Adapt changes from Andreas Tille for porting to Salsa.

[youtubedl] / youtube_dl / extractor / dctp.py

diff --git a/youtube_dl/extractor/dctp.py b/youtube_dl/extractor/dctp.py
index aa2c09eb686f9da5a7bedfdfe57566e9d29a0700..3a6d0560e478cb08445f07d84578d28b3e4bc514 100644 (file)
--- a/youtube_dl/extractor/dctp.py
+++ b/youtube_dl/extractor/dctp.py
@@ -1,61 +1,86 @@
-# encoding: utf-8
+# coding: utf-8
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..compat import compat_str
+from ..utils import (
+    float_or_none,
+    unified_strdate,
+)
 
 
 class DctpTvIE(InfoExtractor):
-    _VALID_URL = r'http://www.dctp.tv/(#/)?filme/(?P<id>.+?)/$'
+    _VALID_URL = r'https?://(?:www\.)?dctp\.tv/(?:#/)?filme/(?P<id>[^/?#&]+)'
     _TEST = {
         'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/',
         'info_dict': {
-            'id': '1324',
+            'id': '95eaa4f33dad413aa17b4ee613cccc6c',
             'display_id': 'videoinstallation-fuer-eine-kaufhausfassade',
             'ext': 'flv',
-            'title': 'Videoinstallation für eine Kaufhausfassade'
+            'title': 'Videoinstallation für eine Kaufhausfassade',
+            'description': 'Kurzfilm',
+            'upload_date': '20110407',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 71.24,
         },
         'params': {
             # rtmp download
             'skip_download': True,
-        }
+        },
     }
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        base_url = 'http://dctp-ivms2-restapi.s3.amazonaws.com/'
-        version_json = self._download_json(
-            base_url + 'version.json',
-            video_id, note='Determining file version')
-        version = version_json['version_name']
-        info_json = self._download_json(
-            '{0}{1}/restapi/slugs/{2}.json'.format(base_url, version, video_id),
-            video_id, note='Fetching object ID')
-        object_id = compat_str(info_json['object_id'])
-        meta_json = self._download_json(
-            '{0}{1}/restapi/media/{2}.json'.format(base_url, version, object_id),
-            video_id, note='Downloading metadata')
-        uuid = meta_json['uuid']
-        title = meta_json['title']
-        wide = meta_json['is_wide']
-        if wide:
-            ratio = '16x9'
+        display_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, display_id)
+
+        video_id = self._html_search_meta(
+            'DC.identifier', webpage, 'video id',
+            default=None) or self._search_regex(
+            r'id=["\']uuid[^>]+>([^<]+)<', webpage, 'video id')
+
+        title = self._og_search_title(webpage)
+
+        servers = self._download_json(
+            'http://www.dctp.tv/streaming_servers/', display_id,
+            note='Downloading server list', fatal=False)
+
+        if servers:
+            endpoint = next(
+                server['endpoint']
+                for server in servers
+                if isinstance(server.get('endpoint'), compat_str) and
+                'cloudfront' in server['endpoint'])
         else:
-            ratio = '4x3'
-        play_path = 'mp4:{0}_dctp_0500_{1}.m4v'.format(uuid, ratio)
+            endpoint = 'rtmpe://s2pqqn4u96e4j8.cloudfront.net/cfx/st/'
+
+        app = self._search_regex(
+            r'^rtmpe?://[^/]+/(?P<app>.*)$', endpoint, 'app')
 
-        servers_json = self._download_json(
-            'http://www.dctp.tv/streaming_servers/',
-            video_id, note='Downloading server list')
-        url = servers_json[0]['endpoint']
+        formats = [{
+            'url': endpoint,
+            'app': app,
+            'play_path': 'mp4:%s_dctp_0500_4x3.m4v' % video_id,
+            'page_url': url,
+            'player_url': 'http://svm-prod-dctptv-static.s3.amazonaws.com/dctptv-relaunch2012-109.swf',
+            'ext': 'flv',
+        }]
+
+        description = self._html_search_meta('DC.description', webpage)
+        upload_date = unified_strdate(
+            self._html_search_meta('DC.date.created', webpage))
+        thumbnail = self._og_search_thumbnail(webpage)
+        duration = float_or_none(self._search_regex(
+            r'id=["\']duration_in_ms[^+]>(\d+)', webpage, 'duration',
+            default=None), scale=1000)
 
         return {
-            'id': object_id,
+            'id': video_id,
             'title': title,
-            'format': 'rtmp',
-            'url': url,
-            'play_path': play_path,
-            'rtmp_real_time': True,
-            'ext': 'flv',
-            'display_id': video_id
+            'formats': formats,
+            'display_id': display_id,
+            'description': description,
+            'upload_date': upload_date,
+            'thumbnail': thumbnail,
+            'duration': duration,
         }