[nrktv] Rework subtitles and eliminate downloading twice

diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py
index 5ed235f1ea4d22b3a4d7a8801b4162a88f0dc16b..8ac38a174b4c64bbc8a3b40ec36c10d6fab595fa 100644 (file)
--- a/youtube_dl/extractor/nrk.py
+++ b/youtube_dl/extractor/nrk.py
@@ -4,6 +4,7 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_urlparse
 from ..utils import (
     ExtractorError,
     float_or_none,
@@ -196,20 +197,6 @@ class NRKTVIE(InfoExtractor):
         }
     ]
 
-    def _debug_print(self, txt):
-        if self._downloader.params.get('verbose', False):
-            self.to_screen('[debug] %s' % txt)
-
-    def _get_subtitles(self, subtitlesurl, video_id, baseurl):
-        url = "%s%s" % (baseurl, subtitlesurl)
-        self._debug_print('%s: Subtitle url: %s' % (video_id, url))
-        captions = self._download_xml(
-            url, video_id, 'Downloading subtitles')
-        lang = captions.get('lang', 'no')
-        return {lang: [
-            {'ext': 'ttml', 'url': url},
-        ]}
-
     def _extract_f4m(self, manifest_url, video_id):
         return self._extract_f4m_formats(
             manifest_url + '?hdcore=3.1.1&plugin=aasp-3.1.1.69.124', video_id, f4m_id='hds')
@@ -218,7 +205,7 @@ class NRKTVIE(InfoExtractor):
         mobj = re.match(self._VALID_URL, url)
         video_id = mobj.group('id')
         part_id = mobj.group('part_id')
-        baseurl = mobj.group('baseurl')
+        base_url = mobj.group('baseurl')
 
         webpage = self._download_webpage(url, video_id)
 
@@ -278,11 +265,14 @@ class NRKTVIE(InfoExtractor):
         self._sort_formats(formats)
 
         subtitles_url = self._html_search_regex(
-            r'data-subtitlesurl[ ]*=[ ]*"([^"]+)"',
-            webpage, 'subtitle URL', default=None)
-        subtitles = None
+            r'data-subtitlesurl\s*=\s*(["\'])(?P<url>.+?)\1',
+            webpage, 'subtitle URL', default=None, group='url')
+        subtitles = {}
         if subtitles_url:
-            subtitles = self.extract_subtitles(subtitles_url, video_id, baseurl)
+            subtitles['no'] = [{
+                'ext': 'ttml',
+                'url': compat_urlparse.urljoin(base_url, subtitles_url),
+            }]
 
         return {
             'id': video_id,