release 2016.02.10

youtube_dl/YoutubeDL.py

@@ -1288,6 +1288,9 @@ class YoutubeDL(object):
 
             if format.get('format_id') is None:
                 format['format_id'] = compat_str(i)
+            else:
+                # Sanitize format_id from characters used in format selector expression
+                format['format_id'] = re.sub('[\s,/+\[\]()]', '_', format['format_id'])
             format_id = format['format_id']
             if format_id not in formats_dict:
                 formats_dict[format_id] = []

youtube_dl/extractor/common.py

@@ -1186,6 +1186,7 @@ class InfoExtractor(object):
         http_count = 0
         m3u8_count = 0
 
+        src_urls = []
         videos = smil.findall(self._xpath_ns('.//video', namespace))
         for video in videos:
             src = video.get('src')
@@ -1222,6 +1223,9 @@ class InfoExtractor(object):
                 continue
 
             src_url = src if src.startswith('http') else compat_urlparse.urljoin(base, src)
+            if src_url in src_urls:
+                continue
+            src_urls.append(src_url)
 
             if proto == 'm3u8' or src_ext == 'm3u8':
                 m3u8_formats = self._extract_m3u8_formats(
@@ -1267,11 +1271,13 @@ class InfoExtractor(object):
         return formats
 
     def _parse_smil_subtitles(self, smil, namespace=None, subtitles_lang='en'):
+        urls = []
         subtitles = {}
         for num, textstream in enumerate(smil.findall(self._xpath_ns('.//textstream', namespace))):
             src = textstream.get('src')
-            if not src:
+            if not src or src in urls:
                 continue
+            urls.append(src)
             ext = textstream.get('ext') or determine_ext(src)
             if not ext:
                 type_ = textstream.get('type')
@@ -1434,6 +1440,8 @@ class InfoExtractor(object):
                             base_url = mpd_base_url + base_url
                         representation_id = representation_attrib.get('id')
                         lang = representation_attrib.get('lang')
+                        url_el = representation.find(_add_ns('BaseURL'))
+                        filesize = int_or_none(url_el.attrib.get('{http://youtube.com/yt/2012/10/10}contentLength') if url_el is not None else None)
                         f = {
                             'format_id': mpd_id or representation_id,
                             'url': base_url,
@@ -1446,6 +1454,7 @@ class InfoExtractor(object):
                             'acodec': 'none' if content_type == 'video' else representation_attrib.get('codecs'),
                             'language': lang if lang not in ('mul', 'und', 'zxx', 'mis') else None,
                             'format_note': 'DASH %s' % content_type,
+                            'filesize': filesize,
                         }
                         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:

youtube_dl/extractor/fox.py

@@ -9,6 +9,7 @@ class FOXIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?fox\.com/watch/(?P<id>[0-9]+)'
     _TEST = {
         'url': 'http://www.fox.com/watch/255180355939/7684182528',
+        'md5': 'ebd296fcc41dd4b19f8115d8461a3165',
         'info_dict': {
             'id': '255180355939',
             'ext': 'mp4',
@@ -17,10 +18,6 @@ class FOXIE(InfoExtractor):
             'duration': 129,
         },
         'add_ie': ['ThePlatform'],
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
     }
 
     def _real_extract(self, url):
@@ -29,7 +26,7 @@ class FOXIE(InfoExtractor):
 
         release_url = self._parse_json(self._search_regex(
             r'"fox_pdk_player"\s*:\s*({[^}]+?})', webpage, 'fox_pdk_player'),
-            video_id)['release_url'] + '&manifest=m3u'
+            video_id)['release_url'] + '&switch=http'
 
         return {
             '_type': 'url_transparent',

youtube_dl/extractor/hotstar.py

@@ -10,8 +10,8 @@ from ..utils import (
 
 
 class HotStarIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?hotstar\.com/.*?[/-](?P<id>\d{10})'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?hotstar\.com/(?:.+?[/-])?(?P<id>\d{10})'
+    _TESTS = [{
         'url': 'http://www.hotstar.com/on-air-with-aib--english-1000076273',
         'info_dict': {
             'id': '1000076273',
@@ -26,7 +26,13 @@ class HotStarIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
-    }
+    }, {
+        'url': 'http://www.hotstar.com/sports/cricket/rajitha-sizzles-on-debut-with-329/2001477583',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.hotstar.com/1000000515',
+        'only_matching': True,
+    }]
 
