[rmcdecouverte] Improve, bypass geo restriction and add support for live (closes...

diff --git a/youtube_dl/extractor/rmcdecouverte.py b/youtube_dl/extractor/rmcdecouverte.py
index c44b90274e36629a1a67717a8a9de925d2f970f7..c3623edcc7adc3fad26d769d79c7088fc98c2593 100644 (file)
--- a/youtube_dl/extractor/rmcdecouverte.py
+++ b/youtube_dl/extractor/rmcdecouverte.py
@@ -1,18 +1,21 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import re
+
 from .common import InfoExtractor
 from .brightcove import BrightcoveLegacyIE
 from ..compat import (
     compat_parse_qs,
     compat_urlparse,
 )
+from ..utils import smuggle_url
 
 
 class RMCDecouverteIE(InfoExtractor):
-    _VALID_URL = r'https?://rmcdecouverte\.bfmtv\.com/.+/program_(?P<id>\d+)'
+    _VALID_URL = r'https?://rmcdecouverte\.bfmtv\.com/(?:(?:[^/]+/)*program_(?P<id>\d+)|(?P<live_id>mediaplayer-direct))'
 
-    _TEST = {
+    _TESTS = [{
         'url': 'https://rmcdecouverte.bfmtv.com/wheeler-dealers-occasions-a-saisir/program_2566/',
         'info_dict': {
             'id': '5983675500001',
@@ -27,12 +30,17 @@ class RMCDecouverteIE(InfoExtractor):
             'skip_download': True,
         },
         'skip': 'only available for a week',
-    }
+    }, {
+        # live, geo restricted, bypassable
+        'url': 'https://rmcdecouverte.bfmtv.com/mediaplayer-direct/',
+        'only_matching': True,
+    }]
     BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1969646226001/default_default/index.html?videoId=%s'
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
+        mobj = re.match(self._VALID_URL, url)
+        display_id = mobj.group('id') or mobj.group('live_id')
+        webpage = self._download_webpage(url, display_id)
         brightcove_legacy_url = BrightcoveLegacyIE._extract_brightcove_url(webpage)
         if brightcove_legacy_url:
             brightcove_id = compat_parse_qs(compat_urlparse.urlparse(
@@ -41,5 +49,7 @@ class RMCDecouverteIE(InfoExtractor):
             brightcove_id = self._search_regex(
                 r'data-video-id=["\'](\d+)', webpage, 'brightcove id')
         return self.url_result(
-            self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, 'BrightcoveNew',
-            brightcove_id)
+            smuggle_url(
+                self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id,
+                {'geo_countries': ['FR']}),
+            'BrightcoveNew', brightcove_id)