Merge branch 'master' of https://github.com/rg3/youtube-dl

[youtube-dl] / youtube_dl / InfoExtractors.py

diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py
index 1443c188f2e21671be59f87a84ba8f0aba7b0c0d..e380f62a1f02b6fec3575876db0931a1fa4a74cb 100755 (executable)
--- a/youtube_dl/InfoExtractors.py
+++ b/youtube_dl/InfoExtractors.py
@@ -3,6 +3,7 @@
 
 from __future__ import absolute_import
 
+import base64
 import datetime
 import netrc
 import os
@@ -105,18 +106,18 @@ class InfoExtractor(object):
     def IE_NAME(self):
         return type(self).__name__[:-2]
 
-    def _download_webpage(self, url, video_id, note=None, errnote=None):
+    def _download_webpage(self, url_or_request, video_id, note=None, errnote=None):
         if note is None:
             note = u'Downloading video webpage'
         self._downloader.to_screen(u'[%s] %s: %s' % (self.IE_NAME, video_id, note))
         try:
-            urlh = compat_urllib_request.urlopen(url)
+            urlh = compat_urllib_request.urlopen(url_or_request)
             webpage_bytes = urlh.read()
             return webpage_bytes.decode('utf-8', 'replace')
         except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
             if errnote is None:
                 errnote = u'Unable to download webpage'
-            raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)))
+            raise ExtractorError(u'%s: %s' % (errnote, compat_str(err)), sys.exc_info()[2])
 
 
 class YoutubeIE(InfoExtractor):
@@ -411,7 +412,7 @@ class YoutubeIE(InfoExtractor):
 
         # uploader_id
         video_uploader_id = None
-        mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/user/([^"]+)">', video_webpage)
+        mobj = re.search(r'<link itemprop="url" href="http://www.youtube.com/(?:user|channel)/([^"]+)">', video_webpage)
         if mobj is not None:
             video_uploader_id = mobj.group(1)
         else:
@@ -674,10 +675,6 @@ class DailymotionIE(InfoExtractor):
     def __init__(self, downloader=None):
         InfoExtractor.__init__(self, downloader)
 
-    def report_download_webpage(self, video_id):
-        """Report webpage download."""
-        self._downloader.to_screen(u'[dailymotion] %s: Downloading webpage' % video_id)
-
     def report_extraction(self, video_id):
         """Report information extraction."""
         self._downloader.to_screen(u'[dailymotion] %s: Extracting information' % video_id)
@@ -696,13 +693,7 @@ class DailymotionIE(InfoExtractor):
         # Retrieve video webpage to extract further information
         request = compat_urllib_request.Request(url)
         request.add_header('Cookie', 'family_filter=off')
-        try:
-            self.report_download_webpage(video_id)
-            webpage_bytes = compat_urllib_request.urlopen(request).read()
-            webpage = webpage_bytes.decode('utf-8')
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            self._downloader.trouble(u'ERROR: unable retrieve video webpage: %s' % compat_str(err))
-            return
+        webpage = self._download_webpage(request, video_id)
 
         # Extract URL, uploader and title from webpage
         self.report_extraction(video_id)
@@ -2820,13 +2811,7 @@ class SoundcloudIE(InfoExtractor):
 
 class InfoQIE(InfoExtractor):
     """Information extractor for infoq.com"""
-
     _VALID_URL = r'^(?:https?://)?(?:www\.)?infoq\.com/[^/]+/[^/]+$'
-    IE_NAME = u'infoq'
-
-    def report_webpage(self, video_id):
-        """Report information extraction."""
-        self._downloader.to_screen(u'[%s] %s: Downloading webpage' % (self.IE_NAME, video_id))
 
     def report_extraction(self, video_id):
         """Report information extraction."""
@@ -2838,38 +2823,29 @@ class InfoQIE(InfoExtractor):
             self._downloader.trouble(u'ERROR: invalid URL: %s' % url)
             return
 
-        self.report_webpage(url)
-
-        request = compat_urllib_request.Request(url)
-        try:
-            webpage = compat_urllib_request.urlopen(request).read()
-        except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err:
-            self._downloader.trouble(u'ERROR: unable to download video webpage: %s' % compat_str(err))
-            return
-
+        webpage = self._download_webpage(url, video_id=url)
         self.report_extraction(url)
 
-
         # Extract video URL
         mobj = re.search(r"jsclassref='([^']*)'", webpage)
         if mobj is None:
             self._downloader.trouble(u'ERROR: unable to extract video url')
             return
-        video_url = 'rtmpe://video.infoq.com/cfx/st/' + compat_urllib_parse.unquote(mobj.group(1).decode('base64'))
-
+        real_id = compat_urllib_parse.unquote(base64.b64decode(mobj.group(1).encode('ascii')).decode('utf-8'))
+        video_url = 'rtmpe://video.infoq.com/cfx/st/' + real_id
 
         # Extract title
         mobj = re.search(r'contentTitle = "(.*?)";', webpage)
         if mobj is None:
             self._downloader.trouble(u'ERROR: unable to extract video title')
             return
-        video_title = mobj.group(1).decode('utf-8')
+        video_title = mobj.group(1)
 
         # Extract description
         video_description = u'No description available.'
         mobj = re.search(r'<meta name="description" content="(.*)"(?:\s*/)?>', webpage)
         if mobj is not None:
-            video_description = mobj.group(1).decode('utf-8')
+            video_description = mobj.group(1)
 
         video_filename = video_url.split('/')[-1]
         video_id, extension = video_filename.split('.')
@@ -3718,8 +3694,8 @@ class SteamIE(InfoExtractor):
         videourl = 'http://store.steampowered.com/video/%s/' % gameID
         webpage = self._download_webpage(videourl, gameID)
         mweb = re.finditer(urlRE, webpage)
-        namesRE = r'<span class=\"title\">(?P<videoName>[\w:/\.\?=\+\s-]+)</span>'
-        titles = list(re.finditer(namesRE, webpage))
+        namesRE = r'<span class="title">(?P<videoName>.+?)</span>'
+        titles = re.finditer(namesRE, webpage)
         videos = []
         for vid,vtitle in zip(mweb,titles):
             video_id = vid.group('videoID')
@@ -3731,7 +3707,7 @@ class SteamIE(InfoExtractor):
                 'id':video_id,
                 'url':video_url,
                 'ext': 'flv',
-                'title': title
+                'title': unescapeHTML(title)
                   }
             videos.append(info)
         return videos