yt_dlp/extractor/douyutv.py

   1 import time
   2 import hashlib
   3 import re
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     ExtractorError,
   8     unescapeHTML,
   9     unified_strdate,
  10     urljoin,
  11 )
  12
  13
  14 class DouyuTVIE(InfoExtractor):
  15     IE_DESC = '斗鱼'
  16     _VALID_URL = r'https?://(?:www\.)?douyu(?:tv)?\.com/(?:[^/]+/)*(?P<id>[A-Za-z0-9]+)'
  17     _TESTS = [{
  18         'url': 'http://www.douyutv.com/iseven',
  19         'info_dict': {
  20             'id': '17732',
  21             'display_id': 'iseven',
  22             'ext': 'flv',
  23             'title': 're:^清晨醒脑！根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  24             'description': r're:.*m7show@163\.com.*',
  25             'thumbnail': r're:^https?://.*\.jpg$',
  26             'uploader': '7师傅',
  27             'is_live': True,
  28         },
  29         'params': {
  30             'skip_download': True,
  31         },
  32     }, {
  33         'url': 'http://www.douyutv.com/85982',
  34         'info_dict': {
  35             'id': '85982',
  36             'display_id': '85982',
  37             'ext': 'flv',
  38             'title': 're:^小漠从零单排记！——CSOL2躲猫猫 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  39             'description': 'md5:746a2f7a253966a06755a912f0acc0d2',
  40             'thumbnail': r're:^https?://.*\.jpg$',
  41             'uploader': 'douyu小漠',
  42             'is_live': True,
  43         },
  44         'params': {
  45             'skip_download': True,
  46         },
  47         'skip': 'Room not found',
  48     }, {
  49         'url': 'http://www.douyutv.com/17732',
  50         'info_dict': {
  51             'id': '17732',
  52             'display_id': '17732',
  53             'ext': 'flv',
  54             'title': 're:^清晨醒脑！根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
  55             'description': r're:.*m7show@163\.com.*',
  56             'thumbnail': r're:^https?://.*\.jpg$',
  57             'uploader': '7师傅',
  58             'is_live': True,
  59         },
  60         'params': {
  61             'skip_download': True,
  62         },
  63     }, {
  64         'url': 'http://www.douyu.com/xiaocang',
  65         'only_matching': True,
  66     }, {
  67         # \"room_id\"
  68         'url': 'http://www.douyu.com/t/lpl',
  69         'only_matching': True,
  70     }]
  71
  72     def _real_extract(self, url):
  73         video_id = self._match_id(url)
  74
  75         if video_id.isdigit():
  76             room_id = video_id
  77         else:
  78             page = self._download_webpage(url, video_id)
  79             room_id = self._html_search_regex(
  80                 r'"room_id\\?"\s*:\s*(\d+),', page, 'room id')
  81
  82         # Grab metadata from mobile API
  83         room = self._download_json(
  84             'http://m.douyu.com/html5/live?roomId=%s' % room_id, video_id,
  85             note='Downloading room info')['data']
  86
  87         # 1 = live, 2 = offline
  88         if room.get('show_status') == '2':
  89             raise ExtractorError('Live stream is offline', expected=True)
  90
  91         # Grab the URL from PC client API
  92         # The m3u8 url from mobile API requires re-authentication every 5 minutes
  93         tt = int(time.time())
  94         signContent = 'lapi/live/thirdPart/getPlay/%s?aid=pcclient&rate=0&time=%d9TUk5fjjUjg9qIMH3sdnh' % (room_id, tt)
  95         sign = hashlib.md5(signContent.encode('ascii')).hexdigest()
  96         video_url = self._download_json(
  97             'http://coapi.douyucdn.cn/lapi/live/thirdPart/getPlay/' + room_id,
  98             video_id, note='Downloading video URL info',
  99             query={'rate': 0}, headers={
 100                 'auth': sign,
 101                 'time': str(tt),
 102                 'aid': 'pcclient'
 103             })['data']['live_url']
 104
 105         title = unescapeHTML(room['room_name'])
 106         description = room.get('show_details')
 107         thumbnail = room.get('room_src')
 108         uploader = room.get('nickname')
 109
 110         return {
 111             'id': room_id,
 112             'display_id': video_id,
 113             'url': video_url,
 114             'title': title,
 115             'description': description,
 116             'thumbnail': thumbnail,
 117             'uploader': uploader,
 118             'is_live': True,
 119         }
 120
 121
 122 class DouyuShowIE(InfoExtractor):
 123     _VALID_URL = r'https?://v(?:mobile)?\.douyu\.com/show/(?P<id>[0-9a-zA-Z]+)'
 124
 125     _TESTS = [{
 126         'url': 'https://v.douyu.com/show/rjNBdvnVXNzvE2yw',
 127         'md5': '0c2cfd068ee2afe657801269b2d86214',
 128         'info_dict': {
 129             'id': 'rjNBdvnVXNzvE2yw',
 130             'ext': 'mp4',
 131             'title': '陈一发儿：砒霜 我有个室友系列！04-01 22点场',
 132             'duration': 7150.08,
 133             'thumbnail': r're:^https?://.*\.jpg$',
 134             'uploader': '陈一发儿',
 135             'uploader_id': 'XrZwYelr5wbK',
 136             'uploader_url': 'https://v.douyu.com/author/XrZwYelr5wbK',
 137             'upload_date': '20170402',
 138         },
 139     }, {
 140         'url': 'https://vmobile.douyu.com/show/rjNBdvnVXNzvE2yw',
 141         'only_matching': True,
 142     }]
 143
 144     def _real_extract(self, url):
 145         url = url.replace('vmobile.', 'v.')
 146         video_id = self._match_id(url)
 147
 148         webpage = self._download_webpage(url, video_id)
 149
 150         room_info = self._parse_json(self._search_regex(
 151             r'var\s+\$ROOM\s*=\s*({.+});', webpage, 'room info'), video_id)
 152
 153         video_info = None
 154
 155         for trial in range(5):
 156             # Sometimes Douyu rejects our request. Let's try it more times
 157             try:
 158                 video_info = self._download_json(
 159                     'https://vmobile.douyu.com/video/getInfo', video_id,
 160                     query={'vid': video_id},
 161                     headers={
 162                         'Referer': url,
 163                         'x-requested-with': 'XMLHttpRequest',
 164                     })
 165                 break
 166             except ExtractorError:
 167                 self._sleep(1, video_id)
 168
 169         if not video_info:
 170             raise ExtractorError('Can\'t fetch video info')
 171
 172         formats = self._extract_m3u8_formats(
 173             video_info['data']['video_url'], video_id,
 174             entry_protocol='m3u8_native', ext='mp4')
 175
 176         upload_date = unified_strdate(self._html_search_regex(
 177             r'<em>上传时间：</em><span>([^<]+)</span>', webpage,
 178             'upload date', fatal=False))
 179
 180         uploader = uploader_id = uploader_url = None
 181         mobj = re.search(
 182             r'(?m)<a[^>]+href="/author/([0-9a-zA-Z]+)".+?<strong[^>]+title="([^"]+)"',
 183             webpage)
 184         if mobj:
 185             uploader_id, uploader = mobj.groups()
 186             uploader_url = urljoin(url, '/author/' + uploader_id)
 187
 188         return {
 189             'id': video_id,
 190             'title': room_info['name'],
 191             'formats': formats,
 192             'duration': room_info.get('duration'),
 193             'thumbnail': room_info.get('pic'),
 194             'upload_date': upload_date,
 195             'uploader': uploader,
 196             'uploader_id': uploader_id,
 197             'uploader_url': uploader_url,
 198         }