[twitch:clips] Fix extraction (Closes #9767)

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

diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py
index c92dcc7b90f151a9ed05e82428dc14691b43e220..5757eb119864c165f4577642cd7a2879e3054a86 100644 (file)
--- a/youtube_dl/extractor/twitch.py
+++ b/youtube_dl/extractor/twitch.py
@@ -16,10 +16,12 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     int_or_none,
+    js_to_json,
     orderedSet,
     parse_duration,
     parse_iso8601,
     sanitized_Request,
+    urlencode_postdata,
 )
 
 
@@ -27,7 +29,7 @@ class TwitchBaseIE(InfoExtractor):
     _VALID_URL_BASE = r'https?://(?:www\.)?twitch\.tv'
 
     _API_BASE = 'https://api.twitch.tv'
-    _USHER_BASE = 'http://usher.twitch.tv'
+    _USHER_BASE = 'https://usher.ttvnw.net'
     _LOGIN_URL = 'http://www.twitch.tv/login'
     _NETRC_MACHINE = 'twitch'
 
@@ -81,7 +83,7 @@ class TwitchBaseIE(InfoExtractor):
             post_url = compat_urlparse.urljoin(redirect_url, post_url)
 
         request = sanitized_Request(
-            post_url, compat_urllib_parse_urlencode(login_form).encode('utf-8'))
+            post_url, urlencode_postdata(login_form))
         request.add_header('Referer', redirect_url)
         response = self._download_webpage(
             request, None, 'Logging in as %s' % username)
@@ -170,6 +172,7 @@ class TwitchVideoIE(TwitchItemBaseIE):
             'title': 'Worlds Semifinals - Star Horn Royal Club vs. OMG',
         },
         'playlist_mincount': 12,
+        'skip': 'HTTP Error 404: Not Found',
     }
 
 
@@ -186,6 +189,7 @@ class TwitchChapterIE(TwitchItemBaseIE):
             'title': 'ACRL Off Season - Sports Cars @ Nordschleife',
         },
         'playlist_mincount': 3,
+        'skip': 'HTTP Error 404: Not Found',
     }, {
         'url': 'http://www.twitch.tv/tsm_theoddone/c/2349361',
         'only_matching': True,
@@ -257,7 +261,7 @@ class TwitchVodIE(TwitchItemBaseIE):
                     'nauth': access_token['token'],
                     'nauthsig': access_token['sig'],
                 })),
-            item_id, 'mp4')
+            item_id, 'mp4', entry_protocol='m3u8_native')
 
         self._prefer_source(formats)
         info['formats'] = formats
@@ -354,31 +358,6 @@ class TwitchPastBroadcastsIE(TwitchPlaylistBaseIE):
     }
 
 
-class TwitchBookmarksIE(TwitchPlaylistBaseIE):
-    IE_NAME = 'twitch:bookmarks'
-    _VALID_URL = r'%s/(?P<id>[^/]+)/profile/bookmarks/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
-    _PLAYLIST_URL = '%s/api/bookmark/?user=%%s&offset=%%d&limit=%%d' % TwitchBaseIE._API_BASE
-    _PLAYLIST_TYPE = 'bookmarks'
-
-    _TEST = {
-        'url': 'http://www.twitch.tv/ognos/profile/bookmarks',
-        'info_dict': {
-            'id': 'ognos',
-            'title': 'Ognos',
-        },
-        'playlist_mincount': 3,
-    }
-
-    def _extract_playlist_page(self, response):
-        entries = []
-        for bookmark in response.get('bookmarks', []):
-            video = bookmark.get('video')
-            if not video:
-                continue
-            entries.append(video['url'])
-        return entries
-
-
 class TwitchStreamIE(TwitchBaseIE):
     IE_NAME = 'twitch:stream'
     _VALID_URL = r'%s/(?P<id>[^/#?]+)/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
@@ -476,3 +455,59 @@ class TwitchStreamIE(TwitchBaseIE):
             'formats': formats,
             'is_live': True,
         }
+
+
+class TwitchClipsIE(InfoExtractor):
+    IE_NAME = 'twitch:clips'
+    _VALID_URL = r'https?://clips\.twitch\.tv/(?:[^/]+/)*(?P<id>[^/?#&]+)'
+
+    _TESTS = [{
+        'url': 'https://clips.twitch.tv/ea/AggressiveCobraPoooound',
+        'md5': '761769e1eafce0ffebfb4089cb3847cd',
+        'info_dict': {
+            'id': 'AggressiveCobraPoooound',
+            'ext': 'mp4',
+            'title': 'EA Play 2016 Live from the Novo Theatre',
+            'thumbnail': 're:^https?://.*\.jpg',
+            'creator': 'EA',
+            'uploader': 'stereotype_',
+            'uploader_id': 'stereotype_',
+        },
+    }, {
+        # multiple formats
+        'url': 'https://clips.twitch.tv/rflegendary/UninterestedBeeDAESuppy',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, video_id)
+
+        clip = self._parse_json(
+            self._search_regex(
+                r'(?s)clipInfo\s*=\s*({.+?});', webpage, 'clip info'),
+            video_id, transform_source=js_to_json)
+
+        title = clip.get('channel_title') or self._og_search_title(webpage)
+
+        formats = [{
+            'url': option['source'],
+            'format_id': option.get('quality'),
+            'height': int_or_none(option.get('quality')),
+        } for option in clip.get('quality_options', []) if option.get('source')]
+
+        if not formats:
+            formats = [{
+                'url': clip['clip_video_url'],
+            }]
+
+        return {
+            'id': video_id,
+            'title': title,
+            'thumbnail': self._og_search_thumbnail(webpage),
+            'creator': clip.get('broadcaster_display_name') or clip.get('broadcaster_login'),
+            'uploader': clip.get('curator_login'),
+            'uploader_id': clip.get('curator_display_name'),
+            'formats': formats,
+        }