[PATCH] diff-highlight: Fix broken multibyte string

From: Yi EungJun <semtlenori@gmail.com>
Date: Mar 30, 2015, 15:55 UTC
Message-ID: <1427730933-26189-1-git-send-email-eungjun.yi@navercorp.com>

From: Yi EungJun <eungjun.yi@navercorp.com>

Highlighted string might be broken if the common subsequence is a proper subset of a multibyte character. For example, if the old string is "진" and the new string is "지", then we expect the diff is rendered as follows:

	-진
	+지

but actually it was rendered as follows:

    -<EC><A7><84>
    +<EC><A7><80>

This fixes the bug by splitting the string by multibyte characters.
---
 contrib/diff-highlight/diff-highlight | 25 +++++++++++++++++++++++--
 1 file changed, 23 insertions(+), 2 deletions(-)

diff --git a/contrib/diff-highlight/diff-highlight b/contrib/diff-highlight/diff-highlight
index 08c88bb..2662c1a 100755
--- a/contrib/diff-highlight/diff-highlight
+++ b/contrib/diff-highlight/diff-highlight
@@ -2,6 +2,9 @@
 
 use warnings FATAL => 'all';
 use strict;
+use File::Basename;
+use File::Spec::Functions qw( catdir );
+use String::Multibyte;
 
 # Highlight by reversing foreground and background. You could do
 # other things like bold or underline if you prefer.
@@ -24,6 +27,8 @@ my @removed;
 my @added;
 my $in_hunk;
 
+my $mbcs = get_mbcs();
+
 # Some scripts may not realize that SIGPIPE is being ignored when launching the
 # pager--for instance scripts written in Python.
 $SIG{PIPE} = 'DEFAULT';
@@ -164,8 +169,8 @@ sub highlight_pair {
 
 sub split_line {
 	local $_ = shift;
-	return map { /$COLOR/ ? $_ : (split //) }
-	       split /($COLOR*)/;
+	return map { /$COLOR/ ? $_ : ($mbcs ? $mbcs->strsplit('', $_) : split //) }
+	       split /($COLOR)/;
 }
 
 sub highlight_line {
@@ -211,3 +216,19 @@ sub is_pair_interesting {
 	       $suffix_a !~ /^$BORING*$/ ||
 	       $suffix_b !~ /^$BORING*$/;
 }
+
+# Returns an instance of String::Multibyte based on the charset defined by
+# i18n.commitencoding or UTF-8, or undef if String::Multibyte doesn't support
+# the charset.
+sub get_mbcs {
+	my $dir = catdir(dirname($INC{'String/Multibyte.pm'}), 'Multibyte');
+	opendir my $dh, $dir or return;
+	my @mbcs_charsets = grep s/[.]pm\z//, readdir $dh;
+	close $dh;
+	my $expected_charset = `git config i18n.commitencoding` || "UTF-8";
+	$expected_charset =~ s/-//g;
+	my @matches = grep {/^$expected_charset$/i} @mbcs_charsets;
+	my $charset = shift @matches;
+
+	return eval 'String::Multibyte->new($charset)';
+}

-- 
2.3.2.209.gd67f9d5.dirty