1 # Grind out a lot of combinatoric tests for folding.
3 binmode STDOUT, ":utf8";
9 require Config; import Config;
10 skip_all_if_miniperl("no dynamic loading on miniperl, no Encode nor POSIX");
11 if ($^O eq 'dec_osf') {
12 skip_all("$^O cannot handle this test");
14 require './loc_tools.pl';
17 use charnames ":full";
19 my $DEBUG = 0; # Outputs extra information for debugging this .t
23 no warnings 'locale'; # Plenty of these would otherwise get generated
27 # Special-cased characters in the .c's that we want to make sure get tested.
28 my %be_sure_to_test = (
29 chr utf8::unicode_to_native(0xDF) => 1, # LATIN_SMALL_LETTER_SHARP_S
30 "\x{1E9E}" => 1, # LATIN_CAPITAL_LETTER_SHARP_S
31 "\x{390}" => 1, # GREEK_SMALL_LETTER_IOTA_WITH_DIALYTIKA_AND_TONOS
32 "\x{3B0}" => 1, # GREEK_SMALL_LETTER_UPSILON_WITH_DIALYTIKA_AND_TONOS
33 "\x{1FD3}" => 1, # GREEK SMALL LETTER IOTA WITH DIALYTIKA AND OXIA
34 "\x{1FE3}" => 1, # GREEK SMALL LETTER UPSILON WITH DIALYTIKA AND OXIA
38 # Tests both unicode and not, so make sure not implicitly testing unicode
39 no feature 'unicode_strings';
41 # Case-insensitive matching is a large and complicated issue. Perl does not
42 # implement it fully, properly. For example, it doesn't include normalization
43 # as part of the equation. To test every conceivable combination is clearly
44 # impossible; these tests are mostly drawn from visual inspection of the code
45 # and experience, trying to exercise all areas.
47 # There are three basic ranges of characters that Perl may treat differently:
48 # 1) Invariants under utf8 which on ASCII-ish machines are ASCII, and are
49 # referred to here as ASCII. On EBCDIC machines, the non-ASCII invariants
50 # are all controls that fold to themselves.
53 # 2) Other characters that fit into a byte but are different in utf8 than not;
54 # here referred to, taking some liberties, as Latin1.
57 # 3) Characters that won't fit in a byte; here referred to as Unicode
60 # Within these basic groups are equivalence classes that testing any character
61 # in is likely to lead to the same results as any other character. This is
62 # used to cut down the number of tests needed, unless PERL_RUN_SLOW_TESTS is
64 my $skip_apparently_redundant = ! $ENV{PERL_RUN_SLOW_TESTS};
66 # Additionally parts of this test run a lot of subtests, outputting the
67 # resulting TAP can be expensive so the tests are summarised internally. The
68 # PERL_DEBUG_FULL_TEST environment variable can be set to produce the full
69 # output for debugging purposes.
74 return $ASCII if utf8::native_to_unicode($ord) < 128;
75 return $Latin1 if $ord < 256;
83 my $list_all_tests = $ENV{PERL_DEBUG_FULL_TEST} || $DEBUG;
84 $| = 1 if $list_all_tests;
86 # Significant time is saved by not outputting each test but grouping the
87 # output into subtests
88 my $okays; # Number of ok's in current subtest
89 my $this_iteration; # Number of possible tests in current subtest
90 my $count = 0; # Number of subtests = number of total tests
93 my ($test, $todo, $do_we_output_locale_name, $debug) = @_;
95 $debug = "" unless $DEBUG;
98 if ($do_we_output_locale_name) {
99 $do_we_output_locale_name = 'setlocale(LC_CTYPE, "'
100 . POSIX::setlocale(&POSIX::LC_CTYPE)
103 if (!$res || $list_all_tests) {
104 # Failed or debug; output the result
106 ok($res, "$do_we_output_locale_name$test; $debug");
108 # Just count the test as passed
114 my %has_test_by_participants; # Makes sure has tests for each range and each
115 # number of characters that fold to the same
117 my %has_test_by_byte_count; # Makes sure has tests for each combination of
118 # n bytes folds to m bytes
120 my %tests; # The set of tests.
