3 # This is a home for regular expression tests that don't fit into
4 # the format supported by op/regexp.t. If you want to add a test
5 # that does fit that format, add it to op/re_tests, not here.
9 # please update note at bottom of file when you change this
17 eval 'use Config'; # Defaults assumed if this fails
21 if ($x =~ /^abc/) {print "ok 1\n";} else {print "not ok 1\n";}
22 if ($x !~ /^def/) {print "ok 2\n";} else {print "not ok 2\n";}
24 # used to be a test for $*
25 if ($x =~ /^def/m) {print "ok 3\n";} else {print "not ok 3\n";}
28 if (/^([0-9][0-9]*)/) {print "ok 4\n";} else {print "not ok 4\n";}
30 if ($x =~ /^xxx/) {print "not ok 5\n";} else {print "ok 5\n";}
31 if ($x !~ /^abc/) {print "not ok 6\n";} else {print "ok 6\n";}
33 if ($x =~ /def/) {print "ok 7\n";} else {print "not ok 7\n";}
34 if ($x !~ /def/) {print "not ok 8\n";} else {print "ok 8\n";}
36 if ($x !~ /.def/) {print "ok 9\n";} else {print "not ok 9\n";}
37 if ($x =~ /.def/) {print "not ok 10\n";} else {print "ok 10\n";}
39 if ($x =~ /\ndef/) {print "ok 11\n";} else {print "not ok 11\n";}
40 if ($x !~ /\ndef/) {print "not ok 12\n";} else {print "ok 12\n";}
43 if (/(a*b*)(c*)/ && $1 eq 'aaabbb' && $2 eq 'ccc') {
48 if (/(a+b+c+)/ && $1 eq 'aaabbbccc') {
54 if (/a+b?c+/) {print "not ok 15\n";} else {print "ok 15\n";}
57 if (/a+b?c+/) {print "ok 16\n";} else {print "not ok 16\n";}
58 if (/a*b+c*/) {print "ok 17\n";} else {print "not ok 17\n";}
61 if (/a*b?c*/) {print "ok 18\n";} else {print "not ok 18\n";}
62 if (/a*b+c*/) {print "not ok 19\n";} else {print "ok 19\n";}
65 if (/bcd|xyz/) {print "ok 20\n";} else {print "not ok 20\n";}
66 if (/xyz|bcd/) {print "ok 21\n";} else {print "not ok 21\n";}
68 if (m|bc/*d|) {print "ok 22\n";} else {print "not ok 22\n";}
70 if (/^$_$/) {print "ok 23\n";} else {print "not ok 23\n";}
72 # used to be a test for $*
73 if ("ab\ncd\n" =~ /^cd/m) {print "ok 24\n";} else {print "not ok 24\n";}
79 @XXX = ('ok 25','not ok 25', 'ok 26','not ok 26','not ok 27');
80 while ($_ = shift(@XXX)) {
81 ?(.*)? && (print $1,"\n");
94 while (($key,$val) = each(%XXX)) {
104 if ($& eq 'xyz') {print "ok 28\n";} else {print "not ok 28\n";}
109 if ($& eq 'xyz') {print "ok 29\n";} else {print "not ok 29\n";}
114 if ($& eq 'xyz') {print "ok 30\n";} else {print "not ok 30\n";}
117 /def/; # optimized up to cmd
118 if ("$`:$&:$'" eq 'abc:def:ghi') {print "ok 31\n";} else {print "not ok 31\n";}
120 /cde/ + 0; # optimized only to spat
121 if ("$`:$&:$'" eq 'ab:cde:fghi') {print "ok 32\n";} else {print "not ok 32\n";}
123 /[d][e][f]/; # not optimized
124 if ("$`:$&:$'" eq 'abc:def:ghi') {print "ok 33\n";} else {print "not ok 33\n";}
126 $_ = 'now is the {time for all} good men to come to.';
128 if ($1 eq 'time for all') {print "ok 34\n";} else {print "not ok 34 $1\n";}
130 $_ = 'xxx {3,4} yyy zzz';
131 print /( {3,4})/ ? "ok 35\n" : "not ok 35\n";
132 print $1 eq ' ' ? "ok 36\n" : "not ok 36\n";
133 print /( {4,})/ ? "not ok 37\n" : "ok 37\n";
134 print /( {2,3}.)/ ? "ok 38\n" : "not ok 38\n";
135 print $1 eq ' y' ? "ok 39\n" : "not ok 39\n";
136 print /(y{2,3}.)/ ? "ok 40\n" : "not ok 40\n";
137 print $1 eq 'yyy ' ? "ok 41\n" : "not ok 41\n";
138 print /x {3,4}/ ? "not ok 42\n" : "ok 42\n";
139 print /^xxx {3,4}/ ? "not ok 43\n" : "ok 43\n";
141 $_ = "now is the time for all good men to come to.";
143 print join(':',@words) eq "now:is:the:time:for:all:good:men:to:come:to"
151 print join(':',@words) eq "now:is:the:time:for:all:good:men:to:come:to"
160 print join(':',@words) eq "to:to"
162 : "not ok 46 `@words'\n";
166 print join(':',@words) eq "to:to"
168 : "not ok 47 `@words'\n";
182 $t1=$t2=$t3=$t4=$t5=$t6=$t7=$t8=$t9=0;
196 $x = "$t1$t2$t3$t4$t5$t6$t7$t8$t9";
197 print $x eq '505550555' ? "ok 48\n" : "not ok 48 $x\n";
200 print "abc" =~ /^abc$|$xyz/ ? "ok 49\n" : "not ok 49\n";
202 # perl 4.009 says "unmatched ()"
203 eval '"abc" =~ /a(bc$)|$xyz/; $result = "$&:$1"';
204 print $@ eq "" ? "ok 50\n" : "not ok 50\n";
205 print $result eq "abc:bc" ? "ok 51\n" : "not ok 51\n";
210 print $` eq "" ? "ok 52\n" : "not ok 52\n" if $x;
212 print $` eq "abcfoo" ? "ok 53\n" : "not ok 53\n" if $x;
214 print $x == 0 ? "ok 54\n" : "not ok 54\n";
217 print $` eq "" ? "ok 55\n" : "not ok 55\n" if $x;
219 print $` eq "abcfoo" ? "ok 56\n" : "not ok 56\n" if $x;
221 print $x == 0 ? "ok 57\n" : "not ok 57\n";
224 print $' eq "fooabcbar" ? "ok 58\n" : "not ok 58\n" if $x;
226 print $' eq "bar" ? "ok 59\n" : "not ok 59\n" if $x;
229 print scalar @x == 2 ? "ok 60\n" : "not ok 60\n";
234 print "not " if (pos $_) != 2;
237 print "not " if defined pos $_;
241 'abc' =~ m'a(?{ $out = 2 })b';
242 print "not " if $out != 2;
246 'abc' =~ m'a(?{ $out = 3 })c';
247 print "not " if $out != 1;
250 $_ = 'foobar1 bar2 foobar3 barfoobar5 foobar6';
251 @out = /(?<!foo)bar./g;
252 print "not " if "@out" ne 'bar2 barf';
255 # Tests which depend on REG_INFTY
256 $reg_infty = defined $Config{reg_infty} ? $Config{reg_infty} : 32767;
257 $reg_infty_m = $reg_infty - 1; $reg_infty_p = $reg_infty + 1;
259 # As well as failing if the pattern matches do unexpected things, the
260 # next three tests will fail if you should have picked up a lower-than-
261 # default value for $reg_infty from Config.pm, but have not.
264 print "not " if eval q(('aaa' =~ /(a{1,$reg_infty_m})/)[0] ne 'aaa') || $@;
268 print "not " if eval q(('a' x $reg_infty_m) !~ /a{$reg_infty_m}/) || $@;
272 print "not " if eval q(('a' x ($reg_infty_m - 1)) =~ /a{$reg_infty_m}/) || $@;
276 eval "'aaa' =~ /a{1,$reg_infty}/";
277 print "not " if $@ !~ m%^\QQuantifier in {,} bigger than%;
280 eval "'aaa' =~ /a{1,$reg_infty_p}/";
282 if $@ !~ m%^\QQuantifier in {,} bigger than%;
286 # Poke a couple more parse failures
288 $context = 'x' x 256;
289 eval qq("${context}y" =~ /(?<=$context)y/);
290 print "not " if $@ !~ m%^\QLookbehind longer than 255 not%;
298 for $l (125, 140, 250, 270, 300000, 30) { # Ordered to free memory
300 print "# length=$l\nnot " unless "ba$a=" =~ /a$a=/;
304 print "not " if "b$a=" =~ /a$a=/;
309 # 20000 nodes, each taking 3 words per string, and 1 per branch
310 $long_constant_len = join '|', 12120 .. 32645;
311 $long_var_len = join '|', 8120 .. 28645;
312 %ans = ( 'ax13876y25677lbc' => 1,
313 'ax13876y25677mcb' => 0, # not b.
314 'ax13876y35677nbc' => 0, # Num too big
315 'ax13876y25677y21378obc' => 1,
316 'ax13876y25677y21378zbc' => 0, # Not followed by [k-o]
317 'ax13876y25677y21378y21378kbc' => 1,
318 'ax13876y25677y21378y21378kcb' => 0, # Not b.
