Merge remote-tracking branch 'origin/master' into pr-bbcnews

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

diff --git a/youtube_dl/extractor/bbc.py b/youtube_dl/extractor/bbc.py
index fed344ea0b6003d63670299e293d60096d610adb..471d865d26c5feb03a66e3eca42fae5ccbe21e39 100644 (file)
--- a/youtube_dl/extractor/bbc.py
+++ b/youtube_dl/extractor/bbc.py
@@ -255,26 +255,11 @@ class BBCCoUkIE(InfoExtractor):
         for connection in self._extract_connections(media):
             captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
             lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en')
-            ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
-            srt = ''
-
-            def _extract_text(p):
-                if p.text is not None:
-                    stripped_text = p.text.strip()
-                    if stripped_text:
-                        return stripped_text
-                return ' '.join(span.text.strip() for span in p.findall('{http://www.w3.org/2006/10/ttaf1}span'))
-            for pos, p in enumerate(ps):
-                srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'), _extract_text(p))
             subtitles[lang] = [
                 {
                     'url': connection.get('href'),
                     'ext': 'ttml',
                 },
-                {
-                    'data': srt,
-                    'ext': 'srt',
-                },
             ]
         return subtitles
 
@@ -428,6 +413,8 @@ class BBCNewsIE(BBCCoUkIE):
             'title': 'Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
             'description': 'Germanwings plane crash site in aerial video - Aerial footage showed the site of the crash in the Alps - courtesy BFM TV',
             'duration': 47,
+            'upload_date': '20150324',
+            'uploader': 'BBC News',
         },
         'params': {
             'skip_download': True,
@@ -438,8 +425,11 @@ class BBCNewsIE(BBCCoUkIE):
         'info_dict': {
             'id': 'NA',
             'ext': 'mp4',
-            'title': 'YPG - Tel Abyad..n tamam. kontrol.m.zde',
+            'title': 'YPG: Tel Abyad\'\u0131n tamam\u0131 kontrol\xfcm\xfczde',
+            'description': 'YPG: Tel Abyad\'\u0131n tamam\u0131 kontrol\xfcm\xfczde',
             'duration': 47,
+            'upload_date': '20150615',
+            'uploader': 'BBC News',
         },
         'params': {
             'skip_download': True,
@@ -450,8 +440,11 @@ class BBCNewsIE(BBCCoUkIE):
         'info_dict': {
             'id': '39275083',
             'ext': 'mp4',
-            'title': 'Honduras militariza sus hospitales por nuevo esc.ndalo de corrupci.n',
+            'title': 'Honduras militariza sus hospitales por nuevo esc\xe1ndalo de corrupci\xf3n',
+            'description': 'Honduras militariza sus hospitales por nuevo esc\xe1ndalo de corrupci\xf3n',
             'duration': 87,
+            'upload_date': '20150619',
+            'uploader': 'BBC News',
         },
         'params': {
             'skip_download': True,
@@ -507,7 +500,9 @@ class BBCNewsIE(BBCCoUkIE):
             title = jent.get('caption',list_title)
 
             duration = parse_duration(jent.get('duration'))
-            description = list_title + ' - ' + jent.get('caption','')
+            description = list_title
+            if jent.get('caption'):
+               description += ' - ' + jent.get('caption')
             thumbnail = None
             if jent.has_key('image'):
                thumbnail=jent['image'].get('href')
@@ -539,8 +534,12 @@ class BBCNewsIE(BBCCoUkIE):
                
             self._sort_formats(formats)
 
+            id = jent.get('id') if programme_id == None else programme_id
+            if id == None:
+               id = 'NA'
+
             ret.append( {
-                'id': jent.get('id') if programme_id == None else programme_id,
+                'id': id,
                 'uploader': 'BBC News',
                 'upload_date': pubdate,
                 'title': title,