v/youtube-dl
1
0
Fork 0
mirror of https://github.com/l1ving/youtube-dl synced 2020-11-18 19:53:54 -08:00
Code Issues Releases Wiki Activity

[bilibili] Support paged videos

Browse Source
This commit is contained in:
Changjae 2018-05-02 17:36:24 +09:00
parent ea1f5e5dbd
commit a0e35315e9
1 changed files with 25 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
youtube_dl/extractor/bilibili.py
View File

@ -23,7 +23,7 @@ from ..utils import (
class BiliBiliIE(InfoExtractor): class BiliBiliIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.|bangumi\.|)bilibili\.(?:tv|com)/(?:video/av|anime/(?P<anime_id>\d+)/play#)(?P<id>\d+)' _VALID_URL = r'https?://(?:www\.|bangumi\.|)bilibili\.(?:tv|com)/(?:video/av|anime/(?P<anime_id>\d+)/play#)(?P<id>\d+)(?:/?\?p=(?P<page>\d+))?'
_TESTS = [{ _TESTS = [{
'url': 'http://www.bilibili.tv/video/av1074402/', 'url': 'http://www.bilibili.tv/video/av1074402/',
@ -91,6 +91,22 @@ class BiliBiliIE(InfoExtractor):
'skip_download': True, # Test metadata only 'skip_download': True, # Test metadata only
}, },
}] }]
}, {
# Test pages
'url': 'http://www.bilibili.com/video/av19608864/?p=3',
'md5': 'e0c3368a3a1d53ac5892da1abe57887d',
'info_dict': {
'id': '19608864',
'ext': 'flv',
'title': '【iu】【penta kill】第七届Gaon Chart K-POP Awards~各种屏录~',
'description': 'md5:d9f59c5658f4f2eafc04a191e1ab0828',
'duration': 100.264,
'timestamp': 1518671057,
'upload_date': '20180215',
'thumbnail': r're:^https?://.+\.jpg',
'uploader': '普通的柒月初叁',
'uploader_id': '260251749',
},
}] }]
_APP_KEY = '84956560bc028eb7' _APP_KEY = '84956560bc028eb7'
@ -110,14 +126,14 @@ class BiliBiliIE(InfoExtractor):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id') video_id = mobj.group('id')
anime_id = mobj.group('anime_id') anime_id = mobj.group('anime_id')
page = mobj.group('page') or 1
webpage = self._download_webpage(url, video_id) webpage = self._download_webpage(url, video_id)
if 'anime/' not in url: if 'anime/' not in url:
cid = self._search_regex( mobj = re.findall(r'cid(?:["\']:|=)(\d+)', webpage)
r'cid(?:["\']:|=)(\d+)', webpage, 'cid', cid = mobj[int(page) - 1] or \
default=None compat_parse_qs(self._search_regex([
) or compat_parse_qs(self._search_regex( r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)',
[r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)',
r'EmbedPlayer\([^)]+,\s*\\"([^"]+)\\"\)', r'EmbedPlayer\([^)]+,\s*\\"([^"]+)\\"\)',
r'<iframe[^>]+src="https://secure\.bilibili\.com/secure,([^"]+)"'], r'<iframe[^>]+src="https://secure\.bilibili\.com/secure,([^"]+)"'],
webpage, 'player parameters'))['cid'][0] webpage, 'player parameters'))['cid'][0]