Merge branch 'douyutv' of https://github.com/bonfy/youtube-dl into bonfy-douyutv

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

diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py
index e950c76dd2f230e86ea94f974dd612ec0478f42e..bff36f9d3f24cad293144d8e216faf5eeefed92c 100644 (file)
--- a/youtube_dl/extractor/nrk.py
+++ b/youtube_dl/extractor/nrk.py
@@ -4,9 +4,11 @@ from __future__ import unicode_literals
 import re
 
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     float_or_none,
+    parse_duration,
     unified_strdate,
 )
 
@@ -72,7 +74,7 @@ class NRKIE(InfoExtractor):
 
 
 class NRKTVIE(InfoExtractor):
-    _VALID_URL = r'http://tv\.nrk(?:super)?\.no/(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
+    _VALID_URL = r'(?P<baseurl>http://tv\.nrk(?:super)?\.no/)(?:serie/[^/]+|program)/(?P<id>[a-zA-Z]{4}\d{8})(?:/\d{2}-\d{2}-\d{4})?(?:#del=(?P<part_id>\d+))?'
 
     _TESTS = [
         {
@@ -111,9 +113,6 @@ class NRKTVIE(InfoExtractor):
                 'upload_date': '20150106',
             },
             'skip': 'Only works from Norway',
-            'params': {
-                'proxy': '127.0.0.1:8118',
-            },
         },
         {
             'url': 'http://tv.nrk.no/serie/tour-de-ski/MSPO40010515/06-01-2015',
@@ -147,12 +146,33 @@ class NRKTVIE(InfoExtractor):
                 'duration': 6947.5199999999995,
             },
             'skip': 'Only works from Norway',
-            'params': {
-                'proxy': '127.0.0.1:8118',
-            },
         }
     ]
 
+    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',
+            transform_source=lambda s: s.replace(r'<br />', '\r\n'))
+        lang = captions.get('lang', 'no')
+        ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/ns/ttml}'))
+        srt = ''
+        for pos, p in enumerate(ps):
+            begin = parse_duration(p.get('begin'))
+            duration = parse_duration(p.get('dur'))
+            starttime = self._subtitles_timecode(begin)
+            endtime = self._subtitles_timecode(begin + duration)
+            srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (compat_str(pos), starttime, endtime, p.text)
+        return {lang: [
+            {'ext': 'ttml', 'url': url},
+            {'ext': 'srt', 'data': srt},
+        ]}
+
     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)
 
@@ -160,6 +180,7 @@ 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')
 
         webpage = self._download_webpage(url, video_id)
 
@@ -216,9 +237,15 @@ class NRKTVIE(InfoExtractor):
         m3u8_url = re.search(r'data-hls-media="([^"]+)"', webpage)
         if m3u8_url:
             formats.extend(self._extract_m3u8_formats(m3u8_url.group(1), video_id, 'mp4'))
-
         self._sort_formats(formats)
 
+        subtitles_url = self._html_search_regex(
+            r'data-subtitlesurl[ ]*=[ ]*"([^"]+)"',
+            webpage, 'subtitle URL', default=None)
+        subtitles = None
+        if subtitles_url:
+            subtitles = self.extract_subtitles(subtitles_url, video_id, baseurl)
+
         return {
             'id': video_id,
             'title': title,
@@ -227,4 +254,5 @@ class NRKTVIE(InfoExtractor):
             'upload_date': upload_date,
             'duration': duration,
             'formats': formats,
+            'subtitles': subtitles,
         }