Internal screen data rewrite for better 256 colour/UTF-8 support.

1 files changed, 53 insertions, 80 deletions

diff --git a/utf8.c b/utf8.c
index 77179e77..5e1c6b94 100644
--- a/utf8.c
+++ b/utf8.c
@@ -1,4 +1,4 @@
-/* $Id: utf8.c,v 1.1 2008-09-09 22:16:37 nicm Exp $ */
+/* $Id: utf8.c,v 1.2 2008-09-25 20:08:56 nicm Exp $ */
 
 /*
  * Copyright (c) 2008 Nicholas Marriott <nicm@users.sourceforge.net>
@@ -22,93 +22,66 @@
 
 #include "tmux.h"
 
-/*
- * UTF8 data structures. Just crappy array + linear search for now.
- */
-
-/* Pack UTF8 index into attr, data. */
-void
-utf8_pack(int idx, u_char *data, u_short *attr)
-{
-	*data = idx & 0xff;
-
-	*attr &= ~(ATTR_UTF8b8|ATTR_UTF8b9);
-	if (idx & 0x100)
-		*attr |= ATTR_UTF8b8;
-	if (idx & 0x200)
-		*attr |= ATTR_UTF8b9;
-	if (idx & 0x400)
-		*attr |= ATTR_UTF8b10;
-	if (idx & 0x800)
-		*attr |= ATTR_UTF8b11;
-}
-
-/* Unpack UTF8 index from attr, data. */
-int
-utf8_unpack(u_char data, u_short attr)
+u_int
+utf8_combine(const u_char data[4])
 {
-	int	idx;
-
-	idx = data;
-	if (attr & ATTR_UTF8b8)
-		idx |= 0x100;
-	if (attr & ATTR_UTF8b9)
-		idx |= 0x200;
-	if (attr & ATTR_UTF8b10)
-		idx |= 0x400;
-	if (attr & ATTR_UTF8b11)
-		idx |= 0x800;
-
-	return (idx);
-}
-
-void
-utf8_init(struct utf8_table *utab, int limit)
-{
-	utab->limit = limit;
-	ARRAY_INIT(&utab->array);
+	u_int	uv;
+
+	if (data[1] == 0xff)
+		uv = data[0];
+	else if (data[2] == 0xff) {
+		uv = data[1] & 0x3f;
+		uv |= (data[0] & 0x1f) << 6;
+	} else if (data[3] == 0xff) {
+		uv = data[2] & 0x3f;
+		uv |= (data[1] & 0x3f) << 6;
+		uv |= (data[0] & 0x0f) << 12;
+	} else {
+		uv = data[3] & 0x3f;
+		uv |= (data[2] & 0x3f) << 6;
+		uv |= (data[1] & 0x3f) << 12;
+		uv |= (data[0] & 0x3f) << 18;
+	}
+	return (uv);
 }
 
 void
-utf8_free(struct utf8_table *utab)
+utf8_split(u_int uv, u_char data[4])
 {
-	ARRAY_FREE(&utab->array);
-}
-
-struct utf8_data *
-utf8_lookup(struct utf8_table *utab, int idx)
-{
-	if (idx < 0 || idx >= (int) ARRAY_LENGTH(&utab->array))
-		return (NULL);
-	return (&ARRAY_ITEM(&utab->array, idx));
-}
-
-int
-utf8_search(struct utf8_table *utab, struct utf8_data *udat)
-{
-	u_int	idx;
-
-	for (idx = 0; idx < ARRAY_LENGTH(&utab->array); idx++) {
-		if (memcmp(udat->data,
-		    ARRAY_ITEM(&utab->array, idx).data, sizeof udat->data) == 0)
-			return (idx);
+	memset(data, 0xff, sizeof data);
+
+	if (uv <= 0x7f)
+		data[0] = uv;
+	else if (uv > 0x7f && uv <= 0x7ff) {
+		data[0] = (uv >> 6) | 0xc0;
+		data[1] = (uv & 0x3f) | 0x80;
+	} else if (uv > 0x7ff && uv <= 0xffff) {
+		data[0] = (uv >> 12) | 0xe0;
+		data[1] = ((uv >> 6) & 0x3f) | 0x80;
+		data[2] = (uv & 0x3f) | 0x80;
+	} else if (uv > 0xffff && uv <= 0x10ffff) {
+		data[0] = (uv >> 18) | 0xf0;
+		data[1] = ((uv >> 12) & 0x3f) | 0x80;
+		data[2] = ((uv >> 6) & 0x3f) | 0x80;
+		data[3] = (uv & 0x3f) | 0x80;
 	}
-	return (-1);
 }
 
 int
-utf8_add(struct utf8_table *utab, struct utf8_data *udat)
+utf8_width(u_int uv)
 {
-	int	idx;
-
-	if (ARRAY_LENGTH(&utab->array) == utab->limit)
-		return (-1);
-
-	if ((idx = utf8_search(utab, udat)) != -1)
-		return (idx);
-
-	ARRAY_EXPAND(&utab->array, 1);
-	memcpy(
-	    &ARRAY_LAST(&utab->array), udat, sizeof ARRAY_LAST(&utab->array));
-	return (ARRAY_LENGTH(&utab->array) - 1);
+	if ((uv >= 0x1100 && uv <= 0x115f) ||
+	    uv == 0x2329 ||
+	    uv == 0x232a ||
+	    (uv >= 0x2e80 && uv <= 0xa4cf && uv != 0x303f) ||
+	    (uv >= 0xac00 && uv <= 0xd7a3) ||
+	    (uv >= 0xf900 && uv <= 0xfaff) ||
+	    (uv >= 0xfe10 && uv <= 0xfe19) ||
+	    (uv >= 0xfe30 && uv <= 0xfe6f) ||
+	    (uv >= 0xff00 && uv <= 0xff60) ||
+	    (uv >= 0xffe0 && uv <= 0xffe6) ||
+	    (uv >= 0x20000 && uv <= 0x2fffd) ||
+	    (uv >= 0x30000 && uv <= 0x3fffd))
+		return (2);
+	return (1);
 }