Import Upstream version 2020.01.24

[youtubedl] / youtube_dl / extractor / cloudflarestream.py

diff --git a/youtube_dl/extractor/cloudflarestream.py b/youtube_dl/extractor/cloudflarestream.py
index 8ff2c6531570ee3a06210832cf967c7033d6ab9f..2fdcfbb3af1fbffb9e66abff56b86e31762ad449 100644 (file)
--- a/youtube_dl/extractor/cloudflarestream.py
+++ b/youtube_dl/extractor/cloudflarestream.py
@@ -1,20 +1,24 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import base64
 import re
 
 from .common import InfoExtractor
 
 
 class CloudflareStreamIE(InfoExtractor):
+    _DOMAIN_RE = r'(?:cloudflarestream\.com|(?:videodelivery|bytehighway)\.net)'
+    _EMBED_RE = r'embed\.%s/embed/[^/]+\.js\?.*?\bvideo=' % _DOMAIN_RE
+    _ID_RE = r'[\da-f]{32}|[\w-]+\.[\w-]+\.[\w-]+'
     _VALID_URL = r'''(?x)
                     https?://
                         (?:
-                            (?:watch\.)?(?:cloudflarestream\.com|videodelivery\.net)/|
-                            embed\.(?:cloudflarestream\.com|videodelivery\.net)/embed/[^/]+\.js\?.*?\bvideo=
+                            (?:watch\.)?%s/|
+                            %s
                         )
-                        (?P<id>[\da-f]+)
-                    '''
+                        (?P<id>%s)
+                    ''' % (_DOMAIN_RE, _EMBED_RE, _ID_RE)
     _TESTS = [{
         'url': 'https://embed.cloudflarestream.com/embed/we4g.fla9.latest.js?video=31c9291ab41fac05471db4e73aa11717',
         'info_dict': {
@@ -41,23 +45,28 @@ class CloudflareStreamIE(InfoExtractor):
         return [
             mobj.group('url')
             for mobj in re.finditer(
-                r'<script[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//embed\.(?:cloudflarestream\.com|videodelivery\.net)/embed/[^/]+\.js\?.*?\bvideo=[\da-f]+?.*?)\1',
+                r'<script[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//%s(?:%s).*?)\1' % (CloudflareStreamIE._EMBED_RE, CloudflareStreamIE._ID_RE),
                 webpage)]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
+        domain = 'bytehighway.net' if 'bytehighway.net/' in url else 'videodelivery.net'
+        base_url = 'https://%s/%s/' % (domain, video_id)
+        if '.' in video_id:
+            video_id = self._parse_json(base64.urlsafe_b64decode(
+                video_id.split('.')[1]), video_id)['sub']
+        manifest_base_url = base_url + 'manifest/video.'
 
         formats = self._extract_m3u8_formats(
-            'https://cloudflarestream.com/%s/manifest/video.m3u8' % video_id,
-            video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls',
-            fatal=False)
+            manifest_base_url + 'm3u8', video_id, 'mp4',
+            'm3u8_native', m3u8_id='hls', fatal=False)
         formats.extend(self._extract_mpd_formats(
-            'https://cloudflarestream.com/%s/manifest/video.mpd' % video_id,
-            video_id, mpd_id='dash', fatal=False))
+            manifest_base_url + 'mpd', video_id, mpd_id='dash', fatal=False))
         self._sort_formats(formats)
 
         return {
             'id': video_id,
             'title': video_id,
+            'thumbnail': base_url + 'thumbnails/thumbnail.jpg',
             'formats': formats,
         }