From a2e301b52cc5bdb44540aa66860dc59fa1fa5a89 Mon Sep 17 00:00:00 2001
From: Paul Eggert <eggert@cs.ucla.edu>
Date: Tue, 6 Jun 2023 13:55:56 -0700
Subject: [PATCH] diff: switch from wchar_t to char32_t

Prefer C11-style char32_t to wchar_t, as char32_t works better on
platforms where wchar_t is only 16 bits.
* .gitignore: Add lib/uchar.h.
* bootstrap.conf (gnulib_modules): Add c32width, mbrtoc32.
Remove mbrtowc.  Sort.
* src/side.c: Include uchar.h instead of wchar.h.
(print_half_line): Use c32width and mbrtowc instead of
wcwidth and mbrtowc.
---
 .gitignore     |  1 +
 bootstrap.conf | 10 +++++-----
 src/side.c     |  8 ++++----
 3 files changed, 10 insertions(+), 9 deletions(-)

diff --git a/.gitignore b/.gitignore
index 9c28f11..0ee17cf 100644
--- a/.gitignore
+++ b/.gitignore
@@ -85,6 +85,7 @@ lib/uniwidth.h
 lib/warn-on-use.h
 lib/wchar.h
 lib/wctype.h
+lib/uchar.h
 man/*.1
 po/*.gmo
 po/*.po
diff --git a/bootstrap.conf b/bootstrap.conf
index 2560846..4b3f02d 100644
--- a/bootstrap.conf
+++ b/bootstrap.conf
@@ -24,11 +24,11 @@ avoided_gnulib_modules='
 # gnulib modules used by this package.
 gnulib_modules='
 announce-gen
-attribute
 argmatch
+attribute
 binary-io
-nullptr
 c-stack
+c32width
 config-h
 count-leading-zeros
 diffseq
@@ -40,10 +40,10 @@ exclude
 exitfail
 extensions
 extern-inline
-flexmember
 fcntl
 file-type
 filenamecat
+flexmember
 fnmatch-gnu
 getopt-gnu
 gettext-h
@@ -65,12 +65,13 @@ largefile
 lstat
 maintainer-makefile
 manywarnings
-mbrtowc
+mbrtoc32
 mempcpy
 minmax
 mkstemp
 mktime
 nstrftime
+nullptr
 perl
 progname
 propername-lite
@@ -101,7 +102,6 @@ update-copyright
 verify
 version-etc
 version-etc-fsf
-wcwidth
 xalloc
 xfreopen
 xmalloca
diff --git a/src/side.c b/src/side.c
index c4042dd..424d933 100644
--- a/src/side.c
+++ b/src/side.c
@@ -22,7 +22,7 @@
 
 #include "diff.h"
 
-#include <wchar.h>
+#include <uchar.h>
 
 static void print_sdiff_common_lines (lin, lin);
 static void print_sdiff_hunk (struct change *);
@@ -135,12 +135,12 @@ print_half_line (char const *const *line, intmax_t indent, intmax_t out_bound)
 
         default:
           {
-            wchar_t wc;
-            size_t bytes = mbrtowc (&wc, tp0, text_limit - tp0, &mbstate);
+            char32_t wc;
+            size_t bytes = mbrtoc32 (&wc, tp0, text_limit - tp0, &mbstate);
 
             if (0 < bytes && bytes < (size_t) -2)
               {
-                int width = wcwidth (wc);
+                int width = c32width (wc);
                 if (0 < width)
                   in_position += width;
                 if (in_position <= out_bound)