This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Revert patch #25993.
[perl5.git] / t / op / pat.t
CommitLineData
8d063cd8 1#!./perl
8d37f932
DD
2#
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.
8d063cd8 6
9133bbab 7$| = 1;
3568d838 8
740266bf 9print "1..1187\n";
8d37f932 10
e4d48cc9
GS
11BEGIN {
12 chdir 't' if -d 't';
20822f61 13 @INC = '../lib';
e4d48cc9 14}
ffbc6a93 15
8d37f932 16eval 'use Config'; # Defaults assumed if this fails
8d063cd8
LW
17
18$x = "abc\ndef\n";
19
20if ($x =~ /^abc/) {print "ok 1\n";} else {print "not ok 1\n";}
21if ($x !~ /^def/) {print "ok 2\n";} else {print "not ok 2\n";}
22
f02c194e
RGS
23# used to be a test for $*
24if ($x =~ /^def/m) {print "ok 3\n";} else {print "not ok 3\n";}
8d063cd8
LW
25
26$_ = '123';
27if (/^([0-9][0-9]*)/) {print "ok 4\n";} else {print "not ok 4\n";}
28
29if ($x =~ /^xxx/) {print "not ok 5\n";} else {print "ok 5\n";}
30if ($x !~ /^abc/) {print "not ok 6\n";} else {print "ok 6\n";}
31
32if ($x =~ /def/) {print "ok 7\n";} else {print "not ok 7\n";}
33if ($x !~ /def/) {print "not ok 8\n";} else {print "ok 8\n";}
34
35if ($x !~ /.def/) {print "ok 9\n";} else {print "not ok 9\n";}
36if ($x =~ /.def/) {print "not ok 10\n";} else {print "ok 10\n";}
37
38if ($x =~ /\ndef/) {print "ok 11\n";} else {print "not ok 11\n";}
39if ($x !~ /\ndef/) {print "not ok 12\n";} else {print "ok 12\n";}
40
41$_ = 'aaabbbccc';
42if (/(a*b*)(c*)/ && $1 eq 'aaabbb' && $2 eq 'ccc') {
43 print "ok 13\n";
44} else {
45 print "not ok 13\n";
46}
47if (/(a+b+c+)/ && $1 eq 'aaabbbccc') {
48 print "ok 14\n";
49} else {
50 print "not ok 14\n";
51}
52
53if (/a+b?c+/) {print "not ok 15\n";} else {print "ok 15\n";}
54
55$_ = 'aaabccc';
56if (/a+b?c+/) {print "ok 16\n";} else {print "not ok 16\n";}
57if (/a*b+c*/) {print "ok 17\n";} else {print "not ok 17\n";}
58
59$_ = 'aaaccc';
60if (/a*b?c*/) {print "ok 18\n";} else {print "not ok 18\n";}
61if (/a*b+c*/) {print "not ok 19\n";} else {print "ok 19\n";}
62
63$_ = 'abcdef';
64if (/bcd|xyz/) {print "ok 20\n";} else {print "not ok 20\n";}
65if (/xyz|bcd/) {print "ok 21\n";} else {print "not ok 21\n";}
66
67if (m|bc/*d|) {print "ok 22\n";} else {print "not ok 22\n";}
378cc40b
LW
68
69if (/^$_$/) {print "ok 23\n";} else {print "not ok 23\n";}
70
f02c194e
RGS
71# used to be a test for $*
72if ("ab\ncd\n" =~ /^cd/m) {print "ok 24\n";} else {print "not ok 24\n";}
378cc40b 73
cb55de95
JH
74$XXX{123} = 123;
75$XXX{234} = 234;
76$XXX{345} = 345;
77
78@XXX = ('ok 25','not ok 25', 'ok 26','not ok 26','not ok 27');
79while ($_ = shift(@XXX)) {
80 ?(.*)? && (print $1,"\n");
81 /not/ && reset;
82 /not ok 26/ && reset 'X';
83}
84
85while (($key,$val) = each(%XXX)) {
86 print "not ok 27\n";
87 exit;
88}
89
90print "ok 27\n";
378cc40b 91
378cc40b
LW
92'cde' =~ /[^ab]*/;
93'xyz' =~ //;
94if ($& eq 'xyz') {print "ok 28\n";} else {print "not ok 28\n";}
95
96$foo = '[^ab]*';
97'cde' =~ /$foo/;
98'xyz' =~ //;
99if ($& eq 'xyz') {print "ok 29\n";} else {print "not ok 29\n";}
100
101$foo = '[^ab]*';
102'cde' =~ /$foo/;
103'xyz' =~ /$null/;
104if ($& eq 'xyz') {print "ok 30\n";} else {print "not ok 30\n";}
a687059c
LW
105
106$_ = 'abcdefghi';
107/def/; # optimized up to cmd
108if ("$`:$&:$'" eq 'abc:def:ghi') {print "ok 31\n";} else {print "not ok 31\n";}
109
110/cde/ + 0; # optimized only to spat
111if ("$`:$&:$'" eq 'ab:cde:fghi') {print "ok 32\n";} else {print "not ok 32\n";}
112
113/[d][e][f]/; # not optimized
114if ("$`:$&:$'" eq 'abc:def:ghi') {print "ok 33\n";} else {print "not ok 33\n";}
115
116$_ = 'now is the {time for all} good men to come to.';
117/ {([^}]*)}/;
118if ($1 eq 'time for all') {print "ok 34\n";} else {print "not ok 34 $1\n";}
119
120$_ = 'xxx {3,4} yyy zzz';
121print /( {3,4})/ ? "ok 35\n" : "not ok 35\n";
122print $1 eq ' ' ? "ok 36\n" : "not ok 36\n";
123print /( {4,})/ ? "not ok 37\n" : "ok 37\n";
124print /( {2,3}.)/ ? "ok 38\n" : "not ok 38\n";
125print $1 eq ' y' ? "ok 39\n" : "not ok 39\n";
126print /(y{2,3}.)/ ? "ok 40\n" : "not ok 40\n";
127print $1 eq 'yyy ' ? "ok 41\n" : "not ok 41\n";
128print /x {3,4}/ ? "not ok 42\n" : "ok 42\n";
129print /^xxx {3,4}/ ? "not ok 43\n" : "ok 43\n";
352d5a3a
LW
130
131$_ = "now is the time for all good men to come to.";
132@words = /(\w+)/g;
133print join(':',@words) eq "now:is:the:time:for:all:good:men:to:come:to"
134 ? "ok 44\n"
135 : "not ok 44\n";
136
137@words = ();
138while (/\w+/g) {
139 push(@words, $&);
140}
141print join(':',@words) eq "now:is:the:time:for:all:good:men:to:come:to"
142 ? "ok 45\n"
143 : "not ok 45\n";
144
145@words = ();
71be2cbc 146pos = 0;
352d5a3a
LW
147while (/to/g) {
148 push(@words, $&);
149}
150print join(':',@words) eq "to:to"
151 ? "ok 46\n"
71be2cbc 152 : "not ok 46 `@words'\n";
352d5a3a 153
71be2cbc 154pos $_ = 0;
352d5a3a
LW
155@words = /to/g;
156print join(':',@words) eq "to:to"
157 ? "ok 47\n"
71be2cbc 158 : "not ok 47 `@words'\n";
352d5a3a
LW
159
160$_ = "abcdefghi";
161
162$pat1 = 'def';
163$pat2 = '^def';
164$pat3 = '.def.';
165$pat4 = 'abc';
166$pat5 = '^abc';
167$pat6 = 'abc$';
168$pat7 = 'ghi';
169$pat8 = '\w*ghi';
170$pat9 = 'ghi$';
171
172$t1=$t2=$t3=$t4=$t5=$t6=$t7=$t8=$t9=0;
173
174for $iter (1..5) {
175 $t1++ if /$pat1/o;
176 $t2++ if /$pat2/o;
177 $t3++ if /$pat3/o;
178 $t4++ if /$pat4/o;
179 $t5++ if /$pat5/o;
180 $t6++ if /$pat6/o;
181 $t7++ if /$pat7/o;
182 $t8++ if /$pat8/o;
183 $t9++ if /$pat9/o;
184}
185
186$x = "$t1$t2$t3$t4$t5$t6$t7$t8$t9";
187print $x eq '505550555' ? "ok 48\n" : "not ok 48 $x\n";
1462b684
LW
188
189$xyz = 'xyz';
190print "abc" =~ /^abc$|$xyz/ ? "ok 49\n" : "not ok 49\n";
191
192# perl 4.009 says "unmatched ()"
193eval '"abc" =~ /a(bc$)|$xyz/; $result = "$&:$1"';
194print $@ eq "" ? "ok 50\n" : "not ok 50\n";
195print $result eq "abc:bc" ? "ok 51\n" : "not ok 51\n";
a0d0e21e
LW
196
197
198$_="abcfooabcbar";
199$x=/abc/g;
200print $` eq "" ? "ok 52\n" : "not ok 52\n" if $x;
201$x=/abc/g;
202print $` eq "abcfoo" ? "ok 53\n" : "not ok 53\n" if $x;
203$x=/abc/g;
204print $x == 0 ? "ok 54\n" : "not ok 54\n";
71be2cbc 205pos = 0;
a0d0e21e
LW
206$x=/ABC/gi;
207print $` eq "" ? "ok 55\n" : "not ok 55\n" if $x;
208$x=/ABC/gi;
209print $` eq "abcfoo" ? "ok 56\n" : "not ok 56\n" if $x;
210$x=/ABC/gi;
211print $x == 0 ? "ok 57\n" : "not ok 57\n";
71be2cbc 212pos = 0;
a0d0e21e
LW
213$x=/abc/g;
214print $' eq "fooabcbar" ? "ok 58\n" : "not ok 58\n" if $x;
215$x=/abc/g;
216print $' eq "bar" ? "ok 59\n" : "not ok 59\n" if $x;
217$_ .= '';
218@x=/abc/g;
219print scalar @x == 2 ? "ok 60\n" : "not ok 60\n";
71be2cbc 220
221$_ = "abdc";
222pos $_ = 2;
c90c0ff4 223/\Gc/gc;
71be2cbc 224print "not " if (pos $_) != 2;
225print "ok 61\n";
c90c0ff4 226/\Gc/g;
227print "not " if defined pos $_;
228print "ok 62\n";
c277df42
IZ
229
230$out = 1;
231'abc' =~ m'a(?{ $out = 2 })b';
232print "not " if $out != 2;
233print "ok 63\n";
234
235$out = 1;
236'abc' =~ m'a(?{ $out = 3 })c';
237print "not " if $out != 1;
459f542a 238print "ok 64\n";
c277df42
IZ
239
240$_ = 'foobar1 bar2 foobar3 barfoobar5 foobar6';
241@out = /(?<!foo)bar./g;
242print "not " if "@out" ne 'bar2 barf';
243print "ok 65\n";
244
8d37f932
DD
245# Tests which depend on REG_INFTY
246$reg_infty = defined $Config{reg_infty} ? $Config{reg_infty} : 32767;
247$reg_infty_m = $reg_infty - 1; $reg_infty_p = $reg_infty + 1;
248
249# As well as failing if the pattern matches do unexpected things, the
250# next three tests will fail if you should have picked up a lower-than-
251# default value for $reg_infty from Config.pm, but have not.
252
253undef $@;
254print "not " if eval q(('aaa' =~ /(a{1,$reg_infty_m})/)[0] ne 'aaa') || $@;
255print "ok 66\n";
256
257undef $@;
258print "not " if eval q(('a' x $reg_infty_m) !~ /a{$reg_infty_m}/) || $@;
259print "ok 67\n";
260
261undef $@;
262print "not " if eval q(('a' x ($reg_infty_m - 1)) =~ /a{$reg_infty_m}/) || $@;
263print "ok 68\n";
264
265undef $@;
266eval "'aaa' =~ /a{1,$reg_infty}/";
9baa0206 267print "not " if $@ !~ m%^\QQuantifier in {,} bigger than%;
8d37f932
DD
268print "ok 69\n";
269
270eval "'aaa' =~ /a{1,$reg_infty_p}/";
271print "not "
9baa0206 272 if $@ !~ m%^\QQuantifier in {,} bigger than%;
8d37f932
DD
273print "ok 70\n";
274undef $@;
275
276# Poke a couple more parse failures
277
278$context = 'x' x 256;
279eval qq("${context}y" =~ /(?<=$context)y/);
9baa0206 280print "not " if $@ !~ m%^\QLookbehind longer than 255 not%;
8d37f932
DD
281print "ok 71\n";
282
b8c5462f 283# removed test
8d37f932
DD
284print "ok 72\n";
285
c277df42 286# Long Monsters
8d37f932 287$test = 73;
c277df42
IZ
288for $l (125, 140, 250, 270, 300000, 30) { # Ordered to free memory
289 $a = 'a' x $l;
290 print "# length=$l\nnot " unless "ba$a=" =~ /a$a=/;
291 print "ok $test\n";
292 $test++;
73d6d589 293
c277df42
IZ
294 print "not " if "b$a=" =~ /a$a=/;
295 print "ok $test\n";
296 $test++;
297}
298
299# 20000 nodes, each taking 3 words per string, and 1 per branch
300$long_constant_len = join '|', 12120 .. 32645;
301$long_var_len = join '|', 8120 .. 28645;
302%ans = ( 'ax13876y25677lbc' => 1,
303 'ax13876y25677mcb' => 0, # not b.
304 'ax13876y35677nbc' => 0, # Num too big
305 'ax13876y25677y21378obc' => 1,
306 'ax13876y25677y21378zbc' => 0, # Not followed by [k-o]
307 'ax13876y25677y21378y21378kbc' => 1,
308 'ax13876y25677y21378y21378kcb' => 0, # Not b.
309 'ax13876y25677y21378y21378y21378kbc' => 0, # 5 runs
310 );
311
312for ( keys %ans ) {
73d6d589 313 print "# const-len `$_' not => $ans{$_}\nnot "
c277df42
IZ
314 if $ans{$_} xor /a(?=([yx]($long_constant_len)){2,4}[k-o]).*b./o;
315 print "ok $test\n";
316 $test++;
73d6d589 317 print "# var-len `$_' not => $ans{$_}\nnot "
c277df42
IZ
318 if $ans{$_} xor /a(?=([yx]($long_var_len)){2,4}[k-o]).*b./o;
319 print "ok $test\n";
320 $test++;
321}
322
323$_ = " a (bla()) and x(y b((l)u((e))) and b(l(e)e)e";
324$expect = "(bla()) ((l)u((e))) (l(e)e)";
325
73d6d589 326sub matchit {
cc6b7395 327 m/
c277df42 328 (
73d6d589 329 \(
c277df42
IZ
330 (?{ $c = 1 }) # Initialize
331 (?:
332 (?(?{ $c == 0 }) # PREVIOUS iteration was OK, stop the loop
333 (?!
334 ) # Fail: will unwind one iteration back
73d6d589 335 )
c277df42
IZ
336 (?:
337 [^()]+ # Match a big chunk
338 (?=
339 [()]
340 ) # Do not try to match subchunks
341 |
73d6d589 342 \(
c277df42
IZ
343 (?{ ++$c })
344 |
73d6d589 345 \)
c277df42
IZ
346 (?{ --$c })
347 )
348 )+ # This may not match with different subblocks
349 )
350 (?(?{ $c != 0 })
351 (?!
