From 3051076dd16643a683a860c8ff5a24b47559261e Mon Sep 17 00:00:00 2001
From: nicm <nicm>
Date: Mon, 3 Nov 2025 09:27:06 +0000
Subject: [PATCH] Ignore Hangul filler character. There doesn't seem to be much
 agreement on what to do with this but ignoring it seems rightand does improve
 things. GitHub issue 3998.

---
 screen-write.c  |  4 ++++
 tmux.h          |  1 +
 utf8-combined.c | 13 +++++++++++--
 3 files changed, 16 insertions(+), 2 deletions(-)

diff --git a/screen-write.c b/screen-write.c
index 1014ae80..8e04681a 100644
--- a/screen-write.c
+++ b/screen-write.c
@@ -2006,6 +2006,10 @@ screen_write_combine(struct screen_write_ctx *ctx, const struct grid_cell *gc)
 	struct tty_ctx		 ttyctx;
 	int			 force_wide = 0, zero_width = 0;
 
+	/* Ignore U+3164 HANGUL_FILLER entirely. */
+	if (utf8_is_hangul_filler(ud))
+		return (1);
+
 	/*
 	 * Is this character which makes no sense without being combined? If
 	 * this is true then flag it here and discard the character (return 1)
diff --git a/tmux.h b/tmux.h
index aa25a408..356d1157 100644
--- a/tmux.h
+++ b/tmux.h
@@ -3499,6 +3499,7 @@ int		 utf8_cstrhas(const char *, const struct utf8_data *);
 int		 utf8_has_zwj(const struct utf8_data *);
 int		 utf8_is_zwj(const struct utf8_data *);
 int		 utf8_is_vs(const struct utf8_data *);
+int		 utf8_is_hangul_filler(const struct utf8_data *);
 int		 utf8_should_combine(const struct utf8_data *,
 		    const struct utf8_data *);
 enum hanguljamo_state hanguljamo_check_state(const struct utf8_data *,
diff --git a/utf8-combined.c b/utf8-combined.c
index 91ddaf75..635ae92c 100644
--- a/utf8-combined.c
+++ b/utf8-combined.c
@@ -55,7 +55,7 @@ utf8_has_zwj(const struct utf8_data *ud)
 	return (memcmp(ud->data + ud->size - 3, "\342\200\215", 3) == 0);
 }
 
-/* Is this a zero width joiner? */
+/* Is this zero width joiner U+200D? */
 int
 utf8_is_zwj(const struct utf8_data *ud)
 {
@@ -64,7 +64,7 @@ utf8_is_zwj(const struct utf8_data *ud)
 	return (memcmp(ud->data, "\342\200\215", 3) == 0);
 }
 
-/* Is this a variation selector? */
+/* Is this variation selector U+FE0F? */
 int
 utf8_is_vs(const struct utf8_data *ud)
 {
@@ -73,6 +73,15 @@ utf8_is_vs(const struct utf8_data *ud)
 	return (memcmp(ud->data, "\357\270\217", 3) == 0);
 }
 
+/* Is this Hangul filler U+3164? */
+int
+utf8_is_hangul_filler(const struct utf8_data *ud)
+{
+	if (ud->size != 3)
+		return (0);
+	return (memcmp(ud->data, "\343\205\244", 3) == 0);
+}
+
 /* Should these two characters combine? */
 int
 utf8_should_combine(const struct utf8_data *with, const struct utf8_data *add)