319 'ax13876y25677y21378y21378y21378kbc' => 0, # 5 runs
323 print "# const-len `$_' not => $ans{$_}\nnot "
324 if $ans{$_} xor /a(?=([yx]($long_constant_len)){2,4}[k-o]).*b./o;
327 print "# var-len `$_' not => $ans{$_}\nnot "
328 if $ans{$_} xor /a(?=([yx]($long_var_len)){2,4}[k-o]).*b./o;
333 $_ = " a (bla()) and x(y b((l)u((e))) and b(l(e)e)e";
334 $expect = "(bla()) ((l)u((e))) (l(e)e)";
340 (?{ $c = 1 }) # Initialize
342 (?(?{ $c == 0 }) # PREVIOUS iteration was OK, stop the loop
344 ) # Fail: will unwind one iteration back
347 [^()]+ # Match a big chunk
350 ) # Do not try to match subchunks
358 )+ # This may not match with different subblocks
363 ) # Otherwise the chunk 1 may succeed with $c>0
368 push @ans, $res while $res = matchit;
370 print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne "1 1 1";
376 print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne $expect;
380 print "not " unless "abc" =~ /^(??{"a"})b/;
385 $matched = qr/\((?:(?>[^()]+)|(??{$matched}))*\)/;
388 push(@ans, $res), push(@ans1, $&) while $res = m/$matched/g;
390 print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne "1 1 1";
394 print "# ans1='@ans1'\n# expect='$expect'\nnot " if "@ans1" ne $expect;
400 print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne $expect;
404 @ans = ('a/b' =~ m%(.*/)?(.*)%); # Stack may be bad
405 print "not " if "@ans" ne 'a/ b';
409 $code = '{$blah = 45}';
412 print "not " unless $@ and $@ =~ /not allowed at runtime/ and $blah == 12;
416 for $code ('{$blah = 45}','=xx') {
418 $res = eval { "xx" =~ /(?$code)/o };
419 if ($code eq '=xx') {
420 print "#'$@','$res','$blah'\nnot " unless not $@ and $res;
422 print "#'$@','$res','$blah'\nnot " unless $@ and $@ =~ /not allowed at runtime/ and $blah == 12;
428 $code = '{$blah = 45}';
431 print "not " if $blah != 45;
437 print "not " if $blah != 45;
443 print "not " unless pos($x) == 2;
448 print "not " unless pos($x) == 2;
458 print "not " unless f(pos($x)) == 4;
463 'foot' =~ /foo(?{$x = 12; 75})[t]/;
464 print "not " unless $^R eq '75';
469 'foot' =~ /foo(?{$x = 12; 75})[xy]/;
470 print "not " unless $^R eq '67' and $x eq '12';
475 'foot' =~ /foo(?{ $^R + 12 })((?{ $x = 12; $^R + 17 })[xy])?/;
476 print "not " unless $^R eq '79' and $x eq '12';
480 print "not " unless qr/\b\v$/i eq '(?i-xsm:\bv$)';
484 print "not " unless qr/\b\v$/s eq '(?s-xim:\bv$)';
488 print "not " unless qr/\b\v$/m eq '(?m-xis:\bv$)';
492 print "not " unless qr/\b\v$/x eq '(?x-ism:\bv$)';
496 print "not " unless qr/\b\v$/xism eq '(?msix:\bv$)';
500 print "not " unless qr/\b\v$/ eq '(?-xism:\bv$)';
505 foreach $ans ('', 'c') {
506 /(?<=(?=a)..)((?=c)|.)/g;
507 print "# \$1 ='$1'\n# \$ans='$ans'\nnot " unless $1 eq $ans;
513 foreach $ans ('', 'a', '') {
515 print "# \$& ='$&'\n# \$ans='$ans'\nnot " unless $& eq $ans;
521 my($v,$a,$b,$res) = @_;
523 print "not " unless $res eq $v;
527 prefixify('/a/b/lib/arch', "/a/b/lib", 'X/lib', 'X/lib/arch');
528 prefixify('/a/b/man/arch', "/a/b/man", 'X/man', 'X/man/arch');
532 print "not " unless $1 and /$1/;
539 print "not " unless $b eq '9';
545 print "not " unless $b eq '11';
552 print "not " unless $b eq '14';
560 my $lex_res = ($lex_b =~ qr/$lex_b(?{ $lex_c = $lex_a++ })/);
561 print "not " unless $lex_res eq '1';
564 print "not " unless $lex_a eq '44';
567 print "not " unless $lex_c eq '43';
573 $match = eval { /$a$c$a/ };
575 unless $b eq '14' and $@ =~ /Eval-group not allowed/ and not $match;
585 my $lex_res = ($lex_b =~ qr/17(?{ $lex_c = $lex_a++ })/);
586 print "not " unless $lex_res eq '1';
589 print "not " unless $lex_a eq '44';
592 print "not " unless $lex_c eq '43';
601 '' =~ /(?{ $c = 4 })/;
602 print "not " unless $c == 4;
606 print "not " unless $c == 3;
611 my $warn_pat = shift;
612 return sub { print "not " unless $_[0] =~ /$warn_pat/ }
616 my ($warn_pat, $code) = @_;
618 eval 'BEGIN { use warnings; $SIG{__WARN__} = $warn_pat };' . $code;
625 my $warn_pat = shift;
626 return sub { must_warn(must_warn_pat($warn_pat)) }
629 my $for_future = make_must_warn('reserved for future extensions');
631 &$for_future('q(a:[b]:) =~ /[x[:foo:]]/');
633 #&$for_future('q(a=[b]=) =~ /[x[=foo=]]/');
634 print "ok $test\n"; $test++; # now a fatal croak
636 #&$for_future('q(a.[b].) =~ /[x[.foo.]]/');
637 print "ok $test\n"; $test++; # now a fatal croak
639 # test if failure of patterns returns empty list
662 print "not " if $#+ != 0 or $#- != 0;
666 print "not " if $+[0] != 2 or $-[0] != 1;
671 if defined $+[1] or defined $-[1] or defined $+[2] or defined $-[2];
676 print "not " if $#+ != 2 or $#- != 2;
680 print "not " if $+[0] != 3 or $-[0] != 0;
684 print "not " if $+[1] != 2 or $-[1] != 1;
688 print "not " if $+[2] != 3 or $-[2] != 2;
693 if defined $+[3] or defined $-[3] or defined $+[4] or defined $-[4];
698 print "not " if $#+ != 3 or $#- != 3;
702 print "not " if $+[0] != 3 or $-[0] != 0;
706 print "not " if $+[1] != 2 or $-[1] != 1;
710 print "not " if $+[3] != 3 or $-[3] != 2;
715 if defined $+[2] or defined $-[2] or defined $+[4] or defined $-[4];
720 print "not " if $#+ != 1 or $#- != 1;
724 print "not " if $+[0] != 2 or $-[0] != 0;
728 print "not " if $+[1] != 2 or $-[1] != 1;
733 if defined $+[2] or defined $-[2] or defined $+[3] or defined $-[3];
737 eval { $+[0] = 13; };
739 if $@ !~ /^Modification of a read-only value attempted/;
743 eval { $-[0] = 13; };
745 if $@ !~ /^Modification of a read-only value attempted/;
749 eval { @+ = (7, 6, 5); };
751 if $@ !~ /^Modification of a read-only value attempted/;
755 eval { @- = qw(foo bar); };
757 if $@ !~ /^Modification of a read-only value attempted/;
762 print "#$#-..$#+\nnot " if $#+ != 2 or $#- != 1;
769 print "not " unless "@a" eq "a a";
776 print "not " if $str =~ /^\G/;
780 print "not " if $str =~ /^.\G/;
784 print "not " unless $str =~ /^..\G/;
788 print "not " if $str =~ /^...\G/;
792 print "not " unless $str =~ /.\G./ and $& eq 'bc';
796 print "not " unless $str =~ /\G../ and $& eq 'cd';
800 undef $foo; undef $bar;
801 print "#'$str','$foo','$bar'\nnot "
802 unless $str =~ /b(?{$foo = $_; $bar = pos})c/
803 and $foo eq 'abcde' and $bar eq 2;
807 undef $foo; undef $bar;
809 print "#'$str','$foo','$bar'\nnot "
810 unless $str =~ /b(?{$foo = $_; $bar = pos})c/g
811 and $foo eq 'abcde' and $bar eq 2 and pos $str eq 3;
817 undef $foo; undef $bar;
818 print "#'$str','$foo','$bar'\nnot "
819 unless /b(?{$foo = $_; $bar = pos})c/
820 and $foo eq 'abcde' and $bar eq 2;
824 undef $foo; undef $bar;
825 print "#'$str','$foo','$bar'\nnot "
826 unless /b(?{$foo = $_; $bar = pos})c/g
827 and $foo eq 'abcde' and $bar eq 2 and pos eq 3;
831 undef $foo; undef $bar;
833 1 while /b(?{$foo = $_; $bar = pos})c/g;
834 print "#'$str','$foo','$bar'\nnot "
835 unless $foo eq 'abcde' and $bar eq 2 and not defined pos;
839 undef $foo; undef $bar;
841 print "#'$str','$foo','$bar','$_'\nnot "
842 unless s/b(?{$foo = $_; $bar = pos})c/x/g and $foo eq 'abcde|abcde'
843 and $bar eq 8 and $_ eq 'axde|axde';
850 @dummy = /([ace]).(?{push @res, $1,$2})([ce])(?{push @res, $1,$2})/g;
851 @res = map {defined $_ ? "'$_'" : 'undef'} @res;
853 print "#'@res' '$_'\nnot "
854 unless "@res" eq "'a' undef 'a' 'c' 'e' undef 'a' undef 'a' 'c'";
859 @dummy = /([ace]).(?{push @res, $`,$&,$'})([ce])(?{push @res, $`,$&,$'})/g;
860 @res = map {defined $_ ? "'$_'" : 'undef'} @res;
862 print "#'@res' '$_'\nnot "
864 "'' 'ab' 'cde|abcde' " .
865 "'' 'abc' 'de|abcde' " .
866 "'abcd' 'e|' 'abcde' " .
867 "'abcde|' 'ab' 'cde' " .
868 "'abcde|' 'abc' 'de'" ;
872 #Some more \G anchor checks
873 $foo='aabbccddeeffgg';
878 print "not " unless($1 eq 'ab');
884 print "not " unless($1 eq 'cc');
890 print "not " unless($1 eq 'de');
894 print "not " unless $foo =~ /\Gef/g;
901 print "not " unless($1 eq 'aa');
906 print "not " unless($1 eq 'bb');
912 print "not " unless($1 eq 'cd');
918 print "not " unless('123||x|123|' eq join '|', @res);
922 # see if matching against temporaries (created via pp_helem()) is safe
923 { foo => "ok $test\n".$^X }->{foo} =~ /^(.*)\n/g;
927 # See if $i work inside (?{}) in the presense of saved substrings and
931 s/(\w)(?{push @b, $1})/,$1,/g for @a;
933 print "# \@b='@b', expect 'f o o b a r'\nnot " unless("@b" eq "f o o b a r");
937 print "not " unless("@a" eq ",f,,o,,o, ,b,,a,,r,");
942 { (?> [^{}]+ | (??{ $brackets }) )* }
946 print "ok $test\n"; # Did we survive?
949 "something { long { and } hairy" =~ $brackets;
950 print "ok $test\n"; # Did we survive?
953 "something { long { and } hairy" =~ m/((??{ $brackets }))/;
954 print "not " unless $1 eq "{ and }";
960 m/^-.*bb/mg and print "not ";
966 $text =~ /\GXb*X/g and print 'not ';
970 $text = "xA\n" x 500;
971 $text =~ /^\s*A/m and print 'not ';
976 @res = ($text =~ /.*?(b).*?\b/g);
977 "@res" eq 'b b' or print 'not ';
984 @c = grep(/[^\s]/,@a);
985 print "not " if "@b" ne "@c";
990 @c = grep(/[\S]/,@a);
991 print "not " if "@b" ne "@c";
996 @c = grep(/[^\S]/,@a);
997 print "not " if "@b" ne "@c";
1002 @c = grep(/[\s]/,@a);
1003 print "not " if "@b" ne "@c";
1008 @c = grep(/[^\d]/,@a);
1009 print "not " if "@b" ne "@c";
1014 @c = grep(/[\D]/,@a);
1015 print "not " if "@b" ne "@c";
1020 @c = grep(/[^\D]/,@a);
1021 print "not " if "@b" ne "@c";
1026 @c = grep(/[\d]/,@a);
1027 print "not " if "@b" ne "@c";
1032 @c = grep(/[^\w]/,@a);
1033 print "not " if "@b" ne "@c";
1038 @c = grep(/[\W]/,@a);
1039 print "not " if "@b" ne "@c";
1044 @c = grep(/[^\W]/,@a);
1045 print "not " if "@b" ne "@c";
1050 @c = grep(/[\w]/,@a);
1051 print "not " if "@b" ne "@c";
1055 # see if backtracking optimization works correctly
1056 "\n\n" =~ /\n $ \n/x or print "not ";
1060 "\n\n" =~ /\n* $ \n/x or print "not ";
1064 "\n\n" =~ /\n+ $ \n/x or print "not ";
1068 [] =~ /^ARRAY/ or print "# [] \nnot ";
1075 use overload '""' => sub { 'Object S' };
1076 sub new { bless [] }
1081 $a and $a =~ /^Object\sS/ or print "# '$a' \nnot ";
1085 # test result of match used as match (!)
1086 'a1b' =~ ('xyz' =~ /y/) and $` eq 'a' or print "not ";
1090 'a1b' =~ ('xyz' =~ /t/) and $` eq 'a' or print "not ";
1096 local $SIG{__WARN__} = sub { $w = 1 };
1098 $w = 1 if ("1\n" x 102) =~ /^\s*\n/m;
1100 print $w ? "not " : "", "ok $test\n";
1103 my %space = ( spc => " ",
1108 # There's no \v but the vertical tabulator seems miraculously
1109 # be 11 both in ASCII and EBCDIC.
1113 my @space0 = sort grep { $space{$_} =~ /\s/ } keys %space;
1114 my @space1 = sort grep { $space{$_} =~ /[[:space:]]/ } keys %space;
1115 my @space2 = sort grep { $space{$_} =~ /[[:blank:]]/ } keys %space;
1117 print "not " unless "@space0" eq "cr ff lf spc tab";
1118 print "ok $test # @space0\n";
1121 print "not " unless "@space1" eq "cr ff lf spc tab vt";
1122 print "ok $test # @space1\n";
1125 print "not " unless "@space2" eq "spc tab";
1126 print "ok $test # @space2\n";
1129 # bugid 20001021.005 - this caused a SEGV
1130 print "not " unless undef =~ /^([^\/]*)(.*)$/;
1134 # bugid 20000731.001
1136 print "not " unless "A \x{263a} B z C" =~ /A . B (??{ "z" }) C/;
1140 my $ordA = ord('A');
1143 if (/(.)(\C)(\C)(.)/) {
1148 print "not ok 233\n";
1150 if ($ordA == 65) { # ASCII (or equivalent), should be UTF-8
1154 print "not ok 234\n";
1159 print "not ok 235\n";
1161 } elsif ($ordA == 193) { # EBCDIC (or equivalent), should be UTF-EBCDIC
1165 print "not ok 234\n";
1170 print "not ok 235\n";
1174 print "not ok $_ # ord('A') == $ordA\n";
1180 print "not ok 236\n";
1184 print "not ok $_\n";
1190 # currently \C are still tagged as UTF-8
1195 print "not ok 238\n";
1197 } elsif ($ordA == 193) {
1201 print "not ok 238\n";
1204 print "not ok 238 # ord('A') == $ordA\n";
1208 print "not ok $_\n";
1213 # currently \C are still tagged as UTF-8
1218 print "not ok 240\n";
1220 } elsif ($ordA == 193) {
1224 print "not ok 240\n";
1227 print "not ok 240 # ord('A') == $ordA\n";
1231 print "not ok $_\n";
1236 # japhy -- added 03/03/2001
1237 () = (my $str = "abc") =~ /(...)/;
1239 print "not " if $1 ne "abc";
1243 # The 242 and 243 go with the 244 and 245.
1244 # The trick is that in EBCDIC the explicit numeric range should match
1245 # (as also in non-EBCDIC) but the explicit alphabetic range should not match.