352 ) # Fail
353 ) # Otherwise the chunk 1 may succeed with $c>0
cc6b7395 354 /xg;
c277df42
IZ
355}
356
0f5d15d6 357@ans = ();
c277df42
IZ
358push @ans, $res while $res = matchit;
359
360print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne "1 1 1";
361print "ok $test\n";
362$test++;
363
364@ans = matchit;
365
366print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne $expect;
367print "ok $test\n";
368$test++;
369
96776eda
GS
370print "not " unless "abc" =~ /^(??{"a"})b/;
371print "ok $test\n";
372$test++;
373
0f5d15d6 374my $matched;
14455d6c 375$matched = qr/\((?:(?>[^()]+)|(??{$matched}))*\)/;
0f5d15d6
IZ
376
377@ans = @ans1 = ();
378push(@ans, $res), push(@ans1, $&) while $res = m/$matched/g;
379
380print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne "1 1 1";
381print "ok $test\n";
382$test++;
383
384print "# ans1='@ans1'\n# expect='$expect'\nnot " if "@ans1" ne $expect;
385print "ok $test\n";
386$test++;
387
388@ans = m/$matched/g;
389
390print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne $expect;
391print "ok $test\n";
392$test++;
393
c277df42
IZ
394@ans = ('a/b' =~ m%(.*/)?(.*)%); # Stack may be bad
395print "not " if "@ans" ne 'a/ b';
396print "ok $test\n";
397$test++;
398
cc6b7395 399$code = '{$blah = 45}';
c277df42 400$blah = 12;
2cd61cdb
IZ
401eval { /(?$code)/ };
402print "not " unless $@ and $@ =~ /not allowed at runtime/ and $blah == 12;
e4d48cc9
GS
403print "ok $test\n";
404$test++;
405
2cd61cdb
IZ
406for $code ('{$blah = 45}','=xx') {
407 $blah = 12;
408 $res = eval { "xx" =~ /(?$code)/o };
409 if ($code eq '=xx') {
410 print "#'$@','$res','$blah'\nnot " unless not $@ and $res;
411 } else {
73d6d589 412 print "#'$@','$res','$blah'\nnot " unless $@ and $@ =~ /not allowed at runtime/ and $blah == 12;
2cd61cdb
IZ
413 }
414 print "ok $test\n";
415 $test++;
416}
417
e4d48cc9
GS
418$code = '{$blah = 45}';
419$blah = 12;
420eval "/(?$code)/";
cc6b7395
IZ
421print "not " if $blah != 45;
422print "ok $test\n";
423$test++;
424
425$blah = 12;
426/(?{$blah = 45})/;
c277df42
IZ
427print "not " if $blah != 45;
428print "ok $test\n";
429$test++;
430
74d6a13a
MB
431$x = 'banana';
432$x =~ /.a/g;
433print "not " unless pos($x) == 2;
434print "ok $test\n";
435$test++;
436
437$x =~ /.z/gc;
438print "not " unless pos($x) == 2;
439print "ok $test\n";
440$test++;
441
442sub f {
443 my $p = $_[0];
444 return $p;
445}
446
447$x =~ /.a/g;
448print "not " unless f(pos($x)) == 4;
449print "ok $test\n";
450$test++;
4599a1de 451
ce862d02
IZ
452$x = $^R = 67;
453'foot' =~ /foo(?{$x = 12; 75})[t]/;
454print "not " unless $^R eq '75';
455print "ok $test\n";
456$test++;
457
458$x = $^R = 67;
459'foot' =~ /foo(?{$x = 12; 75})[xy]/;
460print "not " unless $^R eq '67' and $x eq '12';
461print "ok $test\n";
462$test++;
463
464$x = $^R = 67;
465'foot' =~ /foo(?{ $^R + 12 })((?{ $x = 12; $^R + 17 })[xy])?/;
466print "not " unless $^R eq '79' and $x eq '12';
467print "ok $test\n";
468$test++;
469
8782bef2
GB
470print "not " unless qr/\b\v$/i eq '(?i-xsm:\bv$)';
471print "ok $test\n";
472$test++;
473
474print "not " unless qr/\b\v$/s eq '(?s-xim:\bv$)';
475print "ok $test\n";
476$test++;
477
478print "not " unless qr/\b\v$/m eq '(?m-xis:\bv$)';
479print "ok $test\n";
480$test++;
481
482print "not " unless qr/\b\v$/x eq '(?x-ism:\bv$)';
483print "ok $test\n";
484$test++;
485
486print "not " unless qr/\b\v$/xism eq '(?msix:\bv$)';
487print "ok $test\n";
488$test++;
489
490print "not " unless qr/\b\v$/ eq '(?-xism:\bv$)';
97197631
IZ
491print "ok $test\n";
492$test++;
493
7e5428c5
IZ
494$_ = 'xabcx';
495foreach $ans ('', 'c') {
496 /(?<=(?=a)..)((?=c)|.)/g;
02db2b7b 497 print "# \$1 ='$1'\n# \$ans='$ans'\nnot " unless $1 eq $ans;
7e5428c5
IZ
498 print "ok $test\n";
499 $test++;
500}
501
502$_ = 'a';
503foreach $ans ('', 'a', '') {
504 /^|a|$/g;
02db2b7b 505 print "# \$& ='$&'\n# \$ans='$ans'\nnot " unless $& eq $ans;
7e5428c5
IZ
506 print "ok $test\n";
507 $test++;
508}
509
09f25ae4 510sub prefixify {
73d6d589
NIS
511 my($v,$a,$b,$res) = @_;
512 $v =~ s/\Q$a\E/$b/;
513 print "not " unless $res eq $v;
09f25ae4
IZ
514 print "ok $test\n";
515 $test++;
516}
517prefixify('/a/b/lib/arch', "/a/b/lib", 'X/lib', 'X/lib/arch');
518prefixify('/a/b/man/arch', "/a/b/man", 'X/man', 'X/man/arch');
519
520$_ = 'var="foo"';
521/(\")/;
522print "not " unless $1 and /$1/;
523print "ok $test\n";
524$test++;
525
73d6d589 526$a=qr/(?{++$b})/;
2cd61cdb 527$b = 7;
73d6d589
NIS
528/$a$a/;
529print "not " unless $b eq '9';
2cd61cdb
IZ
530print "ok $test\n";
531$test++;
532
73d6d589
NIS
533$c="$a";
534/$a$a/;
535print "not " unless $b eq '11';
2cd61cdb
IZ
536print "ok $test\n";
537$test++;
538
539{
73d6d589
NIS
540 use re "eval";
541 /$a$c$a/;
542 print "not " unless $b eq '14';
2cd61cdb
IZ
543 print "ok $test\n";
544 $test++;
545
160cb429
JH
546 local $lex_a = 2;
547 my $lex_a = 43;
548 my $lex_b = 17;
549 my $lex_c = 27;
550 my $lex_res = ($lex_b =~ qr/$lex_b(?{ $lex_c = $lex_a++ })/);
551 print "not " unless $lex_res eq '1';
552 print "ok $test\n";
553 $test++;
554 print "not " unless $lex_a eq '44';
555 print "ok $test\n";
556 $test++;
557 print "not " unless $lex_c eq '43';
558 print "ok $test\n";
559 $test++;
560
561
73d6d589 562 no re "eval";
2cd61cdb 563 $match = eval { /$a$c$a/ };
73d6d589 564 print "not "
2cd61cdb
IZ
565 unless $b eq '14' and $@ =~ /Eval-group not allowed/ and not $match;
566 print "ok $test\n";
567 $test++;
568}
cbce877f
IZ
569
570{
160cb429
JH
571 local $lex_a = 2;
572 my $lex_a = 43;
573 my $lex_b = 17;
574 my $lex_c = 27;
575 my $lex_res = ($lex_b =~ qr/17(?{ $lex_c = $lex_a++ })/);
576 print "not " unless $lex_res eq '1';
577 print "ok $test\n";
578 $test++;
579 print "not " unless $lex_a eq '44';
580 print "ok $test\n";
581 $test++;
582 print "not " unless $lex_c eq '43';
583 print "ok $test\n";
584 $test++;
585}
586
587{
cbce877f
IZ
588 package aa;
589 $c = 2;
590 $::c = 3;
591 '' =~ /(?{ $c = 4 })/;
592 print "not " unless $c == 4;
593}
594print "ok $test\n";
595$test++;
596print "not " unless $c == 3;
597print "ok $test\n";
73d6d589
NIS
598$test++;
599
4599a1de
JH
600sub must_warn_pat {
601 my $warn_pat = shift;
602 return sub { print "not " unless $_[0] =~ /$warn_pat/ }
603}
604
605sub must_warn {
606 my ($warn_pat, $code) = @_;
9f1b1f2d
GS
607 local %SIG;
608 eval 'BEGIN { use warnings; $SIG{__WARN__} = $warn_pat };' . $code;
4599a1de
JH
609 print "ok $test\n";
610 $test++;
611}
612
613
614sub make_must_warn {
615 my $warn_pat = shift;
616 return sub { must_warn(must_warn_pat($warn_pat)) }
617}
618
619my $for_future = make_must_warn('reserved for future extensions');
620
621&$for_future('q(a:[b]:) =~ /[x[:foo:]]/');
9baa0206
HS
622
623#&$for_future('q(a=[b]=) =~ /[x[=foo=]]/');
624print "ok $test\n"; $test++; # now a fatal croak
625
626#&$for_future('q(a.[b].) =~ /[x[.foo.]]/');
627print "ok $test\n"; $test++; # now a fatal croak
f7e33566
GS
628
629# test if failure of patterns returns empty list
630$_ = 'aaa';
631@_ = /bbb/;
632print "not " if @_;
633print "ok $test\n";
634$test++;
635
636@_ = /bbb/g;
637print "not " if @_;
638print "ok $test\n";
639$test++;
640
641@_ = /(bbb)/;
642print "not " if @_;
643print "ok $test\n";
644$test++;
645
646@_ = /(bbb)/g;
647print "not " if @_;
648print "ok $test\n";
649$test++;
650
6cef1e77
IZ
651/a(?=.$)/;
652print "not " if $#+ != 0 or $#- != 0;
653print "ok $test\n";
654$test++;
655
656print "not " if $+[0] != 2 or $-[0] != 1;
657print "ok $test\n";
658$test++;
659
73d6d589 660print "not "
6cef1e77
IZ
661 if defined $+[1] or defined $-[1] or defined $+[2] or defined $-[2];
662print "ok $test\n";
663$test++;
664
665/a(a)(a)/;
666print "not " if $#+ != 2 or $#- != 2;
667print "ok $test\n";
668$test++;
669
670print "not " if $+[0] != 3 or $-[0] != 0;
671print "ok $test\n";
672$test++;
673
674print "not " if $+[1] != 2 or $-[1] != 1;
675print "ok $test\n";
676$test++;
677
678print "not " if $+[2] != 3 or $-[2] != 2;
679print "ok $test\n";
680$test++;
681
73d6d589 682print "not "
6cef1e77
IZ
683 if defined $+[3] or defined $-[3] or defined $+[4] or defined $-[4];
684print "ok $test\n";
685$test++;
686
687/.(a)(b)?(a)/;
688print "not " if $#+ != 3 or $#- != 3;
689print "ok $test\n";
690$test++;
691
692print "not " if $+[0] != 3 or $-[0] != 0;
693print "ok $test\n";
694$test++;
695
696print "not " if $+[1] != 2 or $-[1] != 1;
697print "ok $test\n";
698$test++;
699
700print "not " if $+[3] != 3 or $-[3] != 2;
701print "ok $test\n";
702$test++;
703
73d6d589 704print "not "
6cef1e77
IZ
705 if defined $+[2] or defined $-[2] or defined $+[4] or defined $-[4];
706print "ok $test\n";
707$test++;
708
709/.(a)/;
710print "not " if $#+ != 1 or $#- != 1;
711print "ok $test\n";
712$test++;
713
714print "not " if $+[0] != 2 or $-[0] != 0;
715print "ok $test\n";
716$test++;
717
718print "not " if $+[1] != 2 or $-[1] != 1;
719print "ok $test\n";
720$test++;
721
73d6d589 722print "not "
6cef1e77
IZ
723 if defined $+[2] or defined $-[2] or defined $+[3] or defined $-[3];
724print "ok $test\n";
725$test++;
726
03a27ae7 727eval { $+[0] = 13; };
73d6d589 728print "not "
03a27ae7
MG
729 if $@ !~ /^Modification of a read-only value attempted/;
730print "ok $test\n";
731$test++;
732
733eval { $-[0] = 13; };
73d6d589 734print "not "
03a27ae7
MG
735 if $@ !~ /^Modification of a read-only value attempted/;
736print "ok $test\n";
737$test++;
738
739eval { @+ = (7, 6, 5); };
73d6d589 740print "not "
03a27ae7
MG
741 if $@ !~ /^Modification of a read-only value attempted/;
742print "ok $test\n";
743$test++;
744
745eval { @- = qw(foo bar); };
73d6d589 746print "not "
03a27ae7
MG
747 if $@ !~ /^Modification of a read-only value attempted/;
748print "ok $test\n";
749$test++;
750
8f580fb8
IZ
751/.(a)(ba*)?/;
752print "#$#-..$#+\nnot " if $#+ != 2 or $#- != 1;
753print "ok $test\n";
754$test++;
755
ad94a511
IZ
756$_ = 'aaa';
757pos = 1;
758@a = /\Ga/g;
759print "not " unless "@a" eq "a a";
760print "ok $test\n";
761$test++;
762
22e551b9
IZ
763$str = 'abcde';
764pos $str = 2;
765
766print "not " if $str =~ /^\G/;
767print "ok $test\n";
768$test++;
769
770print "not " if $str =~ /^.\G/;
771print "ok $test\n";
772$test++;
773
774print "not " unless $str =~ /^..\G/;
775print "ok $test\n";
776$test++;
777
778print "not " if $str =~ /^...\G/;
779print "ok $test\n";
780$test++;
781
782print "not " unless $str =~ /.\G./ and $& eq 'bc';
783print "ok $test\n";
784$test++;
785
786print "not " unless $str =~ /\G../ and $& eq 'cd';
787print "ok $test\n";
788$test++;
789
9661b544
IZ
790undef $foo; undef $bar;
791print "#'$str','$foo','$bar'\nnot "
73d6d589 792 unless $str =~ /b(?{$foo = $_; $bar = pos})c/
9661b544
IZ
793 and $foo eq 'abcde' and $bar eq 2;
794print "ok $test\n";
795$test++;
796
797undef $foo; undef $bar;
798pos $str = undef;
799print "#'$str','$foo','$bar'\nnot "
73d6d589 800 unless $str =~ /b(?{$foo = $_; $bar = pos})c/g
9661b544
IZ
801 and $foo eq 'abcde' and $bar eq 2 and pos $str eq 3;
802print "ok $test\n";
803$test++;
804
805$_ = $str;
806
807undef $foo; undef $bar;
808print "#'$str','$foo','$bar'\nnot "
73d6d589 809 unless /b(?{$foo = $_; $bar = pos})c/
9661b544
IZ
810 and $foo eq 'abcde' and $bar eq 2;
811print "ok $test\n";
812$test++;
813
814undef $foo; undef $bar;
815print "#'$str','$foo','$bar'\nnot "
73d6d589 816 unless /b(?{$foo = $_; $bar = pos})c/g
9661b544
IZ
817 and $foo eq 'abcde' and $bar eq 2 and pos eq 3;
818print "ok $test\n";
819$test++;
820
821undef $foo; undef $bar;
822pos = undef;
8231 while /b(?{$foo = $_; $bar = pos})c/g;
824print "#'$str','$foo','$bar'\nnot "
825 unless $foo eq 'abcde' and $bar eq 2 and not defined pos;
826print "ok $test\n";
827$test++;
828
829undef $foo; undef $bar;
830$_ = 'abcde|abcde';
831print "#'$str','$foo','$bar','$_'\nnot "
73d6d589 832 unless s/b(?{$foo = $_; $bar = pos})c/x/g and $foo eq 'abcde|abcde'
9661b544
IZ
833 and $bar eq 8 and $_ eq 'axde|axde';
834print "ok $test\n";
835$test++;
836
5c5e4c24
IZ
837@res = ();
838# List context:
839$_ = 'abcde|abcde';
840@dummy = /([ace]).(?{push @res, $1,$2})([ce])(?{push @res, $1,$2})/g;
841@res = map {defined $_ ? "'$_'" : 'undef'} @res;
842$res = "@res";
843print "#'@res' '$_'\nnot "
844 unless "@res" eq "'a' undef 'a' 'c' 'e' undef 'a' undef 'a' 'c'";
845print "ok $test\n";
846$test++;
847
848@res = ();
849@dummy = /([ace]).(?{push @res, $`,$&,$'})([ce])(?{push @res, $`,$&,$'})/g;
850@res = map {defined $_ ? "'$_'" : 'undef'} @res;
851$res = "@res";
852print "#'@res' '$_'\nnot "
853 unless "@res" eq
854 "'' 'ab' 'cde|abcde' " .
855 "'' 'abc' 'de|abcde' " .
856 "'abcd' 'e|' 'abcde' " .
857 "'abcde|' 'ab' 'cde' " .
858 "'abcde|' 'abc' 'de'" ;
859print "ok $test\n";
860$test++;
861
b7a35066
IZ
862#Some more \G anchor checks
863$foo='aabbccddeeffgg';
864
865pos($foo)=1;
866
867$foo=~/.\G(..)/g;
868print "not " unless($1 eq 'ab');
869print "ok $test\n";
870$test++;
871
872pos($foo) += 1;
873$foo=~/.\G(..)/g;
874print "not " unless($1 eq 'cc');
875print "ok $test\n";
876$test++;
877
878pos($foo) += 1;
879$foo=~/.\G(..)/g;
880print "not " unless($1 eq 'de');
881print "ok $test\n";
882$test++;
883
0ef3e39e
HS
884print "not " unless $foo =~ /\Gef/g;
885print "ok $test\n";
886$test++;
887
b7a35066
IZ
888undef pos $foo;
889
890$foo=~/\G(..)/g;
891print "not " unless($1 eq 'aa');
892print "ok $test\n";
893$test++;
894
895$foo=~/\G(..)/g;
896print "not " unless($1 eq 'bb');
897print "ok $test\n";
898$test++;
899
900pos($foo)=5;
901$foo=~/\G(..)/g;
902print "not " unless($1 eq 'cd');
903print "ok $test\n";
904$test++;
905
73d6d589 906$_='123x123';
e60df1fa
IZ
907@res = /(\d*|x)/g;
908print "not " unless('123||x|123|' eq join '|', @res);
909print "ok $test\n";
910$test++;
911
9d080a66
GS
912# see if matching against temporaries (created via pp_helem()) is safe
913{ foo => "ok $test\n".$^X }->{foo} =~ /^(.*)\n/g;
914print "$1\n";
915$test++;
916
cf93c79d
IZ
917# See if $i work inside (?{}) in the presense of saved substrings and
918# changing $_
919@a = qw(foo bar);
920@b = ();
921s/(\w)(?{push @b, $1})/,$1,/g for @a;
922
923print "# \@b='@b', expect 'f o o b a r'\nnot " unless("@b" eq "f o o b a r");
924print "ok $test\n";
925$test++;
926
927print "not " unless("@a" eq ",f,,o,,o, ,b,,a,,r,");
928print "ok $test\n";
929$test++;
930
2c914db6 931$brackets = qr{
14455d6c 932 { (?> [^{}]+ | (??{ $brackets }) )* }
2c914db6
IZ
933 }x;
934
935"{{}" =~ $brackets;
936print "ok $test\n"; # Did we survive?
937$test++;
938
939"something { long { and } hairy" =~ $brackets;
940print "ok $test\n"; # Did we survive?
