From 66c9fa36c10860b380806b9de48f38d628289e03 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Wed, 6 Sep 2017 00:48:37 +0700 Subject: [PATCH] [youtube] Separate methods for embeds extraction --- youtube_dl/extractor/generic.py | 33 ++++------------------------- youtube_dl/extractor/youtube.py | 37 +++++++++++++++++++++++++++++++++ 2 files changed, 41 insertions(+), 29 deletions(-) diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index c81efdc00..b83c18380 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -2243,36 +2243,11 @@ class GenericIE(InfoExtractor): if vid_me_embed_url is not None: return self.url_result(vid_me_embed_url, 'Vidme') - # Look for embedded YouTube player - matches = re.findall(r'''(?x) - (?: - ]+?src=| - data-video-url=| - ]+?src=| - embedSWF\(?:\s*| - ]+data=| - new\s+SWFObject\( - ) - (["\']) - (?P(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/ - (?:embed|v|p)/.+?) - \1''', webpage) - if matches: + # Look for YouTube embeds + youtube_urls = YoutubeIE._extract_urls(webpage) + if youtube_urls: return self.playlist_from_matches( - matches, video_id, video_title, lambda m: unescapeHTML(m[1])) - - # Look for lazyYT YouTube embed - matches = re.findall( - r'class="lazyYT" data-youtube-id="([^"]+)"', webpage) - if matches: - return self.playlist_from_matches(matches, video_id, video_title, lambda m: unescapeHTML(m)) - - # Look for Wordpress "YouTube Video Importer" plugin - matches = re.findall(r'''(?x)]+ - class=(?P[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+ - data-video_id=(?P[\'"])([^\'"]+)(?P=q2)''', webpage) - if matches: - return self.playlist_from_matches(matches, video_id, video_title, lambda m: m[-1]) + youtube_urls, video_id, video_title, ie=YoutubeIE.ie_key()) matches = DailymotionIE._extract_urls(webpage) if matches: diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index 953e38227..ad2e933ee 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -1374,6 +1374,43 @@ class YoutubeIE(YoutubeBaseInfoExtractor): playback_url, video_id, 'Marking watched', 'Unable to mark watched', fatal=False) + @staticmethod + def _extract_urls(webpage): + # Embedded YouTube player + entries = [ + unescapeHTML(mobj.group('url')) + for mobj in re.finditer(r'''(?x) + (?: + ]+?src=| + data-video-url=| + ]+?src=| + embedSWF\(?:\s*| + ]+data=| + new\s+SWFObject\( + ) + (["\']) + (?P(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/ + (?:embed|v|p)/.+?) + \1''', webpage)] + + # lazyYT YouTube embed + entries.extend(list(map( + unescapeHTML, + re.findall(r'class="lazyYT" data-youtube-id="([^"]+)"', webpage)))) + + # Wordpress "YouTube Video Importer" plugin + matches = re.findall(r'''(?x)]+ + class=(?P[\'"])[^\'"]*\byvii_single_video_player\b[^\'"]*(?P=q1)[^>]+ + data-video_id=(?P[\'"])([^\'"]+)(?P=q2)''', webpage) + entries.extend(m[-1] for m in matches) + + return entries + + @staticmethod + def _extract_url(webpage): + urls = YoutubeIE._extract_urls(webpage) + return urls[0] if urls else None + @classmethod def extract_id(cls, url): mobj = re.match(cls._VALID_URL, url, re.VERBOSE)