[youtube] Add support for invidiou.sh (#20309)

[youtube-dl] / youtube_dl / extractor / mgtv.py

diff --git a/youtube_dl/extractor/mgtv.py b/youtube_dl/extractor/mgtv.py
index 8f16a8f4c3ae0c5038c1aeeb36e1db077ce6204d..d53d96aae1799e15d578c765545a929255009d31 100644 (file)
--- a/youtube_dl/extractor/mgtv.py
+++ b/youtube_dl/extractor/mgtv.py
@@ -2,59 +2,65 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import int_or_none
 
 
 class MGTVIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.mgtv\.com/v/(?:[^/]+/)*(?P<id>\d+).html'
+    _VALID_URL = r'https?://(?:www\.)?mgtv\.com/(v|b)/(?:[^/]+/)*(?P<id>\d+)\.html'
+    IE_DESC = '芒果TV'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',
-        'md5': '',
+        'md5': 'b1ffc0fc163152acf6beaa81832c9ee7',
         'info_dict': {
             'id': '3116640',
             'ext': 'mp4',
             'title': '我是歌手第四季双年巅峰会：韩红李玟“双王”领军对抗',
             'description': '我是歌手第四季双年巅峰会',
             'duration': 7461,
-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',
         },
-        'params': {
-            'skip_download': True,  # m3u8 download
-        },
-    }
-
-    _FORMAT_MAP = {
-        '标清': ('Standard', 0),
-        '高清': ('High', 1),
-        '超清': ('SuperHigh', 2),
-    }
+    }, {
+        'url': 'http://www.mgtv.com/b/301817/3826653.html',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         api_data = self._download_json(
-            'http://v.api.mgtv.com/player/video', video_id,
-            query={'video_id': video_id})['data']
+            'http://pcweb.api.mgtv.com/player/video', video_id,
+            query={'video_id': video_id},
+            headers=self.geo_verification_headers())['data']
         info = api_data['info']
+        title = info['title'].strip()
+        stream_domain = api_data['stream_domain'][0]
 
         formats = []
         for idx, stream in enumerate(api_data['stream']):
-            format_name = stream.get('name')
-            format_id, preference = self._FORMAT_MAP.get(format_name, (None, None))
-            format_info = self._download_json(
-                stream['url'], video_id,
-                note='Download video info for format %s' % format_id or '#%d' % idx)
+            stream_path = stream.get('url')
+            if not stream_path:
+                continue
+            format_data = self._download_json(
+                stream_domain + stream_path, video_id,
+                note='Download video info for format #%d' % idx)
+            format_url = format_data.get('info')
+            if not format_url:
+                continue
+            tbr = int_or_none(self._search_regex(
+                r'_(\d+)_mp4/', format_url, 'tbr', default=None))
             formats.append({
-                'format_id': format_id,
-                'url': format_info['info'],
-                'ext': 'mp4',  # These are m3u8 playlists
-                'preference': preference,
+                'format_id': compat_str(tbr or idx),
+                'url': format_url,
+                'ext': 'mp4',
+                'tbr': tbr,
+                'protocol': 'm3u8_native',
             })
         self._sort_formats(formats)
 
         return {
             'id': video_id,
-            'title': info['title'].strip(),
+            'title': title,
             'formats': formats,
             'description': info.get('desc'),
             'duration': int_or_none(info.get('duration')),