941$test++;
942
14455d6c 943"something { long { and } hairy" =~ m/((??{ $brackets }))/;
2c914db6
IZ
944print "not " unless $1 eq "{ and }";
945print "ok $test\n";
946$test++;
947
30944b6d
IZ
948$_ = "a-a\nxbb";
949pos=1;
950m/^-.*bb/mg and print "not ";
951print "ok $test\n";
952$test++;
30382c73
IZ
953
954$text = "aaXbXcc";
955pos($text)=0;
956$text =~ /\GXb*X/g and print 'not ';
957print "ok $test\n";
958$test++;
3cf5c195
IZ
959
960$text = "xA\n" x 500;
961$text =~ /^\s*A/m and print 'not ';
962print "ok $test\n";
963$test++;
d506a20d
IZ
964
965$text = "abc dbf";
966@res = ($text =~ /.*?(b).*?\b/g);
967"@res" eq 'b b' or print 'not ';
968print "ok $test\n";
969$test++;
970
9442cb0e 971@a = map chr,0..255;
aeaf5620
GS
972
973@b = grep(/\S/,@a);
974@c = grep(/[^\s]/,@a);
975print "not " if "@b" ne "@c";
9442cb0e
GS
976print "ok $test\n";
977$test++;
978
aeaf5620
GS
979@b = grep(/\S/,@a);
980@c = grep(/[\S]/,@a);
981print "not " if "@b" ne "@c";
9442cb0e
GS
982print "ok $test\n";
983$test++;
984
aeaf5620
GS
985@b = grep(/\s/,@a);
986@c = grep(/[^\S]/,@a);
987print "not " if "@b" ne "@c";
9442cb0e
GS
988print "ok $test\n";
989$test++;
990
aeaf5620
GS
991@b = grep(/\s/,@a);
992@c = grep(/[\s]/,@a);
993print "not " if "@b" ne "@c";
9442cb0e
GS
994print "ok $test\n";
995$test++;
996
aeaf5620
GS
997@b = grep(/\D/,@a);
998@c = grep(/[^\d]/,@a);
999print "not " if "@b" ne "@c";
9442cb0e
GS
1000print "ok $test\n";
1001$test++;
1002
aeaf5620
GS
1003@b = grep(/\D/,@a);
1004@c = grep(/[\D]/,@a);
1005print "not " if "@b" ne "@c";
9442cb0e
GS
1006print "ok $test\n";
1007$test++;
1008
aeaf5620
GS
1009@b = grep(/\d/,@a);
1010@c = grep(/[^\D]/,@a);
1011print "not " if "@b" ne "@c";
9442cb0e
GS
1012print "ok $test\n";
1013$test++;
1014
aeaf5620
GS
1015@b = grep(/\d/,@a);
1016@c = grep(/[\d]/,@a);
1017print "not " if "@b" ne "@c";
9442cb0e
GS
1018print "ok $test\n";
1019$test++;
1020
aeaf5620
GS
1021@b = grep(/\W/,@a);
1022@c = grep(/[^\w]/,@a);
1023print "not " if "@b" ne "@c";
9442cb0e
GS
1024print "ok $test\n";
1025$test++;
1026
aeaf5620
GS
1027@b = grep(/\W/,@a);
1028@c = grep(/[\W]/,@a);
1029print "not " if "@b" ne "@c";
9442cb0e
GS
1030print "ok $test\n";
1031$test++;
1032
aeaf5620
GS
1033@b = grep(/\w/,@a);
1034@c = grep(/[^\W]/,@a);
1035print "not " if "@b" ne "@c";
9442cb0e
GS
1036print "ok $test\n";
1037$test++;
1038
aeaf5620
GS
1039@b = grep(/\w/,@a);
1040@c = grep(/[\w]/,@a);
1041print "not " if "@b" ne "@c";
9442cb0e
GS
1042print "ok $test\n";
1043$test++;
1aeab75a
GS
1044
1045# see if backtracking optimization works correctly
1046"\n\n" =~ /\n $ \n/x or print "not ";
1047print "ok $test\n";
1048$test++;
1049
1050"\n\n" =~ /\n* $ \n/x or print "not ";
1051print "ok $test\n";
1052$test++;
1053
1054"\n\n" =~ /\n+ $ \n/x or print "not ";
1055print "ok $test\n";
1056$test++;
05b4157f
GS
1057
1058[] =~ /^ARRAY/ or print "# [] \nnot ";
1059print "ok $test\n";
1060$test++;
1061
1062eval << 'EOE';
1063{
1064 package S;
1065 use overload '""' => sub { 'Object S' };
1066 sub new { bless [] }
1067}
1068$a = 'S'->new;
1069EOE
1070
1071$a and $a =~ /^Object\sS/ or print "# '$a' \nnot ";
1072print "ok $test\n";
1073$test++;
815d35b9
MG
1074
1075# test result of match used as match (!)
1076'a1b' =~ ('xyz' =~ /y/) and $` eq 'a' or print "not ";
1077print "ok $test\n";
1078$test++;
1079
1080'a1b' =~ ('xyz' =~ /t/) and $` eq 'a' or print "not ";
1081print "ok $test\n";
1082$test++;
5e39e1e5
HS
1083
1084$w = 0;
1085{
1086 local $SIG{__WARN__} = sub { $w = 1 };
1087 local $^W = 1;
1088 $w = 1 if ("1\n" x 102) =~ /^\s*\n/m;
1089}
1090print $w ? "not " : "", "ok $test\n";
1091$test++;
aaa51d5e
JF
1092
1093my %space = ( spc => " ",
1094 tab => "\t",
1095 cr => "\r",
1096 lf => "\n",
1097 ff => "\f",
75369ccb
JH
1098# There's no \v but the vertical tabulator seems miraculously
1099# be 11 both in ASCII and EBCDIC.
aaa51d5e
JF
1100 vt => chr(11),
1101 false => "space" );
1102
1103my @space0 = sort grep { $space{$_} =~ /\s/ } keys %space;
1104my @space1 = sort grep { $space{$_} =~ /[[:space:]]/ } keys %space;
1105my @space2 = sort grep { $space{$_} =~ /[[:blank:]]/ } keys %space;
1106
1107print "not " unless "@space0" eq "cr ff lf spc tab";
3bec3564 1108print "ok $test # @space0\n";
aaa51d5e
JF
1109$test++;
1110
1111print "not " unless "@space1" eq "cr ff lf spc tab vt";
3bec3564 1112print "ok $test # @space1\n";
aaa51d5e
JF
1113$test++;
1114
1115print "not " unless "@space2" eq "spc tab";
3bec3564 1116print "ok $test # @space2\n";
aaa51d5e 1117$test++;
73d6d589 1118
a1933d95
HS
1119# bugid 20001021.005 - this caused a SEGV
1120print "not " unless undef =~ /^([^\/]*)(.*)$/;
1121print "ok $test\n";
1122$test++;
b91bb191
JH
1123
1124# bugid 20000731.001
1125
1126print "not " unless "A \x{263a} B z C" =~ /A . B (??{ "z" }) C/;
1127print "ok $test\n";
1128$test++;
1129
5ae032e5
JH
1130my $ordA = ord('A');
1131
3baa4c62
JH
1132$_ = "a\x{100}b";
1133if (/(.)(\C)(\C)(.)/) {
1134 print "ok 232\n";
1135 if ($1 eq "a") {
1136 print "ok 233\n";
1137 } else {
1138 print "not ok 233\n";
1139 }
5ae032e5
JH
1140 if ($ordA == 65) { # ASCII (or equivalent), should be UTF-8
1141 if ($2 eq "\xC4") {
1142 print "ok 234\n";
1143 } else {
1144 print "not ok 234\n";
1145 }
1146 if ($3 eq "\x80") {
1147 print "ok 235\n";
1148 } else {
1149 print "not ok 235\n";
1150 }
1151 } elsif ($ordA == 193) { # EBCDIC (or equivalent), should be UTF-EBCDIC
1152 if ($2 eq "\x8C") {
1153 print "ok 234\n";
1154 } else {
1155 print "not ok 234\n";
1156 }
1157 if ($3 eq "\x41") {
1158 print "ok 235\n";
1159 } else {
1160 print "not ok 235\n";
1161 }
3baa4c62 1162 } else {
5ae032e5
JH
1163 for (234..235) {
1164 print "not ok $_ # ord('A') == $ordA\n";
1165 }
3baa4c62
JH
1166 }
1167 if ($4 eq "b") {
1168 print "ok 236\n";
1169 } else {
1170 print "not ok 236\n";
1171 }
1172} else {
1173 for (232..236) {
1174 print "not ok $_\n";
1175 }
1176}
1177$_ = "\x{100}";
1178if (/(\C)/g) {
1179 print "ok 237\n";
73d6d589 1180 # currently \C are still tagged as UTF-8
5ae032e5
JH
1181 if ($ordA == 65) {
1182 if ($1 eq "\xC4") {
1183 print "ok 238\n";
1184 } else {
1185 print "not ok 238\n";
1186 }
1187 } elsif ($ordA == 193) {
1188 if ($1 eq "\x8C") {
1189 print "ok 238\n";
1190 } else {
1191 print "not ok 238\n";
1192 }
3baa4c62 1193 } else {
5ae032e5 1194 print "not ok 238 # ord('A') == $ordA\n";
3baa4c62
JH
1195 }
1196} else {
1197 for (237..238) {
1198 print "not ok $_\n";
1199 }
1200}
1201if (/(\C)/g) {
1202 print "ok 239\n";
73d6d589 1203 # currently \C are still tagged as UTF-8
5ae032e5
JH
1204 if ($ordA == 65) {
1205 if ($1 eq "\x80") {
1206 print "ok 240\n";
1207 } else {
1208 print "not ok 240\n";
1209 }
1210 } elsif ($ordA == 193) {
1211 if ($1 eq "\x41") {
1212 print "ok 240\n";
1213 } else {
1214 print "not ok 240\n";
1215 }
3baa4c62 1216 } else {
5ae032e5 1217 print "not ok 240 # ord('A') == $ordA\n";
3baa4c62
JH
1218 }
1219} else {
1220 for (239..240) {
1221 print "not ok $_\n";
1222 }
1223}
b485d051 1224
db615365
JP
1225{
1226 # japhy -- added 03/03/2001
1227 () = (my $str = "abc") =~ /(...)/;
1228 $str = "def";
1229 print "not " if $1 ne "abc";
fd291da9
JH
1230 print "ok 241\n";
1231}
1232
1233# The 242 and 243 go with the 244 and 245.
1234# The trick is that in EBCDIC the explicit numeric range should match
1235# (as also in non-EBCDIC) but the explicit alphabetic range should not match.
1236
1237if ("\x8e" =~ /[\x89-\x91]/) {
1238 print "ok 242\n";
1239} else {
1240 print "not ok 242\n";
1241}
1242
1243if ("\xce" =~ /[\xc9-\xd1]/) {
db615365 1244 print "ok 243\n";
fd291da9
JH
1245} else {
1246 print "not ok 243\n";
1247}
1248
1249# In most places these tests would succeed since \x8e does not
1250# in most character sets match 'i' or 'j' nor would \xce match
1251# 'I' or 'J', but strictly speaking these tests are here for
1252# the good of EBCDIC, so let's test these only there.
1253if (ord('i') == 0x89 && ord('J') == 0xd1) { # EBCDIC
1254 if ("\x8e" !~ /[i-j]/) {
1255 print "ok 244\n";
1256 } else {
1257 print "not ok 244\n";
1258 }
1259 if ("\xce" !~ /[I-J]/) {
1260 print "ok 245\n";
1261 } else {
1262 print "not ok 245\n";
1263 }
1264} else {
1265 for (244..245) {
60425c38 1266 print "ok $_ # Skip: only in EBCDIC\n";
fd291da9 1267 }
db615365 1268}
4765795a
JH
1269
1270print "not " unless "\x{ab}" =~ /\x{ab}/;
1271print "ok 246\n";
1272
1273print "not " unless "\x{abcd}" =~ /\x{abcd}/;
1274print "ok 247\n";
1275
1276{
f9969324
JH
1277 # bug id 20001008.001
1278
1279 my $test = 248;
1280 my @x = ("stra\337e 138","stra\337e 138");
1281 for (@x) {
1282 s/(\d+)\s*([\w\-]+)/$1 . uc $2/e;
1283 my($latin) = /^(.+)(?:\s+\d)/;
1284 print $latin eq "stra\337e" ? "ok $test\n" : # 248,249
1285 "#latin[$latin]\nnot ok $test\n";
1286 $test++;
1287 $latin =~ s/stra\337e/straße/; # \303\237 after the 2nd a
1288 use utf8; # needed for the raw UTF-8
1289 $latin =~ s!(s)tr(?:aß|s+e)!$1tr.!; # \303\237 after the a
4765795a
JH
1290 }
1291}
1292
1293{
1294 print "not " unless "ba\xd4c" =~ /([a\xd4]+)/ && $1 eq "a\xd4";
1295 print "ok 250\n";
1296
1297 print "not " unless "ba\xd4c" =~ /([a\xd4]+)/ && $1 eq "a\x{d4}";
1298 print "ok 251\n";
1299
1300 print "not " unless "ba\x{d4}c" =~ /([a\xd4]+)/ && $1 eq "a\x{d4}";
1301 print "ok 252\n";
1302
1303 print "not " unless "ba\x{d4}c" =~ /([a\xd4]+)/ && $1 eq "a\xd4";
1304 print "ok 253\n";
1305
1306 print "not " unless "ba\xd4c" =~ /([a\x{d4}]+)/ && $1 eq "a\xd4";
1307 print "ok 254\n";
1308
1309 print "not " unless "ba\xd4c" =~ /([a\x{d4}]+)/ && $1 eq "a\x{d4}";
1310 print "ok 255\n";
1311
1312 print "not " unless "ba\x{d4}c" =~ /([a\x{d4}]+)/ && $1 eq "a\x{d4}";
1313 print "ok 256\n";
1314
1315 print "not " unless "ba\x{d4}c" =~ /([a\x{d4}]+)/ && $1 eq "a\xd4";
1316 print "ok 257\n";
1317}
1318
1319{
1320 # the first half of 20001028.003
1321
1322 my $X = chr(1448);
1323 my ($Y) = $X =~ /(.*)/;
1324 print "not " unless $Y eq v1448 && length($Y) == 1;
1325 print "ok 258\n";
1326}
1327
1328{
1329 # 20001108.001
1330
1331 my $X = "Szab\x{f3},Bal\x{e1}zs";
1332 my $Y = $X;
1333 $Y =~ s/(B)/$1/ for 0..3;
1334 print "not " unless $Y eq $X && $X eq "Szab\x{f3},Bal\x{e1}zs";
1335 print "ok 259\n";
1336}
1337
1338{
1339 # the second half of 20001028.003
1340
3568d838 1341 my $X = '';
4765795a
JH
1342 $X =~ s/^/chr(1488)/e;
1343 print "not " unless length $X == 1 && ord($X) == 1488;
1344 print "ok 260\n";
1345}
1346
1347{
1348 # 20000517.001
1349
1350 my $x = "\x{100}A";
1351
1352 $x =~ s/A/B/;
1353
1354 print "not " unless $x eq "\x{100}B" && length($x) == 2;
1355 print "ok 261\n";
1356}
1357
1358{
1359 # bug id 20001230.002
1360
1361 print "not " unless "École" =~ /^\C\C(.)/ && $1 eq 'c';
1362 print "ok 262\n";
1363
1364 print "not " unless "École" =~ /^\C\C(c)/;
1365 print "ok 263\n";
1366}
1367
8dcad4ad 1368SKIP: {
4765795a
JH
1369 my $test = 264; # till 575
1370
bb99e5f1 1371 use charnames ":full";
4765795a
JH
1372
1373 # This is far from complete testing, there are dozens of character
1374 # classes in Unicode. The mixing of literals and \N{...} is
1375 # intentional so that in non-Latin-1 places we test the native
1376 # characters, not the Unicode code points.
