xref: /openbsd/gnu/usr.bin/perl/t/re/pat_advanced.t (revision 891d7ab6)
1#!./perl
2#
3# This is a home for regular expression tests that don't fit into
4# the format supported by re/regexp.t.  If you want to add a test
5# that does fit that format, add it to re/re_tests, not here.
6
7use strict;
8use warnings;
9use 5.010;
10
11
12sub run_tests;
13
14$| = 1;
15
16
17BEGIN {
18    chdir 't' if -d 't';
19    @INC = ('../lib','.');
20    do "re/ReTest.pl" or die $@;
21}
22
23
24plan tests => 1159;  # Update this when adding/deleting tests.
25
26run_tests() unless caller;
27
28#
29# Tests start here.
30#
31sub run_tests {
32
33  SKIP:
34    {
35        local $Message = '\C matches octet';
36        $_ = "a\x{100}b";
37        ok /(.)(\C)(\C)(.)/ or skip q [\C doesn't match], 4;
38        iseq $1, "a";
39        if ($IS_ASCII) {     # ASCII (or equivalent), should be UTF-8
40            iseq $2, "\xC4";
41            iseq $3, "\x80";
42        }
43        elsif ($IS_EBCDIC) { # EBCDIC (or equivalent), should be UTF-EBCDIC
44            iseq $2, "\x8C";
45            iseq $3, "\x41";
46        }
47        else {
48            SKIP: {
49                ok 0, "Unexpected platform", "ord ('A') = $ordA";
50                skip "Unexpected platform";
51            }
52        }
53        iseq $4, "b";
54    }
55
56
57  SKIP:
58    {
59        local $Message = '\C matches octet';
60        $_ = "\x{100}";
61        ok /(\C)/g or skip q [\C doesn't match], 2;
62        if ($IS_ASCII) {
63            iseq $1, "\xC4";
64        }
65        elsif ($IS_EBCDIC) {
66            iseq $1, "\x8C";
67        }
68        else {
69            ok 0, "Unexpected platform", "ord ('A') = $ordA";
70        }
71        ok /(\C)/g or skip q [\C doesn't match];
72        if ($IS_ASCII) {
73            iseq $1, "\x80";
74        }
75        elsif ($IS_EBCDIC) {
76            iseq $1, "\x41";
77        }
78        else {
79            ok 0, "Unexpected platform", "ord ('A') = $ordA";
80        }
81    }
82
83
84    {
85        # Japhy -- added 03/03/2001
86        () = (my $str = "abc") =~ /(...)/;
87        $str = "def";
88        iseq $1, "abc", 'Changing subject does not modify $1';
89    }
90
91
92  SKIP:
93    {
94        # The trick is that in EBCDIC the explicit numeric range should
95        # match (as also in non-EBCDIC) but the explicit alphabetic range
96        # should not match.
97        ok "\x8e" =~ /[\x89-\x91]/, '"\x8e" =~ /[\x89-\x91]/';
98        ok "\xce" =~ /[\xc9-\xd1]/, '"\xce" =~ /[\xc9-\xd1]/';
99
100        skip "Not an EBCDIC platform", 2 unless ord ('i') == 0x89 &&
101                                                ord ('J') == 0xd1;
102
103        # In most places these tests would succeed since \x8e does not
104        # in most character sets match 'i' or 'j' nor would \xce match
105        # 'I' or 'J', but strictly speaking these tests are here for
106        # the good of EBCDIC, so let's test these only there.
107        nok "\x8e" !~ /[i-j]/, '"\x8e" !~ /[i-j]/';
108        nok "\xce" !~ /[I-J]/, '"\xce" !~ /[I-J]/';
109    }
110
111
112    {
113        ok "\x{ab}"   =~ /\x{ab}/,   '"\x{ab}"   =~ /\x{ab}/  ';
114        ok "\x{abcd}" =~ /\x{abcd}/, '"\x{abcd}" =~ /\x{abcd}/';
115    }
116
117
118    {
119        local $Message = 'bug id 20001008.001';
120
121        my @x = ("stra\337e 138", "stra\337e 138");
122        for (@x) {
123            ok s/(\d+)\s*([\w\-]+)/$1 . uc $2/e;
124            ok my ($latin) = /^(.+)(?:\s+\d)/;
125            iseq $latin, "stra\337e";
126        ok $latin =~ s/stra\337e/straße/;
127            #
128            # Previous code follows, but outcommented - there were no tests.
129            #
130            # $latin =~ s/stra\337e/straße/; # \303\237 after the 2nd a
131            # use utf8; # needed for the raw UTF-8
132            # $latin =~ s!(s)tr(?:aß|s+e)!$1tr.!; # \303\237 after the a
133        }
134    }
135
136
137    {
138        local $Message = 'Test \x escapes';
139        ok "ba\xd4c" =~ /([a\xd4]+)/ && $1 eq "a\xd4";
140        ok "ba\xd4c" =~ /([a\xd4]+)/ && $1 eq "a\x{d4}";
141        ok "ba\x{d4}c" =~ /([a\xd4]+)/ && $1 eq "a\x{d4}";
142        ok "ba\x{d4}c" =~ /([a\xd4]+)/ && $1 eq "a\xd4";
143        ok "ba\xd4c" =~ /([a\x{d4}]+)/ && $1 eq "a\xd4";
144        ok "ba\xd4c" =~ /([a\x{d4}]+)/ && $1 eq "a\x{d4}";
145        ok "ba\x{d4}c" =~ /([a\x{d4}]+)/ && $1 eq "a\x{d4}";
146        ok "ba\x{d4}c" =~ /([a\x{d4}]+)/ && $1 eq "a\xd4";
147    }
148
149
150  SKIP:
151    {
152        local $Message = 'Match code points > 255';
153        $_ = "abc\x{100}\x{200}\x{300}\x{380}\x{400}defg";
154        ok /(.\x{300})./ or skip "No match", 4;
155        ok $` eq "abc\x{100}"            && length ($`) == 4;
156        ok $& eq "\x{200}\x{300}\x{380}" && length ($&) == 3;
157        ok $' eq "\x{400}defg"           && length ($') == 5;
158        ok $1 eq "\x{200}\x{300}"        && length ($1) == 2;
159    }
160
161
162
163    {
164        my $x = "\x{10FFFD}";
165        $x =~ s/(.)/$1/g;
166        ok ord($x) == 0x10FFFD && length($x) == 1, "From Robin Houston";
167    }
168
169
170    {
171        my %d = (
172            "7f" => [0, 0, 0],
173            "80" => [1, 1, 0],
174            "ff" => [1, 1, 0],
175           "100" => [0, 1, 1],
176        );
177      SKIP:
178        while (my ($code, $match) = each %d) {
179            local $Message = "Properties of \\x$code";
180            my $char = eval qq ["\\x{$code}"];
181            my $i = 0;
182            ok (($char =~ /[\x80-\xff]/)            xor !$$match [$i ++]);
183            ok (($char =~ /[\x80-\x{100}]/)         xor !$$match [$i ++]);
184            ok (($char =~ /[\x{100}]/)              xor !$$match [$i ++]);
185        }
186    }
187
188
189    {
190        # From Japhy
191        local $Message;
192        must_warn 'qr/(?c)/',    '^Useless \(\?c\)';
193        must_warn 'qr/(?-c)/',   '^Useless \(\?-c\)';
194        must_warn 'qr/(?g)/',    '^Useless \(\?g\)';
195        must_warn 'qr/(?-g)/',   '^Useless \(\?-g\)';
196        must_warn 'qr/(?o)/',    '^Useless \(\?o\)';
197        must_warn 'qr/(?-o)/',   '^Useless \(\?-o\)';
198
199        # Now test multi-error regexes
200        must_warn 'qr/(?g-o)/',  '^Useless \(\?g\).*\nUseless \(\?-o\)';
201        must_warn 'qr/(?g-c)/',  '^Useless \(\?g\).*\nUseless \(\?-c\)';
202        # (?c) means (?g) error won't be thrown
203        must_warn 'qr/(?o-cg)/', '^Useless \(\?o\).*\nUseless \(\?-c\)';
204        must_warn 'qr/(?ogc)/',  '^Useless \(\?o\).*\nUseless \(\?g\).*\n' .
205                                  'Useless \(\?c\)';
206    }
207
208
209    {
210        local $Message = "/x tests";
211        $_ = "foo";
212        eval_ok <<"        --";
213          /f
214           o\r
215           o
216           \$
217          /x
218        --
219        eval_ok <<"        --";
220          /f
221           o
222           o
223           \$\r
224          /x
225        --
226    }
227
228
229    {
230        local $Message = "/o feature";
231        sub test_o {$_ [0] =~ /$_[1]/o; return $1}
232        iseq test_o ('abc', '(.)..'), 'a';
233        iseq test_o ('abc', '..(.)'), 'a';
234    }
235
236    {
237        # Test basic $^N usage outside of a regex
238        local $Message = '$^N usage outside of a regex';
239        my $x = "abcdef";
240        ok ($x =~ /cde/                  and !defined $^N);
241        ok ($x =~ /(cde)/                and $^N eq "cde");
242        ok ($x =~ /(c)(d)(e)/            and $^N eq   "e");
243        ok ($x =~ /(c(d)e)/              and $^N eq "cde");
244        ok ($x =~ /(foo)|(c(d)e)/        and $^N eq "cde");
245        ok ($x =~ /(c(d)e)|(foo)/        and $^N eq "cde");
246        ok ($x =~ /(c(d)e)|(abc)/        and $^N eq "abc");
247        ok ($x =~ /(c(d)e)|(abc)x/       and $^N eq "cde");
248        ok ($x =~ /(c(d)e)(abc)?/        and $^N eq "cde");
249        ok ($x =~ /(?:c(d)e)/            and $^N eq   "d");
250        ok ($x =~ /(?:c(d)e)(?:f)/       and $^N eq   "d");
251        ok ($x =~ /(?:([abc])|([def]))*/ and $^N eq   "f");
252        ok ($x =~ /(?:([ace])|([bdf]))*/ and $^N eq   "f");
253        ok ($x =~ /(([ace])|([bd]))*/    and $^N eq   "e");
254       {ok ($x =~ /(([ace])|([bdf]))*/   and $^N eq   "f");}
255        ## Test to see if $^N is automatically localized -- it should now
256        ## have the value set in the previous test.
