vis: improve replacement of combining characters

3 files changed, 32 insertions, 6 deletions

diff --git a/text-util.c b/text-util.c
index 2b48719..4d3406b 100644
--- a/text-util.c
+++ b/text-util.c
@@ -1,5 +1,7 @@
 #include "text-util.h"
 #include "util.h"
+#include <wchar.h>
+#include <errno.h>
 
 bool text_range_valid(Filerange *r) {
 	return r->start != EPOS && r->end != EPOS && r->start <= r->end;
@@ -46,3 +48,30 @@ bool text_range_overlap(Filerange *r1, Filerange *r2) {
 bool text_range_contains(Filerange *r, size_t pos) {
 	return text_range_valid(r) && r->start <= pos && pos <= r->end;
 }
+
+int text_char_count(const char *data, size_t len) {
+	int count = 0;
+	mbstate_t ps = { 0 };
+	while (len > 0) {
+		wchar_t wc;
+		size_t wclen = mbrtowc(&wc, data, len, &ps);
+		if (wclen == (size_t)-1 && errno == EILSEQ) {
+			count++;
+			while (!ISUTF8(*data))
+				data++, len--;
+		} else if (wclen == (size_t)-2) {
+			break;
+                } else if (wclen == 0) {
+			count++;
+			data++;
+			len--;
+		} else {
+			int width = wcwidth(wc);
+			if (width != 0)
+				count++;
+			data += wclen;
+			len -= wclen;
+                }
+	}
+	return count;
+}
diff --git a/text-util.h b/text-util.h
index 80affd5..9bfa705 100644
--- a/text-util.h
+++ b/text-util.h
@@ -21,5 +21,7 @@ bool text_range_equal(Filerange*, Filerange*);
 bool text_range_overlap(Filerange*, Filerange*);
 /* test whether a given position is within a certain range */
 bool text_range_contains(Filerange*, size_t pos);
+/* count the number of graphemes in data */
+int text_char_count(const char *data, size_t len);
 
 #endif
diff --git a/vis.c b/vis.c
index c84b415..03dcc61 100644
--- a/vis.c
+++ b/vis.c
@@ -435,14 +435,9 @@ void vis_insert_key(Vis *vis, const char *data, size_t len) {
 }
 
 void vis_replace(Vis *vis, size_t pos, const char *data, size_t len) {
-	size_t chars = 0;
-	for (size_t i = 0; i < len; i++) {
-		if (ISUTF8(data[i]))
-			chars++;
-	}
-
 	Text *txt = vis->win->file->text;
 	Iterator it = text_iterator_get(txt, pos);
+	int chars = text_char_count(data, len);
 	for (char c; chars-- > 0 && text_iterator_byte_get(&it, &c) && c != '\r' && c != '\n'; )
 		text_iterator_char_next(&it, NULL);