1377
1378 my %s = (
1379 "a" => 'Ll',
1380 "\N{CYRILLIC SMALL LETTER A}" => 'Ll',
1381 "A" => 'Lu',
1382 "\N{GREEK CAPITAL LETTER ALPHA}" => 'Lu',
1383 "\N{HIRAGANA LETTER SMALL A}" => 'Lo',
1384 "\N{COMBINING GRAVE ACCENT}" => 'Mn',
1385 "0" => 'Nd',
1386 "\N{ARABIC-INDIC DIGIT ZERO}" => 'Nd',
1387 "_" => 'N',
1388 "!" => 'P',
1389 " " => 'Zs',
1390 "\0" => 'Cc',
1391 );
73d6d589 1392
3568d838
JH
1393 for my $char (map { s/^\S+ //; $_ }
1394 sort map { sprintf("%06x", ord($_))." $_" } keys %s) {
4765795a 1395 my $class = $s{$char};
3568d838
JH
1396 my $code = sprintf("%06x", ord($char));
1397 printf "#\n# 0x$code\n#\n";
4765795a
JH
1398 print "# IsAlpha\n";
1399 if ($class =~ /^[LM]/) {
1400 print "not " unless $char =~ /\p{IsAlpha}/;
1401 print "ok $test\n"; $test++;
1402 print "not " if $char =~ /\P{IsAlpha}/;
1403 print "ok $test\n"; $test++;
1404 } else {
1405 print "not " if $char =~ /\p{IsAlpha}/;
1406 print "ok $test\n"; $test++;
1407 print "not " unless $char =~ /\P{IsAlpha}/;
1408 print "ok $test\n"; $test++;
1409 }
1410 print "# IsAlnum\n";
1411 if ($class =~ /^[LMN]/ && $char ne "_") {
1412 print "not " unless $char =~ /\p{IsAlnum}/;
1413 print "ok $test\n"; $test++;
1414 print "not " if $char =~ /\P{IsAlnum}/;
1415 print "ok $test\n"; $test++;
1416 } else {
1417 print "not " if $char =~ /\p{IsAlnum}/;
1418 print "ok $test\n"; $test++;
1419 print "not " unless $char =~ /\P{IsAlnum}/;
1420 print "ok $test\n"; $test++;
1421 }
1422 print "# IsASCII\n";
24817587
JH
1423 if (ord("A") == 193) {
1424 print "ok $test # Skip: in EBCDIC\n"; $test++;
1425 print "ok $test # Skip: in EBCDIC\n"; $test++;
4765795a 1426 } else {
24817587
JH
1427 if ($code le '00007f') {
1428 print "not " unless $char =~ /\p{IsASCII}/;
1429 print "ok $test\n"; $test++;
1430 print "not " if $char =~ /\P{IsASCII}/;
1431 print "ok $test\n"; $test++;
1432 } else {
1433 print "not " if $char =~ /\p{IsASCII}/;
1434 print "ok $test\n"; $test++;
1435 print "not " unless $char =~ /\P{IsASCII}/;
1436 print "ok $test\n"; $test++;
1437 }
4765795a
JH
1438 }
1439 print "# IsCntrl\n";
1440 if ($class =~ /^C/) {
1441 print "not " unless $char =~ /\p{IsCntrl}/;
1442 print "ok $test\n"; $test++;
1443 print "not " if $char =~ /\P{IsCntrl}/;
1444 print "ok $test\n"; $test++;
1445 } else {
1446 print "not " if $char =~ /\p{IsCntrl}/;
1447 print "ok $test\n"; $test++;
1448 print "not " unless $char =~ /\P{IsCntrl}/;
1449 print "ok $test\n"; $test++;
1450 }
1451 print "# IsBlank\n";
1452 if ($class =~ /^Z[lp]/ || $char eq " ") {
1453 print "not " unless $char =~ /\p{IsBlank}/;
1454 print "ok $test\n"; $test++;
1455 print "not " if $char =~ /\P{IsBlank}/;
1456 print "ok $test\n"; $test++;
1457 } else {
1458 print "not " if $char =~ /\p{IsBlank}/;
1459 print "ok $test\n"; $test++;
1460 print "not " unless $char =~ /\P{IsBlank}/;
1461 print "ok $test\n"; $test++;
1462 }
1463 print "# IsDigit\n";
1464 if ($class =~ /^Nd$/) {
1465 print "not " unless $char =~ /\p{IsDigit}/;
1466 print "ok $test\n"; $test++;
1467 print "not " if $char =~ /\P{IsDigit}/;
1468 print "ok $test\n"; $test++;
1469 } else {
1470 print "not " if $char =~ /\p{IsDigit}/;
1471 print "ok $test\n"; $test++;
1472 print "not " unless $char =~ /\P{IsDigit}/;
1473 print "ok $test\n"; $test++;
1474 }
1475 print "# IsGraph\n";
1476 if ($class =~ /^([LMNPS])|Co/) {
1477 print "not " unless $char =~ /\p{IsGraph}/;
1478 print "ok $test\n"; $test++;
1479 print "not " if $char =~ /\P{IsGraph}/;
1480 print "ok $test\n"; $test++;
1481 } else {
1482 print "not " if $char =~ /\p{IsGraph}/;
1483 print "ok $test\n"; $test++;
1484 print "not " unless $char =~ /\P{IsGraph}/;
1485 print "ok $test\n"; $test++;
1486 }
1487 print "# IsLower\n";
1488 if ($class =~ /^Ll$/) {
1489 print "not " unless $char =~ /\p{IsLower}/;
1490 print "ok $test\n"; $test++;
1491 print "not " if $char =~ /\P{IsLower}/;
1492 print "ok $test\n"; $test++;
1493 } else {
1494 print "not " if $char =~ /\p{IsLower}/;
1495 print "ok $test\n"; $test++;
1496 print "not " unless $char =~ /\P{IsLower}/;
1497 print "ok $test\n"; $test++;
1498 }
1499 print "# IsPrint\n";
1500 if ($class =~ /^([LMNPS])|Co|Zs/) {
1501 print "not " unless $char =~ /\p{IsPrint}/;
1502 print "ok $test\n"; $test++;
1503 print "not " if $char =~ /\P{IsPrint}/;
1504 print "ok $test\n"; $test++;
1505 } else {
1506 print "not " if $char =~ /\p{IsPrint}/;
1507 print "ok $test\n"; $test++;
1508 print "not " unless $char =~ /\P{IsPrint}/;
1509 print "ok $test\n"; $test++;
1510 }
1511 print "# IsPunct\n";
1512 if ($class =~ /^P/ || $char eq "_") {
1513 print "not " unless $char =~ /\p{IsPunct}/;
1514 print "ok $test\n"; $test++;
1515 print "not " if $char =~ /\P{IsPunct}/;
1516 print "ok $test\n"; $test++;
1517 } else {
1518 print "not " if $char =~ /\p{IsPunct}/;
1519 print "ok $test\n"; $test++;
1520 print "not " unless $char =~ /\P{IsPunct}/;
1521 print "ok $test\n"; $test++;
1522 }
1523 print "# IsSpace\n";
1524 if ($class =~ /^Z/ || ($code =~ /^(0009|000A|000B|000C|000D)$/)) {
1525 print "not " unless $char =~ /\p{IsSpace}/;
1526 print "ok $test\n"; $test++;
1527 print "not " if $char =~ /\P{IsSpace}/;
1528 print "ok $test\n"; $test++;
1529 } else {
1530 print "not " if $char =~ /\p{IsSpace}/;
1531 print "ok $test\n"; $test++;
1532 print "not " unless $char =~ /\P{IsSpace}/;
1533 print "ok $test\n"; $test++;
1534 }
1535 print "# IsUpper\n";
1536 if ($class =~ /^L[ut]/) {
1537 print "not " unless $char =~ /\p{IsUpper}/;
1538 print "ok $test\n"; $test++;
1539 print "not " if $char =~ /\P{IsUpper}/;
1540 print "ok $test\n"; $test++;
1541 } else {
1542 print "not " if $char =~ /\p{IsUpper}/;
1543 print "ok $test\n"; $test++;
1544 print "not " unless $char =~ /\P{IsUpper}/;
1545 print "ok $test\n"; $test++;
1546 }
1547 print "# IsWord\n";
1548 if ($class =~ /^[LMN]/ || $char eq "_") {
1549 print "not " unless $char =~ /\p{IsWord}/;
1550 print "ok $test\n"; $test++;
1551 print "not " if $char =~ /\P{IsWord}/;
1552 print "ok $test\n"; $test++;
1553 } else {
1554 print "not " if $char =~ /\p{IsWord}/;
1555 print "ok $test\n"; $test++;
1556 print "not " unless $char =~ /\P{IsWord}/;
1557 print "ok $test\n"; $test++;
1558 }
1559 }
1560}
1561
1562{
1563 $_ = "abc\x{100}\x{200}\x{300}\x{380}\x{400}defg";
1564
1565 if (/(.\x{300})./) {
1566 print "ok 576\n";
1567
1568 print "not " unless $` eq "abc\x{100}" && length($`) == 4;
73d6d589 1569 print "ok 577\n";
4765795a
JH
1570
1571 print "not " unless $& eq "\x{200}\x{300}\x{380}" && length($&) == 3;
73d6d589 1572 print "ok 578\n";
4765795a
JH
1573
1574 print "not " unless $' eq "\x{400}defg" && length($') == 5;
73d6d589 1575 print "ok 579\n";
4765795a
JH
1576
1577 print "not " unless $1 eq "\x{200}\x{300}" && length($1) == 2;
73d6d589 1578 print "ok 580\n";
a8a2fe91
JH
1579 } else {
1580 for (576..580) { print "not ok $_\n" }
4765795a
JH
1581 }
1582}
8269fa76
JH
1583
1584{
1585 # bug id 20010306.008
1586
1587 $a = "a\x{1234}";
1588 # The original bug report had 'no utf8' here but that was irrelevant.
1589 $a =~ m/\w/; # used to core dump
1590
1591 print "ok 581\n";
1592}
b8ef571c
JH
1593
1594{
339e86bc
JH
1595 $test = 582;
1596
b8ef571c
JH
1597 # bugid 20010410.006
1598 for my $rx (
1599 '/(.*?)\{(.*?)\}/csg',
1600 '/(.*?)\{(.*?)\}/cg',
1601 '/(.*?)\{(.*?)\}/sg',
1602 '/(.*?)\{(.*?)\}/g',
1603 '/(.+?)\{(.+?)\}/csg',
1604 )
1605 {
1606 my($input, $i);
1607
1608 $i = 0;
1609 $input = "a{b}c{d}";
1610 eval <<EOT;
1611 while (eval \$input =~ $rx) {
1612 print "# \\\$1 = '\$1' \\\$2 = '\$2'\n";
1613 ++\$i;
1614 }
1615EOT
1616 print "not " unless $i == 2;
1617 print "ok " . $test++ . "\n";
1618 }
1619}
209a9bc1
JH
1620
1621{
1622 # from Robin Houston
1623
b851fbc1 1624 my $x = "\x{10FFFD}";
209a9bc1 1625 $x =~ s/(.)/$1/g;
b851fbc1 1626 print "not " unless ord($x) == 0x10FFFD && length($x) == 1;
209a9bc1
JH
1627 print "ok 587\n";
1628}
3568d838
JH
1629
1630{
1631 my $x = "\x7f";
1632
1633 print "not " if $x =~ /[\x80-\xff]/;
1634 print "ok 588\n";
1635
1636 print "not " if $x =~ /[\x80-\x{100}]/;
1637 print "ok 589\n";
1638
1639 print "not " if $x =~ /[\x{100}]/;
1640 print "ok 590\n";
1641
1642 print "not " if $x =~ /\p{InLatin1Supplement}/;
1643 print "ok 591\n";
1644
1645 print "not " unless $x =~ /\P{InLatin1Supplement}/;
1646 print "ok 592\n";
1647
1648 print "not " if $x =~ /\p{InLatinExtendedA}/;
1649 print "ok 593\n";
1650
1651 print "not " unless $x =~ /\P{InLatinExtendedA}/;
1652 print "ok 594\n";
1653}
1654
1655{
1656 my $x = "\x80";
1657
1658 print "not " unless $x =~ /[\x80-\xff]/;
1659 print "ok 595\n";
1660
1661 print "not " unless $x =~ /[\x80-\x{100}]/;
1662 print "ok 596\n";
1663
1664 print "not " if $x =~ /[\x{100}]/;
1665 print "ok 597\n";
1666
1667 print "not " unless $x =~ /\p{InLatin1Supplement}/;
1668 print "ok 598\n";
1669
1670 print "not " if $x =~ /\P{InLatin1Supplement}/;
1671 print "ok 599\n";
1672
1673 print "not " if $x =~ /\p{InLatinExtendedA}/;
1674 print "ok 600\n";
1675
1676 print "not " unless $x =~ /\P{InLatinExtendedA}/;
1677 print "ok 601\n";
1678}
1679
1680{
1681 my $x = "\xff";
1682
1683 print "not " unless $x =~ /[\x80-\xff]/;
1684 print "ok 602\n";
1685
1686 print "not " unless $x =~ /[\x80-\x{100}]/;
1687 print "ok 603\n";
1688
1689 print "not " if $x =~ /[\x{100}]/;
1690 print "ok 604\n";
1691
cc8040a1
TD
1692 # the next two tests must be ignored on EBCDIC
1693 print "not " unless $x =~ /\p{InLatin1Supplement}/ or ord("A") == 193;
3568d838
JH
1694 print "ok 605\n";
1695
cc8040a1 1696 print "not " if $x =~ /\P{InLatin1Supplement}/ and ord("A") != 193;
3568d838
JH
1697 print "ok 606\n";
1698
1699 print "not " if $x =~ /\p{InLatinExtendedA}/;
1700 print "ok 607\n";
1701
1702 print "not " unless $x =~ /\P{InLatinExtendedA}/;
1703 print "ok 608\n";
1704}
1705
1706{
1707 my $x = "\x{100}";
1708
1709 print "not " if $x =~ /[\x80-\xff]/;
1710 print "ok 609\n";
1711
1712 print "not " unless $x =~ /[\x80-\x{100}]/;
1713 print "ok 610\n";
1714
1715 print "not " unless $x =~ /[\x{100}]/;
1716 print "ok 611\n";
1717
1718 print "not " if $x =~ /\p{InLatin1Supplement}/;
1719 print "ok 612\n";
1720
1721 print "not " unless $x =~ /\P{InLatin1Supplement}/;
1722 print "ok 613\n";
1723
1724 print "not " unless $x =~ /\p{InLatinExtendedA}/;
1725 print "ok 614\n";
1726
1727 print "not " if $x =~ /\P{InLatinExtendedA}/;
1728 print "ok 615\n";
1729}
1730
9d1d55b5
JP
1731{
1732 # from japhy
1733 my $w;
1734 use warnings;
1735 local $SIG{__WARN__} = sub { $w .= shift };
1736
1737 $w = "";
1738 eval 'qr/(?c)/';
1739 print "not " if $w !~ /^Useless \(\?c\)/;
1740 print "ok 616\n";
1741
1742 $w = "";
1743 eval 'qr/(?-c)/';
1744 print "not " if $w !~ /^Useless \(\?-c\)/;
1745 print "ok 617\n";
1746
1747 $w = "";
1748 eval 'qr/(?g)/';
1749 print "not " if $w !~ /^Useless \(\?g\)/;
1750 print "ok 618\n";
1751
1752 $w = "";
1753 eval 'qr/(?-g)/';
1754 print "not " if $w !~ /^Useless \(\?-g\)/;
1755 print "ok 619\n";
1756
1757 $w = "";
1758 eval 'qr/(?o)/';
1759 print "not " if $w !~ /^Useless \(\?o\)/;
1760 print "ok 620\n";
1761
1762 $w = "";
1763 eval 'qr/(?-o)/';
1764 print "not " if $w !~ /^Useless \(\?-o\)/;
1765 print "ok 621\n";
1766
1767 # now test multi-error regexes
1768
1769 $w = "";
1770 eval 'qr/(?g-o)/';
1771 print "not " if $w !~ /^Useless \(\?g\).*\nUseless \(\?-o\)/;
1772 print "ok 622\n";
1773
1774 $w = "";
1775 eval 'qr/(?g-c)/';
1776 print "not " if $w !~ /^Useless \(\?g\).*\nUseless \(\?-c\)/;
1777 print "ok 623\n";
1778
1779 $w = "";
1780 eval 'qr/(?o-cg)/'; # (?c) means (?g) error won't be thrown
1781 print "not " if $w !~ /^Useless \(\?o\).*\nUseless \(\?-c\)/;
1782 print "ok 624\n";
1783
1784 $w = "";
1785 eval 'qr/(?ogc)/';
1786 print "not " if $w !~ /^Useless \(\?o\).*\nUseless \(\?g\).*\nUseless \(\?c\)/;
1787 print "ok 625\n";
1788}
a72deede
JH
1789
1790# More Unicode "class" tests
1791
1792{
1793 use charnames ':full';
1794
1795 print "not " unless "\N{LATIN CAPITAL LETTER A}" =~ /\p{InBasicLatin}/;
1796 print "ok 626\n";
1797
1798 print "not " unless "\N{LATIN CAPITAL LETTER A WITH GRAVE}" =~ /\p{InLatin1Supplement}/;
1799 print "ok 627\n";
1800
1801 print "not " unless "\N{LATIN CAPITAL LETTER A WITH MACRON}" =~ /\p{InLatinExtendedA}/;
1802 print "ok 628\n";
1803
1804 print "not " unless "\N{LATIN SMALL LETTER B WITH STROKE}" =~ /\p{InLatinExtendedB}/;
1805 print "ok 629\n";
1806
1807 print "not " unless "\N{KATAKANA LETTER SMALL A}" =~ /\p{InKatakana}/;
1808 print "ok 630\n";
1809}
1810
6002328a
JH
1811$_ = "foo";
1812
1813eval <<"EOT"; die if $@;
1814 /f
1815 o\r
1816 o
1817 \$
1818 /x && print "ok 631\n";
1819EOT
1820
1821eval <<"EOT"; die if $@;
1822 /f
1823 o
1824 o
1825 \$\r
1826 /x && print "ok 632\n";
1827EOT
1828
569b5e07
AB
1829#test /o feature
1830sub test_o { $_[0] =~/$_[1]/o; return $1}
1831if(test_o('abc','(.)..') eq 'a') {
395ddfe6 1832 print "ok 633\n";
569b5e07 1833} else {
395ddfe6 1834 print "not ok 633\n";
569b5e07
AB
1835}
1836if(test_o('abc','..(.)') eq 'a') {
395ddfe6 1837 print "ok 634\n";
569b5e07 1838} else {
395ddfe6 1839 print "not ok 634\n";
569b5e07
AB
1840}
1841
f79b3095
JH
1842# 635..639: ID 20010619.003 (only the space character is
1843# supposed to be [:print:], not the whole isprint()).
