Fix [3cd9bea1e6]: check-in [43032d7ba3] potential problems. Also don't allow surrogates in \U?????? syntax.

3 files changed, 95 insertions, 87 deletions

diff --git a/generic/tclParse.c b/generic/tclParse.c
index 4f30f8b..b42ff5c 100644
--- a/generic/tclParse.c
+++ b/generic/tclParse.c
@@ -870,7 +870,7 @@ TclParseBackslash(
     count = 2;
     switch (*p) {
 	/*
-	 * Note: in the conversions below, use absolute values (e.g., 0xa)
+	 * Note: in the conversions below, use absolute values (e.g., 0xA)
 	 * rather than symbolic values (e.g. \n) that get converted by the
 	 * compiler. It's possible that compilers on some platforms will do
 	 * the symbolic conversions differently, which could result in
@@ -884,19 +884,19 @@ TclParseBackslash(
 	result = 0x8;
 	break;
     case 'f':
-	result = 0xc;
+	result = 0xC;
 	break;
     case 'n':
-	result = 0xa;
+	result = 0xA;
 	break;
     case 'r':
-	result = 0xd;
+	result = 0xD;
 	break;
     case 't':
 	result = 0x9;
 	break;
     case 'v':
-	result = 0xb;
+	result = 0xB;
 	break;
     case 'x':
 	count += TclParseHex(p+1, (numBytes > 3) ? 2 : numBytes-2, &result);
@@ -921,11 +921,13 @@ TclParseBackslash(
 	     */
 	    result = 'u';
 #if TCL_UTF_MAX > 3
-	} else if (((result & 0xDC00) == 0xD800) && (count == 6) && (p[5] == '\\') && (p[6] == 'u') && (numBytes >= 10)) {
-	    /* If high surrogate is immediately followed by a low surrogate escape, combine them. */
+	} else if (((result & 0xDC00) == 0xD800) && (count == 6)
+		    && (p[5] == '\\') && (p[6] == 'u') && (numBytes >= 10)) {
+	    /* 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);
-	    if ((low & 0xDC00) == 0xDC00) {
+	    if ((count2 == 4) && ((low & 0xDC00) == 0xDC00)) {
 		result = ((result & 0x3FF)<<10 | (low & 0x3FF)) + 0x10000;
 		count += count2 + 2;
 	    }
@@ -939,6 +941,9 @@ TclParseBackslash(
 	     * No hexadigits -> This is just "U".
 	     */
 	    result = 'U';
+	} else if ((result | 0x7FF) == 0xDFFF) {
+	    /* Upper or lower surrogate, not allowed in this syntax. */
+	    result = 0xFFFD;
 	}
 	break;
     case '\n':
diff --git a/tests/string.test b/tests/string.test
index a3590e5..f5defd9 100644
--- a/tests/string.test
+++ b/tests/string.test
@@ -222,7 +222,7 @@ test string-4.15 {string first, ability to two-byte encoded utf-8 chars} {
     # Test for a bug in Tcl 8.3 where test for all-single-byte-encoded
     # strings was incorrect, leading to an index returned by [string first]
     # which pointed past the end of the string.
-    set uchar \u057e    ;# character with two-byte encoding in utf-8
+    set uchar \u057E    ;# character with two-byte encoding in utf-8
     string first % %#$uchar$uchar#$uchar$uchar#% 3
 } 8
 test string-4.17 {string first, corner case} {
@@ -309,9 +309,6 @@ test string-5.19 {string index, bytearray object out of bounds} {
 test string-5.20 {string index, bytearray object out of bounds} {
     string index [binary format I* {0x50515253 0x52}] 20
 } {}
-test string-5.21 {string index, surrogates, bug [11ae2be95dac9417]} tip389 {
-    list [string index a\U100000b 1] [string index a\U100000b 2] [string index a\U100000b 3]
-} [list \U100000 {} b]
 
