From 536f819eda975224666374a9ce83cc3472f5aa5a Mon Sep 17 00:00:00 2001 From: =?utf8?q?Sergey=20M=E2=80=A4?= Date: Sat, 14 Nov 2015 05:51:05 +0600 Subject: [PATCH] [brightcove] Imrove extraction of new embeds --- youtube_dl/extractor/brightcove.py | 42 +++++++++++++++++++++--------- 1 file changed, 29 insertions(+), 13 deletions(-) diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py index 2ad35ec90..d494b8b67 100644 --- a/youtube_dl/extractor/brightcove.py +++ b/youtube_dl/extractor/brightcove.py @@ -354,7 +354,7 @@ class BrightcoveIE(InfoExtractor): class BrightcoveInPageEmbedIE(InfoExtractor): - _VALID_URL = r'https?://players\.brightcove\.net/(?P\d+)/(?P[\da-f-]+)_(?P[a-z]+)/index\.html\?.*videoId=(?P\d+)' + _VALID_URL = r'https?://players\.brightcove\.net/(?P\d+)/(?P[^/]+)_(?P[^/]+)/index\.html\?.*videoId=(?P\d+)' _TEST = { 'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001', 'md5': 'c8100925723840d4b0d243f7025703be', @@ -370,18 +370,34 @@ class BrightcoveInPageEmbedIE(InfoExtractor): } } - @staticmethod - def _extract_url(webpage): - video_attributes = re.search(r'(?s)]*)>.*?', webpage) - if video_attributes: - video_attributes = extract_attributes(video_attributes.group(), r'(?s)\s*data-(account|video-id|playlist-id|policy-key|player|embed)\s*=\s*["\']([^"\']+)["\']') - account_id = video_attributes.get('account') - player_id = video_attributes.get('player') - embed = video_attributes.get('embed') - video_id = video_attributes.get('video-id') - if account_id and player_id and embed and video_id: - return 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s' % (account_id, player_id, embed, video_id) - return None + def _extract_urls(self, webpage): + # Reference: + # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe + # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript) + # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html + + entries = [] + + # Look for iframe embeds [1] + for _, url in re.findall( + r']+src=(["\'])((?:https?:)//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage): + entries.append(self.url_result(self._proto_relative_url(url))) + # Look for embed_in_page embeds [2] + # According to examples from [3] it's unclear whether video id may be optional + # and what to do when it is + for video_id, account_id, player_id, embed in re.findall( + r'''(?sx) + ]+ + data-video-id=["\'](\d+)["\'][^>]*>.*? + .*? + ]+ + src=["\'](?:https?:)?//players\.brightcove\.net/ + (\d+)/([\da-f-]+)_([^/]+)/index\.min\.js + ''', webpage): + entries.append(self.url_result( + 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s' + % (account_id, player_id, embed, video_id))) + return entries def _real_extract(self, url): account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups() -- 2.22.2