diff options
author | Sergey M․ <dstftw@gmail.com> | 2016-07-09 00:28:07 +0700 |
---|---|---|
committer | Sergey M․ <dstftw@gmail.com> | 2016-07-09 00:28:07 +0700 |
commit | cedc70b29277c7ceebee4843bc2f13ebe88109b0 (patch) | |
tree | 87c5e2850629e6950a2e6f47fc2d2a111c2a092f /youtube_dl | |
parent | 07d7689f2eb4698cf98ee837b56489b7ff1924df (diff) |
[facebook] Fix invalid video being extracted (Closes #9851)
Diffstat (limited to 'youtube_dl')
-rw-r--r-- | youtube_dl/extractor/facebook.py | 19 |
1 files changed, 15 insertions, 4 deletions
diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index f5d4f966a..0d43acc4a 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -219,12 +219,23 @@ class FacebookIE(InfoExtractor): BEFORE = '{swf.addParam(param[0], param[1]);});' AFTER = '.forEach(function(variable) {swf.addVariable(variable[0], variable[1]);});' - m = re.search(re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER), webpage) - if m: - swf_params = m.group(1).replace('\\\\', '\\').replace('\\"', '"') + PATTERN = re.escape(BEFORE) + '(?:\n|\\\\n)(.*?)' + re.escape(AFTER) + + for m in re.findall(PATTERN, webpage): + swf_params = m.replace('\\\\', '\\').replace('\\"', '"') data = dict(json.loads(swf_params)) params_raw = compat_urllib_parse_unquote(data['params']) - video_data = json.loads(params_raw)['video_data'] + video_data_candidate = json.loads(params_raw)['video_data'] + for _, f in video_data_candidate.items(): + if not f: + continue + if isinstance(f, dict): + f = [f] + if isinstance(f, list): + continue + if f[0].get('video_id') == video_id: + video_data = video_data_candidate + break def video_data_list2dict(video_data): ret = {} |