From 231980963d3145c5edd80ddff11ee28266f6c290 Mon Sep 17 00:00:00 2001 From: doc.support Date: Fri, 10 Aug 2012 11:09:22 +0200 Subject: [PATCH] Bad unicode chars for snippets display. During the snippets process, indexes are computed on a non unicode chain. The result of the grep call is now converted to unicode before split. --- modules/bibformat/lib/bibformat_utils.py | 4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/modules/bibformat/lib/bibformat_utils.py b/modules/bibformat/lib/bibformat_utils.py index 6483064..a6aa8bd 100644 --- a/modules/bibformat/lib/bibformat_utils.py +++ b/modules/bibformat/lib/bibformat_utils.py @@ -684,6 +684,8 @@ def get_text_snippets(textfile_path, patterns, nb_words_around, max_snippets, \ p1 = Popen(sed_call, stdout=PIPE) p2 = Popen(grep_call, stdin=p1.stdout, stdout=PIPE) output = p2.communicate()[0] + if not isinstance(output, unicode): + output = output.decode('utf-8') result = [] big_snippets = output.split("--") @@ -706,7 +708,7 @@ def get_text_snippets(textfile_path, patterns, nb_words_around, max_snippets, \ out += "
" out += "..." + snippet + "..." count += 1 - return out + return out.encode('utf-8') -- 1.7.2.5