projects / p5sagit/p5-mst-13.2.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | history | HEAD
The [perl #7471] seems to have been fixed; add its test.
[p5sagit/p5-mst-13.2.git] / t / op / pat.t
CommitLineData
8d063cd8 1#!./perl
8d37f932 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$| = 1;
3568d838 8
fbeb8e69 9print "1..979\n";
8d37f932 10
e4d48cc9 11BEGIN {
12 chdir 't' if -d 't';
20822f61 13 @INC = '../lib';
e4d48cc9 14}
ffbc6a93 15
8d37f932 16eval 'use Config'; # Defaults assumed if this fails
8d063cd8 17
18$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
23$* = 1;
24if ($x =~ /^def/) {print "ok 3\n";} else {print "not ok 3\n";}
25$* = 0;
26
27$_ = '123';
28if (/^([0-9][0-9]*)/) {print "ok 4\n";} else {print "not ok 4\n";}
29
30if ($x =~ /^xxx/) {print "not ok 5\n";} else {print "ok 5\n";}
31if ($x !~ /^abc/) {print "not ok 6\n";} else {print "ok 6\n";}
32
33if ($x =~ /def/) {print "ok 7\n";} else {print "not ok 7\n";}
34if ($x !~ /def/) {print "not ok 8\n";} else {print "ok 8\n";}
35
36if ($x !~ /.def/) {print "ok 9\n";} else {print "not ok 9\n";}
37if ($x =~ /.def/) {print "not ok 10\n";} else {print "ok 10\n";}
38
39if ($x =~ /\ndef/) {print "ok 11\n";} else {print "not ok 11\n";}
40if ($x !~ /\ndef/) {print "not ok 12\n";} else {print "ok 12\n";}
41
42$_ = 'aaabbbccc';
43if (/(a*b*)(c*)/ && $1 eq 'aaabbb' && $2 eq 'ccc') {
44 print "ok 13\n";
45} else {
46 print "not ok 13\n";
47}
48if (/(a+b+c+)/ && $1 eq 'aaabbbccc') {
49 print "ok 14\n";
50} else {
51 print "not ok 14\n";
52}
53
54if (/a+b?c+/) {print "not ok 15\n";} else {print "ok 15\n";}
55
56$_ = 'aaabccc';
57if (/a+b?c+/) {print "ok 16\n";} else {print "not ok 16\n";}
58if (/a*b+c*/) {print "ok 17\n";} else {print "not ok 17\n";}
59
60$_ = 'aaaccc';
61if (/a*b?c*/) {print "ok 18\n";} else {print "not ok 18\n";}
62if (/a*b+c*/) {print "not ok 19\n";} else {print "ok 19\n";}
63
64$_ = 'abcdef';
65if (/bcd|xyz/) {print "ok 20\n";} else {print "not ok 20\n";}
66if (/xyz|bcd/) {print "ok 21\n";} else {print "not ok 21\n";}
67
68if (m|bc/*d|) {print "ok 22\n";} else {print "not ok 22\n";}
378cc40b 69
70if (/^$_$/) {print "ok 23\n";} else {print "not ok 23\n";}
71
72$* = 1; # test 3 only tested the optimized version--this one is for real
73if ("ab\ncd\n" =~ /^cd/) {print "ok 24\n";} else {print "not ok 24\n";}
74$* = 0;
75
cb55de95 76$XXX{123} = 123;
77$XXX{234} = 234;
78$XXX{345} = 345;
79
80@XXX = ('ok 25','not ok 25', 'ok 26','not ok 26','not ok 27');
81while ($_ = shift(@XXX)) {
82 ?(.*)? && (print $1,"\n");
83 /not/ && reset;
84 /not ok 26/ && reset 'X';
85}
86
87while (($key,$val) = each(%XXX)) {
88 print "not ok 27\n";
89 exit;
90}
91
92print "ok 27\n";
378cc40b 93
378cc40b 94'cde' =~ /[^ab]*/;
95'xyz' =~ //;
96if ($& eq 'xyz') {print "ok 28\n";} else {print "not ok 28\n";}
97
98$foo = '[^ab]*';
99'cde' =~ /$foo/;
100'xyz' =~ //;
101if ($& eq 'xyz') {print "ok 29\n";} else {print "not ok 29\n";}
102
103$foo = '[^ab]*';
104'cde' =~ /$foo/;
105'xyz' =~ /$null/;
106if ($& eq 'xyz') {print "ok 30\n";} else {print "not ok 30\n";}
a687059c 107
108$_ = 'abcdefghi';
109/def/; # optimized up to cmd
110if ("$`:$&:$'" eq 'abc:def:ghi') {print "ok 31\n";} else {print "not ok 31\n";}
111
112/cde/ + 0; # optimized only to spat
113if ("$`:$&:$'" eq 'ab:cde:fghi') {print "ok 32\n";} else {print "not ok 32\n";}
114
115/[d][e][f]/; # not optimized
116if ("$`:$&:$'" eq 'abc:def:ghi') {print "ok 33\n";} else {print "not ok 33\n";}
117
118$_ = 'now is the {time for all} good men to come to.';
119/ {([^}]*)}/;
120if ($1 eq 'time for all') {print "ok 34\n";} else {print "not ok 34 $1\n";}
121
122$_ = 'xxx {3,4} yyy zzz';
123print /( {3,4})/ ? "ok 35\n" : "not ok 35\n";
124print $1 eq ' ' ? "ok 36\n" : "not ok 36\n";
125print /( {4,})/ ? "not ok 37\n" : "ok 37\n";
126print /( {2,3}.)/ ? "ok 38\n" : "not ok 38\n";
127print $1 eq ' y' ? "ok 39\n" : "not ok 39\n";
128print /(y{2,3}.)/ ? "ok 40\n" : "not ok 40\n";
129print $1 eq 'yyy ' ? "ok 41\n" : "not ok 41\n";
130print /x {3,4}/ ? "not ok 42\n" : "ok 42\n";
131print /^xxx {3,4}/ ? "not ok 43\n" : "ok 43\n";
352d5a3a 132
133$_ = "now is the time for all good men to come to.";
134@words = /(\w+)/g;
135print join(':',@words) eq "now:is:the:time:for:all:good:men:to:come:to"
136 ? "ok 44\n"
137 : "not ok 44\n";
138
139@words = ();
140while (/\w+/g) {
141 push(@words, $&);
142}
143print join(':',@words) eq "now:is:the:time:for:all:good:men:to:come:to"
144 ? "ok 45\n"
145 : "not ok 45\n";
146
147@words = ();
71be2cbc 148pos = 0;
352d5a3a 149while (/to/g) {
150 push(@words, $&);
151}
152print join(':',@words) eq "to:to"
153 ? "ok 46\n"
71be2cbc 154 : "not ok 46 `@words'\n";
352d5a3a 155
71be2cbc 156pos $_ = 0;
352d5a3a 157@words = /to/g;
158print join(':',@words) eq "to:to"
159 ? "ok 47\n"
71be2cbc 160 : "not ok 47 `@words'\n";
352d5a3a 161
162$_ = "abcdefghi";
163
164$pat1 = 'def';
165$pat2 = '^def';
166$pat3 = '.def.';
167$pat4 = 'abc';
168$pat5 = '^abc';
169$pat6 = 'abc$';
170$pat7 = 'ghi';
171$pat8 = '\w*ghi';
172$pat9 = 'ghi$';
173
174$t1=$t2=$t3=$t4=$t5=$t6=$t7=$t8=$t9=0;
175
176for $iter (1..5) {
177 $t1++ if /$pat1/o;
178 $t2++ if /$pat2/o;
179 $t3++ if /$pat3/o;
180 $t4++ if /$pat4/o;
181 $t5++ if /$pat5/o;
182 $t6++ if /$pat6/o;
183 $t7++ if /$pat7/o;
184 $t8++ if /$pat8/o;
185 $t9++ if /$pat9/o;
186}
187
188$x = "$t1$t2$t3$t4$t5$t6$t7$t8$t9";
189print $x eq '505550555' ? "ok 48\n" : "not ok 48 $x\n";
1462b684 190
191$xyz = 'xyz';
192print "abc" =~ /^abc$|$xyz/ ? "ok 49\n" : "not ok 49\n";
193
194# perl 4.009 says "unmatched ()"
195eval '"abc" =~ /a(bc$)|$xyz/; $result = "$&:$1"';
196print $@ eq "" ? "ok 50\n" : "not ok 50\n";
197print $result eq "abc:bc" ? "ok 51\n" : "not ok 51\n";
a0d0e21e 198
199
200$_="abcfooabcbar";
201$x=/abc/g;
202print $` eq "" ? "ok 52\n" : "not ok 52\n" if $x;
203$x=/abc/g;
204print $` eq "abcfoo" ? "ok 53\n" : "not ok 53\n" if $x;
205$x=/abc/g;
206print $x == 0 ? "ok 54\n" : "not ok 54\n";
71be2cbc 207pos = 0;
a0d0e21e 208$x=/ABC/gi;
209print $` eq "" ? "ok 55\n" : "not ok 55\n" if $x;
210$x=/ABC/gi;
211print $` eq "abcfoo" ? "ok 56\n" : "not ok 56\n" if $x;
212$x=/ABC/gi;
213print $x == 0 ? "ok 57\n" : "not ok 57\n";
71be2cbc 214pos = 0;
a0d0e21e 215$x=/abc/g;
216print $' eq "fooabcbar" ? "ok 58\n" : "not ok 58\n" if $x;
217$x=/abc/g;
218print $' eq "bar" ? "ok 59\n" : "not ok 59\n" if $x;
219$_ .= '';
220@x=/abc/g;
221print scalar @x == 2 ? "ok 60\n" : "not ok 60\n";
71be2cbc 222
223$_ = "abdc";
224pos $_ = 2;
c90c0ff4 225/\Gc/gc;
71be2cbc 226print "not " if (pos $_) != 2;
227print "ok 61\n";
c90c0ff4 228/\Gc/g;
229print "not " if defined pos $_;
230print "ok 62\n";
c277df42 231
232$out = 1;
233'abc' =~ m'a(?{ $out = 2 })b';
234print "not " if $out != 2;
235print "ok 63\n";
236
237$out = 1;
238'abc' =~ m'a(?{ $out = 3 })c';
239print "not " if $out != 1;
459f542a 240print "ok 64\n";
c277df42 241
242$_ = 'foobar1 bar2 foobar3 barfoobar5 foobar6';
243@out = /(?<!foo)bar./g;
244print "not " if "@out" ne 'bar2 barf';
245print "ok 65\n";
246
8d37f932 247# Tests which depend on REG_INFTY
248$reg_infty = defined $Config{reg_infty} ? $Config{reg_infty} : 32767;
249$reg_infty_m = $reg_infty - 1; $reg_infty_p = $reg_infty + 1;
250
251# As well as failing if the pattern matches do unexpected things, the
252# next three tests will fail if you should have picked up a lower-than-
253# default value for $reg_infty from Config.pm, but have not.
254
255undef $@;
256print "not " if eval q(('aaa' =~ /(a{1,$reg_infty_m})/)[0] ne 'aaa') || $@;
257print "ok 66\n";
258
259undef $@;
260print "not " if eval q(('a' x $reg_infty_m) !~ /a{$reg_infty_m}/) || $@;
261print "ok 67\n";
262
263undef $@;
264print "not " if eval q(('a' x ($reg_infty_m - 1)) =~ /a{$reg_infty_m}/) || $@;
265print "ok 68\n";
266
267undef $@;
268eval "'aaa' =~ /a{1,$reg_infty}/";
9baa0206 269print "not " if $@ !~ m%^\QQuantifier in {,} bigger than%;
8d37f932 270print "ok 69\n";
271
272eval "'aaa' =~ /a{1,$reg_infty_p}/";
273print "not "
9baa0206 274 if $@ !~ m%^\QQuantifier in {,} bigger than%;
8d37f932 275print "ok 70\n";
276undef $@;
277
278# Poke a couple more parse failures
279
280$context = 'x' x 256;
281eval qq("${context}y" =~ /(?<=$context)y/);
9baa0206 282print "not " if $@ !~ m%^\QLookbehind longer than 255 not%;
8d37f932 283print "ok 71\n";
284
b8c5462f 285# removed test
8d37f932 286print "ok 72\n";
287
c277df42 288# Long Monsters
8d37f932 289$test = 73;
c277df42 290for $l (125, 140, 250, 270, 300000, 30) { # Ordered to free memory
291 $a = 'a' x $l;
292 print "# length=$l\nnot " unless "ba$a=" =~ /a$a=/;
293 print "ok $test\n";
294 $test++;
73d6d589 295
c277df42 296 print "not " if "b$a=" =~ /a$a=/;
297 print "ok $test\n";
298 $test++;
299}
300
301# 20000 nodes, each taking 3 words per string, and 1 per branch
302$long_constant_len = join '|', 12120 .. 32645;
303$long_var_len = join '|', 8120 .. 28645;
304%ans = ( 'ax13876y25677lbc' => 1,
305 'ax13876y25677mcb' => 0, # not b.
306 'ax13876y35677nbc' => 0, # Num too big
307 'ax13876y25677y21378obc' => 1,
308 'ax13876y25677y21378zbc' => 0, # Not followed by [k-o]
309 'ax13876y25677y21378y21378kbc' => 1,
310 'ax13876y25677y21378y21378kcb' => 0, # Not b.
311 'ax13876y25677y21378y21378y21378kbc' => 0, # 5 runs
312 );
313
314for ( keys %ans ) {
73d6d589 315 print "# const-len `$_' not => $ans{$_}\nnot "
c277df42 316 if $ans{$_} xor /a(?=([yx]($long_constant_len)){2,4}[k-o]).*b./o;
317 print "ok $test\n";
318 $test++;
73d6d589 319 print "# var-len `$_' not => $ans{$_}\nnot "
c277df42 320 if $ans{$_} xor /a(?=([yx]($long_var_len)){2,4}[k-o]).*b./o;
321 print "ok $test\n";
322 $test++;
323}
324
325$_ = " a (bla()) and x(y b((l)u((e))) and b(l(e)e)e";
326$expect = "(bla()) ((l)u((e))) (l(e)e)";
327
73d6d589 328sub matchit {
cc6b7395 329 m/
c277df42 330 (
73d6d589 331 \(
c277df42 332 (?{ $c = 1 }) # Initialize
333 (?:
334 (?(?{ $c == 0 }) # PREVIOUS iteration was OK, stop the loop
335 (?!
336 ) # Fail: will unwind one iteration back
73d6d589 337 )
c277df42 338 (?:
339 [^()]+ # Match a big chunk
340 (?=
341 [()]
342 ) # Do not try to match subchunks
343 |
73d6d589 344 \(
c277df42 345 (?{ ++$c })
346 |
73d6d589 347 \)
c277df42 348 (?{ --$c })
349 )
350 )+ # This may not match with different subblocks
351 )
352 (?(?{ $c != 0 })
353 (?!
