youtube-dl

commit c4d9e6731a80cc1ce173d444920d49b424c62887
parent 0d9ec5d9630afa298f8b9acdfae0d85fbaa3fe58
Author: Jaime Marquínez Ferrándiz <jaime.marquinez.ferrandiz@gmail.com>
Date:   Sat,  7 Dec 2013 11:19:59 +0100

[pyvideo] add support for videos that don't come from Youtube

Diffstat:
M
youtube_dl/extractor/pyvideo.py
 | 
30
++++++++++++++++++++++++++----

1 file changed, 26 insertions(+), 4 deletions(-)
diff --git a/youtube_dl/extractor/pyvideo.py b/youtube_dl/extractor/pyvideo.py
@@ -1,11 +1,12 @@
 import re
+import os
 
 from .common import InfoExtractor
 
 
 class PyvideoIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?pyvideo\.org/video/(\d+)/(.*)'
-    _TEST = {
+    _VALID_URL = r'(?:http://)?(?:www\.)?pyvideo\.org/video/(?P<id>\d+)/(.*)'
+    _TESTS = [{
         u'url': u'http://pyvideo.org/video/1737/become-a-logging-expert-in-30-minutes',
         u'file': u'24_4WWkSmNo.mp4',
         u'md5': u'de317418c8bc76b1fd8633e4f32acbc6',
@@ -17,13 +18,34 @@ class PyvideoIE(InfoExtractor):
             u"uploader_id": u"NextDayVideo",
         },
         u'add_ie': ['Youtube'],
-    }
+    },
+    {
+        u'url': u'http://pyvideo.org/video/2542/gloriajw-spotifywitherikbernhardsson182m4v',
+        u'md5': u'5fe1c7e0a8aa5570330784c847ff6d12',
+        u'info_dict': {
+            u'id': u'2542',
+            u'ext': u'm4v',
+            u'title': u'Gloriajw-SpotifyWithErikBernhardsson182',
+        },
+    },
+    ]
 
     def _real_extract(self, url):
         mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group(2)
+        video_id = mobj.group('id')
         webpage = self._download_webpage(url, video_id)
         m_youtube = re.search(r'(https?://www\.youtube\.com/watch\?v=.*)', webpage)
 
         if m_youtube is not None:
             return self.url_result(m_youtube.group(1), 'Youtube')
+
+        title = self._html_search_regex(r'<div class="section">.*?<h3>([^>]+?)</h3>',
+            webpage, u'title', flags=re.DOTALL)
+        video_url = self._search_regex([r'<source src="(.*?)"',
+            r'<dt>Download</dt>.*?<a href="(.+?)"'],
+            webpage, u'video url', flags=re.DOTALL)
+        return {
+            'id': video_id,
+            'title': os.path.splitext(title)[0],
+            'url': video_url,
+        }