1
0
mirror of https://github.com/ytdl-org/youtube-dl.git synced 2025-04-19 21:37:40 +00:00
This commit is contained in:
GitHub Merge Button 2011-09-13 14:38:01 -07:00
commit b0719de04b

View File

@ -39,6 +39,7 @@ import urllib
import urllib2
import warnings
import zlib
import json
if os.name == 'nt':
import ctypes
@ -1519,7 +1520,7 @@ class DailymotionIE(InfoExtractor):
video_id = mobj.group(1)
simple_title = mobj.group(2).decode('utf-8')
video_extension = 'flv'
video_extension = 'mp4'
# Retrieve video webpage to extract further information
request = urllib2.Request(url)
@ -1531,16 +1532,31 @@ 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
sequenceJsonContent = urllib.unquote_plus(mobj.group(1))
video_url = mediaURL
# JSON does not suppot escaping of '.
# Replace every \' by a ' in the JSON string
sequenceJsonContent = sequenceJsonContent.replace("\\'", "'")
# Build the JSON object based on the string
try:
sequenceJson = json.loads(sequenceJsonContent)
except:
self._downloader.trouble(u'ERROR: unable to extract media URL (Bad JSON encoding)')
return
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 (Unable to find the URL)')
return
# '<meta\s+name="title"\s+content="Dailymotion\s*[:\-]\s*(.*?)"\s*\/\s*>'
mobj = re.search(r'(?im)<title>Dailymotion\s*[\-:]\s*(.+?)</title>', webpage)
@ -1550,11 +1566,12 @@ 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