diff options
| author | knagano <knagano@sodan.org> | 2011-05-07 22:53:37 +0900 |
|---|---|---|
| committer | knagano <knagano@sodan.org> | 2011-05-07 22:53:37 +0900 |
| commit | 62a29bbf7bd0b0d9539aa903519f252671e4eebd (patch) | |
| tree | f358c24215f45a12ac0a04100221f0654af7d180 | |
| parent | 4b0d9eed458d862ce938849a9dc88a9a56f57dc2 (diff) | |
Fixed download from Dailymotion.
Fetches FLV URL from "sdURL" in addVariable("sequence") JSON,
instead of addVariable("video") which doesnot exist today.
Supports new title, uploader nickname format.
| -rwxr-xr-x | youtube-dl | 15 |
1 files changed, 10 insertions, 5 deletions
diff --git a/youtube-dl b/youtube-dl index 3ac27a857..377ceff58 100755 --- a/youtube-dl +++ b/youtube-dl @@ -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 |