Prepare to release.

[youtubedl] / youtube_dl / extractor / twitch.py

diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py
index bf57eac01f2f42dd5dd4d4fc7bdce55f08ee7513..e01f11331007d072e6e78272d97fedce599ede4a 100644 (file)
--- a/youtube_dl/extractor/twitch.py
+++ b/youtube_dl/extractor/twitch.py
@@ -8,6 +8,7 @@ import random
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,
 from .common import InfoExtractor
 from ..compat import (
     compat_HTTPError,

+    compat_kwargs,

     compat_parse_qs,
     compat_str,
     compat_urllib_parse_urlencode,
     compat_parse_qs,
     compat_str,
     compat_urllib_parse_urlencode,


@@ -16,11 +17,14 @@ from ..compat import (
 from ..utils import (
     clean_html,
     ExtractorError,
 from ..utils import (
     clean_html,
     ExtractorError,

+    float_or_none,

     int_or_none,
     int_or_none,

-    js_to_json,

     orderedSet,
     parse_duration,
     parse_iso8601,
     orderedSet,
     parse_duration,
     parse_iso8601,

+    qualities,
+    try_get,
+    unified_timestamp,

     update_url_query,
     urlencode_postdata,
     urljoin,
     update_url_query,
     urlencode_postdata,
     urljoin,


@@ -28,7 +32,7 @@ from ..utils import (
 
 
 class TwitchBaseIE(InfoExtractor):
 
 
 class TwitchBaseIE(InfoExtractor):

-    _VALID_URL_BASE = r'https?://(?:(?:www|go)\.)?twitch\.tv'
+    _VALID_URL_BASE = r'https?://(?:(?:www|go|m)\.)?twitch\.tv'

 
     _API_BASE = 'https://api.twitch.tv'
     _USHER_BASE = 'https://usher.ttvnw.net'
 
     _API_BASE = 'https://api.twitch.tv'
     _USHER_BASE = 'https://usher.ttvnw.net'


@@ -45,10 +49,11 @@ class TwitchBaseIE(InfoExtractor):
                 '%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')),
                 expected=True)
 
                 '%s returned error: %s - %s' % (self.IE_NAME, error, response.get('message')),
                 expected=True)
 

-    def _call_api(self, path, item_id, note):
+    def _call_api(self, path, item_id, *args, **kwargs):
+        kwargs.setdefault('headers', {})['Client-ID'] = self._CLIENT_ID

         response = self._download_json(
         response = self._download_json(

-            '%s/%s' % (self._API_BASE, path), item_id, note,
-            headers={'Client-ID': self._CLIENT_ID})
+            '%s/%s' % (self._API_BASE, path), item_id,
+            *args, **compat_kwargs(kwargs))

         self._handle_error(response)
         return response
 
         self._handle_error(response)
         return response
 


@@ -56,7 +61,7 @@ class TwitchBaseIE(InfoExtractor):
         self._login()
 
     def _login(self):
         self._login()
 
     def _login(self):

-        (username, password) = self._get_login_info()
+        username, password = self._get_login_info()

         if username is None:
             return
 
         if username is None:
             return
 


@@ -85,10 +90,15 @@ class TwitchBaseIE(InfoExtractor):
                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
                     response = self._parse_json(
                         e.cause.read().decode('utf-8'), None)
                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
                     response = self._parse_json(
                         e.cause.read().decode('utf-8'), None)

-                    fail(response['message'])
+                    fail(response.get('message') or response['errors'][0])

                 raise
 
                 raise
 

-            redirect_url = urljoin(post_url, response['redirect'])
+            if 'Authenticated successfully' in response.get('message', ''):
+                return None, None
+
+            redirect_url = urljoin(
+                post_url,
+                response.get('redirect') or response['redirect_path'])

             return self._download_webpage_handle(
                 redirect_url, None, 'Downloading login redirect page',
                 headers=headers)
             return self._download_webpage_handle(
                 redirect_url, None, 'Downloading login redirect page',
                 headers=headers)


@@ -106,6 +116,10 @@ class TwitchBaseIE(InfoExtractor):
                 'password': password,
             })
 
                 'password': password,
             })
 