257        iseq $^N, "e", '$^N is automatically localized';
258
259        # Now test inside (?{ ... })
260        local $Message = '$^N usage inside (?{ ... })';
261        our ($y, $z);
262        ok ($x =~ /a([abc])(?{$y=$^N})c/                    and $y eq  "b");
263        ok ($x =~ /a([abc]+)(?{$y=$^N})d/                   and $y eq  "bc");
264        ok ($x =~ /a([abcdefg]+)(?{$y=$^N})d/               and $y eq  "bc");
265        ok ($x =~ /(a([abcdefg]+)(?{$y=$^N})d)(?{$z=$^N})e/ and $y eq  "bc"
266                                                            and $z eq "abcd");
267        ok ($x =~ /(a([abcdefg]+)(?{$y=$^N})de)(?{$z=$^N})/ and $y eq  "bc"
268                                                            and $z eq "abcde");
269
270    }
271
272
273  SKIP:
274    {
275        ## Should probably put in tests for all the POSIX stuff,
276        ## but not sure how to guarantee a specific locale......
277
278        skip "Not an ASCII platform", 2 unless $IS_ASCII;
279        local $Message = 'Test [[:cntrl:]]';
280        my $AllBytes = join "" => map {chr} 0 .. 255;
281        (my $x = $AllBytes) =~ s/[[:cntrl:]]//g;
282        iseq $x, join "", map {chr} 0x20 .. 0x7E, 0x80 .. 0xFF;
283
284        ($x = $AllBytes) =~ s/[^[:cntrl:]]//g;
285        iseq $x, join "", map {chr} 0x00 .. 0x1F, 0x7F;
286    }
287
288
289    {
290        # With /s modifier UTF8 chars were interpreted as bytes
291        local $Message = "UTF-8 chars aren't bytes";
292        my $a = "Hello \x{263A} World";
293        my @a = ($a =~ /./gs);
294        iseq $#a, 12;
295    }
296
297
298    {
299        local $Message = '. matches \n with /s';
300        my $str1 = "foo\nbar";
301        my $str2 = "foo\n\x{100}bar";
302        my ($a, $b) = map {chr} $IS_ASCII ? (0xc4, 0x80) : (0x8c, 0x41);
303        my @a;
304        @a = $str1 =~ /./g;   iseq @a, 6; iseq "@a", "f o o b a r";
305        @a = $str1 =~ /./gs;  iseq @a, 7; iseq "@a", "f o o \n b a r";
306        @a = $str1 =~ /\C/g;  iseq @a, 7; iseq "@a", "f o o \n b a r";
307        @a = $str1 =~ /\C/gs; iseq @a, 7; iseq "@a", "f o o \n b a r";
308        @a = $str2 =~ /./g;   iseq @a, 7; iseq "@a", "f o o \x{100} b a r";
309        @a = $str2 =~ /./gs;  iseq @a, 8; iseq "@a", "f o o \n \x{100} b a r";
310        @a = $str2 =~ /\C/g;  iseq @a, 9; iseq "@a", "f o o \n $a $b b a r";
311        @a = $str2 =~ /\C/gs; iseq @a, 9; iseq "@a", "f o o \n $a $b b a r";
312    }
313
314
315    {
316        no warnings 'digit';
317        # Check that \x## works. 5.6.1 and 5.005_03 fail some of these.
318        my $x;
319        $x = "\x4e" . "E";
320        ok ($x =~ /^\x4EE$/, "Check only 2 bytes of hex are matched.");
321
322        $x = "\x4e" . "i";
323        ok ($x =~ /^\x4Ei$/, "Check that invalid hex digit stops it (2)");
324
325        $x = "\x4" . "j";
326        ok ($x =~ /^\x4j$/,  "Check that invalid hex digit stops it (1)");
327
328        $x = "\x0" . "k";
329        ok ($x =~ /^\xk$/,   "Check that invalid hex digit stops it (0)");
330
331        $x = "\x0" . "x";
332        ok ($x =~ /^\xx$/, "\\xx isn't to be treated as \\0");
333
334        $x = "\x0" . "xa";
335        ok ($x =~ /^\xxa$/, "\\xxa isn't to be treated as \\xa");
336
337        $x = "\x9" . "_b";
338        ok ($x =~ /^\x9_b$/, "\\x9_b isn't to be treated as \\x9b");
339
340        # and now again in [] ranges
341
342        $x = "\x4e" . "E";
343        ok ($x =~ /^[\x4EE]{2}$/, "Check only 2 bytes of hex are matched.");
344
345        $x = "\x4e" . "i";
346        ok ($x =~ /^[\x4Ei]{2}$/, "Check that invalid hex digit stops it (2)");
347
348        $x = "\x4" . "j";
349        ok ($x =~ /^[\x4j]{2}$/,  "Check that invalid hex digit stops it (1)");
350
351        $x = "\x0" . "k";
352        ok ($x =~ /^[\xk]{2}$/,   "Check that invalid hex digit stops it (0)");
353
354        $x = "\x0" . "x";
355        ok ($x =~ /^[\xx]{2}$/, "\\xx isn't to be treated as \\0");
356
357        $x = "\x0" . "xa";
358        ok ($x =~ /^[\xxa]{3}$/, "\\xxa isn't to be treated as \\xa");
359
360        $x = "\x9" . "_b";
361        ok ($x =~ /^[\x9_b]{3}$/, "\\x9_b isn't to be treated as \\x9b");
362
363        # Check that \x{##} works. 5.6.1 fails quite a few of these.
364
365        $x = "\x9b";
366        ok ($x =~ /^\x{9_b}$/, "\\x{9_b} is to be treated as \\x9b");
367
368        $x = "\x9b" . "y";
369        ok ($x =~ /^\x{9_b}y$/, "\\x{9_b} is to be treated as \\x9b (again)");
370
371        $x = "\x9b" . "y";
372        ok ($x =~ /^\x{9b_}y$/, "\\x{9b_} is to be treated as \\x9b");
373
374        $x = "\x9b" . "y";
375        ok ($x =~ /^\x{9_bq}y$/, "\\x{9_bc} is to be treated as \\x9b");
376
377        $x = "\x0" . "y";
378        ok ($x =~ /^\x{x9b}y$/, "\\x{x9b} is to be treated as \\x0");
379
380        $x = "\x0" . "y";
381        ok ($x =~ /^\x{0x9b}y$/, "\\x{0x9b} is to be treated as \\x0");
382
383        $x = "\x9b" . "y";
384        ok ($x =~ /^\x{09b}y$/, "\\x{09b} is to be treated as \\x9b");
385
386        $x = "\x9b";
387        ok ($x =~ /^[\x{9_b}]$/, "\\x{9_b} is to be treated as \\x9b");
388
389        $x = "\x9b" . "y";
390        ok ($x =~ /^[\x{9_b}y]{2}$/,
391                                 "\\x{9_b} is to be treated as \\x9b (again)");
392
393        $x = "\x9b" . "y";
394        ok ($x =~ /^[\x{9b_}y]{2}$/, "\\x{9b_} is to be treated as \\x9b");
395
396        $x = "\x9b" . "y";
397        ok ($x =~ /^[\x{9_bq}y]{2}$/, "\\x{9_bc} is to be treated as \\x9b");
398
399        $x = "\x0" . "y";
400        ok ($x =~ /^[\x{x9b}y]{2}$/, "\\x{x9b} is to be treated as \\x0");
401
402        $x = "\x0" . "y";
403        ok ($x =~ /^[\x{0x9b}y]{2}$/, "\\x{0x9b} is to be treated as \\x0");
404
405        $x = "\x9b" . "y";
406        ok ($x =~ /^[\x{09b}y]{2}$/, "\\x{09b} is to be treated as \\x9b");
407
408    }
409
410
411    {
412        # High bit bug -- japhy
413        my $x = "ab\200d";
414        ok $x =~ /.*?\200/, "High bit fine";
415    }
416
417
418    {
419        # The basic character classes and Unicode
420        ok "\x{0100}" =~ /\w/, 'LATIN CAPITAL LETTER A WITH MACRON in /\w/';
421        ok "\x{0660}" =~ /\d/, 'ARABIC-INDIC DIGIT ZERO in /\d/';
422        ok "\x{1680}" =~ /\s/, 'OGHAM SPACE MARK in /\s/';
423    }
424
425
426    {
427        local $Message = "Folding matches and Unicode";
428        ok "a\x{100}" =~ /A/i;
429        ok "A\x{100}" =~ /a/i;
430        ok "a\x{100}" =~ /a/i;
431        ok "A\x{100}" =~ /A/i;
432        ok "\x{101}a" =~ /\x{100}/i;
433        ok "\x{100}a" =~ /\x{100}/i;
434        ok "\x{101}a" =~ /\x{101}/i;
435        ok "\x{100}a" =~ /\x{101}/i;
436        ok "a\x{100}" =~ /A\x{100}/i;
437        ok "A\x{100}" =~ /a\x{100}/i;
438        ok "a\x{100}" =~ /a\x{100}/i;
439        ok "A\x{100}" =~ /A\x{100}/i;
440        ok "a\x{100}" =~ /[A]/i;
441        ok "A\x{100}" =~ /[a]/i;
442        ok "a\x{100}" =~ /[a]/i;
443        ok "A\x{100}" =~ /[A]/i;
444        ok "\x{101}a" =~ /[\x{100}]/i;
445        ok "\x{100}a" =~ /[\x{100}]/i;
446        ok "\x{101}a" =~ /[\x{101}]/i;
447        ok "\x{100}a" =~ /[\x{101}]/i;
448    }
449
450
451    {
452        use charnames ':full';
453        local $Message = "Folding 'LATIN LETTER A WITH GRAVE'";
454
455        my $lower = "\N{LATIN SMALL LETTER A WITH GRAVE}";
456        my $UPPER = "\N{LATIN CAPITAL LETTER A WITH GRAVE}";
457
458        ok $lower =~ m/$UPPER/i;
459        ok $UPPER =~ m/$lower/i;
460        ok $lower =~ m/[$UPPER]/i;
461        ok $UPPER =~ m/[$lower]/i;
462
463        local $Message = "Folding 'GREEK LETTER ALPHA WITH VRACHY'";
464
465        $lower = "\N{GREEK CAPITAL LETTER ALPHA WITH VRACHY}";
466        $UPPER = "\N{GREEK SMALL LETTER ALPHA WITH VRACHY}";
467
468        ok $lower =~ m/$UPPER/i;
469        ok $UPPER =~ m/$lower/i;
470        ok $lower =~ m/[$UPPER]/i;
471        ok $UPPER =~ m/[$lower]/i;
472
473        local $Message = "Folding 'LATIN LETTER Y WITH DIAERESIS'";
474
475        $lower = "\N{LATIN SMALL LETTER Y WITH DIAERESIS}";
476        $UPPER = "\N{LATIN CAPITAL LETTER Y WITH DIAERESIS}";
477
478        ok $lower =~ m/$UPPER/i;
479        ok $UPPER =~ m/$lower/i;
480        ok $lower =~ m/[$UPPER]/i;
481        ok $UPPER =~ m/[$lower]/i;
482    }
483
484
485    {
486        use charnames ':full';
487        local $PatchId = "13843";
488        local $Message = "GREEK CAPITAL LETTER SIGMA vs " .
