[bilibili] add support for specefic page extraction

This commit is contained in:
remitamine 2015-10-17 23:12:58 +01:00
parent 55af2b26e0
commit 520e753390
1 changed files with 23 additions and 22 deletions

View File

@ -1,18 +1,20 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import re
import json import json
import xml.etree.ElementTree as ET import xml.etree.ElementTree as ET
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import ( from ..utils import (
int_or_none, int_or_none,
unescapeHTML,
ExtractorError, ExtractorError,
) )
class BiliBiliIE(InfoExtractor): class BiliBiliIE(InfoExtractor):
_VALID_URL = r'http://www\.bilibili\.(?:tv|com)/video/av(?P<id>[0-9]+)/' _VALID_URL = r'http://www\.bilibili\.(?:tv|com)/video/av(?P<id>\d+)(?:/index_(?P<page_num>\d+).html)?'
_TESTS = [{ _TESTS = [{
'url': 'http://www.bilibili.tv/video/av1074402/', 'url': 'http://www.bilibili.tv/video/av1074402/',
@ -33,17 +35,31 @@ class BiliBiliIE(InfoExtractor):
'info_dict': { 'info_dict': {
'id': '1041170', 'id': '1041170',
'title': '【BD1080P】刀语【诸神&异域】', 'title': '【BD1080P】刀语【诸神&异域】',
'description': '这是个神奇的故事~每个人不留弹幕不给走哦~切利哦!~',
'uploader': '枫叶逝去',
'timestamp': 1396501299,
}, },
'playlist_count': 12, 'playlist_count': 9,
}] }]
def _extract_video_info(self, cid, view_data, page_num=1, num_pages=1): def _real_extract(self, url):
title = view_data['title'] mobj = re.match(self._VALID_URL, url)
video_id = mobj.group('id')
page_num = mobj.group('page_num') or '1'
view_data = self._download_json(
'http://api.bilibili.com/view?type=json&appkey=8e9fc618fbd41e28&id=%s&page=%s' % (video_id, page_num),
video_id)
if 'error' in view_data:
raise ExtractorError('%s said: %s' % (self.IE_NAME, view_data['error']), expected=True)
cid = view_data['cid']
title = unescapeHTML(view_data['title'])
page = self._download_webpage( page = self._download_webpage(
'http://interface.bilibili.com/v_cdn_play?appkey=8e9fc618fbd41e28&cid=%s' % cid, 'http://interface.bilibili.com/v_cdn_play?appkey=8e9fc618fbd41e28&cid=%s' % cid,
cid, cid,
'Downloading page %d/%d' % (page_num, num_pages) 'Downloading page %s/%s' % (page_num, view_data['pages'])
) )
try: try:
err_info = json.loads(page) err_info = json.loads(page)
@ -76,7 +92,7 @@ class BiliBiliIE(InfoExtractor):
}) })
info = { info = {
'id': cid, 'id': str(cid),
'title': title, 'title': title,
'description': view_data.get('description'), 'description': view_data.get('description'),
'thumbnail': view_data.get('pic'), 'thumbnail': view_data.get('pic'),
@ -92,22 +108,7 @@ class BiliBiliIE(InfoExtractor):
else: else:
info.update({ info.update({
'_type': 'multi_video', '_type': 'multi_video',
'id': video_id,
'entries': entries, 'entries': entries,
}) })
return info return info
def _real_extract(self, url):
video_id = self._match_id(url)
view_data = self._download_json('http://api.bilibili.com/view?type=json&appkey=8e9fc618fbd41e28&id=%s' % video_id, video_id)
num_pages = int_or_none(view_data['pages'])
if num_pages > 1:
play_list_title = view_data['title']
page_list = self._download_json('http://www.bilibili.com/widget/getPageList?aid=%s' % video_id, video_id, 'Downloading page list metadata')
entries = []
for page in page_list:
view_data['title'] = page['pagename']
entries.append(self._extract_video_info(str(page['cid']), view_data, page['page'], num_pages))
return self.playlist_result(entries, video_id, play_list_title, view_data.get('description'))
else:
return self._extract_video_info(str(view_data['cid']), view_data)