github/yt-dlp
1
0
Fork 0
mirror of https://github.com/yt-dlp/yt-dlp.git synced 2024-11-07 20:30:41 -05:00
Code Issues Projects Releases Packages Wiki Activity

[facebook] Fix and caption if title is empty

Browse source
This commit is contained in:
Philipp Hagemeister 2014-09-13 09:01:57 +02:00
parent 6539c91a78
commit 50317dbbad
1 changed files with 15 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

21
youtube_dl/extractor/facebook.py
View file

@ -29,13 +29,13 @@ class FacebookIE(InfoExtractor):
_NETRC_MACHINE = 'facebook' _NETRC_MACHINE = 'facebook'
IE_NAME = 'facebook' IE_NAME = 'facebook'
_TESTS = [{ _TESTS = [{
'url': 'https://www.facebook.com/photo.php?v=120708114770723', 'url': 'https://www.facebook.com/video.php?v=637842556329505&fref=nf',
'md5': '48975a41ccc4b7a581abd68651c1a5a8', 'md5': '6a40d33c0eccbb1af76cf0485a052659',
'info_dict': { 'info_dict': {
'id': '120708114770723', 'id': '637842556329505',
'ext': 'mp4', 'ext': 'mp4',
'duration': 279, 'duration': 38,
'title': 'PEOPLE ARE AWESOME 2013', 'title': 'Did you know Kei Nishikori is the first Asian man to ever reach a Grand Slam fin...',
} }
}, { }, {
'url': 'https://www.facebook.com/video.php?v=10204634152394104', 'url': 'https://www.facebook.com/video.php?v=10204634152394104',
@ -125,7 +125,16 @@ def _real_extract(self, url):
raise ExtractorError('Cannot find video URL') raise ExtractorError('Cannot find video URL')
video_title = self._html_search_regex( video_title = self._html_search_regex(
r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title') r'<h2 class="uiHeaderTitle">([^<]*)</h2>', webpage, 'title',
fatal=False)
if not video_title:
video_title = self._html_search_regex(
r'(?s)<span class="fbPhotosPhotoCaption".*?id="fbPhotoPageCaption"><span class="hasCaption">(.*?)</span>',
webpage, 'alternative title', default=None)
if len(video_title) > 80 + 3:
video_title = video_title[:80] + '...'
if not video_title:
video_title = 'Facebook video #%s' % video_id
return { return {
'id': video_id, 'id': video_id,