Some changes to make doublewidth chars possibly work (or possibly horribly

break whole utf8 support :)


git-svn-id: http://svn.irssi.org/repos/irssi/trunk@2956 dbcabf3a-b0e7-0310-adc4-f8d773084564

4 files changed, 120 insertions, 17 deletions

diff --git a/src/fe-text/gui-entry.c b/src/fe-text/gui-entry.c
index e7e5fcc5..f6bfb27d 100644
--- a/src/fe-text/gui-entry.c
+++ b/src/fe-text/gui-entry.c
@@ -104,7 +104,11 @@ static void gui_entry_draw_from(GUI_ENTRY_REC *entry, int pos)
 
 	p = entry->scrstart + pos < entry->text_len ?
 		entry->text + entry->scrstart + pos : empty_str;
-	for (; *p != '\0' && xpos < end_xpos; p++, xpos++) {
+	for (; *p != '\0'; p++) {
+		xpos += utf8_width(*p);
+		if (xpos > end_xpos)
+			break;
+
 		if (entry->hidden)
                         term_addch(root_window, ' ');
 		else if (*p >= 32 && (entry->utf8 || (*p & 127) >= 32))
diff --git a/src/fe-text/textbuffer-view.c b/src/fe-text/textbuffer-view.c
index a3cbe460..77b21a2e 100644
--- a/src/fe-text/textbuffer-view.c
+++ b/src/fe-text/textbuffer-view.c
@@ -152,8 +152,10 @@ view_update_line_cache(TEXT_BUFFER_VIEW_REC *view, LINE_REC *line)
 	LINE_CACHE_SUB_REC *sub;
 	GSList *lines;
         unsigned char cmd;
-	const unsigned char *ptr, *last_space_ptr;
+	const unsigned char *ptr, *next_ptr, *last_space_ptr;
 	int xpos, pos, indent_pos, last_space, last_color, color, linecount;
+	int char_len;
+	unichar chr;
 
 	g_return_val_if_fail(line->text != NULL, NULL);
 
@@ -196,7 +198,24 @@ view_update_line_cache(TEXT_BUFFER_VIEW_REC *view, LINE_REC *line)
 			continue;
 		}
 
-		if (xpos == view->width && sub != NULL &&
+		if (!view->utf8) {
+			next_ptr = ptr+1;
+			char_len = 1;
+		} else {
+			char_len = 1;
+			while (ptr[char_len] != '\0' && char_len < 6)
+				char_len++;
+
+			next_ptr = ptr;
+			chr = get_utf8_char(&next_ptr, char_len);
+			if (chr < 0)
+				char_len = 1;
+			else
+				char_len = utf8_width(chr);
+			next_ptr++;
+		}
+
+		if (xpos + char_len > view->width && sub != NULL &&
 		    (last_space <= indent_pos || last_space <= 10) &&
 		    view->longword_noindent) {
                         /* long word, remove the indentation from this line */
@@ -204,7 +223,7 @@ view_update_line_cache(TEXT_BUFFER_VIEW_REC *view, LINE_REC *line)
                         sub->indent = 0;
 		}
 
-		if (xpos == view->width) {
+		if (xpos + char_len > view->width) {
 			xpos = indent_func == NULL ? indent_pos :
 				indent_func(view, line, -1);
 
@@ -232,15 +251,14 @@ view_update_line_cache(TEXT_BUFFER_VIEW_REC *view, LINE_REC *line)
 			continue;
 		}
 
-		if (view->utf8)
-			get_utf8_char(&ptr, 6);
-
-		xpos++;
-		if (*ptr++ == ' ') {
-			last_space = xpos-1;
+		if (*ptr == ' ') {
+			last_space = xpos;
 			last_space_ptr = ptr;
 			last_color = color;
 		}
+
+		xpos += char_len;
+		ptr = next_ptr;
 	}
 