+        # Successful login
+        if not redirect_page:
+            return
+

         if re.search(r'(?i)<form[^>]+id="two-factor-submit"', redirect_page) is not None:
             # TODO: Add mechanism to request an SMS or phone call
             tfa_token = self._get_tfa_info('two-factor authentication token')
         if re.search(r'(?i)<form[^>]+id="two-factor-submit"', redirect_page) is not None:
             # TODO: Add mechanism to request an SMS or phone call
             tfa_token = self._get_tfa_info('two-factor authentication token')


@@ -159,6 +173,13 @@ class TwitchItemBaseIE(TwitchBaseIE):
         return self.playlist_result(entries, info['id'], info['title'])
 
     def _extract_info(self, info):
         return self.playlist_result(entries, info['id'], info['title'])
 
     def _extract_info(self, info):

+        status = info.get('status')
+        if status == 'recording':
+            is_live = True
+        elif status == 'recorded':
+            is_live = False
+        else:
+            is_live = None

         return {
             'id': info['_id'],
             'title': info.get('title') or 'Untitled Broadcast',
         return {
             'id': info['_id'],
             'title': info.get('title') or 'Untitled Broadcast',


@@ -169,6 +190,7 @@ class TwitchItemBaseIE(TwitchBaseIE):
             'uploader_id': info.get('channel', {}).get('name'),
             'timestamp': parse_iso8601(info.get('recorded_at')),
             'view_count': int_or_none(info.get('views')),
             'uploader_id': info.get('channel', {}).get('name'),
             'timestamp': parse_iso8601(info.get('recorded_at')),
             'view_count': int_or_none(info.get('views')),

+            'is_live': is_live,

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


@@ -217,7 +239,7 @@ class TwitchVodIE(TwitchItemBaseIE):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
     _VALID_URL = r'''(?x)
                     https?://
                         (?:

-                            (?:(?:www|go)\.)?twitch\.tv/(?:[^/]+/v|videos)/|
+                            (?:(?:www|go|m)\.)?twitch\.tv/(?:[^/]+/v|videos)/|

                             player\.twitch\.tv/\?.*?\bvideo=v
                         )
                         (?P<id>\d+)
                             player\.twitch\.tv/\?.*?\bvideo=v
                         )
                         (?P<id>\d+)


@@ -270,6 +292,9 @@ class TwitchVodIE(TwitchItemBaseIE):
     }, {
         'url': 'https://www.twitch.tv/videos/6528877',
         'only_matching': True,
     }, {
         'url': 'https://www.twitch.tv/videos/6528877',
         'only_matching': True,

+    }, {
+        'url': 'https://m.twitch.tv/beagsandjam/v/247478721',
+        'only_matching': True,

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


@@ -358,9 +383,16 @@ class TwitchPlaylistBaseIE(TwitchBaseIE):
                 break
             offset += limit
         return self.playlist_result(
                 break
             offset += limit
         return self.playlist_result(

-            [self.url_result(entry) for entry in orderedSet(entries)],
+            [self._make_url_result(entry) for entry in orderedSet(entries)],

             channel_id, channel_name)
 
             channel_id, channel_name)
 

+    def _make_url_result(self, url):
+        try:
+            video_id = 'v%s' % TwitchVodIE._match_id(url)
+            return self.url_result(url, TwitchVodIE.ie_key(), video_id=video_id)
+        except AssertionError:
+            return self.url_result(url)
+

     def _extract_playlist_page(self, response):
         videos = response.get('videos')
         return [video['url'] for video in videos] if videos else []
     def _extract_playlist_page(self, response):
         videos = response.get('videos')
         return [video['url'] for video in videos] if videos else []


@@ -374,14 +406,17 @@ class TwitchProfileIE(TwitchPlaylistBaseIE):
     _VALID_URL = r'%s/(?P<id>[^/]+)/profile/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
     _PLAYLIST_TYPE = 'profile'
 
     _VALID_URL = r'%s/(?P<id>[^/]+)/profile/?(?:\#.*)?$' % TwitchBaseIE._VALID_URL_BASE
     _PLAYLIST_TYPE = 'profile'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'http://www.twitch.tv/vanillatv/profile',
         'info_dict': {
             'id': 'vanillatv',
             'title': 'VanillaTV',
         },
         'playlist_mincount': 412,
         'url': 'http://www.twitch.tv/vanillatv/profile',
         'info_dict': {
             'id': 'vanillatv',
             'title': 'VanillaTV',
         },
         'playlist_mincount': 412,

