]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/sina.py
Set upstream metadata fields: Bug-Database, Bug-Submit, Repository, Repository-Browse.
[youtubedl] / youtube_dl / extractor / sina.py
index 2909ef18b51a5ac6dadc4eec33bd05522443da8c..07b766b4a756f67d855ede90ebf72f0997da7e1d 100644 (file)
@@ -5,28 +5,34 @@ import re
 
 from .common import InfoExtractor
 from ..utils import (
-    compat_urllib_request,
-    compat_urllib_parse,
+    HEADRequest,
+    ExtractorError,
+    int_or_none,
+    update_url_query,
+    qualities,
+    get_element_by_attribute,
+    clean_html,
 )
 
 
 class SinaIE(InfoExtractor):
-    _VALID_URL = r'''https?://(.*?\.)?video\.sina\.com\.cn/
-                        (
-                            (.+?/(((?P<pseudo_id>\d+).html)|(.*?(\#|(vid=)|b/)(?P<id>\d+?)($|&|\-))))
-                            |
+    _VALID_URL = r'''(?x)https?://(?:.*?\.)?video\.sina\.com\.cn/
+                        (?:
+                            (?:view/|.*\#)(?P<video_id>\d+)|
+                            .+?/(?P<pseudo_id>[^/?#]+)(?:\.s?html)|
                             # This is used by external sites like Weibo
-                            (api/sinawebApi/outplay.php/(?P<token>.+?)\.swf)
+                            api/sinawebApi/outplay.php/(?P<token>.+?)\.swf
                         )
                   '''
 
     _TESTS = [
         {
-            'url': 'http://video.sina.com.cn/news/vlist/zt/chczlj2013/?opsubject_id=top12#110028898',
-            'file': '110028898.flv',
-            'md5': 'd65dd22ddcf44e38ce2bf58a10c3e71f',
+            'url': 'http://video.sina.com.cn/news/spj/topvideoes20160504/?opsubject_id=top1#250576622',
+            'md5': 'd38433e2fc886007729735650ae4b3e9',
             'info_dict': {
-                'title': '《中国新闻》 朝鲜要求巴拿马立即释放被扣船员',
+                'id': '250576622',
+                'ext': 'mp4',
+                'title': '现场:克鲁兹宣布退选 特朗普将稳获提名',
             }
         },
         {
@@ -36,41 +42,74 @@ class SinaIE(InfoExtractor):
                 'ext': 'flv',
                 'title': '军方提高对朝情报监视级别',
             },
+            'skip': 'the page does not exist or has been deleted',
+        },
+        {
+            'url': 'http://video.sina.com.cn/view/250587748.html',
+            'md5': '3d1807a25c775092aab3bc157fff49b4',
+            'info_dict': {
+                'id': '250587748',
+                'ext': 'mp4',
+                'title': '瞬间泪目:8年前汶川地震珍贵视频首曝光',
+            },
         },
     ]
 
-    @classmethod
-    def suitable(cls, url):
-        return re.match(cls._VALID_URL, url, flags=re.VERBOSE) is not None
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
 
-    def _extract_video(self, video_id):
-        data = compat_urllib_parse.urlencode({'vid': video_id})
-        url_doc = self._download_xml('http://v.iask.com/v_play.php?%s' % data,
-            video_id, 'Downloading video url')
-        image_page = self._download_webpage(
-            'http://interface.video.sina.com.cn/interface/common/getVideoImage.php?%s' % data,
-            video_id, 'Downloading thumbnail info')
+        video_id = mobj.group('video_id')
+        if not video_id:
+            if mobj.group('token') is not None:
+                # The video id is in the redirected url
+                self.to_screen('Getting video id')
+                request = HEADRequest(url)
+                _, urlh = self._download_webpage_handle(request, 'NA', False)
+                return self._real_extract(urlh.geturl())
+            else:
+                pseudo_id = mobj.group('pseudo_id')
+                webpage = self._download_webpage(url, pseudo_id)
+                error = get_element_by_attribute('class', 'errtitle', webpage)
+                if error:
+                    raise ExtractorError('%s said: %s' % (
+                        self.IE_NAME, clean_html(error)), expected=True)
+                video_id = self._search_regex(
+                    r"video_id\s*:\s*'(\d+)'", webpage, 'video id')
 
-        return {'id': video_id,
-                'url': url_doc.find('./durl/url').text,
-                'ext': 'flv',
-                'title': url_doc.find('./vname').text,
-                'thumbnail': image_page.split('=')[1],
-                }
+        video_data = self._download_json(
+            'http://s.video.sina.com.cn/video/h5play',
+            video_id, query={'video_id': video_id})
+        if video_data['code'] != 1:
+            raise ExtractorError('%s said: %s' % (
+                self.IE_NAME, video_data['message']), expected=True)
+        else:
+            video_data = video_data['data']
+            title = video_data['title']
+            description = video_data.get('description')
+            if description:
+                description = description.strip()
 
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
-        video_id = mobj.group('id')
-        if mobj.group('token') is not None:
-            # The video id is in the redirected url
-            self.to_screen('Getting video id')
-            request = compat_urllib_request.Request(url)
-            request.get_method = lambda: 'HEAD'
-            (_, urlh) = self._download_webpage_handle(request, 'NA', False)
-            return self._real_extract(urlh.geturl())
-        elif video_id is None:
-            pseudo_id = mobj.group('pseudo_id')
-            webpage = self._download_webpage(url, pseudo_id)
-            video_id = self._search_regex(r'vid:\'(\d+?)\'', webpage, 'video id')
+            preference = qualities(['cif', 'sd', 'hd', 'fhd', 'ffd'])
+            formats = []
+            for quality_id, quality in video_data.get('videos', {}).get('mp4', {}).items():
+                file_api = quality.get('file_api')
+                file_id = quality.get('file_id')
+                if not file_api or not file_id:
+                    continue
+                formats.append({
+                    'format_id': quality_id,
+                    'url': update_url_query(file_api, {'vid': file_id}),
+                    'preference': preference(quality_id),
+                    'ext': 'mp4',
+                })
+            self._sort_formats(formats)
 
-        return self._extract_video(video_id)
+            return {
+                'id': video_id,
+                'title': title,
+                'description': description,
+                'thumbnail': video_data.get('image'),
+                'duration': int_or_none(video_data.get('length')),
+                'timestamp': int_or_none(video_data.get('create_time')),
+                'formats': formats,
+            }