1844
1845print "not " if "\n" =~ /[[:print:]]/;
1846print "ok 635\n";
1847
1848print "not " if "\t" =~ /[[:print:]]/;
1849print "ok 636\n";
1850
e857312d 1851# Amazingly vertical tabulator is the same in ASCII and EBCDIC.
f79b3095
JH
1852print "not " if "\014" =~ /[[:print:]]/;
1853print "ok 637\n";
1854
1855print "not " if "\r" =~ /[[:print:]]/;
1856print "ok 638\n";
1857
1858print "not " unless " " =~ /[[:print:]]/;
1859print "ok 639\n";
1860
a01268b5
JH
1861##
1862## Test basic $^N usage outside of a regex
1863##
1864$x = "abcdef";
1865$T="ok 640\n";if ($x =~ /cde/ and not defined $^N) {print $T} else {print "not $T"};
1866$T="ok 641\n";if ($x =~ /(cde)/ and $^N eq "cde") {print $T} else {print "not $T"};
1867$T="ok 642\n";if ($x =~ /(c)(d)(e)/ and $^N eq "e") {print $T} else {print "not $T"};
1868$T="ok 643\n";if ($x =~ /(c(d)e)/ and $^N eq "cde") {print $T} else {print "not $T"};
1869$T="ok 644\n";if ($x =~ /(foo)|(c(d)e)/ and $^N eq "cde") {print $T} else {print "not $T"};
1870$T="ok 645\n";if ($x =~ /(c(d)e)|(foo)/ and $^N eq "cde") {print $T} else {print "not $T"};
1871$T="ok 646\n";if ($x =~ /(c(d)e)|(abc)/ and $^N eq "abc") {print $T} else {print "not $T"};
1872$T="ok 647\n";if ($x =~ /(c(d)e)|(abc)x/ and $^N eq "cde") {print $T} else {print "not $T"};
1873$T="ok 648\n";if ($x =~ /(c(d)e)(abc)?/ and $^N eq "cde") {print $T} else {print "not $T"};
1874$T="ok 649\n";if ($x =~ /(?:c(d)e)/ and $^N eq "d" ) {print $T} else {print "not $T"};
1875$T="ok 650\n";if ($x =~ /(?:c(d)e)(?:f)/ and $^N eq "d" ) {print $T} else {print "not $T"};
1876$T="ok 651\n";if ($x =~ /(?:([abc])|([def]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
1877$T="ok 652\n";if ($x =~ /(?:([ace])|([bdf]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
1878$T="ok 653\n";if ($x =~ /(([ace])|([bd]))*/ and $^N eq "e" ){print $T} else {print "not $T"};
1879{
1880 $T="ok 654\n";if($x =~ /(([ace])|([bdf]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
1881}
1882## test to see if $^N is automatically localized -- it should now
1883## have the value set in test 653
1884$T="ok 655\n";if ($^N eq "e" ){print $T} else {print "not $T"};
1885
1886##
1887## Now test inside (?{...})
1888##
1889$T="ok 656\n";if ($x =~ /a([abc])(?{$y=$^N})c/ and $y eq "b" ){print $T} else {print "not $T"};
1890$T="ok 657\n";if ($x =~ /a([abc]+)(?{$y=$^N})d/ and $y eq "bc"){print $T} else {print "not $T"};
1891$T="ok 658\n";if ($x =~ /a([abcdefg]+)(?{$y=$^N})d/ and $y eq "bc"){print $T} else {print "not $T"};
1892$T="ok 659\n";if ($x =~ /(a([abcdefg]+)(?{$y=$^N})d)(?{$z=$^N})e/ and $y eq "bc" and $z eq "abcd")
1893 {print $T} else {print "not $T"};
1894$T="ok 660\n";if ($x =~ /(a([abcdefg]+)(?{$y=$^N})de)(?{$z=$^N})/ and $y eq "bc" and $z eq "abcde")
1895 {print $T} else {print "not $T"};
2796c109
JH
1896
1897# Test the Unicode script classes
1898
cf25bb62 1899print "not " unless chr(0x100) =~ /\p{IsLatin}/; # outside Latin-1
2796c109
JH
1900print "ok 661\n";
1901
cf25bb62 1902print "not " unless chr(0x212b) =~ /\p{IsLatin}/; # Angstrom sign, very outside
2796c109
JH
1903print "ok 662\n";
1904
cf25bb62 1905print "not " unless chr(0x5d0) =~ /\p{IsHebrew}/; # inside InHebrew
2796c109
JH
1906print "ok 663\n";
1907
cf25bb62 1908print "not " unless chr(0xfb4f) =~ /\p{IsHebrew}/; # outside InHebrew
2796c109
JH
1909print "ok 664\n";
1910
7be0dac3
NC
1911# # singleton (not in a range, this test must be ignored on EBCDIC)
1912# print "not " unless chr(0xb5) =~ /\p{IsGreek}/ or ord("A") == 193;
1913# print "ok 665\n";
1914print "ok 665 # 0xb5 moved from Greek to Common with Unicode 4.0.1\n";
5f9563ea 1915
cf25bb62 1916print "not " unless chr(0x37a) =~ /\p{IsGreek}/; # singleton
5f9563ea
JH
1917print "ok 666\n";
1918
cf25bb62 1919print "not " unless chr(0x386) =~ /\p{IsGreek}/; # singleton
5f9563ea
JH
1920print "ok 667\n";
1921
cf25bb62 1922print "not " unless chr(0x387) =~ /\P{IsGreek}/; # not there
5f9563ea
JH
1923print "ok 668\n";
1924
cf25bb62 1925print "not " unless chr(0x388) =~ /\p{IsGreek}/; # range
5f9563ea
JH
1926print "ok 669\n";
1927
cf25bb62 1928print "not " unless chr(0x38a) =~ /\p{IsGreek}/; # range
5f9563ea
JH
1929print "ok 670\n";
1930
cf25bb62 1931print "not " unless chr(0x38b) =~ /\P{IsGreek}/; # not there
5f9563ea
JH
1932print "ok 671\n";
1933
cf25bb62 1934print "not " unless chr(0x38c) =~ /\p{IsGreek}/; # singleton
5f9563ea
JH
1935print "ok 672\n";
1936
550cec39 1937if (ord("A") == 65) {
7be5a6cf
JF
1938##
1939## Test [:cntrl:]...
1940##
1941## Should probably put in tests for all the POSIX stuff, but not sure how to
1942## guarantee a specific locale......
1943##
550cec39
JH
1944 $AllBytes = join('', map { chr($_) } 0..255);
1945 ($x = $AllBytes) =~ s/[[:cntrl:]]//g;
1946 if ($x ne join('', map { chr($_) } 0x20..0x7E, 0x80..0xFF)) {
1947 print "not ";
1948 }
1949 print "ok 673\n";
1950
1951 ($x = $AllBytes) =~ s/[^[:cntrl:]]//g;
1952 if ($x ne join('', map { chr($_) } 0..0x1F, 0x7F)) { print "not " }
1953 print "ok 674\n";
1954} else {
1955 print "ok $_ # Skip: EBCDIC\n" for 673..674;
1956}
f33976b4
DB
1957
1958# With /s modifier UTF8 chars were interpreted as bytes
1959{
1960 my $a = "Hello \x{263A} World";
1961
1962 my @a = ($a =~ /./gs);
1963
1964 print "not " unless $#a == 12;
1965 print "ok 675\n";
1966}
cce850e4
JH
1967
1968@a = ("foo\nbar" =~ /./g);
1969print "ok 676\n" if @a == 6 && "@a" eq "f o o b a r";
1970
1971@a = ("foo\nbar" =~ /./gs);
1972print "ok 677\n" if @a == 7 && "@a" eq "f o o \n b a r";
1973
1974@a = ("foo\nbar" =~ /\C/g);
1975print "ok 678\n" if @a == 7 && "@a" eq "f o o \n b a r";
1976
1977@a = ("foo\nbar" =~ /\C/gs);
1978print "ok 679\n" if @a == 7 && "@a" eq "f o o \n b a r";
1979
1980@a = ("foo\n\x{100}bar" =~ /./g);
1981print "ok 680\n" if @a == 7 && "@a" eq "f o o \x{100} b a r";
1982
1983@a = ("foo\n\x{100}bar" =~ /./gs);
1984print "ok 681\n" if @a == 8 && "@a" eq "f o o \n \x{100} b a r";
1985
1986($a, $b) = map { chr } ord('A') == 65 ? (0xc4, 0x80) : (0x8c, 0x41);
1987
1988@a = ("foo\n\x{100}bar" =~ /\C/g);
1989print "ok 682\n" if @a == 9 && "@a" eq "f o o \n $a $b b a r";
1990
1991@a = ("foo\n\x{100}bar" =~ /\C/gs);
1992print "ok 683\n" if @a == 9 && "@a" eq "f o o \n $a $b b a r";
1993
0af80b60
HS
1994{
1995 # [ID 20010814.004] pos() doesn't work when using =~m// in list context
1996 $_ = "ababacadaea";
1997 $a = join ":", /b./gc;
1998 $b = join ":", /a./gc;
1999 $c = pos;
2000 print "$a $b $c" eq 'ba:ba ad:ae 10' ? "ok 684\n" : "not ok 684\t# $a $b $c\n";
2001}
d9f424b2
JH
2002
2003{
75685a94
JH
2004 # [ID 20010407.006] matching utf8 return values from functions does not work
2005
d9f424b2
JH
2006 package ID_20010407_006;
2007
2008 sub x {
2009 "a\x{1234}";
2010 }
2011
2012 my $x = x;
2013 my $y;
2014
2015 $x =~ /(..)/; $y = $1;
2016 print "not " unless length($y) == 2 && $y eq $x;
75685a94 2017 print "ok 685\n";
d9f424b2
JH
2018
2019 x =~ /(..)/; $y = $1;
2020 print "not " unless length($y) == 2 && $y eq $x;
2021 print "ok 686\n";
2022}
a4c04bdc 2023
e2d8ce26 2024
a4c04bdc
NC
2025my $test = 687;
2026
2027# Force scalar context on the patern match
2028sub ok ($$) {
2029 my($ok, $name) = @_;
2030
2031 printf "%sok %d - %s\n", ($ok ? "" : "not "), $test, $name;
2032
2033 printf "# Failed test at line %d\n", (caller)[2] unless $ok;
2034
2035 $test++;
2036 return $ok;
2037}
2038
2039{
2040 # Check that \x## works. 5.6.1 and 5.005_03 fail some of these.
2041 $x = "\x4e" . "E";
2042 ok ($x =~ /^\x4EE$/, "Check only 2 bytes of hex are matched.");
2043
2044 $x = "\x4e" . "i";
2045 ok ($x =~ /^\x4Ei$/, "Check that invalid hex digit stops it (2)");
2046
2047 $x = "\x4" . "j";
2048 ok ($x =~ /^\x4j$/, "Check that invalid hex digit stops it (1)");
2049
2050 $x = "\x0" . "k";
2051 ok ($x =~ /^\xk$/, "Check that invalid hex digit stops it (0)");
2052
2053 $x = "\x0" . "x";
2054 ok ($x =~ /^\xx$/, "\\xx isn't to be treated as \\0");
2055
2056 $x = "\x0" . "xa";
2057 ok ($x =~ /^\xxa$/, "\\xxa isn't to be treated as \\xa");
2058
2059 $x = "\x9" . "_b";
2060 ok ($x =~ /^\x9_b$/, "\\x9_b isn't to be treated as \\x9b");
2061
2062 print "# and now again in [] ranges\n";
2063
2064 $x = "\x4e" . "E";
2065 ok ($x =~ /^[\x4EE]{2}$/, "Check only 2 bytes of hex are matched.");
2066
2067 $x = "\x4e" . "i";
2068 ok ($x =~ /^[\x4Ei]{2}$/, "Check that invalid hex digit stops it (2)");
2069
2070 $x = "\x4" . "j";
2071 ok ($x =~ /^[\x4j]{2}$/, "Check that invalid hex digit stops it (1)");
2072
2073 $x = "\x0" . "k";
2074 ok ($x =~ /^[\xk]{2}$/, "Check that invalid hex digit stops it (0)");
2075
2076 $x = "\x0" . "x";
2077 ok ($x =~ /^[\xx]{2}$/, "\\xx isn't to be treated as \\0");
2078
2079 $x = "\x0" . "xa";
2080 ok ($x =~ /^[\xxa]{3}$/, "\\xxa isn't to be treated as \\xa");
2081
2082 $x = "\x9" . "_b";
2083 ok ($x =~ /^[\x9_b]{3}$/, "\\x9_b isn't to be treated as \\x9b");
2084
2085}
2086
2087{
2088 # Check that \x{##} works. 5.6.1 fails quite a few of these.
2089
2090 $x = "\x9b";
2091 ok ($x =~ /^\x{9_b}$/, "\\x{9_b} is to be treated as \\x9b");
2092
2093 $x = "\x9b" . "y";
2094 ok ($x =~ /^\x{9_b}y$/, "\\x{9_b} is to be treated as \\x9b (again)");
2095
2096 $x = "\x9b" . "y";
2097 ok ($x =~ /^\x{9b_}y$/, "\\x{9b_} is to be treated as \\x9b");
2098
2099 $x = "\x9b" . "y";
2100 ok ($x =~ /^\x{9_bq}y$/, "\\x{9_bc} is to be treated as \\x9b");
2101
2102 $x = "\x0" . "y";
2103 ok ($x =~ /^\x{x9b}y$/, "\\x{x9b} is to be treated as \\x0");
2104
2105 $x = "\x0" . "y";
2106 ok ($x =~ /^\x{0x9b}y$/, "\\x{0x9b} is to be treated as \\x0");
2107
2108 $x = "\x9b" . "y";
2109 ok ($x =~ /^\x{09b}y$/, "\\x{09b} is to be treated as \\x9b");
2110
2111 print "# and now again in [] ranges\n";
2112
2113 $x = "\x9b";
2114 ok ($x =~ /^[\x{9_b}]$/, "\\x{9_b} is to be treated as \\x9b");
2115
2116 $x = "\x9b" . "y";
2117 ok ($x =~ /^[\x{9_b}y]{2}$/, "\\x{9_b} is to be treated as \\x9b (again)");
2118
2119 $x = "\x9b" . "y";
2120 ok ($x =~ /^[\x{9b_}y]{2}$/, "\\x{9b_} is to be treated as \\x9b");
2121
2122 $x = "\x9b" . "y";
2123 ok ($x =~ /^[\x{9_bq}y]{2}$/, "\\x{9_bc} is to be treated as \\x9b");
2124
2125 $x = "\x0" . "y";
2126 ok ($x =~ /^[\x{x9b}y]{2}$/, "\\x{x9b} is to be treated as \\x0");
2127
2128 $x = "\x0" . "y";
2129 ok ($x =~ /^[\x{0x9b}y]{2}$/, "\\x{0x9b} is to be treated as \\x0");
2130
2131 $x = "\x9b" . "y";
2132 ok ($x =~ /^[\x{09b}y]{2}$/, "\\x{09b} is to be treated as \\x9b");
2133}
e2d8ce26
JP
2134
2135{
d9efae67
JH
2136 # high bit bug -- japhy
2137 my $x = "ab\200d";
2138 $x =~ /.*?\200/ or print "not ";
2139 print "ok 715\n";
e2d8ce26
JP
2140}
2141
4193bef7
JH
2142print "# some Unicode properties\n";
2143
d9efae67 2144{
4193bef7 2145 # Dashes, underbars, case.
d9efae67
JH
2146 print "not " unless "\x80" =~ /\p{in-latin1_SUPPLEMENT}/;
2147 print "ok 716\n";
ab13f0c7 2148
4193bef7 2149 # Complement, leading and trailing whitespace.
ab13f0c7
JH
2150 print "not " unless "\x80" =~ /\P{ ^ In Latin 1 Supplement }/;
2151 print "ok 717\n";
4193bef7 2152
f173cd49
JH
2153 # No ^In, dashes, case, dash, any intervening (word-break) whitespace.
2154 # (well, newlines don't work...)
