git: 160c36eae41a - main - localedata: add some exceptions to utf8proc widths

The branch main has been updated by kevans:

URL: https://cgit.FreeBSD.org/src/commit/?id=160c36eae41afa3c4944ed44778c2b48db8fbb77

commit 160c36eae41afa3c4944ed44778c2b48db8fbb77
Author:     Kyle Evans <kevans@FreeBSD.org>
AuthorDate: 2024-11-13 22:12:42 +0000
Commit:     Kyle Evans <kevans@FreeBSD.org>
CommitDate: 2024-11-13 22:12:42 +0000

    localedata: add some exceptions to utf8proc widths
    
    Hangul Jamo medial vowels and final consonants are reportedly combining
    characters that won't take up any columns on their own and should be
    reported as zero-width, so add an exception for these as well to reflect
    how they work in practice.  This conforms to how other implementations
    (e.g., glibc) treat these characters.
    
    Reviewed by:    bapt (earlier version), jkim
    Sponsored by:   Klara, Inc.
    Differential Revision:  https://reviews.freebsd.org/D47472
---
 tools/tools/locale/tools/getwidths.c | 18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

diff --git a/tools/tools/locale/tools/getwidths.c b/tools/tools/locale/tools/getwidths.c
index 2790b8031912..63c62791253f 100644
--- a/tools/tools/locale/tools/getwidths.c
+++ b/tools/tools/locale/tools/getwidths.c
@@ -28,6 +28,21 @@
 
 #include <utf8proc.h>
 
+static int
+width_of(int32_t wc)
+{
+
+	/*
+	 * Hangul Jamo medial vowels and final consonants are more of
+	 * a combining character, and should be considered zero-width.
+	 */
+	if (wc >= 0x1160 && wc <= 0x11ff)
+		return (0);
+
+	/* No override by default, trust utf8proc's width. */
+	return (utf8proc_charwidth(wc));
+}
+
 int
 main(void)
 {
@@ -43,9 +58,10 @@ main(void)
 		wcc = utf8proc_category(wc);
 		if (wcc == UTF8PROC_CATEGORY_CC)
 			continue;
-		wcw = utf8proc_charwidth(wc);
+		wcw = width_of(wc);
 		if (wcw == 1)
 			continue;
+
 		printf("%04X %d\n", wc, wcw);
 	}