[youtube] Unescape HTML for series (closes #18641)

[youtube-dl] / youtube_dl / extractor / crackle.py

diff --git a/youtube_dl/extractor/crackle.py b/youtube_dl/extractor/crackle.py
index fc014f8b558008f971448153c56fb45354ee641f..f73ef6b63c8bf7a138e8dcaf8a35fc2798322aea 100644 (file)
--- a/youtube_dl/extractor/crackle.py
+++ b/youtube_dl/extractor/crackle.py
@@ -4,23 +4,21 @@ from __future__ import unicode_literals, division
 import re
 
 from .common import InfoExtractor
-from ..compat import (
-    compat_str,
-    compat_HTTPError,
-)
+from ..compat import compat_HTTPError
 from ..utils import (
     determine_ext,
     float_or_none,
     int_or_none,
     parse_age_limit,
     parse_duration,
+    url_or_none,
     ExtractorError
 )
 
 
 class CrackleIE(InfoExtractor):
-    _VALID_URL = r'(?:crackle:|https?://(?:(?:www|m)\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
-    _TEST = {
+    _VALID_URL = r'(?:crackle:|https?://(?:(?:www|m)\.)?(?:sony)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
+    _TESTS = [{
         # geo restricted to CA
         'url': 'https://www.crackle.com/andromeda/2502343',
         'info_dict': {
@@ -45,6 +43,24 @@ class CrackleIE(InfoExtractor):
             # m3u8 download
             'skip_download': True,
         }
+    }, {
+        'url': 'https://www.sonycrackle.com/andromeda/2502343',
+        'only_matching': True,
+    }]
+
+    _MEDIA_FILE_SLOTS = {
+        '360p.mp4': {
+            'width': 640,
+            'height': 360,
+        },
+        '480p.mp4': {
+            'width': 768,
+            'height': 432,
+        },
+        '480p_1mbps.mp4': {
+            'width': 852,
+            'height': 480,
+        },
     }
 
     def _real_extract(self, url):
@@ -83,8 +99,8 @@ class CrackleIE(InfoExtractor):
             for e in media['MediaURLs']:
                 if e.get('UseDRM') is True:
                     continue
-                format_url = e.get('Path')
-                if not format_url or not isinstance(format_url, compat_str):
+                format_url = url_or_none(e.get('Path'))
+                if not format_url:
                     continue
                 ext = determine_ext(format_url)
                 if ext == 'm3u8':
@@ -94,6 +110,20 @@ class CrackleIE(InfoExtractor):
                 elif ext == 'mpd':
                     formats.extend(self._extract_mpd_formats(
                         format_url, video_id, mpd_id='dash', fatal=False))
+                elif format_url.endswith('.ism/Manifest'):
+                    formats.extend(self._extract_ism_formats(
+                        format_url, video_id, ism_id='mss', fatal=False))
+                else:
+                    mfs_path = e.get('Type')
+                    mfs_info = self._MEDIA_FILE_SLOTS.get(mfs_path)
+                    if not mfs_info:
+                        continue
+                    formats.append({
+                        'url': format_url,
+                        'format_id': 'http-' + mfs_path.split('.')[0],
+                        'width': mfs_info['width'],
+                        'height': mfs_info['height'],
+                    })
             self._sort_formats(formats)
 
             description = media.get('Description')
@@ -121,8 +151,8 @@ class CrackleIE(InfoExtractor):
                 for cc_file in cc_files:
                     if not isinstance(cc_file, dict):
                         continue
-                    cc_url = cc_file.get('Path')
-                    if not cc_url or not isinstance(cc_url, compat_str):
+                    cc_url = url_or_none(cc_file.get('Path'))
+                    if not cc_url:
                         continue
                     lang = cc_file.get('Locale') or 'en'
                     subtitles.setdefault(lang, []).append({'url': cc_url})