Backport: Bug [#2093]. Fix exception when inserting DBCS text.

Backport of changeset 7415:a7d3bc738c23.

2 files changed, 17 insertions, 8 deletions

diff --git a/doc/ScintillaHistory.html b/doc/ScintillaHistory.html
index 471fdf42d..3e3b5cacb 100644
--- a/doc/ScintillaHistory.html
+++ b/doc/ScintillaHistory.html
@@ -582,6 +582,10 @@
 	<a href="https://sourceforge.net/p/scintilla/bugs/2087/">Bug #2087</a>.
 	</li>
  	<li>
+	Fix exception when inserting DBCS text.
+	<a href="https://sourceforge.net/p/scintilla/bugs/2093/">Bug #2093</a>.
+	</li>
+ 	<li>
 	Avoid potential long hangs with idle styling for huge documents on Cocoa and GTK.
 	</li>
  	<li>
diff --git a/src/Editor.cxx b/src/Editor.cxx
index d67f97df7..d47ece75b 100644
--- a/src/Editor.cxx
+++ b/src/Editor.cxx
@@ -1900,6 +1900,9 @@ void Editor::FilterSelections() {
 
 // AddCharUTF inserts an array of bytes which may or may not be in UTF-8.
 void Editor::AddCharUTF(const char *s, unsigned int len, bool treatAsDBCS) {
+	if (len == 0) {
+		return;
+	}
 	FilterSelections();
 	{
 		UndoGroup ug(pdoc, (sel.Count() > 1) || !sel.Empty() || inOverstrike);
@@ -1970,12 +1973,14 @@ void Editor::AddCharUTF(const char *s, unsigned int len, bool treatAsDBCS) {
 		SetLastXChosen();
 	}
 
-	if (treatAsDBCS) {
-		NotifyChar((static_cast<unsigned char>(s[0]) << 8) |
-		        static_cast<unsigned char>(s[1]));
-	} else if (len > 0) {
-		int byte = static_cast<unsigned char>(s[0]);
-		if ((byte < 0xC0) || (1 == len)) {
+	int ch = static_cast<unsigned char>(s[0]);
+	if (treatAsDBCS || pdoc->dbcsCodePage != SC_CP_UTF8) {
+		if (len > 1) {
+			// DBCS code page or DBCS font character set.
+			ch |= (ch << 8) | static_cast<unsigned char>(s[1]);
+		}
+	} else {
+		if ((ch < 0xC0) || (1 == len)) {
 			// Handles UTF-8 characters between 0x01 and 0x7F and single byte
 			// characters when not in UTF-8 mode.
 			// Also treats \0 and naked trail bytes 0x80 to 0xBF as valid
@@ -1983,10 +1988,10 @@ void Editor::AddCharUTF(const char *s, unsigned int len, bool treatAsDBCS) {
 		} else {
 			unsigned int utf32[1] = { 0 };
 			UTF32FromUTF8(s, len, utf32, ELEMENTS(utf32));
-			byte = utf32[0];
+			ch = utf32[0];
 		}
-		NotifyChar(byte);
 	}
+	NotifyChar(ch);
 
 	if (recordingMacro) {
 		NotifyMacroRecord(SCI_REPLACESEL, 0, reinterpret_cast<sptr_t>(s));