE(MangomoloBaseIE):
+    IE_NAME = 'mangomolo:live'
+    _VALID_URL = r'https?://admin\.mangomolo\.com/analytics/index\.php/customers/embed/index\?.*?\bchannelid=(?P<id>(?:[A-Za-z0-9+/=]|%2B|%2F|%3D)+)'
+    _IS_LIVE = True
+
+    def _get_real_id(self, page_id):
+        return base64.b64decode(compat_urllib_parse_unquote(page_id).encode()).decode()

youtube_dl/extractor/radiocanada.py

@@ -13,6 +13,7 @@ from ..utils import (
     xpath_element,
     ExtractorError,
     determine_protocol,
+    unsmuggle_url,
 )
 
 
@@ -35,28 +36,51 @@ class RadioCanadaIE(InfoExtractor):
     }
 
     def _real_extract(self, url):
+        url, smuggled_data = unsmuggle_url(url, {})
         app_code, video_id = re.match(self._VALID_URL, url).groups()
 
-        device_types = ['ipad', 'android']
+        metadata = self._download_xml(
+            'http://api.radio-canada.ca/metaMedia/v1/index.ashx',
+            video_id, note='Downloading metadata XML', query={
+                'appCode': app_code,
+                'idMedia': video_id,
+            })
+
+        def get_meta(name):
+            el = find_xpath_attr(metadata, './/Meta', 'name', name)
+            return el.text if el is not None else None
+
+        if get_meta('protectionType'):
+            raise ExtractorError('This video is DRM protected.', expected=True)
+
+        device_types = ['ipad']
         if app_code != 'toutv':
             device_types.append('flash')
+        if not smuggled_data:
+            device_types.append('android')
 
         formats = []
         # TODO: extract f4m formats
         # f4m formats can be extracted using flashhd device_type but they produce unplayable file
         for device_type in device_types:
-            v_data = self._download_xml(
-                'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx',
-                video_id, note='Downloading %s XML' % device_type, query={
-                    'appCode': app_code,
-                    'idMedia': video_id,
-                    'connectionType': 'broadband',
-                    'multibitrate': 'true',
-                    'deviceType': device_type,
+            validation_url = 'http://api.radio-canada.ca/validationMedia/v1/Validation.ashx'
+            query = {
+                'appCode': app_code,
+                'idMedia': video_id,
+                'connectionType': 'broadband',
+                'multibitrate': 'true',
+                'deviceType': device_type,
+            }
+            if smuggled_data:
+                validation_url = 'https://services.radio-canada.ca/media/validation/v2/'
+                query.update(smuggled_data)
+            else:
+                query.update({
                     # paysJ391wsHjbOJwvCs26toz and bypasslock are used to bypass geo-restriction
                     'paysJ391wsHjbOJwvCs26toz': 'CA',
                     'bypasslock': 'NZt5K62gRqfc',
-                }, fatal=False)
+                })
+            v_data = self._download_xml(validation_url, video_id, note='Downloading %s XML' % device_type, query=query, fatal=False)
             v_url = xpath_text(v_data, 'url')
             if not v_url:
                 continue
@@ -101,17 +125,6 @@ class RadioCanadaIE(InfoExtractor):
                                 f4m_id='hds', fatal=False))
         self._sort_formats(formats)
 
-        metadata = self._download_xml(
-            'http://api.radio-canada.ca/metaMedia/v1/index.ashx',
-            video_id, note='Downloading metadata XML', query={
-                'appCode': app_code,
-                'idMedia': video_id,
-            })
-
-        def get_meta(name):
-            el = find_xpath_attr(metadata, './/Meta', 'name', name)
-            return el.text if el is not None else None
-
         return {
             'id': video_id,
             'title': get_meta('Title'),

youtube_dl/extractor/svt.py

@@ -16,7 +16,7 @@ class SVTBaseIE(InfoExtractor):
     def _extract_video(self, video_info, video_id):
         formats = []
         for vr in video_info['videoReferences']:
-            player_type = vr.get('playerType')
+            player_type = vr.get('playerType') or vr.get('format')
             vurl = vr['url']
             ext = determine_ext(vurl)
             if ext == 'm3u8':

youtube_dl/extractor/thisav.py

@@ -3,13 +3,12 @@ from __future__ import unicode_literals
 
 import re
 
-from .common import InfoExtractor
-from ..utils import determine_ext
+from .jwplatform import JWPlatformBaseIE
 
