New upstream version 2019.06.08

[youtubedl] / youtube_dl / extractor / mgtv.py

diff --git a/youtube_dl/extractor/mgtv.py b/youtube_dl/extractor/mgtv.py
index 27bdff8b25cb63e628981ad3a8dbcc7f76db7653..84137df502522a77561fb8d86ebfb8d7febb2b95 100644 (file)
--- a/youtube_dl/extractor/mgtv.py
+++ b/youtube_dl/extractor/mgtv.py
@@ -1,68 +1,90 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
 # coding: utf-8
 from __future__ import unicode_literals
 

+import base64
+import time
+import uuid
+

 from .common import InfoExtractor
 from .common import InfoExtractor

-from ..utils import int_or_none
+from ..compat import (
+    compat_HTTPError,
+    compat_str,
+)
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+)

 
 
 class MGTVIE(InfoExtractor):
 
 
 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'
     IE_DESC = '芒果TV'

+    _GEO_COUNTRIES = ['CN']

 
     _TESTS = [{
         'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',
 
     _TESTS = [{
         'url': 'http://www.mgtv.com/v/1/290525/f/3116640.html',

-        'md5': '1bdadcf760a0b90946ca68ee9a2db41a',

         'info_dict': {
             'id': '3116640',
             'ext': 'mp4',
         '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',
+            'title': '我是歌手 第四季',

             'description': '我是歌手第四季双年巅峰会',
             'duration': 7461,
             'description': '我是歌手第四季双年巅峰会',
             'duration': 7461,

-            'thumbnail': 're:^https?://.*\.jpg$',
+            'thumbnail': r're:^https?://.*\.jpg$',

         },
     }, {
         },
     }, {

-        # no tbr extracted from stream_url
-        'url': 'http://www.mgtv.com/v/1/1/f/3324755.html',
+        'url': 'http://www.mgtv.com/b/301817/3826653.html',

         'only_matching': True,
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         '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},
-            headers=self.geo_verification_headers())['data']
+        try:
+            api_data = self._download_json(
+                'https://pcweb.api.mgtv.com/player/video', video_id, query={
+                    'tk2': base64.urlsafe_b64encode(b'did=%s|pno=1030|ver=0.3.0301|clit=%d' % (compat_str(uuid.uuid4()).encode(), time.time()))[::-1],
+                    'video_id': video_id,
+                }, headers=self.geo_verification_headers())['data']
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                error = self._parse_json(e.cause.read().decode(), None)
+                if error.get('code') == 40005:
+                    self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
+                raise ExtractorError(error['msg'], expected=True)
+            raise

         info = api_data['info']
         info = api_data['info']

+        title = info['title'].strip()
+        stream_data = self._download_json(
+            'https://pcweb.api.mgtv.com/player/getSource', video_id, query={
+                'pm2': api_data['atc']['pm2'],
+                'video_id': video_id,
+            }, headers=self.geo_verification_headers())['data']
+        stream_domain = stream_data['stream_domain'][0]

 
         formats = []
 
         formats = []

-        for idx, stream in enumerate(api_data['stream']):
-            stream_url = stream.get('url')
-            if not stream_url:
+        for idx, stream in enumerate(stream_data['stream']):
+            stream_path = stream.get('url')
+            if not stream_path:

                 continue
                 continue

-            tbr = int_or_none(self._search_regex(
-                r'(\d+)\.mp4', stream_url, 'tbr', default=None))
-
-            def extract_format(stream_url, format_id, idx, query={}):
-                format_info = self._download_json(
-                    stream_url, video_id,
-                    note='Download video info for format %s' % (format_id or '#%d' % idx),
-                    query=query)
-                return {
-                    'format_id': format_id,
-                    'url': format_info['info'],
-                    'ext': 'mp4',
-                    'tbr': tbr,
-                }
-
-            formats.append(extract_format(
-                stream_url, 'hls-%d' % tbr if tbr else None, idx * 2))
-            formats.append(extract_format(stream_url.replace(
-                '/playlist.m3u8', ''), 'http-%d' % tbr if tbr else None, idx * 2 + 1, {'pno': 1031}))
+            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(stream.get('filebitrate') or self._search_regex(
+                r'_(\d+)_mp4/', format_url, 'tbr', default=None))
+            formats.append({
+                '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,
         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')),
             'formats': formats,
             'description': info.get('desc'),
             'duration': int_or_none(info.get('duration')),