Merge with http://www.fi.muni.cz/~xbabinc/elinks/elinks-utf8.git/

without ucdata stuff. UTF-8 code cleanup. Added Pavol Babincak
to the AUTHORS

AUTHORS

@@ -423,6 +423,9 @@ Omar Khayam <omark@cyentec.com>
 <otte@duke.edu>
 	Fix stdin reading on Mac OS X
 
+Pavol Babincak <scroolik@gmail.com>
+	Improved UTF-8 support with double-width chars
+
 Peder Stray <peder@ifi.uio.no>
 	Fix handling of key presses turning up as key prefixes
 

features.conf

@@ -601,6 +601,27 @@ CONFIG_OWN_LIBC=no
 CONFIG_SMALL=no
 
 
+### Unicode UTF-8 support
+#
+# By enabling this option you get better Unicode support. At present only some
+# parts of ELinks are influenced with this. It includes DOM, plain, HTML
+# renderer and user interface. Beside normal Unicode characters there is
+# support for double-width characters (like Japanese, etc.).
+#
+# Some features of Unicode are not handled at all. Combining characters is
+# most visible absence.
+# Some features are partially supported. Like line breaking between
+# double-width characters. There is no other detection for determining when to
+# break or not.
+#
+# Note: This UTF-8 support is experimental.
+#
+# Default: disabled
+
+CONFIG_UTF_8=no
+
+
+
 ### Back-trace Printing
 #
 # Once upon a time, a disaster happens and ELinks crashes. That is a very sad

src/bfu/dialog.c

@@ -100,20 +100,19 @@ redraw_dialog(struct dialog_data *dlg_data, int layout)
 			int titlelen = strlen(title);
 			int titlecells = titlelen;
 			int x, y;
+
 #ifdef CONFIG_UTF_8
 			if (term->utf8)
 				titlecells = utf8_ptr2cells(title, 
-							&title[titlelen]);
+							    &title[titlelen]);
 #endif /* CONFIG_UTF_8 */
 
 			titlecells = int_min(box.width - 2, titlecells);
 
 #ifdef CONFIG_UTF_8
-			if (term->utf8) {
+			if (term->utf8)
-				titlelen = utf8_cells2bytes(title,
+				titlelen = utf8_cells2bytes(title, titlecells,
-							    titlecells,
 							    NULL);
-			}
 #endif /* CONFIG_UTF_8 */
 
 			x = (box.width - titlecells) / 2 + box.x;

src/bfu/text.c

@@ -51,6 +51,7 @@ split_line(unsigned char *text, int max_width, int *cells)
 
 	while (*split && *split != '\n') {
 		unsigned char *next_split;
+
 #ifdef CONFIG_UTF_8
 		if (utf8) {
 			unsigned char *next_char_begin = split 

src/intl/charsets.c

@@ -215,8 +215,7 @@ encode_utf_8(unicode_val_T u)
 #ifdef CONFIG_UTF_8
 /* Number of bytes utf8 character indexed by first byte. Illegal bytes are
  * equal ones and handled different. */
-static char utf8char_len_tab[256] =
+static char utf8char_len_tab[256] = {
-{
 	1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,
 	1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,
 	1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1, 1,1,1,1,1,1,1,1,
@@ -229,11 +228,7 @@ static char utf8char_len_tab[256] =
 
 inline int utf8charlen(const unsigned char *p)
 {
-	int len;
+	return p ? utf8char_len_tab[*p] : 0;
-	if (p==NULL)
-		return 0;
-	len = utf8char_len_tab[*p];
-	return len;
 }
 
 inline int
@@ -260,14 +255,14 @@ strlen_utf8(unsigned char **str)
 inline unsigned char *
 utf8_prevchar(unsigned char *current, int pos, unsigned char *start)
 {
-  if (current == NULL || start == NULL || pos < 0)
+	if (current == NULL || start == NULL || pos < 0)
-    return NULL;
+		return NULL;
-  while (pos > 0 && current != start) {
+	while (pos > 0 && current != start) {
-    current--;
+		current--;
-    if (utf8_islead(*current))
+		if (utf8_islead(*current))
-      pos--;
+			pos--;
-  }
+	}
-  return current;
+	return current;
 }
 
 /* Count number of standard terminal cells needed for displaying UTF-8

src/terminal/draw.c

@@ -118,9 +118,9 @@ draw_char_data(struct terminal *term, int x, int y, unsigned char data)
 #ifdef CONFIG_UTF_8
 #ifdef CONFIG_DEBUG
 	/* Detect attempt to draw double-width char on the last
-	 * collumn of terminal. */
+	 * column of terminal. */
 	if (unicode_to_cell(data) == 2 && x + 1 > term->width)
-		INTERNAL("Attempt to draw double-width glyph on last collumn!");
+		INTERNAL("Attempt to draw double-width glyph on last column!");
 #endif /* CONFIG_DEBUG */
 
 	if (data == UCS_NO_CHAR)
@@ -161,7 +161,7 @@ draw_line(struct terminal *term, int x, int y, int l, struct screen_char *line)
 			screen_char++;
 
 		}
-		/* Instead of displaying double-width character at last collumn
+		/* Instead of displaying double-width character at last column
 		 * display only space. */
 		if (size - 1 > 0 && unicode_to_cell(line[size - 1].data) == 2) {
 			sc = &line[size - 1];

src/terminal/event.c

@@ -251,9 +251,7 @@ handle_interlink_event(struct terminal *term, struct term_event *ev)
 
 	case EVENT_KBD:
 	{
-#ifndef CONFIG_UTF_8
 		int utf8_io = -1;
-#endif /* CONFIG_UTF_8 */
 		int key = get_kbd_key(ev);
 
 		reset_timer();
@@ -267,14 +265,14 @@ handle_interlink_event(struct terminal *term, struct term_event *ev)
 			return 0;
 		}
 
-		if (interlink->utf_8.len) {
 #ifdef CONFIG_UTF_8
-			if ((key & 0xC0) == 0x80 && term->utf8)
+		utf8_io = !!term->utf8;
 #else
-			utf8_io = get_opt_bool_tree(term->spec, "utf_8_io");
+		utf8_io = get_opt_bool_tree(term->spec, "utf_8_io");
-			if ((key & 0xC0) == 0x80 && utf8_io)
 #endif /* CONFIG_UTF_8 */
-			{
+
+		if (interlink->utf_8.len) {
+			if ((key & 0xC0) == 0x80 && utf8_io) {
 				interlink->utf_8.ucs <<= 6;
 				interlink->utf_8.ucs |= key & 0x3F;
 				if (! --interlink->utf_8.len) {
@@ -292,15 +290,7 @@ handle_interlink_event(struct terminal *term, struct term_event *ev)
 			}
 		}
 
-#ifdef CONFIG_UTF_8
+		if (key < 0x80 || key > 0xFF || !utf8_io) {
-		if (key < 0x80 || key > 0xFF || !term->utf8)
-#else
-		if (key < 0x80 || key > 0xFF
-				|| (utf8_io == -1
-					? !get_opt_bool_tree(term->spec, "utf_8_io")
-					: !utf8_io))
-#endif /* CONFIG_UTF_8 */
-		{
 			term_send_event(term, ev);
 			break;