 
-class ThisAVIE(InfoExtractor):
+class ThisAVIE(JWPlatformBaseIE):
     _VALID_URL = r'https?://(?:www\.)?thisav\.com/video/(?P<id>[0-9]+)/.*'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.thisav.com/video/47734/%98%26sup1%3B%83%9E%83%82---just-fit.html',
         'md5': '0480f1ef3932d901f0e0e719f188f19b',
         'info_dict': {
@@ -19,7 +18,17 @@ class ThisAVIE(InfoExtractor):
             'uploader': 'dj7970',
             'uploader_id': 'dj7970'
         }
-    }
+    }, {
+        'url': 'http://www.thisav.com/video/242352/nerdy-18yo-big-ass-tattoos-and-glasses.html',
+        'md5': 'ba90c076bd0f80203679e5b60bf523ee',
+        'info_dict': {
+            'id': '242352',
+            'ext': 'mp4',
+            'title': 'Nerdy 18yo Big Ass Tattoos and Glasses',
+            'uploader': 'cybersluts',
+            'uploader_id': 'cybersluts',
+        },
+    }]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
@@ -28,20 +37,28 @@ class ThisAVIE(InfoExtractor):
         webpage = self._download_webpage(url, video_id)
         title = self._html_search_regex(r'<h1>([^<]*)</h1>', webpage, 'title')
         video_url = self._html_search_regex(
-            r"addVariable\('file','([^']+)'\);", webpage, 'video url')
+            r"addVariable\('file','([^']+)'\);", webpage, 'video url', default=None)
+        if video_url:
+            info_dict = {
+                'formats': [{
+                    'url': video_url,
+                }],
+            }
+        else:
+            info_dict = self._extract_jwplayer_data(
+                webpage, video_id, require_title=False)
         uploader = self._html_search_regex(
             r': <a href="http://www.thisav.com/user/[0-9]+/(?:[^"]+)">([^<]+)</a>',
             webpage, 'uploader name', fatal=False)
         uploader_id = self._html_search_regex(
             r': <a href="http://www.thisav.com/user/[0-9]+/([^"]+)">(?:[^<]+)</a>',
             webpage, 'uploader id', fatal=False)
-        ext = determine_ext(video_url)
 
-        return {
+        info_dict.update({
             'id': video_id,
-            'url': video_url,
             'uploader': uploader,
             'uploader_id': uploader_id,
             'title': title,
-            'ext': ext,
-        }
+        })
+
+        return info_dict

youtube_dl/extractor/toutv.py

@@ -2,12 +2,22 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
-from ..utils import int_or_none
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    ExtractorError,
+    urlencode_postdata,
+    extract_attributes,
+    smuggle_url,
+)
 
 
 class TouTvIE(InfoExtractor):
+    _NETRC_MACHINE = 'toutv'
     IE_NAME = 'tou.tv'
     _VALID_URL = r'https?://ici\.tou\.tv/(?P<id>[a-zA-Z0-9_-]+/S[0-9]+E[0-9]+)'
+    _access_token = None
+    _claims = None
 
     _TEST = {
         'url': 'http://ici.tou.tv/garfield-tout-court/S2015E17',
@@ -22,18 +32,64 @@ class TouTvIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         },
+        'skip': '404 Not Found',
     }
 
