Update upstream source from tag 'upstream/2018.06.18'

[youtubedl] / youtube_dl / extractor / theplatform.py

diff --git a/youtube_dl/extractor/theplatform.py b/youtube_dl/extractor/theplatform.py
index 192d8fa292e0a6f360929590274d06b4745fb8f6..b1a985ff6c12368347d98d95beed6a042e70093c 100644 (file)
--- a/youtube_dl/extractor/theplatform.py
+++ b/youtube_dl/extractor/theplatform.py
@@ -80,14 +80,33 @@ class ThePlatformBaseIE(OnceIE):
                     'url': src,
                 })
 
+        duration = info.get('duration')
+        tp_chapters = info.get('chapters', [])
+        chapters = []
+        if tp_chapters:
+            def _add_chapter(start_time, end_time):
+                start_time = float_or_none(start_time, 1000)
+                end_time = float_or_none(end_time, 1000)
+                if start_time is None or end_time is None:
+                    return
+                chapters.append({
+                    'start_time': start_time,
+                    'end_time': end_time,
+                })
+
+            for chapter in tp_chapters[:-1]:
+                _add_chapter(chapter.get('startTime'), chapter.get('endTime'))
+            _add_chapter(tp_chapters[-1].get('startTime'), tp_chapters[-1].get('endTime') or duration)
+
         return {
             'title': info['title'],
             'subtitles': subtitles,
             'description': info['description'],
             'thumbnail': info['defaultThumbnailUrl'],
-            'duration': int_or_none(info.get('duration'), 1000),
+            'duration': float_or_none(duration, 1000),
             'timestamp': int_or_none(info.get('pubDate'), 1000) or None,
             'uploader': info.get('billingCode'),
+            'chapters': chapters,
         }
 
     def _extract_theplatform_metadata(self, path, video_id):
@@ -179,10 +198,12 @@ class ThePlatformIE(ThePlatformBaseIE, AdobePassIE):
         if m:
             return [m.group('url')]
 
+        # Are whitesapces ignored in URLs?
+        # https://github.com/rg3/youtube-dl/issues/12044
         matches = re.findall(
-            r'<(?:iframe|script)[^>]+src=(["\'])((?:https?:)?//player\.theplatform\.com/p/.+?)\1', webpage)
+            r'(?s)<(?:iframe|script)[^>]+src=(["\'])((?:https?:)?//player\.theplatform\.com/p/.+?)\1', webpage)
         if matches:
-            return list(zip(*matches))[1]
+            return [re.sub(r'\s', '', list(zip(*matches))[1][0])]
 
     @staticmethod
     def _sign_url(url, sig_key, sig_secret, life=600, include_qs=False):
@@ -195,7 +216,7 @@ class ThePlatformIE(ThePlatformBaseIE, AdobePassIE):
         def hex_to_bytes(hex):
             return binascii.a2b_hex(hex.encode('ascii'))
 
-        relative_path = re.match(r'https?://link.theplatform.com/s/([^?]+)', url).group(1)
+        relative_path = re.match(r'https?://link\.theplatform\.com/s/([^?]+)', url).group(1)
         clear_text = hex_to_bytes(flags + expiration_date + str_to_hex(relative_path))
         checksum = hmac.new(sig_key.encode('ascii'), clear_text, hashlib.sha1).hexdigest()
         sig = flags + expiration_date + checksum + str_to_hex(sig_secret)
@@ -306,9 +327,10 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
         },
     }]
 
-    def _extract_feed_info(self, provider_id, feed_id, filter_query, video_id, custom_fields=None, asset_types_query={}):
+    def _extract_feed_info(self, provider_id, feed_id, filter_query, video_id, custom_fields=None, asset_types_query={}, account_id=None):
         real_url = self._URL_TEMPLATE % (self.http_scheme(), provider_id, feed_id, filter_query)
         entry = self._download_json(real_url, video_id)['entries'][0]
+        main_smil_url = 'http://link.theplatform.com/s/%s/media/guid/%d/%s' % (provider_id, account_id, entry['guid']) if account_id else None
 
         formats = []
         subtitles = {}
@@ -333,7 +355,7 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
                 if asset_type in asset_types_query:
                     query.update(asset_types_query[asset_type])
                 cur_formats, cur_subtitles = self._extract_theplatform_smil(update_url_query(
-                    smil_url, query), video_id, 'Downloading SMIL data for %s' % asset_type)
+                    main_smil_url or smil_url, query), video_id, 'Downloading SMIL data for %s' % asset_type)
                 formats.extend(cur_formats)
                 subtitles = self._merge_subtitles(subtitles, cur_subtitles)