354 ) # Fail
355 ) # Otherwise the chunk 1 may succeed with $c>0
cc6b7395 356 /xg;
c277df42 357}
358
0f5d15d6 359@ans = ();
c277df42 360push @ans, $res while $res = matchit;
361
362print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne "1 1 1";
363print "ok $test\n";
364$test++;
365
366@ans = matchit;
367
368print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne $expect;
369print "ok $test\n";
370$test++;
371
96776eda 372print "not " unless "abc" =~ /^(??{"a"})b/;
373print "ok $test\n";
374$test++;
375
0f5d15d6 376my $matched;
14455d6c 377$matched = qr/\((?:(?>[^()]+)|(??{$matched}))*\)/;
0f5d15d6 378
379@ans = @ans1 = ();
380push(@ans, $res), push(@ans1, $&) while $res = m/$matched/g;
381
382print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne "1 1 1";
383print "ok $test\n";
384$test++;
385
386print "# ans1='@ans1'\n# expect='$expect'\nnot " if "@ans1" ne $expect;
387print "ok $test\n";
388$test++;
389
390@ans = m/$matched/g;
391
392print "# ans='@ans'\n# expect='$expect'\nnot " if "@ans" ne $expect;
393print "ok $test\n";
394$test++;
395
c277df42 396@ans = ('a/b' =~ m%(.*/)?(.*)%); # Stack may be bad
397print "not " if "@ans" ne 'a/ b';
398print "ok $test\n";
399$test++;
400
cc6b7395 401$code = '{$blah = 45}';
c277df42 402$blah = 12;
2cd61cdb 403eval { /(?$code)/ };
404print "not " unless $@ and $@ =~ /not allowed at runtime/ and $blah == 12;
e4d48cc9 405print "ok $test\n";
406$test++;
407
2cd61cdb 408for $code ('{$blah = 45}','=xx') {
409 $blah = 12;
410 $res = eval { "xx" =~ /(?$code)/o };
411 if ($code eq '=xx') {
412 print "#'$@','$res','$blah'\nnot " unless not $@ and $res;
413 } else {
73d6d589 414 print "#'$@','$res','$blah'\nnot " unless $@ and $@ =~ /not allowed at runtime/ and $blah == 12;
2cd61cdb 415 }
416 print "ok $test\n";
417 $test++;
418}
419
e4d48cc9 420$code = '{$blah = 45}';
421$blah = 12;
422eval "/(?$code)/";
cc6b7395 423print "not " if $blah != 45;
424print "ok $test\n";
425$test++;
426
427$blah = 12;
428/(?{$blah = 45})/;
c277df42 429print "not " if $blah != 45;
430print "ok $test\n";
431$test++;
432
74d6a13a 433$x = 'banana';
434$x =~ /.a/g;
435print "not " unless pos($x) == 2;
436print "ok $test\n";
437$test++;
438
439$x =~ /.z/gc;
440print "not " unless pos($x) == 2;
441print "ok $test\n";
442$test++;
443
444sub f {
445 my $p = $_[0];
446 return $p;
447}
448
449$x =~ /.a/g;
450print "not " unless f(pos($x)) == 4;
451print "ok $test\n";
452$test++;
4599a1de 453
ce862d02 454$x = $^R = 67;
455'foot' =~ /foo(?{$x = 12; 75})[t]/;
456print "not " unless $^R eq '75';
457print "ok $test\n";
458$test++;
459
460$x = $^R = 67;
461'foot' =~ /foo(?{$x = 12; 75})[xy]/;
462print "not " unless $^R eq '67' and $x eq '12';
463print "ok $test\n";
464$test++;
465
466$x = $^R = 67;
467'foot' =~ /foo(?{ $^R + 12 })((?{ $x = 12; $^R + 17 })[xy])?/;
468print "not " unless $^R eq '79' and $x eq '12';
469print "ok $test\n";
470$test++;
471
8782bef2 472print "not " unless qr/\b\v$/i eq '(?i-xsm:\bv$)';
473print "ok $test\n";
474$test++;
475
476print "not " unless qr/\b\v$/s eq '(?s-xim:\bv$)';
477print "ok $test\n";
478$test++;
479
480print "not " unless qr/\b\v$/m eq '(?m-xis:\bv$)';
481print "ok $test\n";
482$test++;
483
484print "not " unless qr/\b\v$/x eq '(?x-ism:\bv$)';
485print "ok $test\n";
486$test++;
487
488print "not " unless qr/\b\v$/xism eq '(?msix:\bv$)';
489print "ok $test\n";
490$test++;
491
492print "not " unless qr/\b\v$/ eq '(?-xism:\bv$)';
97197631 493print "ok $test\n";
494$test++;
495
7e5428c5 496$_ = 'xabcx';
497foreach $ans ('', 'c') {
498 /(?<=(?=a)..)((?=c)|.)/g;
02db2b7b 499 print "# \$1 ='$1'\n# \$ans='$ans'\nnot " unless $1 eq $ans;
7e5428c5 500 print "ok $test\n";
501 $test++;
502}
503
504$_ = 'a';
505foreach $ans ('', 'a', '') {
506 /^|a|$/g;
02db2b7b 507 print "# \$& ='$&'\n# \$ans='$ans'\nnot " unless $& eq $ans;
7e5428c5 508 print "ok $test\n";
509 $test++;
510}
511
09f25ae4 512sub prefixify {
73d6d589 513 my($v,$a,$b,$res) = @_;
514 $v =~ s/\Q$a\E/$b/;
515 print "not " unless $res eq $v;
09f25ae4 516 print "ok $test\n";
517 $test++;
518}
519prefixify('/a/b/lib/arch', "/a/b/lib", 'X/lib', 'X/lib/arch');
520prefixify('/a/b/man/arch', "/a/b/man", 'X/man', 'X/man/arch');
521
522$_ = 'var="foo"';
523/(\")/;
524print "not " unless $1 and /$1/;
525print "ok $test\n";
526$test++;
527
73d6d589 528$a=qr/(?{++$b})/;
2cd61cdb 529$b = 7;
73d6d589 530/$a$a/;
531print "not " unless $b eq '9';
2cd61cdb 532print "ok $test\n";
533$test++;
534
73d6d589 535$c="$a";
536/$a$a/;
537print "not " unless $b eq '11';
2cd61cdb 538print "ok $test\n";
539$test++;
540
541{
73d6d589 542 use re "eval";
543 /$a$c$a/;
544 print "not " unless $b eq '14';
2cd61cdb 545 print "ok $test\n";
546 $test++;
547
160cb429 548 local $lex_a = 2;
549 my $lex_a = 43;
550 my $lex_b = 17;
551 my $lex_c = 27;
552 my $lex_res = ($lex_b =~ qr/$lex_b(?{ $lex_c = $lex_a++ })/);
553 print "not " unless $lex_res eq '1';
554 print "ok $test\n";
555 $test++;
556 print "not " unless $lex_a eq '44';
557 print "ok $test\n";
558 $test++;
559 print "not " unless $lex_c eq '43';
560 print "ok $test\n";
561 $test++;
562
563
73d6d589 564 no re "eval";
2cd61cdb 565 $match = eval { /$a$c$a/ };
73d6d589 566 print "not "
2cd61cdb 567 unless $b eq '14' and $@ =~ /Eval-group not allowed/ and not $match;
568 print "ok $test\n";
569 $test++;
570}
cbce877f 571
572{
160cb429 573 local $lex_a = 2;
574 my $lex_a = 43;
575 my $lex_b = 17;
576 my $lex_c = 27;
577 my $lex_res = ($lex_b =~ qr/17(?{ $lex_c = $lex_a++ })/);
578 print "not " unless $lex_res eq '1';
579 print "ok $test\n";
580 $test++;
581 print "not " unless $lex_a eq '44';
582 print "ok $test\n";
583 $test++;
584 print "not " unless $lex_c eq '43';
585 print "ok $test\n";
586 $test++;
587}
588
589{
cbce877f 590 package aa;
591 $c = 2;
592 $::c = 3;
593 '' =~ /(?{ $c = 4 })/;
594 print "not " unless $c == 4;
595}
596print "ok $test\n";
597$test++;
598print "not " unless $c == 3;
599print "ok $test\n";
73d6d589 600$test++;
601
4599a1de 602sub must_warn_pat {
603 my $warn_pat = shift;
604 return sub { print "not " unless $_[0] =~ /$warn_pat/ }
605}
606
607sub must_warn {
608 my ($warn_pat, $code) = @_;
9f1b1f2d 609 local %SIG;
610 eval 'BEGIN { use warnings; $SIG{__WARN__} = $warn_pat };' . $code;
4599a1de 611 print "ok $test\n";
612 $test++;
613}
614
615
616sub make_must_warn {
617 my $warn_pat = shift;
618 return sub { must_warn(must_warn_pat($warn_pat)) }
619}
620
621my $for_future = make_must_warn('reserved for future extensions');
622
623&$for_future('q(a:[b]:) =~ /[x[:foo:]]/');
9baa0206 624
625#&$for_future('q(a=[b]=) =~ /[x[=foo=]]/');
626print "ok $test\n"; $test++; # now a fatal croak
627
628#&$for_future('q(a.[b].) =~ /[x[.foo.]]/');
629print "ok $test\n"; $test++; # now a fatal croak
f7e33566 630
631# test if failure of patterns returns empty list
632$_ = 'aaa';
633@_ = /bbb/;
634print "not " if @_;
635print "ok $test\n";
636$test++;
637
638@_ = /bbb/g;
639print "not " if @_;
640print "ok $test\n";
641$test++;
642
643@_ = /(bbb)/;
644print "not " if @_;
645print "ok $test\n";
646$test++;
647
648@_ = /(bbb)/g;
649print "not " if @_;
650print "ok $test\n";
651$test++;
652
6cef1e77 653/a(?=.$)/;
654print "not " if $#+ != 0 or $#- != 0;
655print "ok $test\n";
656$test++;
657
658print "not " if $+[0] != 2 or $-[0] != 1;
659print "ok $test\n";
660$test++;
661
73d6d589 662print "not "
6cef1e77 663 if defined $+[1] or defined $-[1] or defined $+[2] or defined $-[2];
664print "ok $test\n";
665$test++;
666
667/a(a)(a)/;
668print "not " if $#+ != 2 or $#- != 2;
669print "ok $test\n";
670$test++;
671
672print "not " if $+[0] != 3 or $-[0] != 0;
673print "ok $test\n";
674$test++;
675
676print "not " if $+[1] != 2 or $-[1] != 1;
677print "ok $test\n";
678$test++;
679
680print "not " if $+[2] != 3 or $-[2] != 2;
681print "ok $test\n";
682$test++;
683
73d6d589 684print "not "
6cef1e77 685 if defined $+[3] or defined $-[3] or defined $+[4] or defined $-[4];
686print "ok $test\n";
687$test++;
688
689/.(a)(b)?(a)/;
690print "not " if $#+ != 3 or $#- != 3;
691print "ok $test\n";
692$test++;
693
694print "not " if $+[0] != 3 or $-[0] != 0;
695print "ok $test\n";
696$test++;
697
698print "not " if $+[1] != 2 or $-[1] != 1;
699print "ok $test\n";
700$test++;
701
702print "not " if $+[3] != 3 or $-[3] != 2;
703print "ok $test\n";
704$test++;
705
73d6d589 706print "not "
6cef1e77 707 if defined $+[2] or defined $-[2] or defined $+[4] or defined $-[4];
708print "ok $test\n";
709$test++;
710
711/.(a)/;
712print "not " if $#+ != 1 or $#- != 1;
713print "ok $test\n";
714$test++;
715
716print "not " if $+[0] != 2 or $-[0] != 0;
717print "ok $test\n";
718$test++;
719
720print "not " if $+[1] != 2 or $-[1] != 1;
721print "ok $test\n";
722$test++;
723
73d6d589 724print "not "
6cef1e77 725 if defined $+[2] or defined $-[2] or defined $+[3] or defined $-[3];
726print "ok $test\n";
727$test++;
728
03a27ae7 729eval { $+[0] = 13; };
73d6d589 730print "not "
03a27ae7 731 if $@ !~ /^Modification of a read-only value attempted/;
732print "ok $test\n";
733$test++;
734
735eval { $-[0] = 13; };
73d6d589 736print "not "
03a27ae7 737 if $@ !~ /^Modification of a read-only value attempted/;
738print "ok $test\n";
739$test++;
740
741eval { @+ = (7, 6, 5); };
73d6d589 742print "not "
03a27ae7 743 if $@ !~ /^Modification of a read-only value attempted/;
744print "ok $test\n";
745$test++;
746
747eval { @- = qw(foo bar); };
73d6d589 748print "not "
03a27ae7 749 if $@ !~ /^Modification of a read-only value attempted/;
750print "ok $test\n";
751$test++;
752
8f580fb8 753/.(a)(ba*)?/;
754print "#$#-..$#+\nnot " if $#+ != 2 or $#- != 1;
755print "ok $test\n";
756$test++;
757
ad94a511 758$_ = 'aaa';
759pos = 1;
760@a = /\Ga/g;
761print "not " unless "@a" eq "a a";
762print "ok $test\n";
763$test++;
764
22e551b9 765$str = 'abcde';
766pos $str = 2;
767
768print "not " if $str =~ /^\G/;
769print "ok $test\n";
770$test++;
771
772print "not " if $str =~ /^.\G/;
773print "ok $test\n";
774$test++;
775
776print "not " unless $str =~ /^..\G/;
777print "ok $test\n";
778$test++;
779
780print "not " if $str =~ /^...\G/;
781print "ok $test\n";
782$test++;
783
784print "not " unless $str =~ /.\G./ and $& eq 'bc';
785print "ok $test\n";
786$test++;
787
788print "not " unless $str =~ /\G../ and $& eq 'cd';
789print "ok $test\n";
790$test++;
791
9661b544 792undef $foo; undef $bar;
793print "#'$str','$foo','$bar'\nnot "
73d6d589 794 unless $str =~ /b(?{$foo = $_; $bar = pos})c/
9661b544 795 and $foo eq 'abcde' and $bar eq 2;
796print "ok $test\n";
797$test++;
798
799undef $foo; undef $bar;
800pos $str = undef;
801print "#'$str','$foo','$bar'\nnot "
73d6d589 802 unless $str =~ /b(?{$foo = $_; $bar = pos})c/g
9661b544 803 and $foo eq 'abcde' and $bar eq 2 and pos $str eq 3;
804print "ok $test\n";
805$test++;
806
807$_ = $str;
808
809undef $foo; undef $bar;
810print "#'$str','$foo','$bar'\nnot "
73d6d589 811 unless /b(?{$foo = $_; $bar = pos})c/
9661b544 812 and $foo eq 'abcde' and $bar eq 2;
813print "ok $test\n";
814$test++;
815
816undef $foo; undef $bar;
817print "#'$str','$foo','$bar'\nnot "
73d6d589 818 unless /b(?{$foo = $_; $bar = pos})c/g
9661b544 819 and $foo eq 'abcde' and $bar eq 2 and pos eq 3;
820print "ok $test\n";
821$test++;
822
823undef $foo; undef $bar;
824pos = undef;
8251 while /b(?{$foo = $_; $bar = pos})c/g;
826print "#'$str','$foo','$bar'\nnot "
827 unless $foo eq 'abcde' and $bar eq 2 and not defined pos;
828print "ok $test\n";
829$test++;
830
831undef $foo; undef $bar;
832$_ = 'abcde|abcde';
833print "#'$str','$foo','$bar','$_'\nnot "
73d6d589 834 unless s/b(?{$foo = $_; $bar = pos})c/x/g and $foo eq 'abcde|abcde'
9661b544 835 and $bar eq 8 and $_ eq 'axde|axde';
836print "ok $test\n";
837$test++;
838
5c5e4c24 839@res = ();
840# List context:
841$_ = 'abcde|abcde';
842@dummy = /([ace]).(?{push @res, $1,$2})([ce])(?{push @res, $1,$2})/g;
843@res = map {defined $_ ? "'$_'" : 'undef'} @res;
844$res = "@res";
845print "#'@res' '$_'\nnot "
846 unless "@res" eq "'a' undef 'a' 'c' 'e' undef 'a' undef 'a' 'c'";
847print "ok $test\n";
848$test++;
849
850@res = ();
851@dummy = /([ace]).(?{push @res, $`,$&,$'})([ce])(?{push @res, $`,$&,$'})/g;
852@res = map {defined $_ ? "'$_'" : 'undef'} @res;
853$res = "@res";
854print "#'@res' '$_'\nnot "
855 unless "@res" eq
856 "'' 'ab' 'cde|abcde' " .
