Merge 8.6 (Except encoding-12.6 test-case still to be handled)

3 files changed, 16 insertions, 13 deletions

diff --git a/generic/regc_lex.c b/generic/regc_lex.c
index a303ec6..6ef0a83 100644
--- a/generic/regc_lex.c
+++ b/generic/regc_lex.c
@@ -843,12 +843,18 @@ lexescape(
 	if (ISERR()) {
 	    FAILW(REG_EESCAPE);
 	}
-	if (i > 0xFFFF) {
+#if CHRBITS > 16
+	if ((unsigned)i > 0x10FFFF) {
+	    i = 0xFFFD;
+	}
+#else
+	if ((unsigned)i & ~0xFFFF) {
 	    /* TODO: output a Surrogate pair
 	     */
 	    i = 0xFFFD;
 	}
-	RETV(PLAIN, (uchr) i);
+#endif
+	RETV(PLAIN, (uchr)i);
 	break;
     case CHR('v'):
 	RETV(PLAIN, CHR('\v'));
diff --git a/generic/tclEncoding.c b/generic/tclEncoding.c
index 2e4c3f7..4789b7f 100644
--- a/generic/tclEncoding.c
+++ b/generic/tclEncoding.c
@@ -2557,11 +2557,6 @@ UtfToUtf16Proc(
 	}
 	src += TclUtfToUniChar(src, chPtr);
 
-	/*
-	 * Need to handle this in a way that won't cause misalignment by
-	 * casting dst to a Tcl_UniChar. [Bug 1122671]
-	 */
-
 	if (clientData) {
 #if TCL_UTF_MAX > 3
 	    if (*chPtr <= 0xFFFF) {
diff --git a/generic/tclParse.c b/generic/tclParse.c
index adc559d..e95768d 100644
--- a/generic/tclParse.c
+++ b/generic/tclParse.c
@@ -128,6 +128,8 @@ static int		ParseWhiteSpace(const char *src, int numBytes,
 			    int *incompletePtr, char *typePtr);
 static int		ParseAllWhiteSpace(const char *src, int numBytes,
 			    int *incompletePtr);
+static int		ParseHex(const char *src, int numBytes,
+			    int *resultPtr);
 
 /*
  *----------------------------------------------------------------------
@@ -701,7 +703,7 @@ TclParseAllWhiteSpace(
 /*
  *----------------------------------------------------------------------
  *
- * TclParseHex --
+ * ParseHex --
  *
  *	Scans a hexadecimal number as a Tcl_UniChar value (e.g., for parsing
  *	\x and \u escape sequences). At most numBytes bytes are scanned.
@@ -721,7 +723,7 @@ TclParseAllWhiteSpace(
  */
 
 int
-TclParseHex(
+ParseHex(
     const char *src,		/* First character to parse. */
     int numBytes,		/* Max number of byes to scan */
     int *resultPtr)	/* Points to storage provided by caller where
@@ -845,7 +847,7 @@ TclParseBackslash(
 	result = 0xB;
 	break;
     case 'x':
-	count += TclParseHex(p+1, (numBytes > 3) ? 2 : numBytes-2, &result);
+	count += ParseHex(p+1, (numBytes > 3) ? 2 : numBytes-2, &result);
 	if (count == 2) {
 	    /*
 	     * No hexdigits -> This is just "x".
@@ -860,7 +862,7 @@ TclParseBackslash(
 	}
 	break;
     case 'u':
-	count += TclParseHex(p+1, (numBytes > 5) ? 4 : numBytes-2, &result);
+	count += ParseHex(p+1, (numBytes > 5) ? 4 : numBytes-2, &result);
 	if (count == 2) {
 	    /*
 	     * No hexdigits -> This is just "u".
@@ -871,7 +873,7 @@ TclParseBackslash(
 	    /* If high surrogate is immediately followed by a low surrogate
 	     * escape, combine them into one character. */
 	    int low;
-	    int count2 = TclParseHex(p+7, 4, &low);
+	    int count2 = ParseHex(p+7, 4, &low);
 	    if ((count2 == 4) && ((low & 0xDC00) == 0xDC00)) {
 		result = ((result & 0x3FF)<<10 | (low & 0x3FF)) + 0x10000;
 		count += count2 + 2;
@@ -879,7 +881,7 @@ TclParseBackslash(
 	}
 	break;
     case 'U':
-	count += TclParseHex(p+1, (numBytes > 9) ? 8 : numBytes-2, &result);
+	count += ParseHex(p+1, (numBytes > 9) ? 8 : numBytes-2, &result);
 	if (count == 2) {
 	    /*
 	     * No hexdigits -> This is just "U".