From d3cd4ad933b033e750aa92801945f2cac213f88e Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Daniel=20Ekl=C3=B6f?= <daniel@ekloef.se>
Date: Thu, 31 Oct 2024 07:17:35 +0100
Subject: [PATCH] char32: use utf8proc_charwidth() instead of wcwidth(), when
 available

It appears to be slightly more up-to-date with recent Unicode
versions.

In particular, it handles the new "Symbols for Legacy Computing
Supplement" block, introduced in Unicode 16.

Closes #1865
---
 CHANGELOG.md |  5 +++++
 char32.h     | 15 +++++++++++++++
 2 files changed, 20 insertions(+)

diff --git a/CHANGELOG.md b/CHANGELOG.md
index d7e9939d..cc734b81 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -68,6 +68,11 @@
 * OSC-9: sequences beginning with `<number>;` are now ignored. These
   sequences are ConEmu/Windows Terminal sequences, and not intended to
   be notifications.
+* Use `utf8proc_charwidth()` instead of `wcwidth()`+`wcswidth()` when
+  calculating character width, when foot has been built with utf8proc
+  support ([#1865][1865]).
+
+[1865]: https://codeberg.org/dnkl/foot/issues/1865
 
 
 ### Deprecated
diff --git a/char32.h b/char32.h
index 6324c9a0..6a5eb080 100644
--- a/char32.h
+++ b/char32.h
@@ -8,6 +8,10 @@
 #include <wchar.h>
 #include <wctype.h>
 
+#if defined(FOOT_GRAPHEME_CLUSTERING)
+ #include <utf8proc.h>
+#endif
+
 static inline size_t c32len(const char32_t *s) {
     return wcslen((const wchar_t *)s);
 }
@@ -69,11 +73,22 @@ static inline bool isc32graph(char32_t c32) {
 }
 
 static inline int c32width(char32_t c) {
+#if defined(FOOT_GRAPHEME_CLUSTERING)
+    return utf8proc_charwidth((utf8proc_int32_t)c);
+#else
     return wcwidth((wchar_t)c);
+#endif
 }
 
 static inline int c32swidth(const char32_t *s, size_t n) {
+#if defined(FOOT_GRAPHEME_CLUSTERING)
+    int width = 0;
+    for (size_t i = 0; i < n; i++)
+        width += utf8proc_charwidth((utf8proc_int32_t)s[i]);
+    return width;
+#else
     return wcswidth((const wchar_t *)s, n);
+#endif
 }
 
 size_t mbsntoc32(char32_t *dst, const char *src, size_t nms, size_t len);