857 "'' 'abc' 'de|abcde' " .
858 "'abcd' 'e|' 'abcde' " .
859 "'abcde|' 'ab' 'cde' " .
860 "'abcde|' 'abc' 'de'" ;
861print "ok $test\n";
862$test++;
863
b7a35066 864#Some more \G anchor checks
865$foo='aabbccddeeffgg';
866
867pos($foo)=1;
868
869$foo=~/.\G(..)/g;
870print "not " unless($1 eq 'ab');
871print "ok $test\n";
872$test++;
873
874pos($foo) += 1;
875$foo=~/.\G(..)/g;
876print "not " unless($1 eq 'cc');
877print "ok $test\n";
878$test++;
879
880pos($foo) += 1;
881$foo=~/.\G(..)/g;
882print "not " unless($1 eq 'de');
883print "ok $test\n";
884$test++;
885
0ef3e39e 886print "not " unless $foo =~ /\Gef/g;
887print "ok $test\n";
888$test++;
889
b7a35066 890undef pos $foo;
891
892$foo=~/\G(..)/g;
893print "not " unless($1 eq 'aa');
894print "ok $test\n";
895$test++;
896
897$foo=~/\G(..)/g;
898print "not " unless($1 eq 'bb');
899print "ok $test\n";
900$test++;
901
902pos($foo)=5;
903$foo=~/\G(..)/g;
904print "not " unless($1 eq 'cd');
905print "ok $test\n";
906$test++;
907
73d6d589 908$_='123x123';
e60df1fa 909@res = /(\d*|x)/g;
910print "not " unless('123||x|123|' eq join '|', @res);
911print "ok $test\n";
912$test++;
913
9d080a66 914# see if matching against temporaries (created via pp_helem()) is safe
915{ foo => "ok $test\n".$^X }->{foo} =~ /^(.*)\n/g;
916print "$1\n";
917$test++;
918
cf93c79d 919# See if $i work inside (?{}) in the presense of saved substrings and
920# changing $_
921@a = qw(foo bar);
922@b = ();
923s/(\w)(?{push @b, $1})/,$1,/g for @a;
924
925print "# \@b='@b', expect 'f o o b a r'\nnot " unless("@b" eq "f o o b a r");
926print "ok $test\n";
927$test++;
928
929print "not " unless("@a" eq ",f,,o,,o, ,b,,a,,r,");
930print "ok $test\n";
931$test++;
932
2c914db6 933$brackets = qr{
14455d6c 934 { (?> [^{}]+ | (??{ $brackets }) )* }
2c914db6 935 }x;
936
937"{{}" =~ $brackets;
938print "ok $test\n"; # Did we survive?
939$test++;
940
941"something { long { and } hairy" =~ $brackets;
942print "ok $test\n"; # Did we survive?
943$test++;
944
14455d6c 945"something { long { and } hairy" =~ m/((??{ $brackets }))/;
2c914db6 946print "not " unless $1 eq "{ and }";
947print "ok $test\n";
948$test++;
949
30944b6d 950$_ = "a-a\nxbb";
951pos=1;
952m/^-.*bb/mg and print "not ";
953print "ok $test\n";
954$test++;
30382c73 955
956$text = "aaXbXcc";
957pos($text)=0;
958$text =~ /\GXb*X/g and print 'not ';
959print "ok $test\n";
960$test++;
3cf5c195 961
962$text = "xA\n" x 500;
963$text =~ /^\s*A/m and print 'not ';
964print "ok $test\n";
965$test++;
d506a20d 966
967$text = "abc dbf";
968@res = ($text =~ /.*?(b).*?\b/g);
969"@res" eq 'b b' or print 'not ';
970print "ok $test\n";
971$test++;
972
9442cb0e 973@a = map chr,0..255;
aeaf5620 974
975@b = grep(/\S/,@a);
976@c = grep(/[^\s]/,@a);
977print "not " if "@b" ne "@c";
9442cb0e 978print "ok $test\n";
979$test++;
980
aeaf5620 981@b = grep(/\S/,@a);
982@c = grep(/[\S]/,@a);
983print "not " if "@b" ne "@c";
9442cb0e 984print "ok $test\n";
985$test++;
986
aeaf5620 987@b = grep(/\s/,@a);
988@c = grep(/[^\S]/,@a);
989print "not " if "@b" ne "@c";
9442cb0e 990print "ok $test\n";
991$test++;
992
aeaf5620 993@b = grep(/\s/,@a);
994@c = grep(/[\s]/,@a);
995print "not " if "@b" ne "@c";
9442cb0e 996print "ok $test\n";
997$test++;
998
aeaf5620 999@b = grep(/\D/,@a);
1000@c = grep(/[^\d]/,@a);
1001print "not " if "@b" ne "@c";
9442cb0e 1002print "ok $test\n";
1003$test++;
1004
aeaf5620 1005@b = grep(/\D/,@a);
1006@c = grep(/[\D]/,@a);
1007print "not " if "@b" ne "@c";
9442cb0e 1008print "ok $test\n";
1009$test++;
1010
aeaf5620 1011@b = grep(/\d/,@a);
1012@c = grep(/[^\D]/,@a);
1013print "not " if "@b" ne "@c";
9442cb0e 1014print "ok $test\n";
1015$test++;
1016
aeaf5620 1017@b = grep(/\d/,@a);
1018@c = grep(/[\d]/,@a);
1019print "not " if "@b" ne "@c";
9442cb0e 1020print "ok $test\n";
1021$test++;
1022
aeaf5620 1023@b = grep(/\W/,@a);
1024@c = grep(/[^\w]/,@a);
1025print "not " if "@b" ne "@c";
9442cb0e 1026print "ok $test\n";
1027$test++;
1028
aeaf5620 1029@b = grep(/\W/,@a);
1030@c = grep(/[\W]/,@a);
1031print "not " if "@b" ne "@c";
9442cb0e 1032print "ok $test\n";
1033$test++;
1034
aeaf5620 1035@b = grep(/\w/,@a);
1036@c = grep(/[^\W]/,@a);
1037print "not " if "@b" ne "@c";
9442cb0e 1038print "ok $test\n";
1039$test++;
1040
aeaf5620 1041@b = grep(/\w/,@a);
1042@c = grep(/[\w]/,@a);
1043print "not " if "@b" ne "@c";
9442cb0e 1044print "ok $test\n";
1045$test++;
1aeab75a 1046
1047# see if backtracking optimization works correctly
1048"\n\n" =~ /\n $ \n/x or print "not ";
1049print "ok $test\n";
1050$test++;
1051
1052"\n\n" =~ /\n* $ \n/x or print "not ";
1053print "ok $test\n";
1054$test++;
1055
1056"\n\n" =~ /\n+ $ \n/x or print "not ";
1057print "ok $test\n";
1058$test++;
05b4157f 1059
1060[] =~ /^ARRAY/ or print "# [] \nnot ";
1061print "ok $test\n";
1062$test++;
1063
1064eval << 'EOE';
1065{
1066 package S;
1067 use overload '""' => sub { 'Object S' };
1068 sub new { bless [] }
1069}
1070$a = 'S'->new;
1071EOE
1072
1073$a and $a =~ /^Object\sS/ or print "# '$a' \nnot ";
1074print "ok $test\n";
1075$test++;
815d35b9 1076
1077# test result of match used as match (!)
1078'a1b' =~ ('xyz' =~ /y/) and $` eq 'a' or print "not ";
1079print "ok $test\n";
1080$test++;
1081
1082'a1b' =~ ('xyz' =~ /t/) and $` eq 'a' or print "not ";
1083print "ok $test\n";
1084$test++;
5e39e1e5 1085
1086$w = 0;
1087{
1088 local $SIG{__WARN__} = sub { $w = 1 };
1089 local $^W = 1;
1090 $w = 1 if ("1\n" x 102) =~ /^\s*\n/m;
1091}
1092print $w ? "not " : "", "ok $test\n";
1093$test++;
aaa51d5e 1094
1095my %space = ( spc => " ",
1096 tab => "\t",
1097 cr => "\r",
1098 lf => "\n",
1099 ff => "\f",
75369ccb 1100# There's no \v but the vertical tabulator seems miraculously
1101# be 11 both in ASCII and EBCDIC.
aaa51d5e 1102 vt => chr(11),
1103 false => "space" );
1104
1105my @space0 = sort grep { $space{$_} =~ /\s/ } keys %space;
1106my @space1 = sort grep { $space{$_} =~ /[[:space:]]/ } keys %space;
1107my @space2 = sort grep { $space{$_} =~ /[[:blank:]]/ } keys %space;
1108
1109print "not " unless "@space0" eq "cr ff lf spc tab";
3bec3564 1110print "ok $test # @space0\n";
aaa51d5e 1111$test++;
1112
1113print "not " unless "@space1" eq "cr ff lf spc tab vt";
3bec3564 1114print "ok $test # @space1\n";
aaa51d5e 1115$test++;
1116
1117print "not " unless "@space2" eq "spc tab";
3bec3564 1118print "ok $test # @space2\n";
aaa51d5e 1119$test++;
73d6d589 1120
a1933d95 1121# bugid 20001021.005 - this caused a SEGV
1122print "not " unless undef =~ /^([^\/]*)(.*)$/;
1123print "ok $test\n";
1124$test++;
b91bb191 1125
1126# bugid 20000731.001
1127
1128print "not " unless "A \x{263a} B z C" =~ /A . B (??{ "z" }) C/;
1129print "ok $test\n";
1130$test++;
1131
5ae032e5 1132my $ordA = ord('A');
1133
3baa4c62 1134$_ = "a\x{100}b";
1135if (/(.)(\C)(\C)(.)/) {
1136 print "ok 232\n";
1137 if ($1 eq "a") {
1138 print "ok 233\n";
1139 } else {
1140 print "not ok 233\n";
1141 }
5ae032e5 1142 if ($ordA == 65) { # ASCII (or equivalent), should be UTF-8
1143 if ($2 eq "\xC4") {
1144 print "ok 234\n";
1145 } else {
1146 print "not ok 234\n";
1147 }
1148 if ($3 eq "\x80") {
1149 print "ok 235\n";
1150 } else {
1151 print "not ok 235\n";
1152 }
1153 } elsif ($ordA == 193) { # EBCDIC (or equivalent), should be UTF-EBCDIC
1154 if ($2 eq "\x8C") {
1155 print "ok 234\n";
1156 } else {
1157 print "not ok 234\n";
1158 }
1159 if ($3 eq "\x41") {
1160 print "ok 235\n";
1161 } else {
1162 print "not ok 235\n";
1163 }
3baa4c62 1164 } else {
5ae032e5 1165 for (234..235) {
1166 print "not ok $_ # ord('A') == $ordA\n";
1167 }
3baa4c62 1168 }
1169 if ($4 eq "b") {
1170 print "ok 236\n";
1171 } else {
1172 print "not ok 236\n";
1173 }
1174} else {
1175 for (232..236) {
1176 print "not ok $_\n";
1177 }
1178}
1179$_ = "\x{100}";
1180if (/(\C)/g) {
1181 print "ok 237\n";
73d6d589 1182 # currently \C are still tagged as UTF-8
5ae032e5 1183 if ($ordA == 65) {
1184 if ($1 eq "\xC4") {
1185 print "ok 238\n";
1186 } else {
1187 print "not ok 238\n";
1188 }
1189 } elsif ($ordA == 193) {
1190 if ($1 eq "\x8C") {
1191 print "ok 238\n";
1192 } else {
1193 print "not ok 238\n";
1194 }
3baa4c62 1195 } else {
5ae032e5 1196 print "not ok 238 # ord('A') == $ordA\n";
3baa4c62 1197 }
1198} else {
1199 for (237..238) {
1200 print "not ok $_\n";
1201 }
1202}
1203if (/(\C)/g) {
1204 print "ok 239\n";
73d6d589 1205 # currently \C are still tagged as UTF-8
5ae032e5 1206 if ($ordA == 65) {
1207 if ($1 eq "\x80") {
1208 print "ok 240\n";
1209 } else {
1210 print "not ok 240\n";
1211 }
1212 } elsif ($ordA == 193) {
1213 if ($1 eq "\x41") {
1214 print "ok 240\n";
1215 } else {
1216 print "not ok 240\n";
1217 }
3baa4c62 1218 } else {
5ae032e5 1219 print "not ok 240 # ord('A') == $ordA\n";
3baa4c62 1220 }
1221} else {
1222 for (239..240) {
1223 print "not ok $_\n";
1224 }
1225}
b485d051 1226
db615365 1227{
1228 # japhy -- added 03/03/2001
1229 () = (my $str = "abc") =~ /(...)/;
1230 $str = "def";
1231 print "not " if $1 ne "abc";
fd291da9 1232 print "ok 241\n";
1233}
1234
1235# The 242 and 243 go with the 244 and 245.
1236# The trick is that in EBCDIC the explicit numeric range should match
1237# (as also in non-EBCDIC) but the explicit alphabetic range should not match.
1238
1239if ("\x8e" =~ /[\x89-\x91]/) {
1240 print "ok 242\n";
1241} else {
1242 print "not ok 242\n";
1243}
1244
1245if ("\xce" =~ /[\xc9-\xd1]/) {
db615365 1246 print "ok 243\n";
fd291da9 1247} else {
1248 print "not ok 243\n";
1249}
1250
1251# In most places these tests would succeed since \x8e does not
1252# in most character sets match 'i' or 'j' nor would \xce match
1253# 'I' or 'J', but strictly speaking these tests are here for
1254# the good of EBCDIC, so let's test these only there.