 	rec = g_malloc(sizeof(LINE_CACHE_REC)-sizeof(LINE_CACHE_SUB_REC) +
@@ -309,9 +327,9 @@ static int view_line_draw(TEXT_BUFFER_VIEW_REC *view, LINE_REC *line,
 {
         INDENT_FUNC indent_func;
 	LINE_CACHE_REC *cache;
-        const unsigned char *text, *text_newline;
+        const unsigned char *text, *end, *text_newline;
 	unsigned char *tmp;
-	int xpos, color, drawcount, first, need_move, need_clrtoeol;
+	int xpos, color, drawcount, first, need_move, need_clrtoeol, char_width;
 
 	if (view->dirty) /* don't bother drawing anything - redraw is coming */
                 return 0;
@@ -401,11 +419,16 @@ static int view_line_draw(TEXT_BUFFER_VIEW_REC *view, LINE_REC *line,
 			continue;
 		}
 
-		if (xpos < term_width) {
-			const unsigned char *end = text;
-			if (view->utf8)
-				get_utf8_char(&end, 6);
+		end = text;
+		if (view->utf8) {
+			unichar chr = get_utf8_char(&end, 6);
+			char_width = utf8_width(chr);
+		} else {
+			char_width = 1;
+		}
 
+		xpos += char_width;
+		if (xpos <= term_width) {
 			if (*text >= 32 &&
 			    (end != text || (*text & 127) >= 32)) {
 				for (; text < end; text++)
@@ -419,7 +442,6 @@ static int view_line_draw(TEXT_BUFFER_VIEW_REC *view, LINE_REC *line,
 			}
 		}
 		text++;
-		xpos++;
 	}
 
