From 1abb9b9b690870c5f72d5c323f0213573b8edca4 Mon Sep 17 00:00:00 2001
From: Andrej Lajovic <s57ln@hamradio.si>
Date: Thu, 4 Oct 2012 12:55:12 -0500
Subject: [PATCH] UTF-8 wide characters

  * Fix put_echo_char() to handle UTF-8 characters wider than two bytes
  * Reimplement Fl_Text_Buffer_mod::get_char_at() to handle UTF-8 chars
    wider than two bytes
---
 src/dialogs/fl_digi.cxx                | 11 ++++++--
 src/widgets/Fl_Text_Buffer_mod_1_3.cxx | 39 +++++++++++++-------------
 2 files changed, 28 insertions(+), 22 deletions(-)

diff --git a/src/dialogs/fl_digi.cxx b/src/dialogs/fl_digi.cxx
index 59850d9d..744e3907 100644
--- a/src/dialogs/fl_digi.cxx
+++ b/src/dialogs/fl_digi.cxx
@@ -6364,9 +6364,14 @@ void put_echo_char(unsigned int data, int style)
 	if (asc != NULL) { // MAIL / ARQ / RTTY / CW
 		sch.assign(asc[data & 0xFF]);
 	} else if (data & 0x8000) { //UTF-8 extended character
-		sch.assign("  ");
-		sch[0] = (data >> 8) & 0xFF;
-		sch[1] = (data & 0xFF);
+		unsigned int shiftdata = data;
+		while (!(shiftdata & 0xff000000))
+			shiftdata <<= 8;
+		unsigned char c;
+		while ((c = (shiftdata >> 24) & 0xff)) {
+			sch += c;
+			shiftdata <<= 8;
+		}
 	} else { // keyboard character including MSB set chars
 		sch.assign(" ");
 		sch[0] = data;
diff --git a/src/widgets/Fl_Text_Buffer_mod_1_3.cxx b/src/widgets/Fl_Text_Buffer_mod_1_3.cxx
index a346d2c9..20c36129 100644
--- a/src/widgets/Fl_Text_Buffer_mod_1_3.cxx
+++ b/src/widgets/Fl_Text_Buffer_mod_1_3.cxx
@@ -258,29 +258,30 @@ unsigned int Fl_Text_Buffer_mod::char_at(int pos) const {
  Return a UTF-8 character at the given index.
  Pos must be at a character boundary.
  */
+
 unsigned int Fl_Text_Buffer_mod::get_char_at(int pos, int &len) const {  
-	if (pos < 0 || pos >= mLength) {
-		len = 1;
-		return 0;
-	}
+  if (pos < 0 || pos >= mLength)
+    return '\0';
   
-	IS_UTF8_ALIGNED2(this, (pos))
+  IS_UTF8_ALIGNED2(this, (pos))
   
-	const char *src = address(pos);
-	unsigned int code;
-	int codelen;
-	if (*src & 0x80) {              // what should be a multibyte encoding
-		fl_utf8decode(src, src+2, &codelen);
-		if (codelen == 2)
-			code = (*src << 8) | (*(src+1) & 0xFF);
-		else
-			code = *src & 0xFF;
-    } else {                      // handle the 1-byte utf8 encoding:
-      code = (*src & 0xFF);
-      codelen = 1;
+  const char *src = address(pos);
+  unsigned int code = 0;
+
+  if (*src & 0x80) {
+    fl_utf8decode(src, 0, &len);
+    
+    for (int i = 0; i < len && i < 4; i++) {
+      code <<= 8;
+      code |= (unsigned char)*(src + i) & 0xFF;
     }
-    len = codelen;
-	return code;
+  }
+  else {
+    code = *src;
+    len = 1;
+  }
+  
+  return code;
 }