1255if (ord('i') == 0x89 && ord('J') == 0xd1) { # EBCDIC
1256 if ("\x8e" !~ /[i-j]/) {
1257 print "ok 244\n";
1258 } else {
1259 print "not ok 244\n";
1260 }
1261 if ("\xce" !~ /[I-J]/) {
1262 print "ok 245\n";
1263 } else {
1264 print "not ok 245\n";
1265 }
1266} else {
1267 for (244..245) {
60425c38 1268 print "ok $_ # Skip: only in EBCDIC\n";
fd291da9 1269 }
db615365 1270}
4765795a 1271
1272print "not " unless "\x{ab}" =~ /\x{ab}/;
1273print "ok 246\n";
1274
1275print "not " unless "\x{abcd}" =~ /\x{abcd}/;
1276print "ok 247\n";
1277
1278{
f9969324 1279 # bug id 20001008.001
1280
1281 my $test = 248;
1282 my @x = ("stra\337e 138","stra\337e 138");
1283 for (@x) {
1284 s/(\d+)\s*([\w\-]+)/$1 . uc $2/e;
1285 my($latin) = /^(.+)(?:\s+\d)/;
1286 print $latin eq "stra\337e" ? "ok $test\n" : # 248,249
1287 "#latin[$latin]\nnot ok $test\n";
1288 $test++;
1289 $latin =~ s/stra\337e/straße/; # \303\237 after the 2nd a
1290 use utf8; # needed for the raw UTF-8
1291 $latin =~ s!(s)tr(?:aß|s+e)!$1tr.!; # \303\237 after the a
4765795a 1292 }
1293}
1294
1295{
1296 print "not " unless "ba\xd4c" =~ /([a\xd4]+)/ && $1 eq "a\xd4";
1297 print "ok 250\n";
1298
1299 print "not " unless "ba\xd4c" =~ /([a\xd4]+)/ && $1 eq "a\x{d4}";
1300 print "ok 251\n";
1301
1302 print "not " unless "ba\x{d4}c" =~ /([a\xd4]+)/ && $1 eq "a\x{d4}";
1303 print "ok 252\n";
1304
1305 print "not " unless "ba\x{d4}c" =~ /([a\xd4]+)/ && $1 eq "a\xd4";
1306 print "ok 253\n";
1307
1308 print "not " unless "ba\xd4c" =~ /([a\x{d4}]+)/ && $1 eq "a\xd4";
1309 print "ok 254\n";
1310
1311 print "not " unless "ba\xd4c" =~ /([a\x{d4}]+)/ && $1 eq "a\x{d4}";
1312 print "ok 255\n";
1313
1314 print "not " unless "ba\x{d4}c" =~ /([a\x{d4}]+)/ && $1 eq "a\x{d4}";
1315 print "ok 256\n";
1316
1317 print "not " unless "ba\x{d4}c" =~ /([a\x{d4}]+)/ && $1 eq "a\xd4";
1318 print "ok 257\n";
1319}
1320
1321{
1322 # the first half of 20001028.003
1323
1324 my $X = chr(1448);
1325 my ($Y) = $X =~ /(.*)/;
1326 print "not " unless $Y eq v1448 && length($Y) == 1;
1327 print "ok 258\n";
1328}
1329
1330{
1331 # 20001108.001
1332
1333 my $X = "Szab\x{f3},Bal\x{e1}zs";
1334 my $Y = $X;
1335 $Y =~ s/(B)/$1/ for 0..3;
1336 print "not " unless $Y eq $X && $X eq "Szab\x{f3},Bal\x{e1}zs";
1337 print "ok 259\n";
1338}
1339
1340{
1341 # the second half of 20001028.003
1342
3568d838 1343 my $X = '';
4765795a 1344 $X =~ s/^/chr(1488)/e;
1345 print "not " unless length $X == 1 && ord($X) == 1488;
1346 print "ok 260\n";
1347}
1348
1349{
1350 # 20000517.001
1351
1352 my $x = "\x{100}A";
1353
1354 $x =~ s/A/B/;
1355
1356 print "not " unless $x eq "\x{100}B" && length($x) == 2;
1357 print "ok 261\n";
1358}
1359
1360{
1361 # bug id 20001230.002
1362
1363 print "not " unless "École" =~ /^\C\C(.)/ && $1 eq 'c';
1364 print "ok 262\n";
1365
1366 print "not " unless "École" =~ /^\C\C(c)/;
1367 print "ok 263\n";
1368}
1369
1370{
1371 my $test = 264; # till 575
1372
1373 use charnames ':full';
1374
1375 # This is far from complete testing, there are dozens of character
1376 # classes in Unicode. The mixing of literals and \N{...} is
1377 # intentional so that in non-Latin-1 places we test the native
1378 # characters, not the Unicode code points.
1379
1380 my %s = (
1381 "a" => 'Ll',
1382 "\N{CYRILLIC SMALL LETTER A}" => 'Ll',
1383 "A" => 'Lu',
1384 "\N{GREEK CAPITAL LETTER ALPHA}" => 'Lu',
1385 "\N{HIRAGANA LETTER SMALL A}" => 'Lo',
1386 "\N{COMBINING GRAVE ACCENT}" => 'Mn',
1387 "0" => 'Nd',
1388 "\N{ARABIC-INDIC DIGIT ZERO}" => 'Nd',
1389 "_" => 'N',
1390 "!" => 'P',
1391 " " => 'Zs',
1392 "\0" => 'Cc',
1393 );
73d6d589 1394
3568d838 1395 for my $char (map { s/^\S+ //; $_ }
1396 sort map { sprintf("%06x", ord($_))." $_" } keys %s) {
4765795a 1397 my $class = $s{$char};
3568d838 1398 my $code = sprintf("%06x", ord($char));
1399 printf "#\n# 0x$code\n#\n";
4765795a 1400 print "# IsAlpha\n";
1401 if ($class =~ /^[LM]/) {
1402 print "not " unless $char =~ /\p{IsAlpha}/;
1403 print "ok $test\n"; $test++;
1404 print "not " if $char =~ /\P{IsAlpha}/;
1405 print "ok $test\n"; $test++;
1406 } else {
1407 print "not " if $char =~ /\p{IsAlpha}/;
1408 print "ok $test\n"; $test++;
1409 print "not " unless $char =~ /\P{IsAlpha}/;
1410 print "ok $test\n"; $test++;
1411 }
1412 print "# IsAlnum\n";
1413 if ($class =~ /^[LMN]/ && $char ne "_") {
1414 print "not " unless $char =~ /\p{IsAlnum}/;
1415 print "ok $test\n"; $test++;
1416 print "not " if $char =~ /\P{IsAlnum}/;
1417 print "ok $test\n"; $test++;
1418 } else {
1419 print "not " if $char =~ /\p{IsAlnum}/;
1420 print "ok $test\n"; $test++;
1421 print "not " unless $char =~ /\P{IsAlnum}/;
1422 print "ok $test\n"; $test++;
1423 }
1424 print "# IsASCII\n";
24817587 1425 if (ord("A") == 193) {
1426 print "ok $test # Skip: in EBCDIC\n"; $test++;
1427 print "ok $test # Skip: in EBCDIC\n"; $test++;
4765795a 1428 } else {
24817587 1429 if ($code le '00007f') {
1430 print "not " unless $char =~ /\p{IsASCII}/;
1431 print "ok $test\n"; $test++;
1432 print "not " if $char =~ /\P{IsASCII}/;
1433 print "ok $test\n"; $test++;
1434 } else {
1435 print "not " if $char =~ /\p{IsASCII}/;
1436 print "ok $test\n"; $test++;
1437 print "not " unless $char =~ /\P{IsASCII}/;
1438 print "ok $test\n"; $test++;
1439 }
4765795a 1440 }
1441 print "# IsCntrl\n";
1442 if ($class =~ /^C/) {
1443 print "not " unless $char =~ /\p{IsCntrl}/;
1444 print "ok $test\n"; $test++;
1445 print "not " if $char =~ /\P{IsCntrl}/;
1446 print "ok $test\n"; $test++;
1447 } else {
1448 print "not " if $char =~ /\p{IsCntrl}/;
1449 print "ok $test\n"; $test++;
1450 print "not " unless $char =~ /\P{IsCntrl}/;
1451 print "ok $test\n"; $test++;
1452 }
1453 print "# IsBlank\n";
1454 if ($class =~ /^Z[lp]/ || $char eq " ") {
1455 print "not " unless $char =~ /\p{IsBlank}/;
1456 print "ok $test\n"; $test++;
1457 print "not " if $char =~ /\P{IsBlank}/;
1458 print "ok $test\n"; $test++;
1459 } else {
1460 print "not " if $char =~ /\p{IsBlank}/;
1461 print "ok $test\n"; $test++;
1462 print "not " unless $char =~ /\P{IsBlank}/;
1463 print "ok $test\n"; $test++;
1464 }
1465 print "# IsDigit\n";
1466 if ($class =~ /^Nd$/) {
1467 print "not " unless $char =~ /\p{IsDigit}/;
1468 print "ok $test\n"; $test++;
1469 print "not " if $char =~ /\P{IsDigit}/;
1470 print "ok $test\n"; $test++;
1471 } else {
1472 print "not " if $char =~ /\p{IsDigit}/;
1473 print "ok $test\n"; $test++;
1474 print "not " unless $char =~ /\P{IsDigit}/;
1475 print "ok $test\n"; $test++;
1476 }
1477 print "# IsGraph\n";
1478 if ($class =~ /^([LMNPS])|Co/) {
1479 print "not " unless $char =~ /\p{IsGraph}/;
1480 print "ok $test\n"; $test++;
1481 print "not " if $char =~ /\P{IsGraph}/;
1482 print "ok $test\n"; $test++;
1483 } else {
1484 print "not " if $char =~ /\p{IsGraph}/;
1485 print "ok $test\n"; $test++;
1486 print "not " unless $char =~ /\P{IsGraph}/;
1487 print "ok $test\n"; $test++;
1488 }
1489 print "# IsLower\n";
1490 if ($class =~ /^Ll$/) {
1491 print "not " unless $char =~ /\p{IsLower}/;
1492 print "ok $test\n"; $test++;
1493 print "not " if $char =~ /\P{IsLower}/;
1494 print "ok $test\n"; $test++;
1495 } else {
1496 print "not " if $char =~ /\p{IsLower}/;
1497 print "ok $test\n"; $test++;
1498 print "not " unless $char =~ /\P{IsLower}/;
1499 print "ok $test\n"; $test++;
1500 }
1501 print "# IsPrint\n";
1502 if ($class =~ /^([LMNPS])|Co|Zs/) {
1503 print "not " unless $char =~ /\p{IsPrint}/;
1504 print "ok $test\n"; $test++;
1505 print "not " if $char =~ /\P{IsPrint}/;
1506 print "ok $test\n"; $test++;
1507 } else {
1508 print "not " if $char =~ /\p{IsPrint}/;
1509 print "ok $test\n"; $test++;
1510 print "not " unless $char =~ /\P{IsPrint}/;
1511 print "ok $test\n"; $test++;
1512 }
1513 print "# IsPunct\n";
1514 if ($class =~ /^P/ || $char eq "_") {
1515 print "not " unless $char =~ /\p{IsPunct}/;
1516 print "ok $test\n"; $test++;
1517 print "not " if $char =~ /\P{IsPunct}/;
1518 print "ok $test\n"; $test++;
1519 } else {
1520 print "not " if $char =~ /\p{IsPunct}/;
1521 print "ok $test\n"; $test++;
1522 print "not " unless $char =~ /\P{IsPunct}/;
1523 print "ok $test\n"; $test++;
1524 }
1525 print "# IsSpace\n";
1526 if ($class =~ /^Z/ || ($code =~ /^(0009|000A|000B|000C|000D)$/)) {
1527 print "not " unless $char =~ /\p{IsSpace}/;
1528 print "ok $test\n"; $test++;
1529 print "not " if $char =~ /\P{IsSpace}/;
1530 print "ok $test\n"; $test++;
1531 } else {
1532 print "not " if $char =~ /\p{IsSpace}/;
1533 print "ok $test\n"; $test++;
1534 print "not " unless $char =~ /\P{IsSpace}/;
1535 print "ok $test\n"; $test++;
1536 }
1537 print "# IsUpper\n";
1538 if ($class =~ /^L[ut]/) {
1539 print "not " unless $char =~ /\p{IsUpper}/;
1540 print "ok $test\n"; $test++;
1541 print "not " if $char =~ /\P{IsUpper}/;
1542 print "ok $test\n"; $test++;
1543 } else {
1544 print "not " if $char =~ /\p{IsUpper}/;
1545 print "ok $test\n"; $test++;
1546 print "not " unless $char =~ /\P{IsUpper}/;
1547 print "ok $test\n"; $test++;
1548 }
1549 print "# IsWord\n";
1550 if ($class =~ /^[LMN]/ || $char eq "_") {
1551 print "not " unless $char =~ /\p{IsWord}/;
1552 print "ok $test\n"; $test++;
1553 print "not " if $char =~ /\P{IsWord}/;
1554 print "ok $test\n"; $test++;
1555 } else {
1556 print "not " if $char =~ /\p{IsWord}/;
1557 print "ok $test\n"; $test++;
1558 print "not " unless $char =~ /\P{IsWord}/;
1559 print "ok $test\n"; $test++;
1560 }
1561 }
1562}
1563
1564{
1565 $_ = "abc\x{100}\x{200}\x{300}\x{380}\x{400}defg";
1566
1567 if (/(.\x{300})./) {
1568 print "ok 576\n";
1569
1570 print "not " unless $` eq "abc\x{100}" && length($`) == 4;
73d6d589 1571 print "ok 577\n";
4765795a 1572
1573 print "not " unless $& eq "\x{200}\x{300}\x{380}" && length($&) == 3;
73d6d589 1574 print "ok 578\n";
4765795a 1575
1576 print "not " unless $' eq "\x{400}defg" && length($') == 5;
73d6d589 1577 print "ok 579\n";
4765795a 1578
1579 print "not " unless $1 eq "\x{200}\x{300}" && length($1) == 2;
73d6d589 1580 print "ok 580\n";
a8a2fe91 1581 } else {
1582 for (576..580) { print "not ok $_\n" }
4765795a 1583 }
1584}
8269fa76 1585
1586{
1587 # bug id 20010306.008
1588
1589 $a = "a\x{1234}";
1590 # The original bug report had 'no utf8' here but that was irrelevant.
