From 2e5d60b7db7020b726cd54ee4cad8f2afbd1479d Mon Sep 17 00:00:00 2001 From: Ismael Mejia Date: Thu, 21 Feb 2013 20:51:35 +0100 Subject: [PATCH] Removed conversion from youtube closed caption format to srt since youtube api supports the 'srt' format --- test/test_youtube_subtitles.py | 4 ++-- youtube_dl/InfoExtractors.py | 24 ++++-------------------- 2 files changed, 6 insertions(+), 22 deletions(-) diff --git a/test/test_youtube_subtitles.py b/test/test_youtube_subtitles.py index 5d3566a35f..ff09ea459c 100644 --- a/test/test_youtube_subtitles.py +++ b/test/test_youtube_subtitles.py @@ -43,7 +43,7 @@ class TestYoutubeSubtitles(unittest.TestCase): DL.params['writesubtitles'] = True IE = YoutubeIE(DL) info_dict = IE.extract('QRS8MkLhQmM') - self.assertEqual(md5(info_dict[0]['subtitles']), 'c3228550d59116f3c29fba370b55d033') + self.assertEqual(md5(info_dict[0]['subtitles']), '4cd9278a35ba2305f47354ee13472260') def test_youtube_subtitles_it(self): DL = FakeDownloader() @@ -51,7 +51,7 @@ class TestYoutubeSubtitles(unittest.TestCase): DL.params['subtitleslang'] = 'it' IE = YoutubeIE(DL) info_dict = IE.extract('QRS8MkLhQmM') - self.assertEqual(md5(info_dict[0]['subtitles']), '132a88a0daf8e1520f393eb58f1f646a') + self.assertEqual(md5(info_dict[0]['subtitles']), '164a51f16f260476a05b50fe4c2f161d') if __name__ == '__main__': unittest.main() diff --git a/youtube_dl/InfoExtractors.py b/youtube_dl/InfoExtractors.py index d3c3ac2640..e3998fbe88 100755 --- a/youtube_dl/InfoExtractors.py +++ b/youtube_dl/InfoExtractors.py @@ -228,23 +228,6 @@ class YoutubeIE(InfoExtractor): """Indicate the download will use the RTMP protocol.""" self._downloader.to_screen(u'[youtube] RTMP download detected') - def _closed_captions_xml_to_srt(self, xml_string): - srt = '' - texts = re.findall(r'([^<]+)', xml_string, re.MULTILINE) - # TODO parse xml instead of regex - for n, (start, dur_tag, dur, caption) in enumerate(texts): - if not dur: dur = '4' - start = float(start) - end = start + float(dur) - start = "%02i:%02i:%02i,%03i" %(start/(60*60), start/60%60, start%60, start%1*1000) - end = "%02i:%02i:%02i,%03i" %(end/(60*60), end/60%60, end%60, end%1*1000) - caption = unescapeHTML(caption) - caption = unescapeHTML(caption) # double cycle, intentional - srt += str(n+1) + '\n' - srt += start + ' --> ' + end + '\n' - srt += caption + '\n\n' - return srt - def _extract_subtitles(self, video_id): self.report_video_subtitles_download(video_id) request = compat_urllib_request.Request('http://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id) @@ -268,15 +251,16 @@ class YoutubeIE(InfoExtractor): 'lang': srt_lang, 'name': srt_lang_list[srt_lang].encode('utf-8'), 'v': video_id, + 'fmt': 'srt', }) url = 'http://www.youtube.com/api/timedtext?' + params try: - srt_xml = compat_urllib_request.urlopen(url).read().decode('utf-8') + srt = compat_urllib_request.urlopen(url).read().decode('utf-8') except (compat_urllib_error.URLError, compat_http_client.HTTPException, socket.error) as err: return (u'WARNING: unable to download video subtitles: %s' % compat_str(err), None) - if not srt_xml: + if not srt: return (u'WARNING: Did not fetch video subtitles', None) - return (None, self._closed_captions_xml_to_srt(srt_xml)) + return (None, srt) def _print_formats(self, formats): print('Available formats:')