+    def _real_initialize(self):
+        email, password = self._get_login_info()
+        if email is None:
+            return
+        state = 'http://ici.tou.tv//'
+        webpage = self._download_webpage(state, None, 'Downloading homepage')
+        toutvlogin = self._parse_json(self._search_regex(
+            r'(?s)toutvlogin\s*=\s*({.+?});', webpage, 'toutvlogin'), None, js_to_json)
+        authorize_url = toutvlogin['host'] + '/auth/oauth/v2/authorize'
+        login_webpage = self._download_webpage(
+            authorize_url, None, 'Downloading login page', query={
+                'client_id': toutvlogin['clientId'],
+                'redirect_uri': 'https://ici.tou.tv/login/loginCallback',
+                'response_type': 'token',
+                'scope': 'media-drmt openid profile email id.write media-validation.read.privileged',
+                'state': state,
+            })
+        login_form = self._search_regex(
+            r'(?s)(<form[^>]+id="Form-login".+?</form>)', login_webpage, 'login form')
+        form_data = self._hidden_inputs(login_form)
+        form_data.update({
+            'login-email': email,
+            'login-password': password,
+        })
+        post_url = extract_attributes(login_form).get('action') or authorize_url
+        _, urlh = self._download_webpage_handle(
+            post_url, None, 'Logging in', data=urlencode_postdata(form_data))
+        self._access_token = self._search_regex(
+            r'access_token=([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})',
+            urlh.geturl(), 'access token')
+        self._claims = self._download_json(
+            'https://services.radio-canada.ca/media/validation/v2/getClaims',
+            None, 'Extracting Claims', query={
+                'token': self._access_token,
+                'access_token': self._access_token,
+            })['claims']
+
     def _real_extract(self, url):
         path = self._match_id(url)
         metadata = self._download_json('http://ici.tou.tv/presentation/%s' % path, path)
+        if metadata.get('IsDrm'):
+            raise ExtractorError('This video is DRM protected.', expected=True)
         video_id = metadata['IdMedia']
         details = metadata['Details']
         title = details['OriginalTitle']
+        video_url = 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id)
+        if self._access_token and self._claims:
+            video_url = smuggle_url(video_url, {
+                'access_token': self._access_token,
+                'claims': self._claims,
+            })
 
         return {
             '_type': 'url_transparent',
-            'url': 'radiocanada:%s:%s' % (metadata.get('AppCode', 'toutv'), video_id),
+            'url': video_url,
             'id': video_id,
             'title': title,
             'thumbnail': details.get('ImageUrl'),

youtube_dl/extractor/tv4.py

@@ -2,9 +2,13 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
+    int_or_none,
     parse_iso8601,
+    try_get,
+    update_url_query,
 )
 
 
@@ -65,36 +69,47 @@ class TV4IE(InfoExtractor):
         video_id = self._match_id(url)
 
         info = self._download_json(
-            'http://www.tv4play.se/player/assets/%s.json' % video_id, video_id, 'Downloading video info JSON')
+            'http://www.tv4play.se/player/assets/%s.json' % video_id,
+            video_id, 'Downloading video info JSON')
 
         # If is_geo_restricted is true, it doesn't necessarily mean we can't download it
-        if info['is_geo_restricted']:
+        if info.get('is_geo_restricted'):
             self.report_warning('This content might not be available in your country due to licensing restrictions.')
-        if info['requires_subscription']:
+        if info.get('requires_subscription'):
             raise ExtractorError('This content requires subscription.', expected=True)
 
-        sources_data = self._download_json(
-            'https://prima.tv4play.se/api/web/asset/%s/play.json?protocol=http&videoFormat=MP4' % video_id, video_id, 'Downloading sources JSON')
-        sources = sources_data['playback']
+        title = info['title']
 
         formats = []
-        for item in sources.get('items', {}).get('item', []):
-            ext, bitrate = item['mediaFormat'], item['bitrate']
-            formats.append({
-                'format_id': '%s_%s' % (ext, bitrate),
-                'tbr': bitrate,
-                'ext': ext,
-                'url': item['url'],
-            })
+        # http formats are linked with unresolvable host
+        for kind in ('hls', ''):
+            data = self._download_json(
+                'https://prima.tv4play.se/api/web/asset/%s/play.json' % video_id,
+                video_id, 'Downloading sources JSON', query={
+                    'protocol': kind,
+                    'videoFormat': 'MP4+WEBVTTS+WEBVTT',
+                })
+            item = try_get(data, lambda x: x['playback']['items']['item'], dict)
+            manifest_url = item.get('url')
+            if not isinstance(manifest_url, compat_str):
+                continue
+            if kind == 'hls':
+                formats.extend(self._extract_m3u8_formats(
+                    manifest_url, video_id, 'mp4', entry_protocol='m3u8_native',
+                    m3u8_id=kind, fatal=False))
+            else:
+                formats.extend(self._extract_f4m_formats(
+                    update_url_query(manifest_url, {'hdcore': '3.8.0'}),
+                    video_id, f4m_id='hds', fatal=False))
         self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'title': info['title'],
+            'title': title,
             'formats': formats,
             'description': info.get('description'),
             'timestamp': parse_iso8601(info.get('broadcast_date_time')),
-            'duration': info.get('duration'),
+            'duration': int_or_none(info.get('duration')),
             'thumbnail': info.get('image'),
-            'is_live': sources.get('live'),
+            'is_live': info.get('is_live') is True,
         }

youtube_dl/extractor/vier.py

@@ -48,8 +48,8 @@ class VierIE(InfoExtractor):
             [r'data-filename="([^"]+)"', r'"filename"\s*:\s*"([^"]+)"'],
             webpage, 'filename')
 
