[brightcove:new] Add support for non numeric ref: preffixed video ids

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

diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py
index 9ae724c2b88aeed4f6544ae4e86b575b7a42977c..9760e606f23481afb7376bdf1f7552831b2c06c1 100644 (file)
--- a/youtube_dl/extractor/brightcove.py
+++ b/youtube_dl/extractor/brightcove.py
@@ -11,9 +11,9 @@ from ..compat import (
     compat_str,
     compat_urllib_parse,
     compat_urllib_parse_urlparse,
-    compat_urllib_request,
     compat_urlparse,
     compat_xml_parse_error,
+    compat_HTTPError,
 )
 from ..utils import (
     determine_ext,
@@ -24,6 +24,7 @@ from ..utils import (
     js_to_json,
     int_or_none,
     parse_iso8601,
+    sanitized_Request,
     unescapeHTML,
     unsmuggle_url,
 )
@@ -250,7 +251,7 @@ 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 +356,7 @@ 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>\d+|ref:[^&]+)'
     _TESTS = [{
         'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
         'md5': 'c8100925723840d4b0d243f7025703be',
@@ -387,14 +388,28 @@ 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,
+    }, {
+        # non numeric ref: prefixed video id
+        'url': 'http://players.brightcove.net/710858724001/default_default/index.html?videoId=ref:event-stream-356',
+        '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 = []
 
@@ -402,21 +417,24 @@ 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/
-                        (\d+)/([\da-f-]+)_([^/]+)/index\.min\.js
+                        (\d+)/([\da-f-]+)_([^/]+)/index(?:\.min)?\.js
                 ''', webpage):
             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):
@@ -441,11 +459,17 @@ 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})
-        json_data = self._download_json(req, video_id)
+        try:
+            json_data = self._download_json(req, video_id)
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
+                json_data = self._parse_json(e.cause.read().decode(), video_id)
+                raise ExtractorError(json_data[0]['message'], expected=True)
+            raise
 
         title = json_data['name']
 
@@ -456,11 +480,9 @@ class BrightcoveNewIE(InfoExtractor):
             if source_type == 'application/x-mpegURL':
                 if not src:
                     continue
-                m3u8_formats = self._extract_m3u8_formats(
+                formats.extend(self._extract_m3u8_formats(
                     src, video_id, 'mp4', entry_protocol='m3u8_native',
-                    m3u8_id='hls', fatal=False)
-                if m3u8_formats:
-                    formats.extend(m3u8_formats)
+                    m3u8_id='hls', fatal=False))
             else:
                 streaming_src = source.get('streaming_src')
                 stream_name, app_name = source.get('stream_name'), source.get('app_name')