[internetvideoarchive] fix extraction

youtube_dl/extractor/internetvideoarchive.py

@@ -1,15 +1,13 @@
 from __future__ import unicode_literals
 
+import json
+import re
+
 from .common import InfoExtractor
 from ..compat import (
     compat_parse_qs,
     compat_urlparse,
 )
-from ..utils import (
-    determine_ext,
-    int_or_none,
-    xpath_text,
-)
 
 
 class InternetVideoArchiveIE(InfoExtractor):
@@ -20,7 +18,7 @@ class InternetVideoArchiveIE(InfoExtractor):
         'info_dict': {
             'id': '194487',
             'ext': 'mp4',
-            'title': 'KICK-ASS 2',
+            'title': 'Kick-Ass 2',
             'description': 'md5:c189d5b7280400630a1d3dd17eaa8d8a',
         },
         'params': {
@@ -33,68 +31,34 @@ class InternetVideoArchiveIE(InfoExtractor):
     def _build_json_url(query):
         return 'http://video.internetvideoarchive.net/player/6/configuration.ashx?' + query
 
-    @staticmethod
-    def _build_xml_url(query):
-        return 'http://video.internetvideoarchive.net/flash/players/flashconfiguration.aspx?' + query
-
     def _real_extract(self, url):
-        query = compat_urlparse.urlparse(url).query
-        query_dic = compat_parse_qs(query)
-        video_id = query_dic['publishedid'][0]
-
-        if '/player/' in url:
-            configuration = self._download_json(url, video_id)
-
-            # There are multiple videos in the playlist whlie only the first one
-            # matches the video played in browsers
-            video_info = configuration['playlist'][0]
-            title = video_info['title']
-
-            formats = []
-            for source in video_info['sources']:
-                file_url = source['file']
-                if determine_ext(file_url) == 'm3u8':
-                    m3u8_formats = self._extract_m3u8_formats(
-                        file_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)
-                    if m3u8_formats:
-                        formats.extend(m3u8_formats)
-                        file_url = m3u8_formats[0]['url']
+        query = compat_parse_qs(compat_urlparse.urlparse(url).query)
+        video_id = query['publishedid'][0]
+        data = self._download_json(
+            'https://video.internetvideoarchive.net/videojs7/videojs7.ivasettings.ashx',
+            video_id, data=json.dumps({
+                'customerid': query['customerid'][0],
+                'publishedid': video_id,
+            }).encode())
+        title = data['Title']
+        formats = self._extract_m3u8_formats(
+            data['VideoUrl'], video_id, 'mp4',
+            'm3u8_native', m3u8_id='hls', fatal=False)
+        file_url = formats[0]['url']
+        if '.ism/' in file_url:
+            replace_url = lambda x: re.sub(r'\.ism/[^?]+', '.ism/' + x, file_url)
             formats.extend(self._extract_f4m_formats(
-                            file_url.replace('.m3u8', '.f4m'),
-                            video_id, f4m_id='hds', fatal=False))
+                replace_url('.f4m'), video_id, f4m_id='hds', fatal=False))
             formats.extend(self._extract_mpd_formats(
-                            file_url.replace('.m3u8', '.mpd'),
-                            video_id, mpd_id='dash', fatal=False))
-                else:
-                    a_format = {
-                        'url': file_url,
-                    }
-
-                    if source.get('label') and source['label'][-4:] == ' kbs':
-                        tbr = int_or_none(source['label'][:-4])
-                        a_format.update({
-                            'tbr': tbr,
-                            'format_id': 'http-%d' % tbr,
-                        })
-                        formats.append(a_format)
-
+                replace_url('.mpd'), video_id, mpd_id='dash', fatal=False))
+            formats.extend(self._extract_ism_formats(
+                replace_url('Manifest'), video_id, ism_id='mss', fatal=False))
         self._sort_formats(formats)
 
-            description = video_info.get('description')
-            thumbnail = video_info.get('image')
-        else:
-            configuration = self._download_xml(url, video_id)
-            formats = [{
-                'url': xpath_text(configuration, './file', 'file URL', fatal=True),
-            }]
-            thumbnail = xpath_text(configuration, './image', 'thumbnail')
-            title = 'InternetVideoArchive video %s' % video_id
-            description = None
-
         return {
             'id': video_id,
             'title': title,
             'formats': formats,
-            'thumbnail': thumbnail,
-            'description': description,
+            'thumbnail': data.get('PosterUrl'),
+            'description': data.get('Description'),
         }