X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/9fb40a4e5f59ba243dee2edad23a2c5555338113..fe6cc1b2aebf52f3fcdcc22ca8e846b67a406cc0:/youtube_dl/extractor/iqiyi.py diff --git a/youtube_dl/extractor/iqiyi.py b/youtube_dl/extractor/iqiyi.py index 2af6a6d..cd11aa7 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 @@ -235,7 +239,7 @@ class IqiyiIE(InfoExtractor): return ohdave_rsa_encrypt(data, e, N) def _login(self): - (username, password) = self._get_login_info() + username, password = self._get_login_info() # No authentication to be performed if not username: @@ -337,15 +341,18 @@ class IqiyiIE(InfoExtractor): url, 'temp_id', note='download video page') # There's no simple way to determine whether an URL is a playlist or not - # So detect it - playlist_result = self._extract_playlist(webpage) - if playlist_result: - return playlist_result - + # 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') + 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: + return playlist_result + 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): @@ -376,8 +383,9 @@ class IqiyiIE(InfoExtractor): self._sleep(5, video_id) self._sort_formats(formats) - title = (get_element_by_id('widget-videotitle', webpage) or - clean_html(get_element_by_attribute('class', 'mod-play-tit', webpage))) + title = (get_element_by_id('widget-videotitle', webpage) + or 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,