From 38f59e2793dfcb5f493977857304ab50b784e6ee Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Mon, 19 Mar 2018 23:40:19 +0700 Subject: [PATCH] [canalc2] Add support for HTML5 videos (closes #15916, closes #15919) --- youtube_dl/extractor/canalc2.py | 28 ++++++++++++++++++---------- 1 file changed, 18 insertions(+), 10 deletions(-) diff --git a/youtube_dl/extractor/canalc2.py b/youtube_dl/extractor/canalc2.py index acd87e371a..407cc80848 100644 --- a/youtube_dl/extractor/canalc2.py +++ b/youtube_dl/extractor/canalc2.py @@ -31,6 +31,10 @@ def _real_extract(self, url): webpage = self._download_webpage( 'http://www.canalc2.tv/video/%s' % video_id, video_id) + title = self._html_search_regex( + r'(?s)class="[^"]*col_description[^"]*">.*?

(.+?)

', + webpage, 'title') + formats = [] for _, video_url in re.findall(r'file\s*=\s*(["\'])(.+?)\1', webpage): if video_url.startswith('rtmp://'): @@ -49,17 +53,21 @@ def _real_extract(self, url): 'url': video_url, 'format_id': 'http', }) - self._sort_formats(formats) - title = self._html_search_regex( - r'(?s)class="[^"]*col_description[^"]*">.*?

(.*?)

', webpage, 'title') - duration = parse_duration(self._search_regex( - r'id=["\']video_duree["\'][^>]*>([^<]+)', - webpage, 'duration', fatal=False)) + if formats: + info = { + 'formats': formats, + } + else: + info = self._parse_html5_media_entries(url, webpage, url)[0] - return { + self._sort_formats(info['formats']) + + info.update({ 'id': video_id, 'title': title, - 'duration': duration, - 'formats': formats, - } + 'duration': parse_duration(self._search_regex( + r'id=["\']video_duree["\'][^>]*>([^<]+)', + webpage, 'duration', fatal=False)), + }) + return info