489                         "COMBINING GREEK PERISPOMENI";
490
491        my $SIGMA = "\N{GREEK CAPITAL LETTER SIGMA}";
492        my $char  = "\N{COMBINING GREEK PERISPOMENI}";
493
494        may_not_warn sub {ok "_:$char:_" !~ m/_:$SIGMA:_/i};
495    }
496
497
498    {
499        local $Message = '\X';
500        use charnames ':full';
501
502        ok "a!"                          =~ /^(\X)!/ && $1 eq "a";
503        ok "\xDF!"                       =~ /^(\X)!/ && $1 eq "\xDF";
504        ok "\x{100}!"                    =~ /^(\X)!/ && $1 eq "\x{100}";
505        ok "\x{100}\x{300}!"             =~ /^(\X)!/ && $1 eq "\x{100}\x{300}";
506        ok "\N{LATIN CAPITAL LETTER E}!" =~ /^(\X)!/ &&
507               $1 eq "\N{LATIN CAPITAL LETTER E}";
508        ok "\N{LATIN CAPITAL LETTER E}\N{COMBINING GRAVE ACCENT}!"
509                                         =~ /^(\X)!/ &&
510               $1 eq "\N{LATIN CAPITAL LETTER E}\N{COMBINING GRAVE ACCENT}";
511
512        local $Message = '\C and \X';
513        ok "!abc!" =~ /a\Cc/;
514        ok "!abc!" =~ /a\Xc/;
515    }
516
517
518    {
519        local $Message = "Final Sigma";
520
521        my $SIGMA = "\x{03A3}"; # CAPITAL
522        my $Sigma = "\x{03C2}"; # SMALL FINAL
523        my $sigma = "\x{03C3}"; # SMALL
524
525        ok $SIGMA =~ /$SIGMA/i;
526        ok $SIGMA =~ /$Sigma/i;
527        ok $SIGMA =~ /$sigma/i;
528
529        ok $Sigma =~ /$SIGMA/i;
530        ok $Sigma =~ /$Sigma/i;
531        ok $Sigma =~ /$sigma/i;
532
533        ok $sigma =~ /$SIGMA/i;
534        ok $sigma =~ /$Sigma/i;
535        ok $sigma =~ /$sigma/i;
536
537        ok $SIGMA =~ /[$SIGMA]/i;
538        ok $SIGMA =~ /[$Sigma]/i;
539        ok $SIGMA =~ /[$sigma]/i;
540
541        ok $Sigma =~ /[$SIGMA]/i;
542        ok $Sigma =~ /[$Sigma]/i;
543        ok $Sigma =~ /[$sigma]/i;
544
545        ok $sigma =~ /[$SIGMA]/i;
546        ok $sigma =~ /[$Sigma]/i;
547        ok $sigma =~ /[$sigma]/i;
548
549        local $Message = "More final Sigma";
550
551        my $S3 = "$SIGMA$Sigma$sigma";
552
553        ok ":$S3:" =~ /:(($SIGMA)+):/i   && $1 eq $S3 && $2 eq $sigma;
554        ok ":$S3:" =~ /:(($Sigma)+):/i   && $1 eq $S3 && $2 eq $sigma;
555        ok ":$S3:" =~ /:(($sigma)+):/i   && $1 eq $S3 && $2 eq $sigma;
556
557        ok ":$S3:" =~ /:(([$SIGMA])+):/i && $1 eq $S3 && $2 eq $sigma;
558        ok ":$S3:" =~ /:(([$Sigma])+):/i && $1 eq $S3 && $2 eq $sigma;
559        ok ":$S3:" =~ /:(([$sigma])+):/i && $1 eq $S3 && $2 eq $sigma;
560    }
561
562
563    {
564        use charnames ':full';
565        local $Message = "Parlez-Vous " .
566                         "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais?";
567
568        ok "Fran\N{LATIN SMALL LETTER C}ais" =~ /Fran.ais/ &&
569            $& eq "Francais";
570        ok "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~ /Fran.ais/ &&
571            $& eq "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais";
572        ok "Fran\N{LATIN SMALL LETTER C}ais" =~ /Fran\Cais/ &&
573            $& eq "Francais";
574        # COMBINING CEDILLA is two bytes when encoded
575        ok "Franc\N{COMBINING CEDILLA}ais" =~ /Franc\C\Cais/;
576        ok "Fran\N{LATIN SMALL LETTER C}ais" =~ /Fran\Xais/ &&
577            $& eq "Francais";
578        ok "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~ /Fran\Xais/  &&
579            $& eq "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais";
580        ok "Franc\N{COMBINING CEDILLA}ais" =~ /Fran\Xais/ &&
581            $& eq "Franc\N{COMBINING CEDILLA}ais";
582        ok "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~
583           /Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais/  &&
584            $& eq "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais";
585        ok "Franc\N{COMBINING CEDILLA}ais" =~ /Franc\N{COMBINING CEDILLA}ais/ &&
586            $& eq "Franc\N{COMBINING CEDILLA}ais";
587
588        my @f = (
589            ["Fran\N{LATIN SMALL LETTER C}ais",                    "Francais"],
590            ["Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais",
591                               "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais"],
592            ["Franc\N{COMBINING CEDILLA}ais", "Franc\N{COMBINING CEDILLA}ais"],
593        );
594        foreach my $entry (@f) {
595            my ($subject, $match) = @$entry;
596            ok $subject =~ /Fran(?:c\N{COMBINING CEDILLA}?|
597                    \N{LATIN SMALL LETTER C WITH CEDILLA})ais/x &&
598               $& eq $match;
599        }
600    }
601
602
603    {
604        local $Message = "Lingering (and useless) UTF8 flag doesn't mess up /i";
605        my $pat = "ABcde";
606        my $str = "abcDE\x{100}";
607        chop $str;
608        ok $str =~ /$pat/i;
609
610        $pat = "ABcde\x{100}";
611        $str = "abcDE";
612        chop $pat;
613        ok $str =~ /$pat/i;
614
615        $pat = "ABcde\x{100}";
616        $str = "abcDE\x{100}";
617        chop $pat;
618        chop $str;
619        ok $str =~ /$pat/i;
620    }
621
622
623    {
624        use charnames ':full';
625        local $Message = "LATIN SMALL LETTER SHARP S " .
626                         "(\N{LATIN SMALL LETTER SHARP S})";
627
628        ok "\N{LATIN SMALL LETTER SHARP S}" =~
629                                            /\N{LATIN SMALL LETTER SHARP S}/;
630        ok "\N{LATIN SMALL LETTER SHARP S}" =~
631                                            /\N{LATIN SMALL LETTER SHARP S}/i;
632        ok "\N{LATIN SMALL LETTER SHARP S}" =~
633                                           /[\N{LATIN SMALL LETTER SHARP S}]/;
634        ok "\N{LATIN SMALL LETTER SHARP S}" =~
635                                           /[\N{LATIN SMALL LETTER SHARP S}]/i;
636
637        ok "ss" =~  /\N{LATIN SMALL LETTER SHARP S}/i;
638        ok "SS" =~  /\N{LATIN SMALL LETTER SHARP S}/i;
639        ok "ss" =~ /[\N{LATIN SMALL LETTER SHARP S}]/i;
640        ok "SS" =~ /[\N{LATIN SMALL LETTER SHARP S}]/i;
641
642        ok "\N{LATIN SMALL LETTER SHARP S}" =~ /ss/i;
643        ok "\N{LATIN SMALL LETTER SHARP S}" =~ /SS/i;
644
645        local $Message = "Unoptimized named sequence in class";
646        ok "ss" =~ /[\N{LATIN SMALL LETTER SHARP S}x]/i;
647        ok "SS" =~ /[\N{LATIN SMALL LETTER SHARP S}x]/i;
648        ok "\N{LATIN SMALL LETTER SHARP S}" =~
649          /[\N{LATIN SMALL LETTER SHARP S}x]/;
650        ok "\N{LATIN SMALL LETTER SHARP S}" =~
651          /[\N{LATIN SMALL LETTER SHARP S}x]/i;
652    }
653
654
655    {
656        # More whitespace: U+0085, U+2028, U+2029\n";
657
658        # U+0085, U+00A0 need to be forced to be Unicode, the \x{100} does that.
659      SKIP: {
660          skip "EBCDIC platform", 4 if $IS_EBCDIC;
661          # Do \x{0015} and \x{0041} match \s in EBCDIC?
