Fixed 4tube.com extractor to pull metadata from associated Javascript and not the HTML of the desired page.

This commit is contained in:
kitty 2015-10-08 06:31:23 -07:00 committed by Sergey M․
parent c725e2c8b0
commit b90b0c4ffa

View file

@ -45,11 +45,9 @@ class FourTubeIE(InfoExtractor):
'uploadDate', webpage)) 'uploadDate', webpage))
thumbnail = self._html_search_meta('thumbnailUrl', webpage) thumbnail = self._html_search_meta('thumbnailUrl', webpage)
uploader_id = self._html_search_regex( uploader_id = self._html_search_regex(
r'<a class="img-avatar" href="[^"]+/channels/([^/"]+)" title="Go to [^"]+ page">', r'<a class="img-avatar" href="[^"]+/users/([^/"]+)" title="Go to [^"]+ page">', webpage, 'uploader id')
webpage, 'uploader id')
uploader = self._html_search_regex( uploader = self._html_search_regex(
r'<a class="img-avatar" href="[^"]+/channels/[^/"]+" title="Go to ([^"]+) page">', r'<a class="img-avatar" href="[^"]+/users/[^/"]+" title="Go to ([^"]+) page">', webpage, 'uploader')
webpage, 'uploader')
categories_html = self._search_regex( categories_html = self._search_regex(
r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>', r'(?s)><i class="icon icon-tag"></i>\s*Categories / Tags\s*.*?<ul class="list">(.*?)</ul>',
@ -68,9 +66,12 @@ class FourTubeIE(InfoExtractor):
webpage, 'like count', fatal=False)) webpage, 'like count', fatal=False))
duration = parse_duration(self._html_search_meta('duration', webpage)) duration = parse_duration(self._html_search_meta('duration', webpage))
player_url = self._search_regex(r'<script id="playerembed" src="([^"]+)">',webpage,'player javascript')
player_js = self._download_webpage(player_url,video_id,'Downloading player Javascript')
params_js = self._search_regex( params_js = self._search_regex(
r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)', r'\$\.ajax\(url,\ opts\);\s*\}\s*\}\)\(([0-9,\[\] ]+)\)',
webpage, 'initialization parameters' player_js, 'initialization parameters'
) )
params = self._parse_json('[%s]' % params_js, video_id) params = self._parse_json('[%s]' % params_js, video_id)
media_id = params[0] media_id = params[0]