Merge remote-tracking branch 'upstream/master' into bliptv
[youtube-dl] / youtube_dl / extractor / brightcove.py
index 2ad35ec902c06607093b6cf7b0ea31630e98dcff..f5ebae1e68e456c158476a02d9df49feac97e02a 100644 (file)
@@ -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,
 )
@@ -20,17 +19,18 @@ from ..utils import (
     ExtractorError,
     find_xpath_attr,
     fix_xml_ampersands,
-    unescapeHTML,
-    unsmuggle_url,
-    js_to_json,
     float_or_none,
+    js_to_json,
     int_or_none,
     parse_iso8601,
-    extract_attributes,
+    sanitized_Request,
+    unescapeHTML,
+    unsmuggle_url,
 )
 
 
-class BrightcoveIE(InfoExtractor):
+class BrightcoveLegacyIE(InfoExtractor):
+    IE_NAME = 'brightcove:legacy'
     _VALID_URL = r'(?:https?://.*brightcove\.com/(services|viewer).*?\?|brightcove:)(?P<query>.*)'
     _FEDERATED_URL_TEMPLATE = 'http://c.brightcove.com/services/viewer/htmlFederated?%s'
 
@@ -250,7 +250,7 @@ class BrightcoveIE(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]
@@ -353,9 +353,10 @@ class BrightcoveIE(InfoExtractor):
         return info
 
 
-class BrightcoveInPageEmbedIE(InfoExtractor):
-    _VALID_URL = r'https?://players\.brightcove\.net/(?P<account_id>\d+)/(?P<player_id>[\da-f-]+)_(?P<embed>[a-z]+)/index\.html\?.*videoId=(?P<video_id>\d+)'
-    _TEST = {
+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+)'
+    _TESTS = [{
         'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
         'md5': 'c8100925723840d4b0d243f7025703be',
         'info_dict': {
@@ -363,25 +364,62 @@ class BrightcoveInPageEmbedIE(InfoExtractor):
             'ext': 'mp4',
             'title': 'Meet the man behind Popcorn Time',
             'description': 'md5:eac376a4fe366edc70279bfb681aea16',
+            'duration': 165.768,
             'timestamp': 1441391203,
             'upload_date': '20150904',
-            'duration': 165768,
             'uploader_id': '929656772001',
+            'formats': 'mincount:22',
+        },
+    }, {
+        # with rtmp streams
+        'url': 'http://players.brightcove.net/4036320279001/5d112ed9-283f-485f-a7f9-33f42e8bc042_default/index.html?videoId=4279049078001',
+        'info_dict': {
+            'id': '4279049078001',
+            'ext': 'mp4',
+            'title': 'Titansgrave: Chapter 0',
+            'description': 'Titansgrave: Chapter 0',
+            'duration': 1242.058,
+            'timestamp': 1433556729,
+            'upload_date': '20150606',
+            'uploader_id': '4036320279001',
+            'formats': 'mincount:41',
+        },
+        'params': {
+            'skip_download': True,
         }
-    }
+    }]
 
     @staticmethod
-    def _extract_url(webpage):
-        video_attributes = re.search(r'(?s)<video([^>]*)>.*?</(?:video|audio)>', webpage)
-        if video_attributes:
-            video_attributes = extract_attributes(video_attributes.group(), r'(?s)\s*data-(account|video-id|playlist-id|policy-key|player|embed)\s*=\s*["\']([^"\']+)["\']')
-            account_id = video_attributes.get('account')
-            player_id = video_attributes.get('player')
-            embed = video_attributes.get('embed')
-            video_id = video_attributes.get('video-id')
-            if account_id and player_id and embed and video_id:
-                return 'http://players.brightcove.net/%s/%s_%s/index.html?videoId=%s' % (account_id, player_id, embed, video_id)
-        return None
+    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)
+        # 3. http://docs.brightcove.com/en/video-cloud/brightcove-player/guides/embed-in-page.html
+
+        entries = []
+
+        # Look for iframe embeds [1]
+        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]
+        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
+                r'''(?sx)
+                    <video[^>]+
+                        data-video-id=["\'](\d+)["\'][^>]*>.*?
+                    </video>.*?
+                    <script[^>]+
+                        src=["\'](?:https?:)?//players\.brightcove\.net/
+                        (\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):
         account_id, player_id, embed, video_id = re.match(self._VALID_URL, url).groups()
@@ -405,7 +443,7 @@ class BrightcoveInPageEmbedIE(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})