 
 proc largest_int {} {
@@ -395,7 +392,7 @@ test string-6.24 {string is digit, true} {
     string is digit 0123456789
 } 1
 test string-6.25 {string is digit, false} {
-    list [string is digit -fail var 0123\u00dc567] $var
+    list [string is digit -fail var 0123\u00DC567] $var
 } {0 4}
 test string-6.26 {string is digit, false} {
     list [string is digit -fail var +123567] $var
@@ -518,7 +515,7 @@ test string-6.60 {string is lower, true} {
     string is lower abc
 } 1
 test string-6.61 {string is lower, unicode true} {
-    string is lower abc\u00fcue
+    string is lower abc\u00FCue
 } 1
 test string-6.62 {string is lower, false} {
     list [string is lower -fail var aBc] $var
@@ -527,7 +524,7 @@ test string-6.63 {string is lower, false} {
     list [string is lower -fail var abc1] $var
 } {0 3}
 test string-6.64 {string is lower, unicode false} {
-    list [string is lower -fail var ab\u00dcUE] $var
+    list [string is lower -fail var ab\u00DCUE] $var
 } {0 2}
 test string-6.65 {string is space, true} {
     string is space " \t\n\v\f"
@@ -565,7 +562,7 @@ test string-6.75 {string is upper, true} {
     string is upper ABC
 } 1
 test string-6.76 {string is upper, unicode true} {
-    string is upper ABC\u00dcUE
+    string is upper ABC\u00DCUE
 } 1
 test string-6.77 {string is upper, false} {
     list [string is upper -fail var AbC] $var
@@ -574,13 +571,13 @@ test string-6.78 {string is upper, false} {
     list [string is upper -fail var AB2C] $var
 } {0 2}
 test string-6.79 {string is upper, unicode false} {
-    list [string is upper -fail var ABC\u00fcue] $var
+    list [string is upper -fail var ABC\u00FCue] $var
 } {0 3}
 test string-6.80 {string is wordchar, true} {
     string is wordchar abc_123
 } 1
 test string-6.81 {string is wordchar, unicode true} {
-    string is wordchar abc\u00fcab\u00dcAB\u5001
+    string is wordchar abc\u00FCab\u00DCAB\u5001
 } 1
 test string-6.82 {string is wordchar, false} {
     list [string is wordchar -fail var abcd.ef] $var
@@ -606,7 +603,7 @@ test string-6.87 {string is print} {
 } {0 15}
 test string-6.88 {string is punct} {
     ## any graph char that isn't alnum
-    list [string is punct -fail var "_!@#\u00beq0"] $var
+    list [string is punct -fail var "_!@#\u00BEq0"] $var
 } {0 4}
 test string-6.89 {string is xdigit} {
     list [string is xdigit -fail var 0123456789\u0061bcdefABCDEFg] $var
@@ -699,7 +696,7 @@ test string-6.108 {string is double, Bug 1382287} {
     string is double $x
 } 0
 test string-6.109 {string is double, Bug 1360532} {
-    string is double 1\u00a0
+    string is double 1\u00A0
 } 0
 test string-6.110 {string is entier, true} {
     string is entier +1234567890
@@ -1278,8 +1275,8 @@ test string-12.19 {string range, bytearray object} {
     string equal $r1 $r2
 } 1
 test string-12.20 {string range, out of bounds indices} {
-    string range \u00ff 0 1
-} \u00ff
+    string range \u00FF 0 1
+} \u00FF
 # Bug 1410553
 test string-12.21 {string range, regenerates correct reps, bug 1410553} {
     set bytes "\x00 \x03 \x41"
@@ -1490,18 +1487,14 @@ test string-17.5 {string totitle} {
     string totitle {123#$&*()}
 } {123#$&*()}
 test string-17.6 {string totitle, unicode} {
-    string totitle ABCabc\xc7\xe7
-} "Abcabc\xe7\xe7"
+    string totitle ABCabc\xC7\xE7
+} "Abcabc\xE7\xE7"
 test string-17.7 {string totitle, unicode} {
-    string totitle \u01f3BCabc\xc7\xe7
-} "\u01f2bcabc\xe7\xe7"
+    string totitle \u01F3BCabc\xc7\xe7
+} "\u01F2bcabc\xe7\xe7"
 test string-17.8 {string totitle, compiled} {
     lindex [string totitle [list aa bb [list cc]]] 0
 } Aa
-test string-17.9 {string totitle, surrogates, bug [11ae2be95dac9417]} tip389 {
-    list [string totitle a\U118c0c 1 1] [string totitle a\U118c0c 2 2] \
-	[string totitle a\U118c0c 3 3]
-} [list a\U118a0c a\U118c0C a\U118c0C]
 
