Merge tag 'upstream/2017.02.24.1'

[youtubedl] / youtube_dl / extractor / mailru.py

diff --git a/youtube_dl/extractor/mailru.py b/youtube_dl/extractor/mailru.py
index 7460d81cd501b8c52dcce3caae8313f6854b571a..f7cc3c83289f1101207c385d5bfed2055c7b7f67 100644 (file)
--- a/youtube_dl/extractor/mailru.py
+++ b/youtube_dl/extractor/mailru.py
@@ -1,22 +1,26 @@
-# encoding: utf-8
+# coding: utf-8

 from __future__ import unicode_literals
 
 import re
 
 from .common import InfoExtractor
 from __future__ import unicode_literals
 
 import re
 
 from .common import InfoExtractor

+from ..utils import (
+    int_or_none,
+    remove_end,
+)

 
 
 class MailRuIE(InfoExtractor):
     IE_NAME = 'mailru'
     IE_DESC = 'Видео@Mail.Ru'
 
 
 class MailRuIE(InfoExtractor):
     IE_NAME = 'mailru'
     IE_DESC = 'Видео@Mail.Ru'

-    _VALID_URL = r'http://(?:www\.)?my\.mail\.ru/(?:video/.*#video=/?(?P<idv1>(?:[^/]+/){3}\d+)|(?:(?P<idv2prefix>(?:[^/]+/){2})video/(?P<idv2suffix>[^/]+/\d+))\.html)'
+    _VALID_URL = r'https?://(?:(?:www|m)\.)?my\.mail\.ru/(?:video/.*#video=/?(?P<idv1>(?:[^/]+/){3}\d+)|(?:(?P<idv2prefix>(?:[^/]+/){2})video/(?P<idv2suffix>[^/]+/\d+))\.html)'

 
     _TESTS = [
         {
             'url': 'http://my.mail.ru/video/top#video=/mail/sonypicturesrus/75/76',
             'md5': 'dea205f03120046894db4ebb6159879a',
             'info_dict': {
 
     _TESTS = [
         {
             'url': 'http://my.mail.ru/video/top#video=/mail/sonypicturesrus/75/76',
             'md5': 'dea205f03120046894db4ebb6159879a',
             'info_dict': {

-                'id': '46301138',
+                'id': '46301138_76',

                 'ext': 'mp4',
                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
                 'timestamp': 1393232740,
                 'ext': 'mp4',
                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
                 'timestamp': 1393232740,


@@ -25,21 +29,43 @@ class MailRuIE(InfoExtractor):
                 'uploader_id': 'sonypicturesrus@mail.ru',
                 'duration': 184,
             },
                 'uploader_id': 'sonypicturesrus@mail.ru',
                 'duration': 184,
             },

+            'skip': 'Not accessible from Travis CI server',

         },
         {
             'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html',
             'md5': '00a91a58c3402204dcced523777b475f',
             'info_dict': {
         },
         {
             'url': 'http://my.mail.ru/corp/hitech/video/news_hi-tech_mail_ru/1263.html',
             'md5': '00a91a58c3402204dcced523777b475f',
             'info_dict': {

-                'id': '46843144',
+                'id': '46843144_1263',

                 'ext': 'mp4',
                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
                 '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,
             },
                 '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',
+        },
+        {
+            'url': 'http://m.my.mail.ru/mail/3sktvtr/video/_myvideo/138.html',
+            'only_matching': True,
+        }

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


@@ -49,38 +75,64 @@ class MailRuIE(InfoExtractor):
         if not video_id:
             video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
 
         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)
+
+        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)

 
 

-        author = video_data['author']
-        uploader = author['name']
-        uploader_id = author['id']
+        meta_data = video_data['meta']
+        title = remove_end(meta_data['title'], '.mp4')

 
 

-        movie = video_data['movie']
-        content_id = str(movie['contentId'])
-        title = movie['title']
-        if title.endswith('.mp4'):
-            title = title[:-4]
-        thumbnail = movie['poster']
-        duration = movie['duration']
+        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'))

 
 

-        view_count = video_data['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

 
 

-        formats = [
-            {
-                'url': video['url'],
-                'format_id': video['name'],
-            } for video in video_data['videos']
-        ]
+        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,
             'title': title,
             'thumbnail': thumbnail,
 
         return {
             'id': content_id,
             'title': title,
             'thumbnail': thumbnail,

-            'timestamp': video_data['timestamp'],
+            'timestamp': timestamp,

             'uploader': uploader,
             'uploader_id': uploader_id,
             'duration': duration,
             'view_count': view_count,
             'formats': formats,
             'uploader': uploader,
             'uploader_id': uploader_id,
             'duration': duration,
             'view_count': view_count,
             'formats': formats,

-        }
\ No newline at end of file
+        }