From 01cb57016ff747ab07c506a708093dfd633d58f6 Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Tue, 25 Apr 2017 22:23:57 +0800 Subject: [PATCH] [iqiyi] Fix extraction of Yule videos --- ChangeLog | 1 + youtube_dl/extractor/iqiyi.py | 13 +++++++++---- 2 files changed, 10 insertions(+), 4 deletions(-) diff --git a/ChangeLog b/ChangeLog index 50daa1a69..d38039663 100644 --- a/ChangeLog +++ b/ChangeLog @@ -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) diff --git a/youtube_dl/extractor/iqiyi.py b/youtube_dl/extractor/iqiyi.py index 8a711afb9..fdfa7de9e 100644 --- a/youtube_dl/extractor/iqiyi.py +++ b/youtube_dl/extractor/iqiyi.py @@ -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 @@ class IqiyiIE(InfoExtractor): # 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 @@ class IqiyiIE(InfoExtractor): 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 @@ class IqiyiIE(InfoExtractor): 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']+data-videochanged-title="word"[^>]*>([^<]+)', webpage, 'title')) return { 'id': video_id,