Merge core-8-branch. Also, use a different value for TCL_STUB_MAGIC when TCL_UTF_MAX>4.

5 files changed, 28 insertions, 14 deletions

diff --git a/doc/ToUpper.3 b/doc/ToUpper.3
index b06b793..1c7a0c2 100644
--- a/doc/ToUpper.3
+++ b/doc/ToUpper.3
@@ -33,7 +33,7 @@ int
 .SH ARGUMENTS
 .AS char *str in/out
 .AP int ch in
-The character to be converted.
+The Unicode character to be converted.
 .AP char *str in/out
 Pointer to UTF-8 string to be converted in place.
 .BE
diff --git a/doc/UniCharIsAlpha.3 b/doc/UniCharIsAlpha.3
index e1d23ab..16b1517 100644
--- a/doc/UniCharIsAlpha.3
+++ b/doc/UniCharIsAlpha.3
@@ -53,7 +53,7 @@ The character to be examined.
 
 .SH DESCRIPTION
 .PP
-All of the routines described examine characters and return a
+All of the routines described examine Unicode characters and return a
 boolean value. A non-zero return value means that the character does
 belong to the character class associated with the called routine. The
 rest of this document just describes the character classes associated
diff --git a/doc/Utf.3 b/doc/Utf.3
index de9545d..78d795e 100644
--- a/doc/Utf.3
+++ b/doc/Utf.3
@@ -77,7 +77,7 @@ int
 Buffer in which the UTF-8 representation of the Tcl_UniChar is stored.  At most
 \fBTCL_UTF_MAX\fR bytes are stored in the buffer.
 .AP int ch in
-The character to be converted or examined.
+The Unicode character to be converted or examined.
 .AP Tcl_UniChar *chPtr out
 Filled with the Tcl_UniChar represented by the head of the UTF-8 string.
 .AP "const char" *src in
diff --git a/generic/tcl.h b/generic/tcl.h
index 36850d5..77dc51b 100644
--- a/generic/tcl.h
+++ b/generic/tcl.h
@@ -2391,10 +2391,10 @@ typedef int (Tcl_NRPostProc) (ClientData data[], Tcl_Interp *interp,
 /*
  *----------------------------------------------------------------------------
  * The following constant is used to test for older versions of Tcl in the
- * stubs tables.
+ * stubs tables. If TCL_UTF_MAX>4 use a different value.
  */
 
-#define TCL_STUB_MAGIC		((int) 0xFCA3BACF)
+#define TCL_STUB_MAGIC		((int) 0xFCA3BACF + (TCL_UTF_MAX>4))
 
 /*
  * The following function is required to be defined in all stubs aware
diff --git a/generic/tclUtf.c b/generic/tclUtf.c
index aff10c1..23acc8a 100644
--- a/generic/tclUtf.c
+++ b/generic/tclUtf.c
@@ -562,10 +562,10 @@ Tcl_UtfFindFirst(
 #if TCL_UTF_MAX == 4
 	if (!len) {
 	    len += TclUtfToUniChar(src, &find);
-	    fullchar = (((fullchar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
+	    fullchar = (((fullchar & 0x3ff) << 10) | (find & 0x3ff)) + 0x10000;
 	}
 #endif
-	if (find == fullchar) {
+	if (fullchar == ch) {
 	    return src;
 	}
 	if (*src == '\0') {
@@ -610,10 +610,10 @@ Tcl_UtfFindLast(
 #if TCL_UTF_MAX == 4
 	if (!len) {
 	    len += TclUtfToUniChar(src, &find);
-	    fullchar = (((fullchar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
+	    fullchar = (((fullchar & 0x3ff) << 10) | (find & 0x3ff)) + 0x10000;
 	}
 #endif
-	if (find == fullchar) {
+	if (fullchar == ch) {
 	    last = src;
 	}
 	if (*src == '\0') {
@@ -730,12 +730,27 @@ Tcl_UniCharAtIndex(
     register int index)		/* The position of the desired character. */
 {
     Tcl_UniChar ch = 0;
+    int fullchar = 0;
+#if TCL_UTF_MAX == 4
+	int len = 1;
+#endif
 
-    while (index >= 0) {
-	index--;
+    while (index-- >= 0) {
+#if TCL_UTF_MAX == 4
+	src += (len = TclUtfToUniChar(src, &ch));
+#else
 	src += TclUtfToUniChar(src, &ch);
+#endif
+    }
+    fullchar = ch;
+#if TCL_UTF_MAX == 4
+     if (!len) {
+	/* If last Tcl_UniChar was an upper surrogate, combine with lower surrogate */
+	(void)TclUtfToUniChar(src, &ch);
+	fullchar = (((fullchar & 0x3ff) << 10) | (ch & 0x3ff)) + 0x10000;
     }
-    return ch;
+#endif
+    return fullchar;
 }
 
 /*
@@ -762,8 +777,7 @@ Tcl_UtfAtIndex(
 {
     Tcl_UniChar ch = 0;
 
-    while (index > 0) {
-	index--;
+    while (index-- > 0) {
 	src += TclUtfToUniChar(src, &ch);
     }
     return src;