debian/control: Update the repository to github.com.

[youtubedl] / youtube_dl / extractor / odnoklassniki.py

diff --git a/youtube_dl/extractor/odnoklassniki.py b/youtube_dl/extractor/odnoklassniki.py
index 215ffe87b55db126300f0c18c98d9c5bfd920ed7..854b6800c18bd47d2ed1b40cace0969a22eebc75 100644 (file)
--- a/youtube_dl/extractor/odnoklassniki.py
+++ b/youtube_dl/extractor/odnoklassniki.py
@@ -2,8 +2,14 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from __future__ import unicode_literals
 
 from .common import InfoExtractor

-from ..compat import compat_urllib_parse_unquote
+from ..compat import (
+    compat_etree_fromstring,
+    compat_parse_qs,
+    compat_urllib_parse_unquote,
+    compat_urllib_parse_urlparse,
+)

 from ..utils import (
 from ..utils import (

+    ExtractorError,

     unified_strdate,
     int_or_none,
     qualities,
     unified_strdate,
     int_or_none,
     qualities,


@@ -12,47 +18,107 @@ from ..utils import (
 
 
 class OdnoklassnikiIE(InfoExtractor):
 
 
 class OdnoklassnikiIE(InfoExtractor):

-    _VALID_URL = r'https?://(?:odnoklassniki|ok)\.ru/(?:video|web-api/video/moviePlayer)/(?P<id>[\d-]+)'
+    _VALID_URL = r'https?://(?:(?:www|m|mobile)\.)?(?:odnoklassniki|ok)\.ru/(?:video(?:embed)?|web-api/video/moviePlayer)/(?P<id>[\d-]+)'

     _TESTS = [{
         # metadata in JSON
         'url': 'http://ok.ru/video/20079905452',
     _TESTS = [{
         # metadata in JSON
         'url': 'http://ok.ru/video/20079905452',

-        'md5': '8e24ad2da6f387948e7a7d44eb8668fe',
+        'md5': '6ba728d85d60aa2e6dd37c9e70fdc6bc',

         'info_dict': {
             'id': '20079905452',
             'ext': 'mp4',
             'title': 'Культура меняет нас (прекрасный ролик!))',
             'duration': 100,
         '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,
             'uploader_id': '330537914540',
             'uploader': 'Виталий Добровольский',
             'like_count': int,

+            'age_limit': 0,

         },
         },

+        'skip': 'Video has been blocked',

     }, {
         # metadataUrl
     }, {
         # metadataUrl

-        'url': 'http://ok.ru/video/63567059965189-0',
-        'md5': '9676cf86eff5391d35dea675d224e131',
+        'url': 'http://ok.ru/video/63567059965189-0?fromTime=5',
+        'md5': '6ff470ea2dd51d5d18c295a355b0b6bc',

         'info_dict': {
             'id': '63567059965189-0',
             'ext': 'mp4',
             'title': 'Девушка без комплексов ...',
             'duration': 191,
         'info_dict': {
             'id': '63567059965189-0',
             'ext': 'mp4',
             'title': 'Девушка без комплексов ...',
             'duration': 191,

+            'upload_date': '20150518',

             'uploader_id': '534380003155',
             'uploader_id': '534380003155',

-            'uploader': 'Андрей Мещанинов',
+            'uploader': '☭ Андрей Мещанинов ☭',

             'like_count': int,
             'like_count': int,

+            'age_limit': 0,
+            'start_time': 5,
+        },
+    }, {
+        # YouTube embed (metadataUrl, provider == USER_YOUTUBE)
+        'url': 'http://ok.ru/video/64211978996595-1',
+        'md5': '2f206894ffb5dbfcce2c5a14b909eea5',
+        'info_dict': {
+            'id': '64211978996595-1',
+            'ext': 'mp4',
+            'title': 'Космическая среда от 26 августа 2015',
+            'description': 'md5:848eb8b85e5e3471a3a803dae1343ed0',
+            'duration': 440,
+            'upload_date': '20150826',
+            'uploader_id': 'tvroscosmos',
+            'uploader': 'Телестудия Роскосмоса',
+            'age_limit': 0,

         },
         },

+    }, {
+        # YouTube embed (metadata, provider == USER_YOUTUBE, no metadata.movie.title field)
+        'url': 'http://ok.ru/video/62036049272859-0',
+        'info_dict': {
+            'id': '62036049272859-0',
+            'ext': 'mp4',
+            'title': 'МУЗЫКА     ДОЖДЯ .',
+            'description': 'md5:6f1867132bd96e33bf53eda1091e8ed0',
+            'upload_date': '20120106',
+            'uploader_id': '473534735899',
+            'uploader': 'МARINA D',
+            'age_limit': 0,
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'skip': 'Video has not been found',