-        playlist_url = 'http://vod.streamcloud.be/%s/mp4:_definst_/%s.mp4/playlist.m3u8' % (application, filename)
-        formats = self._extract_m3u8_formats(playlist_url, display_id, 'mp4')
+        playlist_url = 'http://vod.streamcloud.be/%s/_definst_/mp4:%s.mp4/playlist.m3u8' % (application, filename)
+        formats = self._extract_wowza_formats(playlist_url, display_id)
         self._sort_formats(formats)
 
         title = self._og_search_title(webpage, default=display_id)

youtube_dl/extractor/vodplatform.py

@@ -25,29 +25,8 @@ class VODPlatformIE(InfoExtractor):
         title = unescapeHTML(self._og_search_title(webpage))
         hidden_inputs = self._hidden_inputs(webpage)
 
-        base_url = self._search_regex(
-            '(.*/)(?:playlist.m3u8|manifest.mpd)',
-            hidden_inputs.get('HiddenmyhHlsLink') or hidden_inputs['HiddenmyDashLink'],
-            'base url')
-        formats = self._extract_m3u8_formats(
-            base_url + 'playlist.m3u8', video_id, 'mp4',
-            'm3u8_native', m3u8_id='hls', fatal=False)
-        formats.extend(self._extract_mpd_formats(
-            base_url + 'manifest.mpd', video_id,
-            mpd_id='dash', fatal=False))
-        rtmp_formats = self._extract_smil_formats(
-            base_url + 'jwplayer.smil', video_id, fatal=False)
-        for rtmp_format in rtmp_formats:
-            rtsp_format = rtmp_format.copy()
-            rtsp_format['url'] = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path'])
-            del rtsp_format['play_path']
-            del rtsp_format['ext']
-            rtsp_format.update({
-                'url': rtsp_format['url'].replace('rtmp://', 'rtsp://'),
-                'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'),
-                'protocol': 'rtsp',
-            })
-            formats.extend([rtmp_format, rtsp_format])
+        formats = self._extract_wowza_formats(
+            hidden_inputs.get('HiddenmyhHlsLink') or hidden_inputs['HiddenmyDashLink'], video_id, skip_protocols=['f4m', 'smil'])
         self._sort_formats(formats)
 
         return {

youtube_dl/extractor/vrt.py

@@ -5,7 +5,6 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    determine_ext,
     float_or_none,
 )
 
@@ -75,7 +74,6 @@ class VRTIE(InfoExtractor):
         },
         {
             'url': 'http://cobra.canvas.be/cm/cobra/videozone/rubriek/film-videozone/1.2377055',
-            'md5': '',
             'info_dict': {
                 'id': '2377055',
                 'ext': 'mp4',
@@ -119,39 +117,17 @@ class VRTIE(InfoExtractor):
                 video_id, 'mp4', m3u8_id='hls', fatal=False))
 
