]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/sohu.py
Prepare for release.
[youtubedl] / youtube_dl / extractor / sohu.py
index 77bb0a8dc3b5555e28eda831ca11c2d5eff96f79..c04791997f3672cdb643870086c7ed7f52db54c1 100644 (file)
 # encoding: utf-8
+from __future__ import unicode_literals
 
-import json
 import re
 
 from .common import InfoExtractor
-from ..utils import ExtractorError
+from .common import compat_str
 
 
 class SohuIE(InfoExtractor):
-    _VALID_URL = r'https?://tv\.sohu\.com/\d+?/n(?P<id>\d+)\.shtml.*?'
+    _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
 
     _TEST = {
-        u'url': u'http://tv.sohu.com/20130724/n382479172.shtml#super',
-        u'file': u'382479172.mp4',
-        u'md5': u'bde8d9a6ffd82c63a1eefaef4eeefec7',
-        u'info_dict': {
-            u'title': u'MV:Far East Movement《The Illest》',
+        'url': 'http://tv.sohu.com/20130724/n382479172.shtml#super',
+        'md5': 'bde8d9a6ffd82c63a1eefaef4eeefec7',
+        'info_dict': {
+            'id': '382479172',
+            'ext': 'mp4',
+            'title': 'MV:Far East Movement《The Illest》',
         },
+        'skip': 'Only available from China',
     }
 
     def _real_extract(self, url):
 
-        def _fetch_data(vid_id):
-            base_data_url = u'http://hot.vrs.sohu.com/vrs_flash.action?vid='
-            data_url = base_data_url + str(vid_id)
-            data_json = self._download_webpage(
-                data_url, video_id,
-                note=u'Downloading JSON data for ' + str(vid_id))
-            return json.loads(data_json)
+        def _fetch_data(vid_id, mytv=False):
+            if mytv:
+                base_data_url = 'http://my.tv.sohu.com/play/videonew.do?vid='
+            else:
+                base_data_url = 'http://hot.vrs.sohu.com/vrs_flash.action?vid='
+
+            return self._download_json(
+                base_data_url + vid_id, video_id,
+                'Downloading JSON data for %s' % vid_id)
 
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
+        mytv = mobj.group('mytv') is not None
 
         webpage = self._download_webpage(url, video_id)
-        raw_title = self._html_search_regex(r'(?s)<title>(.+?)</title>',
-                                            webpage, u'video title')
+        raw_title = self._html_search_regex(
+            r'(?s)<title>(.+?)</title>',
+            webpage, 'video title')
         title = raw_title.partition('-')[0].strip()
 
-        vid = self._html_search_regex(r'var vid="(\d+)"', webpage,
-                                      u'video path')
-        data = _fetch_data(vid)
-
-        QUALITIES = ('ori', 'super', 'high', 'nor')
-        vid_ids = [data['data'][q + 'Vid']
-                   for q in QUALITIES
-                   if data['data'][q + 'Vid'] != 0]
-        if not vid_ids:
-            raise ExtractorError(u'No formats available for this video')
+        vid = self._html_search_regex(
+            r'var vid ?= ?["\'](\d+)["\']',
+            webpage, 'video path')
+        vid_data = _fetch_data(vid, mytv)
 
-        # For now, we just pick the highest available quality
-        vid_id = vid_ids[-1]
+        formats_json = {}
+        for format_id in ('nor', 'high', 'super', 'ori', 'h2644k', 'h2654k'):
+            vid_id = vid_data['data'].get('%sVid' % format_id)
+            if not vid_id:
+                continue
+            vid_id = compat_str(vid_id)
+            formats_json[format_id] = vid_data if vid == vid_id else _fetch_data(vid_id, mytv)
 
-        format_data = data if vid == vid_id else _fetch_data(vid_id)
-        part_count = format_data['data']['totalBlocks']
-        allot = format_data['allot']
-        prot = format_data['prot']
-        clipsURL = format_data['data']['clipsURL']
-        su = format_data['data']['su']
+        part_count = vid_data['data']['totalBlocks']
 
         playlist = []
         for i in range(part_count):
-            part_url = ('http://%s/?prot=%s&file=%s&new=%s' %
-                        (allot, prot, clipsURL[i], su[i]))
-            part_str = self._download_webpage(
-                part_url, video_id,
-                note=u'Downloading part %d of %d' % (i+1, part_count))
-
-            part_info = part_str.split('|')
-            video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3])
-
-            video_info = {
-                'id': '%s_part%02d' % (video_id, i + 1),
+            formats = []
+            for format_id, format_data in formats_json.items():
+                allot = format_data['allot']
+                prot = format_data['prot']
+
+                data = format_data['data']
+                clips_url = data['clipsURL']
+                su = data['su']
+
+                part_str = self._download_webpage(
+                    'http://%s/?prot=%s&file=%s&new=%s' %
+                    (allot, prot, clips_url[i], su[i]),
+                    video_id,
+                    'Downloading %s video URL part %d of %d'
+                    % (format_id, i + 1, part_count))
+
+                part_info = part_str.split('|')
+                video_url = '%s%s?key=%s' % (part_info[0], su[i], part_info[3])
+
+                formats.append({
+                    'url': video_url,
+                    'format_id': format_id,
+                    'filesize': data['clipsBytes'][i],
+                    'width': data['width'],
+                    'height': data['height'],
+                    'fps': data['fps'],
+                })
+            self._sort_formats(formats)
+
+            playlist.append({
+                'id': '%s_part%d' % (video_id, i + 1),
                 'title': title,
-                'url': video_url,
-                'ext': 'mp4',
-            }
-            playlist.append(video_info)
+                'duration': vid_data['data']['clipsDuration'][i],
+                'formats': formats,
+            })
 
         if len(playlist) == 1:
             info = playlist[0]