Add support for Howcast.com - closes #835

[youtube-dl] / youtube_dl / InfoExtractors.py

diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 6d6203a1cd24c4aac90dbf52da95ced8a47a740d..938d2d805d2e3ad8ca881677a8a2c18709ce3201 100755 (executable)
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -3301,18 +3301,26 @@ class UstreamIE(InfoExtractor):
         video_id = m.group('videoID')
         video_url = u'http://tcdn.ustream.tv/video/%s' % video_id
         webpage = self._download_webpage(url, video_id)
-        m = re.search(r'data-title="(?P<title>.+)"',webpage)
-        title = m.group('title')
-        m = re.search(r'<a class="state" data-content-type="channel" data-content-id="(?P<uploader>\d+)"',webpage)
-        uploader = m.group('uploader')
+        self.report_extraction(video_id)
+        try:
+            m = re.search(r'data-title="(?P<title>.+)"',webpage)
+            title = m.group('title')
+            m = re.search(r'data-content-type="channel".*?>(?P<uploader>.*?)</a>',
+                          webpage, re.DOTALL)
+            uploader = unescapeHTML(m.group('uploader').strip())
+            m = re.search(r'<link rel="image_src" href="(?P<thumb>.*?)"', webpage)
+            thumb = m.group('thumb')
+        except AttributeError:
+            raise ExtractorError(u'Unable to extract info')
         info = {
                 'id':video_id,
                 'url':video_url,
                 'ext': 'flv',
                 'title': title,
-                'uploader': uploader
+                'uploader': uploader,
+                'thumbnail': thumb,
                   }
-        return [info]
+        return info
 
 class WorldStarHipHopIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www|m)\.worldstar(?:candy|hiphop)\.com/videos/video\.php\?v=(?P<id>.*)'
@@ -4033,7 +4041,7 @@ class RedTubeIE(InfoExtractor):
         
 class InaIE(InfoExtractor):
     """Information Extractor for Ina.fr"""
-    _VALID_URL = r'(?:http://)?(?:www.)?ina\.fr/video/(?P<id>I[0-9]+)/.*'
+    _VALID_URL = r'(?:http://)?(?:www\.)?ina\.fr/video/(?P<id>I[0-9]+)/.*'
 
     def _real_extract(self,url):
         mobj = re.match(self._VALID_URL, url)
@@ -4060,6 +4068,42 @@ class InaIE(InfoExtractor):
             'title':    video_title,
         }]
 
+class HowcastIE(InfoExtractor):
+    """Information Extractor for Ina.fr"""
+    _VALID_URL = r'(?:https?://)?(?:www\.)?howcast\.com/videos/(?P<id>[\d]+)'
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+
+        video_id = mobj.group('id')
+        webpage_url = 'http://www.howcast.com/videos/' + video_id
+        webpage = self._download_webpage(webpage_url, video_id)
+
+        mobj = re.search(r'\'file\': "(http://mobile-media\.howcast\.com/\d+\.mp4)"', webpage)
+        if mobj is None:
+            raise ExtractorError(u'Unable to extract video URL')
+        video_url = mobj.group(1)
+
+        mobj = re.search(r'<meta content=(?:"([^"]+)"|\'([^\']+)\') property=\'og:title\'', webpage)
+        if mobj is None:
+            raise ExtractorError(u'Unable to extract title')
+        video_title = mobj.group(1) or mobj.group(2)
+
+        mobj = re.search(r'<meta content=(?:"([^"]+)"|\'([^\']+)\') name=\'description\'', webpage)
+        if mobj is None:
+            self._downloader.report_warning(u'unable to extract description')
+            video_description = None
+        else:
+            video_description = mobj.group(1) or mobj.group(2)
+
+        return [{
+            'id':       video_id,
+            'url':      video_url,
+            'ext':      'mp4',
+            'title':    video_title,
+            'description': video_description,
+        }]
+
 def gen_extractors():
     """ Return a list of an instance of every supported extractor.
     The order does matter; the first extractor matched is the one handling the URL.
@@ -4117,6 +4161,7 @@ def gen_extractors():
         BandcampIE(),
         RedTubeIE(),
         InaIE(),
+        HowcastIE(),
         GenericIE()
     ]