     _GET_CONTENT_TEMPLATE = 'http://account.hotstar.com/AVS/besc?action=GetAggregatedContentDetails&channel=PCTV&contentId=%s'
     _GET_CDN_TEMPLATE = 'http://getcdn.hotstar.com/AVS/besc?action=GetCDN&asJson=Y&channel=%s&id=%s&type=%s'

youtube_dl/extractor/mailru.py

@@ -4,6 +4,10 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    remove_end,
+)
 
 
 class MailRuIE(InfoExtractor):
@@ -34,14 +38,30 @@ class MailRuIE(InfoExtractor):
                 'id': '46843144_1263',
                 'ext': 'mp4',
                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
-                'timestamp': 1397217632,
-                'upload_date': '20140411',
-                'uploader': 'hitech',
+                'timestamp': 1397039888,
+                'upload_date': '20140409',
+                'uploader': 'hitech@corp.mail.ru',
                 'uploader_id': 'hitech@corp.mail.ru',
                 'duration': 245,
             },
             'skip': 'Not accessible from Travis CI server',
         },
+        {
+            # only available via metaUrl API
+            'url': 'http://my.mail.ru/mail/720pizle/video/_myvideo/502.html',
+            'md5': '3b26d2491c6949d031a32b96bd97c096',
+            'info_dict': {
+                'id': '56664382_502',
+                'ext': 'mp4',
+                'title': ':8336',
+                'timestamp': 1449094163,
+                'upload_date': '20151202',
+                'uploader': '720pizle@mail.ru',
+                'uploader_id': '720pizle@mail.ru',
+                'duration': 6001,
+            },
+            'skip': 'Not accessible from Travis CI server',
+        }
     ]
 
     def _real_extract(self, url):
@@ -51,32 +71,55 @@ class MailRuIE(InfoExtractor):
         if not video_id:
             video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
 
-        video_data = self._download_json(
-            'http://api.video.mail.ru/videos/%s.json?new=1' % video_id, video_id, 'Downloading video JSON')
+        webpage = self._download_webpage(url, video_id)
 
-        author = video_data['author']
-        uploader = author['name']
-        uploader_id = author.get('id') or author.get('email')
-        view_count = video_data.get('views_count')
+        video_data = None
+
+        page_config = self._parse_json(self._search_regex(
+            r'(?s)<script[^>]+class="sp-video__page-config"[^>]*>(.+?)</script>',
+            webpage, 'page config', default='{}'), video_id, fatal=False)
+        if page_config:
+            meta_url = page_config.get('metaUrl') or page_config.get('video', {}).get('metaUrl')
+            if meta_url:
+                video_data = self._download_json(
+                    meta_url, video_id, 'Downloading video meta JSON', fatal=False)
+
+        # Fallback old approach
+        if not video_data:
+            video_data = self._download_json(
+                'http://api.video.mail.ru/videos/%s.json?new=1' % video_id,
+                video_id, 'Downloading video JSON')
+
+        formats = []
+        for f in video_data['videos']:
+            video_url = f.get('url')
+            if not video_url:
+                continue
+            format_id = f.get('key')
+            height = int_or_none(self._search_regex(
+                r'^(\d+)[pP]$', format_id, 'height', default=None)) if format_id else None
+            formats.append({
+                'url': video_url,
+                'format_id': format_id,
+                'height': height,
+            })
+        self._sort_formats(formats)
 
         meta_data = video_data['meta']
-        content_id = '%s_%s' % (
-            meta_data.get('accId', ''), meta_data['itemId'])
-        title = meta_data['title']
-        if title.endswith('.mp4'):
-            title = title[:-4]
-        thumbnail = meta_data['poster']
-        duration = meta_data['duration']
-        timestamp = meta_data['timestamp']
+        title = remove_end(meta_data['title'], '.mp4')
 
-        formats = [
-            {
-                'url': video['url'],
-                'format_id': video['key'],
-                'height': int(video['key'].rstrip('p'))
-            } for video in video_data['videos']
-        ]
-        self._sort_formats(formats)
+        author = video_data.get('author')
+        uploader = author.get('name')
+        uploader_id = author.get('id') or author.get('email')
+        view_count = int_or_none(video_data.get('viewsCount') or video_data.get('views_count'))
+
+        acc_id = meta_data.get('accId')
+        item_id = meta_data.get('itemId')
+        content_id = '%s_%s' % (acc_id, item_id) if acc_id and item_id else video_id
+
+        thumbnail = meta_data.get('poster')
+        duration = int_or_none(meta_data.get('duration'))
+        timestamp = int_or_none(meta_data.get('timestamp'))
 
         return {
             'id': content_id,

youtube_dl/extractor/pbs.py

@@ -8,6 +8,7 @@ from ..utils import (
     ExtractorError,
     determine_ext,
     int_or_none,
+    js_to_json,
     strip_jsonp,
     unified_strdate,
     US_RATINGS,
@@ -432,9 +433,20 @@ class PBSIE(InfoExtractor):
                 for vid_id in video_id]
             return self.playlist_result(entries, display_id)
 
