X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fdouyutv.py;h=82d8a042f58e738422a4c64b7fa90f8ac127f5fb;hb=54b960f340ed5398136ef0206d17cafba2575678;hp=d7956e6e4a20d733947f70a2933e06c2f0d144de;hpb=048fdc2292cc4c4add2fd3317a74d9f7d7ced227;p=youtube-dl

diff --git a/youtube_dl/extractor/douyutv.py b/youtube_dl/extractor/douyutv.py
index d7956e6e4..82d8a042f 100644
--- a/youtube_dl/extractor/douyutv.py
+++ b/youtube_dl/extractor/douyutv.py
@@ -1,77 +1,119 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import time
+import hashlib
+
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from ..utils import (
+    ExtractorError,
+    unescapeHTML,
+)
 
 
 class DouyuTVIE(InfoExtractor):
-    _VALID_URL = r'http://(?:www\.)?douyutv\.com/(?P<id>[A-Za-z0-9]+)'
-    _TEST = {
+    IE_DESC = '斗鱼'
+    _VALID_URL = r'https?://(?:www\.)?douyu(?:tv)?\.com/(?:[^/]+/)*(?P<id>[A-Za-z0-9]+)'
+    _TESTS = [{
         'url': 'http://www.douyutv.com/iseven',
         'info_dict': {
-            'id': 'iseven',
+            'id': '17732',
+            'display_id': 'iseven',
             'ext': 'flv',
-            'title': 're:^清晨醒脑！T-ara根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
-            'description': 'md5:9e525642c25a0a24302869937cf69d17',
-            'thumbnail': 're:^https?://.*\.jpg$',
+            'title': 're:^清晨醒脑！T-ARA根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'description': r're:.*m7show@163\.com.*',
+            'thumbnail': r're:^https?://.*\.jpg$',
             'uploader': '7师傅',
-            'uploader_id': '431925',
             'is_live': True,
         },
         'params': {
             'skip_download': True,
-        }
-    }
+        },
+    }, {
+        'url': 'http://www.douyutv.com/85982',
+        'info_dict': {
+            'id': '85982',
+            'display_id': '85982',
+            'ext': 'flv',
+            '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}$',
+            'description': 'md5:746a2f7a253966a06755a912f0acc0d2',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'douyu小漠',
+            'is_live': True,
+        },
+        'params': {
+            'skip_download': True,
+        },
+        'skip': 'Room not found',
+    }, {
+        'url': 'http://www.douyutv.com/17732',
+        'info_dict': {
+            'id': '17732',
+            'display_id': '17732',
+            'ext': 'flv',
+            'title': 're:^清晨醒脑！T-ARA根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'description': r're:.*m7show@163\.com.*',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': '7师傅',
+            'is_live': True,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }, {
+        'url': 'http://www.douyu.com/xiaocang',
+        'only_matching': True,
+    }, {
+        # \"room_id\"
+        'url': 'http://www.douyu.com/t/lpl',
+        'only_matching': True,
+    }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        config = self._download_json(
-            'http://www.douyutv.com/api/client/room/%s' % video_id, video_id)
-
-        data = config['data']
+        if video_id.isdigit():
+            room_id = video_id
+        else:
+            page = self._download_webpage(url, video_id)
+            room_id = self._html_search_regex(
+                r'"room_id\\?"\s*:\s*(\d+),', page, 'room id')
 
-        error_code = config.get('error', 0)
-        show_status = data.get('show_status')
-        if error_code is not 0:
-            raise ExtractorError(
-                'Server reported error %i' % error_code, expected=True)
+        # Grab metadata from mobile API
+        room = self._download_json(
+            'http://m.douyu.com/html5/live?roomId=%s' % room_id, video_id,
+            note='Downloading room info')['data']
 
         # 1 = live, 2 = offline
-        if show_status == '2':
-            raise ExtractorError(
-                'Live stream is offline', expected=True)
-
-        base_url = data['rtmp_url']
-        live_path = data['rtmp_live']
-
-        title = self._live_title(data['room_name'])
-        description = data.get('show_details')
-        thumbnail = data.get('room_src')
-
-        uploader = data.get('nickname')
-        uploader_id = data.get('owner_uid')
+        if room.get('show_status') == '2':
+            raise ExtractorError('Live stream is offline', expected=True)
 
-        multi_formats = data.get('rtmp_multi_bitrate')
-        if not isinstance(multi_formats, dict):
-            multi_formats = {}
-        multi_formats['live'] = live_path
+        # Grab the URL from PC client API
+        # The m3u8 url from mobile API requires re-authentication every 5 minutes
+        tt = int(time.time())
+        signContent = 'lapi/live/thirdPart/getPlay/%s?aid=pcclient&rate=0&time=%d9TUk5fjjUjg9qIMH3sdnh' % (room_id, tt)
+        sign = hashlib.md5(signContent.encode('ascii')).hexdigest()
+        video_url = self._download_json(
+            'http://coapi.douyucdn.cn/lapi/live/thirdPart/getPlay/' + room_id,
+            video_id, note='Downloading video URL info',
+            query={'rate': 0}, headers={
+                'auth': sign,
+                'time': str(tt),
+                'aid': 'pcclient'
+            })['data']['live_url']
 
-        formats = [{
-            'url': '%s/%s' % (base_url, format_path),
-            'format_id': format_id,
-            'preference': 1 if format_id == 'live' else 0,
-        } for format_id, format_path in multi_formats.items()]
-        self._sort_formats(formats)
+        title = self._live_title(unescapeHTML(room['room_name']))
+        description = room.get('show_details')
+        thumbnail = room.get('room_src')
+        uploader = room.get('nickname')
 
         return {
-            'id': video_id,
+            'id': room_id,
+            'display_id': video_id,
+            'url': video_url,
             'title': title,
             'description': description,
             'thumbnail': thumbnail,
             'uploader': uploader,
-            'uploader_id': uploader_id,
-            'formats': formats,
             'is_live': True,
         }