1
0
Fork 0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2024-11-14 05:15:37 +00:00

Fixed dailymotion video nd uploader extractor

This commit is contained in:
Laurent Raufaste 2011-05-29 12:36:12 -04:00
parent 4b0d9eed45
commit 368cb1fd0a

View file

@ -30,6 +30,7 @@ import time
import urllib
import urllib2
import zlib
import json
# parse_qs was moved from the cgi module to the urlparse module recently.
try:
@ -1328,16 +1329,21 @@ class DailymotionIE(InfoExtractor):
return
# Extract URL, uploader and title from webpage
# First we need to get the sequence urlencoded json variable
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))
# if needed add http://www.dailymotion.com/ if relative URL
video_url = mediaURL
sequenceJsonContent = urllib.unquote_plus(mobj.group(1))
sequenceJson = json.loads(sequenceJsonContent)
try:
video_url = sequenceJson[0]["layerList"][0]["sequenceList"][1]["layerList"][2]["param"]["videoPluginParameters"]["hqURL"]
except:
self._downloader.trouble(u'ERROR: unable to extract media URL')
return
# '<meta\s+name="title"\s+content="Dailymotion\s*[:\-]\s*(.*?)"\s*\/\s*>'
mobj = re.search(r'(?im)<title>Dailymotion\s*[\-:]\s*(.+?)</title>', webpage)
@ -1347,11 +1353,12 @@ class DailymotionIE(InfoExtractor):
video_title = mobj.group(1).decode('utf-8')
video_title = sanitize_title(video_title)
mobj = re.search(r'(?im)<Attribute name="owner">(.+?)</Attribute>', webpage)
if mobj is None:
# Extract the video uploader nickname from the sequence JSON
try:
video_uploader = sequenceJson[0]["layerList"][0]["sequenceList"][1]["layerList"][0]["param"]["metadata"]["uploader"]
except:
self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
return
video_uploader = mobj.group(1)
try:
# Process video information