2155 print "not " unless "\x80" =~ /\p{latin-1 supplement}/;
4193bef7
JH
2156 print "ok 718\n";
2157}
2158
2159{
2160 print "not " unless "a" =~ /\pL/;
2161 print "ok 719\n";
4193bef7 2162
4193bef7
JH
2163 print "not " unless "a" =~ /\p{IsLl}/;
2164 print "ok 720\n";
4193bef7 2165
c87b7cc2 2166 print "not " if "a" =~ /\p{IsLu}/;
4193bef7 2167 print "ok 721\n";
4193bef7 2168
61247495 2169 print "not " unless "a" =~ /\p{Ll}/;
4193bef7 2170 print "ok 722\n";
c87b7cc2 2171
61247495 2172 print "not " if "a" =~ /\p{Lu}/;
c87b7cc2
JH
2173 print "ok 723\n";
2174
61247495 2175 print "not " unless "A" =~ /\pL/;
c87b7cc2
JH
2176 print "ok 724\n";
2177
61247495 2178 print "not " unless "A" =~ /\p{IsLu}/;
c87b7cc2
JH
2179 print "ok 725\n";
2180
61247495 2181 print "not " if "A" =~ /\p{IsLl}/;
c87b7cc2
JH
2182 print "ok 726\n";
2183
61247495 2184 print "not " unless "A" =~ /\p{Lu}/;
c87b7cc2
JH
2185 print "ok 727\n";
2186
61247495 2187 print "not " if "A" =~ /\p{Ll}/;
c87b7cc2
JH
2188 print "ok 728\n";
2189
61247495 2190 print "not " if "a" =~ /\PL/;
c87b7cc2
JH
2191 print "ok 729\n";
2192
61247495 2193 print "not " if "a" =~ /\P{IsLl}/;
c87b7cc2 2194 print "ok 730\n";
61247495
JH
2195
2196 print "not " unless "a" =~ /\P{IsLu}/;
2197 print "ok 731\n";
2198
2199 print "not " if "a" =~ /\P{Ll}/;
2200 print "ok 732\n";
2201
2202 print "not " unless "a" =~ /\P{Lu}/;
2203 print "ok 733\n";
2204
2205 print "not " if "A" =~ /\PL/;
2206 print "ok 734\n";
2207
2208 print "not " if "A" =~ /\P{IsLu}/;
2209 print "ok 735\n";
2210
2211 print "not " unless "A" =~ /\P{IsLl}/;
2212 print "ok 736\n";
2213
2214 print "not " if "A" =~ /\P{Lu}/;
2215 print "ok 737\n";
2216
2217 print "not " unless "A" =~ /\P{Ll}/;
2218 print "ok 738\n";
2219
4193bef7 2220}
9b4e380a
JH
2221
2222{
2223 print "not " if "a" =~ /\p{Common}/;
2224 print "ok 739\n";
2225
2226 print "not " unless "1" =~ /\p{Common}/;
2227 print "ok 740\n";
2228}
2229
2230{
2231 print "not " if "a" =~ /\p{Inherited}/;
2232 print "ok 741\n";
2233
2234 print "not " unless "\x{300}" =~ /\p{Inherited}/;
2235 print "ok 742\n";
2236}
2237
2238{
12ac2576
JP
2239 # L& and LC are the same
2240 print "not " unless "a" =~ /\p{LC}/ and "a" =~ /\p{L&}/;
9b4e380a
JH
2241 print "ok 743\n";
2242
12ac2576 2243 print "not " if "1" =~ /\p{LC}/ or "1" =~ /\p{L&}/;
9b4e380a
JH
2244 print "ok 744\n";
2245}
d73e5302
JH
2246
2247{
904bc114 2248 print "not " unless "a" =~ /\p{Lowercase Letter}/;
d73e5302
JH
2249 print "ok 745\n";
2250
904bc114 2251 print "not " if "A" =~ /\p{lowercaseletter}/;
d73e5302
JH
2252 print "ok 746\n";
2253}
2254
2255{
cf25bb62 2256 print "not " unless "\x{AC00}" =~ /\p{HangulSyllables}/;
d73e5302
JH
2257 print "ok 747\n";
2258}
71d929cb
JH
2259
2260{
701a277b
JH
2261 # Script=, Block=, Category=
2262
71d929cb
JH
2263 print "not " unless "\x{0100}" =~ /\p{Script=Latin}/;
2264 print "ok 748\n";
2265
2266 print "not " unless "\x{0100}" =~ /\p{Block=LatinExtendedA}/;
2267 print "ok 749\n";
2268
2269 print "not " unless "\x{0100}" =~ /\p{Category=UppercaseLetter}/;
2270 print "ok 750\n";
2271}
2272
ef54fa25 2273{
701a277b
JH
2274 print "# the basic character classes and Unicode \n";
2275
ef54fa25
JH
2276 # 0100;LATIN CAPITAL LETTER A WITH MACRON;Lu;0;L;0041 0304;;;;N;LATIN CAPITAL LETTER A MACRON;;;0101;
2277 print "not " unless "\x{0100}" =~ /\w/;
2278 print "ok 751\n";
2279
2280 # 0660;ARABIC-INDIC DIGIT ZERO;Nd;0;AN;;0;0;0;N;;;;;
2281 print "not " unless "\x{0660}" =~ /\d/;
2282 print "ok 752\n";
2283
2284 # 1680;OGHAM SPACE MARK;Zs;0;WS;;;;;N;;;;;
2285 print "not " unless "\x{1680}" =~ /\s/;
2286 print "ok 753\n";
2287}
701a277b
JH
2288
2289{
2290 print "# folding matches and Unicode\n";
2291
2292 print "not " unless "a\x{100}" =~ /A/i;
2293 print "ok 754\n";
2294
27e28630 2295 print "not " unless "A\x{100}" =~ /a/i;
701a277b
JH
2296 print "ok 755\n";
2297
2298 print "not " unless "a\x{100}" =~ /a/i;
2299 print "ok 756\n";
2300
2301 print "not " unless "A\x{100}" =~ /A/i;
2302 print "ok 757\n";
bc517b45
JH
2303
2304 print "not " unless "\x{101}a" =~ /\x{100}/i;
2305 print "ok 758\n";
2306
2307 print "not " unless "\x{100}a" =~ /\x{100}/i;
2308 print "ok 759\n";
2309
2310 print "not " unless "\x{101}a" =~ /\x{101}/i;
2311 print "ok 760\n";
2312
2313 print "not " unless "\x{100}a" =~ /\x{101}/i;
2314 print "ok 761\n";
2315
2316 print "not " unless "a\x{100}" =~ /A\x{100}/i;
2317 print "ok 762\n";
2318
27e28630 2319 print "not " unless "A\x{100}" =~ /a\x{100}/i;
bc517b45
JH
2320 print "ok 763\n";
2321
2322 print "not " unless "a\x{100}" =~ /a\x{100}/i;
2323 print "ok 764\n";
2324
2325 print "not " unless "A\x{100}" =~ /A\x{100}/i;
2326 print "ok 765\n";
2327
2328 print "not " unless "a\x{100}" =~ /[A]/i;
2329 print "ok 766\n";
2330
27e28630 2331 print "not " unless "A\x{100}" =~ /[a]/i;
bc517b45
JH
2332 print "ok 767\n";
2333
2334 print "not " unless "a\x{100}" =~ /[a]/i;
2335 print "ok 768\n";
2336
2337 print "not " unless "A\x{100}" =~ /[A]/i;
2338 print "ok 769\n";
2339
2340 print "not " unless "\x{101}a" =~ /[\x{100}]/i;
2341 print "ok 770\n";
2342
2343 print "not " unless "\x{100}a" =~ /[\x{100}]/i;
2344 print "ok 771\n";
2345
2346 print "not " unless "\x{101}a" =~ /[\x{101}]/i;
2347 print "ok 772\n";
2348
2349 print "not " unless "\x{100}a" =~ /[\x{101}]/i;
2350 print "ok 773\n";
2351
701a277b 2352}
a5961de5
JH
2353
2354{
2355 use charnames ':full';
2356
2357 print "# LATIN LETTER A WITH GRAVE\n";
2358 my $lower = "\N{LATIN SMALL LETTER A WITH GRAVE}";
2359 my $UPPER = "\N{LATIN CAPITAL LETTER A WITH GRAVE}";
2360
bc517b45
JH
2361 print $lower =~ m/$UPPER/i ? "ok 774\n" : "not ok 774\n";
2362 print $UPPER =~ m/$lower/i ? "ok 775\n" : "not ok 775\n";
2363 print $lower =~ m/[$UPPER]/i ? "ok 776\n" : "not ok 776\n";
2364 print $UPPER =~ m/[$lower]/i ? "ok 777\n" : "not ok 777\n";
a5961de5
JH
2365
2366 print "# GREEK LETTER ALPHA WITH VRACHY\n";
2367
2368 $lower = "\N{GREEK CAPITAL LETTER ALPHA WITH VRACHY}";
2369 $UPPER = "\N{GREEK SMALL LETTER ALPHA WITH VRACHY}";
2370
bc517b45
JH
2371 print $lower =~ m/$UPPER/i ? "ok 778\n" : "not ok 778\n";
2372 print $UPPER =~ m/$lower/i ? "ok 779\n" : "not ok 779\n";
2373 print $lower =~ m/[$UPPER]/i ? "ok 780\n" : "not ok 780\n";
2374 print $UPPER =~ m/[$lower]/i ? "ok 781\n" : "not ok 781\n";
a5961de5
JH
2375
2376 print "# LATIN LETTER Y WITH DIAERESIS\n";
2377
2378 $lower = "\N{LATIN CAPITAL LETTER Y WITH DIAERESIS}";
2379 $UPPER = "\N{LATIN SMALL LETTER Y WITH DIAERESIS}";
bc517b45
JH
2380 print $lower =~ m/$UPPER/i ? "ok 782\n" : "not ok 782\n";
2381 print $UPPER =~ m/$lower/i ? "ok 783\n" : "not ok 783\n";
2382 print $lower =~ m/[$UPPER]/i ? "ok 784\n" : "not ok 784\n";
2383 print $UPPER =~ m/[$lower]/i ? "ok 785\n" : "not ok 785\n";
a5961de5 2384}
55da9344
JH
2385
2386{
2387 use warnings;
2388 use charnames ':full';
2389
2390 print "# GREEK CAPITAL LETTER SIGMA vs COMBINING GREEK PERISPOMENI\n";
2391
2392 my $SIGMA = "\N{GREEK CAPITAL LETTER SIGMA}";
d07ddd77 2393 my $char = "\N{COMBINING GREEK PERISPOMENI}";
55da9344 2394
bc517b45
JH
2395 # Before #13843 this was failing by matching falsely.
2396 print "_:$char:_" =~ m/_:$SIGMA:_/i ? "not ok 786\n" : "ok 786\n";
55da9344 2397}
b7c83a7e
JH
2398
2399{
2400 print "# \\X\n";
2401
2402 use charnames ':full';
2403
eb08e2da
JH
2404 print "a!" =~ /^(\X)!/ && $1 eq "a" ?
2405 "ok 787\n" : "not ok 787 # $1\n";
2406 print "\xDF!" =~ /^(\X)!/ && $1 eq "\xDF" ?
2407 "ok 788\n" : "not ok 788 # $1\n";
2408 print "\x{100}!" =~ /^(\X)!/ && $1 eq "\x{100}" ?
2409 "ok 789\n" : "not ok 789 # $1\n";
2410 print "\x{100}\x{300}!" =~ /^(\X)!/ && $1 eq "\x{100}\x{300}" ?
2411 "ok 790\n" : "not ok 790 # $1\n";
2412 print "\N{LATIN CAPITAL LETTER E}!" =~ /^(\X)!/ &&
2413 $1 eq "\N{LATIN CAPITAL LETTER E}" ?
2414 "ok 791\n" : "not ok 791 # $1\n";
2415 print "\N{LATIN CAPITAL LETTER E}\N{COMBINING GRAVE ACCENT}!" =~
2416 /^(\X)!/ &&
2417 $1 eq "\N{LATIN CAPITAL LETTER E}\N{COMBINING GRAVE ACCENT}" ?
2418 "ok 792\n" : "not ok 792 # $1\n";
b7c83a7e 2419}
112bedeb
JH
2420
2421{
2422 print "#\\C and \\X\n";
2423
2424 print "!abc!" =~ /a\Cc/ ? "ok 793\n" : "not ok 793\n";
2425 print "!abc!" =~ /a\Xc/ ? "ok 794\n" : "not ok 794\n";
2426}
09091399
JH
2427
2428{
2429 print "# FINAL SIGMA\n";
2430
2431 my $SIGMA = "\x{03A3}"; # CAPITAL
2432 my $Sigma = "\x{03C2}"; # SMALL FINAL
2433 my $sigma = "\x{03C3}"; # SMALL
2434
2435 print $SIGMA =~ /$SIGMA/i ? "ok 795\n" : "not ok 795\n";
2436 print $SIGMA =~ /$Sigma/i ? "ok 796\n" : "not ok 796\n";
2437 print $SIGMA =~ /$sigma/i ? "ok 797\n" : "not ok 797\n";
2438
2439 print $Sigma =~ /$SIGMA/i ? "ok 798\n" : "not ok 798\n";
2440 print $Sigma =~ /$Sigma/i ? "ok 799\n" : "not ok 799\n";
2441 print $Sigma =~ /$sigma/i ? "ok 800\n" : "not ok 800\n";
2442
2443 print $sigma =~ /$SIGMA/i ? "ok 801\n" : "not ok 801\n";
2444 print $sigma =~ /$Sigma/i ? "ok 802\n" : "not ok 802\n";
2445 print $sigma =~ /$sigma/i ? "ok 803\n" : "not ok 803\n";
2446
2447 print $SIGMA =~ /[$SIGMA]/i ? "ok 804\n" : "not ok 804\n";
2448 print $SIGMA =~ /[$Sigma]/i ? "ok 805\n" : "not ok 805\n";
2449 print $SIGMA =~ /[$sigma]/i ? "ok 806\n" : "not ok 806\n";
2450
2451 print $Sigma =~ /[$SIGMA]/i ? "ok 807\n" : "not ok 807\n";
2452 print $Sigma =~ /[$Sigma]/i ? "ok 808\n" : "not ok 808\n";
2453 print $Sigma =~ /[$sigma]/i ? "ok 809\n" : "not ok 809\n";
2454
2455 print $sigma =~ /[$SIGMA]/i ? "ok 810\n" : "not ok 810\n";
2456 print $sigma =~ /[$Sigma]/i ? "ok 811\n" : "not ok 811\n";
2457 print $sigma =~ /[$sigma]/i ? "ok 812\n" : "not ok 812\n";
2458}
e036fef9
JF
2459
2460{
2461 print "# parlez-vous?\n";
2462
2463 use charnames ':full';
2464
2465 print "fran\N{LATIN SMALL LETTER C}ais" =~
2466 /fran.ais/ &&
2467 $& eq "francais" ?
2468 "ok 813\n" : "not ok 813\n";
2469
2470 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2471 /fran.ais/ &&
2472 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2473 "ok 814\n" : "not ok 814\n";
2474
2475 print "fran\N{LATIN SMALL LETTER C}ais" =~
2476 /fran\Cais/ &&
2477 $& eq "francais" ?
2478 "ok 815\n" : "not ok 815\n";
2479
2480 print "franc\N{COMBINING CEDILLA}ais" =~
2481 /franc\C\Cais/ ? # COMBINING CEDILLA is two bytes when encoded
2482 "ok 816\n" : "not ok 816\n";
2483
2484 print "fran\N{LATIN SMALL LETTER C}ais" =~
2485 /fran\Xais/ &&
2486 $& eq "francais" ?
2487 "ok 817\n" : "not ok 817\n";
2488
2489 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2490 /fran\Xais/ &&
2491 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2492 "ok 818\n" : "not ok 818\n";
2493
2494 print "franc\N{COMBINING CEDILLA}ais" =~
2495 /fran\Xais/ &&
2496 $& eq "franc\N{COMBINING CEDILLA}ais" ?
2497 "ok 819\n" : "not ok 819\n";
2498
2499 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2500 /fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais/ &&
2501 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2502 "ok 820\n" : "not ok 820\n";
2503
2504 print "franc\N{COMBINING CEDILLA}ais" =~
2505 /franc\N{COMBINING CEDILLA}ais/ &&
2506 $& eq "franc\N{COMBINING CEDILLA}ais" ?
2507 "ok 821\n" : "not ok 821\n";
2508
2509 print "fran\N{LATIN SMALL LETTER C}ais" =~
2510 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2511 $& eq "francais" ?
2512 "ok 822\n" : "not ok 822\n";
2513
2514 print "fran\N{LATIN SMALL LETTER C}ais" =~
2515 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2516 $& eq "francais" ?
2517 "ok 823\n" : "not ok 823\n";
2518
2519 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2520 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2521 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2522 "ok 824\n" : "not ok 824\n";
2523
2524 print "franc\N{COMBINING CEDILLA}ais" =~
2525 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2526 $& eq "franc\N{COMBINING CEDILLA}ais" ?
2527 "ok 825\n" : "not ok 825\n";
2528}
ffce6cc2
JH
2529
2530{
2531 print "# Does lingering (and useless) UTF8 flag mess up /i matching?\n";
2532
2533 {
2534 my $regex = "ABcde";
2535 my $string = "abcDE\x{100}";
2536 chop($string);
2537 if ($string =~ m/$regex/i) {
2538 print "ok 826\n";
2539 } else {
2540 print "not ok 826\n";
2541 }
2542 }
2543
2544 {
2545 my $regex = "ABcde\x{100}";
2546 my $string = "abcDE";
2547 chop($regex);
2548 if ($string =~ m/$regex/i) {
2549 print "ok 827\n";
2550 } else {
2551 print "not ok 827\n";
2552 }
2553 }
2554
2555 {
2556 my $regex = "ABcde\x{100}";
2557 my $string = "abcDE\x{100}";
2558 chop($regex);
2559 chop($string);
2560 if ($string =~ m/$regex/i) {
2561 print "ok 828\n";
2562 } else {
2563 print "not ok 828\n";
2564 }
2565 }
2566}
d07ddd77
JH
2567
2568{
2569 print "# more SIGMAs\n";
2570
2571 my $SIGMA = "\x{03A3}"; # CAPITAL
2572 my $Sigma = "\x{03C2}"; # SMALL FINAL
2573 my $sigma = "\x{03C3}"; # SMALL
2574
2575 my $S3 = "$SIGMA$Sigma$sigma";
2576
2577 print ":$S3:" =~ /:(($SIGMA)+):/i && $1 eq $S3 && $2 eq $sigma ?
2578 "ok 829\n" : "not ok 829\n";
2579 print ":$S3:" =~ /:(($Sigma)+):/i && $1 eq $S3 && $2 eq $sigma ?
2580 "ok 830\n" : "not ok 830\n";
2581 print ":$S3:" =~ /:(($sigma)+):/i && $1 eq $S3 && $2 eq $sigma ?
2582 "ok 831\n" : "not ok 831\n";
2583
2584 print ":$S3:" =~ /:(([$SIGMA])+):/i && $1 eq $S3 && $2 eq $sigma ?
2585 "ok 832\n" : "not ok 832\n";
2586 print ":$S3:" =~ /:(([$Sigma])+):/i && $1 eq $S3 && $2 eq $sigma ?
2587 "ok 833\n" : "not ok 833\n";
2588 print ":$S3:" =~ /:(([$sigma])+):/i && $1 eq $S3 && $2 eq $sigma ?
2589 "ok 834\n" : "not ok 834\n";
2590}
925f9e00
JH
2591
2592{
2593 print "# LATIN SMALL LETTER SHARP S\n";
2594
2595 use charnames ':full';
2596
2597 print "\N{LATIN SMALL LETTER SHARP S}" =~
2598 /\N{LATIN SMALL LETTER SHARP S}/ ? "ok 835\n" : "not ok 835\n";
2599
2600 print "\N{LATIN SMALL LETTER SHARP S}" =~
2601 /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 836\n" : "not ok 836\n";
2602
2603 print "\N{LATIN SMALL LETTER SHARP S}" =~
2604 /[\N{LATIN SMALL LETTER SHARP S}]/ ? "ok 837\n" : "not ok 837\n";
2605
2606 print "\N{LATIN SMALL LETTER SHARP S}" =~
2607 /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 838\n" : "not ok 838\n";
2608
2609 print "ss" =~
2610 /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 839\n" : "not ok 839\n";
2611
2612 print "SS" =~
2613 /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 840\n" : "not ok 840\n";
2614
e0f9d4a8
JH
2615 print "ss" =~
2616 /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 841\n" : "not ok 841\n";
2617
2618 print "SS" =~
2619 /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 842\n" : "not ok 842\n";
5486206c
JH
2620
2621 print "\N{LATIN SMALL LETTER SHARP S}" =~ /ss/i ?