         if src:
-            if determine_ext(src) == 'm3u8':
-                formats.extend(self._extract_m3u8_formats(
-                    src, video_id, 'mp4', entry_protocol='m3u8_native',
-                    m3u8_id='hls', fatal=False))
-                formats.extend(self._extract_f4m_formats(
-                    src.replace('playlist.m3u8', 'manifest.f4m'),
-                    video_id, f4m_id='hds', fatal=False))
-                if 'data-video-geoblocking="true"' not in webpage:
-                    rtmp_formats = self._extract_smil_formats(
-                        src.replace('playlist.m3u8', 'jwplayer.smil'),
-                        video_id, fatal=False)
-                    formats.extend(rtmp_formats)
-                    for rtmp_format in rtmp_formats:
-                        rtmp_format_c = rtmp_format.copy()
-                        rtmp_format_c['url'] = '%s/%s' % (rtmp_format['url'], rtmp_format['play_path'])
-                        del rtmp_format_c['play_path']
-                        del rtmp_format_c['ext']
-                        http_format = rtmp_format_c.copy()
+            formats = self._extract_wowza_formats(src, video_id)
+            if 'data-video-geoblocking="true"' not in webpage:
+                for f in formats:
+                    if f['url'].startswith('rtsp://'):
+                        http_format = f.copy()
                         http_format.update({
-                            'url': rtmp_format_c['url'].replace('rtmp://', 'http://').replace('vod.', 'download.').replace('/_definst_/', '/').replace('mp4:', ''),
-                            'format_id': rtmp_format['format_id'].replace('rtmp', 'http'),
+                            'url': f['url'].replace('rtsp://', 'http://').replace('vod.', 'download.').replace('/_definst_/', '/').replace('mp4:', ''),
+                            'format_id': f['format_id'].replace('rtsp', 'http'),
                             'protocol': 'http',
                         })
-                        rtsp_format = rtmp_format_c.copy()
-                        rtsp_format.update({
-                            'url': rtsp_format['url'].replace('rtmp://', 'rtsp://'),
-                            'format_id': rtmp_format['format_id'].replace('rtmp', 'rtsp'),
-                            'protocol': 'rtsp',
-                        })
-                        formats.extend([http_format, rtsp_format])
-            else:
-                formats.extend(self._extract_f4m_formats(
-                    '%s/manifest.f4m' % src, video_id, f4m_id='hds', fatal=False))
+                        formats.append(http_format)
 
         if not formats and 'data-video-geoblocking="true"' in webpage:
             self.raise_geo_restricted('This video is only available in Belgium')

youtube_dl/extractor/vyborymos.py

@@ -0,0 +1,55 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+
+
+class VyboryMosIE(InfoExtractor):
+    _VALID_URL = r'https?://vybory\.mos\.ru/(?:#precinct/|account/channels\?.*?\bstation_id=)(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'http://vybory.mos.ru/#precinct/13636',
+        'info_dict': {
+            'id': '13636',
+            'ext': 'mp4',
+            'title': 're:^Участковая избирательная комиссия №2231 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'description': 'Россия, Москва, улица Введенского, 32А',
+            'is_live': True,
+        },
+        'params': {
+            'skip_download': True,
+        }
+    }, {
+        'url': 'http://vybory.mos.ru/account/channels?station_id=13636',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        station_id = self._match_id(url)
+
+        channels = self._download_json(
+            'http://vybory.mos.ru/account/channels?station_id=%s' % station_id,
+            station_id)
+
+        formats = []
+        for cam_num, (sid, hosts, name, _) in enumerate(channels, 1):
+            for num, host in enumerate(hosts, 1):
+                formats.append({
+                    'url': 'http://%s/master.m3u8?sid=%s' % (host, sid),
+                    'ext': 'mp4',
+                    'format_id': 'camera%d-host%d' % (cam_num, num),
+                    'format_note': '%s, %s' % (name, host),
+                })
+
+        info = self._download_json(
+            'http://vybory.mos.ru/json/voting_stations/136/%s.json' % station_id,
+            station_id, 'Downloading station info')
+
+        title = info['name']
+
+        return {
+            'id': station_id,
+            'title': self._live_title(title),
+            'description': info.get('address'),
+            'is_live': True,
+            'formats': formats,
+        }

youtube_dl/extractor/xfileshare.py

@@ -124,12 +124,14 @@ class XFileShareIE(InfoExtractor):
             webpage = self._download_webpage(req, video_id, 'Downloading video page')
 
         title = (self._search_regex(
-            [r'style="z-index: [0-9]+;">([^<]+)</span>',
+            (r'style="z-index: [0-9]+;">([^<]+)</span>',
              r'<td nowrap>([^<]+)</td>',
              r'h4-fine[^>]*>([^<]+)<',
              r'>Watch (.+) ',
-             r'<h2 class="video-page-head">([^<]+)</h2>'],
-            webpage, 'title', default=None) or self._og_search_title(webpage)).strip()
+             r'<h2 class="video-page-head">([^<]+)</h2>',
+             r'<h2 style="[^"]*color:#403f3d[^"]*"[^>]*>([^<]+)<'),  # streamin.to
+            webpage, 'title', default=None) or self._og_search_title(
+            webpage, default=None) or video_id).strip()
 
         def extract_video_url(default=NO_DEFAULT):
             return self._search_regex(

youtube_dl/postprocessor/ffmpeg.py

@@ -139,6 +139,30 @@ class FFmpegPostProcessor(PostProcessor):
     def probe_executable(self):
         return self._paths[self.probe_basename]
 