 	if (need_clrtoeol && xpos < term_width) {
diff --git a/src/fe-text/utf8.c b/src/fe-text/utf8.c
index 2d72fe2b..4049991d 100644
--- a/src/fe-text/utf8.c
+++ b/src/fe-text/utf8.c
@@ -7,6 +7,10 @@
  * Copyright (C) 1999 Tom Tromey
  * Copyright (C) 2000 Red Hat, Inc.
  *
+ * UTF-8 width tables based on locale data from GNU libc by
+ *
+ * Copyright (C) 1991-2002 Free Software Foundation, Inc.
+ *
  * This library is free software; you can redistribute it and/or
  * modify it under the terms of the GNU Lesser General Public
  * License as published by the Free Software Foundation; either
@@ -175,3 +179,73 @@ void utf16_to_utf8(const unichar *str, char *out)
 	}
 	*out = '\0';
 }
+
+static const unichar wcc[] = {
+	0x0, 0x300, 0x34F, 0x360, 0x363, 0x483, 0x487, 0x488, 0x48A, 0x591,
+	0x5A2, 0x5A3, 0x5BA, 0x5BB, 0x5BE, 0x5BF, 0x5C0, 0x5C1, 0x5C3, 0x5C4,
+	0x5C5, 0x64B, 0x656, 0x670, 0x671, 0x6D6, 0x6E5, 0x6E7, 0x6E9, 0x6EA,
+	0x6EE, 0x70F, 0x710, 0x711, 0x712, 0x730, 0x74B, 0x7A6, 0x7B1, 0x901,
+	0x903, 0x93C, 0x93D, 0x941, 0x949, 0x94D, 0x94E, 0x951, 0x955, 0x962,
+	0x964, 0x981, 0x982, 0x9BC, 0x9BD, 0x9C1, 0x9C5, 0x9CD, 0x9CE, 0x9E2,
+	0x9E4, 0xA02, 0xA03, 0xA3C, 0xA3D, 0xA41, 0xA43, 0xA47, 0xA49, 0xA4B,
+	0xA4E, 0xA70, 0xA72, 0xA81, 0xA83, 0xABC, 0xABD, 0xAC1, 0xAC6, 0xAC7,
+	0xAC9, 0xACD, 0xACE, 0xB01, 0xB02, 0xB3C, 0xB3D, 0xB3F, 0xB40, 0xB41,
+	0xB44, 0xB4D, 0xB4E, 0xB56, 0xB57, 0xB82, 0xB83, 0xBC0, 0xBC1, 0xBCD,
+	0xBCE, 0xC3E, 0xC41, 0xC46, 0xC49, 0xC4A, 0xC4E, 0xC55, 0xC57, 0xCBF,
+	0xCC0, 0xCC6, 0xCC7, 0xCCC, 0xCCE, 0xD41, 0xD44, 0xD4D, 0xD4E, 0xDCA,
+	0xDCB, 0xDD2, 0xDD5, 0xDD6, 0xDD7, 0xE31, 0xE32, 0xE34, 0xE3B, 0xE47,
+	0xE4F, 0xEB1, 0xEB2, 0xEB4, 0xEBA, 0xEBB, 0xEBD, 0xEC8, 0xECE, 0xF18,
+	0xF1A, 0xF35, 0xF36, 0xF37, 0xF38, 0xF39, 0xF3A, 0xF71, 0xF7F, 0xF80,
+	0xF85, 0xF86, 0xF88, 0xF90, 0xF98, 0xF99, 0xFBD, 0xFC6, 0xFC7, 0x102D,
+	0x1031, 0x1032, 0x1033, 0x1036, 0x1038, 0x1039, 0x103A, 0x1058, 0x105A,
+	0x1100, 0x1160, 0x17B7, 0x17BE, 0x17C6, 0x17C7, 0x17C9, 0x17D4, 0x180B,
+	0x180F, 0x18A9, 0x18AA, 0x200B, 0x2010, 0x202A, 0x202F, 0x206A, 0x2070,
+	0x20D0, 0x20E4, 0x2E80, 0x3008, 0x300C, 0x3014, 0x3016, 0x3018, 0x301C,
+	0x302A, 0x3030, 0x303F, 0x3041, 0x3095, 0x3099, 0x309B, 0xA4C7, 0xAC00,
+	0xD7A4, 0xF8F0, 0xF900, 0xFA2E, 0xFB1E, 0xFB1F, 0xFE20, 0xFE24, 0xFE30,
+	0xFE6C, 0xFEFF, 0xFF00, 0xFF01, 0xFF5F, 0xFFE0, 0xFFE7, 0xFFF9, 0xFFFC,
+#if 1
+	0x1D167, 0x1D16A, 0x1D173, 0x1D183, 0x1D185, 0x1D18C, 0x1D1AA, 0x1D1AE,
+	0x20000, 0x2A6D7, 0x2F800, 0x2FA1E, 0xE0001, 0xE0002, 0xE0020, 0xE0080
+#endif
+};
+
+static const int wccnum = sizeof(wcc) / sizeof(wcc[0]) - 1;
+
+static const char wws[] = {
+	1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0,
+	1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0,
+	1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0,
+	1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0,
+	1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0,
+	1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0,
+	1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0,
+	1, 2, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 2, 1, 2,
+	1, 2, 1, 2, 0, 2, 1, 2, 1, 0, 2, 1, 2, 1, 0, 2, 1, 0, 1, 0, 1, 2, 1, 0,
+	1, 2, 1, 2, 1, 0, 1, 0, 1, 0, 1, 0, 1, 0, 1, 2, 1, 2, 1, 0, 1, 0, 1, -1
+};
+
+int utf8_width(unichar c)
+{
+	int p, q, r;
+	unichar d;
+
+	if (c < wcc[1])
+		return 1;
+
+	p = 0;
+	q = wccnum;
+
+	while (p < q - 1) {
+		r = (p + q)/2;
+		d = wcc[r];
+		if (d < c)
+			p = r;
+		else if (d > c)
+			q = r;
+		else
+			return wws[r];
+	}
+
+	return wws[p];
+}
diff --git a/src/fe-text/utf8.h b/src/fe-text/utf8.h
index efc6af08..64c0e347 100644
--- a/src/fe-text/utf8.h
+++ b/src/fe-text/utf8.h
@@ -18,4 +18,7 @@ int utf16_char_to_utf8(unichar c, char *outbuf);
    Make sure out is at least 6 x length of str. */
 void utf16_to_utf8(const unichar *str, char *out);
 
+/* Returns width for character (0-2). */
+int utf8_width(unichar c);
+
 #endif