1247 if ("\x8e" =~ /[\x89-\x91]/) {
1250 print "not ok 242\n";
1253 if ("\xce" =~ /[\xc9-\xd1]/) {
1256 print "not ok 243\n";
1259 # In most places these tests would succeed since \x8e does not
1260 # in most character sets match 'i' or 'j' nor would \xce match
1261 # 'I' or 'J', but strictly speaking these tests are here for
1262 # the good of EBCDIC, so let's test these only there.
1263 if (ord('i') == 0x89 && ord('J') == 0xd1) { # EBCDIC
1264 if ("\x8e" !~ /[i-j]/) {
1267 print "not ok 244\n";
1269 if ("\xce" !~ /[I-J]/) {
1272 print "not ok 245\n";
1276 print "ok $_ # Skip: only in EBCDIC\n";
1280 print "not " unless "\x{ab}" =~ /\x{ab}/;
1283 print "not " unless "\x{abcd}" =~ /\x{abcd}/;
1287 # bug id 20001008.001
1290 my @x = ("stra\337e 138","stra\337e 138");
1292 s/(\d+)\s*([\w\-]+)/$1 . uc $2/e;
1293 my($latin) = /^(.+)(?:\s+\d)/;
1294 print $latin eq "stra\337e" ? "ok $test\n" : # 248,249
1295 "#latin[$latin]\nnot ok $test\n";
1297 $latin =~ s/stra\337e/straße/; # \303\237 after the 2nd a
1298 use utf8; # needed for the raw UTF-8
1299 $latin =~ s!(s)tr(?:aß|s+e)!$1tr.!; # \303\237 after the a
1304 print "not " unless "ba\xd4c" =~ /([a\xd4]+)/ && $1 eq "a\xd4";
1307 print "not " unless "ba\xd4c" =~ /([a\xd4]+)/ && $1 eq "a\x{d4}";
1310 print "not " unless "ba\x{d4}c" =~ /([a\xd4]+)/ && $1 eq "a\x{d4}";
1313 print "not " unless "ba\x{d4}c" =~ /([a\xd4]+)/ && $1 eq "a\xd4";
1316 print "not " unless "ba\xd4c" =~ /([a\x{d4}]+)/ && $1 eq "a\xd4";
1319 print "not " unless "ba\xd4c" =~ /([a\x{d4}]+)/ && $1 eq "a\x{d4}";
1322 print "not " unless "ba\x{d4}c" =~ /([a\x{d4}]+)/ && $1 eq "a\x{d4}";
1325 print "not " unless "ba\x{d4}c" =~ /([a\x{d4}]+)/ && $1 eq "a\xd4";
1330 # the first half of 20001028.003
1333 my ($Y) = $X =~ /(.*)/;
1334 print "not " unless $Y eq v1448 && length($Y) == 1;
1341 my $X = "Szab\x{f3},Bal\x{e1}zs";
1343 $Y =~ s/(B)/$1/ for 0..3;
1344 print "not " unless $Y eq $X && $X eq "Szab\x{f3},Bal\x{e1}zs";
1349 # the second half of 20001028.003
1352 $X =~ s/^/chr(1488)/e;
1353 print "not " unless length $X == 1 && ord($X) == 1488;
1364 print "not " unless $x eq "\x{100}B" && length($x) == 2;
1369 # bug id 20001230.002
1371 print "not " unless "École" =~ /^\C\C(.)/ && $1 eq 'c';
1374 print "not " unless "École" =~ /^\C\C(c)/;
1379 my $test = 264; # till 575
1381 use charnames ":full";
1383 # This is far from complete testing, there are dozens of character
1384 # classes in Unicode. The mixing of literals and \N{...} is
1385 # intentional so that in non-Latin-1 places we test the native
1386 # characters, not the Unicode code points.
1390 "\N{CYRILLIC SMALL LETTER A}" => 'Ll',
1392 "\N{GREEK CAPITAL LETTER ALPHA}" => 'Lu',
1393 "\N{HIRAGANA LETTER SMALL A}" => 'Lo',
1394 "\N{COMBINING GRAVE ACCENT}" => 'Mn',
1396 "\N{ARABIC-INDIC DIGIT ZERO}" => 'Nd',
1403 for my $char (map { s/^\S+ //; $_ }
1404 sort map { sprintf("%06x", ord($_))." $_" } keys %s) {
1405 my $class = $s{$char};
1406 my $code = sprintf("%06x", ord($char));
1407 printf "#\n# 0x$code\n#\n";
1408 print "# IsAlpha\n";
1409 if ($class =~ /^[LM]/) {
1410 print "not " unless $char =~ /\p{IsAlpha}/;
1411 print "ok $test\n"; $test++;
1412 print "not " if $char =~ /\P{IsAlpha}/;
1413 print "ok $test\n"; $test++;
1415 print "not " if $char =~ /\p{IsAlpha}/;
1416 print "ok $test\n"; $test++;
1417 print "not " unless $char =~ /\P{IsAlpha}/;
1418 print "ok $test\n"; $test++;
1420 print "# IsAlnum\n";
1421 if ($class =~ /^[LMN]/ && $char ne "_") {
1422 print "not " unless $char =~ /\p{IsAlnum}/;
1423 print "ok $test\n"; $test++;
1424 print "not " if $char =~ /\P{IsAlnum}/;
1425 print "ok $test\n"; $test++;
1427 print "not " if $char =~ /\p{IsAlnum}/;
1428 print "ok $test\n"; $test++;
1429 print "not " unless $char =~ /\P{IsAlnum}/;
1430 print "ok $test\n"; $test++;
1432 print "# IsASCII\n";
1433 if (ord("A") == 193) {
1434 print "ok $test # Skip: in EBCDIC\n"; $test++;
1435 print "ok $test # Skip: in EBCDIC\n"; $test++;
1437 if ($code le '00007f') {
1438 print "not " unless $char =~ /\p{IsASCII}/;
1439 print "ok $test\n"; $test++;
1440 print "not " if $char =~ /\P{IsASCII}/;
1441 print "ok $test\n"; $test++;
1443 print "not " if $char =~ /\p{IsASCII}/;
1444 print "ok $test\n"; $test++;
1445 print "not " unless $char =~ /\P{IsASCII}/;
1446 print "ok $test\n"; $test++;
1449 print "# IsCntrl\n";
1450 if ($class =~ /^C/) {
1451 print "not " unless $char =~ /\p{IsCntrl}/;
1452 print "ok $test\n"; $test++;
1453 print "not " if $char =~ /\P{IsCntrl}/;
1454 print "ok $test\n"; $test++;
1456 print "not " if $char =~ /\p{IsCntrl}/;
1457 print "ok $test\n"; $test++;
1458 print "not " unless $char =~ /\P{IsCntrl}/;
1459 print "ok $test\n"; $test++;
1461 print "# IsBlank\n";
1462 if ($class =~ /^Z[lp]/ || $char eq " ") {
1463 print "not " unless $char =~ /\p{IsBlank}/;
1464 print "ok $test\n"; $test++;
1465 print "not " if $char =~ /\P{IsBlank}/;
1466 print "ok $test\n"; $test++;
1468 print "not " if $char =~ /\p{IsBlank}/;
1469 print "ok $test\n"; $test++;
1470 print "not " unless $char =~ /\P{IsBlank}/;
1471 print "ok $test\n"; $test++;
1473 print "# IsDigit\n";
1474 if ($class =~ /^Nd$/) {
1475 print "not " unless $char =~ /\p{IsDigit}/;
1476 print "ok $test\n"; $test++;
1477 print "not " if $char =~ /\P{IsDigit}/;
1478 print "ok $test\n"; $test++;
1480 print "not " if $char =~ /\p{IsDigit}/;
1481 print "ok $test\n"; $test++;
1482 print "not " unless $char =~ /\P{IsDigit}/;
1483 print "ok $test\n"; $test++;
1485 print "# IsGraph\n";
1486 if ($class =~ /^([LMNPS])|Co/) {
1487 print "not " unless $char =~ /\p{IsGraph}/;
1488 print "ok $test\n"; $test++;
1489 print "not " if $char =~ /\P{IsGraph}/;
1490 print "ok $test\n"; $test++;
1492 print "not " if $char =~ /\p{IsGraph}/;
1493 print "ok $test\n"; $test++;
1494 print "not " unless $char =~ /\P{IsGraph}/;
1495 print "ok $test\n"; $test++;
1497 print "# IsLower\n";
1498 if ($class =~ /^Ll$/) {
1499 print "not " unless $char =~ /\p{IsLower}/;
1500 print "ok $test\n"; $test++;
1501 print "not " if $char =~ /\P{IsLower}/;
1502 print "ok $test\n"; $test++;
1504 print "not " if $char =~ /\p{IsLower}/;
1505 print "ok $test\n"; $test++;
1506 print "not " unless $char =~ /\P{IsLower}/;
1507 print "ok $test\n"; $test++;
1509 print "# IsPrint\n";
1510 if ($class =~ /^([LMNPS])|Co|Zs/) {
1511 print "not " unless $char =~ /\p{IsPrint}/;
1512 print "ok $test\n"; $test++;
1513 print "not " if $char =~ /\P{IsPrint}/;
1514 print "ok $test\n"; $test++;
1516 print "not " if $char =~ /\p{IsPrint}/;
1517 print "ok $test\n"; $test++;
1518 print "not " unless $char =~ /\P{IsPrint}/;
1519 print "ok $test\n"; $test++;
1521 print "# IsPunct\n";
1522 if ($class =~ /^P/ || $char eq "_") {
1523 print "not " unless $char =~ /\p{IsPunct}/;
1524 print "ok $test\n"; $test++;
1525 print "not " if $char =~ /\P{IsPunct}/;
1526 print "ok $test\n"; $test++;
1528 print "not " if $char =~ /\p{IsPunct}/;
1529 print "ok $test\n"; $test++;
1530 print "not " unless $char =~ /\P{IsPunct}/;
1531 print "ok $test\n"; $test++;
1533 print "# IsSpace\n";
1534 if ($class =~ /^Z/ || ($code =~ /^(0009|000A|000B|000C|000D)$/)) {
1535 print "not " unless $char =~ /\p{IsSpace}/;
1536 print "ok $test\n"; $test++;
1537 print "not " if $char =~ /\P{IsSpace}/;
1538 print "ok $test\n"; $test++;
1540 print "not " if $char =~ /\p{IsSpace}/;
1541 print "ok $test\n"; $test++;
1542 print "not " unless $char =~ /\P{IsSpace}/;
1543 print "ok $test\n"; $test++;
1545 print "# IsUpper\n";
1546 if ($class =~ /^L[ut]/) {
1547 print "not " unless $char =~ /\p{IsUpper}/;
1548 print "ok $test\n"; $test++;
1549 print "not " if $char =~ /\P{IsUpper}/;
1550 print "ok $test\n"; $test++;
1552 print "not " if $char =~ /\p{IsUpper}/;
1553 print "ok $test\n"; $test++;
1554 print "not " unless $char =~ /\P{IsUpper}/;
1555 print "ok $test\n"; $test++;
1558 if ($class =~ /^[LMN]/ || $char eq "_") {
1559 print "not " unless $char =~ /\p{IsWord}/;
1560 print "ok $test\n"; $test++;
1561 print "not " if $char =~ /\P{IsWord}/;
1562 print "ok $test\n"; $test++;
1564 print "not " if $char =~ /\p{IsWord}/;
1565 print "ok $test\n"; $test++;
1566 print "not " unless $char =~ /\P{IsWord}/;
1567 print "ok $test\n"; $test++;
1573 $_ = "abc\x{100}\x{200}\x{300}\x{380}\x{400}defg";
1575 if (/(.\x{300})./) {
1578 print "not " unless $` eq "abc\x{100}" && length($`) == 4;
1581 print "not " unless $& eq "\x{200}\x{300}\x{380}" && length($&) == 3;
1584 print "not " unless $' eq "\x{400}defg" && length($') == 5;
1587 print "not " unless $1 eq "\x{200}\x{300}" && length($1) == 2;
1590 for (576..580) { print "not ok $_\n" }
1595 # bug id 20010306.008
1598 # The original bug report had 'no utf8' here but that was irrelevant.