662          ok "<\x{100}\x{0085}>" =~ /<\x{100}\s>/, '\x{0085} in \s';
663          ok        "<\x{0085}>" =~        /<\v>/, '\x{0085} in \v';
664          ok "<\x{100}\x{00A0}>" =~ /<\x{100}\s>/, '\x{00A0} in \s';
665          ok        "<\x{00A0}>" =~        /<\h>/, '\x{00A0} in \h';
666        }
667        my @h = map {sprintf "%05x" => $_} 0x01680, 0x0180E, 0x02000 .. 0x0200A,
668                                           0x0202F, 0x0205F, 0x03000;
669        my @v = map {sprintf "%05x" => $_} 0x02028, 0x02029;
670
671        my @H = map {sprintf "%05x" => $_} 0x01361,   0x0200B, 0x02408, 0x02420,
672                                           0x0303F,   0xE0020;
673        my @V = map {sprintf "%05x" => $_} 0x0008A .. 0x0008D, 0x00348, 0x10100,
674                                           0xE005F,   0xE007C;
675
676        for my $hex (@h) {
677            my $str = eval qq ["<\\x{$hex}>"];
678            ok $str =~ /<\s>/, "\\x{$hex} in \\s";
679            ok $str =~ /<\h>/, "\\x{$hex} in \\h";
680            ok $str !~ /<\v>/, "\\x{$hex} not in \\v";
681        }
682
683        for my $hex (@v) {
684            my $str = eval qq ["<\\x{$hex}>"];
685            ok $str =~ /<\s>/, "\\x{$hex} in \\s";
686            ok $str =~ /<\v>/, "\\x{$hex} in \\v";
687            ok $str !~ /<\h>/, "\\x{$hex} not in \\h";
688        }
689
690        for my $hex (@H) {
691            my $str = eval qq ["<\\x{$hex}>"];
692            ok $str =~ /<\S>/, "\\x{$hex} in \\S";
693            ok $str =~ /<\H>/, "\\x{$hex} in \\H";
694        }
695
696        for my $hex (@V) {
697            my $str = eval qq ["<\\x{$hex}>"];
698            ok $str =~ /<\S>/, "\\x{$hex} in \\S";
699            ok $str =~ /<\V>/, "\\x{$hex} in \\V";
700        }
701    }
702
703
704    {
705        # . with /s should work on characters, as opposed to bytes
706        local $Message = ". with /s works on characters, not bytes";
707
708        my $s = "\x{e4}\x{100}";
709        # This is not expected to match: the point is that
710        # neither should we get "Malformed UTF-8" warnings.
711        may_not_warn sub {$s =~ /\G(.+?)\n/gcs}, "No 'Malformed UTF-8' warning";
712
713        my @c;
714        push @c => $1 while $s =~ /\G(.)/gs;
715
716        local $" = "";
717        iseq "@c", $s;
718
719        # Test only chars < 256
720        my $t1 = "Q003\n\n\x{e4}\x{f6}\n\nQ004\n\n\x{e7}";
721        my $r1 = "";
722        while ($t1 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) {
723        $r1 .= $1 . $2;
724        }
725
726        my $t2 = $t1 . "\x{100}"; # Repeat with a larger char
727        my $r2 = "";
728        while ($t2 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) {
729        $r2 .= $1 . $2;
730        }
731        $r2 =~ s/\x{100}//;
732
733        iseq $r1, $r2;
734    }
735
736
737    {
738        local $Message = "Unicode lookbehind";
739        ok "A\x{100}B"        =~ /(?<=A.)B/;
740        ok "A\x{200}\x{300}B" =~ /(?<=A..)B/;
741        ok "\x{400}AB"        =~ /(?<=\x{400}.)B/;
742        ok "\x{500}\x{600}B"  =~ /(?<=\x{500}.)B/;
743
744        # Original code also contained:
745        # ok "\x{500\x{600}}B"  =~ /(?<=\x{500}.)B/;
746        # but that looks like a typo.
747    }
748
749
750    {
751        local $Message = 'UTF-8 hash keys and /$/';
752        # http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters
753        #                                         /2002-01/msg01327.html
754
755        my $u = "a\x{100}";
756        my $v = substr ($u, 0, 1);
757        my $w = substr ($u, 1, 1);
758        my %u = ($u => $u, $v => $v, $w => $w);
759        for (keys %u) {
760            my $m1 =            /^\w*$/ ? 1 : 0;
761            my $m2 = $u {$_} =~ /^\w*$/ ? 1 : 0;
762            iseq $m1, $m2;
763        }
764    }
765
766
767    {
768        local $Message = "No SEGV in s/// and UTF-8";
769        my $s = "s#\x{100}" x 4;
770        ok $s =~ s/[^\w]/ /g;
771        if ( 1 or $ENV{PERL_TEST_LEGACY_POSIX_CC} ) {
772            iseq $s, "s \x{100}" x 4;
773        }
774        else {
775            iseq $s, "s  " x 4;
776        }
777    }
778
779
780    {
781        local $Message = "UTF-8 bug (maybe already known?)";
782        my $u = "foo";
783        $u =~ s/./\x{100}/g;
784        iseq $u, "\x{100}\x{100}\x{100}";
785
786        $u = "foobar";
787        $u =~ s/[ao]/\x{100}/g;
788        iseq $u, "f\x{100}\x{100}b\x{100}r";
789
790        $u =~ s/\x{100}/e/g;
791        iseq $u, "feeber";
792    }
793
794
795    {
796        local $Message = "UTF-8 bug with s///";
797        # check utf8/non-utf8 mixtures
798        # try to force all float/anchored check combinations
799
800        my $c = "\x{100}";
801        my $subst;
802        for my $re ("xx.*$c", "x.*$c$c", "$c.*xx", "$c$c.*x",
803                    "xx.*(?=$c)", "(?=$c).*xx",) {
804            ok "xxx" !~ /$re/;
805            ok +($subst = "xxx") !~ s/$re//;
806        }
807        for my $re ("xx.*$c*", "$c*.*xx") {
808            ok "xxx" =~ /$re/;
809            ok +($subst = "xxx") =~ s/$re//;
810            iseq $subst, "";
811        }
812        for my $re ("xxy*", "y*xx") {
813            ok "xx$c" =~ /$re/;
814            ok +($subst = "xx$c") =~ s/$re//;
815            iseq $subst, $c;
816            ok "xy$c" !~ /$re/;
817            ok +($subst = "xy$c") !~ s/$re//;
818        }
819        for my $re ("xy$c*z", "x$c*yz") {
820            ok "xyz" =~ /$re/;
821            ok +($subst = "xyz") =~ s/$re//;
822            iseq $subst, "";
823        }
824    }
825
826
827    {
828        local $Message = "qr /.../x";
829        my $R = qr / A B C # D E/x;
830        ok "ABCDE" =~    $R   && $& eq "ABC";
831        ok "ABCDE" =~   /$R/  && $& eq "ABC";
832        ok "ABCDE" =~  m/$R/  && $& eq "ABC";
833        ok "ABCDE" =~  /($R)/ && $1 eq "ABC";
834        ok "ABCDE" =~ m/($R)/ && $1 eq "ABC";
835    }
836
837
838
839
840    {
841        local $\;
842        $_ = 'aaaaaaaaaa';
843        utf8::upgrade($_); chop $_; $\="\n";
844        ok /[^\s]+/, 'm/[^\s]/ utf8';
845        ok /[^\d]+/, 'm/[^\d]/ utf8';
846        ok +($a = $_, $_ =~ s/[^\s]+/./g), 's/[^\s]/ utf8';
847        ok +($a = $_, $a =~ s/[^\d]+/./g), 's/[^\s]/ utf8';
848    }
849
850
851
852
853    {
854        # Subject: Odd regexp behavior
855        # From: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk>
856        # Date: Wed, 26 Feb 2003 16:53:12 +0000
857        # Message-Id: <E18o4nw-0008Ly-00@wisbech.cl.cam.ac.uk>
858        # To: perl-unicode@perl.org
859
860        local $Message = 'Markus Kuhn 2003-02-26';
861
862        my $x = "\x{2019}\nk";
863        ok $x =~ s/(\S)\n(\S)/$1 $2/sg;
864        ok $x eq "\x{2019} k";
865
866        $x = "b\nk";
867        ok $x =~ s/(\S)\n(\S)/$1 $2/sg;
868        ok $x eq "b k";
869
870        ok "\x{2019}" =~ /\S/;
871    }
872
873
874    {
875        # XXX DAPM 13-Apr-06. Recursive split is still broken. It's only luck it
876        # hasn't been crashing. Disable this test until it is fixed properly.
877        # XXX also check what it returns rather than just doing ok(1,...)
878        # split /(?{ split "" })/, "abc";
879        local $TODO = "Recursive split is still broken";
880        ok 0, 'cache_re & "(?{": it dumps core in 5.6.1 & 5.8.0';
881    }
882
883
884    {
885        ok "\x{100}\n" =~ /\x{100}\n$/, "UTF-8 length cache and fbm_compile";
886    }
887
888
889    {
890        package Str;
891        use overload q /""/ => sub {${$_ [0]};};
892        sub new {my ($c, $v) = @_; bless \$v, $c;}
893
894        package main;
895        $_ = Str -> new ("a\x{100}/\x{100}b");
896        ok join (":", /\b(.)\x{100}/g) eq "a:/", "re_intuit_start and PL_bostr";
897    }
898
899    {
900        my $re = qq /^([^X]*)X/;
901        utf8::upgrade ($re);
902        ok "\x{100}X" =~ /$re/, "S_cl_and ANYOF_UNICODE & ANYOF_INVERTED";
903    }
904
905    {
906        ok "123\x{100}" =~ /^.*1.*23\x{100}$/,
907           'UTF-8 + multiple floating substr';
908    }
909
910    {
911        local $Message = '<20030808193656.5109.1@llama.ni-s.u-net.com>';
912
913        # LATIN SMALL/CAPITAL LETTER A WITH MACRON
914        ok "  \x{101}" =~ qr/\x{100}/i;
915
916        # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW
917        ok "  \x{1E01}" =~ qr/\x{1E00}/i;
918
919        # DESERET SMALL/CAPITAL LETTER LONG I
920        ok "  \x{10428}" =~ qr/\x{10400}/i;
921
922        # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X'
923        ok "  \x{1E01}x" =~ qr/\x{1E00}X/i;
924    }
925
926    {
927        for (120 .. 130) {
928            my $head = 'x' x $_;
929            local $Message = q [Don't misparse \x{...} in regexp ] .
930                             q [near 127 char EXACT limit];
931            for my $tail ('\x{0061}', '\x{1234}', '\x61') {
932                eval_ok qq ["$head$tail" =~ /$head$tail/];
933            }
934            local $Message = q [Don't misparse \N{...} in regexp ] .
