debian/watch: Skip pgpsigurlmangle for the moment.

[youtubedl] / youtube_dl / extractor / kaltura.py

diff --git a/youtube_dl/extractor/kaltura.py b/youtube_dl/extractor/kaltura.py
index d2873049202813be7587067b629503dc2da0f877..ccbc39c665412980e6b6104e83ffaf2e8574517f 100644 (file)
--- a/youtube_dl/extractor/kaltura.py
+++ b/youtube_dl/extractor/kaltura.py
@@ -2,23 +2,41 @@
 from __future__ import unicode_literals
 
 import re
 from __future__ import unicode_literals
 
 import re

+import base64

 
 from .common import InfoExtractor
 
 from .common import InfoExtractor

-from ..compat import compat_urllib_parse
+from ..compat import (
+    compat_urllib_parse,
+    compat_urlparse,
+)

 from ..utils import (
 from ..utils import (

+    clean_html,

     ExtractorError,
     int_or_none,
     ExtractorError,
     int_or_none,

+    unsmuggle_url,

 )
 
 
 class KalturaIE(InfoExtractor):
     _VALID_URL = r'''(?x)
 )
 
 
 class KalturaIE(InfoExtractor):
     _VALID_URL = r'''(?x)

-    (?:kaltura:|
-       https?://(:?(?:www|cdnapisec)\.)?kaltura\.com/index\.php/kwidget/(?:[^/]+/)*?wid/_
-    )(?P<partner_id>\d+)
-    (?::|
-       /(?:[^/]+/)*?entry_id/
-    )(?P<id>[0-9a-z_]+)'''
+                (?:
+                    kaltura:(?P<partner_id_s>\d+):(?P<id_s>[0-9a-z_]+)|
+                    https?://
+                        (:?(?:www|cdnapi(?:sec)?)\.)?kaltura\.com/
+                        (?:
+                            (?:
+                                # flash player
+                                index\.php/kwidget/
+                                (?:[^/]+/)*?wid/_(?P<partner_id>\d+)/
+                                (?:[^/]+/)*?entry_id/(?P<id>[0-9a-z_]+)|
+                                # html5 player
+                                html5/html5lib/
+                                (?:[^/]+/)*?entry_id/(?P<id_html5>[0-9a-z_]+)
+                                .*\?.*\bwid=_(?P<partner_id_html5>\d+)
+                            )
+                        )
+                )
+                '''

     _API_BASE = 'http://cdnapi.kaltura.com/api_v3/index.php?'
     _TESTS = [
         {
     _API_BASE = 'http://cdnapi.kaltura.com/api_v3/index.php?'
     _TESTS = [
         {


@@ -27,7 +45,7 @@ class KalturaIE(InfoExtractor):
             'info_dict': {
                 'id': '1_1jc2y3e4',
                 'ext': 'mp4',
             'info_dict': {
                 'id': '1_1jc2y3e4',
                 'ext': 'mp4',

-                'title': 'Track 4',
+                'title': 'Straight from the Heart',

                 'upload_date': '20131219',
                 'uploader_id': 'mlundberg@wolfgangsvault.com',
                 'description': 'The Allman Brothers Band, 12/16/1981',
                 'upload_date': '20131219',
                 'uploader_id': 'mlundberg@wolfgangsvault.com',
                 'description': 'The Allman Brothers Band, 12/16/1981',


@@ -43,6 +61,10 @@ class KalturaIE(InfoExtractor):
             'url': 'https://cdnapisec.kaltura.com/index.php/kwidget/wid/_557781/uiconf_id/22845202/entry_id/1_plr1syf3',
             'only_matching': True,
         },
             'url': 'https://cdnapisec.kaltura.com/index.php/kwidget/wid/_557781/uiconf_id/22845202/entry_id/1_plr1syf3',
             'only_matching': True,
         },

+        {
+            'url': 'https://cdnapisec.kaltura.com/html5/html5lib/v2.30.2/mwEmbedFrame.php/p/1337/uiconf_id/20540612/entry_id/1_sf5ovm7u?wid=_243342',
+            'only_matching': True,
+        }

     ]
 
     def _kaltura_api_call(self, video_id, actions, *args, **kwargs):
     ]
 
     def _kaltura_api_call(self, video_id, actions, *args, **kwargs):


