Update upstream source from tag 'upstream/2020.05.08'

[youtubedl] / youtube_dl / extractor / discovery.py

diff --git a/youtube_dl/extractor/discovery.py b/youtube_dl/extractor/discovery.py
index 91449dcd8549e992afed748651ad0e3312812721..e0139cc862d74bc3c20a9d6567747b96b642c730 100644 (file)
--- a/youtube_dl/extractor/discovery.py
+++ b/youtube_dl/extractor/discovery.py
@@ -5,34 +5,43 @@ import re
 import string
 
 from .discoverygo import DiscoveryGoBaseIE
-from ..compat import compat_str
-from ..utils import (
-    ExtractorError,
-    try_get,
-)
+from ..compat import compat_urllib_parse_unquote
+from ..utils import ExtractorError
 from ..compat import compat_HTTPError
 
 
 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>[^./?#]+))'''
+    _VALID_URL = r'''(?x)https?://
+        (?P<site>
+            go\.discovery|
+            www\.
+                (?:
+                    investigationdiscovery|
+                    discoverylife|
+                    animalplanet|
+                    ahctv|
+                    destinationamerica|
+                    sciencechannel|
+                    tlc
+                )|
+            watch\.
+                (?:
+                    hgtv|
+                    foodnetwork|
+                    travelchannel|
+                    diynetwork|
+                    cookingchanneltv|
+                    motortrend
+                )
+        )\.com/tv-shows/(?P<show_slug>[^/]+)/(?:video|full-episode)s/(?P<id>[^./?#]+)'''
     _TESTS = [{
-        'url': 'https://www.discovery.com/tv-shows/cash-cab/videos/dave-foley',
+        'url': 'https://go.discovery.com/tv-shows/cash-cab/videos/riding-with-matthew-perry',
         'info_dict': {
-            'id': '5a2d9b4d6b66d17a5026e1fd',
+            'id': '5a2f35ce6b66d17a5026e29e',
             'ext': 'mp4',
-            'title': 'Dave Foley',
-            'description': 'md5:4b39bcafccf9167ca42810eb5f28b01f',
-            'duration': 608,
+            'title': 'Riding with Matthew Perry',
+            'description': 'md5:a34333153e79bc4526019a5129e7f878',
+            'duration': 84,
         },
         'params': {
             'skip_download': True,  # requires ffmpeg
@@ -40,39 +49,61 @@ class DiscoveryIE(DiscoveryGoBaseIE):
     }, {
         'url': 'https://www.investigationdiscovery.com/tv-shows/final-vision/full-episodes/final-vision',
         'only_matching': True,
+    }, {
+        'url': 'https://go.discovery.com/tv-shows/alaskan-bush-people/videos/follow-your-own-road',
+        'only_matching': True,
+    }, {
+        # using `show_slug` is important to get the correct video data
+        'url': 'https://www.sciencechannel.com/tv-shows/mythbusters-on-science/full-episodes/christmas-special',
+        'only_matching': True,
     }]
     _GEO_COUNTRIES = ['US']
     _GEO_BYPASS = False
+    _API_BASE_URL = 'https://api.discovery.com/v1/'
 
     def _real_extract(self, url):
-        site, path, display_id = re.match(self._VALID_URL, url).groups()
-        webpage = self._download_webpage(url, display_id)
+        site, show_slug, display_id = re.match(self._VALID_URL, url).groups()
+
+        access_token = None
+        cookies = self._get_cookies(url)
+
+        # prefer Affiliate Auth Token over Anonymous Auth Token
+        auth_storage_cookie = cookies.get('eosAf') or cookies.get('eosAn')
+        if auth_storage_cookie and auth_storage_cookie.value:
+            auth_storage = self._parse_json(compat_urllib_parse_unquote(
+                compat_urllib_parse_unquote(auth_storage_cookie.value)),
+                display_id, fatal=False) or {}
+            access_token = auth_storage.get('a') or auth_storage.get('access_token')
 
-        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']
+        if not access_token:
+            access_token = self._download_json(
+                'https://%s.com/anonymous' % site, display_id,
+                'Downloading token JSON metadata', query={
+                    'authRel': 'authorization',
+                    'client_id': '3020a40c2356a645b4b4',
+                    'nonce': ''.join([random.choice(string.ascii_letters) for _ in range(32)]),
+                    'redirectUri': 'https://www.discovery.com/',
+                })['access_token']
 
-        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']
+        headers = self.geo_verification_headers()
+        headers['Authorization'] = 'Bearer ' + access_token
 
         try:
+            video = self._download_json(
+                self._API_BASE_URL + 'content/videos',
+                display_id, 'Downloading content JSON metadata',
+                headers=headers, query={
+                    'embed': 'show.name',
+                    'fields': 'authenticated,description.detailed,duration,episodeNumber,id,name,parental.rating,season.number,show,tags',
+                    'slug': display_id,
+                    'show_slug': show_slug,
+                })[0]
+            video_id = video['id']
             stream = self._download_json(
-                'https://api.discovery.com/v1/streaming/video/' + video_id,
-                display_id, headers={
-                    'Authorization': 'Bearer ' + access_token,
-                })
+                self._API_BASE_URL + 'streaming/video/' + video_id,
+                display_id, 'Downloading streaming JSON metadata', headers=headers)
         except ExtractorError as e:
-            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code in (401, 403):
                 e_description = self._parse_json(
                     e.cause.read().decode(), display_id)['description']
                 if 'resource not available for country' in e_description: