Merge remote-tracking branch 'upstream/master' into mix_20240309

1 files changed, 8 insertions, 5 deletions

diff --git a/src/nvim/mbyte.c b/src/nvim/mbyte.c
index 01e720283e..65f718f925 100644
--- a/src/nvim/mbyte.c
+++ b/src/nvim/mbyte.c
@@ -454,9 +454,6 @@ static bool prop_is_emojilike(const utf8proc_property_t *prop)
 /// Is only correct for characters >= 0x80.
 /// When p_ambw is "double", return 2 for a character with East Asian Width
 /// class 'A'(mbiguous).
-///
-/// @note Tables `doublewidth` and `ambiguous` are generated by
-///       gen_unicode_tables.lua, which must be manually invoked as needed.
 int utf_char2cells(int c)
 {
   if (c < 0x80) {
@@ -842,6 +839,13 @@ bool utf_composinglike(const char *p1, const char *p2, GraphemeState *state)
   return arabic_combine(first, second);
 }
 
+/// same as utf_composinglike but operating on UCS-4 values
+bool utf_iscomposing(int c1, int c2, GraphemeState *state)
+{
+  return (!utf8proc_grapheme_break_stateful(c1, c2, state)
+          || arabic_combine(c1, c2));
+}
+
 /// Get the screen char at the beginning of a string
 ///
 /// Caller is expected to check for things like unprintable chars etc
@@ -1855,8 +1859,7 @@ StrCharInfo utfc_next_impl(StrCharInfo cur)
   while (true) {
     uint8_t const next_len = utf8len_tab[*next];
     int32_t const next_code = utf_ptr2CharInfo_impl(next, (uintptr_t)next_len);
-    if (utf8proc_grapheme_break_stateful(prev_code, next_code, &state)
-        && !arabic_combine(prev_code, next_code)) {
+    if (!utf_iscomposing(prev_code, next_code, &state)) {
       return (StrCharInfo){
         .ptr = (char *)next,
         .chr = (CharInfo){ .value = next_code, .len = (next_code < 0 ? 1 : next_len) },