From 3869028ffb6be6ab719e5cf1004276dfdfd1216d Mon Sep 17 00:00:00 2001 From: Yen Chi Hsuan Date: Sat, 16 Sep 2017 12:18:38 +0800 Subject: [PATCH] [utils] Use bytes-like objects in dfxp2srt This fixes handling of non-UTF8 TTML subtitles Closes #14191 --- ChangeLog | 6 ++++++ test/test_utils.py | 26 +++++++++++++++++++++++--- youtube_dl/postprocessor/ffmpeg.py | 2 +- youtube_dl/utils.py | 18 +++++++++++------- 4 files changed, 41 insertions(+), 11 deletions(-) diff --git a/ChangeLog b/ChangeLog index 041dfd7b9..ba9260e3e 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,3 +1,9 @@ +version + +Core +* [utils] Fix handling raw TTML subtitles (#14191) + + version 2017.09.15 Core diff --git a/test/test_utils.py b/test/test_utils.py index e50f3764e..efa73d0f4 100644 --- a/test/test_utils.py +++ b/test/test_utils.py @@ -1064,7 +1064,7 @@ ffmpeg version 2.4.4 Copyright (c) 2000-2014 the FFmpeg ...'''), '2.4.4')

Ignored, three

- ''' + '''.encode('utf-8') srt_data = '''1 00:00:00,000 --> 00:00:01,000 The following line contains Chinese characters and special symbols @@ -1089,7 +1089,7 @@ Line

The first line

- ''' + '''.encode('utf-8') srt_data = '''1 00:00:00,000 --> 00:00:01,000 The first line @@ -1115,7 +1115,7 @@ The first line

inner
style

-''' +'''.encode('utf-8') srt_data = '''1 00:00:02,080 --> 00:00:05,839 default stylecustom style @@ -1138,6 +1138,26 @@ part 3 ''' self.assertEqual(dfxp2srt(dfxp_data_with_style), srt_data) + dfxp_data_non_utf8 = ''' + + +
+

Line 1

+

第二行

+
+ + '''.encode('utf-16') + srt_data = '''1 +00:00:00,000 --> 00:00:01,000 +Line 1 + +2 +00:00:01,000 --> 00:00:02,000 +第二行 + +''' + self.assertEqual(dfxp2srt(dfxp_data_non_utf8), srt_data) + def test_cli_option(self): self.assertEqual(cli_option({'proxy': '127.0.0.1:3128'}, '--proxy', 'proxy'), ['--proxy', '127.0.0.1:3128']) self.assertEqual(cli_option({'proxy': None}, '--proxy', 'proxy'), []) diff --git a/youtube_dl/postprocessor/ffmpeg.py b/youtube_dl/postprocessor/ffmpeg.py index 51256a3fb..f71d413b5 100644 --- a/youtube_dl/postprocessor/ffmpeg.py +++ b/youtube_dl/postprocessor/ffmpeg.py @@ -585,7 +585,7 @@ class FFmpegSubtitlesConvertorPP(FFmpegPostProcessor): dfxp_file = old_file srt_file = subtitles_filename(filename, lang, 'srt') - with io.open(dfxp_file, 'rt', encoding='utf-8') as f: + with open(dfxp_file, 'rb') as f: srt_data = dfxp2srt(f.read()) with io.open(srt_file, 'wt', encoding='utf-8') as f: diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py index 9e4492d40..b724e0b70 100644 --- a/youtube_dl/utils.py +++ b/youtube_dl/utils.py @@ -2572,14 +2572,18 @@ def srt_subtitles_timecode(seconds): def dfxp2srt(dfxp_data): + ''' + @param dfxp_data A bytes-like object containing DFXP data + @returns A unicode object containing converted SRT data + ''' LEGACY_NAMESPACES = ( - ('http://www.w3.org/ns/ttml', [ - 'http://www.w3.org/2004/11/ttaf1', - 'http://www.w3.org/2006/04/ttaf1', - 'http://www.w3.org/2006/10/ttaf1', + (b'http://www.w3.org/ns/ttml', [ + b'http://www.w3.org/2004/11/ttaf1', + b'http://www.w3.org/2006/04/ttaf1', + b'http://www.w3.org/2006/10/ttaf1', ]), - ('http://www.w3.org/ns/ttml#styling', [ - 'http://www.w3.org/ns/ttml#style', + (b'http://www.w3.org/ns/ttml#styling', [ + b'http://www.w3.org/ns/ttml#style', ]), ) @@ -2674,7 +2678,7 @@ def dfxp2srt(dfxp_data): for ns in v: dfxp_data = dfxp_data.replace(ns, k) - dfxp = compat_etree_fromstring(dfxp_data.encode('utf-8')) + dfxp = compat_etree_fromstring(dfxp_data) out = [] paras = dfxp.findall(_x('.//ttml:p')) or dfxp.findall('.//p')