1591 $a =~ m/\w/; # used to core dump
1592
1593 print "ok 581\n";
1594}
b8ef571c 1595
1596{
339e86bc 1597 $test = 582;
1598
b8ef571c 1599 # bugid 20010410.006
1600 for my $rx (
1601 '/(.*?)\{(.*?)\}/csg',
1602 '/(.*?)\{(.*?)\}/cg',
1603 '/(.*?)\{(.*?)\}/sg',
1604 '/(.*?)\{(.*?)\}/g',
1605 '/(.+?)\{(.+?)\}/csg',
1606 )
1607 {
1608 my($input, $i);
1609
1610 $i = 0;
1611 $input = "a{b}c{d}";
1612 eval <<EOT;
1613 while (eval \$input =~ $rx) {
1614 print "# \\\$1 = '\$1' \\\$2 = '\$2'\n";
1615 ++\$i;
1616 }
1617EOT
1618 print "not " unless $i == 2;
1619 print "ok " . $test++ . "\n";
1620 }
1621}
209a9bc1 1622
1623{
1624 # from Robin Houston
1625
b851fbc1 1626 my $x = "\x{10FFFD}";
209a9bc1 1627 $x =~ s/(.)/$1/g;
b851fbc1 1628 print "not " unless ord($x) == 0x10FFFD && length($x) == 1;
209a9bc1 1629 print "ok 587\n";
1630}
3568d838 1631
1632{
1633 my $x = "\x7f";
1634
1635 print "not " if $x =~ /[\x80-\xff]/;
1636 print "ok 588\n";
1637
1638 print "not " if $x =~ /[\x80-\x{100}]/;
1639 print "ok 589\n";
1640
1641 print "not " if $x =~ /[\x{100}]/;
1642 print "ok 590\n";
1643
1644 print "not " if $x =~ /\p{InLatin1Supplement}/;
1645 print "ok 591\n";
1646
1647 print "not " unless $x =~ /\P{InLatin1Supplement}/;
1648 print "ok 592\n";
1649
1650 print "not " if $x =~ /\p{InLatinExtendedA}/;
1651 print "ok 593\n";
1652
1653 print "not " unless $x =~ /\P{InLatinExtendedA}/;
1654 print "ok 594\n";
1655}
1656
1657{
1658 my $x = "\x80";
1659
1660 print "not " unless $x =~ /[\x80-\xff]/;
1661 print "ok 595\n";
1662
1663 print "not " unless $x =~ /[\x80-\x{100}]/;
1664 print "ok 596\n";
1665
1666 print "not " if $x =~ /[\x{100}]/;
1667 print "ok 597\n";
1668
1669 print "not " unless $x =~ /\p{InLatin1Supplement}/;
1670 print "ok 598\n";
1671
1672 print "not " if $x =~ /\P{InLatin1Supplement}/;
1673 print "ok 599\n";
1674
1675 print "not " if $x =~ /\p{InLatinExtendedA}/;
1676 print "ok 600\n";
1677
1678 print "not " unless $x =~ /\P{InLatinExtendedA}/;
1679 print "ok 601\n";
1680}
1681
1682{
1683 my $x = "\xff";
1684
1685 print "not " unless $x =~ /[\x80-\xff]/;
1686 print "ok 602\n";
1687
1688 print "not " unless $x =~ /[\x80-\x{100}]/;
1689 print "ok 603\n";
1690
1691 print "not " if $x =~ /[\x{100}]/;
1692 print "ok 604\n";
1693
1694 print "not " unless $x =~ /\p{InLatin1Supplement}/;
1695 print "ok 605\n";
1696
1697 print "not " if $x =~ /\P{InLatin1Supplement}/;
1698 print "ok 606\n";
1699
1700 print "not " if $x =~ /\p{InLatinExtendedA}/;
1701 print "ok 607\n";
1702
1703 print "not " unless $x =~ /\P{InLatinExtendedA}/;
1704 print "ok 608\n";
1705}
1706
1707{
1708 my $x = "\x{100}";
1709
1710 print "not " if $x =~ /[\x80-\xff]/;
1711 print "ok 609\n";
1712
1713 print "not " unless $x =~ /[\x80-\x{100}]/;
1714 print "ok 610\n";
1715
1716 print "not " unless $x =~ /[\x{100}]/;
1717 print "ok 611\n";
1718
1719 print "not " if $x =~ /\p{InLatin1Supplement}/;
1720 print "ok 612\n";
1721
1722 print "not " unless $x =~ /\P{InLatin1Supplement}/;
1723 print "ok 613\n";
1724
1725 print "not " unless $x =~ /\p{InLatinExtendedA}/;
1726 print "ok 614\n";
1727
1728 print "not " if $x =~ /\P{InLatinExtendedA}/;
1729 print "ok 615\n";
1730}
1731
9d1d55b5 1732{
1733 # from japhy
1734 my $w;
1735 use warnings;
1736 local $SIG{__WARN__} = sub { $w .= shift };
1737
1738 $w = "";
1739 eval 'qr/(?c)/';
1740 print "not " if $w !~ /^Useless \(\?c\)/;
1741 print "ok 616\n";
1742
1743 $w = "";
1744 eval 'qr/(?-c)/';
1745 print "not " if $w !~ /^Useless \(\?-c\)/;
1746 print "ok 617\n";
1747
1748 $w = "";
1749 eval 'qr/(?g)/';
1750 print "not " if $w !~ /^Useless \(\?g\)/;
1751 print "ok 618\n";
1752
1753 $w = "";
1754 eval 'qr/(?-g)/';
1755 print "not " if $w !~ /^Useless \(\?-g\)/;
1756 print "ok 619\n";
1757
1758 $w = "";
1759 eval 'qr/(?o)/';
1760 print "not " if $w !~ /^Useless \(\?o\)/;
1761 print "ok 620\n";
1762
1763 $w = "";
1764 eval 'qr/(?-o)/';
1765 print "not " if $w !~ /^Useless \(\?-o\)/;
1766 print "ok 621\n";
1767
1768 # now test multi-error regexes
1769
1770 $w = "";
1771 eval 'qr/(?g-o)/';
1772 print "not " if $w !~ /^Useless \(\?g\).*\nUseless \(\?-o\)/;
1773 print "ok 622\n";
1774
1775 $w = "";
1776 eval 'qr/(?g-c)/';
1777 print "not " if $w !~ /^Useless \(\?g\).*\nUseless \(\?-c\)/;
1778 print "ok 623\n";
1779
1780 $w = "";
1781 eval 'qr/(?o-cg)/'; # (?c) means (?g) error won't be thrown
1782 print "not " if $w !~ /^Useless \(\?o\).*\nUseless \(\?-c\)/;
1783 print "ok 624\n";
1784
1785 $w = "";
1786 eval 'qr/(?ogc)/';
1787 print "not " if $w !~ /^Useless \(\?o\).*\nUseless \(\?g\).*\nUseless \(\?c\)/;
1788 print "ok 625\n";
1789}
a72deede 1790
1791# More Unicode "class" tests
1792
1793{
1794 use charnames ':full';
1795
1796 print "not " unless "\N{LATIN CAPITAL LETTER A}" =~ /\p{InBasicLatin}/;
1797 print "ok 626\n";
1798
1799 print "not " unless "\N{LATIN CAPITAL LETTER A WITH GRAVE}" =~ /\p{InLatin1Supplement}/;
1800 print "ok 627\n";
1801
1802 print "not " unless "\N{LATIN CAPITAL LETTER A WITH MACRON}" =~ /\p{InLatinExtendedA}/;
1803 print "ok 628\n";
1804
1805 print "not " unless "\N{LATIN SMALL LETTER B WITH STROKE}" =~ /\p{InLatinExtendedB}/;
1806 print "ok 629\n";
1807
1808 print "not " unless "\N{KATAKANA LETTER SMALL A}" =~ /\p{InKatakana}/;
1809 print "ok 630\n";
1810}
1811
6002328a 1812$_ = "foo";
1813
1814eval <<"EOT"; die if $@;
1815 /f
1816 o\r
1817 o
1818 \$
1819 /x && print "ok 631\n";
1820EOT
1821
1822eval <<"EOT"; die if $@;
1823 /f
1824 o
1825 o
1826 \$\r
1827 /x && print "ok 632\n";
1828EOT
1829
569b5e07 1830#test /o feature
1831sub test_o { $_[0] =~/$_[1]/o; return $1}
1832if(test_o('abc','(.)..') eq 'a') {
395ddfe6 1833 print "ok 633\n";
569b5e07 1834} else {
395ddfe6 1835 print "not ok 633\n";
569b5e07 1836}
1837if(test_o('abc','..(.)') eq 'a') {
395ddfe6 1838 print "ok 634\n";
569b5e07 1839} else {
395ddfe6 1840 print "not ok 634\n";
569b5e07 1841}
1842
f79b3095 1843# 635..639: ID 20010619.003 (only the space character is
1844# supposed to be [:print:], not the whole isprint()).
1845
1846print "not " if "\n" =~ /[[:print:]]/;
1847print "ok 635\n";
1848
1849print "not " if "\t" =~ /[[:print:]]/;
1850print "ok 636\n";
1851
e857312d 1852# Amazingly vertical tabulator is the same in ASCII and EBCDIC.
f79b3095 1853print "not " if "\014" =~ /[[:print:]]/;
1854print "ok 637\n";
1855
1856print "not " if "\r" =~ /[[:print:]]/;
1857print "ok 638\n";
1858
1859print "not " unless " " =~ /[[:print:]]/;
1860print "ok 639\n";
1861
a01268b5 1862##
1863## Test basic $^N usage outside of a regex
1864##
1865$x = "abcdef";
1866$T="ok 640\n";if ($x =~ /cde/ and not defined $^N) {print $T} else {print "not $T"};
1867$T="ok 641\n";if ($x =~ /(cde)/ and $^N eq "cde") {print $T} else {print "not $T"};
1868$T="ok 642\n";if ($x =~ /(c)(d)(e)/ and $^N eq "e") {print $T} else {print "not $T"};
1869$T="ok 643\n";if ($x =~ /(c(d)e)/ and $^N eq "cde") {print $T} else {print "not $T"};
1870$T="ok 644\n";if ($x =~ /(foo)|(c(d)e)/ and $^N eq "cde") {print $T} else {print "not $T"};
1871$T="ok 645\n";if ($x =~ /(c(d)e)|(foo)/ and $^N eq "cde") {print $T} else {print "not $T"};
1872$T="ok 646\n";if ($x =~ /(c(d)e)|(abc)/ and $^N eq "abc") {print $T} else {print "not $T"};
1873$T="ok 647\n";if ($x =~ /(c(d)e)|(abc)x/ and $^N eq "cde") {print $T} else {print "not $T"};
1874$T="ok 648\n";if ($x =~ /(c(d)e)(abc)?/ and $^N eq "cde") {print $T} else {print "not $T"};
1875$T="ok 649\n";if ($x =~ /(?:c(d)e)/ and $^N eq "d" ) {print $T} else {print "not $T"};
1876$T="ok 650\n";if ($x =~ /(?:c(d)e)(?:f)/ and $^N eq "d" ) {print $T} else {print "not $T"};
1877$T="ok 651\n";if ($x =~ /(?:([abc])|([def]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
1878$T="ok 652\n";if ($x =~ /(?:([ace])|([bdf]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
1879$T="ok 653\n";if ($x =~ /(([ace])|([bd]))*/ and $^N eq "e" ){print $T} else {print "not $T"};
1880{
1881 $T="ok 654\n";if($x =~ /(([ace])|([bdf]))*/ and $^N eq "f" ){print $T} else {print "not $T"};
1882}
1883## test to see if $^N is automatically localized -- it should now
1884## have the value set in test 653
1885$T="ok 655\n";if ($^N eq "e" ){print $T} else {print "not $T"};
1886
1887##
1888## Now test inside (?{...})
1889##
1890$T="ok 656\n";if ($x =~ /a([abc])(?{$y=$^N})c/ and $y eq "b" ){print $T} else {print "not $T"};
1891$T="ok 657\n";if ($x =~ /a([abc]+)(?{$y=$^N})d/ and $y eq "bc"){print $T} else {print "not $T"};
1892$T="ok 658\n";if ($x =~ /a([abcdefg]+)(?{$y=$^N})d/ and $y eq "bc"){print $T} else {print "not $T"};
1893$T="ok 659\n";if ($x =~ /(a([abcdefg]+)(?{$y=$^N})d)(?{$z=$^N})e/ and $y eq "bc" and $z eq "abcd")
1894 {print $T} else {print "not $T"};
1895$T="ok 660\n";if ($x =~ /(a([abcdefg]+)(?{$y=$^N})de)(?{$z=$^N})/ and $y eq "bc" and $z eq "abcde")
1896 {print $T} else {print "not $T"};
2796c109 1897
1898# Test the Unicode script classes
1899
cf25bb62 1900print "not " unless chr(0x100) =~ /\p{IsLatin}/; # outside Latin-1
2796c109 1901print "ok 661\n";
1902
cf25bb62 1903print "not " unless chr(0x212b) =~ /\p{IsLatin}/; # Angstrom sign, very outside
2796c109 1904print "ok 662\n";
1905
cf25bb62 1906print "not " unless chr(0x5d0) =~ /\p{IsHebrew}/; # inside InHebrew
2796c109 1907print "ok 663\n";
1908
cf25bb62 1909print "not " unless chr(0xfb4f) =~ /\p{IsHebrew}/; # outside InHebrew
2796c109 1910print "ok 664\n";
1911
cf25bb62 1912print "not " unless chr(0xb5) =~ /\p{IsGreek}/; # singleton (not in a range)
5f9563ea 1913print "ok 665\n";
1914
cf25bb62 1915print "not " unless chr(0x37a) =~ /\p{IsGreek}/; # singleton
5f9563ea 1916print "ok 666\n";
1917
cf25bb62 1918print "not " unless chr(0x386) =~ /\p{IsGreek}/; # singleton
5f9563ea 1919print "ok 667\n";
1920
cf25bb62 1921print "not " unless chr(0x387) =~ /\P{IsGreek}/; # not there
5f9563ea 1922print "ok 668\n";
1923
cf25bb62 1924print "not " unless chr(0x388) =~ /\p{IsGreek}/; # range
5f9563ea 1925print "ok 669\n";
1926
cf25bb62 1927print "not " unless chr(0x38a) =~ /\p{IsGreek}/; # range
5f9563ea 1928print "ok 670\n";
1929
cf25bb62 1930print "not " unless chr(0x38b) =~ /\P{IsGreek}/; # not there
5f9563ea 1931print "ok 671\n";
1932
cf25bb62 1933print "not " unless chr(0x38c) =~ /\p{IsGreek}/; # singleton
5f9563ea 1934print "ok 672\n";
1935
550cec39 1936if (ord("A") == 65) {
7be5a6cf 1937##
1938## Test [:cntrl:]...
1939##
1940## Should probably put in tests for all the POSIX stuff, but not sure how to
1941## guarantee a specific locale......
1942##
550cec39 1943 $AllBytes = join('', map { chr($_) } 0..255);
1944 ($x = $AllBytes) =~ s/[[:cntrl:]]//g;
1945 if ($x ne join('', map { chr($_) } 0x20..0x7E, 0x80..0xFF)) {
1946 print "not ";
1947 }
1948 print "ok 673\n";
1949
1950 ($x = $AllBytes) =~ s/[^[:cntrl:]]//g;
1951 if ($x ne join('', map { chr($_) } 0..0x1F, 0x7F)) { print "not " }
1952 print "ok 674\n";
1953} else {
1954 print "ok $_ # Skip: EBCDIC\n" for 673..674;
1955}
f33976b4 1956
1957# With /s modifier UTF8 chars were interpreted as bytes
1958{
1959 my $a = "Hello \x{263A} World";
1960
1961 my @a = ($a =~ /./gs);
1962
1963 print "not " unless $#a == 12;
1964 print "ok 675\n";
1965}
cce850e4 1966
1967@a = ("foo\nbar" =~ /./g);
1968print "ok 676\n" if @a == 6 && "@a" eq "f o o b a r";
1969
1970@a = ("foo\nbar" =~ /./gs);
1971print "ok 677\n" if @a == 7 && "@a" eq "f o o \n b a r";
1972
1973@a = ("foo\nbar" =~ /\C/g);
1974print "ok 678\n" if @a == 7 && "@a" eq "f o o \n b a r";
1975
1976@a = ("foo\nbar" =~ /\C/gs);
1977print "ok 679\n" if @a == 7 && "@a" eq "f o o \n b a r";
1978
1979@a = ("foo\n\x{100}bar" =~ /./g);
1980print "ok 680\n" if @a == 7 && "@a" eq "f o o \x{100} b a r";
1981
1982@a = ("foo\n\x{100}bar" =~ /./gs);
1983print "ok 681\n" if @a == 8 && "@a" eq "f o o \n \x{100} b a r";
1984
1985($a, $b) = map { chr } ord('A') == 65 ? (0xc4, 0x80) : (0x8c, 0x41);
1986
1987@a = ("foo\n\x{100}bar" =~ /\C/g);
1988print "ok 682\n" if @a == 9 && "@a" eq "f o o \n $a $b b a r";
1989
1990@a = ("foo\n\x{100}bar" =~ /\C/gs);
1991print "ok 683\n" if @a == 9 && "@a" eq "f o o \n $a $b b a r";
1992
0af80b60 1993{
1994 # [ID 20010814.004] pos() doesn't work when using =~m// in list context
1995 $_ = "ababacadaea";
1996 $a = join ":", /b./gc;
1997 $b = join ":", /a./gc;
1998 $c = pos;
1999 print "$a $b $c" eq 'ba:ba ad:ae 10' ? "ok 684\n" : "not ok 684\t# $a $b $c\n";
2000}
d9f424b2 2001
2002{
75685a94 2003 # [ID 20010407.006] matching utf8 return values from functions does not work
2004
d9f424b2 2005 package ID_20010407_006;
2006
2007 sub x {
2008 "a\x{1234}";
2009 }
2010
2011 my $x = x;
2012 my $y;
2013
2014 $x =~ /(..)/; $y = $1;
2015 print "not " unless length($y) == 2 && $y eq $x;
75685a94 2016 print "ok 685\n";
d9f424b2 2017
2018 x =~ /(..)/; $y = $1;
2019 print "not " unless length($y) == 2 && $y eq $x;
2020 print "ok 686\n";
2021}
a4c04bdc 2022
e2d8ce26 2023
a4c04bdc 2024my $test = 687;
2025
2026# Force scalar context on the patern match
2027sub ok ($$) {
2028 my($ok, $name) = @_;
2029
2030 printf "%sok %d - %s\n", ($ok ? "" : "not "), $test, $name;
2031
2032 printf "# Failed test at line %d\n", (caller)[2] unless $ok;
2033
2034 $test++;
2035 return $ok;
2036}
2037
2038{
2039 # Check that \x## works. 5.6.1 and 5.005_03 fail some of these.