 test string-18.1 {string trim} {
     list [catch {string trim} msg] $msg
@@ -1537,7 +1530,7 @@ test string-18.11 {string trim, unicode} {
     string trim "\xe7\xe8 AB\xe7C \xe8\xe7" \xe7\xe8
 } " AB\xe7C "
 test string-18.12 {string trim, unicode default} {
-    string trim \ufeff\x00\u0085\u00a0\u1680\u180eABC\u1361\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200a\u200b\u2028\u2029\u202f\u205f\u3000
+    string trim \uFEFF\x00\u0085\u00A0\u1680\u180EABC\u1361\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200A\u200B\u2028\u2029\u202F\u205F\u3000
 } ABC\u1361
 
 test string-19.1 {string trimleft} {
@@ -1547,7 +1540,7 @@ test string-19.2 {string trimleft} {
     string trimleft "    XYZ      "
 } {XYZ      }
 test string-19.3 {string trimleft, unicode default} {
-    string trimleft \ufeff\u0085\u00a0\x00\u1680\u180e\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200a\u200b\u2028\u2029\u202f\u205f\u3000\u1361ABC
+    string trimleft \uFEFF\u0085\u00A0\x00\u1680\u180E\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200A\u200B\u2028\u2029\u202F\u205F\u3000\u1361ABC
 } \u1361ABC
 
 test string-20.1 {string trimright errors} {
@@ -1566,7 +1559,7 @@ test string-20.5 {string trimright} {
     string trimright ""
 } {}
 test string-20.6 {string trimright, unicode default} {
-    string trimright ABC\u1361\u0085\x00\u00a0\u1680\u180e\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200a\u200b\u2028\u2029\u202f\u205f\u3000
+    string trimright ABC\u1361\u0085\x00\u00A0\u1680\u180E\u2000\u2001\u2002\u2003\u2004\u2005\u2006\u2007\u2008\u2009\u200A\u200B\u2028\u2029\u202F\u205F\u3000
 } ABC\u1361
 
 test string-21.1 {string wordend} {
@@ -1597,19 +1590,19 @@ test string-21.9 {string wordend} {
     string worde "x.y" end-1
 } 2
 test string-21.10 {string wordend, unicode} {
-    string wordend "xyz\u00c7de fg" 0
+    string wordend "xyz\u00C7de fg" 0
 } 6
 test string-21.11 {string wordend, unicode} {
-    string wordend "xyz\uc700de fg" 0
+    string wordend "xyz\uC700de fg" 0
 } 6
 test string-21.12 {string wordend, unicode} {
-    string wordend "xyz\u203fde fg" 0
+    string wordend "xyz\u203Fde fg" 0
 } 6
 test string-21.13 {string wordend, unicode} {
     string wordend "xyz\u2045de fg" 0
 } 3
 test string-21.14 {string wordend, unicode} {
-    string wordend "\uc700\uc700 abc" 8
+    string wordend "\uC700\uC700 abc" 8
 } 6
 
