From af9cdee9cba610aa3924f90a8a3fcd7dd43c65eb Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Sergey=20M=E2=80=A4?= Date: Sat, 13 Jun 2015 19:53:32 +0600 Subject: [PATCH] [brightcove] Improve and generalize brightcove URL extraction from JS --- youtube_dl/extractor/brightcove.py | 35 +++++++++++++++++++++--------- 1 file changed, 25 insertions(+), 10 deletions(-) diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py index 20a6ed965..d768f99e6 100644 --- a/youtube_dl/extractor/brightcove.py +++ b/youtube_dl/extractor/brightcove.py @@ -156,6 +156,28 @@ def find_param(name): linkBase = find_param('linkBaseURL') if linkBase is not None: params['linkBaseURL'] = linkBase + return cls._make_brightcove_url(params) + + @classmethod + def _build_brighcove_url_from_js(cls, object_js): + # The layout of JS is as follows: + # customBC.createVideo = function (width, height, playerID, playerKey, videoPlayer, VideoRandomID) { + # // build Brightcove XML + # } + m = re.search( + r'''(?x)customBC.\createVideo\( + .*? # skipping width and height + ["\'](?P\d+)["\']\s*,\s* # playerID + ["\'](?PAQ[^"\']{48})[^"\']*["\']\s*,\s* # playerKey begins with AQ and is 50 characters + # in length, however it's appended to itself + # in places, so truncate + ["\'](?P\d+)["\'] # @videoPlayer + ''', object_js) + if m: + return cls._make_brightcove_url(m.groupdict()) + + @classmethod + def _make_brightcove_url(cls, params): data = compat_urllib_parse.urlencode(params) return cls._FEDERATED_URL_TEMPLATE % data @@ -191,16 +213,9 @@ def _extract_brightcove_urls(cls, webpage): if matches: return list(filter(None, [cls._build_brighcove_url(m) for m in matches])) - custombcs = re.findall(r'customBC.\createVideo\((.+?)\);',webpage) - if custombcs: - urls = [] - for match in custombcs: - # brightcove playerkey begins with AQ and is 50 characters in length, - # however it's appended to itself in places, so truncate. - f = re.search(r'["\'](AQ[^"\']{48}).*?["\'](\d+)["\']', match) - if f: - urls.append('brightcove:playerKey='+f.group(1)+'&%40videoPlayer='+f.group(2)) - return urls + return list(filter(None, [ + cls._build_brighcove_url_from_js(custom_bc) + for custom_bc in re.findall(r'(customBC\.createVideo\(.+?\);)', webpage)])) def _real_extract(self, url): url, smuggled_data = unsmuggle_url(url, {})