Browse Source

Dailymotion: fix the download of the video in the max quality (closes #986)

rtmp_test
Jaime Marquínez Ferrándiz 12 years ago
parent
commit
b27c856fbc
1 changed files with 20 additions and 20 deletions
  1. +20
    -20
      youtube_dl/extractor/dailymotion.py

+ 20
- 20
youtube_dl/extractor/dailymotion.py View File

@ -1,9 +1,9 @@
import re import re
import json
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
compat_urllib_request, compat_urllib_request,
compat_urllib_parse,
ExtractorError, ExtractorError,
unescapeHTML, unescapeHTML,
@ -39,26 +39,7 @@ class DailymotionIE(InfoExtractor):
# Extract URL, uploader and title from webpage # Extract URL, uploader and title from webpage
self.report_extraction(video_id) self.report_extraction(video_id)
mobj = re.search(r'\s*var flashvars = (.*)', webpage)
if mobj is None:
raise ExtractorError(u'Unable to extract media URL')
flashvars = compat_urllib_parse.unquote(mobj.group(1))
for key in ['hd1080URL', 'hd720URL', 'hqURL', 'sdURL', 'ldURL', 'video_url']:
if key in flashvars:
max_quality = key
self.to_screen(u'Using %s' % key)
break
else:
raise ExtractorError(u'Unable to extract video URL')
mobj = re.search(r'"' + max_quality + r'":"(.+?)"', flashvars)
if mobj is None:
raise ExtractorError(u'Unable to extract video URL')
video_url = compat_urllib_parse.unquote(mobj.group(1)).replace('\\/', '/')
# TODO: support choosing qualities
mobj = re.search(r'<meta property="og:title" content="(?P<title>[^"]*)" />', webpage) mobj = re.search(r'<meta property="og:title" content="(?P<title>[^"]*)" />', webpage)
if mobj is None: if mobj is None:
@ -76,6 +57,25 @@ class DailymotionIE(InfoExtractor):
if mobj is not None: if mobj is not None:
video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1) video_upload_date = mobj.group(3) + mobj.group(2) + mobj.group(1)
embed_url = 'http://www.dailymotion.com/embed/video/%s' % video_id
embed_page = self._download_webpage(embed_url, video_id,
u'Downloading embed page')
info = self._search_regex(r'var info = ({.*?}),', embed_page, 'video info')
info = json.loads(info)
# TODO: support choosing qualities
for key in ['stream_h264_hd1080_url','stream_h264_hd_url',
'stream_h264_hq_url','stream_h264_url',
'stream_h264_ld_url']:
if info.get(key):#key in info and info[key]:
max_quality = key
self.to_screen(u'Using %s' % key)
break
else:
raise ExtractorError(u'Unable to extract video URL')
video_url = info[max_quality]
return [{ return [{
'id': video_id, 'id': video_id,
'url': video_url, 'url': video_url,


Loading…
Cancel
Save