935                             q [near 127 char EXACT limit];
936            for my $tail ('\N{SNOWFLAKE}') {
937                eval_ok qq [use charnames ':full';
938                           "$head$tail" =~ /$head$tail/];
939            }
940        }
941    }
942
943    {   # TRIE related
944        our @got = ();
945        "words" =~ /(word|word|word)(?{push @got, $1})s$/;
946        iseq @got, 1, "TRIE optimation";
947
948        @got = ();
949        "words" =~ /(word|word|word)(?{push @got,$1})s$/i;
950        iseq @got, 1,"TRIEF optimisation";
951
952        my @nums = map {int rand 1000} 1 .. 100;
953        my $re = "(" . (join "|", @nums) . ")";
954        $re = qr/\b$re\b/;
955
956        foreach (@nums) {
957            ok $_ =~ /$re/, "Trie nums";
958        }
959
960        $_ = join " ", @nums;
961        @got = ();
962        push @got, $1 while /$re/g;
963
964        my %count;
965        $count {$_} ++ for @got;
966        my $ok = 1;
967        for (@nums) {
968            $ok = 0 if --$count {$_} < 0;
969        }
970        ok $ok, "Trie min count matches";
971    }
972
973
974    {
975        # TRIE related
976        # LATIN SMALL/CAPITAL LETTER A WITH MACRON
977        ok "foba  \x{101}foo" =~ qr/(foo|\x{100}foo|bar)/i &&
978           $1 eq "\x{101}foo",
979           "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH MACRON";
980
981        # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW
982        ok "foba  \x{1E01}foo" =~ qr/(foo|\x{1E00}foo|bar)/i &&
983           $1 eq "\x{1E01}foo",
984           "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH RING BELOW";
985
986        # DESERET SMALL/CAPITAL LETTER LONG I
987        ok "foba  \x{10428}foo" =~ qr/(foo|\x{10400}foo|bar)/i &&
988           $1 eq "\x{10428}foo",
989           "TRIEF + DESERET SMALL/CAPITAL LETTER LONG I";
990
991        # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X'
992        ok "foba  \x{1E01}xfoo" =~ qr/(foo|\x{1E00}Xfoo|bar)/i &&
993           $1 eq "\x{1E01}xfoo",
994           "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X'";
995
996        use charnames ':full';
997
998        my $s = "\N{LATIN SMALL LETTER SHARP S}";
999        ok "foba  ba$s" =~ qr/(foo|Ba$s|bar)/i &&  $1 eq "ba$s",
1000           "TRIEF + LATIN SMALL LETTER SHARP S =~ ss";
1001        ok "foba  ba$s" =~ qr/(Ba$s|foo|bar)/i &&  $1 eq "ba$s",
1002           "TRIEF + LATIN SMALL LETTER SHARP S =~ ss";
1003        ok "foba  ba$s" =~ qr/(foo|bar|Ba$s)/i &&  $1 eq "ba$s",
1004           "TRIEF + LATIN SMALL LETTER SHARP S =~ ss";
1005
1006        ok "foba  ba$s" =~ qr/(foo|Bass|bar)/i &&  $1 eq "ba$s",
1007           "TRIEF + LATIN SMALL LETTER SHARP S =~ ss";
1008
1009        ok "foba  ba$s" =~ qr/(foo|BaSS|bar)/i &&  $1 eq "ba$s",
1010           "TRIEF + LATIN SMALL LETTER SHARP S =~ SS";
1011
1012        ok "foba  ba${s}pxySS$s$s" =~ qr/(b(?:a${s}t|a${s}f|a${s}p)[xy]+$s*)/i
1013            &&  $1 eq "ba${s}pxySS$s$s",
1014           "COMMON PREFIX TRIEF + LATIN SMALL LETTER SHARP S";
1015    }
1016
1017
1018
1019
1020    {
1021    BEGIN {
1022        unshift @INC, 'lib';
1023    }
1024        use Cname;
1025
1026        ok 'fooB'  =~ /\N{foo}[\N{B}\N{b}]/, "Passthrough charname";
1027        #
1028        # Why doesn't must_warn work here?
1029        #
1030        my $w;
1031        local $SIG {__WARN__} = sub {$w .= "@_"};
1032        eval 'q(xxWxx) =~ /[\N{WARN}]/';
1033        ok $w && $w =~ /Using just the first character returned by \\N{} in character class/,
1034                 "single character in [\\N{}] warning";
1035
1036        undef $w;
1037        eval q [ok "\0" !~ /[\N{EMPTY-STR}XY]/,
1038                   "Zerolength charname in charclass doesn't match \\0"];
1039        ok $w && $w =~ /Ignoring zero length/,
1040                 'Ignoring zero length \N{} in character class warning';
1041
1042        ok 'AB'  =~ /(\N{EVIL})/ && $1 eq 'A', 'Charname caching $1';
1043        ok 'ABC' =~ /(\N{EVIL})/,              'Charname caching $1';
1044        ok 'xy'  =~ /x\N{EMPTY-STR}y/,
1045                    'Empty string charname produces NOTHING node';
1046        ok ''    =~ /\N{EMPTY-STR}/,
1047                    'Empty string charname produces NOTHING node';
1048        ok "\N{LONG-STR}" =~ /^\N{LONG-STR}$/, 'Verify that long string works';
1049        ok "\N{LONG-STR}" =~ /^\N{LONG-STR}$/i, 'Verify under folding that long string works';
1050
1051        # If remove the limitation in regcomp code these should work
1052        # differently
1053        undef $w;
1054        eval q [ok "\N{LONG-STR}" =~ /^\N{TOO-LONG-STR}$/, 'Verify that too long a string fails gracefully'];
1055        ok $w && $w =~ /Using just the first characters returned/, 'Verify that got too-long string warning in \N{} that exceeds the limit';
1056        undef $w;
1057        eval q [ok "\N{LONG-STR}" =~ /^\N{TOO-LONG-STR}$/i, 'Verify under folding that too long a string fails gracefully'];
1058        ok $w && $w =~ /Using just the first characters returned/, 'Verify under folding that got too-long string warning in \N{} that exceeds the limit';
1059        undef $w;
1060        eval q [ok "\N{TOO-LONG-STR}" !~ /^\N{TOO-LONG-STR}$/, 'Verify that too long a string doesnt work'];
1061        ok $w && $w =~ /Using just the first characters returned/, 'Verify that got too-long string warning in \N{} that exceeds the limit';
1062        undef $w;
1063        eval q [ok "\N{TOO-LONG-STR}" !~ /^\N{TOO-LONG-STR}$/i, 'Verify under folding that too long a string doesnt work'];
1064        ok $w && $w =~ /Using just the first characters returned/i, 'Verify under folding that got too-long string warning in \N{} that exceeds the limit';
1065        undef $w;
1066        eval 'q(syntax error) =~ /\N{MALFORMED}/';
1067        ok $@ && $@ =~ /Malformed/, 'Verify that malformed utf8 gives an error';
1068        undef $w;
1069        eval 'q() =~ /\N{4F}/';
1070        ok $w && $w =~ /Deprecated/, 'Verify that leading digit in name gives warning';
1071        undef $w;
1072        eval 'q() =~ /\N{COM,MA}/';
1073        ok $w && $w =~ /Deprecated/, 'Verify that comma in name gives warning';
1074        undef $w;
1075        my $name = "A\x{D7}O";
1076        eval "q(W) =~ /\\N{$name}/";
1077        ok $w && $w =~ /Deprecated/, 'Verify that latin1 symbol in name gives warning';
1078        undef $w;
1079        $name = "A\x{D1}O";
1080        eval "q(W) =~ /\\N{$name}/";
1081        ok ! $w, 'Verify that latin1 letter in name doesnt give warning';
1082
1083    }
1084
1085
1086    {
1087        use charnames ':full';
1088
1089        ok 'aabc' !~ /a\N{PLUS SIGN}b/, '/a\N{PLUS SIGN}b/ against aabc';
1090        ok 'a+bc' =~ /a\N{PLUS SIGN}b/, '/a\N{PLUS SIGN}b/ against a+bc';
1091
1092        ok ' A B' =~ /\N{SPACE}\N{U+0041}\N{SPACE}\N{U+0042}/,
1093            'Intermixed named and unicode escapes';
1094        ok "\N{SPACE}\N{U+0041}\N{SPACE}\N{U+0042}" =~
1095           /\N{SPACE}\N{U+0041}\N{SPACE}\N{U+0042}/,
1096            'Intermixed named and unicode escapes';
1097        ok "\N{SPACE}\N{U+0041}\N{SPACE}\N{U+0042}" =~
1098           /[\N{SPACE}\N{U+0041}][\N{SPACE}\N{U+0042}]/,
1099            'Intermixed named and unicode escapes';
1100        ok "\0" =~ /^\N{NULL}$/, 'Verify that \N{NULL} works; is not confused with an error';
1101    }
1102
1103
1104    {
1105        our $brackets;
1106        $brackets = qr{
1107            {  (?> [^{}]+ | (??{ $brackets }) )* }
1108        }x;
1109
1110        ok "{b{c}d" !~ m/^((??{ $brackets }))/, "Bracket mismatch";
1111
1112        SKIP: {
1113            our @stack = ();
1114            my @expect = qw(
1115                stuff1
1116                stuff2
1117                <stuff1>and<stuff2>
1118                right
1119                <right>
1120                <<right>>
1121                <<<right>>>
1122                <<stuff1>and<stuff2>><<<<right>>>>
1123            );
1124
1125            local $_ = '<<<stuff1>and<stuff2>><<<<right>>>>>';
1126            ok /^(<((?:(?>[^<>]+)|(?1))*)>(?{push @stack, $2 }))$/,
1127                "Recursion matches";
1128            iseq @stack, @expect, "Right amount of matches"
1129                 or skip "Won't test individual results as count isn't equal",
1130                          0 + @expect;
1131            my $idx = 0;
1132            foreach my $expect (@expect) {
1133                iseq $stack [$idx], $expect,
1134                    "Expecting '$expect' at stack pos #$idx";
1135                $idx ++;
1136            }
1137        }
1138    }
1139
1140
1141    {
1142        my $s = '123453456';
1143        $s =~ s/(?<digits>\d+)\k<digits>/$+{digits}/;
1144        ok $s eq '123456', 'Named capture (angle brackets) s///';
1145        $s = '123453456';