1599 $a =~ m/\w/; # used to core dump
1607 # bugid 20010410.006
1609 '/(.*?)\{(.*?)\}/csg',
1610 '/(.*?)\{(.*?)\}/cg',
1611 '/(.*?)\{(.*?)\}/sg',
1612 '/(.*?)\{(.*?)\}/g',
1613 '/(.+?)\{(.+?)\}/csg',
1619 $input = "a{b}c{d}";
1621 while (eval \$input =~ $rx) {
1622 print "# \\\$1 = '\$1' \\\$2 = '\$2'\n";
1626 print "not " unless $i == 2;
1627 print "ok " . $test++ . "\n";
1632 # from Robin Houston
1634 my $x = "\x{10FFFD}";
1636 print "not " unless ord($x) == 0x10FFFD && length($x) == 1;
1643 print "not " if $x =~ /[\x80-\xff]/;
1646 print "not " if $x =~ /[\x80-\x{100}]/;
1649 print "not " if $x =~ /[\x{100}]/;
1652 print "not " if $x =~ /\p{InLatin1Supplement}/;
1655 print "not " unless $x =~ /\P{InLatin1Supplement}/;
1658 print "not " if $x =~ /\p{InLatinExtendedA}/;
1661 print "not " unless $x =~ /\P{InLatinExtendedA}/;
1668 print "not " unless $x =~ /[\x80-\xff]/;
1671 print "not " unless $x =~ /[\x80-\x{100}]/;
1674 print "not " if $x =~ /[\x{100}]/;
1677 print "not " unless $x =~ /\p{InLatin1Supplement}/;
1680 print "not " if $x =~ /\P{InLatin1Supplement}/;
1683 print "not " if $x =~ /\p{InLatinExtendedA}/;
1686 print "not " unless $x =~ /\P{InLatinExtendedA}/;
1693 print "not " unless $x =~ /[\x80-\xff]/;
1696 print "not " unless $x =~ /[\x80-\x{100}]/;
1699 print "not " if $x =~ /[\x{100}]/;
1702 # the next two tests must be ignored on EBCDIC
1703 print "not " unless $x =~ /\p{InLatin1Supplement}/ or ord("A") == 193;
1706 print "not " if $x =~ /\P{InLatin1Supplement}/ and ord("A") != 193;
1709 print "not " if $x =~ /\p{InLatinExtendedA}/;
1712 print "not " unless $x =~ /\P{InLatinExtendedA}/;
1719 print "not " if $x =~ /[\x80-\xff]/;
1722 print "not " unless $x =~ /[\x80-\x{100}]/;
1725 print "not " unless $x =~ /[\x{100}]/;
1728 print "not " if $x =~ /\p{InLatin1Supplement}/;
1731 print "not " unless $x =~ /\P{InLatin1Supplement}/;
1734 print "not " unless $x =~ /\p{InLatinExtendedA}/;
1737 print "not " if $x =~ /\P{InLatinExtendedA}/;
1745 local $SIG{__WARN__} = sub { $w .= shift };
1749 print "not " if $w !~ /^Useless \(\?c\)/;
1754 print "not " if $w !~ /^Useless \(\?-c\)/;
1759 print "not " if $w !~ /^Useless \(\?g\)/;
1764 print "not " if $w !~ /^Useless \(\?-g\)/;
1769 print "not " if $w !~ /^Useless \(\?o\)/;
1774 print "not " if $w !~ /^Useless \(\?-o\)/;
1777 # now test multi-error regexes
1781 print "not " if $w !~ /^Useless \(\?g\).*\nUseless \(\?-o\)/;
1786 print "not " if $w !~ /^Useless \(\?g\).*\nUseless \(\?-c\)/;
1790 eval 'qr/(?o-cg)/'; # (?c) means (?g) error won't be thrown
1791 print "not " if $w !~ /^Useless \(\?o\).*\nUseless \(\?-c\)/;
1796 print "not " if $w !~ /^Useless \(\?o\).*\nUseless \(\?g\).*\nUseless \(\?c\)/;
1800 # More Unicode "class" tests
1803 use charnames ':full';
1805 print "not " unless "\N{LATIN CAPITAL LETTER A}" =~ /\p{InBasicLatin}/;
1808 print "not " unless "\N{LATIN CAPITAL LETTER A WITH GRAVE}" =~ /\p{InLatin1Supplement}/;
1811 print "not " unless "\N{LATIN CAPITAL LETTER A WITH MACRON}" =~ /\p{InLatinExtendedA}/;
1814 print "not " unless "\N{LATIN SMALL LETTER B WITH STROKE}" =~ /\p{InLatinExtendedB}/;
1817 print "not " unless "\N{KATAKANA LETTER SMALL A}" =~ /\p{InKatakana}/;
1823 eval <<"EOT"; die if $@;
1828 /x && print "ok 631\n";
1831 eval <<"EOT"; die if $@;
1836 /x && print "ok 632\n";
1840 sub test_o { $_[0] =~/$_[1]/o; return $1}
1841 if(test_o('abc','(.)..') eq 'a') {
1844 print "not ok 633\n";
1846 if(test_o('abc','..(.)') eq 'a') {
1849 print "not ok 634\n";
1852 # 635..639: ID 20010619.003 (only the space character is
1853 # supposed to be [:print:], not the whole isprint()).
1855 print "not " if "\n" =~ /[[:print:]]/;
1858 print "not " if "\t" =~ /[[:print:]]/;
1861 # Amazingly vertical tabulator is the same in ASCII and EBCDIC.
1862 print "not " if "\014" =~ /[[:print:]]/;
1865 print "not " if "\r" =~ /[[:print:]]/;
1868 print "not " unless " " =~ /[[:print:]]/;
1872 ## Test basic $^N usage outside of a regex
1875 $T="ok 640\n";if ($x =~ /cde/ and not defined $^N) {print $T} else {print "not $T"};
1876 $T="ok 641\n";if ($x =~ /(cde)/ and $^N eq "cde") {print $T} else {print "not $T"};
1877 $T="ok 642\n";if ($x =~ /(c)(d)(e)/ and $^N eq "e") {print $T} else {print "not $T"};
1878 $T="ok 643\n";if ($x =~ /(c(d)e)/ and $^N eq "cde") {print $T} else {print "not $T"};
1879 $T="ok 644\n";if ($x =~ /(foo)|(c(d)e)/ and $^N eq "cde") {print $T} else {print "not $T"};
1880 $T="ok 645\n";if ($x =~ /(c(d)e)|(foo)/ and $^N eq "cde") {print $T} else {print "not $T"};
1881 $T="ok 646\n";if ($x =~ /(c(d)e)|(abc)/ and $^N eq "abc") {print $T} else {print "not $T"};
1882 $T="ok 647\n";if ($x =~ /(c(d)e)|(abc)x/ and $^N eq "cde") {print $T} else {print "not $T"};
1883 $T="ok 648\n";if ($x =~ /(c(d)e)(abc)?/ and $^N eq "cde") {print $T} else {print "not $T"};
1884 $T="ok 649\n";if ($x =~ /(?:c(d)e)/ and $^N eq "d" ) {print $T} else {print "not $T"};
1885 $T="ok 650\n";if ($x =~ /(?:c(d)e)(?:f)/ and $^N eq "d" ) {print $T} else {print "not $T"};
1886 $T="ok 651\n";if ($x =~ /(?:([abc])|([def]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
1887 $T="ok 652\n";if ($x =~ /(?:([ace])|([bdf]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
1888 $T="ok 653\n";if ($x =~ /(([ace])|([bd]))*/ and $^N eq "e" ){print $T} else {print "not $T"};
1890 $T="ok 654\n";if($x =~ /(([ace])|([bdf]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
1892 ## test to see if $^N is automatically localized -- it should now
1893 ## have the value set in test 653
1894 $T="ok 655\n";if ($^N eq "e" ){print $T} else {print "not $T"};
1897 ## Now test inside (?{...})
1899 $T="ok 656\n";if ($x =~ /a([abc])(?{$y=$^N})c/ and $y eq "b" ){print $T} else {print "not $T"};
1900 $T="ok 657\n";if ($x =~ /a([abc]+)(?{$y=$^N})d/ and $y eq "bc"){print $T} else {print "not $T"};
1901 $T="ok 658\n";if ($x =~ /a([abcdefg]+)(?{$y=$^N})d/ and $y eq "bc"){print $T} else {print "not $T"};
1902 $T="ok 659\n";if ($x =~ /(a([abcdefg]+)(?{$y=$^N})d)(?{$z=$^N})e/ and $y eq "bc" and $z eq "abcd")
1903 {print $T} else {print "not $T"};
1904 $T="ok 660\n";if ($x =~ /(a([abcdefg]+)(?{$y=$^N})de)(?{$z=$^N})/ and $y eq "bc" and $z eq "abcde")
1905 {print $T} else {print "not $T"};
1907 # Test the Unicode script classes
1909 print "not " unless chr(0x100) =~ /\p{IsLatin}/; # outside Latin-1
1912 print "not " unless chr(0x212b) =~ /\p{IsLatin}/; # Angstrom sign, very outside
1915 print "not " unless chr(0x5d0) =~ /\p{IsHebrew}/; # inside InHebrew
1918 print "not " unless chr(0xfb4f) =~ /\p{IsHebrew}/; # outside InHebrew
1921 # # singleton (not in a range, this test must be ignored on EBCDIC)
1922 # print "not " unless chr(0xb5) =~ /\p{IsGreek}/ or ord("A") == 193;
1924 print "ok 665 # 0xb5 moved from Greek to Common with Unicode 4.0.1\n";
1926 print "not " unless chr(0x37a) =~ /\p{IsGreek}/; # singleton
1929 print "not " unless chr(0x386) =~ /\p{IsGreek}/; # singleton
1932 print "not " unless chr(0x387) =~ /\P{IsGreek}/; # not there
1935 print "not " unless chr(0x388) =~ /\p{IsGreek}/; # range
1938 print "not " unless chr(0x38a) =~ /\p{IsGreek}/; # range
1941 print "not " unless chr(0x38b) =~ /\P{IsGreek}/; # not there
1944 print "not " unless chr(0x38c) =~ /\p{IsGreek}/; # singleton
1947 if (ord("A") == 65) {
1949 ## Test [:cntrl:]...
1951 ## Should probably put in tests for all the POSIX stuff, but not sure how to
1952 ## guarantee a specific locale......
1954 $AllBytes = join('', map { chr($_) } 0..255);
1955 ($x = $AllBytes) =~ s/[[:cntrl:]]//g;
1956 if ($x ne join('', map { chr($_) } 0x20..0x7E, 0x80..0xFF)) {
1961 ($x = $AllBytes) =~ s/[^[:cntrl:]]//g;
1962 if ($x ne join('', map { chr($_) } 0..0x1F, 0x7F)) { print "not " }
1965 print "ok $_ # Skip: EBCDIC\n" for 673..674;
1968 # With /s modifier UTF8 chars were interpreted as bytes
1970 my $a = "Hello \x{263A} World";
1972 my @a = ($a =~ /./gs);
1974 print "not " unless $#a == 12;
1978 @a = ("foo\nbar" =~ /./g);
1979 print "ok 676\n" if @a == 6 && "@a" eq "f o o b a r";
1981 @a = ("foo\nbar" =~ /./gs);
1982 print "ok 677\n" if @a == 7 && "@a" eq "f o o \n b a r";
1984 @a = ("foo\nbar" =~ /\C/g);
1985 print "ok 678\n" if @a == 7 && "@a" eq "f o o \n b a r";
1987 @a = ("foo\nbar" =~ /\C/gs);
1988 print "ok 679\n" if @a == 7 && "@a" eq "f o o \n b a r";
1990 @a = ("foo\n\x{100}bar" =~ /./g);
1991 print "ok 680\n" if @a == 7 && "@a" eq "f o o \x{100} b a r";
1993 @a = ("foo\n\x{100}bar" =~ /./gs);
1994 print "ok 681\n" if @a == 8 && "@a" eq "f o o \n \x{100} b a r";
1996 ($a, $b) = map { chr } ord('A') == 65 ? (0xc4, 0x80) : (0x8c, 0x41);
1998 @a = ("foo\n\x{100}bar" =~ /\C/g);
1999 print "ok 682\n" if @a == 9 && "@a" eq "f o o \n $a $b b a r";
2001 @a = ("foo\n\x{100}bar" =~ /\C/gs);
2002 print "ok 683\n" if @a == 9 && "@a" eq "f o o \n $a $b b a r";
2005 # [ID 20010814.004] pos() doesn't work when using =~m// in list context
2007 $a = join ":", /b./gc;
2008 $b = join ":", /a./gc;
2010 print "$a $b $c" eq 'ba:ba ad:ae 10' ? "ok 684\n" : "not ok 684\t# $a $b $c\n";
2014 # [ID 20010407.006] matching utf8 return values from functions does not work
2016 package ID_20010407_006;
2025 $x =~ /(..)/; $y = $1;
2026 print "not " unless length($y) == 2 && $y eq $x;
2029 x =~ /(..)/; $y = $1;
2030 print "not " unless length($y) == 2 && $y eq $x;
2037 # Force scalar context on the patern match
2039 my($ok, $name) = @_;
2041 printf "%sok %d - %s\n", ($ok ? "" : "not "), $test, $name;
2043 printf "# Failed test at line %d\n", (caller)[2] unless $ok;
2050 # Check that \x## works. 5.6.1 and 5.005_03 fail some of these.