121 # Each key is a code point that folds to something else.
122 # Each value is a list of things that the key folds to. If the 'thing' is a
123 # single code point, it is that ordinal. If it is a multi-char fold, it is an
124 # ordered list of the code points in that fold. Here's an example for 'S':
125 # '83' => [ 115, 383 ]
127 # And one for a multi-char fold: \xDF
137 # [ # LATIN SMALL LETTER LONG S
141 # 7838 # LATIN_CAPITAL_LETTER_SHARP_S
144 my %folds; # keys are code points that fold;
145 # values are each a list of code points the key folds to
146 my %inverse_folds; # keys are strings of the folded-to;
147 # values are lists of characters that fold to them
150 my ($to, @from) = @_;
152 # Called to cause the input to be tested by adding to %tests. @from is
153 # the list of characters that fold to the string $to. @from should be
154 # sorted so the lowest code point is first....
155 # The input is in string form; %tests uses code points, so have to
158 my $to_chars = length $to;
159 my @test_to; # List of tests for $to
161 if ($to_chars == 1) {
165 push @test_to, [ map { ord $_ } split "", $to ];
167 # For multi-char folds, we also test that things that can fold to each
168 # individual character in the fold also work. If we were testing
169 # comprehensively, we would try every combination of upper and lower
170 # case in the fold, but it will have to suffice to avoid running
171 # forever to make sure that each thing that folds to these is tested
172 # at least once. Because of complement matching ([^...]), we need to
173 # do both the folded, and the folded-from.
174 # We first look at each character in the multi-char fold, and save how
175 # many characters fold to it; and also the maximum number of such
177 my @folds_to_count; # 0th char in fold is index 0 ...
178 my $max_folds_to = 0;
180 for (my $i = 0; $i < $to_chars; $i++) {
181 my $to_char = substr($to, $i, 1);
182 if (exists $inverse_folds{$to_char}) {
183 $folds_to_count[$i] = scalar @{$inverse_folds{$to_char}};
184 $max_folds_to = $folds_to_count[$i] if $max_folds_to < $folds_to_count[$i];
187 $folds_to_count[$i] = 0;
191 # We will need to generate as many tests as the maximum number of
192 # folds, so that each fold will have at least one test.
193 # For example, consider character X which folds to the three character
194 # string 'xyz'. If 2 things fold to x (X and x), 4 to y (Y, Y'
195 # (Y-prime), Y'' (Y-prime-prime), and y), and 1 thing to z (itself), 4
196 # tests will be generated:
201 for (my $i = 0; $i < $max_folds_to; $i++) {
202 my @this_test_to; # Assemble a single test
204 # For each character in the multi-char fold ...
205 for (my $j = 0; $j < $to_chars; $j++) {
206 my $this_char = substr($to, $j, 1);
208 # Use its corresponding inverse fold, if available.
209 if ($i < $folds_to_count[$j]) {
210 push @this_test_to, ord $inverse_folds{$this_char}[$i];
212 else { # Or else itself.
213 push @this_test_to, ord $this_char;
217 # Add this test to the list
218 push @test_to, [ @this_test_to ];
221 # Here, have assembled all the tests for the multi-char fold. Sort so
222 # lowest code points are first for consistency and aesthetics in
223 # output. We know there are at least two characters in the fold, but
224 # I haven't bothered to worry about sorting on an optional third
225 # character if the first two are identical.
226 @test_to = sort { ($a->[0] == $b->[0])
227 ? $a->[1] <=> $b->[1]
228 : $a->[0] <=> $b->[0]
233 # This test is from n bytes to m bytes. Record that so won't try to add
234 # another test that does the same.
