youtube-dl

commit de02d1f4e94a50eaad0f268fffec6880e9d61d2f
parent e1f93a0a76f054f537c9103a8958e1c1ab631fa5
Author: Sergey M․ <dstftw@gmail.com>
Date:   Mon,  8 Aug 2016 04:58:02 +0700

[rozhlas] Fix regexes and improve extraction (Closes #10253)

Diffstat:
M
youtube_dl/extractor/rozhlas.py
 | 
30
+++++++++++++++++++++++-------

1 file changed, 23 insertions(+), 7 deletions(-)
diff --git a/youtube_dl/extractor/rozhlas.py b/youtube_dl/extractor/rozhlas.py
@@ -2,11 +2,15 @@
 from __future__ import unicode_literals
 
 from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    remove_start,
+)
 
 
 class RozhlasIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?prehravac\.rozhlas\.cz/audio/(?P<id>[0-9]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'http://prehravac.rozhlas.cz/audio/3421320',
         'md5': '504c902dbc9e9a1fd50326eccf02a7e2',
         'info_dict': {
@@ -15,20 +19,32 @@ class RozhlasIE(InfoExtractor):
             'title': 'Echo Pavla Klusáka (30.06.2015 21:00)',
             'description': 'Osmdesátiny Terryho Rileyho jsou skvělou příležitostí proletět se elektronickými i akustickými díly zakladatatele minimalismu, který je aktivní už přes padesát let'
         }
-    }
+    }, {
+        'url': 'http://prehravac.rozhlas.cz/audio/3421320/embed',
+        'skip_download': True,
+    }]
 
     def _real_extract(self, url):
         audio_id = self._match_id(url)
-        webpage = self._download_webpage(url, audio_id)
 
-        title = self._html_search_regex(r'<h3>(.+?)</h3>', webpage, 'title')
-        description = self._html_search_regex(r'<p title="(.+?)">', webpage, 'description', fatal=False)
+        webpage = self._download_webpage(
+            'http://prehravac.rozhlas.cz/audio/%s' % audio_id, audio_id)
 
-        url = 'http://media.rozhlas.cz/_audio/' + audio_id + '.mp3'
+        title = self._html_search_regex(
+            r'<h3>(.+?)</h3>\s*<p[^>]*>.*?</p>\s*<div[^>]+id=["\']player-track',
+            webpage, 'title', default=None) or remove_start(
+            self._og_search_title(webpage), 'Radio Wave - ')
+        description = self._html_search_regex(
+            r'<p[^>]+title=(["\'])(?P<url>(?:(?!\1).)+)\1[^>]*>.*?</p>\s*<div[^>]+id=["\']player-track',
+            webpage, 'description', fatal=False, group='url')
+        duration = int_or_none(self._search_regex(
+            r'data-duration=["\'](\d+)', webpage, 'duration', default=None))
 
         return {
             'id': audio_id,
-            'url': url,
+            'url': 'http://media.rozhlas.cz/_audio/%s.mp3' % audio_id,
             'title': title,
             'description': description,
+            'duration': duration,
+            'vcodec': 'none',
         }