2052 ok ($x =~ /^\x4EE$/, "Check only 2 bytes of hex are matched.");
2055 ok ($x =~ /^\x4Ei$/, "Check that invalid hex digit stops it (2)");
2058 ok ($x =~ /^\x4j$/, "Check that invalid hex digit stops it (1)");
2061 ok ($x =~ /^\xk$/, "Check that invalid hex digit stops it (0)");
2064 ok ($x =~ /^\xx$/, "\\xx isn't to be treated as \\0");
2067 ok ($x =~ /^\xxa$/, "\\xxa isn't to be treated as \\xa");
2070 ok ($x =~ /^\x9_b$/, "\\x9_b isn't to be treated as \\x9b");
2072 print "# and now again in [] ranges\n";
2075 ok ($x =~ /^[\x4EE]{2}$/, "Check only 2 bytes of hex are matched.");
2078 ok ($x =~ /^[\x4Ei]{2}$/, "Check that invalid hex digit stops it (2)");
2081 ok ($x =~ /^[\x4j]{2}$/, "Check that invalid hex digit stops it (1)");
2084 ok ($x =~ /^[\xk]{2}$/, "Check that invalid hex digit stops it (0)");
2087 ok ($x =~ /^[\xx]{2}$/, "\\xx isn't to be treated as \\0");
2090 ok ($x =~ /^[\xxa]{3}$/, "\\xxa isn't to be treated as \\xa");
2093 ok ($x =~ /^[\x9_b]{3}$/, "\\x9_b isn't to be treated as \\x9b");
2098 # Check that \x{##} works. 5.6.1 fails quite a few of these.
2101 ok ($x =~ /^\x{9_b}$/, "\\x{9_b} is to be treated as \\x9b");
2104 ok ($x =~ /^\x{9_b}y$/, "\\x{9_b} is to be treated as \\x9b (again)");
2107 ok ($x =~ /^\x{9b_}y$/, "\\x{9b_} is to be treated as \\x9b");
2110 ok ($x =~ /^\x{9_bq}y$/, "\\x{9_bc} is to be treated as \\x9b");
2113 ok ($x =~ /^\x{x9b}y$/, "\\x{x9b} is to be treated as \\x0");
2116 ok ($x =~ /^\x{0x9b}y$/, "\\x{0x9b} is to be treated as \\x0");
2119 ok ($x =~ /^\x{09b}y$/, "\\x{09b} is to be treated as \\x9b");
2121 print "# and now again in [] ranges\n";
2124 ok ($x =~ /^[\x{9_b}]$/, "\\x{9_b} is to be treated as \\x9b");
2127 ok ($x =~ /^[\x{9_b}y]{2}$/, "\\x{9_b} is to be treated as \\x9b (again)");
2130 ok ($x =~ /^[\x{9b_}y]{2}$/, "\\x{9b_} is to be treated as \\x9b");
2133 ok ($x =~ /^[\x{9_bq}y]{2}$/, "\\x{9_bc} is to be treated as \\x9b");
2136 ok ($x =~ /^[\x{x9b}y]{2}$/, "\\x{x9b} is to be treated as \\x0");
2139 ok ($x =~ /^[\x{0x9b}y]{2}$/, "\\x{0x9b} is to be treated as \\x0");
2142 ok ($x =~ /^[\x{09b}y]{2}$/, "\\x{09b} is to be treated as \\x9b");
2146 # high bit bug -- japhy
2148 $x =~ /.*?\200/ or print "not ";
2152 print "# some Unicode properties\n";
2155 # Dashes, underbars, case.
2156 print "not " unless "\x80" =~ /\p{in-latin1_SUPPLEMENT}/;
2159 # Complement, leading and trailing whitespace.
2160 print "not " unless "\x80" =~ /\P{ ^ In Latin 1 Supplement }/;
2163 # No ^In, dashes, case, dash, any intervening (word-break) whitespace.
2164 # (well, newlines don't work...)
2165 print "not " unless "\x80" =~ /\p{latin-1 supplement}/;
2170 print "not " unless "a" =~ /\pL/;
2173 print "not " unless "a" =~ /\p{IsLl}/;
2176 print "not " if "a" =~ /\p{IsLu}/;
2179 print "not " unless "a" =~ /\p{Ll}/;
2182 print "not " if "a" =~ /\p{Lu}/;
2185 print "not " unless "A" =~ /\pL/;
2188 print "not " unless "A" =~ /\p{IsLu}/;
2191 print "not " if "A" =~ /\p{IsLl}/;
2194 print "not " unless "A" =~ /\p{Lu}/;
2197 print "not " if "A" =~ /\p{Ll}/;
2200 print "not " if "a" =~ /\PL/;
2203 print "not " if "a" =~ /\P{IsLl}/;
2206 print "not " unless "a" =~ /\P{IsLu}/;
2209 print "not " if "a" =~ /\P{Ll}/;
2212 print "not " unless "a" =~ /\P{Lu}/;
2215 print "not " if "A" =~ /\PL/;
2218 print "not " if "A" =~ /\P{IsLu}/;
2221 print "not " unless "A" =~ /\P{IsLl}/;
2224 print "not " if "A" =~ /\P{Lu}/;
2227 print "not " unless "A" =~ /\P{Ll}/;
2233 print "not " if "a" =~ /\p{Common}/;
2236 print "not " unless "1" =~ /\p{Common}/;
2241 print "not " if "a" =~ /\p{Inherited}/;
2244 print "not " unless "\x{300}" =~ /\p{Inherited}/;
2249 # L& and LC are the same
2250 print "not " unless "a" =~ /\p{LC}/ and "a" =~ /\p{L&}/;
2253 print "not " if "1" =~ /\p{LC}/ or "1" =~ /\p{L&}/;
2258 print "not " unless "a" =~ /\p{Lowercase Letter}/;
2261 print "not " if "A" =~ /\p{lowercaseletter}/;
2266 print "not " unless "\x{AC00}" =~ /\p{HangulSyllables}/;
2271 # Script=, Block=, Category=
2273 print "not " unless "\x{0100}" =~ /\p{Script=Latin}/;
2276 print "not " unless "\x{0100}" =~ /\p{Block=LatinExtendedA}/;
2279 print "not " unless "\x{0100}" =~ /\p{Category=UppercaseLetter}/;
2284 print "# the basic character classes and Unicode \n";
2286 # 0100;LATIN CAPITAL LETTER A WITH MACRON;Lu;0;L;0041 0304;;;;N;LATIN CAPITAL LETTER A MACRON;;;0101;
2287 print "not " unless "\x{0100}" =~ /\w/;
2290 # 0660;ARABIC-INDIC DIGIT ZERO;Nd;0;AN;;0;0;0;N;;;;;
2291 print "not " unless "\x{0660}" =~ /\d/;
2294 # 1680;OGHAM SPACE MARK;Zs;0;WS;;;;;N;;;;;
2295 print "not " unless "\x{1680}" =~ /\s/;
2300 print "# folding matches and Unicode\n";
2302 print "not " unless "a\x{100}" =~ /A/i;
2305 print "not " unless "A\x{100}" =~ /a/i;
2308 print "not " unless "a\x{100}" =~ /a/i;
2311 print "not " unless "A\x{100}" =~ /A/i;
2314 print "not " unless "\x{101}a" =~ /\x{100}/i;
2317 print "not " unless "\x{100}a" =~ /\x{100}/i;
2320 print "not " unless "\x{101}a" =~ /\x{101}/i;
2323 print "not " unless "\x{100}a" =~ /\x{101}/i;
2326 print "not " unless "a\x{100}" =~ /A\x{100}/i;
2329 print "not " unless "A\x{100}" =~ /a\x{100}/i;
2332 print "not " unless "a\x{100}" =~ /a\x{100}/i;
2335 print "not " unless "A\x{100}" =~ /A\x{100}/i;
2338 print "not " unless "a\x{100}" =~ /[A]/i;
2341 print "not " unless "A\x{100}" =~ /[a]/i;
2344 print "not " unless "a\x{100}" =~ /[a]/i;
2347 print "not " unless "A\x{100}" =~ /[A]/i;
2350 print "not " unless "\x{101}a" =~ /[\x{100}]/i;
2353 print "not " unless "\x{100}a" =~ /[\x{100}]/i;
2356 print "not " unless "\x{101}a" =~ /[\x{101}]/i;
2359 print "not " unless "\x{100}a" =~ /[\x{101}]/i;
2365 use charnames ':full';
2367 print "# LATIN LETTER A WITH GRAVE\n";
2368 my $lower = "\N{LATIN SMALL LETTER A WITH GRAVE}";
2369 my $UPPER = "\N{LATIN CAPITAL LETTER A WITH GRAVE}";
2371 print $lower =~ m/$UPPER/i ? "ok 774\n" : "not ok 774\n";
2372 print $UPPER =~ m/$lower/i ? "ok 775\n" : "not ok 775\n";
2373 print $lower =~ m/[$UPPER]/i ? "ok 776\n" : "not ok 776\n";
2374 print $UPPER =~ m/[$lower]/i ? "ok 777\n" : "not ok 777\n";
2376 print "# GREEK LETTER ALPHA WITH VRACHY\n";
2378 $lower = "\N{GREEK CAPITAL LETTER ALPHA WITH VRACHY}";
2379 $UPPER = "\N{GREEK SMALL LETTER ALPHA WITH VRACHY}";
2381 print $lower =~ m/$UPPER/i ? "ok 778\n" : "not ok 778\n";
2382 print $UPPER =~ m/$lower/i ? "ok 779\n" : "not ok 779\n";
2383 print $lower =~ m/[$UPPER]/i ? "ok 780\n" : "not ok 780\n";
2384 print $UPPER =~ m/[$lower]/i ? "ok 781\n" : "not ok 781\n";
2386 print "# LATIN LETTER Y WITH DIAERESIS\n";
2388 $lower = "\N{LATIN CAPITAL LETTER Y WITH DIAERESIS}";
2389 $UPPER = "\N{LATIN SMALL LETTER Y WITH DIAERESIS}";
2390 print $lower =~ m/$UPPER/i ? "ok 782\n" : "not ok 782\n";
2391 print $UPPER =~ m/$lower/i ? "ok 783\n" : "not ok 783\n";
2392 print $lower =~ m/[$UPPER]/i ? "ok 784\n" : "not ok 784\n";
2393 print $UPPER =~ m/[$lower]/i ? "ok 785\n" : "not ok 785\n";
2398 use charnames ':full';
2400 print "# GREEK CAPITAL LETTER SIGMA vs COMBINING GREEK PERISPOMENI\n";
2402 my $SIGMA = "\N{GREEK CAPITAL LETTER SIGMA}";
2403 my $char = "\N{COMBINING GREEK PERISPOMENI}";
2405 # Before #13843 this was failing by matching falsely.
2406 print "_:$char:_" =~ m/_:$SIGMA:_/i ? "not ok 786\n" : "ok 786\n";
2412 use charnames ':full';
2414 print "a!" =~ /^(\X)!/ && $1 eq "a" ?
2415 "ok 787\n" : "not ok 787 # $1\n";
2416 print "\xDF!" =~ /^(\X)!/ && $1 eq "\xDF" ?
2417 "ok 788\n" : "not ok 788 # $1\n";
2418 print "\x{100}!" =~ /^(\X)!/ && $1 eq "\x{100}" ?
2419 "ok 789\n" : "not ok 789 # $1\n";
2420 print "\x{100}\x{300}!" =~ /^(\X)!/ && $1 eq "\x{100}\x{300}" ?
2421 "ok 790\n" : "not ok 790 # $1\n";
2422 print "\N{LATIN CAPITAL LETTER E}!" =~ /^(\X)!/ &&
2423 $1 eq "\N{LATIN CAPITAL LETTER E}" ?
2424 "ok 791\n" : "not ok 791 # $1\n";
2425 print "\N{LATIN CAPITAL LETTER E}\N{COMBINING GRAVE ACCENT}!" =~
2427 $1 eq "\N{LATIN CAPITAL LETTER E}\N{COMBINING GRAVE ACCENT}" ?