2040 $x = "\x4e" . "E";
2041 ok ($x =~ /^\x4EE$/, "Check only 2 bytes of hex are matched.");
2042
2043 $x = "\x4e" . "i";
2044 ok ($x =~ /^\x4Ei$/, "Check that invalid hex digit stops it (2)");
2045
2046 $x = "\x4" . "j";
2047 ok ($x =~ /^\x4j$/, "Check that invalid hex digit stops it (1)");
2048
2049 $x = "\x0" . "k";
2050 ok ($x =~ /^\xk$/, "Check that invalid hex digit stops it (0)");
2051
2052 $x = "\x0" . "x";
2053 ok ($x =~ /^\xx$/, "\\xx isn't to be treated as \\0");
2054
2055 $x = "\x0" . "xa";
2056 ok ($x =~ /^\xxa$/, "\\xxa isn't to be treated as \\xa");
2057
2058 $x = "\x9" . "_b";
2059 ok ($x =~ /^\x9_b$/, "\\x9_b isn't to be treated as \\x9b");
2060
2061 print "# and now again in [] ranges\n";
2062
2063 $x = "\x4e" . "E";
2064 ok ($x =~ /^[\x4EE]{2}$/, "Check only 2 bytes of hex are matched.");
2065
2066 $x = "\x4e" . "i";
2067 ok ($x =~ /^[\x4Ei]{2}$/, "Check that invalid hex digit stops it (2)");
2068
2069 $x = "\x4" . "j";
2070 ok ($x =~ /^[\x4j]{2}$/, "Check that invalid hex digit stops it (1)");
2071
2072 $x = "\x0" . "k";
2073 ok ($x =~ /^[\xk]{2}$/, "Check that invalid hex digit stops it (0)");
2074
2075 $x = "\x0" . "x";
2076 ok ($x =~ /^[\xx]{2}$/, "\\xx isn't to be treated as \\0");
2077
2078 $x = "\x0" . "xa";
2079 ok ($x =~ /^[\xxa]{3}$/, "\\xxa isn't to be treated as \\xa");
2080
2081 $x = "\x9" . "_b";
2082 ok ($x =~ /^[\x9_b]{3}$/, "\\x9_b isn't to be treated as \\x9b");
2083
2084}
2085
2086{
2087 # Check that \x{##} works. 5.6.1 fails quite a few of these.
2088
2089 $x = "\x9b";
2090 ok ($x =~ /^\x{9_b}$/, "\\x{9_b} is to be treated as \\x9b");
2091
2092 $x = "\x9b" . "y";
2093 ok ($x =~ /^\x{9_b}y$/, "\\x{9_b} is to be treated as \\x9b (again)");
2094
2095 $x = "\x9b" . "y";
2096 ok ($x =~ /^\x{9b_}y$/, "\\x{9b_} is to be treated as \\x9b");
2097
2098 $x = "\x9b" . "y";
2099 ok ($x =~ /^\x{9_bq}y$/, "\\x{9_bc} is to be treated as \\x9b");
2100
2101 $x = "\x0" . "y";
2102 ok ($x =~ /^\x{x9b}y$/, "\\x{x9b} is to be treated as \\x0");
2103
2104 $x = "\x0" . "y";
2105 ok ($x =~ /^\x{0x9b}y$/, "\\x{0x9b} is to be treated as \\x0");
2106
2107 $x = "\x9b" . "y";
2108 ok ($x =~ /^\x{09b}y$/, "\\x{09b} is to be treated as \\x9b");
2109
2110 print "# and now again in [] ranges\n";
2111
2112 $x = "\x9b";
2113 ok ($x =~ /^[\x{9_b}]$/, "\\x{9_b} is to be treated as \\x9b");
2114
2115 $x = "\x9b" . "y";
2116 ok ($x =~ /^[\x{9_b}y]{2}$/, "\\x{9_b} is to be treated as \\x9b (again)");
2117
2118 $x = "\x9b" . "y";
2119 ok ($x =~ /^[\x{9b_}y]{2}$/, "\\x{9b_} is to be treated as \\x9b");
2120
2121 $x = "\x9b" . "y";
2122 ok ($x =~ /^[\x{9_bq}y]{2}$/, "\\x{9_bc} is to be treated as \\x9b");
2123
2124 $x = "\x0" . "y";
2125 ok ($x =~ /^[\x{x9b}y]{2}$/, "\\x{x9b} is to be treated as \\x0");
2126
2127 $x = "\x0" . "y";
2128 ok ($x =~ /^[\x{0x9b}y]{2}$/, "\\x{0x9b} is to be treated as \\x0");
2129
2130 $x = "\x9b" . "y";
2131 ok ($x =~ /^[\x{09b}y]{2}$/, "\\x{09b} is to be treated as \\x9b");
2132}
e2d8ce26 2133
2134{
d9efae67 2135 # high bit bug -- japhy
2136 my $x = "ab\200d";
2137 $x =~ /.*?\200/ or print "not ";
2138 print "ok 715\n";
e2d8ce26 2139}
2140
4193bef7 2141print "# some Unicode properties\n";
2142
d9efae67 2143{
4193bef7 2144 # Dashes, underbars, case.
d9efae67 2145 print "not " unless "\x80" =~ /\p{in-latin1_SUPPLEMENT}/;
2146 print "ok 716\n";
ab13f0c7 2147
4193bef7 2148 # Complement, leading and trailing whitespace.
ab13f0c7 2149 print "not " unless "\x80" =~ /\P{ ^ In Latin 1 Supplement }/;
2150 print "ok 717\n";
4193bef7 2151
f173cd49 2152 # No ^In, dashes, case, dash, any intervening (word-break) whitespace.
2153 # (well, newlines don't work...)
2154 print "not " unless "\x80" =~ /\p{latin-1 supplement}/;
4193bef7 2155 print "ok 718\n";
2156}
2157
2158{
2159 print "not " unless "a" =~ /\pL/;
2160 print "ok 719\n";
4193bef7 2161
4193bef7 2162 print "not " unless "a" =~ /\p{IsLl}/;
2163 print "ok 720\n";
4193bef7 2164
c87b7cc2 2165 print "not " if "a" =~ /\p{IsLu}/;
4193bef7 2166 print "ok 721\n";
4193bef7 2167
61247495 2168 print "not " unless "a" =~ /\p{Ll}/;
4193bef7 2169 print "ok 722\n";
c87b7cc2 2170
61247495 2171 print "not " if "a" =~ /\p{Lu}/;
c87b7cc2 2172 print "ok 723\n";
2173
61247495 2174 print "not " unless "A" =~ /\pL/;
c87b7cc2 2175 print "ok 724\n";
2176
61247495 2177 print "not " unless "A" =~ /\p{IsLu}/;
c87b7cc2 2178 print "ok 725\n";
2179
61247495 2180 print "not " if "A" =~ /\p{IsLl}/;
c87b7cc2 2181 print "ok 726\n";
2182
61247495 2183 print "not " unless "A" =~ /\p{Lu}/;
c87b7cc2 2184 print "ok 727\n";
2185
61247495 2186 print "not " if "A" =~ /\p{Ll}/;
c87b7cc2 2187 print "ok 728\n";
2188
61247495 2189 print "not " if "a" =~ /\PL/;
c87b7cc2 2190 print "ok 729\n";
2191
61247495 2192 print "not " if "a" =~ /\P{IsLl}/;
c87b7cc2 2193 print "ok 730\n";
61247495 2194
2195 print "not " unless "a" =~ /\P{IsLu}/;
2196 print "ok 731\n";
2197
2198 print "not " if "a" =~ /\P{Ll}/;
2199 print "ok 732\n";
2200
2201 print "not " unless "a" =~ /\P{Lu}/;
2202 print "ok 733\n";
2203
2204 print "not " if "A" =~ /\PL/;
2205 print "ok 734\n";
2206
2207 print "not " if "A" =~ /\P{IsLu}/;
2208 print "ok 735\n";
2209
2210 print "not " unless "A" =~ /\P{IsLl}/;
2211 print "ok 736\n";
2212
2213 print "not " if "A" =~ /\P{Lu}/;
2214 print "ok 737\n";
2215
2216 print "not " unless "A" =~ /\P{Ll}/;
2217 print "ok 738\n";
2218
4193bef7 2219}
9b4e380a 2220
2221{
2222 print "not " if "a" =~ /\p{Common}/;
2223 print "ok 739\n";
2224
2225 print "not " unless "1" =~ /\p{Common}/;
2226 print "ok 740\n";
2227}
2228
2229{
2230 print "not " if "a" =~ /\p{Inherited}/;
2231 print "ok 741\n";
2232
2233 print "not " unless "\x{300}" =~ /\p{Inherited}/;
2234 print "ok 742\n";
2235}
2236
2237{
2238 print "not " unless "a" =~ /\p{L&}/;
2239 print "ok 743\n";
2240
2241 print "not " if "1" =~ /\p{L&}/;
2242 print "ok 744\n";
2243}
d73e5302 2244
2245{
904bc114 2246 print "not " unless "a" =~ /\p{Lowercase Letter}/;
d73e5302 2247 print "ok 745\n";
2248
904bc114 2249 print "not " if "A" =~ /\p{lowercaseletter}/;
d73e5302 2250 print "ok 746\n";
2251}
2252
2253{
cf25bb62 2254 print "not " unless "\x{AC00}" =~ /\p{HangulSyllables}/;
d73e5302 2255 print "ok 747\n";
2256}
71d929cb 2257
2258{
701a277b 2259 # Script=, Block=, Category=
2260
71d929cb 2261 print "not " unless "\x{0100}" =~ /\p{Script=Latin}/;
2262 print "ok 748\n";
2263
2264 print "not " unless "\x{0100}" =~ /\p{Block=LatinExtendedA}/;
2265 print "ok 749\n";
2266
2267 print "not " unless "\x{0100}" =~ /\p{Category=UppercaseLetter}/;
2268 print "ok 750\n";
2269}
2270
ef54fa25 2271{
701a277b 2272 print "# the basic character classes and Unicode \n";
2273
ef54fa25 2274 # 0100;LATIN CAPITAL LETTER A WITH MACRON;Lu;0;L;0041 0304;;;;N;LATIN CAPITAL LETTER A MACRON;;;0101;
2275 print "not " unless "\x{0100}" =~ /\w/;
2276 print "ok 751\n";
2277
2278 # 0660;ARABIC-INDIC DIGIT ZERO;Nd;0;AN;;0;0;0;N;;;;;
2279 print "not " unless "\x{0660}" =~ /\d/;
2280 print "ok 752\n";
2281
2282 # 1680;OGHAM SPACE MARK;Zs;0;WS;;;;;N;;;;;
2283 print "not " unless "\x{1680}" =~ /\s/;
2284 print "ok 753\n";
2285}
701a277b 2286
2287{
2288 print "# folding matches and Unicode\n";
2289
2290 print "not " unless "a\x{100}" =~ /A/i;
2291 print "ok 754\n";
2292
27e28630 2293 print "not " unless "A\x{100}" =~ /a/i;
701a277b 2294 print "ok 755\n";
2295
2296 print "not " unless "a\x{100}" =~ /a/i;
2297 print "ok 756\n";
2298
2299 print "not " unless "A\x{100}" =~ /A/i;
2300 print "ok 757\n";
bc517b45 2301
2302 print "not " unless "\x{101}a" =~ /\x{100}/i;
2303 print "ok 758\n";
2304
2305 print "not " unless "\x{100}a" =~ /\x{100}/i;
2306 print "ok 759\n";
2307
2308 print "not " unless "\x{101}a" =~ /\x{101}/i;
2309 print "ok 760\n";
2310
2311 print "not " unless "\x{100}a" =~ /\x{101}/i;
2312 print "ok 761\n";
2313
2314 print "not " unless "a\x{100}" =~ /A\x{100}/i;
2315 print "ok 762\n";
2316
27e28630 2317 print "not " unless "A\x{100}" =~ /a\x{100}/i;
bc517b45 2318 print "ok 763\n";
2319
2320 print "not " unless "a\x{100}" =~ /a\x{100}/i;
2321 print "ok 764\n";
2322
2323 print "not " unless "A\x{100}" =~ /A\x{100}/i;
2324 print "ok 765\n";
2325
2326 print "not " unless "a\x{100}" =~ /[A]/i;
2327 print "ok 766\n";
2328
27e28630 2329 print "not " unless "A\x{100}" =~ /[a]/i;
bc517b45 2330 print "ok 767\n";
2331
2332 print "not " unless "a\x{100}" =~ /[a]/i;
2333 print "ok 768\n";
2334
2335 print "not " unless "A\x{100}" =~ /[A]/i;
2336 print "ok 769\n";
2337
2338 print "not " unless "\x{101}a" =~ /[\x{100}]/i;
2339 print "ok 770\n";
2340
2341 print "not " unless "\x{100}a" =~ /[\x{100}]/i;
2342 print "ok 771\n";
2343
2344 print "not " unless "\x{101}a" =~ /[\x{101}]/i;
2345 print "ok 772\n";
2346
2347 print "not " unless "\x{100}a" =~ /[\x{101}]/i;
2348 print "ok 773\n";
2349
701a277b 2350}
a5961de5 2351
2352{
2353 use charnames ':full';
2354
2355 print "# LATIN LETTER A WITH GRAVE\n";
2356 my $lower = "\N{LATIN SMALL LETTER A WITH GRAVE}";
2357 my $UPPER = "\N{LATIN CAPITAL LETTER A WITH GRAVE}";
2358
bc517b45 2359 print $lower =~ m/$UPPER/i ? "ok 774\n" : "not ok 774\n";
2360 print $UPPER =~ m/$lower/i ? "ok 775\n" : "not ok 775\n";
2361 print $lower =~ m/[$UPPER]/i ? "ok 776\n" : "not ok 776\n";
2362 print $UPPER =~ m/[$lower]/i ? "ok 777\n" : "not ok 777\n";
a5961de5 2363
2364 print "# GREEK LETTER ALPHA WITH VRACHY\n";
2365
2366 $lower = "\N{GREEK CAPITAL LETTER ALPHA WITH VRACHY}";
2367 $UPPER = "\N{GREEK SMALL LETTER ALPHA WITH VRACHY}";
2368
bc517b45 2369 print $lower =~ m/$UPPER/i ? "ok 778\n" : "not ok 778\n";
2370 print $UPPER =~ m/$lower/i ? "ok 779\n" : "not ok 779\n";
2371 print $lower =~ m/[$UPPER]/i ? "ok 780\n" : "not ok 780\n";
2372 print $UPPER =~ m/[$lower]/i ? "ok 781\n" : "not ok 781\n";
a5961de5 2373
2374 print "# LATIN LETTER Y WITH DIAERESIS\n";
2375
2376 $lower = "\N{LATIN CAPITAL LETTER Y WITH DIAERESIS}";
2377 $UPPER = "\N{LATIN SMALL LETTER Y WITH DIAERESIS}";
bc517b45 2378 print $lower =~ m/$UPPER/i ? "ok 782\n" : "not ok 782\n";
2379 print $UPPER =~ m/$lower/i ? "ok 783\n" : "not ok 783\n";
2380 print $lower =~ m/[$UPPER]/i ? "ok 784\n" : "not ok 784\n";
2381 print $UPPER =~ m/[$lower]/i ? "ok 785\n" : "not ok 785\n";
a5961de5 2382}
55da9344 2383
2384{
2385 use warnings;
2386 use charnames ':full';
2387
2388 print "# GREEK CAPITAL LETTER SIGMA vs COMBINING GREEK PERISPOMENI\n";
2389
2390 my $SIGMA = "\N{GREEK CAPITAL LETTER SIGMA}";
d07ddd77 2391 my $char = "\N{COMBINING GREEK PERISPOMENI}";
55da9344 2392
bc517b45 2393 # Before #13843 this was failing by matching falsely.
