[youtube] Unescape HTML for series (closes #18641)

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

diff --git a/youtube_dl/extractor/npo.py b/youtube_dl/extractor/npo.py
index f756331f878165694611a0338b83f829037e7830..c2cb85a734f256330b9cbb9bfd77910a6f962c45 100644 (file)
--- a/youtube_dl/extractor/npo.py
+++ b/youtube_dl/extractor/npo.py
@@ -36,8 +36,8 @@ class NPOIE(NPOBaseIE):
                         https?://
                             (?:www\.)?
                             (?:
-                                npo\.nl/(?!(?:live|radio)/)(?:[^/]+/){2}|
-                                ntr\.nl/(?:[^/]+/){2,}|
+                                npo\.nl/(?:[^/]+/)*|
+                                (?:ntr|npostart)\.nl/(?:[^/]+/){2,}|
                                 omroepwnl\.nl/video/fragment/[^/]+__|
                                 (?:zapp|npo3)\.nl/(?:[^/]+/){2,}
                             )
@@ -160,8 +160,20 @@ class NPOIE(NPOBaseIE):
     }, {
         'url': 'https://www.zapp.nl/1803-skelterlab/instructie-video-s/740-instructievideo-s/POMS_AT_11736927',
         'only_matching': True,
+    }, {
+        'url': 'https://www.npostart.nl/broodje-gezond-ei/28-05-2018/KN_1698996',
+        'only_matching': True,
+    }, {
+        'url': 'https://npo.nl/KN_1698996',
+        'only_matching': True,
     }]
 
+    @classmethod
+    def suitable(cls, url):
+        return (False if any(ie.suitable(url)
+                for ie in (NPOLiveIE, NPORadioIE, NPORadioFragmentIE))
+                else super(NPOIE, cls).suitable(url))
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
         return self._get_info(video_id)
@@ -195,6 +207,10 @@ class NPOIE(NPOBaseIE):
         formats = []
         urls = set()
 
+        def is_legal_url(format_url):
+            return format_url and format_url not in urls and re.match(
+                r'^(?:https?:)?//', format_url)
+
         QUALITY_LABELS = ('Laag', 'Normaal', 'Hoog')
         QUALITY_FORMATS = ('adaptive', 'wmv_sb', 'h264_sb', 'wmv_bb', 'h264_bb', 'wvc1_std', 'h264_std')
 
@@ -208,7 +224,7 @@ class NPOIE(NPOBaseIE):
             })['items'][0]
         for num, item in enumerate(items):
             item_url = item.get('url')
-            if not item_url or item_url in urls:
+            if not is_legal_url(item_url):
                 continue
             urls.add(item_url)
             format_id = self._search_regex(
@@ -229,7 +245,7 @@ class NPOIE(NPOBaseIE):
                     quality = quality_from_format_id(format_id)
                     f_id = format_id
                 else:
-                    quality, f_id = None
+                    quality, f_id = [None] * 2
                 formats.append({
                     'url': format_url,
                     'format_id': f_id,
@@ -266,7 +282,7 @@ class NPOIE(NPOBaseIE):
                 video_url = stream_info.get('url')
             if not video_url or video_url in urls:
                 continue
-            urls.add(item_url)
+            urls.add(video_url)
             if determine_ext(video_url) == 'm3u8':
                 formats.extend(self._extract_m3u8_formats(
                     video_url, video_id, ext='mp4',
@@ -279,7 +295,7 @@ class NPOIE(NPOBaseIE):
         if not is_live:
             for num, stream in enumerate(metadata.get('streams', [])):
                 stream_url = stream.get('url')
-                if not stream_url or stream_url in urls:
+                if not is_legal_url(stream_url):
                     continue
                 urls.add(stream_url)
                 # smooth streaming is not supported
@@ -385,7 +401,7 @@ class NPOLiveIE(NPOBaseIE):
 
 class NPORadioIE(InfoExtractor):
     IE_NAME = 'npo.nl:radio'
-    _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)/?$'
+    _VALID_URL = r'https?://(?:www\.)?npo\.nl/radio/(?P<id>[^/]+)'
 
     _TEST = {
         'url': 'http://www.npo.nl/radio/radio-1',
@@ -400,6 +416,10 @@ class NPORadioIE(InfoExtractor):
         }
     }
 
+    @classmethod
+    def suitable(cls, url):
+        return False if NPORadioFragmentIE.suitable(url) else super(NPORadioIE, cls).suitable(url)
+
     @staticmethod
     def _html_get_attribute_regex(attribute):
         return r'{0}\s*=\s*\'([^\']+)\''.format(attribute)