Merge branch 'lecture2go' of https://github.com/nichdu/youtube-dl into nichdu-lecture2go

[youtube-dl] / youtube_dl / extractor / youku.py

diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py
index d5b73ebcef71ae52dc9c39367240a908249966a9..78caeb8b36e0be8cf4e97365d9e28251723059b7 100644 (file)
--- a/youtube_dl/extractor/youku.py
+++ b/youtube_dl/extractor/youku.py
@@ -9,11 +9,13 @@ from ..utils import ExtractorError
 from ..compat import (
     compat_urllib_parse,
     compat_ord,
+    compat_urllib_request,
 )
 
 
 class YoukuIE(InfoExtractor):
     IE_NAME = 'youku'
+    IE_DESC = '优酷'
     _VALID_URL = r'''(?x)
         (?:
             http://(?:v|player)\.youku\.com/(?:v_show/id_|player\.php/sid/)|
@@ -39,6 +41,14 @@ class YoukuIE(InfoExtractor):
             'title': '武媚娘传奇 85',
         },
         'playlist_count': 11,
+    }, {
+        'url': 'http://v.youku.com/v_show/id_XMTI1OTczNDM5Mg==.html',
+        'info_dict': {
+            'id': 'XMTI1OTczNDM5Mg',
+            'title': '花千骨 04',
+        },
+        'playlist_count': 13,
+        'skip': 'Available in China only',
     }]
 
     def construct_video_urls(self, data1, data2):
@@ -165,24 +175,35 @@ class YoukuIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        # request basic data
-        data1_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id
-        data2_url = 'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id
+        def retrieve_data(req_url, note):
+            req = compat_urllib_request.Request(req_url)
+
+            cn_verification_proxy = self._downloader.params.get('cn_verification_proxy')
+            if cn_verification_proxy:
+                req.add_header('Ytdl-request-proxy', cn_verification_proxy)
 
-        raw_data1 = self._download_json(data1_url, video_id)
-        raw_data2 = self._download_json(data2_url, video_id)
-        data1 = raw_data1['data'][0]
-        data2 = raw_data2['data'][0]
+            raw_data = self._download_json(req, video_id, note=note)
+            return raw_data['data'][0]
+
+        # request basic data
+        data1 = retrieve_data(
+            'http://v.youku.com/player/getPlayList/VideoIDS/%s' % video_id,
+            'Downloading JSON metadata 1')
+        data2 = retrieve_data(
+            'http://v.youku.com/player/getPlayList/VideoIDS/%s/Pf/4/ctype/12/ev/1' % video_id,
+            'Downloading JSON metadata 2')
 
         error_code = data1.get('error_code')
         if error_code:
-            # -8 means blocked outside China.
-            # Chinese and English, separated by newline.
             error = data1.get('error')
-            raise ExtractorError(
-                error or 'Server reported error %i' %
-                error_code,
-                expected=True)
+            if error is not None and '因版权原因无法观看此视频' in error:
+                raise ExtractorError(
+                    'Youku said: Sorry, this video is available in China only', expected=True)
+            else:
+                msg = 'Youku server reported error %i' % error_code
+                if error is not None:
+                    msg += ': ' + error
+                raise ExtractorError(msg)
 
         title = data1['title']
 
@@ -190,28 +211,22 @@ class YoukuIE(InfoExtractor):
         video_urls_dict = self.construct_video_urls(data1, data2)
 
         # construct info
-        entries = []
+        entries = [{
+            'id': '%s_part%d' % (video_id, i + 1),
+            'title': title,
+            'formats': [],
+            # some formats are not available for all parts, we have to detect
+            # which one has all
+        } for i in range(max(len(v) for v in data1['segs'].values()))]
         for fm in data1['streamtypes']:
             video_urls = video_urls_dict[fm]
-            for i in range(len(video_urls)):
-                if len(entries) < i + 1:
-                    entries.append({'formats': []})
-                entries[i]['formats'].append(
-                    {
-                        'url': video_urls[i],
-                        'format_id': self.get_format_name(fm),
-                        'ext': self.parse_ext_l(fm),
-                        'filesize': int(data1['segs'][fm][i]['size'])
-                    }
-                )
-
-        for i in range(len(entries)):
-            entries[i].update(
-                {
-                    'id': '%s_part%d' % (video_id, i + 1),
-                    'title': title,
-                }
-            )
+            for video_url, seg, entry in zip(video_urls, data1['segs'][fm], entries):
+                entry['formats'].append({
+                    'url': video_url,
+                    'format_id': self.get_format_name(fm),
+                    'ext': self.parse_ext_l(fm),
+                    'filesize': int(seg['size']),
+                })
 
         return {
             '_type': 'multi_video',