Start new release.

[youtubedl] / youtube_dl / extractor / mailru.py

diff --git a/youtube_dl/extractor/mailru.py b/youtube_dl/extractor/mailru.py
index ab1300185099a20835fdde08b88635ceacafe191..6b7c5e3e03dc5b2cb6012a6e03d23a71fe359fee 100644 (file)
--- a/youtube_dl/extractor/mailru.py
+++ b/youtube_dl/extractor/mailru.py
@@ -1,16 +1,27 @@
-# 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'''(?x)
+                    https?://
+                        (?:(?:www|m)\.)?my\.mail\.ru/
+                        (?:
+                            video/.*\#video=/?(?P<idv1>(?:[^/]+/){3}\d+)|
+                            (?:(?P<idv2prefix>(?:[^/]+/){2})video/(?P<idv2suffix>[^/]+/\d+))\.html|
+                            (?:video/embed|\+/video/meta)/(?P<metaid>\d+)
+                        )
+                    '''

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


@@ -19,7 +30,7 @@ class MailRuIE(InfoExtractor):
                 'id': '46301138_76',
                 'ext': 'mp4',
                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',
                 'id': '46301138_76',
                 'ext': 'mp4',
                 'title': 'Новый Человек-Паук. Высокое напряжение. Восстание Электро',

-                'timestamp': 1393232740,
+                'timestamp': 1393235077,

                 'upload_date': '20140224',
                 'uploader': 'sonypicturesrus',
                 'uploader_id': 'sonypicturesrus@mail.ru',
                 'upload_date': '20140224',
                 'uploader': 'sonypicturesrus',
                 'uploader_id': 'sonypicturesrus@mail.ru',


@@ -34,49 +45,106 @@ class MailRuIE(InfoExtractor):
                 'id': '46843144_1263',
                 'ext': 'mp4',
                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',
                 'id': '46843144_1263',
                 'ext': 'mp4',
                 'title': 'Samsung Galaxy S5 Hammer Smash Fail Battery Explosion',

-                'timestamp': 1397217632,
-                'upload_date': '20140411',
+                'timestamp': 1397039888,
+                'upload_date': '20140409',

                 'uploader': 'hitech',
                 'uploader_id': 'hitech@corp.mail.ru',
                 'duration': 245,
             },
             'skip': 'Not accessible from Travis CI server',
         },
                 'uploader': 'hitech',
                 '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,
+        },
+        {
+            'url': 'https://my.mail.ru/video/embed/7949340477499637815',
+            'only_matching': True,
+        },
+        {
+            'url': 'http://my.mail.ru/+/video/meta/7949340477499637815',
+            'only_matching': True,
+        }

     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
     ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)

-        video_id = mobj.group('idv1')
+        meta_id = mobj.group('metaid')

 
 

-        if not video_id:
-            video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
+        video_id = None
+        if meta_id:
+            meta_url = 'https://my.mail.ru/+/video/meta/%s' % meta_id
+        else:
+            video_id = mobj.group('idv1')
+            if not video_id:
+                video_id = mobj.group('idv2prefix') + mobj.group('idv2suffix')
+            webpage = self._download_webpage(url, video_id)
+            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')
+            else:
+                meta_url = None

 
 

-        video_data = self._download_json(
-            'http://api.video.mail.ru/videos/%s.json?new=1' % video_id, video_id, 'Downloading video JSON')
+        video_data = None
+        if meta_url:
+            video_data = self._download_json(
+                meta_url, video_id or meta_id, 'Downloading video meta JSON',
+                fatal=not 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')
+        # 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']
 
         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,
 
         return {
             'id': content_id,