-    }
+    }, {
+        'url': 'http://m.twitch.tv/vanillatv/profile',
+        'only_matching': True,
+    }]

 
 
 class TwitchVideosBaseIE(TwitchPlaylistBaseIE):
 
 
 class TwitchVideosBaseIE(TwitchPlaylistBaseIE):


@@ -395,14 +430,17 @@ class TwitchAllVideosIE(TwitchVideosBaseIE):
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive,upload,highlight'
     _PLAYLIST_TYPE = 'all videos'
 
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive,upload,highlight'
     _PLAYLIST_TYPE = 'all videos'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'https://www.twitch.tv/spamfish/videos/all',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 869,
         'url': 'https://www.twitch.tv/spamfish/videos/all',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 869,

-    }
+    }, {
+        'url': 'https://m.twitch.tv/spamfish/videos/all',
+        'only_matching': True,
+    }]

 
 
 class TwitchUploadsIE(TwitchVideosBaseIE):
 
 
 class TwitchUploadsIE(TwitchVideosBaseIE):


@@ -411,14 +449,17 @@ class TwitchUploadsIE(TwitchVideosBaseIE):
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'upload'
     _PLAYLIST_TYPE = 'uploads'
 
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'upload'
     _PLAYLIST_TYPE = 'uploads'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'https://www.twitch.tv/spamfish/videos/uploads',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 0,
         'url': 'https://www.twitch.tv/spamfish/videos/uploads',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 0,

-    }
+    }, {
+        'url': 'https://m.twitch.tv/spamfish/videos/uploads',
+        'only_matching': True,
+    }]

 
 
 class TwitchPastBroadcastsIE(TwitchVideosBaseIE):
 
 
 class TwitchPastBroadcastsIE(TwitchVideosBaseIE):


@@ -427,14 +468,17 @@ class TwitchPastBroadcastsIE(TwitchVideosBaseIE):
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive'
     _PLAYLIST_TYPE = 'past broadcasts'
 
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'archive'
     _PLAYLIST_TYPE = 'past broadcasts'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'https://www.twitch.tv/spamfish/videos/past-broadcasts',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 0,
         'url': 'https://www.twitch.tv/spamfish/videos/past-broadcasts',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 0,

-    }
+    }, {
+        'url': 'https://m.twitch.tv/spamfish/videos/past-broadcasts',
+        'only_matching': True,
+    }]

 
 
 class TwitchHighlightsIE(TwitchVideosBaseIE):
 
 
 class TwitchHighlightsIE(TwitchVideosBaseIE):


@@ -443,14 +487,17 @@ class TwitchHighlightsIE(TwitchVideosBaseIE):
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'highlight'
     _PLAYLIST_TYPE = 'highlights'
 
     _PLAYLIST_PATH = TwitchVideosBaseIE._PLAYLIST_PATH + 'highlight'
     _PLAYLIST_TYPE = 'highlights'
 

-    _TEST = {
+    _TESTS = [{

         'url': 'https://www.twitch.tv/spamfish/videos/highlights',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 805,
         'url': 'https://www.twitch.tv/spamfish/videos/highlights',
         'info_dict': {
             'id': 'spamfish',
             'title': 'Spamfish',
         },
         'playlist_mincount': 805,

-    }
+    }, {
+        'url': 'https://m.twitch.tv/spamfish/videos/highlights',
+        'only_matching': True,
+    }]

 
 
 class TwitchStreamIE(TwitchBaseIE):
 
 
 class TwitchStreamIE(TwitchBaseIE):


@@ -458,7 +505,7 @@ class TwitchStreamIE(TwitchBaseIE):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
     _VALID_URL = r'''(?x)
                     https?://
                         (?:

-                            (?:(?:www|go)\.)?twitch\.tv/|
+                            (?:(?:www|go|m)\.)?twitch\.tv/|

                             player\.twitch\.tv/\?.*?\bchannel=
                         )
                         (?P<id>[^/#?]+)
                             player\.twitch\.tv/\?.*?\bchannel=
                         )
                         (?P<id>[^/#?]+)


@@ -492,6 +539,9 @@ class TwitchStreamIE(TwitchBaseIE):
     }, {
         'url': 'https://go.twitch.tv/food',
         'only_matching': True,
     }, {
         'url': 'https://go.twitch.tv/food',
         'only_matching': True,