2394 print "_:$char:_" =~ m/_:$SIGMA:_/i ? "not ok 786\n" : "ok 786\n";
55da9344 2395}
b7c83a7e 2396
2397{
2398 print "# \\X\n";
2399
2400 use charnames ':full';
2401
eb08e2da 2402 print "a!" =~ /^(\X)!/ && $1 eq "a" ?
2403 "ok 787\n" : "not ok 787 # $1\n";
2404 print "\xDF!" =~ /^(\X)!/ && $1 eq "\xDF" ?
2405 "ok 788\n" : "not ok 788 # $1\n";
2406 print "\x{100}!" =~ /^(\X)!/ && $1 eq "\x{100}" ?
2407 "ok 789\n" : "not ok 789 # $1\n";
2408 print "\x{100}\x{300}!" =~ /^(\X)!/ && $1 eq "\x{100}\x{300}" ?
2409 "ok 790\n" : "not ok 790 # $1\n";
2410 print "\N{LATIN CAPITAL LETTER E}!" =~ /^(\X)!/ &&
2411 $1 eq "\N{LATIN CAPITAL LETTER E}" ?
2412 "ok 791\n" : "not ok 791 # $1\n";
2413 print "\N{LATIN CAPITAL LETTER E}\N{COMBINING GRAVE ACCENT}!" =~
2414 /^(\X)!/ &&
2415 $1 eq "\N{LATIN CAPITAL LETTER E}\N{COMBINING GRAVE ACCENT}" ?
2416 "ok 792\n" : "not ok 792 # $1\n";
b7c83a7e 2417}
112bedeb 2418
2419{
2420 print "#\\C and \\X\n";
2421
2422 print "!abc!" =~ /a\Cc/ ? "ok 793\n" : "not ok 793\n";
2423 print "!abc!" =~ /a\Xc/ ? "ok 794\n" : "not ok 794\n";
2424}
09091399 2425
2426{
2427 print "# FINAL SIGMA\n";
2428
2429 my $SIGMA = "\x{03A3}"; # CAPITAL
2430 my $Sigma = "\x{03C2}"; # SMALL FINAL
2431 my $sigma = "\x{03C3}"; # SMALL
2432
2433 print $SIGMA =~ /$SIGMA/i ? "ok 795\n" : "not ok 795\n";
2434 print $SIGMA =~ /$Sigma/i ? "ok 796\n" : "not ok 796\n";
2435 print $SIGMA =~ /$sigma/i ? "ok 797\n" : "not ok 797\n";
2436
2437 print $Sigma =~ /$SIGMA/i ? "ok 798\n" : "not ok 798\n";
2438 print $Sigma =~ /$Sigma/i ? "ok 799\n" : "not ok 799\n";
2439 print $Sigma =~ /$sigma/i ? "ok 800\n" : "not ok 800\n";
2440
2441 print $sigma =~ /$SIGMA/i ? "ok 801\n" : "not ok 801\n";
2442 print $sigma =~ /$Sigma/i ? "ok 802\n" : "not ok 802\n";
2443 print $sigma =~ /$sigma/i ? "ok 803\n" : "not ok 803\n";
2444
2445 print $SIGMA =~ /[$SIGMA]/i ? "ok 804\n" : "not ok 804\n";
2446 print $SIGMA =~ /[$Sigma]/i ? "ok 805\n" : "not ok 805\n";
2447 print $SIGMA =~ /[$sigma]/i ? "ok 806\n" : "not ok 806\n";
2448
2449 print $Sigma =~ /[$SIGMA]/i ? "ok 807\n" : "not ok 807\n";
2450 print $Sigma =~ /[$Sigma]/i ? "ok 808\n" : "not ok 808\n";
2451 print $Sigma =~ /[$sigma]/i ? "ok 809\n" : "not ok 809\n";
2452
2453 print $sigma =~ /[$SIGMA]/i ? "ok 810\n" : "not ok 810\n";
2454 print $sigma =~ /[$Sigma]/i ? "ok 811\n" : "not ok 811\n";
2455 print $sigma =~ /[$sigma]/i ? "ok 812\n" : "not ok 812\n";
2456}
e036fef9 2457
2458{
2459 print "# parlez-vous?\n";
2460
2461 use charnames ':full';
2462
2463 print "fran\N{LATIN SMALL LETTER C}ais" =~
2464 /fran.ais/ &&
2465 $& eq "francais" ?
2466 "ok 813\n" : "not ok 813\n";
2467
2468 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2469 /fran.ais/ &&
2470 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2471 "ok 814\n" : "not ok 814\n";
2472
2473 print "fran\N{LATIN SMALL LETTER C}ais" =~
2474 /fran\Cais/ &&
2475 $& eq "francais" ?
2476 "ok 815\n" : "not ok 815\n";
2477
2478 print "franc\N{COMBINING CEDILLA}ais" =~
2479 /franc\C\Cais/ ? # COMBINING CEDILLA is two bytes when encoded
2480 "ok 816\n" : "not ok 816\n";
2481
2482 print "fran\N{LATIN SMALL LETTER C}ais" =~
2483 /fran\Xais/ &&
2484 $& eq "francais" ?
2485 "ok 817\n" : "not ok 817\n";
2486
2487 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2488 /fran\Xais/ &&
2489 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2490 "ok 818\n" : "not ok 818\n";
2491
2492 print "franc\N{COMBINING CEDILLA}ais" =~
2493 /fran\Xais/ &&
2494 $& eq "franc\N{COMBINING CEDILLA}ais" ?
2495 "ok 819\n" : "not ok 819\n";
2496
2497 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2498 /fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais/ &&
2499 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2500 "ok 820\n" : "not ok 820\n";
2501
2502 print "franc\N{COMBINING CEDILLA}ais" =~
2503 /franc\N{COMBINING CEDILLA}ais/ &&
2504 $& eq "franc\N{COMBINING CEDILLA}ais" ?
2505 "ok 821\n" : "not ok 821\n";
2506
2507 print "fran\N{LATIN SMALL LETTER C}ais" =~
2508 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2509 $& eq "francais" ?
2510 "ok 822\n" : "not ok 822\n";
2511
2512 print "fran\N{LATIN SMALL LETTER C}ais" =~
2513 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2514 $& eq "francais" ?
2515 "ok 823\n" : "not ok 823\n";
2516
2517 print "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
2518 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2519 $& eq "fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" ?
2520 "ok 824\n" : "not ok 824\n";
2521
2522 print "franc\N{COMBINING CEDILLA}ais" =~
2523 /fran(?:c\N{COMBINING CEDILLA}?|\N{LATIN SMALL LETTER C WITH CEDILLA})ais/ &&
2524 $& eq "franc\N{COMBINING CEDILLA}ais" ?
2525 "ok 825\n" : "not ok 825\n";
2526}
ffce6cc2 2527
2528{
2529 print "# Does lingering (and useless) UTF8 flag mess up /i matching?\n";
2530
2531 {
2532 my $regex = "ABcde";
2533 my $string = "abcDE\x{100}";
2534 chop($string);
2535 if ($string =~ m/$regex/i) {
2536 print "ok 826\n";
2537 } else {
2538 print "not ok 826\n";
2539 }
2540 }
2541
2542 {
2543 my $regex = "ABcde\x{100}";
2544 my $string = "abcDE";
2545 chop($regex);
2546 if ($string =~ m/$regex/i) {
2547 print "ok 827\n";
2548 } else {
2549 print "not ok 827\n";
2550 }
2551 }
2552
2553 {
2554 my $regex = "ABcde\x{100}";
2555 my $string = "abcDE\x{100}";
2556 chop($regex);
2557 chop($string);
2558 if ($string =~ m/$regex/i) {
2559 print "ok 828\n";
2560 } else {
2561 print "not ok 828\n";
2562 }
2563 }
2564}
d07ddd77 2565
2566{
2567 print "# more SIGMAs\n";
2568
2569 my $SIGMA = "\x{03A3}"; # CAPITAL
2570 my $Sigma = "\x{03C2}"; # SMALL FINAL
2571 my $sigma = "\x{03C3}"; # SMALL
2572
2573 my $S3 = "$SIGMA$Sigma$sigma";
2574
2575 print ":$S3:" =~ /:(($SIGMA)+):/i && $1 eq $S3 && $2 eq $sigma ?
2576 "ok 829\n" : "not ok 829\n";
2577 print ":$S3:" =~ /:(($Sigma)+):/i && $1 eq $S3 && $2 eq $sigma ?
2578 "ok 830\n" : "not ok 830\n";
2579 print ":$S3:" =~ /:(($sigma)+):/i && $1 eq $S3 && $2 eq $sigma ?
2580 "ok 831\n" : "not ok 831\n";
2581
2582 print ":$S3:" =~ /:(([$SIGMA])+):/i && $1 eq $S3 && $2 eq $sigma ?
2583 "ok 832\n" : "not ok 832\n";
2584 print ":$S3:" =~ /:(([$Sigma])+):/i && $1 eq $S3 && $2 eq $sigma ?
2585 "ok 833\n" : "not ok 833\n";
2586 print ":$S3:" =~ /:(([$sigma])+):/i && $1 eq $S3 && $2 eq $sigma ?
2587 "ok 834\n" : "not ok 834\n";
2588}
925f9e00 2589
2590{
2591 print "# LATIN SMALL LETTER SHARP S\n";
2592
2593 use charnames ':full';
2594
2595 print "\N{LATIN SMALL LETTER SHARP S}" =~
2596 /\N{LATIN SMALL LETTER SHARP S}/ ? "ok 835\n" : "not ok 835\n";
2597
2598 print "\N{LATIN SMALL LETTER SHARP S}" =~
2599 /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 836\n" : "not ok 836\n";
2600
2601 print "\N{LATIN SMALL LETTER SHARP S}" =~
2602 /[\N{LATIN SMALL LETTER SHARP S}]/ ? "ok 837\n" : "not ok 837\n";
2603
2604 print "\N{LATIN SMALL LETTER SHARP S}" =~
2605 /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 838\n" : "not ok 838\n";
2606
2607 print "ss" =~
2608 /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 839\n" : "not ok 839\n";
2609
2610 print "SS" =~
2611 /\N{LATIN SMALL LETTER SHARP S}/i ? "ok 840\n" : "not ok 840\n";
2612
e0f9d4a8 2613 print "ss" =~
2614 /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 841\n" : "not ok 841\n";
2615
2616 print "SS" =~
2617 /[\N{LATIN SMALL LETTER SHARP S}]/i ? "ok 842\n" : "not ok 842\n";
5486206c 2618
2619 print "\N{LATIN SMALL LETTER SHARP S}" =~ /ss/i ?
2620 "ok 843\n" : "not ok 843\n";
2621
2622 print "\N{LATIN SMALL LETTER SHARP S}" =~ /SS/i ?
2623 "ok 844\n" : "not ok 844\n";
925f9e00 2624}
d8f6a732 2625
2626{
2627 print "# more whitespace: U+0085, U+2028, U+2029\n";
2628
2629 # U+0085 needs to be forced to be Unicode, the \x{100} does that.
5486206c 2630 print "<\x{100}\x{0085}>" =~ /<\x{100}\s>/ ? "ok 845\n" : "not ok 845\n";
2631 print "<\x{2028}>" =~ /<\s>/ ? "ok 846\n" : "not ok 846\n";
2632 print "<\x{2029}>" =~ /<\s>/ ? "ok 847\n" : "not ok 847\n";
d8f6a732 2633}
2634
def8e4ea 2635{
229196e5 2636 print "# . with /s should work on characters, as opposed to bytes\n";
def8e4ea 2637
2638 my $s = "\x{e4}\x{100}";
2639
2640 # This is not expected to match: the point is that
2641 # neither should we get "Malformed UTF-8" warnings.
2642 print $s =~ /\G(.+?)\n/gcs ?
5486206c 2643 "not ok 848\n" : "ok 848\n";
def8e4ea 2644
2645 my @c;
2646
2647 while ($s =~ /\G(.)/gs) {
2648 push @c, $1;
2649 }
2650
5486206c 2651 print join("", @c) eq $s ? "ok 849\n" : "not ok 849\n";
a0804c9e 2652
2653 my $t1 = "Q003\n\n\x{e4}\x{f6}\n\nQ004\n\n\x{e7}"; # test only chars < 256
2654 my $r1 = "";
2655 while ($t1 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) {
2656 $r1 .= $1 . $2;
2657 }
2658
2659 my $t2 = $t1 . "\x{100}"; # repeat with a larger char
2660 my $r2 = "";
2661 while ($t2 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) {
2662 $r2 .= $1 . $2;
2663 }
2664 $r2 =~ s/\x{100}//;
5486206c 2665 print $r1 eq $r2 ? "ok 850\n" : "not ok 850\n";
def8e4ea 2666}
e54858b0 2667
2668{
2669 print "# Unicode lookbehind\n";
2670
5486206c 2671 print "A\x{100}B" =~ /(?<=A.)B/ ? "ok 851\n" : "not ok 851\n";
2672 print "A\x{200}\x{300}B" =~ /(?<=A..)B/ ? "ok 852\n" : "not ok 852\n";
2673 print "\x{400}AB" =~ /(?<=\x{400}.)B/ ? "ok 853\n" : "not ok 853\n";
2674 print "\x{500\x{600}}B" =~ /(?<=\x{500}.)B/ ? "ok 854\n" : "not ok 854\n";
e54858b0 2675}
c46248c1 2676
2677{
85fd1718 2678 print "# UTF-8 hash keys and /\$/\n";
2679 # http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2002-01/msg01327.html
2680
2681 my $u = "a\x{100}";
2682 my $v = substr($u,0,1);
2683 my $w = substr($u,1,1);
2684 my %u = ( $u => $u, $v => $v, $w => $w );
8d21bda2 2685 my $i = 855;
85fd1718 2686 for (keys %u) {
2687 my $m1 = /^\w*$/ ? 1 : 0;
2688 my $m2 = $u{$_}=~/^\w*$/ ? 1 : 0;
2689 print $m1 == $m2 ? "ok $i\n" : "not ok $i # $m1 $m2\n";
2690 $i++;
2691 }
2692}
8d21bda2 2693
2694{
2695 print "# [ID 20020124.005]\n";
2696 # Fixed by #14795.
2697 my $i = 858;
2698 for my $char ("a", "\x{df}", "\x{100}"){
2699 $x = "$char b $char";
2700 $x =~ s{($char)}{
2701 "c" =~ /c/;
2702 "x";
2703 }ge;
2704 print substr($x,0,1) eq substr($x,-1,1) ?