1146        $s =~ s/(?'digits'\d+)\k'digits'/$+{digits}/;
1147        ok $s eq '123456', 'Named capture (single quotes) s///';
1148    }
1149
1150
1151    {
1152        my @ary = (
1153            pack('U', 0x00F1),            # n-tilde
1154            '_'.pack('U', 0x00F1),        # _ + n-tilde
1155            'c'.pack('U', 0x0327),        # c + cedilla
1156            pack('U*', 0x00F1, 0x0327),   # n-tilde + cedilla
1157            pack('U', 0x0391),            # ALPHA
1158            pack('U', 0x0391).'2',        # ALPHA + 2
1159            pack('U', 0x0391).'_',        # ALPHA + _
1160        );
1161
1162        for my $uni (@ary) {
1163            my ($r1, $c1, $r2, $c2) = eval qq {
1164                use utf8;
1165                scalar ("..foo foo.." =~ /(?'${uni}'foo) \\k'${uni}'/),
1166                        \$+{${uni}},
1167                scalar ("..bar bar.." =~ /(?<${uni}>bar) \\k<${uni}>/),
1168                        \$+{${uni}};
1169            };
1170            ok $r1,                         "Named capture UTF (?'')";
1171            ok defined $c1 && $c1 eq 'foo', "Named capture UTF \%+";
1172            ok $r2,                         "Named capture UTF (?<>)";
1173            ok defined $c2 && $c2 eq 'bar', "Named capture UTF \%+";
1174        }
1175    }
1176
1177    {
1178        my $s = 'foo bar baz';
1179        my @res;
1180        if ('1234' =~ /(?<A>1)(?<B>2)(?<A>3)(?<B>4)/) {
1181            foreach my $name (sort keys(%-)) {
1182                my $ary = $- {$name};
1183                foreach my $idx (0 .. $#$ary) {
1184                    push @res, "$name:$idx:$ary->[$idx]";
1185                }
1186            }
1187        }
1188        my @expect = qw (A:0:1 A:1:3 B:0:2 B:1:4);
1189        iseq "@res", "@expect", "Check %-";
1190        eval'
1191            no warnings "uninitialized";
1192            print for $- {this_key_doesnt_exist};
1193        ';
1194        ok !$@,'lvalue $- {...} should not throw an exception';
1195    }
1196
1197    {
1198        # \, breaks {3,4}
1199        ok "xaaay"    !~ /xa{3\,4}y/, '\, in a pattern';
1200        ok "xa{3,4}y" =~ /xa{3\,4}y/, '\, in a pattern';
1201
1202        # \c\ followed by _
1203        ok "x\c_y"    !~ /x\c\_y/,    '\_ in a pattern';
1204        ok "x\c\_y"   =~ /x\c\_y/,    '\_ in a pattern';
1205
1206        # \c\ followed by other characters
1207        for my $c ("z", "\0", "!", chr(254), chr(256)) {
1208            my $targ = "a\034$c";
1209            my $reg  = "a\\c\\$c";
1210            ok eval ("qq/$targ/ =~ /$reg/"), "\\c\\ in pattern";
1211        }
1212    }
1213
1214    {   # Test the (*PRUNE) pattern
1215        our $count = 0;
1216        'aaab' =~ /a+b?(?{$count++})(*FAIL)/;
1217        iseq $count, 9, "Expect 9 for no (*PRUNE)";
1218        $count = 0;
1219        'aaab' =~ /a+b?(*PRUNE)(?{$count++})(*FAIL)/;
1220        iseq $count, 3, "Expect 3 with (*PRUNE)";
1221        local $_ = 'aaab';
1222        $count = 0;
1223        1 while /.(*PRUNE)(?{$count++})(*FAIL)/g;
1224        iseq $count, 4, "/.(*PRUNE)/";
1225        $count = 0;
1226        'aaab' =~ /a+b?(??{'(*PRUNE)'})(?{$count++})(*FAIL)/;
1227        iseq $count, 3, "Expect 3 with (*PRUNE)";
1228        local $_ = 'aaab';
1229        $count = 0;
1230        1 while /.(??{'(*PRUNE)'})(?{$count++})(*FAIL)/g;
1231        iseq $count, 4, "/.(*PRUNE)/";
1232    }
1233
1234
1235    {   # Test the (*SKIP) pattern
1236        our $count = 0;
1237        'aaab' =~ /a+b?(*SKIP)(?{$count++})(*FAIL)/;
1238        iseq $count, 1, "Expect 1 with (*SKIP)";
1239        local $_ = 'aaab';
1240        $count = 0;
1241        1 while /.(*SKIP)(?{$count++})(*FAIL)/g;
1242        iseq $count, 4, "/.(*SKIP)/";
1243        $_ = 'aaabaaab';
1244        $count = 0;
1245        our @res = ();
1246        1 while /(a+b?)(*SKIP)(?{$count++; push @res,$1})(*FAIL)/g;
1247        iseq $count, 2, "Expect 2 with (*SKIP)";
1248        iseq "@res", "aaab aaab", "Adjacent (*SKIP) works as expected";
1249    }
1250
1251
1252    {   # Test the (*SKIP) pattern
1253        our $count = 0;
1254        'aaab' =~ /a+b?(*MARK:foo)(*SKIP)(?{$count++})(*FAIL)/;
1255        iseq $count, 1, "Expect 1 with (*SKIP)";
1256        local $_ = 'aaab';
1257        $count = 0;
1258        1 while /.(*MARK:foo)(*SKIP)(?{$count++})(*FAIL)/g;
1259        iseq $count, 4, "/.(*SKIP)/";
1260        $_ = 'aaabaaab';
1261        $count = 0;
1262        our @res = ();
1263        1 while /(a+b?)(*MARK:foo)(*SKIP)(?{$count++; push @res,$1})(*FAIL)/g;
1264        iseq $count, 2, "Expect 2 with (*SKIP)";
1265        iseq "@res", "aaab aaab", "Adjacent (*SKIP) works as expected";
1266    }
1267
1268
1269    {   # Test the (*SKIP) pattern
1270        our $count = 0;
1271        'aaab' =~ /a*(*MARK:a)b?(*MARK:b)(*SKIP:a)(?{$count++})(*FAIL)/;
1272        iseq $count, 3, "Expect 3 with *MARK:a)b?(*MARK:b)(*SKIP:a)";
1273        local $_ = 'aaabaaab';
1274        $count = 0;
1275        our @res = ();
1276        1 while
1277        /(a*(*MARK:a)b?)(*MARK:x)(*SKIP:a)(?{$count++; push @res,$1})(*FAIL)/g;
1278        iseq $count, 5, "Expect 5 with (*MARK:a)b?)(*MARK:x)(*SKIP:a)";
1279        iseq "@res", "aaab b aaab b ",
1280             "Adjacent (*MARK:a)b?)(*MARK:x)(*SKIP:a) works as expected";
1281    }
1282
1283
1284    {   # Test the (*COMMIT) pattern
1285        our $count = 0;
1286        'aaabaaab' =~ /a+b?(*COMMIT)(?{$count++})(*FAIL)/;
1287        iseq $count, 1, "Expect 1 with (*COMMIT)";
1288        local $_ = 'aaab';
1289        $count = 0;
1290        1 while /.(*COMMIT)(?{$count++})(*FAIL)/g;
1291        iseq $count, 1, "/.(*COMMIT)/";
1292        $_ = 'aaabaaab';
1293        $count = 0;
1294        our @res = ();
1295        1 while /(a+b?)(*COMMIT)(?{$count++; push @res,$1})(*FAIL)/g;
1296        iseq $count, 1, "Expect 1 with (*COMMIT)";
1297        iseq "@res", "aaab", "Adjacent (*COMMIT) works as expected";
1298    }
1299
1300
1301    {
1302        # Test named commits and the $REGERROR var
1303        our $REGERROR;
1304        for my $name ('', ':foo') {
1305            for my $pat ("(*PRUNE$name)",
1306                         ($name ? "(*MARK$name)" : "") . "(*SKIP$name)",
1307                         "(*COMMIT$name)") {
1308                for my $suffix ('(*FAIL)', '') {
1309                    'aaaab' =~ /a+b$pat$suffix/;
1310                    iseq $REGERROR,
1311                         ($suffix ? ($name ? 'foo' : "1") : ""),
1312                        "Test $pat and \$REGERROR $suffix";
1313                }
1314            }
1315        }
1316    }
1317
1318
1319    {
1320        # Test named commits and the $REGERROR var
1321        package Fnorble;
1322        our $REGERROR;
1323        for my $name ('', ':foo') {
1324            for my $pat ("(*PRUNE$name)",
1325                         ($name ? "(*MARK$name)" : "") . "(*SKIP$name)",
1326                         "(*COMMIT$name)") {
1327                for my $suffix ('(*FAIL)','') {
1328                    'aaaab' =~ /a+b$pat$suffix/;
1329                  ::iseq $REGERROR,
1330                         ($suffix ? ($name ? 'foo' : "1") : ""),
1331                        "Test $pat and \$REGERROR $suffix";
1332                }
1333            }
1334        }
1335    }
1336
1337
1338    {
1339        # Test named commits and the $REGERROR var
1340        local $Message = '$REGERROR';
1341        our $REGERROR;
1342        for my $word (qw (bar baz bop)) {
1343            $REGERROR = "";
1344            "aaaaa$word" =~
1345              /a+(?:bar(*COMMIT:bar)|baz(*COMMIT:baz)|bop(*COMMIT:bop))(*FAIL)/;
1346            iseq $REGERROR, $word;
1347        }
1348    }
1349
1350    {
1351        #Mindnumbingly simple test of (*THEN)
1352        for ("ABC","BAX") {
1353            ok /A (*THEN) X | B (*THEN) C/x, "Simple (*THEN) test";
1354        }
1355    }
1356
1357
1358    {
1359        local $Message = "Relative Recursion";
1360        my $parens = qr/(\((?:[^()]++|(?-1))*+\))/;
1361        local $_ = 'foo((2*3)+4-3) + bar(2*(3+4)-1*(2-3))';
1362        my ($all, $one, $two) = ('', '', '');
1363        ok m/foo $parens \s* \+ \s* bar $parens/x;
1364        iseq $1, '((2*3)+4-3)';
1365        iseq $2, '(2*(3+4)-1*(2-3))';
1366        iseq $&, 'foo((2*3)+4-3) + bar(2*(3+4)-1*(2-3))';
1367        iseq $&, $_;
1368    }
1369
1370    {
1371        my $spaces="      ";
1372        local $_ = join 'bar', $spaces, $spaces;
1373        our $count = 0;
1374        s/(?>\s+bar)(?{$count++})//g;
1375        iseq $_, $spaces, "SUSPEND final string";
1376        iseq $count, 1, "Optimiser should have prevented more than one match";
1377    }
1378
1379
1380    {
1381        # From Message-ID: <877ixs6oa6.fsf@k75.linux.bogus>
1382        my $dow_name = "nada";
1383        my $parser = "(\$dow_name) = \$time_string =~ /(D\x{e9}\\ " .
