Merge pull request #8092 from bpfoley/twitter-thumbnail

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

diff --git a/youtube_dl/extractor/pyvideo.py b/youtube_dl/extractor/pyvideo.py
index e28ca3fb62e9d60e3f7816f71f6efdea1bd21e04..30a5f2de4475a934cfa467764d0ce559d3e68a74 100644 (file)
--- a/youtube_dl/extractor/pyvideo.py
+++ b/youtube_dl/extractor/pyvideo.py
@@ -1,29 +1,59 @@
+from __future__ import unicode_literals
+
 import re
+import os
 
 from .common import InfoExtractor
 
 
 class PyvideoIE(InfoExtractor):
-    _VALID_URL = r'(?:http://)?(?:www\.)?pyvideo\.org/video/(\d+)/(.*)'
-    _TEST = {
-        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',
-        u'info_dict': {
-            u"title": u"Become a logging expert in 30 minutes",
-            u"description": u"md5:9665350d466c67fb5b1598de379021f7",
-            u"upload_date": u"20130320",
-            u"uploader": u"NextDayVideo",
-            u"uploader_id": u"NextDayVideo",
+    _VALID_URL = r'http://(?:www\.)?pyvideo\.org/video/(?P<id>\d+)/(.*)'
+
+    _TESTS = [
+        {
+            'url': 'http://pyvideo.org/video/1737/become-a-logging-expert-in-30-minutes',
+            'md5': '520915673e53a5c5d487c36e0c4d85b5',
+            'info_dict': {
+                'id': '24_4WWkSmNo',
+                'ext': 'webm',
+                'title': 'Become a logging expert in 30 minutes',
+                'description': 'md5:9665350d466c67fb5b1598de379021f7',
+                'upload_date': '20130320',
+                'uploader': 'Next Day Video',
+                'uploader_id': 'NextDayVideo',
+            },
+            'add_ie': ['Youtube'],
         },
-        u'add_ie': ['Youtube'],
-    }
+        {
+            'url': 'http://pyvideo.org/video/2542/gloriajw-spotifywitherikbernhardsson182m4v',
+            'md5': '5fe1c7e0a8aa5570330784c847ff6d12',
+            'info_dict': {
+                'id': '2542',
+                'ext': 'm4v',
+                'title': '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)
 
+        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">\s*<h3(?:\s+class="[^"]*"[^>]*)?>([^>]+?)</h3>',
+            webpage, 'title', flags=re.DOTALL)
+        video_url = self._search_regex(
+            [r'<source src="(.*?)"', r'<dt>Download</dt>.*?<a href="(.+?)"'],
+            webpage, 'video url', flags=re.DOTALL)
+
+        return {
+            'id': video_id,
+            'title': os.path.splitext(title)[0],
+            'url': video_url,
+        }