Merge pull request #7045 from remitamine/ign

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

diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py
index 9ae724c2b88aeed4f6544ae4e86b575b7a42977c..03a4f446e9d80808b4e4b3df3ad5b3d0f8a550e3 100644 (file)
--- a/youtube_dl/extractor/brightcove.py
+++ b/youtube_dl/extractor/brightcove.py
@@ -11,7 +11,6 @@ from ..compat import (
     compat_str,
     compat_urllib_parse,
     compat_urllib_parse_urlparse,
-    compat_urllib_request,
     compat_urlparse,
     compat_xml_parse_error,
 )
@@ -24,6 +23,7 @@ from ..utils import (
     js_to_json,
     int_or_none,
     parse_iso8601,
+    sanitized_Request,
     unescapeHTML,
     unsmuggle_url,
 )
@@ -250,7 +250,7 @@ class BrightcoveLegacyIE(InfoExtractor):
 
     def _get_video_info(self, video_id, query_str, query, referer=None):
         request_url = self._FEDERATED_URL_TEMPLATE % query_str
-        req = compat_urllib_request.Request(request_url)
+        req = sanitized_Request(request_url)
         linkBase = query.get('linkBaseURL')
         if linkBase is not None:
             referer = linkBase[0]
@@ -355,7 +355,7 @@ class BrightcoveLegacyIE(InfoExtractor):
 
 class BrightcoveNewIE(InfoExtractor):
     IE_NAME = 'brightcove:new'
-    _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*videoId=(?P<video_id>\d+)'
+    _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[^/]+)_(?P<embed>[^/]+)/index\.html\?.*videoId=(?P<video_id>(?:ref:)?\d+)'
     _TESTS = [{
         'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
         'md5': 'c8100925723840d4b0d243f7025703be',
@@ -387,14 +387,24 @@ class BrightcoveNewIE(InfoExtractor):
         'params': {
             'skip_download': True,
         }
+    }, {
+        # ref: prefixed video id
+        'url': 'http://players.brightcove.net/3910869709001/21519b5c-4b3b-4363-accb-bdc8f358f823_default/index.html?videoId=ref:7069442',
+        'only_matching': True,
     }]
 
+    @staticmethod
+    def _extract_url(webpage):
+        urls = BrightcoveNewIE._extract_urls(webpage)
+        return urls[0] if urls else None
+
     @staticmethod
     def _extract_urls(webpage):
         # Reference:
         # 1. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideoiniframe
-        # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript)
+        # 2. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/publish-video.html#setvideousingjavascript
         # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html
+        # 4. https://support.brightcove.com/en/video-cloud/docs/dynamically-assigning-videos-player
 
         entries = []
 
@@ -402,13 +412,15 @@ class BrightcoveNewIE(InfoExtractor):
         for _, url in re.findall(
                 r'<iframe[^>]+src=(["\'])((?:https?:)//players\.brightcove\.net/\d+/[^/]+/index\.html.+?)\1', webpage):
             entries.append(url)
+
         # Look for embed_in_page embeds [2]
-        # According to examples from [3] it's unclear whether video id may be optional
-        # and what to do when it is
         for video_id, account_id, player_id, embed in re.findall(
+                # According to examples from [3] it's unclear whether video id
+                # may be optional and what to do when it is
+                # According to [4] data-video-id may be prefixed with ref:
                 r'''(?sx)
                     <video[^>]+
-                        data-video-id=["\'](\d+)["\'][^>]*>.*?
+                        data-video-id=["\']((?:ref:)?\d+)["\'][^>]*>.*?
                     </video>.*?
                     <script[^>]+
                         src=["\'](?:https?:)?//players\.brightcove\.net/
@@ -417,6 +429,7 @@ class BrightcoveNewIE(InfoExtractor):
             entries.append(
                 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s'
                 % (account_id, player_id, embed, video_id))
+
         return entries
 
     def _real_extract(self, url):
@@ -441,7 +454,7 @@ class BrightcoveNewIE(InfoExtractor):
                 r'policyKey\s*:\s*(["\'])(?P<pk>.+?)\1',
                 webpage, 'policy key', group='pk')
 
-        req = compat_urllib_request.Request(
+        req = sanitized_Request(
             'https://edge.api.brightcove.com/playback/v1/accounts/%s/videos/%s'
             % (account_id, video_id),
             headers={'Accept': 'application/json;pk=%s' % policy_key})