[crackle] Bypass geo restriction

youtube_dl/extractor/crackle.py

@@ -4,20 +4,24 @@ from __future__ import unicode_literals, division
 import re
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_str,
+    compat_HTTPError,
+)
 from ..utils import (
     determine_ext,
     float_or_none,
     int_or_none,
     parse_age_limit,
     parse_duration,
+    ExtractorError
 )
 
 
 class CrackleIE(InfoExtractor):
-    _GEO_COUNTRIES = ['US']
     _VALID_URL = r'(?:crackle:|https?://(?:(?:www|m)\.)?crackle\.com/(?:playlist/\d+/|(?:[^/]+/)+))(?P<id>\d+)'
     _TEST = {
+        # geo restricted to CA
         'url': 'https://www.crackle.com/andromeda/2502343',
         'info_dict': {
             'id': '2502343',
@@ -46,12 +50,32 @@ class CrackleIE(InfoExtractor):
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
+        country_code = self._downloader.params.get('geo_bypass_country', None)
+        countries = [country_code] if country_code else (
+            'US', 'AU', 'CA', 'AS', 'FM', 'GU', 'MP', 'PR', 'PW', 'MH', 'VI')
+
+        last_e = None
+
+        for country in countries:
+            try:
                 media = self._download_json(
                     'https://web-api-us.crackle.com/Service.svc/details/media/%s/%s'
-            % (video_id, self._GEO_COUNTRIES[0]), video_id, query={
+                    % (video_id, country), video_id,
+                    'Downloading media JSON as %s' % country,
+                    'Unable to download media JSON', query={
                         'disableProtocols': 'true',
                         'format': 'json'
                     })
+            except ExtractorError as e:
+                # 401 means geo restriction, trying next country
+                if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                    last_e = e
+                    continue
+                raise
+
+            media_urls = media.get('MediaURLs')
+            if not media_urls or not isinstance(media_urls, list):
+                continue
 
             title = media['Title']
 
@@ -136,3 +160,5 @@ class CrackleIE(InfoExtractor):
                 'subtitles': subtitles,
                 'formats': formats,
             }
+
+        raise last_e