 test string-22.1 {string wordstart} {
@@ -1643,13 +1636,13 @@ test string-22.10 {string wordstart} {
     string wordstart "one two three" end-5
 } 7
 test string-22.11 {string wordstart, unicode} {
-    string wordstart "one tw\u00c7o three" 7
+    string wordstart "one tw\u00C7o three" 7
 } 4
 test string-22.12 {string wordstart, unicode} {
-    string wordstart "ab\uc700\uc700 cdef ghi" 12
+    string wordstart "ab\uC700\uC700 cdef ghi" 12
 } 10
 test string-22.13 {string wordstart, unicode} {
-    string wordstart "\uc700\uc700 abc" 8
+    string wordstart "\uC700\uC700 abc" 8
 } 3
 
 test string-23.0 {string is boolean, Bug 1187123} testindexobj {
@@ -1720,40 +1713,40 @@ test string-24.4 {string reverse command - unshared string} {
     string reverse $x$y
 } edcba
 test string-24.5 {string reverse command - shared unicode string} {
-    set x abcde\ud0ad
+    set x abcde\uD0AD
     string reverse $x
-} \ud0adedcba
+} \uD0ADedcba
 test string-24.6 {string reverse command - unshared string} {
     set x abc
-    set y de\ud0ad
+    set y de\uD0AD
     string reverse $x$y
-} \ud0adedcba
+} \uD0ADedcba
 test string-24.7 {string reverse command - simple case} {
     string reverse a
 } a
 test string-24.8 {string reverse command - simple case} {
-    string reverse \ud0ad
-} \ud0ad
+    string reverse \uD0AD
+} \uD0AD
 test string-24.9 {string reverse command - simple case} {
     string reverse {}
 } {}
 test string-24.10 {string reverse command - corner case} {
-    set x \ubeef\ud0ad
+    set x \uBEEF\uD0AD
     string reverse $x
-} \ud0ad\ubeef
+} \uD0AD\uBEEF
 test string-24.11 {string reverse command - corner case} {
-    set x \ubeef
-    set y \ud0ad
+    set x \uBEEF
+    set y \uD0AD
     string reverse $x$y
-} \ud0ad\ubeef
+} \uD0AD\uBEEF
 test string-24.12 {string reverse command - corner case} {
-    set x \ubeef
-    set y \ud0ad
+    set x \uBEEF
+    set y \uD0AD
     string is ascii [string reverse $x$y]
 } 0
 test string-24.13 {string reverse command - pure Unicode string} {
-    string reverse [string range \ubeef\ud0ad\ubeef\ud0ad\ubeef\ud0ad 1 5]
-} \ud0ad\ubeef\ud0ad\ubeef\ud0ad
+    string reverse [string range \uBEEF\uD0AD\uBEEF\uD0AD\uBEEF\uD0AD 1 5]
+} \uD0AD\uBEEF\uD0AD\uBEEF\uD0AD
 test string-24.14 {string reverse command - pure bytearray} {
     binary scan [string reverse [binary format H* 010203]] H* x
     set x
@@ -1809,7 +1802,7 @@ test string-25.13 {string is list} {
 } {0 2}
 test string-25.14 {string is list} {
     set x {}
-    list [string is list -failindex x "\uabcd {b c}d e"] $x
+    list [string is list -failindex x "\uABCD {b c}d e"] $x
 } {0 2}
 
 test string-26.1 {tcl::prefix, too few args} -body {
diff --git a/tests/utf.test b/tests/utf.test
index 1e257d3..2e4882d 100644
--- a/tests/utf.test
+++ b/tests/utf.test
@@ -22,6 +22,7 @@ catch {unset x}
 