+    def get_audio_codec(self, path):
+        if not self.probe_available:
+            raise PostProcessingError('ffprobe or avprobe not found. Please install one.')
+        try:
+            cmd = [
+                encodeFilename(self.probe_executable, True),
+                encodeArgument('-show_streams'),
+                encodeFilename(self._ffmpeg_filename_argument(path), True)]
+            if self._downloader.params.get('verbose', False):
+                self._downloader.to_screen('[debug] %s command line: %s' % (self.basename, shell_quote(cmd)))
+            handle = subprocess.Popen(cmd, stderr=compat_subprocess_get_DEVNULL(), stdout=subprocess.PIPE, stdin=subprocess.PIPE)
+            output = handle.communicate()[0]
+            if handle.wait() != 0:
+                return None
+        except (IOError, OSError):
+            return None
+        audio_codec = None
+        for line in output.decode('ascii', 'ignore').split('\n'):
+            if line.startswith('codec_name='):
+                audio_codec = line.split('=')[1].strip()
+            elif line.strip() == 'codec_type=audio' and audio_codec is not None:
+                return audio_codec
+        return None
+
     def run_ffmpeg_multiple_files(self, input_paths, out_path, opts):
         self.check_version()
 
@@ -188,31 +212,6 @@ class FFmpegExtractAudioPP(FFmpegPostProcessor):
         self._preferredquality = preferredquality
         self._nopostoverwrites = nopostoverwrites
 
-    def get_audio_codec(self, path):
-
-        if not self.probe_available:
-            raise PostProcessingError('ffprobe or avprobe not found. Please install one.')
-        try:
-            cmd = [
-                encodeFilename(self.probe_executable, True),
-                encodeArgument('-show_streams'),
-                encodeFilename(self._ffmpeg_filename_argument(path), True)]
-            if self._downloader.params.get('verbose', False):
-                self._downloader.to_screen('[debug] %s command line: %s' % (self.basename, shell_quote(cmd)))
-            handle = subprocess.Popen(cmd, stderr=compat_subprocess_get_DEVNULL(), stdout=subprocess.PIPE, stdin=subprocess.PIPE)
-            output = handle.communicate()[0]
-            if handle.wait() != 0:
-                return None
-        except (IOError, OSError):
-            return None
-        audio_codec = None
-        for line in output.decode('ascii', 'ignore').split('\n'):
-            if line.startswith('codec_name='):
-                audio_codec = line.split('=')[1].strip()
-            elif line.strip() == 'codec_type=audio' and audio_codec is not None:
-                return audio_codec
-        return None
-
     def run_ffmpeg(self, path, out_path, codec, more_opts):
         if codec is None:
             acodec_opts = []
@@ -504,15 +503,15 @@ class FFmpegFixupM4aPP(FFmpegPostProcessor):
 class FFmpegFixupM3u8PP(FFmpegPostProcessor):
     def run(self, info):
         filename = info['filepath']
-        temp_filename = prepend_extension(filename, 'temp')
+        if self.get_audio_codec(filename) == 'aac':
+            temp_filename = prepend_extension(filename, 'temp')
 
-        options = ['-c', 'copy', '-f', 'mp4', '-bsf:a', 'aac_adtstoasc']
-        self._downloader.to_screen('[ffmpeg] Fixing malformated aac bitstream in "%s"' % filename)
-        self.run_ffmpeg(filename, temp_filename, options)
-
-        os.remove(encodeFilename(filename))
-        os.rename(encodeFilename(temp_filename), encodeFilename(filename))
+            options = ['-c', 'copy', '-f', 'mp4', '-bsf:a', 'aac_adtstoasc']
+            self._downloader.to_screen('[ffmpeg] Fixing malformated aac bitstream in "%s"' % filename)
+            self.run_ffmpeg(filename, temp_filename, options)
 
+            os.remove(encodeFilename(filename))
+            os.rename(encodeFilename(temp_filename), encodeFilename(filename))
         return [], info
 
 

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.09.15'
+__version__ = '2016.09.18'