+    }, {
+        'url': 'https://m.twitch.tv/food',
+        'only_matching': True,

     }]
 
     @classmethod
     }]
 
     @classmethod


@@ -577,21 +627,23 @@ class TwitchStreamIE(TwitchBaseIE):
         }
 
 
         }
 
 

-class TwitchClipsIE(InfoExtractor):
+class TwitchClipsIE(TwitchBaseIE):

     IE_NAME = 'twitch:clips'
     _VALID_URL = r'https?://clips\.twitch\.tv/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
     _TESTS = [{
     IE_NAME = 'twitch:clips'
     _VALID_URL = r'https?://clips\.twitch\.tv/(?:[^/]+/)*(?P<id>[^/?#&]+)'
 
     _TESTS = [{

-        'url': 'https://clips.twitch.tv/ea/AggressiveCobraPoooound',
+        'url': 'https://clips.twitch.tv/FaintLightGullWholeWheat',

         'md5': '761769e1eafce0ffebfb4089cb3847cd',
         'info_dict': {
         'md5': '761769e1eafce0ffebfb4089cb3847cd',
         'info_dict': {

-            'id': 'AggressiveCobraPoooound',
+            'id': '42850523',

             'ext': 'mp4',
             'title': 'EA Play 2016 Live from the Novo Theatre',
             'thumbnail': r're:^https?://.*\.jpg',
             'ext': 'mp4',
             'title': 'EA Play 2016 Live from the Novo Theatre',
             'thumbnail': r're:^https?://.*\.jpg',

+            'timestamp': 1465767393,
+            'upload_date': '20160612',

             'creator': 'EA',
             'uploader': 'stereotype_',
             'creator': 'EA',
             'uploader': 'stereotype_',

-            'uploader_id': 'stereotype_',
+            'uploader_id': '43566419',

         },
     }, {
         # multiple formats
         },
     }, {
         # multiple formats


@@ -602,34 +654,63 @@ class TwitchClipsIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
     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)
+        status = self._download_json(
+            'https://clips.twitch.tv/api/v2/clips/%s/status' % video_id,
+            video_id)

 
 

-        title = clip.get('title') or clip.get('channel_title') or self._og_search_title(webpage)
+        formats = []

 
 

-        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'],
-            }]
+        for option in status['quality_options']:
+            if not isinstance(option, dict):
+                continue
+            source = option.get('source')
+            if not source or not isinstance(source, compat_str):
+                continue
+            formats.append({
+                'url': source,
+                'format_id': option.get('quality'),
+                'height': int_or_none(option.get('quality')),
+                'fps': int_or_none(option.get('frame_rate')),
+            })

 
         self._sort_formats(formats)
 
 
         self._sort_formats(formats)
 

-        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'),
+        info = {

             'formats': formats,
         }
             'formats': formats,
         }

+
+        clip = self._call_api(
+            'kraken/clips/%s' % video_id, video_id, fatal=False, headers={
+                'Accept': 'application/vnd.twitchtv.v5+json',
+            })
+
+        if clip:
+            quality_key = qualities(('tiny', 'small', 'medium'))
+            thumbnails = []
+            thumbnails_dict = clip.get('thumbnails')
+            if isinstance(thumbnails_dict, dict):
+                for thumbnail_id, thumbnail_url in thumbnails_dict.items():
+                    thumbnails.append({
+                        'id': thumbnail_id,
+                        'url': thumbnail_url,
+                        'preference': quality_key(thumbnail_id),
+                    })
+
+            info.update({
+                'id': clip.get('tracking_id') or video_id,
+                'title': clip.get('title') or video_id,
+                'duration': float_or_none(clip.get('duration')),
+                'views': int_or_none(clip.get('views')),
+                'timestamp': unified_timestamp(clip.get('created_at')),
+                'thumbnails': thumbnails,
+                'creator': try_get(clip, lambda x: x['broadcaster']['display_name'], compat_str),
+                'uploader': try_get(clip, lambda x: x['curator']['display_name'], compat_str),
+                'uploader_id': try_get(clip, lambda x: x['curator']['id'], compat_str),
+            })
+        else:
+            info.update({
+                'title': video_id,
+                'id': video_id,
+            })
+
+        return info