2705 "ok $i\n" : "not ok $i # debug: $x\n";
2706 $i++;
2707 }
2708}
6e602e29 2709
2710{
2711 print "# SEGV in s/// and UTF-8\n";
2712 $s = "s#\x{100}" x 4;
2713 $s =~ s/[^\w]/ /g;
2714 print $s eq "s \x{100}" x 4 ? "ok 861\n" : "not ok 861\n";
2715}
f272994b 2716
2717{
2718 print "# UTF-8 bug (maybe alreayd known?)\n";
2719 my $u;
2720
2721 $u = "foo";
2722 $u =~ s/./\x{100}/g;
2723 print $u eq "\x{100}\x{100}\x{100}" ? "ok 862\n" : "not ok 862\n";
2724
2725 $u = "foobar";
2726 $u =~ s/[ao]/\x{100}/g;
2727 print $u eq "f\x{100}\x{100}b\x{100}r" ? "ok 863\n" : "not ok 863\n";
2728
2729 $u =~ s/\x{100}/e/g;
2730 print $u eq "feeber" ? "ok 864\n" : "not ok 864\n";
2731}
2732
33b8afdf 2733{
2734 print "# UTF-8 bug with s///\n";
2735 # check utf8/non-utf8 mixtures
2736 # try to force all float/anchored check combinations
2737 my $c = "\x{100}";
2738 my $test = 865;
2739 my $subst;
2740 for my $re (
2741 "xx.*$c", "x.*$c$c", "$c.*xx", "$c$c.*x", "xx.*(?=$c)", "(?=$c).*xx",
2742 ) {
2743 print "xxx" =~ /$re/ ? "not ok $test\n" : "ok $test\n";
2744 ++$test;
2745 print +($subst = "xxx") =~ s/$re// ? "not ok $test\n" : "ok $test\n";
2746 ++$test;
2747 }
2748 for my $re ("xx.*$c*", "$c*.*xx") {
2749 print "xxx" =~ /$re/ ? "ok $test\n" : "not ok $test\n";
2750 ++$test;
2751 ($subst = "xxx") =~ s/$re//;
2752 print $subst eq '' ? "ok $test\n" : "not ok $test\t# $subst\n";
2753 ++$test;
2754 }
2755 for my $re ("xxy*", "y*xx") {
2756 print "xx$c" =~ /$re/ ? "ok $test\n" : "not ok $test\n";
2757 ++$test;
2758 ($subst = "xx$c") =~ s/$re//;
2759 print $subst eq $c ? "ok $test\n" : "not ok $test\n";
2760 ++$test;
2761 print "xy$c" =~ /$re/ ? "not ok $test\n" : "ok $test\n";
2762 ++$test;
2763 print +($subst = "xy$c") =~ /$re/ ? "not ok $test\n" : "ok $test\n";
2764 ++$test;
2765 }
2766 for my $re ("xy$c*z", "x$c*yz") {
2767 print "xyz" =~ /$re/ ? "ok $test\n" : "not ok $test\n";
2768 ++$test;
2769 ($subst = "xyz") =~ s/$re//;
2770 print $subst eq '' ? "ok $test\n" : "not ok $test\n";
2771 ++$test;
2772 }
2773}
574c8022 2774
ff385a1b 2775{
2776 print "# qr/.../x\n";
4c79aee6 2777 my $test = 893;
ff385a1b 2778
2779 my $R = qr/ A B C # D E/x;
2780
2781 print eval {"ABCDE" =~ $R} ? "ok $test\n" : "not ok $test\n";
2782 $test++;
2783
2784 print eval {"ABCDE" =~ m/$R/} ? "ok $test\n" : "not ok $test\n";
2785 $test++;
2786
2787 print eval {"ABCDE" =~ m/($R)/} ? "ok $test\n" : "not ok $test\n";
2788 $test++;
2789}
bc45ce41 2790
2791{
2792 print "# illegal Unicode properties\n";
4c79aee6 2793 my $test = 896;
bc45ce41 2794
2795 print eval { "a" =~ /\pq / } ? "not ok $test\n" : "ok $test\n";
2796 $test++;
2797
2798 print eval { "a" =~ /\p{qrst} / } ? "not ok $test\n" : "ok $test\n";
2799 $test++;
2800}
d65afb4b 2801
2802{
491fd90a 2803 print "# [ID 20020412.005] wrong pmop flags checked when empty pattern\n";
d65afb4b 2804 # requires reuse of last successful pattern
2805 my $test = 898;
2806 $test =~ /\d/;
2807 for (0 .. 1) {
2808 my $match = ?? + 0;
2809 if ($match != $_) {
2810 print "ok $test\n";
2811 } else {
2812 printf "not ok %s\t# 'match once' %s on %s iteration\n", $test,
2813 $match ? 'succeeded' : 'failed', $_ ? 'second' : 'first';
2814 }
2815 ++$test;
2816 }
2817 $test =~ /(\d)/;
2818 my $result = join '', $test =~ //g;
2819 if ($result eq $test) {
2820 print "ok $test\n";
2821 } else {
2822 printf "not ok %s\t# expected '%s', got '%s'\n", $test, $test, $result;
2823 }
2824 ++$test;
2825}
491fd90a 2826
2827print "# user-defined character properties\n";
2828
2829sub InKana1 {
2830 return <<'END';
28313040 309F
283230A0 30FF
2833END
2834}
2835
2836sub InKana2 {
2837 return <<'END';
2838+utf8::InHiragana
2839+utf8::InKatakana
2840END
2841}
2842
2843sub InKana3 {
2844 return <<'END';
2845+utf8::InHiragana
2846+utf8::InKatakana
2847-utf8::IsCn
2848END
2849}
2850
2851sub InNotKana {
2852 return <<'END';
2853!utf8::InHiragana
2854-utf8::InKatakana
2855+utf8::IsCn
2856END
2857}
2858
2859$test = 901;
2860
2861print "\x{3040}" =~ /\p{InKana1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2862print "\x{303F}" =~ /\P{InKana1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2863
2864print "\x{3040}" =~ /\p{InKana2}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2865print "\x{303F}" =~ /\P{InKana2}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2866
2867print "\x{3041}" =~ /\p{InKana3}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2868print "\x{3040}" =~ /\P{InKana3}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2869
2870print "\x{3040}" =~ /\p{InNotKana}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2871print "\x{3041}" =~ /\P{InNotKana}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2872
11ef8fdd 2873sub InConsonant { # Not EBCDIC-aware.
2874 return <<EOF;
28750061 007f
2876-0061
2877-0065
2878-0069
2879-006f
2880-0075
2881EOF
2882}
2883
2884print "d" =~ /\p{InConsonant}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2885print "e" =~ /\P{InConsonant}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2886
faf11cac 2887{
2888 print "# [ID 20020630.002] utf8 regex only matches 32k\n";
2889 $test = 911;
2890 for ([ 'byte', "\x{ff}" ], [ 'utf8', "\x{1ff}" ]) {
2891 my($type, $char) = @$_;
2892 for my $len (32000, 32768, 33000) {
2893 my $s = $char . "f" x $len;
2894 my $r = $s =~ /$char([f]*)/gc;
2895 print $r ? "ok $test\n" : "not ok $test\t# <$type x $len> fail\n";
2896 ++$test;
2897 print +(!$r or pos($s) == $len + 1) ? "ok $test\n"
2898 : "not ok $test\t# <$type x $len> pos @{[ pos($s) ]}\n";
2899 ++$test;
2900 }
2901 }
2902}
2903
2904$test = 923;
f3b1e556 2905
2906$a = bless qr/foo/, 'Foo';
2907print(('goodfood' =~ $a ? '' : 'not '),
2908 "ok $test\t# reblessed qr// matches\n");
2909++$test;
2910
2911print(($a eq '(?-xism:foo)' ? '' : 'not '),
2912 "ok $test\t# reblessed qr// stringizes\n");
2913++$test;
446eaa42 2914
2915$x = "\x{3fe}";
cb50f42d 2916$z=$y = "\317\276"; # $y is byte representation of $x
2917
446eaa42 2918$a = qr/$x/;
2919print(($x =~ $a ? '' : 'not '), "ok $test - utf8 interpolation in qr//\n");
2920++$test;
2921
2922print(("a$a" =~ $x ? '' : 'not '),
cb50f42d 2923 "ok $test - stringifed qr// preserves utf8\n");
2924++$test;
2925
2926print(("a$x" =~ /^a$a\z/ ? '' : 'not '),
2927 "ok $test - interpolated qr// preserves utf8\n");
2928++$test;
2929
2930print(("a$x" =~ /^a(??{$a})\z/ ? '' : 'not '),
2931 "ok $test - postponed interpolation of qr// preserves utf8\n");
2932++$test;
2933
ab01544f 2934print((length(qr/##/x) == 12 ? '' : 'not '),
2935 "ok $test - ## in qr// doesn't corrupt memory [perl #17776]\n");
2936++$test;
2937
cb50f42d 2938{ use re 'eval';
2939
2940print(("$x$x" =~ /^$x(??{$x})\z/ ? '' : 'not '),
2941 "ok $test - postponed utf8 string in utf8 re matches utf8\n");
2942++$test;
2943
2944print(("$y$x" =~ /^$y(??{$x})\z/ ? '' : 'not '),
2945 "ok $test - postponed utf8 string in non-utf8 re matches utf8\n");
446eaa42 2946++$test;
2947
cb50f42d 2948print(("$y$x" !~ /^$y(??{$y})\z/ ? '' : 'not '),
2949 "ok $test - postponed non-utf8 string in non-utf8 re doesn't match utf8\n");
446eaa42 2950++$test;
2951
cb50f42d 2952print(("$x$x" !~ /^$x(??{$y})\z/ ? '' : 'not '),
2953 "ok $test - postponed non-utf8 string in utf8 re doesn't match utf8\n");
446eaa42 2954++$test;
2955
cb50f42d 2956print(("$y$y" =~ /^$y(??{$y})\z/ ? '' : 'not '),
2957 "ok $test - postponed non-utf8 string in non-utf8 re matches non-utf8\n");
2958++$test;
2959
2960print(("$x$y" =~ /^$x(??{$y})\z/ ? '' : 'not '),
2961 "ok $test - postponed non-utf8 string in utf8 re matches non-utf8\n");
2962++$test;
2963$y = $z; # reset $y after upgrade
2964
2965print(("$x$y" !~ /^$x(??{$x})\z/ ? '' : 'not '),
2966 "ok $test - postponed utf8 string in utf8 re doesn't match non-utf8\n");
2967++$test;
2968$y = $z; # reset $y after upgrade
2969
2970print(("$y$y" !~ /^$y(??{$x})\z/ ? '' : 'not '),
2971 "ok $test - postponed utf8 string in non-utf8 re doesn't match non-utf8\n");
2972++$test;
2973
2974} # no re 'eval'
2975
2d92f8a0 2976print "# more user-defined character properties\n";
2977
2978sub IsSyriac1 {
2979 return <<'END';
29800712 072C
29810730 074A
2982END
2983}
2984
2985print "\x{0712}" =~ /\p{IsSyriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2986print "\x{072F}" =~ /\P{IsSyriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2987
2988sub Syriac1 {
2989 return <<'END';
29900712 072C
29910730 074A
2992END
2993}
2994
2995print "\x{0712}" =~ /\p{Syriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2996print "\x{072F}" =~ /\P{Syriac1}/ ? "ok $test\n" : "not ok $test\n"; $test++;
2997
0e933229 2998{
3a2263fe 2999 print "# Change #18179\n";
3000 # previously failed with "panic: end_shift
0e933229 3001 my $s = "\x{100}" x 5;
3002 my $ok = $s =~ /(\x{100}{4})/;
3003 my($ord, $len) = (ord $1, length $1);
3004 print +($ok && $ord == 0x100 && $len == 4)
3005 ? "ok $test\n" : "not ok $test\t# $ok/$ord/$len\n";
3006 ++$test;
3007}
3008
f14c76ed 3009{
3010 print "# [perl #15763]\n";
3011
3012 $a = "x\x{100}";
3013 chop $a; # but leaves the UTF-8 flag
3014 $a .= "y"; # 1 byte before "y"
3015
3016 ok($a =~ /^\C/, 'match one \C on 1-byte UTF-8');
3017 ok($a =~ /^\C{1}/, 'match \C{1}');
3018
3019 ok($a =~ /^\Cy/, 'match \Cy');
3020 ok($a =~ /^\C{1}y/, 'match \C{1}y');
3021
3022 $a = "\x{100}y"; # 2 bytes before "y"
3023
3024 ok($a =~ /^\C/, 'match one \C on 2-byte UTF-8');
3025 ok($a =~ /^\C{1}/, 'match \C{1}');
3026 ok($a =~ /^\C\C/, 'match two \C');
3027 ok($a =~ /^\C{2}/, 'match \C{2}');
3028
3029 ok($a =~ /^\C\C\C/, 'match three \C on 2-byte UTF-8 and a byte');
3030 ok($a =~ /^\C{3}/, 'match \C{3}');
3031
3032 ok($a =~ /^\C\Cy/, 'match two \C');
3033 ok($a =~ /^\C{2}y/, 'match \C{2}');
3034
836995da 3035 ok($a !~ /^\C\C\Cy/, q{don't match three \Cy});
3036 ok($a !~ /^\C{2}\Cy/, q{don't match \C{3}y});
f14c76ed 3037
3038 $a = "\x{1000}y"; # 3 bytes before "y"
3039
3040 ok($a =~ /^\C/, 'match one \C on three-byte UTF-8');
3041 ok($a =~ /^\C{1}/, 'match \C{1}');
3042 ok($a =~ /^\C\C/, 'match two \C');
3043 ok($a =~ /^\C{2}/, 'match \C{2}');
3044 ok($a =~ /^\C\C\C/, 'match three \C');
3045 ok($a =~ /^\C{3}/, 'match \C{3}');
3046
3047 ok($a =~ /^\C\C\C\C/, 'match four \C on three-byte UTF-8 and a byte');
3048 ok($a =~ /^\C{4}/, 'match \C{4}');
3049
3050 ok($a =~ /^\C\C\Cy/, 'match three \Cy');
3051 ok($a =~ /^\C{3}y/, 'match \C{3}y');
3052
836995da 3053 ok($a !~ /^\C\C\C\C\y/, q{don't match four \Cy});
3054 ok($a !~ /^\C{4}y/, q{don't match \C{4}y});
f14c76ed 3055}
3056
5dec093f 3057$_ = 'aaaaaaaaaa';
3058utf8::upgrade($_); chop $_; $\="\n";
3059ok(/[^\s]+/, "m/[^\s]/ utf8");
3060ok(/[^\d]+/, "m/[^\d]/ utf8");
3061ok(($a = $_, $_ =~ s/[^\s]+/./g), "s/[^\s]/ utf8");
3062ok(($a = $_, $a =~ s/[^\d]+/./g), "s/[^\s]/ utf8");
f14c76ed 3063
ce97f9d6 3064ok("\x{100}" =~ /\x{100}/, "[perl #15397]");
3065ok("\x{100}" =~ /(\x{100})/, "[perl #15397]");
3066ok("\x{100}" =~ /(\x{100}){1}/, "[perl #15397]");
3067ok("\x{100}\x{100}" =~ /(\x{100}){2}/, "[perl #15397]");
3068ok("\x{100}\x{100}" =~ /(\x{100})(\x{100})/, "[perl #15397]");
3069
fbeb8e69 3070$x = "CD";
3071$x =~ /(AB)*?CD/;
3072ok(!defined $1, "[perl #7471]");
ce97f9d6 3073
fbeb8e69 3074$x = "CD";
3075$x =~ /(AB)*CD/;
3076ok(!defined $1, "[perl #7471]");
3077
3078# last test 979
Unnamed repository; edit this file to name it for gitweb.