mollusk/youtube-dl
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Fixed download from Dailymotion.

Browse Source 
Fetches FLV URL from "sdURL" in addVariable("sequence") JSON,
instead of addVariable("video") which doesnot exist today.
Supports new title, uploader nickname format.
This commit is contained in:
knagano
2011-05-07 22:53:37 +09:00
parent 4b0d9eed45
commit 62a29bbf7b
1 changed files with 10 additions and 5 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
youtube-dl
View File

@@ -1320,6 +1320,7 @@ class DailymotionIE(InfoExtractor):
# Retrieve video webpage to extract further information
request = urllib2.Request(url)
request.add_header('Cookie', 'family_filter=off')
try:
self.report_download_webpage(video_id)
webpage = urllib2.urlopen(request).read()
@@ -1329,25 +1330,29 @@ class DailymotionIE(InfoExtractor):
# Extract URL, uploader and title from webpage
self.report_extraction(video_id)
mobj = re.search(r'(?i)addVariable\(\"video\"\s*,\s*\"([^\"]*)\"\)', webpage)
mobj = re.search(r'(?i)addVariable\(\"sequence\"\s*,\s*\"([^\"]+?)\"\)', webpage)
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract media URL')
return
mediaURL = urllib.unquote(mobj.group(1))
sequence = urllib.unquote(mobj.group(1))
mobj = re.search(r',\"sdURL\"\:\"([^\"]+?)\",', sequence)
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract media URL')
return
mediaURL = urllib.unquote(mobj.group(1)).replace('\\', '')
# if needed add http://www.dailymotion.com/ if relative URL
video_url = mediaURL
# '<meta\s+name="title"\s+content="Dailymotion\s*[:\-]\s*(.*?)"\s*\/\s*>'
mobj = re.search(r'(?im)<title>Dailymotion\s*[\-:]\s*(.+?)</title>', webpage)
mobj = re.search(r'(?im)<title>Dailymotion\s*-\s*(.+)\s*-\s*[^<]+?</title>', webpage)
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract title')
return
video_title = mobj.group(1).decode('utf-8')
video_title = sanitize_title(video_title)
mobj = re.search(r'(?im)<Attribute name="owner">(.+?)</Attribute>', webpage)
mobj = re.search(r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a></span>', webpage)
if mobj is None:
self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
return