diff options
| author | Yen Chi Hsuan <yan12125@gmail.com> | 2016-08-27 01:17:12 +0800 | 
|---|---|---|
| committer | Yen Chi Hsuan <yan12125@gmail.com> | 2016-08-27 01:17:12 +0800 | 
| commit | d181cff685aa889864a0356fa546e2d5605f27d2 (patch) | |
| tree | e08897e842976b75c44a5986214858859548e5da | |
| parent | 906b87cf5f6ccf28ebd75d6a92367d7c238f2ad9 (diff) | |
| parent | 3b4b82d4cec702fc06e2d6b38a44dd0c7bd77a5b (diff) | |
Merge branch 'steven7851-patch-2'
| -rw-r--r-- | youtube_dl/extractor/douyutv.py | 109 | 
1 files changed, 50 insertions, 59 deletions
diff --git a/youtube_dl/extractor/douyutv.py b/youtube_dl/extractor/douyutv.py index ce6962755..e366e17e6 100644 --- a/youtube_dl/extractor/douyutv.py +++ b/youtube_dl/extractor/douyutv.py @@ -3,9 +3,17 @@ from __future__ import unicode_literals  import hashlib  import time +import uuid +  from .common import InfoExtractor -from ..utils import (ExtractorError, unescapeHTML) -from ..compat import (compat_str, compat_basestring) +from ..compat import ( +    compat_str, +    compat_urllib_parse_urlencode, +) +from ..utils import ( +    ExtractorError, +    unescapeHTML, +)  class DouyuTVIE(InfoExtractor): @@ -21,7 +29,6 @@ class DouyuTVIE(InfoExtractor):              'description': 're:.*m7show@163\.com.*',              'thumbnail': 're:^https?://.*\.jpg$',              'uploader': '7师傅', -            'uploader_id': '431925',              'is_live': True,          },          'params': { @@ -37,7 +44,6 @@ class DouyuTVIE(InfoExtractor):              'description': 'md5:746a2f7a253966a06755a912f0acc0d2',              'thumbnail': 're:^https?://.*\.jpg$',              'uploader': 'douyu小漠', -            'uploader_id': '3769985',              'is_live': True,          },          'params': { @@ -54,7 +60,6 @@ class DouyuTVIE(InfoExtractor):              'description': 're:.*m7show@163\.com.*',              'thumbnail': 're:^https?://.*\.jpg$',              'uploader': '7师傅', -            'uploader_id': '431925',              'is_live': True,          },          'params': { @@ -65,6 +70,10 @@ class DouyuTVIE(InfoExtractor):          '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):          video_id = self._match_id(url) @@ -75,74 +84,56 @@ class DouyuTVIE(InfoExtractor):              room_id = self._html_search_regex(                  r'"room_id"\s*:\s*(\d+),', page, 'room id') -        config = None -        # Douyu API sometimes returns error "Unable to load the requested class: eticket_redis_cache" -        # Retry with different parameters - same parameters cause same errors -        for i in range(5): -            prefix = 'room/%s?aid=android&client_sys=android&time=%d' % ( -                room_id, int(time.time())) -            auth = hashlib.md5((prefix + '1231').encode('ascii')).hexdigest() - -            config_page = self._download_webpage( -                'http://www.douyutv.com/api/v1/%s&auth=%s' % (prefix, auth), -                video_id) -            try: -                config = self._parse_json(config_page, 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 config is None: -            raise ExtractorError('Unable to fetch API result') - -        data = config['data'] - -        error_code = config.get('error', 0) -        if error_code is not 0: -            error_desc = 'Server reported error %i' % error_code -            if isinstance(data, (compat_str, compat_basestring)): -                error_desc += ': ' + data -            raise ExtractorError(error_desc, expected=True) +        room = self._download_json( +            'http://m.douyu.com/html5/live?roomId=%s' % room_id, video_id, +            note='Downloading room info')['data'] -        show_status = data.get('show_status')          # 1 = live, 2 = offline -        if show_status == '2': -            raise ExtractorError( -                'Live stream is offline', expected=True) +        if room.get('show_status') == '2': +            raise ExtractorError('Live stream is offline', expected=True) + +        tt = compat_str(int(time.time() / 60)) +        did = uuid.uuid4().hex.upper() -        base_url = data['rtmp_url'] -        live_path = data['rtmp_live'] +        sign_content = ''.join((room_id, did, self._API_KEY, tt)) +        sign = hashlib.md5((sign_content).encode('utf-8')).hexdigest() -        title = self._live_title(unescapeHTML(data['room_name'])) -        description = data.get('show_details') -        thumbnail = data.get('room_src') +        flv_data = compat_urllib_parse_urlencode({ +            'cdn': 'ws', +            'rate': '0', +            'tt': tt, +            'did': did, +            'sign': sign, +        }) + +        video_info = self._download_json( +            'http://www.douyu.com/lapi/live/getPlay/%s' % room_id, video_id, +            data=flv_data, note='Downloading video info', +            headers={'Content-Type': 'application/x-www-form-urlencoded'}) + +        error_code = video_info.get('error', 0) +        if error_code is not 0: +            raise ExtractorError( +                '%s reported error %i' % (self.IE_NAME, error_code), +                expected=True) -        uploader = data.get('nickname') -        uploader_id = data.get('owner_uid') +        base_url = video_info['data']['rtmp_url'] +        live_path = video_info['data']['rtmp_live'] -        multi_formats = data.get('rtmp_multi_bitrate') -        if not isinstance(multi_formats, dict): -            multi_formats = {} -        multi_formats['live'] = live_path +        video_url = '%s/%s' % (base_url, live_path) -        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('notice') +        thumbnail = room.get('room_src') +        uploader = room.get('nickname')          return {              '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,          }  | 
