diff options
author | Sergey M․ <dstftw@gmail.com> | 2017-05-19 23:11:02 +0700 |
---|---|---|
committer | Sergey M․ <dstftw@gmail.com> | 2017-05-19 23:11:02 +0700 |
commit | 8fffac6927b4144bfb3e473b58fa52313042ecbb (patch) | |
tree | 34e3f9bf33db2ac99c4e840dddafcc3b2d239852 /youtube_dl/extractor/njpwworld.py | |
parent | 5f6fbcea08bc6473cf24c8f2ffee38bc01c80de0 (diff) |
[njpwworld] Fix extraction (closes #13162)
Diffstat (limited to 'youtube_dl/extractor/njpwworld.py')
-rw-r--r-- | youtube_dl/extractor/njpwworld.py | 22 |
1 files changed, 15 insertions, 7 deletions
diff --git a/youtube_dl/extractor/njpwworld.py b/youtube_dl/extractor/njpwworld.py index f5e3f6815..9b5ad5a9f 100644 --- a/youtube_dl/extractor/njpwworld.py +++ b/youtube_dl/extractor/njpwworld.py @@ -6,6 +6,7 @@ import re from .common import InfoExtractor from ..compat import compat_urlparse from ..utils import ( + extract_attributes, get_element_by_class, urlencode_postdata, ) @@ -56,17 +57,24 @@ class NJPWWorldIE(InfoExtractor): webpage = self._download_webpage(url, video_id) formats = [] - for player_url, kind in re.findall(r'<a[^>]+href="(/player[^"]+)".+?<img[^>]+src="[^"]+qf_btn_([^".]+)', webpage): - player_url = compat_urlparse.urljoin(url, player_url) - + for mobj in re.finditer(r'<a[^>]+\bhref=(["\'])/player.+?[^>]*>', webpage): + player = extract_attributes(mobj.group(0)) + player_path = player.get('href') + if not player_path: + continue + kind = self._search_regex( + r'(low|high)$', player.get('class') or '', 'kind', + default='low') + player_url = compat_urlparse.urljoin(url, player_path) player_page = self._download_webpage( player_url, video_id, note='Downloading player page') - entries = self._parse_html5_media_entries( player_url, player_page, video_id, m3u8_id='hls-%s' % kind, - m3u8_entry_protocol='m3u8_native', - preference=2 if 'hq' in kind else 1) - formats.extend(entries[0]['formats']) + m3u8_entry_protocol='m3u8_native') + kind_formats = entries[0]['formats'] + for f in kind_formats: + f['quality'] = 2 if kind == 'high' else 1 + formats.extend(kind_formats) self._sort_formats(formats) |