[douyutv] Simplify

This commit is contained in:
Yen Chi Hsuan 2016-08-27 01:16:39 +08:00
parent 545ef4f531
commit 3b4b82d4ce
No known key found for this signature in database
GPG Key ID: 3FDDD575826C5C30
1 changed files with 39 additions and 47 deletions

View File

@ -4,9 +4,16 @@ from __future__ import unicode_literals
import hashlib import hashlib
import time import time
import uuid import uuid
from .common import InfoExtractor from .common import InfoExtractor
from ..utils import (ExtractorError, unescapeHTML) from ..compat import (
from ..compat import (compat_str, compat_basestring, compat_urllib_parse_urlencode) compat_str,
compat_urllib_parse_urlencode,
)
from ..utils import (
ExtractorError,
unescapeHTML,
)
class DouyuTVIE(InfoExtractor): class DouyuTVIE(InfoExtractor):
@ -63,6 +70,10 @@ class DouyuTVIE(InfoExtractor):
'only_matching': True, 'only_matching': True,
}] }]
# Decompile core.swf in webpage by ffdec "Search SWFs in memory". core.swf
# is encrypted originally, but ffdec can dump memory to get the decrypted one.
_API_KEY = 'A12Svb&%1UUmf@hC'
def _real_extract(self, url): def _real_extract(self, url):
video_id = self._match_id(url) video_id = self._match_id(url)
@ -73,60 +84,41 @@ class DouyuTVIE(InfoExtractor):
room_id = self._html_search_regex( room_id = self._html_search_regex(
r'"room_id"\s*:\s*(\d+),', page, 'room id') r'"room_id"\s*:\s*(\d+),', page, 'room id')
room_url = 'http://m.douyu.com/html5/live?roomId=%s' % room_id room = self._download_json(
room_content = self._download_webpage(room_url, video_id) 'http://m.douyu.com/html5/live?roomId=%s' % room_id, video_id,
room_json = self._parse_json(room_content, video_id, fatal=False) note='Downloading room info')['data']
room = room_json['data']
show_status = room.get('show_status')
# 1 = live, 2 = offline # 1 = live, 2 = offline
if show_status == '2': if room.get('show_status') == '2':
raise ExtractorError( raise ExtractorError('Live stream is offline', expected=True)
'Live stream is offline', expected=True)
flv_json = None tt = compat_str(int(time.time() / 60))
# Douyu API sometimes returns error "Unable to load the requested class: eticket_redis_cache" did = uuid.uuid4().hex.upper()
# Retry with different parameters - same parameters cause same errors
for i in range(5):
tt = int(time.time() / 60)
did = uuid.uuid4().hex.upper()
# Decompile core.swf in webpage by ffdec "Search SWFs in memory" sign_content = ''.join((room_id, did, self._API_KEY, tt))
# core.swf is encrypted originally, but ffdec can dump memory to get the decrypted one sign = hashlib.md5((sign_content).encode('utf-8')).hexdigest()
# If API changes in the future, just use this way to update
sign_content = '{room_id}{did}A12Svb&%1UUmf@hC{tt}'.format(room_id = room_id, did = did, tt = tt)
sign = hashlib.md5((sign_content).encode('utf-8')).hexdigest()
payload = {'cdn': 'ws', 'rate': '0', 'tt': tt, 'did': did, 'sign': sign} flv_data = compat_urllib_parse_urlencode({
flv_data = compat_urllib_parse_urlencode(payload) 'cdn': 'ws',
'rate': '0',
'tt': tt,
'did': did,
'sign': sign,
})
flv_request_url = 'http://www.douyu.com/lapi/live/getPlay/%s' % room_id video_info = self._download_json(
flv_content = self._download_webpage(flv_request_url, video_id, data=flv_data, 'http://www.douyu.com/lapi/live/getPlay/%s' % room_id, video_id,
headers={'Content-Type': 'application/x-www-form-urlencoded'}) data=flv_data, note='Downloading video info',
try: headers={'Content-Type': 'application/x-www-form-urlencoded'})
flv_json = self._parse_json(flv_content, video_id, fatal=False)
except ExtractorError:
# Wait some time before retrying to get a different time() value
self._sleep(1, video_id, msg_template='%(video_id)s: Error occurs. '
'Waiting for %(timeout)s seconds before retrying')
continue
else:
break
if flv_json is None:
raise ExtractorError('Unable to fetch API result')
flv = flv_json['data'] error_code = video_info.get('error', 0)
error_code = flv_json.get('error', 0)
if error_code is not 0: if error_code is not 0:
error_desc = 'Server reported error %i' % error_code raise ExtractorError(
if isinstance(flv, (compat_str, compat_basestring)): '%s reported error %i' % (self.IE_NAME, error_code),
error_desc += ': ' + flv expected=True)
raise ExtractorError(error_desc, expected=True)
base_url = flv['rtmp_url'] base_url = video_info['data']['rtmp_url']
live_path = flv['rtmp_live'] live_path = video_info['data']['rtmp_live']
video_url = '%s/%s' % (base_url, live_path) video_url = '%s/%s' % (base_url, live_path)