 # Some tests require support for 4-byte UTF-8 sequences
 testConstraint fullutf [expr {[format %c 0x010000] != "\uFFFD"}]
+testConstraint tip389 [expr {[string length \U010000] == 2}]
 
 test utf-1.1 {Tcl_UniCharToUtf: 1 byte sequences} testbytestring {
     expr {"\x01" eq [testbytestring "\x01"]}
@@ -59,6 +60,9 @@ test utf-1.11 {Tcl_UniCharToUtf: 3 byte sequence, low surrogate} testbytestring
 test utf-1.12 {Tcl_UniCharToUtf: 4 byte sequence, high/low surrogate} {fullutf testbytestring} {
     expr {"\uD842\uDC42" eq [testbytestring "\xF0\xA0\xA1\x82"]}
 } 1
+test utf-1.13 {Tcl_UniCharToUtf: Invalid surrogate} testbytestring {
+    expr {"\UD842" eq [testbytestring "\xEF\xBF\xBD"]}
+} 1
 
 test utf-2.1 {Tcl_UtfToUniChar: low ascii} {
     string length "abc"
@@ -81,10 +85,10 @@ test utf-2.6 {Tcl_UtfToUniChar: lead (3-byte) followed by 1 trail} testbytestrin
 test utf-2.7 {Tcl_UtfToUniChar: lead (3-byte) followed by 2 trail} testbytestring {
     string length [testbytestring "\xE4\xB9\x8E"]
 } {1}
-test utf-2.8 {Tcl_UtfToUniChar: lead (4-byte) followed by 3 trail} -constraints {fullutf testbytestring} -body {
+test utf-2.8 {Tcl_UtfToUniChar: lead (4-byte) followed by 3 trail} -constraints {tip389 testbytestring} -body {
     string length [testbytestring "\xF0\x90\x80\x80"]
 } -result {2}
-test utf-2.9 {Tcl_UtfToUniChar: lead (4-byte) followed by 3 trail} -constraints {fullutf testbytestring} -body {
+test utf-2.9 {Tcl_UtfToUniChar: lead (4-byte) followed by 3 trail} -constraints {tip389 testbytestring} -body {
     string length [testbytestring "\xF4\x8F\xBF\xBF"]
 } -result {2}
 test utf-2.10 {Tcl_UtfToUniChar: lead (4-byte) followed by 3 trail, underflow} testbytestring {
@@ -160,7 +164,7 @@ test utf-8.3 {Tcl_UniCharAtIndex: index > 0} {
 } {c}
 test utf-8.4 {Tcl_UniCharAtIndex: index > 0} {
     string index \u4E4E\u25A\xFF\u543 2
-} "\uff"
+} "\uFF"
 
 test utf-9.1 {Tcl_UtfAtIndex: index = 0} {
     string range abcd 0 2
@@ -186,6 +190,12 @@ test utf-10.4 {Tcl_UtfBackslash: stops at first non-hex} testbytestring {
 test utf-10.5 {Tcl_UtfBackslash: stops after 4 hex chars} testbytestring {
     expr {"\u4E216" eq "[testbytestring \xE4\xB8\xA1]6"}
 } 1
+test utf-10.6 {Tcl_UtfBackslash: stops after 5 hex chars} {fullutf testbytestring} {
+    expr {"\U1E2165" eq "[testbytestring \xF0\x9E\x88\x96]5"}
+} 1
+test utf-10.7 {Tcl_UtfBackslash: stops after 6 hex chars} {fullutf testbytestring} {
+    expr {"\U10E2165" eq "[testbytestring \xF4\x8E\x88\x96]5"}
+} 1
 proc bsCheck {char num} {
     global errNum
     test utf-10.$errNum {backslash substitution} {
@@ -194,7 +204,7 @@ proc bsCheck {char num} {
     } $num
     incr errNum
 }
-set errNum 6
+set errNum 8
 bsCheck \b	8
 bsCheck \e	101
 bsCheck \f	12
@@ -260,10 +270,10 @@ test utf-11.2 {Tcl_UtfToUpper} {
     string toupper abc
 } ABC
 test utf-11.3 {Tcl_UtfToUpper} {
-    string toupper \u00E3ab
+    string toupper \u00E3AB
 } \u00C3AB
 test utf-11.4 {Tcl_UtfToUpper} {
-    string toupper \u01E3ab
+    string toupper \u01E3AB
 } \u01E2AB
 test utf-11.5 {Tcl_UtfToUpper Georgian (new in Unicode 11)} {
     string toupper \u10D0\u1C90
@@ -292,10 +302,10 @@ test utf-13.2 {Tcl_UtfToTitle} {
     string totitle abc
 } Abc
 test utf-13.3 {Tcl_UtfToTitle} {
-    string totitle \u00E3ab
+    string totitle \u00E3AB
 } \u00C3ab
 test utf-13.4 {Tcl_UtfToTitle} {
-    string totitle \u01F3ab
+    string totitle \u01F3AB
 } \u01F2ab
 test utf-13.5 {Tcl_UtfToTitle Georgian (new in Unicode 11)} {
     string totitle \u10D0\u1C90
@@ -321,7 +331,7 @@ test utf-15.1 {Tcl_UniCharToUpper, negative delta} {
     string toupper aA
 } AA
 test utf-15.2 {Tcl_UniCharToUpper, positive delta} {
-    string toupper \u0178\u00ff
+    string toupper \u0178\u00FF
 } \u0178\u0178
 test utf-15.3 {Tcl_UniCharToUpper, no delta} {
     string toupper !
@@ -331,24 +341,24 @@ test utf-16.1 {Tcl_UniCharToLower, negative delta} {
     string tolower aA
 } aa
 test utf-16.2 {Tcl_UniCharToLower, positive delta} {
-    string tolower \u0178\u00ff\uA78D\u01c5
-} \u00ff\u00ff\u0265\u01c6
+    string tolower \u0178\u00FF\uA78D\u01C5
+} \u00FF\u00FF\u0265\u01C6
 
