diff options
| author | George Brighton <george@gebn.co.uk> | 2015-08-02 19:21:10 +0100 |
|---|---|---|
| committer | Sergey M․ <dstftw@gmail.com> | 2015-08-03 02:35:06 +0600 |
| commit | 754e70cf3e74218ae5d840985fbf07bbe274332a (patch) | |
| tree | f73567f7944cda7d43880f6619b3e147463a44b8 /youtube_dl | |
| parent | 84bc4dcb0f678f0a8c9f993e101b9769e3959f76 (diff) | |
[pornhub] Fix video url regular expression.
PornHub seems to have subtly changed their JavaScript. Before, video URL strings were embedded directly in the video's `flashvars_*` object, but they are now assigned to variables of the form `player_quality_*`, which are then added to this object later under the relevant quality key.
Diffstat (limited to 'youtube_dl')
| -rw-r--r-- | youtube_dl/extractor/pornhub.py | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 0b7886840..fbaa830d6 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -81,7 +81,7 @@ class PornHubIE(InfoExtractor): comment_count = self._extract_count( r'All Comments\s*<span>\(([\d,.]+)\)', webpage, 'comment') - video_urls = list(map(compat_urllib_parse_unquote, re.findall(r'"quality_[0-9]{3}p":"([^"]+)', webpage))) + video_urls = list(map(compat_urllib_parse_unquote, re.findall(r"var player_quality_[0-9]{3}p = '([^']+)'", webpage))) if webpage.find('"encrypted":true') != -1: password = compat_urllib_parse_unquote_plus( self._search_regex(r'"video_title":"([^"]+)', webpage, 'password')) |