     }, {
         'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452',
         'only_matching': True,
     }, {
         'url': 'http://ok.ru/web-api/video/moviePlayer/20079905452',
         'only_matching': True,

+    }, {
+        'url': 'http://www.ok.ru/video/20648036891',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.ok.ru/videoembed/20648036891',
+        'only_matching': True,
+    }, {
+        'url': 'http://m.ok.ru/video/20079905452',
+        'only_matching': True,
+    }, {
+        'url': 'http://mobile.ok.ru/video/20079905452',
+        'only_matching': True,

     }]
 
     def _real_extract(self, url):
     }]
 
     def _real_extract(self, url):

+        start_time = int_or_none(compat_parse_qs(
+            compat_urllib_parse_urlparse(url).query).get('fromTime', [None])[0])
+

         video_id = self._match_id(url)
 
         webpage = self._download_webpage(
             'http://ok.ru/video/%s' % video_id, video_id)
 
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(
             'http://ok.ru/video/%s' % video_id, video_id)
 

+        error = self._search_regex(
+            r'[^>]+class="vp_video_stub_txt"[^>]*>([^<]+)<',
+            webpage, 'error', default=None)
+        if error:
+            raise ExtractorError(error, expected=True)
+

         player = self._parse_json(
             unescapeHTML(self._search_regex(
         player = self._parse_json(
             unescapeHTML(self._search_regex(

-                r'data-attributes="([^"]+)"', webpage, 'player')),
+                r'data-options=(?P<quote>["\'])(?P<player>{.+?%s.+?})(?P=quote)' % video_id,
+                webpage, 'player', group='player')),

             video_id)
 
         flashvars = player['flashvars']
             video_id)
 
         flashvars = player['flashvars']


@@ -66,7 +132,14 @@ class OdnoklassnikiIE(InfoExtractor):
                 video_id, 'Downloading metadata JSON')
 
         movie = metadata['movie']
                 video_id, 'Downloading metadata JSON')
 
         movie = metadata['movie']

-        title = movie['title']
+
+        # Some embedded videos may not contain title in movie dict (e.g.
+        # http://ok.ru/video/62036049272859-0) thus we allow missing title
+        # here and it's going to be extracted later by an extractor that
+        # will process the actual embed.
+        provider = metadata.get('provider')
+        title = movie['title'] if provider == 'UPLOADED_ODKL' else movie.get('title')
+

         thumbnail = movie.get('poster')
         duration = int_or_none(movie.get('duration'))
 
         thumbnail = movie.get('poster')
         duration = int_or_none(movie.get('duration'))
 


@@ -85,16 +158,7 @@ class OdnoklassnikiIE(InfoExtractor):
 
         like_count = int_or_none(metadata.get('likeCount'))
 
 
         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 {
+        info = {

             'id': video_id,
             'title': title,
             'thumbnail': thumbnail,
             'id': video_id,
             'title': title,
             'thumbnail': thumbnail,


@@ -104,5 +168,43 @@ class OdnoklassnikiIE(InfoExtractor):
             'uploader_id': uploader_id,
             'like_count': like_count,
             'age_limit': age_limit,
             'uploader_id': uploader_id,
             'like_count': like_count,
             'age_limit': age_limit,

-            'formats': formats,
+            'start_time': start_time,

         }
         }

+
+        if provider == 'USER_YOUTUBE':
+            info.update({
+                '_type': 'url_transparent',
+                'url': movie['contentId'],
+            })
+            return info
+
+        quality = qualities(('4', '0', '1', '2', '3', '5'))
+
+        formats = [{
+            'url': f['url'],
+            'ext': 'mp4',
+            'format_id': f['name'],
+        } for f in metadata['videos']]
+
+        m3u8_url = metadata.get('hlsManifestUrl')
+        if m3u8_url:
+            formats.extend(self._extract_m3u8_formats(
+                m3u8_url, video_id, 'mp4', 'm3u8_native',
+                m3u8_id='hls', fatal=False))
+
+        dash_manifest = metadata.get('metadataEmbedded')
+        if dash_manifest:
+            formats.extend(self._parse_mpd_formats(
+                compat_etree_fromstring(dash_manifest), 'mpd'))
+
+        for fmt in formats:
+            fmt_type = self._search_regex(
+                r'\btype[/=](\d)', fmt['url'],
+                'format type', default=None)
+            if fmt_type:
+                fmt['quality'] = quality(fmt_type)
+
+        self._sort_formats(formats)
+
+        info['formats'] = formats
+        return info