 test utf-17.1 {Tcl_UniCharToLower, no delta} {
     string tolower !
 } !
 
 test utf-18.1 {Tcl_UniCharToTitle, add one for title} {
-    string totitle \u01c4
-} \u01c5
+    string totitle \u01C4
+} \u01C5
 test utf-18.2 {Tcl_UniCharToTitle, subtract one for title} {
-    string totitle \u01c6
-} \u01c5
+    string totitle \u01C6
+} \u01C5
 test utf-18.3 {Tcl_UniCharToTitle, subtract delta for title (positive)} {
-    string totitle \u017f
+    string totitle \u017F
 } \u0053
 test utf-18.4 {Tcl_UniCharToTitle, subtract delta for title (negative)} {
-    string totitle \u00ff
+    string totitle \u00FF
 } \u0178
 test utf-18.5 {Tcl_UniCharToTitle, no delta} {
     string totitle !
@@ -365,15 +375,15 @@ test utf-20.1 {TclUniCharNcmp} {
 
 test utf-21.1 {TclUniCharIsAlnum} {
     # this returns 1 with Unicode 7 compliance
-    string is alnum \u1040\u021f\u0220
+    string is alnum \u1040\u021F\u0220
 } {1}
 test utf-21.2 {unicode alnum char in regc_locale.c} {
     # this returns 1 with Unicode 7 compliance
-    list [regexp {^[[:alnum:]]+$} \u1040\u021f\u0220] [regexp {^\w+$} \u1040\u021f\u0220_\u203f\u2040\u2054\ufe33\ufe34\ufe4d\ufe4e\ufe4f\uff3f]
+    list [regexp {^[[:alnum:]]+$} \u1040\u021F\u0220] [regexp {^\w+$} \u1040\u021F\u0220_\u203F\u2040\u2054\uFE33\uFE34\uFE4D\uFE4E\uFE4F\uFF3F]
 } {1 1}
 test utf-21.3 {unicode print char in regc_locale.c} {
     # this returns 1 with Unicode 7 compliance
-    regexp {^[[:print:]]+$} \ufbc1
+    regexp {^[[:print:]]+$} \uFBC1
 } 1
 test utf-21.4 {TclUniCharIsGraph} {
     # [Bug 3464428]
@@ -385,11 +395,11 @@ test utf-21.5 {unicode graph char in regc_locale.c} {
 } {1}
 test utf-21.6 {TclUniCharIsGraph} {
     # [Bug 3464428]
-    string is graph \u00a0
+    string is graph \u00A0
 } {0}
 test utf-21.7 {unicode graph char in regc_locale.c} {
     # [Bug 3464428]
-    regexp {[[:graph:]]} \u0020\u00a0\u2028\u2029
+    regexp {[[:graph:]]} \u0020\u00A0\u2028\u2029
 } {0}
 test utf-21.8 {TclUniCharIsPrint} {
     # [Bug 3464428]
@@ -405,45 +415,45 @@ test utf-21.10 {unicode print char in regc_locale.c} {
 } {0}
 test utf-21.11 {TclUniCharIsControl} {
     # [Bug 3464428]