1384                     "C\x{e9}adaoin|D\x{e9}\\ Sathairn|\\w+|\x{100})/";
1385        my $time_string = "D\x{e9} C\x{e9}adaoin";
1386        eval $parser;
1387        ok !$@, "Test Eval worked";
1388        iseq $dow_name, $time_string, "UTF-8 trie common prefix extraction";
1389    }
1390
1391
1392    {
1393        my $v;
1394        ($v = 'bar') =~ /(\w+)/g;
1395        $v = 'foo';
1396        iseq "$1", 'bar', '$1 is safe after /g - may fail due ' .
1397                          'to specialized config in pp_hot.c'
1398    }
1399
1400
1401    {
1402        local $Message = "http://nntp.perl.org/group/perl.perl5.porters/118663";
1403        my $qr_barR1 = qr/(bar)\g-1/;
1404        ok "foobarbarxyz" =~ $qr_barR1;
1405        ok "foobarbarxyz" =~ qr/foo${qr_barR1}xyz/;
1406        ok "foobarbarxyz" =~ qr/(foo)${qr_barR1}xyz/;
1407        ok "foobarbarxyz" =~ qr/(foo)(bar)\g{-1}xyz/;
1408        ok "foobarbarxyz" =~ qr/(foo${qr_barR1})xyz/;
1409        ok "foobarbarxyz" =~ qr/(foo(bar)\g{-1})xyz/;
1410    }
1411
1412    {
1413        local $Message = '$REGMARK';
1414        our @r = ();
1415        our ($REGMARK, $REGERROR);
1416        ok 'foofoo' =~ /foo (*MARK:foo) (?{push @r,$REGMARK}) /x;
1417        iseq "@r","foo";
1418        iseq $REGMARK, "foo";
1419        ok 'foofoo' !~ /foo (*MARK:foo) (*FAIL) /x;
1420        ok !$REGMARK;
1421        iseq $REGERROR, 'foo';
1422    }
1423
1424
1425    {
1426        local $Message = '\K test';
1427        my $x;
1428        $x = "abc.def.ghi.jkl";
1429        $x =~ s/.*\K\..*//;
1430        iseq $x, "abc.def.ghi";
1431
1432        $x = "one two three four";
1433        $x =~ s/o+ \Kthree//g;
1434        iseq $x, "one two  four";
1435
1436        $x = "abcde";
1437        $x =~ s/(.)\K/$1/g;
1438        iseq $x, "aabbccddee";
1439    }
1440
1441
1442    {
1443        sub kt {
1444            return '4' if $_[0] eq '09028623';
1445        }
1446        # Nested EVAL using PL_curpm (via $1 or friends)
1447        my $re;
1448        our $grabit = qr/ ([0-6][0-9]{7}) (??{ kt $1 }) [890] /x;
1449        $re = qr/^ ( (??{ $grabit }) ) $ /x;
1450        my @res = '0902862349' =~ $re;
1451        iseq join ("-", @res), "0902862349",
1452            'PL_curpm is set properly on nested eval';
1453
1454        our $qr = qr/ (o) (??{ $1 }) /x;
1455        ok 'boob'=~/( b (??{ $qr }) b )/x && 1, "PL_curpm, nested eval";
1456    }
1457
1458
1459    {
1460        use charnames ":full";
1461        ok "\N{ROMAN NUMERAL ONE}" =~ /\p{Alphabetic}/, "I =~ Alphabetic";
1462        ok "\N{ROMAN NUMERAL ONE}" =~ /\p{Uppercase}/,  "I =~ Uppercase";
1463        ok "\N{ROMAN NUMERAL ONE}" !~ /\p{Lowercase}/,  "I !~ Lowercase";
1464        ok "\N{ROMAN NUMERAL ONE}" =~ /\p{IDStart}/,    "I =~ ID_Start";
1465        ok "\N{ROMAN NUMERAL ONE}" =~ /\p{IDContinue}/, "I =~ ID_Continue";
1466        ok "\N{SMALL ROMAN NUMERAL ONE}" =~ /\p{Alphabetic}/, "i =~ Alphabetic";
1467        ok "\N{SMALL ROMAN NUMERAL ONE}" !~ /\p{Uppercase}/,  "i !~ Uppercase";
1468        ok "\N{SMALL ROMAN NUMERAL ONE}" =~ /\p{Lowercase}/,  "i =~ Lowercase";
1469        ok "\N{SMALL ROMAN NUMERAL ONE}" =~ /\p{IDStart}/,    "i =~ ID_Start";
1470        ok "\N{SMALL ROMAN NUMERAL ONE}" =~ /\p{IDContinue}/, "i =~ ID_Continue"
1471    }
1472
1473
1474    {
1475        # requirement of Unicode Technical Standard #18, 1.7 Code Points
1476        # cf. http://www.unicode.org/reports/tr18/#Supplementary_Characters
1477        for my $u (0x7FF, 0x800, 0xFFFF, 0x10000) {
1478            no warnings 'utf8'; # oops
1479            my $c = chr $u;
1480            my $x = sprintf '%04X', $u;
1481            ok "A${c}B" =~ /A[\0-\x{10000}]B/, "Unicode range - $x";
1482        }
1483    }
1484
1485
1486    {
1487        my $res="";
1488
1489        if ('1' =~ /(?|(?<digit>1)|(?<digit>2))/) {
1490            $res = "@{$- {digit}}";
1491        }
1492        iseq $res, "1",
1493            "Check that (?|...) doesnt cause dupe entries in the names array";
1494
1495        $res = "";
1496        if ('11' =~ /(?|(?<digit>1)|(?<digit>2))(?&digit)/) {
1497            $res = "@{$- {digit}}";
1498        }
1499        iseq $res, "1", "Check that (?&..) to a buffer inside " .
1500                        "a (?|...) goes to the leftmost";
1501    }
1502
1503
1504    {
1505        use warnings;
1506        local $Message = "ASCII pattern that really is UTF-8";
1507        my @w;
1508        local $SIG {__WARN__} = sub {push @w, "@_"};
1509        my $c = qq (\x{DF});
1510        ok $c =~ /${c}|\x{100}/;
1511        ok @w == 0;
1512    }
1513
1514
1515    {
1516        local $Message = "Corruption of match results of qr// across scopes";
1517        my $qr = qr/(fo+)(ba+r)/;
1518        'foobar' =~ /$qr/;
1519        iseq "$1$2", "foobar";
1520        {
1521            'foooooobaaaaar' =~ /$qr/;
1522            iseq "$1$2", 'foooooobaaaaar';
1523        }
1524        iseq "$1$2", "foobar";
1525    }
1526
1527
1528    {
1529        local $Message = "HORIZWS";
1530        local $_ = "\t \r\n \n \t".chr(11)."\n";
1531        s/\H/H/g;
1532        s/\h/h/g;
1533        iseq $_, "hhHHhHhhHH";
1534        $_ = "\t \r\n \n \t" . chr (11) . "\n";
1535        utf8::upgrade ($_);
1536        s/\H/H/g;
1537        s/\h/h/g;
1538        iseq $_, "hhHHhHhhHH";
1539    }
1540
1541
1542    {
1543        local $Message = "Various whitespace special patterns";
1544        my @h = map {chr $_}   0x09,   0x20,   0xa0, 0x1680, 0x180e, 0x2000,
1545                             0x2001, 0x2002, 0x2003, 0x2004, 0x2005, 0x2006,
1546                             0x2007, 0x2008, 0x2009, 0x200a, 0x202f, 0x205f,
1547                             0x3000;
1548        my @v = map {chr $_}   0x0a,   0x0b,   0x0c,   0x0d,   0x85, 0x2028,
1549                             0x2029;
1550        my @lb = ("\x0D\x0A", map {chr $_} 0x0A .. 0x0D, 0x85, 0x2028, 0x2029);
1551        foreach my $t ([\@h,  qr/\h/, qr/\h+/],
1552                       [\@v,  qr/\v/, qr/\v+/],
1553                       [\@lb, qr/\R/, qr/\R+/],) {
1554            my $ary = shift @$t;
1555            foreach my $pat (@$t) {
1556                foreach my $str (@$ary) {
1557                    ok $str =~ /($pat)/, $pat;
1558                    iseq $1, $str, $pat;
1559                    utf8::upgrade ($str);
1560                    ok $str =~ /($pat)/, "Upgraded string - $pat";
1561                    iseq $1, $str, "Upgraded string - $pat";
1562                }
1563            }
1564        }
1565    }
1566
1567
1568    {
1569        local $Message = "Check that \\xDF match properly in its various forms";
1570        # Test that \xDF matches properly. this is pretty hacky stuff,
1571        # but its actually needed. The malarky with '-' is to prevent
1572        # compilation caching from playing any role in the test.
1573        my @df = (chr (0xDF), '-', chr (0xDF));
1574        utf8::upgrade ($df [2]);
1575        my @strs = ('ss', 'sS', 'Ss', 'SS', chr (0xDF));
1576        my @ss = map {("$_", "$_")} @strs;
1577        utf8::upgrade ($ss [$_ * 2 + 1]) for 0 .. $#strs;
1578
1579        for my $ssi (0 .. $#ss) {
1580            for my $dfi (0 .. $#df) {
1581                my $pat = $df [$dfi];
1582                my $str = $ss [$ssi];
1583                my $utf_df = ($dfi > 1) ? 'utf8' : '';
1584                my $utf_ss = ($ssi % 2) ? 'utf8' : '';
1585                (my $sstr = $str) =~ s/\xDF/\\xDF/;
1586
1587                if ($utf_df || $utf_ss || length ($ss [$ssi]) == 1) {
1588                    my $ret = $str =~ /$pat/i;
1589                    next if $pat eq '-';
1590                    ok $ret, "\"$sstr\" =~ /\\xDF/i " .