2428 "ok 792\n" : "not ok 792 # $1\n";
2432 print "#\\C and \\X\n";
2434 print "!abc!" =~ /a\Cc/ ? "ok 793\n" : "not ok 793\n";
2435 print "!abc!" =~ /a\Xc/ ? "ok 794\n" : "not ok 794\n";
2439 print "# FINAL SIGMA\n";
2441 my $SIGMA = "\x{03A3}"; # CAPITAL
2442 my $Sigma = "\x{03C2}"; # SMALL FINAL
2443 my $sigma = "\x{03C3}"; # SMALL
2445 print $SIGMA =~ /$SIGMA/i ? "ok 795\n" : "not ok 795\n";
2446 print $SIGMA =~ /$Sigma/i ? "ok 796\n" : "not ok 796\n";
2447 print $SIGMA =~ /$sigma/i ? "ok 797\n" : "not ok 797\n";
2449 print $Sigma =~ /$SIGMA/i ? "ok 798\n" : "not ok 798\n";
2450 print $Sigma =~ /$Sigma/i ? "ok 799\n" : "not ok 799\n";
2451 print $Sigma =~ /$sigma/i ? "ok 800\n" : "not ok 800\n";
2453 print $sigma =~ /$SIGMA/i ? "ok 801\n" : "not ok 801\n";
2454 print $sigma =~ /$Sigma/i ? "ok 802\n" : "not ok 802\n";
2455 print $sigma =~ /$sigma/i ? "ok 803\n" : "not ok 803\n";
2457 print $SIGMA =~ /[$SIGMA]/i ? "ok 804\n" : "not ok 804\n";
2458 print $SIGMA =~ /[$Sigma]/i ? "ok 805\n" : "not ok 805\n";
2459 print $SIGMA =~ /[$sigma]/i ? "ok 806\n" : "not ok 806\n";
2461 print $Sigma =~ /[$SIGMA]/i ? "ok 807\n" : "not ok 807\n";
2462 print $Sigma =~ /[$Sigma]/i ? "ok 808\n" : "not ok 808\n";
2463 print $Sigma =~ /[$sigma]/i ? "ok 809\n" : "not ok 809\n";
2465 print $sigma =~ /[$SIGMA]/i ? "ok 810\n" : "not ok 810\n";
2466 print $sigma =~ /[$Sigma]/i ? "ok 811\n" : "not ok 811\n";
2467 print $sigma =~ /[$sigma]/i ? "ok 812\n" : "not ok 812\n";
2471 print "# parlez-vous?\n";
2473 use charnames ':full';
2475 print "fran\N{LATIN SMALL LETTER C}ais" =~
2478 "ok 813\n" : "not ok 813\n";
2480 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2482 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2483 "ok 814\n" : "not ok 814\n";
2485 print "fran\N{LATIN SMALL LETTER C}ais" =~
2488 "ok 815\n" : "not ok 815\n";
2490 print "franc\N{COMBINING CEDILLA}ais" =~
2491 /franc\C\Cais/ ? # COMBINING CEDILLA is two bytes when encoded
2492 "ok 816\n" : "not ok 816\n";
2494 print "fran\N{LATIN SMALL LETTER C}ais" =~
2497 "ok 817\n" : "not ok 817\n";
2499 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2501 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2502 "ok 818\n" : "not ok 818\n";
2504 print "franc\N{COMBINING CEDILLA}ais" =~
2506 $& eq "franc\N{COMBINING CEDILLA}ais" ?
2507 "ok 819\n" : "not ok 819\n";
2509 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2510 /fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais/ &&
2511 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2512 "ok 820\n" : "not ok 820\n";
2514 print "franc\N{COMBINING CEDILLA}ais" =~
2515 /franc\N{COMBINING CEDILLA}ais/ &&
2516 $& eq "franc\N{COMBINING CEDILLA}ais" ?
2517 "ok 821\n" : "not ok 821\n";
2519 print "fran\N{LATIN SMALL LETTER C}ais" =~
2520 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2522 "ok 822\n" : "not ok 822\n";
2524 print "fran\N{LATIN SMALL LETTER C}ais" =~
2525 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2527 "ok 823\n" : "not ok 823\n";
2529 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2530 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2531 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2532 "ok 824\n" : "not ok 824\n";
2534 print "franc\N{COMBINING CEDILLA}ais" =~
2535 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2536 $& eq "franc\N{COMBINING CEDILLA}ais" ?
2537 "ok 825\n" : "not ok 825\n";
2541 print "# Does lingering (and useless) UTF8 flag mess up /i matching?\n";
2544 my $regex = "ABcde";
2545 my $string = "abcDE\x{100}";
2547 if ($string =~ m/$regex/i) {
2550 print "not ok 826\n";
2555 my $regex = "ABcde\x{100}";
2556 my $string = "abcDE";
2558 if ($string =~ m/$regex/i) {
2561 print "not ok 827\n";
2566 my $regex = "ABcde\x{100}";
2567 my $string = "abcDE\x{100}";
2570 if ($string =~ m/$regex/i) {
2573 print "not ok 828\n";
2579 print "# more SIGMAs\n";
2581 my $SIGMA = "\x{03A3}"; # CAPITAL
2582 my $Sigma = "\x{03C2}"; # SMALL FINAL
2583 my $sigma = "\x{03C3}"; # SMALL
2585 my $S3 = "$SIGMA$Sigma$sigma";
2587 print ":$S3:" =~ /:(($SIGMA)+):/i && $1 eq $S3 && $2 eq $sigma ?
2588 "ok 829\n" : "not ok 829\n";
2589 print ":$S3:" =~ /:(($Sigma)+):/i && $1 eq $S3 && $2 eq $sigma ?
2590 "ok 830\n" : "not ok 830\n";
2591 print ":$S3:" =~ /:(($sigma)+):/i && $1 eq $S3 && $2 eq $sigma ?
2592 "ok 831\n" : "not ok 831\n";
2594 print ":$S3:" =~ /:(([$SIGMA])+):/i && $1 eq $S3 && $2 eq $sigma ?
2595 "ok 832\n" : "not ok 832\n";
2596 print ":$S3:" =~ /:(([$Sigma])+):/i && $1 eq $S3 && $2 eq $sigma ?
2597 "ok 833\n" : "not ok 833\n";
2598 print ":$S3:" =~ /:(([$sigma])+):/i && $1 eq $S3 && $2 eq $sigma ?
2599 "ok 834\n" : "not ok 834\n";
2603 print "# LATIN SMALL LETTER SHARP S\n";
2605 use charnames ':full';
2607 print "\N{LATIN SMALL LETTER SHARP S}" =~
2608 /\N{LATIN SMALL LETTER SHARP S}/ ? "ok 835\n" : "not ok 835\n";
2610 print "\N{LATIN SMALL LETTER SHARP S}" =~
2611 /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 836\n" : "not ok 836\n";
2613 print "\N{LATIN SMALL LETTER SHARP S}" =~
2614 /[\N{LATIN SMALL LETTER SHARP S}]/ ? "ok 837\n" : "not ok 837\n";
2616 print "\N{LATIN SMALL LETTER SHARP S}" =~
2617 /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 838\n" : "not ok 838\n";
2620 /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 839\n" : "not ok 839\n";
2623 /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 840\n" : "not ok 840\n";
2626 /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 841\n" : "not ok 841\n";
2629 /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 842\n" : "not ok 842\n";
2631 print "\N{LATIN SMALL LETTER SHARP S}" =~ /ss/i ?
2632 "ok 843\n" : "not ok 843\n";
2634 print "\N{LATIN SMALL LETTER SHARP S}" =~ /SS/i ?
2635 "ok 844\n" : "not ok 844\n";
2639 print "# more whitespace: U+0085, U+2028, U+2029\n";
2641 # U+0085 needs to be forced to be Unicode, the \x{100} does that.
2643 print "<\x{100}\x{0085}>" =~ /<\x{100}e>/ ? "ok 845\n" : "not ok 845\n";
2645 print "<\x{100}\x{0085}>" =~ /<\x{100}\s>/ ? "ok 845\n" : "not ok 845\n";
2647 print "<\x{2028}>" =~ /<\s>/ ? "ok 846\n" : "not ok 846\n";
2648 print "<\x{2029}>" =~ /<\s>/ ? "ok 847\n" : "not ok 847\n";
2652 print "# . with /s should work on characters, as opposed to bytes\n";
2654 my $s = "\x{e4}\x{100}";
2656 # This is not expected to match: the point is that
2657 # neither should we get "Malformed UTF-8" warnings.
2658 print $s =~ /\G(.+?)\n/gcs ?
2659 "not ok 848\n" : "ok 848\n";
2663 while ($s =~ /\G(.)/gs) {
2667 print join("", @c) eq $s ? "ok 849\n" : "not ok 849\n";
2669 my $t1 = "Q003\n\n\x{e4}\x{f6}\n\nQ004\n\n\x{e7}"; # test only chars < 256
2671 while ($t1 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) {
2675 my $t2 = $t1 . "\x{100}"; # repeat with a larger char
2677 while ($t2 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) {
2681 print $r1 eq $r2 ? "ok 850\n" : "not ok 850\n";
2685 print "# Unicode lookbehind\n";
2687 print "A\x{100}B" =~ /(?<=A.)B/ ? "ok 851\n" : "not ok 851\n";
2688 print "A\x{200}\x{300}B" =~ /(?<=A..)B/ ? "ok 852\n" : "not ok 852\n";
2689 print "\x{400}AB" =~ /(?<=\x{400}.)B/ ? "ok 853\n" : "not ok 853\n";
2690 print "\x{500\x{600}}B" =~ /(?<=\x{500}.)B/ ? "ok 854\n" : "not ok 854\n";
2694 print "# UTF-8 hash keys and /\$/\n";
2695 # http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2002-01/msg01327.html
2698 my $v = substr($u,0,1);
2699 my $w = substr($u,1,1);
2700 my %u = ( $u => $u, $v => $v, $w => $w );
2703 my $m1 = /^\w*$/ ? 1 : 0;
2704 my $m2 = $u{$_}=~/^\w*$/ ? 1 : 0;
2705 print $m1 == $m2 ? "ok $i\n" : "not ok $i # $m1 $m2\n";
2711 print "# [ID 20020124.005]\n";
2714 for my $char ("a", "\x{df}", "\x{100}"){
2715 $x = "$char b $char";
2720 print substr($x,0,1) eq substr($x,-1,1) ?
2721 "ok $i\n" : "not ok $i # debug: $x\n";
2727 print "# SEGV in s/// and UTF-8\n";
2728 $s = "s#\x{100}" x 4;
2730 print $s eq "s \x{100}" x 4 ? "ok 861\n" : "not ok 861\n";
2734 print "# UTF-8 bug (maybe alreayd known?)\n";
2738 $u =~ s/./\x{100}/g;
2739 print $u eq "\x{100}\x{100}\x{100}" ? "ok 862\n" : "not ok 862\n";
2742 $u =~ s/[ao]/\x{100}/g;
2743 print $u eq "f\x{100}\x{100}b\x{100}r" ? "ok 863\n" : "not ok 863\n";
2745 $u =~ s/\x{100}/e/g;
2746 print $u eq "feeber" ? "ok 864\n" : "not ok 864\n";
2750 print "# UTF-8 bug with s///\n";
2751 # check utf8/non-utf8 mixtures
2752 # try to force all float/anchored check combinations
2757 "xx.*$c", "x.*$c$c", "$c.*xx", "$c$c.*x", "xx.*(?=$c)", "(?=$c).*xx",
2759 print "xxx" =~ /$re/ ? "not ok $test\n" : "ok $test\n";
2761 print +($subst = "xxx") =~ s/$re// ? "not ok $test\n" : "ok $test\n";
2764 for my $re ("xx.*$c*", "$c*.*xx") {
2765 print "xxx" =~ /$re/ ? "ok $test\n" : "not ok $test\n";
2767 ($subst = "xxx") =~ s/$re//;
2768 print $subst eq '' ? "ok $test\n" : "not ok $test\t# $subst\n";
2771 for my $re ("xxy*", "y*xx") {
2772 print "xx$c" =~ /$re/ ? "ok $test\n" : "not ok $test\n";
2774 ($subst = "xx$c") =~ s/$re//;
2775 print $subst eq $c ? "ok $test\n" : "not ok $test\n";
2777 print "xy$c" =~ /$re/ ? "not ok $test\n" : "ok $test\n";
2779 print +($subst = "xy$c") =~ /$re/ ? "not ok $test\n" : "ok $test\n";
2782 for my $re ("xy$c*z", "x$c*yz") {
2783 print "xyz" =~ /$re/ ? "ok $test\n" : "not ok $test\n";
2785 ($subst = "xyz") =~ s/$re//;
2786 print $subst eq '' ? "ok $test\n" : "not ok $test\n";
2792 print "# qr/.../x\n";
2795 my $R = qr/ A B C # D E/x;
2797 print eval {"ABCDE" =~ $R} ? "ok $test\n" : "not ok $test\n";
2800 print eval {"ABCDE" =~ m/$R/} ? "ok $test\n" : "not ok $test\n";
2803 print eval {"ABCDE" =~ m/($R)/} ? "ok $test\n" : "not ok $test\n";
2808 print "# illegal Unicode properties\n";
2811 print eval { "a" =~ /\pq / } ? "not ok $test\n" : "ok $test\n";
2814 print eval { "a" =~ /\p{qrst} / } ? "not ok $test\n" : "ok $test\n";
2819 print "# [ID 20020412.005] wrong pmop flags checked when empty pattern\n";
2820 # requires reuse of last successful pattern
2828 printf "not ok %s\t# 'match once' %s on %s iteration\n", $test,
2829 $match ? 'succeeded' : 'failed', $_ ? 'second' : 'first';
2834 my $result = join '', $test =~ //g;
2835 if ($result eq $test) {
2838 printf "not ok %s\t# expected '%s', got '%s'\n", $test, $test, $result;
2843 print "# user-defined character properties\n";
2877 print "\x{3040}" =~ /\p{InKana1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2878 print "\x{303F}" =~ /\P{InKana1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2880 print "\x{3040}" =~ /\p{InKana2}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2881 print "\x{303F}" =~ /\P{InKana2}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2883 print "\x{3041}" =~ /\p{InKana3}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2884 print "\x{3040}" =~ /\P{InKana3}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2886 print "\x{3040}" =~ /\p{InNotKana}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2887 print "\x{3041}" =~ /\P{InNotKana}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2889 sub InConsonant { # Not EBCDIC-aware.
