Merge branch 'moviefap' of https://github.com/gebn/youtube-dl into gebn-moviefap

diff --git a/youtube_dl/extractor/__init__.py b/youtube_dl/extractor/__init__.py
index b70c1b57ed27e3ea79a764ce544dd38d443664cb..d41d277c91b5bd712d788793a33be1f0a1e5f395 100644 (file)
--- a/youtube_dl/extractor/__init__.py
+++ b/youtube_dl/extractor/__init__.py
@@ -622,6 +622,7 @@ from .twitch import (
     TwitchBookmarksIE,
     TwitchStreamIE,
 )
+from .twitter import TwitterCardIE
 from .ubu import UbuIE
 from .udemy import (
     UdemyIE,

diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py
index 94bd6345da18815a50b72502a8b91ae4e30ae2b5..b56ee2959cd92c3bb26ca57c471c049236ea63ca 100644 (file)
--- a/youtube_dl/extractor/twitch.py
+++ b/youtube_dl/extractor/twitch.py
@@ -189,17 +189,17 @@ class TwitchVodIE(TwitchItemBaseIE):
     _ITEM_SHORTCUT = 'v'
 
     _TEST = {
-        'url': 'http://www.twitch.tv/ksptv/v/3622000',
+        'url': 'http://www.twitch.tv/riotgames/v/6528877',
         'info_dict': {
-            'id': 'v3622000',
+            'id': 'v6528877',
             'ext': 'mp4',
-            'title': '''KSPTV: Squadcast: "Everyone's on vacation so here's Dahud" Edition!''',
+            'title': 'LCK Summer Split - Week 6 Day 1',
             'thumbnail': 're:^https?://.*\.jpg$',
-            'duration': 6951,
-            'timestamp': 1419028564,
-            'upload_date': '20141219',
-            'uploader': 'KSPTV',
-            'uploader_id': 'ksptv',
+            'duration': 17208,
+            'timestamp': 1435131709,
+            'upload_date': '20150624',
+            'uploader': 'Riot Games',
+            'uploader_id': 'riotgames',
             'view_count': int,
         },
         'params': {
@@ -215,7 +215,7 @@ class TwitchVodIE(TwitchItemBaseIE):
             '%s/api/vods/%s/access_token' % (self._API_BASE, item_id), item_id,
             'Downloading %s access token' % self._ITEM_TYPE)
         formats = self._extract_m3u8_formats(
-            '%s/vod/%s?nauth=%s&nauthsig=%s'
+            '%s/vod/%s?nauth=%s&nauthsig=%s&allow_source=true'
             % (self._USHER_BASE, item_id, access_token['token'], access_token['sig']),
             item_id, 'mp4')
         self._prefer_source(formats)

diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py
new file mode 100644 (file)
index 0000000..1aaa063
--- /dev/null
+++ b/youtube_dl/extractor/twitter.py
@@ -0,0 +1,72 @@
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..compat import compat_urllib_request
+from ..utils import (
+    float_or_none,
+    unescapeHTML,
+)
+
+
+class TwitterCardIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?twitter\.com/i/cards/tfw/v1/(?P<id>\d+)'
+    _TEST = {
+        'url': 'https://twitter.com/i/cards/tfw/v1/560070183650213889',
+        'md5': 'a74f50b310c83170319ba16de6955192',
+        'info_dict': {
+            'id': '560070183650213889',
+            'ext': 'mp4',
+            'title': 'TwitterCard',
+            'thumbnail': 're:^https?://.*\.jpg$',
+            'duration': 30.033,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+
+        # Different formats served for different User-Agents
+        USER_AGENTS = [
+            'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/20.0 (Chrome)',  # mp4
+            'Mozilla/5.0 (Windows NT 5.2; WOW64; rv:38.0) Gecko/20100101 Firefox/38.0',  # webm
+        ]
+
+        config = None
+        formats = []
+        for user_agent in USER_AGENTS:
+            request = compat_urllib_request.Request(url)
+            request.add_header('User-Agent', user_agent)
+            webpage = self._download_webpage(request, video_id)
+
+            config = self._parse_json(
+                unescapeHTML(self._search_regex(
+                    r'data-player-config="([^"]+)"', webpage, 'data player config')),
+                video_id)
+
+            video_url = config['playlist'][0]['source']
+
+            f = {
+                'url': video_url,
+            }
+
+            m = re.search(r'/(?P<width>\d+)x(?P<height>\d+)/', video_url)
+            if m:
+                f.update({
+                    'width': int(m.group('width')),
+                    'height': int(m.group('height')),
+                })
+            formats.append(f)
+        self._sort_formats(formats)
+
+        thumbnail = config.get('posterImageUrl')
+        duration = float_or_none(config.get('duration'))
+
+        return {
+            'id': video_id,
+            'title': 'TwitterCard',
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'formats': formats,
+        }