Merge pull request #7045 from remitamine/ign

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

diff --git a/youtube_dl/extractor/ustream.py b/youtube_dl/extractor/ustream.py
index a395dc5dac85d0dabe5aea59e25275942f7e28a8..73b05ecab82a10a6c80360b0b980f285dfbb9c45 100644 (file)
--- a/youtube_dl/extractor/ustream.py
+++ b/youtube_dl/extractor/ustream.py
@@ -1,40 +1,59 @@
 from __future__ import unicode_literals
 
-import json
 import re
 
 from .common import InfoExtractor
 from ..compat import (
     compat_urlparse,
 )
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    float_or_none,
+)
 
 
 class UstreamIE(InfoExtractor):
-    _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<videoID>\d+)'
+    _VALID_URL = r'https?://www\.ustream\.tv/(?P<type>recorded|embed|embed/recorded)/(?P<id>\d+)'
     IE_NAME = 'ustream'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://www.ustream.tv/recorded/20274954',
         'md5': '088f151799e8f572f84eb62f17d73e5c',
         'info_dict': {
             'id': '20274954',
             'ext': 'flv',
-            'uploader': 'Young Americans for Liberty',
             'title': 'Young Americans for Liberty February 7, 2012 2:28 AM',
+            'description': 'Young Americans for Liberty February 7, 2012 2:28 AM',
+            'timestamp': 1328577035,
+            'upload_date': '20120207',
+            'uploader': 'yaliberty',
+            'uploader_id': '6780869',
         },
-    }
+    }, {
+        # From http://sportscanada.tv/canadagames/index.php/week2/figure-skating/444
+        # Title and uploader available only from params JSON
+        'url': 'http://www.ustream.tv/embed/recorded/59307601?ub=ff0000&lc=ff0000&oc=ffffff&uc=ffffff&v=3&wmode=direct',
+        'md5': '5a2abf40babeac9812ed20ae12d34e10',
+        'info_dict': {
+            'id': '59307601',
+            'ext': 'flv',
+            'title': '-CG11- Canada Games Figure Skating',
+            'uploader': 'sportscanadatv',
+        },
+        'skip': 'This Pro Broadcaster has chosen to remove this video from the ustream.tv site.',
+    }]
 
     def _real_extract(self, url):
         m = re.match(self._VALID_URL, url)
-        video_id = m.group('videoID')
+        video_id = m.group('id')
 
         # some sites use this embed format (see: http://github.com/rg3/youtube-dl/issues/2990)
         if m.group('type') == 'embed/recorded':
-            video_id = m.group('videoID')
+            video_id = m.group('id')
             desktop_url = 'http://www.ustream.tv/recorded/' + video_id
             return self.url_result(desktop_url, 'Ustream')
         if m.group('type') == 'embed':
-            video_id = m.group('videoID')
+            video_id = m.group('id')
             webpage = self._download_webpage(url, video_id)
             desktop_video_id = self._html_search_regex(
                 r'ContentVideoIds=\["([^"]*?)"\]', webpage, 'desktop_video_id')
@@ -42,46 +61,50 @@ class UstreamIE(InfoExtractor):
             return self.url_result(desktop_url, 'Ustream')
 
         params = self._download_json(
-            'http://cdngw.ustream.tv/rgwjson/Viewer.getVideo/' + json.dumps({
-                'brandId': 1,
-                'videoId': int(video_id),
-                'autoplay': False,
-            }), video_id)
+            'https://api.ustream.tv/videos/%s.json' % video_id, video_id)
 
-        if 'error' in params:
-            raise ExtractorError(params['error']['message'], expected=True)
+        error = params.get('error')
+        if error:
+            raise ExtractorError(
+                '%s returned error: %s' % (self.IE_NAME, error), expected=True)
 
-        video_url = params['flv']
+        video = params['video']
 
-        webpage = self._download_webpage(url, video_id)
+        title = video['title']
+        filesize = float_or_none(video.get('file_size'))
 
-        self.report_extraction(video_id)
-
-        video_title = self._html_search_regex(r'data-title="(?P<title>.+)"',
-                                              webpage, 'title', default=None)
-
-        if not video_title:
-            try:
-                video_title = params['moduleConfig']['meta']['title']
-            except KeyError:
-                pass
+        formats = [{
+            'id': video_id,
+            'url': video_url,
+            'ext': format_id,
+            'filesize': filesize,
+        } for format_id, video_url in video['media_urls'].items()]
+        self._sort_formats(formats)
 
-        if not video_title:
-            video_title = 'Ustream video ' + video_id
+        description = video.get('description')
+        timestamp = int_or_none(video.get('created_at'))
+        duration = float_or_none(video.get('length'))
+        view_count = int_or_none(video.get('views'))
 
-        uploader = self._html_search_regex(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>',
-                                           webpage, 'uploader', fatal=False, flags=re.DOTALL)
+        uploader = video.get('owner', {}).get('username')
+        uploader_id = video.get('owner', {}).get('id')
 
-        thumbnail = self._html_search_regex(r'<link rel="image_src" href="(?P<thumb>.*?)"',
-                                            webpage, 'thumbnail', fatal=False)
+        thumbnails = [{
+            'id': thumbnail_id,
+            'url': thumbnail_url,
+        } for thumbnail_id, thumbnail_url in video.get('thumbnail', {}).items()]
 
         return {
             'id': video_id,
-            'url': video_url,
-            'ext': 'flv',
-            'title': video_title,
+            'title': title,
+            'description': description,
+            'thumbnails': thumbnails,
+            'timestamp': timestamp,
+            'duration': duration,
+            'view_count': view_count,
             'uploader': uploader,
-            'thumbnail': thumbnail,
+            'uploader_id': uploader_id,
+            'formats': formats,
         }