236 my $to_bytes = length $to;
237 foreach my $from_map (@from) {
238 $has_test_by_byte_count{length $from_map}{$to_bytes} = $to;
242 my $ord_smallest_from = ord shift @from;
243 if (exists $tests{$ord_smallest_from}) {
244 die "There are already tests for $ord_smallest_from"
247 # Add in the fold tests,
248 push @{$tests{$ord_smallest_from}}, @test_to;
250 # Then any remaining froms in the equivalence class.
251 push @{$tests{$ord_smallest_from}}, map { ord $_ } @from;
254 # Get the Unicode rules and construct inverse mappings from them
257 my $file="../lib/unicore/CaseFolding.txt";
259 # Use the Unicode data file if we are on an ASCII platform (which its data is
260 # for), and it is in the modern format (starting in Unicode 3.1.0) and it is
261 # available. This avoids being affected by potential bugs introduced by other
264 && pack("C*", split /\./, Unicode::UCD::UnicodeVersion()) ge v3.1.0
265 && open my $fh, "<", $file)
270 # Lines look like (though without the initial '#')
271 #0130; F; 0069 0307; # LATIN CAPITAL LETTER I WITH DOT ABOVE
273 # Get rid of comments, ignore blank or comment-only lines
274 my $line = $_ =~ s/ (?: \s* \# .* )? $ //rx;
275 next unless length $line;
276 my ($hex_from, $fold_type, @hex_folded) = split /[\s;]+/, $line;
278 next if $fold_type =~ / ^ [IT] $/x; # Perl doesn't do Turkish folding
279 next if $fold_type eq 'S'; # If Unicode's tables are correct, the F
280 # should be a superset of S
282 my $from = hex $hex_from;
283 my @to = map { hex $_ } @hex_folded;
284 @{$folds{$from}} = @to;
285 my $folded_str = pack ("U0U*", @to);
286 push @{$inverse_folds{$folded_str}}, chr $from;
289 else { # Here, can't use the .txt file: read the Unicode rules file and
290 # construct inverse mappings from it
292 my ($invlist_ref, $invmap_ref, undef, $default)
293 = Unicode::UCD::prop_invmap('Case_Folding');
294 for my $i (0 .. @$invlist_ref - 1 - 1) {
295 next if $invmap_ref->[$i] == $default;
297 # Make into an array if not so already, so can treat uniformly below
298 $invmap_ref->[$i] = [ $invmap_ref->[$i] ] if ! ref $invmap_ref->[$i];
300 # Each subsequent element of the range requires adjustment of +1 from
301 # the previous element
303 for my $j ($invlist_ref->[$i] .. $invlist_ref->[$i+1] -1) {
305 my @to = map { $_ + $adjust } @{$invmap_ref->[$i]};
306 push @{$folds{$j}}, @to;
307 my $folded_str = join "", map { chr } @to;
308 utf8::upgrade($folded_str);
309 #note (sprintf "%d: %04X: %s", __LINE__, $j, join " ",
310 # map { sprintf "%04X", $_ + $adjust } @{$invmap_ref->[$i]});
311 push @{$inverse_folds{$folded_str}}, chr $j;
316 # Analyze the data and generate tests to get adequate test coverage. We sort
317 # things so that smallest code points are done first.
319 foreach my $to (sort { (length $a == length $b)
321 : length $a <=> length $b
322 } keys %inverse_folds)
325 # Within each fold, sort so that the smallest code points are done first
326 @{$inverse_folds{$to}} = sort { $a cmp $b } @{$inverse_folds{$to}};
327 my @from = @{$inverse_folds{$to}};
329 # Just add it to the tests if doing complete coverage
330 if (! $skip_apparently_redundant) {
331 add_test($to, @from);
335 my $to_chars = length $to;
336 my $to_range_type = range_type(substr($to, 0, 1));
338 # If this is required to be tested, do so. We check for these first, as
339 # they will take up slots of byte-to-byte combinations that we otherwise
340 # would have to have other tests to get.