2622 "ok 843\n" : "not ok 843\n";
2623
2624 print "\N{LATIN SMALL LETTER SHARP S}" =~ /SS/i ?
2625 "ok 844\n" : "not ok 844\n";
925f9e00 2626}
d8f6a732
JH
2627
2628{
2629 print "# more whitespace: U+0085, U+2028, U+2029\n";
2630
2631 # U+0085 needs to be forced to be Unicode, the \x{100} does that.
5486206c
JH
2632 print "<\x{100}\x{0085}>" =~ /<\x{100}\s>/ ? "ok 845\n" : "not ok 845\n";
2633 print "<\x{2028}>" =~ /<\s>/ ? "ok 846\n" : "not ok 846\n";
2634 print "<\x{2029}>" =~ /<\s>/ ? "ok 847\n" : "not ok 847\n";
d8f6a732
JH
2635}
2636
def8e4ea 2637{
229196e5 2638 print "# . with /s should work on characters, as opposed to bytes\n";
def8e4ea
JH
2639
2640 my $s = "\x{e4}\x{100}";
2641
2642 # This is not expected to match: the point is that
2643 # neither should we get "Malformed UTF-8" warnings.
2644 print $s =~ /\G(.+?)\n/gcs ?
5486206c 2645 "not ok 848\n" : "ok 848\n";
def8e4ea
JH
2646
2647 my @c;
2648
2649 while ($s =~ /\G(.)/gs) {
2650 push @c, $1;
2651 }
2652
5486206c 2653 print join("", @c) eq $s ? "ok 849\n" : "not ok 849\n";
a0804c9e
A
2654
2655 my $t1 = "Q003\n\n\x{e4}\x{f6}\n\nQ004\n\n\x{e7}"; # test only chars < 256
2656 my $r1 = "";
2657 while ($t1 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) {
2658 $r1 .= $1 . $2;
2659 }
2660
2661 my $t2 = $t1 . "\x{100}"; # repeat with a larger char
2662 my $r2 = "";
2663 while ($t2 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) {
2664 $r2 .= $1 . $2;
2665 }
2666 $r2 =~ s/\x{100}//;
5486206c 2667 print $r1 eq $r2 ? "ok 850\n" : "not ok 850\n";
def8e4ea 2668}
e54858b0
JH
2669
2670{
2671 print "# Unicode lookbehind\n";
2672
5486206c
JH
2673 print "A\x{100}B" =~ /(?<=A.)B/ ? "ok 851\n" : "not ok 851\n";
2674 print "A\x{200}\x{300}B" =~ /(?<=A..)B/ ? "ok 852\n" : "not ok 852\n";
2675 print "\x{400}AB" =~ /(?<=\x{400}.)B/ ? "ok 853\n" : "not ok 853\n";
2676 print "\x{500\x{600}}B" =~ /(?<=\x{500}.)B/ ? "ok 854\n" : "not ok 854\n";
e54858b0 2677}
c46248c1
JH
2678
2679{
85fd1718
JH
2680 print "# UTF-8 hash keys and /\$/\n";
2681 # http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2002-01/msg01327.html
2682
2683 my $u = "a\x{100}";
2684 my $v = substr($u,0,1);
2685 my $w = substr($u,1,1);
2686 my %u = ( $u => $u, $v => $v, $w => $w );
8d21bda2 2687 my $i = 855;
85fd1718
JH
2688 for (keys %u) {
2689 my $m1 = /^\w*$/ ? 1 : 0;
2690 my $m2 = $u{$_}=~/^\w*$/ ? 1 : 0;
2691 print $m1 == $m2 ? "ok $i\n" : "not ok $i # $m1 $m2\n";
2692 $i++;
2693 }
2694}
8d21bda2
JH
2695
2696{
2697 print "# [ID 20020124.005]\n";
2698 # Fixed by #14795.
2699 my $i = 858;
2700 for my $char ("a", "\x{df}", "\x{100}"){
2701 $x = "$char b $char";
2702 $x =~ s{($char)}{
2703 "c" =~ /c/;
2704 "x";
2705 }ge;
2706 print substr($x,0,1) eq substr($x,-1,1) ?
2707 "ok $i\n" : "not ok $i # debug: $x\n";
2708 $i++;
2709 }
2710}
6e602e29
A
2711
2712{
2713 print "# SEGV in s/// and UTF-8\n";
2714 $s = "s#\x{100}" x 4;
2715 $s =~ s/[^\w]/ /g;
2716 print $s eq "s \x{100}" x 4 ? "ok 861\n" : "not ok 861\n";
2717}
f272994b
A
2718
2719{
2720 print "# UTF-8 bug (maybe alreayd known?)\n";
2721 my $u;
2722
2723 $u = "foo";
2724 $u =~ s/./\x{100}/g;
2725 print $u eq "\x{100}\x{100}\x{100}" ? "ok 862\n" : "not ok 862\n";
2726
2727 $u = "foobar";
2728 $u =~ s/[ao]/\x{100}/g;
2729 print $u eq "f\x{100}\x{100}b\x{100}r" ? "ok 863\n" : "not ok 863\n";
2730
2731 $u =~ s/\x{100}/e/g;
2732 print $u eq "feeber" ? "ok 864\n" : "not ok 864\n";
2733}
2734
33b8afdf
JH
2735{
2736 print "# UTF-8 bug with s///\n";
2737 # check utf8/non-utf8 mixtures
2738 # try to force all float/anchored check combinations
2739 my $c = "\x{100}";
2740 my $test = 865;
2741 my $subst;
2742 for my $re (
2743 "xx.*$c", "x.*$c$c", "$c.*xx", "$c$c.*x", "xx.*(?=$c)", "(?=$c).*xx",
2744 ) {
2745 print "xxx" =~ /$re/ ? "not ok $test\n" : "ok $test\n";
2746 ++$test;
2747 print +($subst = "xxx") =~ s/$re// ? "not ok $test\n" : "ok $test\n";
2748 ++$test;
2749 }
2750 for my $re ("xx.*$c*", "$c*.*xx") {
2751 print "xxx" =~ /$re/ ? "ok $test\n" : "not ok $test\n";
2752 ++$test;
2753 ($subst = "xxx") =~ s/$re//;
2754 print $subst eq '' ? "ok $test\n" : "not ok $test\t# $subst\n";
2755 ++$test;
2756 }
2757 for my $re ("xxy*", "y*xx") {
2758 print "xx$c" =~ /$re/ ? "ok $test\n" : "not ok $test\n";
2759 ++$test;
2760 ($subst = "xx$c") =~ s/$re//;
2761 print $subst eq $c ? "ok $test\n" : "not ok $test\n";
2762 ++$test;
2763 print "xy$c" =~ /$re/ ? "not ok $test\n" : "ok $test\n";
2764 ++$test;
2765 print +($subst = "xy$c") =~ /$re/ ? "not ok $test\n" : "ok $test\n";
2766 ++$test;
2767 }
2768 for my $re ("xy$c*z", "x$c*yz") {
2769 print "xyz" =~ /$re/ ? "ok $test\n" : "not ok $test\n";
2770 ++$test;
2771 ($subst = "xyz") =~ s/$re//;
2772 print $subst eq '' ? "ok $test\n" : "not ok $test\n";
2773 ++$test;
2774 }
2775}
574c8022 2776
ff385a1b
JF
2777{
2778 print "# qr/.../x\n";
4c79aee6 2779 my $test = 893;
ff385a1b
JF
2780
2781 my $R = qr/ A B C # D E/x;
2782
2783 print eval {"ABCDE" =~ $R} ? "ok $test\n" : "not ok $test\n";
2784 $test++;
2785
2786 print eval {"ABCDE" =~ m/$R/} ? "ok $test\n" : "not ok $test\n";
2787 $test++;
2788
2789 print eval {"ABCDE" =~ m/($R)/} ? "ok $test\n" : "not ok $test\n";
2790 $test++;
2791}
bc45ce41
JH
2792
2793{
2794 print "# illegal Unicode properties\n";
4c79aee6 2795 my $test = 896;
bc45ce41
JH
2796
2797 print eval { "a" =~ /\pq / } ? "not ok $test\n" : "ok $test\n";
2798 $test++;
2799
2800 print eval { "a" =~ /\p{qrst} / } ? "not ok $test\n" : "ok $test\n";
2801 $test++;
2802}
d65afb4b
HS
2803
2804{
491fd90a 2805 print "# [ID 20020412.005] wrong pmop flags checked when empty pattern\n";
d65afb4b
HS
2806 # requires reuse of last successful pattern
2807 my $test = 898;
2808 $test =~ /\d/;
2809 for (0 .. 1) {
2810 my $match = ?? + 0;
2811 if ($match != $_) {
2812 print "ok $test\n";
2813 } else {
2814 printf "not ok %s\t# 'match once' %s on %s iteration\n", $test,
2815 $match ? 'succeeded' : 'failed', $_ ? 'second' : 'first';
2816 }
2817 ++$test;
2818 }
2819 $test =~ /(\d)/;
2820 my $result = join '', $test =~ //g;
2821 if ($result eq $test) {
2822 print "ok $test\n";
2823 } else {
2824 printf "not ok %s\t# expected '%s', got '%s'\n", $test, $test, $result;
2825 }
2826 ++$test;
2827}
491fd90a
JH
2828
2829print "# user-defined character properties\n";
2830
2831sub InKana1 {
2832 return <<'END';
28333040 309F
283430A0 30FF
2835END
2836}
2837
2838sub InKana2 {
2839 return <<'END';
2840+utf8::InHiragana
2841+utf8::InKatakana
2842END
2843}
2844
2845sub InKana3 {
2846 return <<'END';
2847+utf8::InHiragana
2848+utf8::InKatakana
2849-utf8::IsCn
2850END
2851}
2852
2853sub InNotKana {
2854 return <<'END';
2855!utf8::InHiragana
2856-utf8::InKatakana
2857+utf8::IsCn
2858END
2859}
2860
2861$test = 901;
2862
2863print "\x{3040}" =~ /\p{InKana1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2864print "\x{303F}" =~ /\P{InKana1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2865
2866print "\x{3040}" =~ /\p{InKana2}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2867print "\x{303F}" =~ /\P{InKana2}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2868
2869print "\x{3041}" =~ /\p{InKana3}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2870print "\x{3040}" =~ /\P{InKana3}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2871
2872print "\x{3040}" =~ /\p{InNotKana}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2873print "\x{3041}" =~ /\P{InNotKana}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2874
11ef8fdd
JH
2875sub InConsonant { # Not EBCDIC-aware.
2876 return <<EOF;
28770061 007f
2878-0061
2879-0065
2880-0069
2881-006f
2882-0075
2883EOF
2884}
2885
2886print "d" =~ /\p{InConsonant}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2887print "e" =~ /\P{InConsonant}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2888
faf11cac
HS
2889{
2890 print "# [ID 20020630.002] utf8 regex only matches 32k\n";
2891 $test = 911;
2892 for ([ 'byte', "\x{ff}" ], [ 'utf8', "\x{1ff}" ]) {
2893 my($type, $char) = @$_;
2894 for my $len (32000, 32768, 33000) {
2895 my $s = $char . "f" x $len;
2896 my $r = $s =~ /$char([f]*)/gc;
2897 print $r ? "ok $test\n" : "not ok $test\t# <$type x $len> fail\n";
2898 ++$test;
2899 print +(!$r or pos($s) == $len + 1) ? "ok $test\n"
2900 : "not ok $test\t# <$type x $len> pos @{[ pos($s) ]}\n";
2901 ++$test;
2902 }
2903 }
2904}
2905
2906$test = 923;
f3b1e556
YST
2907
2908$a = bless qr/foo/, 'Foo';
2909print(('goodfood' =~ $a ? '' : 'not '),
2910 "ok $test\t# reblessed qr// matches\n");
2911++$test;
2912
2913print(($a eq '(?-xism:foo)' ? '' : 'not '),
2914 "ok $test\t# reblessed qr// stringizes\n");
2915++$test;
446eaa42
YST
2916
2917$x = "\x{3fe}";
cb50f42d
YST
2918$z=$y = "\317\276"; # $y is byte representation of $x
2919
446eaa42
YST
2920$a = qr/$x/;
2921print(($x =~ $a ? '' : 'not '), "ok $test - utf8 interpolation in qr//\n");
2922++$test;
2923
2924print(("a$a" =~ $x ? '' : 'not '),
cb50f42d
YST
2925 "ok $test - stringifed qr// preserves utf8\n");
2926++$test;
2927
2928print(("a$x" =~ /^a$a\z/ ? '' : 'not '),
2929 "ok $test - interpolated qr// preserves utf8\n");
2930++$test;
2931
2932print(("a$x" =~ /^a(??{$a})\z/ ? '' : 'not '),
2933 "ok $test - postponed interpolation of qr// preserves utf8\n");
2934++$test;
2935
ab01544f
RGS
2936print((length(qr/##/x) == 12 ? '' : 'not '),
2937 "ok $test - ## in qr// doesn't corrupt memory [perl #17776]\n");
2938++$test;
2939
cb50f42d
YST
2940{ use re 'eval';
2941
2942print(("$x$x" =~ /^$x(??{$x})\z/ ? '' : 'not '),
2943 "ok $test - postponed utf8 string in utf8 re matches utf8\n");
2944++$test;
2945
2946print(("$y$x" =~ /^$y(??{$x})\z/ ? '' : 'not '),
2947 "ok $test - postponed utf8 string in non-utf8 re matches utf8\n");
446eaa42
YST
2948++$test;
2949
cb50f42d
YST
2950print(("$y$x" !~ /^$y(??{$y})\z/ ? '' : 'not '),
2951 "ok $test - postponed non-utf8 string in non-utf8 re doesn't match utf8\n");
446eaa42
YST
2952++$test;
2953
cb50f42d
YST
2954print(("$x$x" !~ /^$x(??{$y})\z/ ? '' : 'not '),
2955 "ok $test - postponed non-utf8 string in utf8 re doesn't match utf8\n");
446eaa42
YST
2956++$test;
2957
cb50f42d
YST
2958print(("$y$y" =~ /^$y(??{$y})\z/ ? '' : 'not '),
2959 "ok $test - postponed non-utf8 string in non-utf8 re matches non-utf8\n");
2960++$test;
2961
2962print(("$x$y" =~ /^$x(??{$y})\z/ ? '' : 'not '),
2963 "ok $test - postponed non-utf8 string in utf8 re matches non-utf8\n");
2964++$test;
2965$y = $z; # reset $y after upgrade
2966
2967print(("$x$y" !~ /^$x(??{$x})\z/ ? '' : 'not '),
2968 "ok $test - postponed utf8 string in utf8 re doesn't match non-utf8\n");
2969++$test;
2970$y = $z; # reset $y after upgrade
2971
2972print(("$y$y" !~ /^$y(??{$x})\z/ ? '' : 'not '),
2973 "ok $test - postponed utf8 string in non-utf8 re doesn't match non-utf8\n");
2974++$test;
2975
2976} # no re 'eval'
2977
2d92f8a0
JH
2978print "# more user-defined character properties\n";
2979
2980sub IsSyriac1 {
2981 return <<'END';
29820712 072C
29830730 074A
2984END
2985}
2986
2987print "\x{0712}" =~ /\p{IsSyriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2988print "\x{072F}" =~ /\P{IsSyriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2989
2990sub Syriac1 {
2991 return <<'END';
29920712 072C
29930730 074A
2994END
2995}
2996
2997print "\x{0712}" =~ /\p{Syriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2998print "\x{072F}" =~ /\P{Syriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2999
0e933229 3000{
3a2263fe
RGS
3001 print "# Change #18179\n";
3002 # previously failed with "panic: end_shift
0e933229
IH
3003 my $s = "\x{100}" x 5;
3004 my $ok = $s =~ /(\x{100}{4})/;
3005 my($ord, $len) = (ord $1, length $1);
3006 print +($ok && $ord == 0x100 && $len == 4)
3007 ? "ok $test\n" : "not ok $test\t# $ok/$ord/$len\n";
3008 ++$test;
3009}
3010
f14c76ed
RGS
3011{
3012 print "# [perl #15763]\n";
3013
3014 $a = "x\x{100}";
3015 chop $a; # but leaves the UTF-8 flag
3016 $a .= "y"; # 1 byte before "y"
3017
3018 ok($a =~ /^\C/, 'match one \C on 1-byte UTF-8');
3019 ok($a =~ /^\C{1}/, 'match \C{1}');
3020
3021 ok($a =~ /^\Cy/, 'match \Cy');
3022 ok($a =~ /^\C{1}y/, 'match \C{1}y');
3023
3024 $a = "\x{100}y"; # 2 bytes before "y"
3025
3026 ok($a =~ /^\C/, 'match one \C on 2-byte UTF-8');
3027 ok($a =~ /^\C{1}/, 'match \C{1}');
3028 ok($a =~ /^\C\C/, 'match two \C');
3029 ok($a =~ /^\C{2}/, 'match \C{2}');
3030
