Merge branch 'remitamine-brightcove_in_page_embed'

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

diff --git a/youtube_dl/extractor/brightcove.py b/youtube_dl/extractor/brightcove.py
index 2ad35ec902c06607093b6cf7b0ea31630e98dcff..6b184157cc28663c3c2d01c5360590b95e41769a 100644 (file)
--- a/youtube_dl/extractor/brightcove.py
+++ b/youtube_dl/extractor/brightcove.py
@@ -20,17 +20,17 @@ 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,
+    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'
 
@@ -353,8 +353,9 @@ 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+)'
+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+)'
     _TEST = {
         'url': 'http://players.brightcove.net/929656772001/e41d32dc-ec74-459e-a845-6c69f7b724ea_default/index.html?videoId=4463358922001',
         'md5': 'c8100925723840d4b0d243f7025703be',
@@ -365,23 +366,40 @@ class BrightcoveInPageEmbedIE(InfoExtractor):
             'description': 'md5:eac376a4fe366edc70279bfb681aea16',
             'timestamp': 1441391203,
             'upload_date': '20150904',
-            'duration': 165768,
+            'duration': 165.768,
             'uploader_id': '929656772001',
         }
     }
 
     @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]
+        # 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(
+                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()