d/p/remove-autoupdate-mechanism.patch: Update patch with metadata at the top.

[youtubedl] / youtube_dl / extractor / discovery.py

diff --git a/youtube_dl/extractor/discovery.py b/youtube_dl/extractor/discovery.py
index 52c2d7ddf99873779b7f3223b0acfe4563e2b5d9..91449dcd8549e992afed748651ad0e3312812721 100644 (file)
--- a/youtube_dl/extractor/discovery.py
+++ b/youtube_dl/extractor/discovery.py
@@ -1,47 +1,87 @@
 from __future__ import unicode_literals
 
 from __future__ import unicode_literals
 

+import random

 import re
 import re

-import json
+import string

 
 

-from .common import InfoExtractor
+from .discoverygo import DiscoveryGoBaseIE
+from ..compat import compat_str
+from ..utils import (
+    ExtractorError,
+    try_get,
+)
+from ..compat import compat_HTTPError

 
 
 
 

-class DiscoveryIE(InfoExtractor):
-    _VALID_URL = r'http://www\.discovery\.com\/[a-zA-Z0-9\-]*/[a-zA-Z0-9\-]*/videos/(?P<id>[a-zA-Z0-9\-]*)(.htm)?'
-    _TEST = {
-        'url': 'http://www.discovery.com/tv-shows/mythbusters/videos/mission-impossible-outtakes.htm',
-        'md5': 'e12614f9ee303a6ccef415cb0793eba2',
+class DiscoveryIE(DiscoveryGoBaseIE):
+    _VALID_URL = r'''(?x)https?://(?:www\.)?(?P<site>
+            discovery|
+            investigationdiscovery|
+            discoverylife|
+            animalplanet|
+            ahctv|
+            destinationamerica|
+            sciencechannel|
+            tlc|
+            velocity
+        )\.com(?P<path>/tv-shows/[^/]+/(?:video|full-episode)s/(?P<id>[^./?#]+))'''
+    _TESTS = [{
+        'url': 'https://www.discovery.com/tv-shows/cash-cab/videos/dave-foley',

         'info_dict': {
         'info_dict': {

-            'id': '614784',
+            'id': '5a2d9b4d6b66d17a5026e1fd',

             'ext': 'mp4',
             'ext': 'mp4',

-            'title': 'MythBusters: Mission Impossible Outtakes',
-            'description': ('Watch Jamie Hyneman and Adam Savage practice being'
-                            ' each other -- to the point of confusing Jamie\'s dog -- and '
-                            'don\'t miss Adam moon-walking as Jamie ... behind Jamie\'s'
-                            ' back.'),
-            'duration': 156,
+            'title': 'Dave Foley',
+            'description': 'md5:4b39bcafccf9167ca42810eb5f28b01f',
+            'duration': 608,

         },
         },

-    }
+        'params': {
+            'skip_download': True,  # requires ffmpeg
+        }
+    }, {
+        'url': 'https://www.investigationdiscovery.com/tv-shows/final-vision/full-episodes/final-vision',
+        'only_matching': True,
+    }]
+    _GEO_COUNTRIES = ['US']
+    _GEO_BYPASS = False

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

-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        webpage = self._download_webpage(url, video_id)
-
-        video_list_json = self._search_regex(r'var videoListJSON = ({.*?});',
-                                             webpage, 'video list', flags=re.DOTALL)
-        video_list = json.loads(video_list_json)
-        info = video_list['clips'][0]
-        formats = []
-        for f in info['mp4']:
-            formats.append(
-                {'url': f['src'], 'ext': 'mp4', 'tbr': int(f['bitrate'][:-1])})
-
-        return {
-            'id': info['contentId'],
-            'title': video_list['name'],
-            'formats': formats,
-            'description': info['videoCaption'],
-            'thumbnail': info.get('videoStillURL') or info.get('thumbnailURL'),
-            'duration': info['duration'],
-        }
+        site, path, display_id = re.match(self._VALID_URL, url).groups()
+        webpage = self._download_webpage(url, display_id)
+
+        react_data = self._parse_json(self._search_regex(
+            r'window\.__reactTransmitPacket\s*=\s*({.+?});',
+            webpage, 'react data'), display_id)
+        content_blocks = react_data['layout'][path]['contentBlocks']
+        video = next(cb for cb in content_blocks if cb.get('type') == 'video')['content']['items'][0]
+        video_id = video['id']
+
+        access_token = self._download_json(
+            'https://www.%s.com/anonymous' % site, display_id, query={
+                'authRel': 'authorization',
+                'client_id': try_get(
+                    react_data, lambda x: x['application']['apiClientId'],
+                    compat_str) or '3020a40c2356a645b4b4',
+                'nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
+                'redirectUri': 'https://fusion.ddmcdn.com/app/mercury-sdk/180/redirectHandler.html?https://www.%s.com' % site,
+            })['access_token']
+
+        try:
+            stream = self._download_json(
+                'https://api.discovery.com/v1/streaming/video/' + video_id,
+                display_id, headers={
+                    'Authorization': 'Bearer ' + access_token,
+                })
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                e_description = self._parse_json(
+                    e.cause.read().decode(), display_id)['description']
+                if 'resource not available for country' in e_description:
+                    self.raise_geo_restricted(countries=self._GEO_COUNTRIES)
+                if 'Authorized Networks' in e_description:
+                    raise ExtractorError(
+                        'This video is only available via cable service provider subscription that'
+                        ' is not currently supported. You may want to use --cookies.', expected=True)
+                raise ExtractorError(e_description)
+            raise
+
+        return self._extract_video_info(video, stream, display_id)