1
0
Fork 0
mirror of https://github.com/blackjack4494/yt-dlc.git synced 2024-12-29 19:26:51 +00:00
yt-dlc/youtube_dl/extractor/vuclip.py
2016-07-11 00:52:25 +08:00

70 lines
2.2 KiB
Python

from __future__ import unicode_literals
import re
from .common import InfoExtractor
from ..compat import (
compat_urllib_parse_urlparse,
)
from ..utils import (
ExtractorError,
parse_duration,
remove_end,
)
class VuClipIE(InfoExtractor):
_VALID_URL = r'https?://(?:m\.)?vuclip\.com/w\?.*?cid=(?P<id>[0-9]+)'
_TEST = {
'url': 'http://m.vuclip.com/w?cid=922692425&fid=70295&z=1010&nvar&frm=index.html',
'info_dict': {
'id': '922692425',
'ext': '3gp',
'title': 'The Toy Soldiers - Hollywood Movie Trailer',
'duration': 177,
}
}
def _real_extract(self, url):
video_id = self._match_id(url)
webpage = self._download_webpage(url, video_id)
ad_m = re.search(
r'''value="No.*?" onClick="location.href='([^"']+)'"''', webpage)
if ad_m:
urlr = compat_urllib_parse_urlparse(url)
adfree_url = urlr.scheme + '://' + urlr.netloc + ad_m.group(1)
webpage = self._download_webpage(
adfree_url, video_id, note='Download post-ad page')
error_msg = self._html_search_regex(
r'<p class="message">(.*?)</p>', webpage, 'error message',
default=None)
if error_msg:
raise ExtractorError(
'%s said: %s' % (self.IE_NAME, error_msg), expected=True)
# These clowns alternate between two page types
video_url = self._search_regex(
r'<a[^>]+href="([^"]+)"[^>]*><img[^>]+src="[^"]*/play\.gif',
webpage, 'video URL', default=None)
if video_url:
formats = [{
'url': video_url,
}]
else:
formats = self._parse_html5_media_entries(url, webpage)[0]['formats']
title = remove_end(self._html_search_regex(
r'<title>(.*?)-\s*Vuclip</title>', webpage, 'title').strip(), ' - Video')
duration = parse_duration(self._html_search_regex(
r'[(>]([0-9]+:[0-9]+)(?:<span|\))', webpage, 'duration', fatal=False))
return {
'id': video_id,
'formats': formats,
'title': title,
'duration': duration,
}