2900 print "d" =~ /\p{InConsonant}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2901 print "e" =~ /\P{InConsonant}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2904 print "# [ID 20020630.002] utf8 regex only matches 32k\n";
2906 for ([ 'byte', "\x{ff}" ], [ 'utf8', "\x{1ff}" ]) {
2907 my($type, $char) = @$_;
2908 for my $len (32000, 32768, 33000) {
2909 my $s = $char . "f" x $len;
2910 my $r = $s =~ /$char([f]*)/gc;
2911 print $r ? "ok $test\n" : "not ok $test\t# <$type x $len> fail\n";
2913 print +(!$r or pos($s) == $len + 1) ? "ok $test\n"
2914 : "not ok $test\t# <$type x $len> pos @{[ pos($s) ]}\n";
2922 $a = bless qr/foo/, 'Foo';
2923 print(('goodfood' =~ $a ? '' : 'not '),
2924 "ok $test\t# reblessed qr// matches\n");
2927 print(($a eq '(?-xism:foo)' ? '' : 'not '),
2928 "ok $test\t# reblessed qr// stringizes\n");
2932 $z=$y = "\317\276"; # $y is byte representation of $x
2935 print(($x =~ $a ? '' : 'not '), "ok $test - utf8 interpolation in qr//\n");
2938 print(("a$a" =~ $x ? '' : 'not '),
2939 "ok $test - stringifed qr// preserves utf8\n");
2942 print(("a$x" =~ /^a$a\z/ ? '' : 'not '),
2943 "ok $test - interpolated qr// preserves utf8\n");
2946 print(("a$x" =~ /^a(??{$a})\z/ ? '' : 'not '),
2947 "ok $test - postponed interpolation of qr// preserves utf8\n");
2950 print((length(qr/##/x) == 12 ? '' : 'not '),
2951 "ok $test - ## in qr// doesn't corrupt memory [perl #17776]\n");
2956 print(("$x$x" =~ /^$x(??{$x})\z/ ? '' : 'not '),
2957 "ok $test - postponed utf8 string in utf8 re matches utf8\n");
2960 print(("$y$x" =~ /^$y(??{$x})\z/ ? '' : 'not '),
2961 "ok $test - postponed utf8 string in non-utf8 re matches utf8\n");
2964 print(("$y$x" !~ /^$y(??{$y})\z/ ? '' : 'not '),
2965 "ok $test - postponed non-utf8 string in non-utf8 re doesn't match utf8\n");
2968 print(("$x$x" !~ /^$x(??{$y})\z/ ? '' : 'not '),
2969 "ok $test - postponed non-utf8 string in utf8 re doesn't match utf8\n");
2972 print(("$y$y" =~ /^$y(??{$y})\z/ ? '' : 'not '),
2973 "ok $test - postponed non-utf8 string in non-utf8 re matches non-utf8\n");
2976 print(("$x$y" =~ /^$x(??{$y})\z/ ? '' : 'not '),
2977 "ok $test - postponed non-utf8 string in utf8 re matches non-utf8\n");
2979 $y = $z; # reset $y after upgrade
2981 print(("$x$y" !~ /^$x(??{$x})\z/ ? '' : 'not '),
2982 "ok $test - postponed utf8 string in utf8 re doesn't match non-utf8\n");
2984 $y = $z; # reset $y after upgrade
2986 print(("$y$y" !~ /^$y(??{$x})\z/ ? '' : 'not '),
2987 "ok $test - postponed utf8 string in non-utf8 re doesn't match non-utf8\n");
2992 print "# more user-defined character properties\n";
3001 ok("\x{0712}" =~ /\p{IsSyriac1}/, '\x{0712}, \p{IsSyriac1}');
3002 ok("\x{072F}" =~ /\P{IsSyriac1}/, '\x{072F}, \P{IsSyriac1}');
3011 ok("\x{0712}" =~ /\p{Syriac1}/, '\x{0712}, \p{Syriac1}');
3012 ok("\x{072F}" =~ /\P{Syriac1}/, '\x{072F}, \p{Syriac1}');
3014 print "# user-defined character properties may lack \\n at the end\n";
3015 sub InGreekSmall { return "03B1\t03C9" }
3016 sub InGreekCapital { return "0391\t03A9\n-03A2" }
3018 ok("\x{03C0}" =~ /\p{InGreekSmall}/, "Small pi");
3019 ok("\x{03C2}" =~ /\p{InGreekSmall}/, "Final sigma");
3020 ok("\x{03A0}" =~ /\p{InGreekCapital}/, "Capital PI");
3021 ok("\x{03A2}" =~ /\P{InGreekCapital}/, "Reserved");
3023 sub AsciiHexAndDash {
3025 +utf8::ASCII_Hex_Digit
3030 ok("-" =~ /\p{Dash}/, "'-' is Dash");
3031 ok("A" =~ /\p{ASCII_Hex_Digit}/, "'A' is ASCII_Hex_Digit");
3032 ok("-" =~ /\p{AsciiHexAndDash}/, "'-' is AsciiHexAndDash");
3033 ok("A" =~ /\p{AsciiHexAndDash}/, "'A' is AsciiHexAndDash");
3036 print "# Change #18179\n";
3037 # previously failed with "panic: end_shift
3038 my $s = "\x{100}" x 5;
3039 my $ok = $s =~ /(\x{100}{4})/;
3040 my($ord, $len) = (ord $1, length $1);
3041 print +($ok && $ord == 0x100 && $len == 4)
3042 ? "ok $test\n" : "not ok $test\t# $ok/$ord/$len\n";
3047 print "# [perl #15763]\n";
3050 chop $a; # but leaves the UTF-8 flag
3051 $a .= "y"; # 1 byte before "y"
3053 ok($a =~ /^\C/, 'match one \C on 1-byte UTF-8');
3054 ok($a =~ /^\C{1}/, 'match \C{1}');
3056 ok($a =~ /^\Cy/, 'match \Cy');
3057 ok($a =~ /^\C{1}y/, 'match \C{1}y');
3059 $a = "\x{100}y"; # 2 bytes before "y"
3061 ok($a =~ /^\C/, 'match one \C on 2-byte UTF-8');
3062 ok($a =~ /^\C{1}/, 'match \C{1}');
3063 ok($a =~ /^\C\C/, 'match two \C');
3064 ok($a =~ /^\C{2}/, 'match \C{2}');
3066 ok($a =~ /^\C\C\C/, 'match three \C on 2-byte UTF-8 and a byte');
3067 ok($a =~ /^\C{3}/, 'match \C{3}');
3069 ok($a =~ /^\C\Cy/, 'match two \C');
3070 ok($a =~ /^\C{2}y/, 'match \C{2}');
3072 ok($a !~ /^\C\C\Cy/, q{don't match three \Cy});
3073 ok($a !~ /^\C{2}\Cy/, q{don't match \C{3}y});
3075 $a = "\x{1000}y"; # 3 bytes before "y"
3077 ok($a =~ /^\C/, 'match one \C on three-byte UTF-8');
3078 ok($a =~ /^\C{1}/, 'match \C{1}');
3079 ok($a =~ /^\C\C/, 'match two \C');
3080 ok($a =~ /^\C{2}/, 'match \C{2}');
3081 ok($a =~ /^\C\C\C/, 'match three \C');
3082 ok($a =~ /^\C{3}/, 'match \C{3}');
3084 ok($a =~ /^\C\C\C\C/, 'match four \C on three-byte UTF-8 and a byte');
3085 ok($a =~ /^\C{4}/, 'match \C{4}');
3087 ok($a =~ /^\C\C\Cy/, 'match three \Cy');
3088 ok($a =~ /^\C{3}y/, 'match \C{3}y');
3090 ok($a !~ /^\C\C\C\C\y/, q{don't match four \Cy});
3091 ok($a !~ /^\C{4}y/, q{don't match \C{4}y});
3095 utf8::upgrade($_); chop $_; $\="\n";
3096 ok(/[^\s]+/, "m/[^\s]/ utf8");
3097 ok(/[^\d]+/, "m/[^\d]/ utf8");
3098 ok(($a = $_, $_ =~ s/[^\s]+/./g), "s/[^\s]/ utf8");
3099 ok(($a = $_, $a =~ s/[^\d]+/./g), "s/[^\s]/ utf8");
3101 ok("\x{100}" =~ /\x{100}/, "[perl #15397]");
3102 ok("\x{100}" =~ /(\x{100})/, "[perl #15397]");
3103 ok("\x{100}" =~ /(\x{100}){1}/, "[perl #15397]");
3104 ok("\x{100}\x{100}" =~ /(\x{100}){2}/, "[perl #15397]");
3105 ok("\x{100}\x{100}" =~ /(\x{100})(\x{100})/, "[perl #15397]");
3109 ok(!defined $1, "[perl #7471]");
3113 ok(!defined $1, "[perl #7471]");
3115 $pattern = "^(b+?|a){1,2}c";
3116 ok("bac" =~ /$pattern/ && $1 eq 'a', "[perl #3547]");
3117 ok("bbac" =~ /$pattern/ && $1 eq 'a', "[perl #3547]");
3118 ok("bbbac" =~ /$pattern/ && $1 eq 'a', "[perl #3547]");
3119 ok("bbbbac" =~ /$pattern/ && $1 eq 'a', "[perl #3547]");
3124 ok( $1 eq "\x{100}", '$1 is utf-8 [perl #18232]' );
3126 ok( $1 eq "\x{100}", '$1 is still utf-8' );
3127 ok( $1 ne "\xC4\x80", '$1 is not non-utf-8' );
3132 my $attr = 'Name-1' ;
3134 my $NormalChar = qr/[\p{IsDigit}\p{IsLower}\p{IsUpper}]/;
3135 my $NormalWord = qr/${NormalChar}+?/;
3136 my $PredNameHyphen = qr/^${NormalWord}(\-${NormalWord})*?$/;
3139 ok( $attr =~ $PredNameHyphen, "[perl #19767] original test" );
3145 ok ( "0" =~ /\p{N}+\z/, "[perl #19767] variant test" );
3152 $p++ if /(??{ $p })/
3154 ok ($p == 5, "[perl #20683] (??{ }) returns stale values");
3155 { package P; $a=1; sub TIESCALAR { bless[] } sub FETCH { $a++ } }
3160 ok ( $p == 5, "(??{ }) returns stale values");
3164 # Subject: Odd regexp behavior
3165 # From: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>
3166 # Date: Wed, 26 Feb 2003 16:53:12 +0000
3167 # Message-Id: <E18o4nw-0008Ly-00@wisbech.cl.cam.ac.uk>
3168 # To: perl-unicode@perl.org
3170 $x = "\x{2019}\nk"; $x =~ s/(\S)\n(\S)/$1 $2/sg;
3171 ok($x eq "\x{2019} k", "Markus Kuhn 2003-02-26");
3173 $x = "b\nk"; $x =~ s/(\S)\n(\S)/$1 $2/sg;
3174 ok($x eq "b k", "Markus Kuhn 2003-02-26");
3176 ok("\x{2019}" =~ /\S/, "Markus Kuhn 2003-02-26");
3181 ok('-1-3-5-' eq join('', split /((??{$i++}))/, '-1-3-5-'),
3182 "[perl #21411] (??{ .. }) corrupts split's stack");
3183 split /(?{'WOW'})/, 'abc';
3184 ok('a|b|c' eq join ('|', @_),
3185 "[perl #21411] (?{ .. }) version of the above");
3189 # XXX DAPM 13-Apr-06. Recursive split is still broken. It's only luck it
3190 # hasn't been crashing. Disable this test until it is fixed properly.
3191 # XXX also check what it returns rather than just doing ok(1,...)
