From 2b6587fc6cf0bc144393a46fcf12967e8cdcbde6 Mon Sep 17 00:00:00 2001
From: hobbs <hobbs>
Date: Thu, 28 Jun 2001 01:10:15 +0000
Subject: 	* tests/subst.test: 	* generic/tclUtf.c (Tcl_UtfBackslash):
 Corrected backslash 	handling of multibyte utf-8 chars. [Bug #217987]

---
 generic/tclUtf.c | 23 +++++++++++++++++------
 tests/subst.test |  7 ++++++-
 2 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/generic/tclUtf.c b/generic/tclUtf.c
index 2a6c217..8b39d5f 100644
--- a/generic/tclUtf.c
+++ b/generic/tclUtf.c
@@ -8,7 +8,7 @@
  * See the file "license.terms" for information on usage and redistribution
  * of this file, and for a DISCLAIMER OF ALL WARRANTIES.
  *
- * RCS: @(#) $Id: tclUtf.c,v 1.15 2001/04/06 10:50:00 dkf Exp $
+ * RCS: @(#) $Id: tclUtf.c,v 1.16 2001/06/28 01:10:15 hobbs Exp $
  */
 
 #include "tclInt.h"
@@ -111,7 +111,7 @@ static int UtfCount _ANSI_ARGS_((int ch));
  *---------------------------------------------------------------------------
  */
  
-static int
+INLINE static int
 UtfCount(ch)
     int ch;			/* The Tcl_UniChar whose size is returned. */
 {
@@ -781,7 +781,8 @@ Tcl_UtfBackslash(src, readPtr, dst)
 				 * backslash sequence. */
 {
     register CONST char *p = src+1;
-    int result, count, n;
+    Tcl_UniChar result;
+    int count, n;
     char buf[TCL_UTF_MAX];
 
     if (dst == NULL) {
@@ -883,15 +884,25 @@ Tcl_UtfBackslash(src, readPtr, dst)
 		result = (unsigned char)((result << 3) + (*p - '0'));
 		break;
 	    }
-	    result = *p;
-	    count = 2;
+	    if (UCHAR(*p) < UNICODE_SELF) {
+		result = *p;
+		count = 2;
+	    } else {
+		/*
+		 * We have to convert here because the user has put a
+		 * backslash in front of a multi-byte utf-8 character.
+		 * While this means nothing special, we shouldn't break up
+		 * a correct utf-8 character. [Bug #217987] test subst-3.2
+		 */
+		count = Tcl_UtfToUniChar(p, &result) + 1; /* +1 for '\' */
+	    }
 	    break;
     }
 
     if (readPtr != NULL) {
 	*readPtr = count;
     }
-    return Tcl_UniCharToUtf(result, dst);
+    return Tcl_UniCharToUtf((int) result, dst);
 }
 
 /*
diff --git a/tests/subst.test b/tests/subst.test
index 1ebceb7..21b0d7e 100644
--- a/tests/subst.test
+++ b/tests/subst.test
@@ -11,7 +11,7 @@
 # See the file "license.terms" for information on usage and redistribution
 # of this file, and for a DISCLAIMER OF ALL WARRANTIES.
 #
-# RCS: @(#) $Id: subst.test,v 1.7 2000/11/01 22:13:39 hobbs Exp $
+# RCS: @(#) $Id: subst.test,v 1.8 2001/06/28 01:10:15 hobbs Exp $
 
 if {[lsearch [namespace children] ::tcltest] == -1} {
     package require tcltest
@@ -38,6 +38,11 @@ test subst-2.3 {simple strings} {
 test subst-3.1 {backslash substitutions} {
     subst {\x\$x\[foo bar]\\}
 } "x\$x\[foo bar]\\"
+test subst-3.2 {backslash substitutions with utf chars} {
+    # 'j' is just a char that doesn't mean anything, and \344 is 'ä'
+    # that also doesn't mean anything, but is multi-byte in UTF-8.
+    list [subst \j] [subst \\j] [subst \\344] [subst \\\344]
+} "j j \344 \344"
 
 test subst-4.1 {variable substitutions} {
     set a 44
-- 
cgit v0.12