[iqiyi] Fix extraction of Yule videos

This commit is contained in:
Yen Chi Hsuan 2017-04-25 22:23:57 +08:00
parent 290f64dbaa
commit 01cb57016f
No known key found for this signature in database
GPG key ID: 7F902A182457CA23
2 changed files with 10 additions and 4 deletions

View file

@ -5,6 +5,7 @@ Core
* [socks] Handle cases where credentials are required but missing
Extractors
* [iqiyi] Fix extraction of Yule videos
* [iqiyi] Fix playlist detection (#12504)
- [azubu] Remove extractor (#12813)
* [porn91] Fix extraction (#12814)

View file

@ -189,7 +189,11 @@ class IqiyiIE(InfoExtractor):
'only_matching': True,
}, {
'url': 'http://yule.iqiyi.com/pcb.html',
'only_matching': True,
'info_dict': {
'id': '4a0af228fddb55ec96398a364248ed7f',
'ext': 'mp4',
'title': '第2017-04-21期 女艺人频遭极端粉丝骚扰',
},
}, {
# VIP-only video. The first 2 parts (6 minutes) are available without login
# MD5 sums omitted as values are different on Travis CI and my machine
@ -340,7 +344,7 @@ def _real_extract(self, url):
# Sometimes there are playlist links in individual videos, so treat it
# as a single video first
tvid = self._search_regex(
r'data-player-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid', default=None)
r'data-(?:player|shareplattrigger)-tvid\s*=\s*[\'"](\d+)', webpage, 'tvid', default=None)
if tvid is None:
playlist_result = self._extract_playlist(webpage)
if playlist_result:
@ -348,7 +352,7 @@ def _real_extract(self, url):
raise ExtractorError('Can\'t find any video')
video_id = self._search_regex(
r'data-player-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
r'data-(?:player|shareplattrigger)-videoid\s*=\s*[\'"]([a-f\d]+)', webpage, 'video_id')
formats = []
for _ in range(5):
@ -380,7 +384,8 @@ def _real_extract(self, url):
self._sort_formats(formats)
title = (get_element_by_id('widget-videotitle', webpage) or
clean_html(get_element_by_attribute('class', 'mod-play-tit', webpage)))
clean_html(get_element_by_attribute('class', 'mod-play-tit', webpage)) or
self._html_search_regex(r'<span[^>]+data-videochanged-title="word"[^>]*>([^<]+)</span>', webpage, 'title'))
return {
'id': video_id,