Rewrite combined character handling to be more consistent and to support

newer Unicode combined characters (which we have to "know" are combined
since they are not width zero). GitHub issue 3600.

1 files changed, 4 insertions, 4 deletions

diff --git a/utf8.c b/utf8.c
index 10ccf422..d26a49e4 100644
--- a/utf8.c
+++ b/utf8.c
@@ -136,8 +136,8 @@ utf8_from_data(const struct utf8_data *ud, utf8_char *uc)
 		goto fail;
 	if (ud->size <= 3) {
 		index = (((utf8_char)ud->data[2] << 16)|
-		          ((utf8_char)ud->data[1] << 8)|
-		          ((utf8_char)ud->data[0]));
+			  ((utf8_char)ud->data[1] << 8)|
+			  ((utf8_char)ud->data[0]));
 	} else if (utf8_put_item(ud->data, ud->size, &index) != 0)
 		goto fail;
 	*uc = UTF8_SET_SIZE(ud->size)|UTF8_SET_WIDTH(ud->width)|index;
@@ -226,9 +226,9 @@ utf8_width(struct utf8_data *ud, int *width)
 	case 0:
 		return (UTF8_ERROR);
 	}
-	log_debug("UTF-8 %.*s is %08X", (int)ud->size, ud->data, (u_int)wc);
+	log_debug("UTF-8 %.*s is %05X", (int)ud->size, ud->data, (u_int)wc);
 	*width = wcwidth(wc);
-	log_debug("wcwidth(%08X) returned %d", (u_int)wc, *width);
+	log_debug("wcwidth(%05X) returned %d", (u_int)wc, *width);
 	if (*width < 0) {
 		/*
 		 * C1 control characters are nonprintable, so they are always