-    string is control \u0000\u001f\u00ad\u0605\u061c\u180e\u2066\ufeff
+    string is control \u0000\u001F\u00AD\u0605\u061C\u180E\u2066\uFEFF
 } {1}
 test utf-21.12 {unicode control char in regc_locale.c} {
     # [Bug 3464428], [Bug a876646efe]
-    regexp {^[[:cntrl:]]*$} \u0000\u001f\u00ad\u0605\u061c\u180e\u2066\ufeff
+    regexp {^[[:cntrl:]]*$} \u0000\u001F\u00AD\u0605\u061C\u180E\u2066\uFEFF
 } {1}
 
 test utf-22.1 {TclUniCharIsWordChar} {
     string wordend "xyz123_bar fg" 0
 } 10
 test utf-22.2 {TclUniCharIsWordChar} {
-    string wordend "x\u5080z123_bar\u203c fg" 0
+    string wordend "x\u5080z123_bar\u203C fg" 0
 } 10
 
 test utf-23.1 {TclUniCharIsAlpha} {
     # this returns 1 with Unicode 7 compliance
-    string is alpha \u021f\u0220\u037f\u052f
+    string is alpha \u021F\u0220\u037F\u052F
 } {1}
 test utf-23.2 {unicode alpha char in regc_locale.c} {
     # this returns 1 with Unicode 7 compliance
-    regexp {^[[:alpha:]]+$} \u021f\u0220\u037f\u052f
+    regexp {^[[:alpha:]]+$} \u021F\u0220\u037F\u052F
 } {1}
 
 test utf-24.1 {TclUniCharIsDigit} {
     # this returns 1 with Unicode 7 compliance
-    string is digit \u1040\uabf0
+    string is digit \u1040\uABF0
 } {1}
 test utf-24.2 {unicode digit char in regc_locale.c} {
     # this returns 1 with Unicode 7 compliance
-    list [regexp {^[[:digit:]]+$} \u1040\uabf0] [regexp {^\d+$} \u1040\uabf0]
+    list [regexp {^[[:digit:]]+$} \u1040\uABF0] [regexp {^\d+$} \u1040\uABF0]
 } {1 1}
 
 test utf-24.3 {TclUniCharIsSpace} {
     # this returns 1 with Unicode 7/TIP 413 compliance
-    string is space \u0085\u1680\u180e\u200b\u202f\u2060
+    string is space \u0085\u1680\u180E\u200B\u202F\u2060
 } {1}
 test utf-24.4 {unicode space char in regc_locale.c} {
     # this returns 1 with Unicode 7/TIP 413 compliance
-    list [regexp {^[[:space:]]+$} \u0085\u1680\u180e\u200b\u202f\u2060] [regexp {^\s+$} \u0085\u1680\u180e\u200b\u202f\u2060]
+    list [regexp {^[[:space:]]+$} \u0085\u1680\u180E\u200B\u202F\u2060] [regexp {^\s+$} \u0085\u1680\u180E\u200B\u202F\u2060]
 } {1 1}
 
 testConstraint teststringobj [llength [info commands teststringobj]]