[bbccouk] Convert to new subtitles system

I haven't found any video available outside the UK, so I haven't added tests.

I have updated how the srt file is build, because (at least for www.bbc.co.uk/programmes/p02j9b69) the subtitles is inside 'span' elements.
This commit is contained in:
Jaime Marquínez Ferrándiz 2015-02-19 16:46:41 +01:00
parent 8807f1277f
commit f13b1e7d7f

View file

@ -2,12 +2,12 @@
import xml.etree.ElementTree import xml.etree.ElementTree
from .subtitles import SubtitlesInfoExtractor from .common import InfoExtractor
from ..utils import ExtractorError from ..utils import ExtractorError
from ..compat import compat_HTTPError from ..compat import compat_HTTPError
class BBCCoUkIE(SubtitlesInfoExtractor): class BBCCoUkIE(InfoExtractor):
IE_NAME = 'bbc.co.uk' IE_NAME = 'bbc.co.uk'
IE_DESC = 'BBC iPlayer' IE_DESC = 'BBC iPlayer'
_VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})' _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/(?:(?:(?:programmes|iplayer(?:/[^/]+)?/(?:episode|playlist))/)|music/clips[/#])(?P<id>[\da-z]{8})'
@ -215,17 +215,32 @@ def _extract_audio(self, media, programme_id):
formats.extend(conn_formats) formats.extend(conn_formats)
return formats return formats
def _extract_captions(self, media, programme_id): def _get_subtitles(self, media, programme_id):
subtitles = {} subtitles = {}
for connection in self._extract_connections(media): for connection in self._extract_connections(media):
captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions') captions = self._download_xml(connection.get('href'), programme_id, 'Downloading captions')
lang = captions.get('{http://www.w3.org/XML/1998/namespace}lang', 'en') 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}')) ps = captions.findall('./{0}body/{0}div/{0}p'.format('{http://www.w3.org/2006/10/ttaf1}'))
srt = '' 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): 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'), srt += '%s\r\n%s --> %s\r\n%s\r\n\r\n' % (str(pos), p.get('begin'), p.get('end'), _extract_text(p))
p.text.strip() if p.text is not None else '') subtitles[lang] = [
subtitles[lang] = srt {
'url': connection.get('href'),
'ext': 'ttml',
},
{
'data': srt,
'ext': 'srt',
},
]
return subtitles return subtitles
def _download_media_selector(self, programme_id): def _download_media_selector(self, programme_id):
@ -249,7 +264,7 @@ def _download_media_selector(self, programme_id):
elif kind == 'video': elif kind == 'video':
formats.extend(self._extract_video(media, programme_id)) formats.extend(self._extract_video(media, programme_id))
elif kind == 'captions': elif kind == 'captions':
subtitles = self._extract_captions(media, programme_id) subtitles = self.extract_subtitles(media, programme_id)
return formats, subtitles return formats, subtitles
@ -324,10 +339,6 @@ def _real_extract(self, url):
else: else:
programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id) programme_id, title, description, duration, formats, subtitles = self._download_playlist(group_id)
if self._downloader.params.get('listsubtitles', False):
self._list_available_subtitles(programme_id, subtitles)
return
self._sort_formats(formats) self._sort_formats(formats)
return { return {