1591                             "(str is @{[$utf_ss||'latin']}, pat is " .
1592                             "@{[$utf_df||'latin']})";
1593                }
1594                else {
1595                    my $ret = $str !~ /$pat/i;
1596                    next if $pat eq '-';
1597                    ok $ret, "\"$sstr\" !~ /\\xDF/i " .
1598                             "(str is @{[$utf_ss||'latin']}, pat is " .
1599                             "@{[$utf_df||'latin']})";
1600                }
1601            }
1602        }
1603    }
1604
1605
1606    {
1607        local $Message = "BBC(Bleadperl Breaks CPAN) Today: String::Multibyte";
1608        my $re  = qr/(?:[\x00-\xFF]{4})/;
1609        my $hyp = "\0\0\0-";
1610        my $esc = "\0\0\0\\";
1611
1612        my $str = "$esc$hyp$hyp$esc$esc";
1613        my @a = ($str =~ /\G(?:\Q$esc$esc\E|\Q$esc$hyp\E|$re)/g);
1614
1615        iseq @a,3;
1616        local $" = "=";
1617        iseq "@a","$esc$hyp=$hyp=$esc$esc";
1618    }
1619
1620
1621    {
1622        # Test for keys in %+ and %-
1623        local $Message = 'Test keys in %+ and %-';
1624        no warnings 'uninitialized';
1625        my $_ = "abcdef";
1626        /(?<foo>a)|(?<foo>b)/;
1627        iseq ((join ",", sort keys %+), "foo");
1628        iseq ((join ",", sort keys %-), "foo");
1629        iseq ((join ",", sort values %+), "a");
1630        iseq ((join ",", sort map "@$_", values %-), "a ");
1631        /(?<bar>a)(?<bar>b)(?<quux>.)/;
1632        iseq ((join ",", sort keys %+), "bar,quux");
1633        iseq ((join ",", sort keys %-), "bar,quux");
1634        iseq ((join ",", sort values %+), "a,c"); # leftmost
1635        iseq ((join ",", sort map "@$_", values %-), "a b,c");
1636        /(?<un>a)(?<deux>c)?/; # second buffer won't capture
1637        iseq ((join ",", sort keys %+), "un");
1638        iseq ((join ",", sort keys %-), "deux,un");
1639        iseq ((join ",", sort values %+), "a");
1640        iseq ((join ",", sort map "@$_", values %-), ",a");
1641    }
1642
1643
1644    {
1645        # length() on captures, the numbered ones end up in Perl_magic_len
1646        my $_ = "aoeu \xe6var ook";
1647        /^ \w+ \s (?<eek>\S+)/x;
1648
1649        iseq length ($`),      0, q[length $`];
1650        iseq length ($'),      4, q[length $'];
1651        iseq length ($&),      9, q[length $&];
1652        iseq length ($1),      4, q[length $1];
1653        iseq length ($+{eek}), 4, q[length $+{eek} == length $1];
1654    }
1655
1656
1657    {
1658        my $ok = -1;
1659
1660        $ok = exists ($-{x}) ? 1 : 0 if 'bar' =~ /(?<x>foo)|bar/;
1661        iseq $ok, 1, '$-{x} exists after "bar"=~/(?<x>foo)|bar/';
1662        iseq scalar (%+), 0, 'scalar %+ == 0 after "bar"=~/(?<x>foo)|bar/';
1663        iseq scalar (%-), 1, 'scalar %- == 1 after "bar"=~/(?<x>foo)|bar/';
1664
1665        $ok = -1;
1666        $ok = exists ($+{x}) ? 1 : 0 if 'bar' =~ /(?<x>foo)|bar/;
1667        iseq $ok, 0, '$+{x} not exists after "bar"=~/(?<x>foo)|bar/';
1668        iseq scalar (%+), 0, 'scalar %+ == 0 after "bar"=~/(?<x>foo)|bar/';
1669        iseq scalar (%-), 1, 'scalar %- == 1 after "bar"=~/(?<x>foo)|bar/';
1670
1671        $ok = -1;
1672        $ok = exists ($-{x}) ? 1 : 0 if 'foo' =~ /(?<x>foo)|bar/;
1673        iseq $ok, 1, '$-{x} exists after "foo"=~/(?<x>foo)|bar/';
1674        iseq scalar (%+), 1, 'scalar %+ == 1 after "foo"=~/(?<x>foo)|bar/';
1675        iseq scalar (%-), 1, 'scalar %- == 1 after "foo"=~/(?<x>foo)|bar/';
1676
1677        $ok = -1;
1678        $ok = exists ($+{x}) ? 1 : 0 if 'foo'=~/(?<x>foo)|bar/;
1679        iseq $ok, 1, '$+{x} exists after "foo"=~/(?<x>foo)|bar/';
1680    }
1681
1682
1683    {
1684        local $_;
1685        ($_ = 'abc') =~ /(abc)/g;
1686        $_ = '123';
1687        iseq "$1", 'abc', "/g leads to unsafe match vars: $1";
1688    }
1689
1690
1691    {
1692        local $Message = 'Message-ID: <20070818091501.7eff4831@r2d2>';
1693        my $str = "";
1694        for (0 .. 5) {
1695            my @x;
1696            $str .= "@x"; # this should ALWAYS be the empty string
1697            'a' =~ /(a|)/;
1698            push @x, 1;
1699        }
1700        iseq length ($str), 0, "Trie scope error, string should be empty";
1701        $str = "";
1702        my @foo = ('a') x 5;
1703        for (@foo) {
1704            my @bar;
1705            $str .= "@bar";
1706            s/a|/push @bar, 1/e;
1707        }
1708        iseq length ($str), 0, "Trie scope error, string should be empty";
1709    }
1710
1711
1712    {
1713# more TRIE/AHOCORASICK problems with mixed utf8 / latin-1 and case folding
1714    for my $chr (160 .. 255) {
1715        my $chr_byte = chr($chr);
1716        my $chr_utf8 = chr($chr); utf8::upgrade($chr_utf8);
1717        my $rx = qr{$chr_byte|X}i;
1718        ok($chr_utf8 =~ $rx, "utf8/latin, codepoint $chr");
1719    }
1720    }
1721
1722    {
1723        our $a = 3; "" =~ /(??{ $a })/;
1724        our $b = $a;
1725        iseq $b, $a, "Copy of scalar used for postponed subexpression";
1726    }
1727
1728
1729    {
1730        our @ctl_n = ();
1731        our @plus = ();
1732        our $nested_tags;
1733        $nested_tags = qr{
1734            <
1735               (\w+)
1736               (?{
1737                       push @ctl_n,$^N;
1738                       push @plus,$+;
1739               })
1740            >
1741            (??{$nested_tags})*
1742            </\s* \w+ \s*>
1743        }x;
1744
1745        my $match = '<bla><blubb></blubb></bla>' =~ m/^$nested_tags$/;
1746        ok $match, 'nested construct matches';
1747        iseq "@ctl_n", "bla blubb", '$^N inside of (?{}) works as expected';
1748        iseq "@plus",  "bla blubb", '$+  inside of (?{}) works as expected';
1749    }
1750
1751
1752    SKIP: {
1753        # XXX: This set of tests is essentially broken, POSIX character classes
1754        # should not have differing definitions under Unicode.
1755        # There are property names for that.
1756        skip "Tests assume ASCII", 4 unless $IS_ASCII;
1757
1758        my @notIsPunct = grep {/[[:punct:]]/ and not /\p{IsPunct}/}
1759                                map {chr} 0x20 .. 0x7f;
1760        iseq join ('', @notIsPunct), '$+<=>^`|~',
1761            '[:punct:] disagress with IsPunct on Symbols';
1762
1763        my @isPrint = grep {not /[[:print:]]/ and /\p{IsPrint}/}
1764                            map {chr} 0 .. 0x1f, 0x7f .. 0x9f;
1765        iseq join ('', @isPrint), "",
1766            'IsPrint agrees with [:print:] on control characters';
1767
1768        my @isPunct = grep {/[[:punct:]]/ != /\p{IsPunct}/}
1769                            map {chr} 0x80 .. 0xff;
1770        iseq join ('', @isPunct), "\xa1\xab\xb7\xbb\xbf",    # ¡ « · » ¿
1771            'IsPunct disagrees with [:punct:] outside ASCII';
1772
1773        my @isPunctLatin1 = eval q {
1774            use encoding 'latin1';
1775            grep {/[[:punct:]]/ != /\p{IsPunct}/} map {chr} 0x80 .. 0xff;
1776        };
1777        skip "Eval failed ($@)", 1 if $@;
1778        skip "PERL_LEGACY_UNICODE_CHARCLASS_MAPPINGS set to 0", 1
1779              if !$ENV{PERL_TEST_LEGACY_POSIX_CC};
1780        iseq join ('', @isPunctLatin1), '',
1781            'IsPunct agrees with [:punct:] with explicit Latin1';
1782    }
1783
1784    #
1785    # Keep the following tests last -- they may crash perl
1786    #
1787    print "# Tests that follow may crash perl\n";
1788    {
1789        eval '/\k/';
1790        ok $@ =~ /\QSequence \k... not terminated in regex;\E/,
1791           'Lone \k not allowed';
1792    }
1793
1794    {
1795        local $Message = "Substitution with lookahead (possible segv)";
1796        $_ = "ns1ns1ns1";
1797        s/ns(?=\d)/ns_/g;
1798        iseq $_, "ns_1ns_1ns_1";
1799        $_ = "ns1";
1800        s/ns(?=\d)/ns_/;
1801        iseq $_, "ns_1";
1802        $_ = "123";
1803        s/(?=\d+)|(?<=\d)/!Bang!/g;
1804        iseq $_, "!Bang!1!Bang!2!Bang!3!Bang!";
1805    }
1806
1807    { 
1808        # Earlier versions of Perl said this was fatal.
1809        local $Message = "U+0FFFF shouldn't crash the regex engine";
1810        no warnings 'utf8';
1811        my $a = eval "chr(65535)";
1812        use warnings;
1813        my $warning_message;
1814        local $SIG{__WARN__} = sub { $warning_message = $_[0] };
1815        eval $a =~ /[a-z]/;
1816        ok(1);  # If it didn't crash, it worked.
1817    }
1818} # End of sub run_tests
1819
18201;
1821