3031 ok($a =~ /^\C\C\C/, 'match three \C on 2-byte UTF-8 and a byte');
3032 ok($a =~ /^\C{3}/, 'match \C{3}');
3033
3034 ok($a =~ /^\C\Cy/, 'match two \C');
3035 ok($a =~ /^\C{2}y/, 'match \C{2}');
3036
836995da
RGS
3037 ok($a !~ /^\C\C\Cy/, q{don't match three \Cy});
3038 ok($a !~ /^\C{2}\Cy/, q{don't match \C{3}y});
f14c76ed
RGS
3039
3040 $a = "\x{1000}y"; # 3 bytes before "y"
3041
3042 ok($a =~ /^\C/, 'match one \C on three-byte UTF-8');
3043 ok($a =~ /^\C{1}/, 'match \C{1}');
3044 ok($a =~ /^\C\C/, 'match two \C');
3045 ok($a =~ /^\C{2}/, 'match \C{2}');
3046 ok($a =~ /^\C\C\C/, 'match three \C');
3047 ok($a =~ /^\C{3}/, 'match \C{3}');
3048
3049 ok($a =~ /^\C\C\C\C/, 'match four \C on three-byte UTF-8 and a byte');
3050 ok($a =~ /^\C{4}/, 'match \C{4}');
3051
3052 ok($a =~ /^\C\C\Cy/, 'match three \Cy');
3053 ok($a =~ /^\C{3}y/, 'match \C{3}y');
3054
836995da
RGS
3055 ok($a !~ /^\C\C\C\C\y/, q{don't match four \Cy});
3056 ok($a !~ /^\C{4}y/, q{don't match \C{4}y});
f14c76ed
RGS
3057}
3058
5dec093f
AE
3059$_ = 'aaaaaaaaaa';
3060utf8::upgrade($_); chop $_; $\="\n";
3061ok(/[^\s]+/, "m/[^\s]/ utf8");
3062ok(/[^\d]+/, "m/[^\d]/ utf8");
3063ok(($a = $_, $_ =~ s/[^\s]+/./g), "s/[^\s]/ utf8");
3064ok(($a = $_, $a =~ s/[^\d]+/./g), "s/[^\s]/ utf8");
f14c76ed 3065
ce97f9d6
JH
3066ok("\x{100}" =~ /\x{100}/, "[perl #15397]");
3067ok("\x{100}" =~ /(\x{100})/, "[perl #15397]");
3068ok("\x{100}" =~ /(\x{100}){1}/, "[perl #15397]");
3069ok("\x{100}\x{100}" =~ /(\x{100}){2}/, "[perl #15397]");
3070ok("\x{100}\x{100}" =~ /(\x{100})(\x{100})/, "[perl #15397]");
3071
fbeb8e69
JH
3072$x = "CD";
3073$x =~ /(AB)*?CD/;
3074ok(!defined $1, "[perl #7471]");
ce97f9d6 3075
fbeb8e69
JH
3076$x = "CD";
3077$x =~ /(AB)*CD/;
3078ok(!defined $1, "[perl #7471]");
3079
ec391688
JH
3080$pattern = "^(b+?|a){1,2}c";
3081ok("bac" =~ /$pattern/ && $1 eq 'a', "[perl #3547]");
3082ok("bbac" =~ /$pattern/ && $1 eq 'a', "[perl #3547]");
3083ok("bbbac" =~ /$pattern/ && $1 eq 'a', "[perl #3547]");
3084ok("bbbbac" =~ /$pattern/ && $1 eq 'a', "[perl #3547]");
3085
a30b2f1f
RGS
3086{
3087 # [perl #18232]
3088 "\x{100}" =~ /(.)/;
3089 ok( $1 eq "\x{100}", '$1 is utf-8 [perl #18232]' );
3090 { 'a' =~ /./; }
3091 ok( $1 eq "\x{100}", '$1 is still utf-8' );
3092 ok( $1 ne "\xC4\x80", '$1 is not non-utf-8' );
3093}
3094
f119b0fb
JH
3095{
3096 use utf8;
3097 my $attr = 'Name-1' ;
3098
3099 my $NormalChar = qr/[\p{IsDigit}\p{IsLower}\p{IsUpper}]/;
3100 my $NormalWord = qr/${NormalChar}+?/;
3101 my $PredNameHyphen = qr/^${NormalWord}(\-${NormalWord})*?$/;
3102
3103 $attr =~ /^$/;
3104 ok( $attr =~ $PredNameHyphen, "[perl #19767] original test" );
3105}
3106
3107{
3108 use utf8;
3109 "a" =~ m/[b]/;
3110 ok ( "0" =~ /\p{N}+\z/, "[perl #19767] variant test" );
3111}
3112
faf82a0b
AE
3113{
3114
3115 $p = 1;
3116 foreach (1,2,3,4) {
3117 $p++ if /(??{ $p })/
3118 }
3119 ok ($p == 5, "[perl #20683] (??{ }) returns stale values");
3120 { package P; $a=1; sub TIESCALAR { bless[] } sub FETCH { $a++ } }
3121 tie $p, P;
3122 foreach (1,2,3,4) {
3123 /(??{ $p })/
3124 }
3125 ok ( $p == 5, "(??{ }) returns stale values");
3126}
3127
351208f1
JH
3128{
3129 # Subject: Odd regexp behavior
3130 # From: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>
3131 # Date: Wed, 26 Feb 2003 16:53:12 +0000
3132 # Message-Id: <E18o4nw-0008Ly-00@wisbech.cl.cam.ac.uk>
3133 # To: perl-unicode@perl.org
3134
3135 $x = "\x{2019}\nk"; $x =~ s/(\S)\n(\S)/$1 $2/sg;
3136 ok($x eq "\x{2019} k", "Markus Kuhn 2003-02-26");
3137
3138 $x = "b\nk"; $x =~ s/(\S)\n(\S)/$1 $2/sg;
3139 ok($x eq "b k", "Markus Kuhn 2003-02-26");
3140
3141 ok("\x{2019}" =~ /\S/, "Markus Kuhn 2003-02-26");
3142}
3143
91e09a61
AE
3144{
3145 my $i;
3146 ok('-1-3-5-' eq join('', split /((??{$i++}))/, '-1-3-5-'),
080c2dec
AE
3147 "[perl #21411] (??{ .. }) corrupts split's stack");
3148 split /(?{'WOW'})/, 'abc';
3149 ok('a|b|c' eq join ('|', @_),
3150 "[perl #21411] (?{ .. }) version of the above");
3151}
3152
3153{
3154 split /(?{ split "" })/, "abc";
3155 ok(1,'cache_re & "(?{": it dumps core in 5.6.1 & 5.8.0');
91e09a61
AE
3156}
3157
c517dc2b
JH
3158{
3159 ok("\x{100}\n" =~ /\x{100}\n$/, "UTF8 length cache and fbm_compile");
3160}
ec391688 3161
16e1b944
JH
3162{
3163 package Str;
3164 use overload q/""/ => sub { ${$_[0]}; };
3165 sub new { my ($c, $v) = @_; bless \$v, $c; }
3166
3167 package main;
3168 $_ = Str->new("a\x{100}/\x{100}b");
3169 ok(join(":", /\b(.)\x{100}/g) eq "a:/", "re_intuit_start and PL_bostr");
3170}
3171
7ef91622
IH
3172{
3173 $_ = "code: 'x' { '...' }\n"; study;
3174 my @x; push @x, $& while m/'[^\']*'/gx;
3175 ok(join(":", @x) eq "'x':'...'",
3176 "[perl #17757] Parse::RecDescent triggers infinite loop");
3177}
3178
14ebb1a2
JH
3179{
3180 my $re = qq/^([^X]*)X/;
3181 utf8::upgrade($re);
3182 ok("\x{100}X" =~ /$re/, "S_cl_and ANYOF_UNICODE & ANYOF_INVERTED");
3183}
3184
f02c194e
RGS
3185# bug #22354
3186sub func ($) {
3187 ok( "a\nb" !~ /^b/, $_[0] );
3188 ok( "a\nb" =~ /^b/m, "$_[0] - with /m" );
3189}
3190func "standalone";
3191$_ = "x"; s/x/func "in subst"/e;
3192$_ = "x"; s/x/func "in multiline subst"/em;
3193#$_ = "x"; /x(?{func "in regexp"})/;
3194#$_ = "x"; /x(?{func "in multiline regexp"})/m;
3195
5aca4364
AG
3196# bug #19049
3197$_="abcdef\n";
3198@x = m/./g;
973dddac 3199ok("abcde" eq "$`", '# TODO #19049 - global match not setting $`');
5aca4364 3200
6b43b216
DM
3201ok("123\x{100}" =~ /^.*1.*23\x{100}$/, 'uft8 + multiple floating substr');
3202
89d0f8e1 3203# LATIN SMALL/CAPITAL LETTER A WITH MACRON
c1e0e3d2
NIS
3204ok(" \x{101}" =~ qr/\x{100}/i,
3205 "<20030808193656.5109.1@llama.ni-s.u-net.com>");
3206
89d0f8e1 3207# LATIN SMALL/CAPITAL LETTER A WITH RING BELOW
5dab1207
NIS
3208ok(" \x{1E01}" =~ qr/\x{1E00}/i,
3209 "<20030808193656.5109.1@llama.ni-s.u-net.com>");
3210
89d0f8e1
JH
3211# DESERET SMALL/CAPITAL LETTER LONG I
3212ok(" \x{10428}" =~ qr/\x{10400}/i,
3213 "<20030808193656.5109.1@llama.ni-s.u-net.com>");
3214
3215# LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X'
3216ok(" \x{1E01}x" =~ qr/\x{1E00}X/i,
3217 "<20030808193656.5109.1@llama.ni-s.u-net.com>");
3218
a7913593
JH
3219{
3220 # [perl #23769] Unicode regex broken on simple example
3221 # regrepeat() didn't handle UTF-8 EXACT case right.
3222
3223 my $s = "\x{a0}\x{a0}\x{a0}\x{100}"; chop $s;
3224
3225 ok($s =~ /\x{a0}/, "[perl #23769]");
3226 ok($s =~ /\x{a0}+/, "[perl #23769]");
3227 ok($s =~ /\x{a0}\x{a0}/, "[perl #23769]");
090f7165
JH
3228
3229 ok("aaa\x{100}" =~ /(a+)/, "[perl #23769] easy invariant");
3230 ok($1 eq "aaa", "[perl #23769]");
3231
3232 ok("\xa0\xa0\xa0\x{100}" =~ /(\xa0+)/, "[perl #23769] regrepeat invariant");
3233 ok($1 eq "\xa0\xa0\xa0", "[perl #23769]");
3234
3235 ok("ababab\x{100} " =~ /((?:ab)+)/, "[perl #23769] hard invariant");
3236 ok($1 eq "ababab", "[perl #23769]");
3237
3238 ok("\xa0\xa1\xa0\xa1\xa0\xa1\x{100}" =~ /((?:\xa0\xa1)+)/, "[perl #23769] hard variant");
3239 ok($1 eq "\xa0\xa1\xa0\xa1\xa0\xa1", "[perl #23769]");
3240
3241 ok("aaa\x{100} " =~ /(a+?)/, "[perl #23769] easy invariant");
3242 ok($1 eq "a", "[perl #23769]");
3243
3244 ok("\xa0\xa0\xa0\x{100} " =~ /(\xa0+?)/, "[perl #23769] regrepeat variant");
3245 ok($1 eq "\xa0", "[perl #23769]");
3246
3247 ok("ababab\x{100} " =~ /((?:ab)+?)/, "[perl #23769] hard invariant");
3248 ok($1 eq "ab", "[perl #23769]");
3249
3250 ok("\xa0\xa1\xa0\xa1\xa0\xa1\x{100}" =~ /((?:\xa0\xa1)+?)/, "[perl #23769] hard variant");
3251 ok($1 eq "\xa0\xa1", "[perl #23769]");
3252
3253 ok("\xc4\xc4\xc4" !~ /(\x{100}+)/, "[perl #23769] don't match first byte of utf8 representation");
3254 ok("\xc4\xc4\xc4" !~ /(\x{100}+?)/, "[perl #23769] don't match first byte of utf8 representation");
a7913593
JH
3255}
3256
4be7a33f
HS
3257for (120 .. 130) {
3258 my $head = 'x' x $_;
3259 for my $tail ('\x{0061}', '\x{1234}') {
3260 ok(
3261 eval qq{ "$head$tail" =~ /$head$tail/ },
3262 '\x{...} misparsed in regexp near 127 char EXACT limit'
3263 );
3264 }
3265}
3266
289555fe
HS
3267# perl #25269: panic: pp_match start/end pointers
3268ok("a-bc" eq eval {
3269 my($x, $y) = "bca" =~ /^(?=.*(a)).*(bc)/;
3270 "$x-$y";
3271}, 'captures can move backwards in string');
3272
b6d5fef8
RGS
3273# perl #27940: \cA not recognized in character classes
3274ok("a\cAb" =~ /\cA/, '\cA in pattern');
3275ok("a\cAb" =~ /[\cA]/, '\cA in character class');
3276ok("a\cAb" =~ /[\cA-\cB]/, '\cA in character class range');
3277ok("abc" =~ /[^\cA-\cB]/, '\cA in negated character class range');
3278ok("a\cBb" =~ /[\cA-\cC]/, '\cB in character class range');
3279ok("a\cCbc" =~ /[^\cA-\cB]/, '\cC in negated character class range');
3280ok("a\cAb" =~ /(??{"\cA"})/, '\cA in ??{} pattern');
5dab1207 3281
0e788c72
HS
3282# perl #28532: optional zero-width match at end of string is ignored
3283ok(("abc" =~ /^abc(\z)?/) && defined($1),
3284 'optional zero-width match at end of string');
3285ok(("abc" =~ /^abc(\z)??/) && !defined($1),
3286 'optional zero-width match at end of string');
3287
a3621e74
YO
3288
3289
3290{ # TRIE related
3291 my @got=();
3292 "words"=~/(word|word|word)(?{push @got,$1})s$/;
3293 ok(@got==1,"TRIE optimation is working") or warn "# @got";
3294 @got=();
3295 "words"=~/(word|word|word)(?{push @got,$1})s$/i;
3296 ok(@got==1,"TRIEF optimisation is working") or warn "# @got";
3297
3298 my @nums=map {int rand 1000} 1..100;
3299 my $re="(".(join "|",@nums).")";
3300 $re=qr/\b$re\b/;
3301
3302 foreach (@nums) {
3303 ok($_=~/$re/,"Trie nums");
3304 }
3305 $_=join " ", @nums;
3306 @got=();
3307 push @got,$1 while /$re/g;
3308
3309 my %count;
3310 $count{$_}++ for @got;
3311 my $ok=1;
3312 for (@nums) {
3313 $ok=0 if --$count{$_}<0;
3314 }
3315 ok($ok,"Trie min count matches");
3316}
3317
3318
3319# TRIE related
3320# LATIN SMALL/CAPITAL LETTER A WITH MACRON
3321ok(("foba \x{101}foo" =~ qr/(foo|\x{100}foo|bar)/i) && $1 eq "\x{101}foo",
3322 "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH MACRON");
3323
3324# LATIN SMALL/CAPITAL LETTER A WITH RING BELOW
3325ok(("foba \x{1E01}foo" =~ qr/(foo|\x{1E00}foo|bar)/i) && $1 eq "\x{1E01}foo",
3326 "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH RING BELOW");
3327
3328# DESERET SMALL/CAPITAL LETTER LONG I
3329ok(("foba \x{10428}foo" =~ qr/(foo|\x{10400}foo|bar)/i) && $1 eq "\x{10428}foo",
3330 "TRIEF + DESERET SMALL/CAPITAL LETTER LONG I");
3331
3332# LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X'
3333ok(("foba \x{1E01}xfoo" =~ qr/(foo|\x{1E00}Xfoo|bar)/i) && $1 eq "\x{1E01}xfoo",
3334 "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X'");
3335
3336{# TRIE related
3337
3338use charnames ':full';
3339
3340$s="\N{LATIN SMALL LETTER SHARP S}";
3341ok(("foba ba$s" =~ qr/(foo|Ba$s|bar)/i)
3342 && $1 eq "ba$s",
3343 "TRIEF + LATIN SMALL LETTER SHARP S =~ ss");
3344ok(("foba ba$s" =~ qr/(Ba$s|foo|bar)/i)
3345 && $1 eq "ba$s",
3346 "TRIEF + LATIN SMALL LETTER SHARP S =~ ss");
3347ok(("foba ba$s" =~ qr/(foo|bar|Ba$s)/i)
3348 && $1 eq "ba$s",
3349 "TRIEF + LATIN SMALL LETTER SHARP S =~ ss");
3350
3351ok(("foba ba$s" =~ qr/(foo|Bass|bar)/i)
3352 && $1 eq "ba$s",
3353 "TRIEF + LATIN SMALL LETTER SHARP S =~ ss");
3354
3355ok(("foba ba$s" =~ qr/(foo|BaSS|bar)/i)
3356 && $1 eq "ba$s",
3357 "TRIEF + LATIN SMALL LETTER SHARP S =~ SS");
3358}
3359
3360
3361
3362{
3363 my @normal=qw(these are some normal words);
3364 my $psycho=join "|",@normal,map chr $_,255..20000;
3365 ok(('these'=~/($psycho)/) && $1 eq 'these','Pyscho');
3366}
809e8e66
DM
3367
3368# [perl #36207] mixed utf8 / latin-1 and case folding
3369
3370{
108bb1ad
YO
3371 my $utf8 = "\xe9\x{100}"; chop $utf8;
3372 my $latin1 = "\xe9";
3373
3374 ok($utf8 =~ /\xe9/i, "utf8/latin");
3375 ok($utf8 =~ /$latin1/i, "utf8/latin runtime");
3376 ok($utf8 =~ /(abc|\xe9)/i, "utf8/latin trie");
3377 ok($utf8 =~ /(abc|$latin1)/i, "utf8/latin trie runtime");
3378
3379 ok("\xe9" =~ /$utf8/i, "# TODO latin/utf8");
3380 ok("\xe9" =~ /(abc|$utf8)/i, "# latin/utf8 trie");
3381 ok($latin1 =~ /$utf8/i, "# TODO latin/utf8 runtime");
3382 ok($latin1 =~ /(abc|$utf8)/i, "# latin/utf8 trie runtime");
809e8e66
DM
3383}
3384
740266bf
NC
3385# [perl #37038] Global regular matches generate invalid pointers
3386
3387{
3388 my $s = "abcd";
3389 $s =~ /(..)(..)/g;
3390 $s = $1;
3391 $s = $2;
3392 ok($s eq 'cd',
973dddac 3393 "# TODO assigning to original string should not corrupt match vars");
740266bf
NC
3394}
3395
3396# last test 1187
3397