X-Git-Url: http://git.bitcoin.ninja/index.cgi?a=blobdiff_plain;f=youtube_dl%2Fextractor%2Fweibo.py;h=b835f8975c528f94fa884553ed5c5316f9d75330;hb=0c69958844a446bc3373f45f8f750cbc3202d14e;hp=0757495bd8a11e22c98c9307d734f9e83529a631;hpb=6d69d03bac08f8381031de721167103697bf3fed;p=youtube-dl

diff --git a/youtube_dl/extractor/weibo.py b/youtube_dl/extractor/weibo.py
index 0757495bd..b835f8975 100644
--- a/youtube_dl/extractor/weibo.py
+++ b/youtube_dl/extractor/weibo.py
@@ -1,48 +1,105 @@
 # coding: utf-8
-
-import re
-import json
+from __future__ import unicode_literals
 
 from .common import InfoExtractor
 
-class WeiboIE(InfoExtractor):
-    """
-    The videos in Weibo come from different sites, this IE just finds the link
-    to the external video and returns it.
-    """
-    _VALID_URL = r'https?://video\.weibo\.com/v/weishipin/t_(?P<id>.+?)\.htm'
+from urllib.request import Request
+from urllib.parse import urlencode
+from urllib import parse
+import json
+import random as rnd
+from os import path
 
+class WeiboIE(InfoExtractor):
+    _VALID_URL = r'https?://weibo\.com/[0-9]+/(?P<id>[a-zA-Z0-9]+)'
     _TEST = {
-        u'url': u'http://video.weibo.com/v/weishipin/t_zjUw2kZ.htm',
-        u'file': u'98322879.flv',
-        u'info_dict': {
-            u'title': u'魔声耳机最新广告“All Eyes On Us”',
-        },
-        u'note': u'Sina video',
-        u'params': {
-            u'skip_download': True,
-        },
-    }
-
-    # Additional example videos from different sites
-    # Youku: http://video.weibo.com/v/weishipin/t_zQGDWQ8.htm
-    # 56.com: http://video.weibo.com/v/weishipin/t_zQ44HxN.htm
+            'url': 'https://weibo.com/6275294458/Fp6RGfbff?type=comment',
+            'info_dict': {
+                'id': 'Fp6RGfbff',
+                'ext': 'mp4',
+                'title': 'You should have servants to massage you,... 来自Hosico_猫 - 微博',
+                }
+            }
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url, flags=re.VERBOSE)
-        video_id = mobj.group('id')
-        info_url = 'http://video.weibo.com/?s=v&a=play_list&format=json&mix_video_id=t_%s' % video_id
-        info_page = self._download_webpage(info_url, video_id)
-        info = json.loads(info_page)
-
-        videos_urls = map(lambda v: v['play_page_url'], info['result']['data'])
-        #Prefer sina video since they have thumbnails
-        videos_urls = sorted(videos_urls, key=lambda u: u'video.sina.com' in u)
-        player_url = videos_urls[-1]
-        m_sina = re.match(r'https?://video.sina.com.cn/v/b/(\d+)-\d+.html', player_url)
-        if m_sina is not None:
-            self.to_screen('Sina video detected')
-            sina_id = m_sina.group(1)
-            player_url = 'http://you.video.sina.com.cn/swf/quotePlayer.swf?vid=%s' % sina_id
-        return self.url_result(player_url)
+        video_id = self._match_id(url)
+        headers = {
+            'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8',
+            'Accept-Encoding': 'gzip, deflate, br',
+            'Accept-Language': 'en,zh-CN;q=0.9,zh;q=0.8',
+            'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_13_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/63.0.3239.84 Safari/537.36',
+            'Upgrade-Insecure-Requests': '1',
+        }
+        # to get Referer url for genvisitor
+        webpage,urlh = self._download_webpage_handle(url, video_id, headers=headers, note="first visit the page")
+
+        visitor_url = urlh.geturl()
+
+        data = urlencode({
+            "cb": "gen_callback",
+            "fp": '{"os":"2","browser":"Gecko57,0,0,0","fonts":"undefined","screenInfo":"1440*900*24","plugins":""}',
+            }).encode()
+        headers = {
+                'Accept-Encoding': 'gzip, deflate, br',
+                'Accept': '*/*',
+                'Referer': visitor_url,
+                }
+
+        r_genvisitor = Request(
+            'https://passport.weibo.com/visitor/genvisitor',
+            data = data,
+            headers = headers,
+            method = 'POST'
+            )
+        webpage,urlh = self._download_webpage_handle(r_genvisitor, video_id, note="gen visitor")
+
+        p = webpage.split("&&")[1] # split "gen_callback && gen_callback(...)"
+        i1 = p.find('{')
+        i2 = p.rfind('}')
+        j = p[i1:i2+1] # get JSON object
+        d = json.loads(j)
+        tid = d["data"]["tid"]
+        cnfd = "%03d" % d["data"]["confidence"]
+
+        param = urlencode({
+            'a': 'incarnate',
+            't': tid,
+            'w': 2,
+            'c': cnfd,
+            'cb': 'cross_domain',
+            'from': 'weibo',
+            '_rand': rnd.random()
+            })
+        gencallback_url = "https://passport.weibo.com/visitor/visitor?" + param
+        webpage,urlh = self._download_webpage_handle(gencallback_url, video_id, note="gen callback")
+
+        webpage,urlh = self._download_webpage_handle(url, video_id, headers=headers, note="retry to visit the page")
+
+        # TODO more code goes here, for example ...
+        title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title')
+
+        video_sources_text = self._search_regex("video-sources=\\\\\"(.+?)\"", webpage, 'video_sources')
+        
+        video_formats = parse.parse_qs(video_sources_text)
 
+        formats = []
+        supported_resolutions = ['720', '480']
+        for res in supported_resolutions:
+            f = video_formats.get(res)
+            if isinstance(f, list):
+                if len(f) > 0:
+                    vid_url = f[0]
+                    formats.append({
+                        'url': vid_url,
+                        'format': 'mp4',
+                        'height': int(res),
+                        })
+        self._sort_formats(formats)
+        uploader = self._og_search_property('nick-name', webpage, 'uploader', default = None)
+        return {
+                'id': video_id,
+                'title': title,
+                'uploader': uploader,
+                'formats': formats
+                # TODO more properties (see youtube_dl/extractor/common.py)
+                }