New upstream version 2018.09.10

[youtubedl] / youtube_dl / extractor / dctp.py

diff --git a/youtube_dl/extractor/dctp.py b/youtube_dl/extractor/dctp.py
index 9099f5046a14ad7c769a6da50d813076f8b9231e..769a219dffe36ec73dab5d1e0fc18ab9bfc61993 100644 (file)
--- a/youtube_dl/extractor/dctp.py
+++ b/youtube_dl/extractor/dctp.py
@@ -1,61 +1,115 @@
-# encoding: utf-8
+# coding: utf-8

 from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..compat import compat_str
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
 from ..compat import compat_str

+from ..utils import (
+    float_or_none,
+    int_or_none,
+    unified_timestamp,
+    url_or_none,
+)

 
 
 class DctpTvIE(InfoExtractor):
 
 
 class DctpTvIE(InfoExtractor):

-    _VALID_URL = r'https?://www.dctp.tv/(#/)?filme/(?P<id>.+?)/$'
-    _TEST = {
+    _VALID_URL = r'https?://(?:www\.)?dctp\.tv/(?:#/)?filme/(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        # 4x3

         'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/',
         'info_dict': {
         'url': 'http://www.dctp.tv/filme/videoinstallation-fuer-eine-kaufhausfassade/',
         'info_dict': {

-            'id': '1324',
+            'id': '95eaa4f33dad413aa17b4ee613cccc6c',

             'display_id': 'videoinstallation-fuer-eine-kaufhausfassade',
             'ext': 'flv',
             'display_id': 'videoinstallation-fuer-eine-kaufhausfassade',
             'ext': 'flv',

-            'title': 'Videoinstallation für eine Kaufhausfassade'
+            'title': 'Videoinstallation für eine Kaufhausfassade',
+            'description': 'Kurzfilm',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 71.24,
+            'timestamp': 1302172322,
+            'upload_date': '20110407',

         },
         'params': {
             # rtmp download
             'skip_download': True,
         },
         'params': {
             # rtmp download
             'skip_download': True,

-        }
-    }
+        },
+    }, {
+        # 16x9
+        'url': 'http://www.dctp.tv/filme/sind-youtuber-die-besseren-lehrer/',
+        'only_matching': True,
+    }]
+
+    _BASE_URL = 'http://dctp-ivms2-restapi.s3.amazonaws.com'

 
     def _real_extract(self, url):
 
     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)
+
+        version = self._download_json(
+            '%s/version.json' % self._BASE_URL, display_id,
+            'Downloading version JSON')
+
+        restapi_base = '%s/%s/restapi' % (
+            self._BASE_URL, version['version_name'])
+
+        info = self._download_json(
+            '%s/slugs/%s.json' % (restapi_base, display_id), display_id,
+            'Downloading video info JSON')
+
+        media = self._download_json(
+            '%s/media/%s.json' % (restapi_base, compat_str(info['object_id'])),
+            display_id, 'Downloading media JSON')
+
+        uuid = media['uuid']
+        title = media['title']
+        ratio = '16x9' if media.get('is_wide') else '4x3'
+        play_path = 'mp4:%s_dctp_0500_%s.m4v' % (uuid, ratio)
+
+        servers = self._download_json(
+            'http://www.dctp.tv/streaming_servers/', display_id,
+            note='Downloading server list JSON', fatal=False)
+
+        if servers:
+            endpoint = next(
+                server['endpoint']
+                for server in servers
+                if url_or_none(server.get('endpoint')) and
+                'cloudfront' in server['endpoint'])

         else:
         else:

-            ratio = '4x3'
-        play_path = 'mp4:{0}_dctp_0500_{1}.m4v'.format(uuid, ratio)
+            endpoint = 'rtmpe://s2pqqn4u96e4j8.cloudfront.net/cfx/st/'

 
 

-        servers_json = self._download_json(
-            'http://www.dctp.tv/streaming_servers/',
-            video_id, note='Downloading server list')
-        url = servers_json[0]['endpoint']
+        app = self._search_regex(
+            r'^rtmpe?://[^/]+/(?P<app>.*)$', endpoint, 'app')

 
 

-        return {
-            'id': object_id,
-            'title': title,
-            'format': 'rtmp',
-            'url': url,
+        formats = [{
+            'url': endpoint,
+            'app': app,

             'play_path': play_path,
             'play_path': play_path,

-            'rtmp_real_time': True,
+            'page_url': url,
+            'player_url': 'http://svm-prod-dctptv-static.s3.amazonaws.com/dctptv-relaunch2012-110.swf',

             'ext': 'flv',
             'ext': 'flv',

-            'display_id': video_id
+        }]
+
+        thumbnails = []
+        images = media.get('images')
+        if isinstance(images, list):
+            for image in images:
+                if not isinstance(image, dict):
+                    continue
+                image_url = url_or_none(image.get('url'))
+                if not image_url:
+                    continue
+                thumbnails.append({
+                    'url': image_url,
+                    'width': int_or_none(image.get('width')),
+                    'height': int_or_none(image.get('height')),
+                })
+
+        return {
+            'id': uuid,
+            'display_id': display_id,
+            'title': title,
+            'alt_title': media.get('subtitle'),
+            'description': media.get('description') or media.get('teaser'),
+            'timestamp': unified_timestamp(media.get('created')),
+            'duration': float_or_none(media.get('duration_in_ms'), scale=1000),
+            'thumbnails': thumbnails,
+            'formats': formats,

         }
         }