+        player = self._download_webpage(
+            'http://player.pbs.org/portalplayer/%s' % video_id, display_id)
+
+        info = self._parse_json(
+            self._search_regex(
+                r'(?s)PBS\.videoData\s*=\s*({.+?});\n',
+                player, 'video data', default='{}'),
+            display_id, transform_source=js_to_json, fatal=False)
+
+        # Fallback to old videoInfo API
+        if not info:
             info = self._download_json(
                 'http://player.pbs.org/videoInfo/%s?format=json&type=partner' % video_id,
-            display_id)
+                display_id, 'Downloading video info JSON')
 
         formats = []
         for encoding_name in ('recommended_encoding', 'alternate_encoding'):

youtube_dl/extractor/viddler.py

@@ -1,6 +1,10 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import (
+    compat_urllib_parse,
+    compat_urlparse,
+)
 from ..utils import (
     float_or_none,
     int_or_none,
@@ -12,10 +16,10 @@ class ViddlerIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?viddler\.com/(?:v|embed|player)/(?P<id>[a-z0-9]+)'
     _TESTS = [{
         'url': 'http://www.viddler.com/v/43903784',
-        'md5': 'ae43ad7cb59431ce043f0ff7fa13cbf4',
+        'md5': '9eee21161d2c7f5b39690c3e325fab2f',
         'info_dict': {
             'id': '43903784',
-            'ext': 'mp4',
+            'ext': 'mov',
             'title': 'Video Made Easy',
             'description': 'md5:6a697ebd844ff3093bd2e82c37b409cd',
             'uploader': 'viddler',
@@ -29,10 +33,10 @@ class ViddlerIE(InfoExtractor):
         }
     }, {
         'url': 'http://www.viddler.com/v/4d03aad9/',
-        'md5': 'faa71fbf70c0bee7ab93076fd007f4b0',
+        'md5': 'f12c5a7fa839c47a79363bfdf69404fb',
         'info_dict': {
             'id': '4d03aad9',
-            'ext': 'mp4',
+            'ext': 'ts',
             'title': 'WALL-TO-GORTAT',
             'upload_date': '20150126',
             'uploader': 'deadspin',
@@ -42,10 +46,10 @@ class ViddlerIE(InfoExtractor):
         }
     }, {
         'url': 'http://www.viddler.com/player/221ebbbd/0/',
-        'md5': '0defa2bd0ea613d14a6e9bd1db6be326',
+        'md5': '740511f61d3d1bb71dc14a0fe01a1c10',
         'info_dict': {
             'id': '221ebbbd',
-            'ext': 'mp4',
+            'ext': 'mov',
             'title': 'LETeens-Grammar-snack-third-conditional',
             'description': ' ',
             'upload_date': '20140929',
@@ -54,16 +58,42 @@ class ViddlerIE(InfoExtractor):
             'view_count': int,
             'comment_count': int,
         }
+    }, {
+        # secret protected
+        'url': 'http://www.viddler.com/v/890c0985?secret=34051570',
+        'info_dict': {
+            'id': '890c0985',
+            'ext': 'mp4',
+            'title': 'Complete Property Training - Traineeships',
+            'description': ' ',
+            'upload_date': '20130606',
+            'uploader': 'TiffanyBowtell',
+            'timestamp': 1370496993,
+            'view_count': int,
+            'comment_count': int,
+        },
+        'params': {
+            'skip_download': True,
+        },
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        json_url = (
-            'http://api.viddler.com/api/v2/viddler.videos.getPlaybackDetails.json?video_id=%s&key=v0vhrt7bg2xq1vyxhkct' %
-            video_id)
+        query = {
+            'video_id': video_id,
+            'key': 'v0vhrt7bg2xq1vyxhkct',
+        }
+
+        qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query)
+        secret = qs.get('secret', [None])[0]
+        if secret:
+            query['secret'] = secret
+
         headers = {'Referer': 'http://static.cdn-ec.viddler.com/js/arpeggio/v2/embed.html'}
-        request = sanitized_Request(json_url, None, headers)
+        request = sanitized_Request(
+            'http://api.viddler.com/api/v2/viddler.videos.getPlaybackDetails.json?%s'
+            % compat_urllib_parse.urlencode(query), None, headers)
         data = self._download_json(request, video_id)['video']
 
         formats = []

youtube_dl/extractor/youku.py

@@ -229,6 +229,9 @@ class YoukuIE(InfoExtractor):
             if error_note is not None and '因版权原因无法观看此视频' in error_note:
                 raise ExtractorError(
                     'Youku said: Sorry, this video is available in China only', expected=True)
+            elif error_note and '该视频被设为私密' in error_note:
+                raise ExtractorError(
+                    'Youku said: Sorry, this video is private', expected=True)
             else:
                 msg = 'Youku server reported error %i' % error.get('code')
                 if error_note is not None:

youtube_dl/version.py

@@ -1,3 +1,3 @@
 from __future__ import unicode_literals
 
-__version__ = '2016.02.09'
+__version__ = '2016.02.10'