3192 # split /(?{ split "" })/, "abc";
3193 ok(1,'cache_re & "(?{": it dumps core in 5.6.1 & 5.8.0');
3197 ok("\x{100}\n" =~ /\x{100}\n$/, "UTF8 length cache and fbm_compile");
3202 use overload q/""/ => sub { ${$_[0]}; };
3203 sub new { my ($c, $v) = @_; bless \$v, $c; }
3206 $_ = Str->new("a\x{100}/\x{100}b");
3207 ok(join(":", /\b(.)\x{100}/g) eq "a:/", "re_intuit_start and PL_bostr");
3211 $_ = "code: 'x' { '...' }\n"; study;
3212 my @x; push @x, $& while m/'[^\']*'/gx;
3213 ok(join(":", @x) eq "'x':'...'",
3214 "[perl #17757] Parse::RecDescent triggers infinite loop");
3218 my $re = qq/^([^X]*)X/;
3220 ok("\x{100}X" =~ /$re/, "S_cl_and ANYOF_UNICODE & ANYOF_INVERTED");
3225 ok( "a\nb" !~ /^b/, $_[0] );
3226 ok( "a\nb" =~ /^b/m, "$_[0] - with /m" );
3229 $_ = "x"; s/x/func "in subst"/e;
3230 $_ = "x"; s/x/func "in multiline subst"/em;
3231 #$_ = "x"; /x(?{func "in regexp"})/;
3232 #$_ = "x"; /x(?{func "in multiline regexp"})/m;
3237 ok("abcde" eq "$`", '# TODO #19049 - global match not setting $`');
3239 ok("123\x{100}" =~ /^.*1.*23\x{100}$/, 'uft8 + multiple floating substr');
3241 # LATIN SMALL/CAPITAL LETTER A WITH MACRON
3242 ok(" \x{101}" =~ qr/\x{100}/i,
3243 "<20030808193656.5109.1@llama.ni-s.u-net.com>");
3245 # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW
3246 ok(" \x{1E01}" =~ qr/\x{1E00}/i,
3247 "<20030808193656.5109.1@llama.ni-s.u-net.com>");
3249 # DESERET SMALL/CAPITAL LETTER LONG I
3250 ok(" \x{10428}" =~ qr/\x{10400}/i,
3251 "<20030808193656.5109.1@llama.ni-s.u-net.com>");
3253 # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X'
3254 ok(" \x{1E01}x" =~ qr/\x{1E00}X/i,
3255 "<20030808193656.5109.1@llama.ni-s.u-net.com>");
3258 # [perl #23769] Unicode regex broken on simple example
3259 # regrepeat() didn't handle UTF-8 EXACT case right.
3261 my $s = "\x{a0}\x{a0}\x{a0}\x{100}"; chop $s;
3263 ok($s =~ /\x{a0}/, "[perl #23769]");
3264 ok($s =~ /\x{a0}+/, "[perl #23769]");
3265 ok($s =~ /\x{a0}\x{a0}/, "[perl #23769]");
3267 ok("aaa\x{100}" =~ /(a+)/, "[perl #23769] easy invariant");
3268 ok($1 eq "aaa", "[perl #23769]");
3270 ok("\xa0\xa0\xa0\x{100}" =~ /(\xa0+)/, "[perl #23769] regrepeat invariant");
3271 ok($1 eq "\xa0\xa0\xa0", "[perl #23769]");
3273 ok("ababab\x{100} " =~ /((?:ab)+)/, "[perl #23769] hard invariant");
3274 ok($1 eq "ababab", "[perl #23769]");
3276 ok("\xa0\xa1\xa0\xa1\xa0\xa1\x{100}" =~ /((?:\xa0\xa1)+)/, "[perl #23769] hard variant");
3277 ok($1 eq "\xa0\xa1\xa0\xa1\xa0\xa1", "[perl #23769]");
3279 ok("aaa\x{100} " =~ /(a+?)/, "[perl #23769] easy invariant");
3280 ok($1 eq "a", "[perl #23769]");
3282 ok("\xa0\xa0\xa0\x{100} " =~ /(\xa0+?)/, "[perl #23769] regrepeat variant");
3283 ok($1 eq "\xa0", "[perl #23769]");
3285 ok("ababab\x{100} " =~ /((?:ab)+?)/, "[perl #23769] hard invariant");
3286 ok($1 eq "ab", "[perl #23769]");
3288 ok("\xa0\xa1\xa0\xa1\xa0\xa1\x{100}" =~ /((?:\xa0\xa1)+?)/, "[perl #23769] hard variant");
3289 ok($1 eq "\xa0\xa1", "[perl #23769]");
3291 ok("\xc4\xc4\xc4" !~ /(\x{100}+)/, "[perl #23769] don't match first byte of utf8 representation");
3292 ok("\xc4\xc4\xc4" !~ /(\x{100}+?)/, "[perl #23769] don't match first byte of utf8 representation");
3296 my $head = 'x' x $_;
3297 for my $tail ('\x{0061}', '\x{1234}') {
3299 eval qq{ "$head$tail" =~ /$head$tail/ },
3300 '\x{...} misparsed in regexp near 127 char EXACT limit'
3305 # perl #25269: panic: pp_match start/end pointers
3307 my($x, $y) = "bca" =~ /^(?=.*(a)).*(bc)/;
3309 }, 'captures can move backwards in string');
3311 # perl #27940: \cA not recognized in character classes
3312 ok("a\cAb" =~ /\cA/, '\cA in pattern');
3313 ok("a\cAb" =~ /[\cA]/, '\cA in character class');
3314 ok("a\cAb" =~ /[\cA-\cB]/, '\cA in character class range');
3315 ok("abc" =~ /[^\cA-\cB]/, '\cA in negated character class range');
3316 ok("a\cBb" =~ /[\cA-\cC]/, '\cB in character class range');
3317 ok("a\cCbc" =~ /[^\cA-\cB]/, '\cC in negated character class range');
3318 ok("a\cAb" =~ /(??{"\cA"})/, '\cA in ??{} pattern');
3319 ok("ab" !~ /a\cIb/x, '\cI in pattern');
3321 # perl #28532: optional zero-width match at end of string is ignored
3322 ok(("abc" =~ /^abc(\z)?/) && defined($1),
3323 'optional zero-width match at end of string');
3324 ok(("abc" =~ /^abc(\z)??/) && !defined($1),
3325 'optional zero-width match at end of string');
3331 "words"=~/(word|word|word)(?{push @got,$1})s$/;
3332 ok(@got==1,"TRIE optimation is working") or warn "# @got";
3334 "words"=~/(word|word|word)(?{push @got,$1})s$/i;
3335 ok(@got==1,"TRIEF optimisation is working") or warn "# @got";
3337 my @nums=map {int rand 1000} 1..100;
3338 my $re="(".(join "|",@nums).")";
3342 ok($_=~/$re/,"Trie nums");
3346 push @got,$1 while /$re/g;
3349 $count{$_}++ for @got;
3352 $ok=0 if --$count{$_}<0;
3354 ok($ok,"Trie min count matches");
3359 # LATIN SMALL/CAPITAL LETTER A WITH MACRON
3360 ok(("foba \x{101}foo" =~ qr/(foo|\x{100}foo|bar)/i) && $1 eq "\x{101}foo",
3361 "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH MACRON");
3363 # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW
3364 ok(("foba \x{1E01}foo" =~ qr/(foo|\x{1E00}foo|bar)/i) && $1 eq "\x{1E01}foo",
3365 "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH RING BELOW");
3367 # DESERET SMALL/CAPITAL LETTER LONG I
3368 ok(("foba \x{10428}foo" =~ qr/(foo|\x{10400}foo|bar)/i) && $1 eq "\x{10428}foo",
3369 "TRIEF + DESERET SMALL/CAPITAL LETTER LONG I");
3371 # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X'
3372 ok(("foba \x{1E01}xfoo" =~ qr/(foo|\x{1E00}Xfoo|bar)/i) && $1 eq "\x{1E01}xfoo",
3373 "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X'");
3377 use charnames ':full';
3379 $s="\N{LATIN SMALL LETTER SHARP S}";
3380 ok(("foba ba$s" =~ qr/(foo|Ba$s|bar)/i)
3382 "TRIEF + LATIN SMALL LETTER SHARP S =~ ss");
3383 ok(("foba ba$s" =~ qr/(Ba$s|foo|bar)/i)
3385 "TRIEF + LATIN SMALL LETTER SHARP S =~ ss");
3386 ok(("foba ba$s" =~ qr/(foo|bar|Ba$s)/i)
3388 "TRIEF + LATIN SMALL LETTER SHARP S =~ ss");
3390 ok(("foba ba$s" =~ qr/(foo|Bass|bar)/i)
3392 "TRIEF + LATIN SMALL LETTER SHARP S =~ ss");
3394 ok(("foba ba$s" =~ qr/(foo|BaSS|bar)/i)
3396 "TRIEF + LATIN SMALL LETTER SHARP S =~ SS");
3402 my @normal=qw(these are some normal words);
3403 my $psycho=join "|",@normal,map chr $_,255..20000;
3404 ok(('these'=~/($psycho)/) && $1 eq 'these','Pyscho');
3407 # [perl #36207] mixed utf8 / latin-1 and case folding
3410 my $utf8 = "\xe9\x{100}"; chop $utf8;
3411 my $latin1 = "\xe9";
3413 ok($utf8 =~ /\xe9/i, "utf8/latin");
3414 ok($utf8 =~ /$latin1/i, "utf8/latin runtime");
3415 ok($utf8 =~ /(abc|\xe9)/i, "utf8/latin trie");
3416 ok($utf8 =~ /(abc|$latin1)/i, "utf8/latin trie runtime");
3418 ok("\xe9" =~ /$utf8/i, "# TODO latin/utf8");
3419 ok("\xe9" =~ /(abc|$utf8)/i, "# latin/utf8 trie");
3420 ok($latin1 =~ /$utf8/i, "# TODO latin/utf8 runtime");
3421 ok($latin1 =~ /(abc|$utf8)/i, "# latin/utf8 trie runtime");
3424 # [perl #37038] Global regular matches generate invalid pointers
3432 "# assigning to original string should not corrupt match vars");
3442 my $aeek = bless {}, 'wooosh';
3443 eval {$aeek->gloople() =~ /(.)/g;};
3444 ok($@ eq "", "//g match against return value of sub") or print "# $@\n";
3451 eval {gloople() =~ /(.)/g;};
3452 ok($@ eq "", "# 26410 didn't affect sub calls for some reason")
3459 sub variable : lvalue { $var }
3462 my $o = bless [], "lv";
3464 eval { for (1..2) { $f .= $1 if $o->variable =~ /(.)/g } };
3465 ok($f eq "ab", "pos retained between calls # TODO") or print "# $@\n";
3470 sub variable : lvalue { $var }
3473 eval { for (1..2) { $f .= $1 if variable() =~ /(.)/g } };
3474 ok($f eq "ab", "pos retained between calls # TODO") or print "# $@\n";
3477 # [perl #37836] Simple Regex causes SEGV when run on specific data
3479 print "ok $test # Skip: in EBCDIC\n"; $test++;
3482 $_ = pack('U0C2', 0xa2, 0xf8); # ill-formed UTF-8
3484 eval { $ret = s/[\0]+//g };
3485 ok($ret == 0, "ill-formed UTF-8 doesn't match NUL in class");
3488 { # [perl #38293] chr(65535) should be allowed in regexes
3489 no warnings 'utf8'; # to allow non-characters
3494 ok($c eq "", "U+FFFF, parsed as atom");
3499 ok($c eq "", "U+FFFF backslashed, parsed as atom");
3503 ok($c eq "", "U+FFFF, parsed in class");
3508 ok($c eq "", "U+FFFF backslashed, parsed in class");
3511 $s =~ s/\x{ffff}//i;
3512 ok($s eq "AB", "U+FFFF, EXACTF");
3516 ok($s eq "\x{ffff}", "U+FFFF, BOUND");
3520 ok($s eq "\x{ffff}", "U+FFFF, NBOUND");
3521 } # non-characters end
3524 # https://rt.perl.org/rt3/Ticket/Display.html?id=39583
3526 # The printing characters
3527 my @chars = ("A".."Z");
3529 my $size = 32771 - 4;
3532 # create some random junk. Inefficient, but it works.
3533 for ($i = 0 ; $i < $size ; $i++) {
3534 $test .= $chars[int(rand(@chars))];
3537 $test .= ($delim x 4);
3540 if ($test =~ s/^(.*?)${delim}{4}//s) {
3544 ok($matched,'pattern matches');
3545 ok(length($test)==0,"Empty string");
3546 ok(defined($res) && length($res)==$size,"\$1 is correct size");
3549 # Keep the following test last -- it may crash perl
3551 ok(("a" x (2**15 - 10)) =~ /^()(a|bb)*$/, "Recursive stack cracker: #24274")
3552 or print "# Unexpected outcome: should pass or crash perl\n";