@@ -93,43 +115,65 @@ class KalturaIE(InfoExtractor):
                 'version': '-1',
             },
             {
                 'version': '-1',
             },
             {

-                'action': 'getContextData',
-                'contextDataParams:objectType': 'KalturaEntryContextDataParams',
-                'contextDataParams:referrer': 'http://www.kaltura.com/',
-                'contextDataParams:streamerType': 'http',
+                'action': 'getbyentryid',

                 'entryId': video_id,
                 'entryId': video_id,

-                'service': 'baseentry',
+                'service': 'flavorAsset',

             },
         ]
         return self._kaltura_api_call(
             video_id, actions, note='Downloading video info JSON')
 
     def _real_extract(self, url):
             },
         ]
         return self._kaltura_api_call(
             video_id, actions, note='Downloading video info JSON')
 
     def _real_extract(self, url):

-        video_id = self._match_id(url)
+        url, smuggled_data = unsmuggle_url(url, {})
+

         mobj = re.match(self._VALID_URL, url)
         mobj = re.match(self._VALID_URL, url)

-        partner_id, entry_id = mobj.group('partner_id'), mobj.group('id')
-
-        info, source_data = self._get_video_info(entry_id, partner_id)
-
-        formats = [{
-            'format_id': '%(fileExt)s-%(bitrate)s' % f,
-            'ext': f['fileExt'],
-            'tbr': f['bitrate'],
-            'fps': f.get('frameRate'),
-            'filesize_approx': int_or_none(f.get('size'), invscale=1024),
-            'container': f.get('containerFormat'),
-            'vcodec': f.get('videoCodecId'),
-            'height': f.get('height'),
-            'width': f.get('width'),
-            'url': '%s/flavorId/%s' % (info['dataUrl'], f['id']),
-        } for f in source_data['flavorAssets']]
+        partner_id = mobj.group('partner_id_s') or mobj.group('partner_id') or mobj.group('partner_id_html5')
+        entry_id = mobj.group('id_s') or mobj.group('id') or mobj.group('id_html5')
+
+        info, flavor_assets = self._get_video_info(entry_id, partner_id)
+
+        source_url = smuggled_data.get('source_url')
+        if source_url:
+            referrer = base64.b64encode(
+                '://'.join(compat_urlparse.urlparse(source_url)[:2])
+                .encode('utf-8')).decode('utf-8')
+        else:
+            referrer = None
+
+        formats = []
+        for f in flavor_assets:
+            # Continue if asset is not ready
+            if f['status'] != 2:
+                continue
+            video_url = '%s/flavorId/%s' % (info['dataUrl'], f['id'])
+            if referrer:
+                video_url += '?referrer=%s' % referrer
+            formats.append({
+                'format_id': '%(fileExt)s-%(bitrate)s' % f,
+                'ext': f.get('fileExt'),
+                'tbr': int_or_none(f['bitrate']),
+                'fps': int_or_none(f.get('frameRate')),
+                'filesize_approx': int_or_none(f.get('size'), invscale=1024),
+                'container': f.get('containerFormat'),
+                'vcodec': f.get('videoCodecId'),
+                'height': int_or_none(f.get('height')),
+                'width': int_or_none(f.get('width')),
+                'url': video_url,
+            })
+        m3u8_url = info['dataUrl'].replace('format/url', 'format/applehttp')
+        if referrer:
+            m3u8_url += '?referrer=%s' % referrer
+        formats.extend(self._extract_m3u8_formats(
+            m3u8_url, entry_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False))
+
+        self._check_formats(formats, entry_id)

         self._sort_formats(formats)
 
         return {
         self._sort_formats(formats)
 
         return {

-            'id': video_id,
+            'id': entry_id,

             'title': info['name'],
             'formats': formats,
             'title': info['name'],
             'formats': formats,

-            'description': info.get('description'),
+            'description': clean_html(info.get('description')),

             'thumbnail': info.get('thumbnailUrl'),
             'duration': info.get('duration'),
             'timestamp': info.get('createdAt'),
             'thumbnail': info.get('thumbnailUrl'),
             'duration': info.get('duration'),
             'timestamp': info.get('createdAt'),