New upstream version 2017.11.06

[youtubedl] / youtube_dl / extractor / odnoklassniki.py

diff --git a/youtube_dl/extractor/odnoklassniki.py b/youtube_dl/extractor/odnoklassniki.py
index f9e064a60e445668200b759ca4e0ad1a6f7c28ab..854b6800c18bd47d2ed1b40cace0969a22eebc75 100644 (file)
--- a/youtube_dl/extractor/odnoklassniki.py
+++ b/youtube_dl/extractor/odnoklassniki.py
@@ -2,7 +2,12 @@
 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 (
     ExtractorError,
     unified_strdate,
@@ -32,8 +37,8 @@ class OdnoklassnikiIE(InfoExtractor):
         'skip': 'Video has been blocked',
     }, {
         # 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',
@@ -44,11 +49,12 @@ class OdnoklassnikiIE(InfoExtractor):
             'uploader': '☭ Андрей Мещанинов ☭',
             'like_count': int,
             'age_limit': 0,
+            'start_time': 5,
         },
     }, {
         # YouTube embed (metadataUrl, provider == USER_YOUTUBE)
         'url': 'http://ok.ru/video/64211978996595-1',
-        'md5': '5d7475d428845cd2e13bae6f1a992278',
+        'md5': '2f206894ffb5dbfcce2c5a14b909eea5',
         'info_dict': {
             'id': '64211978996595-1',
             'ext': 'mp4',
@@ -56,10 +62,27 @@ class OdnoklassnikiIE(InfoExtractor):
             'description': 'md5:848eb8b85e5e3471a3a803dae1343ed0',
             'duration': 440,
             'upload_date': '20150826',
-            'uploader_id': '750099571',
-            'uploader': 'Ð\90лина Ð\9f',
+            'uploader_id': 'tvroscosmos',
+            'uploader': 'ТелеÑ\81Ñ\82Ñ\83диÑ\8f РоÑ\81коÑ\81моÑ\81а',
             '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,
@@ -78,6 +101,9 @@ class OdnoklassnikiIE(InfoExtractor):
     }]
 
     def _real_extract(self, url):
+        start_time = int_or_none(compat_parse_qs(
+            compat_urllib_parse_urlparse(url).query).get('fromTime', [None])[0])
+
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(
@@ -106,7 +132,14 @@ class OdnoklassnikiIE(InfoExtractor):
                 video_id, 'Downloading metadata JSON')
 
         movie = metadata['movie']
-        title = movie['title']
+
+        # Some embedded videos may not contain title in movie dict (e.g.
+        # http://ok.ru/video/62036049272859-0) thus we allow missing title
+        # here and it's going to be extracted later by an extractor that
+        # will process the actual embed.
+        provider = metadata.get('provider')
+        title = movie['title'] if provider == 'UPLOADED_ODKL' else movie.get('title')
+
         thumbnail = movie.get('poster')
         duration = int_or_none(movie.get('duration'))
 
@@ -135,23 +168,42 @@ class OdnoklassnikiIE(InfoExtractor):
             'uploader_id': uploader_id,
             'like_count': like_count,
             'age_limit': age_limit,
+            'start_time': start_time,
         }
 
-        if metadata.get('provider') == 'USER_YOUTUBE':
+        if provider == 'USER_YOUTUBE':
             info.update({
                 '_type': 'url_transparent',
                 'url': movie['contentId'],
             })
             return info
 
-        quality = qualities(('mobile', 'lowest', 'low', 'sd', 'hd'))
+        quality = qualities(('4', '0', '1', '2', '3', '5'))
 
         formats = [{
             'url': f['url'],
             'ext': 'mp4',
             'format_id': f['name'],
-            'quality': quality(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