341 foreach my $from_map (@from) {
342 if (exists $be_sure_to_test{$from_map}) {
343 add_test($to, @from);
348 # If the fold contains heterogeneous range types, is suspect and should be
351 foreach my $char (split "", $to) {
352 if (range_type($char) != $to_range_type) {
353 add_test($to, @from);
359 # If the mapping crosses range types, is suspect and should be tested
360 foreach my $from_map (@from) {
361 if (range_type($from_map) != $to_range_type) {
362 add_test($to, @from);
367 # Here, all components of the mapping are in the same range type. For
368 # single character folds, we test one case in each range type that has 2
369 # particpants, 3 particpants, etc.
370 if ($to_chars == 1) {
371 if (! exists $has_test_by_participants{scalar @from}{$to_range_type}) {
372 add_test($to, @from);
373 $has_test_by_participants{scalar @from}{$to_range_type} = $to;
378 # We also test all combinations of mappings from m to n bytes. This is
379 # because the regex optimizer cares. (Don't bother worrying about that
380 # Latin1 chars will occupy a different number of bytes under utf8, as
381 # there are plenty of other cases that catch these byte numbers.)
383 my $to_bytes = length $to;
384 foreach my $from_map (@from) {
385 if (! exists $has_test_by_byte_count{length $from_map}{$to_bytes}) {
386 add_test($to, @from);
392 # For each range type, test additionally a character that folds to itself
394 add_test(chr utf8::unicode_to_native(0xF7), chr utf8::unicode_to_native(0xF7));
395 add_test(chr 0x2C7, chr 0x2C7);
397 # To cut down on the number of tests
398 my $has_tested_aa_above_latin1;
399 my $has_tested_latin1_aa;
400 my $has_tested_ascii_aa;
401 my $has_tested_l_above_latin1;
402 my $has_tested_above_latin1_l;
403 my $has_tested_ascii_l;
404 my $has_tested_above_latin1_d;
405 my $has_tested_ascii_d;
406 my $has_tested_non_latin1_d;
407 my $has_tested_above_latin1_a;
408 my $has_tested_ascii_a;
409 my $has_tested_non_latin1_a;
411 # For use by pairs() in generating combinations
417 # Returns all ordered combinations of pairs of elements from the input array.
418 # It doesn't return pairs like (a, a), (b, b). Change the slice to an array
419 # to do that. This was just to have fewer tests.
421 #print STDERR __LINE__, ": ", join(" XXX ", map { sprintf "%04X", $_ } @_), "\n";
422 map { prefix $_[$_], @_[0..$_-1, $_+1..$#_] } 0..$#_
427 my @charsets = qw(d u a aa);
428 if (locales_enabled('LC_CTYPE')) {
429 my $current_locale = POSIX::setlocale( &POSIX::LC_CTYPE, "C") // "";
430 if ($current_locale eq 'C') {
433 # Some implementations don't have the 128-255 range characters all
434 # mean nothing under the C locale (an example being VMS). This is
435 # legal, but since we don't know what the right answers should be,
436 # skip the locale tests in that situation.
437 for my $i (128 .. 255) {
438 my $char = chr(utf8::unicode_to_native($i));
439 goto skip_C_locale_tests if uc($char) ne $char || lc($char) ne $char;
445 # Look for utf8 locale. We use the pseudo-modifier 'L' to indicate
446 # that we really want /l, but change to a UTF-8 locale.
447 $utf8_locale = find_utf8_ctype_locale();
448 push @charsets, 'L' if defined $utf8_locale;
452 # Finally ready to do the tests
453 foreach my $test (sort { numerically } keys %tests) {
456 my $previous_pattern;
457 my @pairs = pairs(sort numerically $test, @{$tests{$test}});
459 # Each fold can be viewed as a closure of all the characters that
460 # participate in it. Look at each possible pairing from a closure, with the
461 # first member of the pair the target string to match against, and the
462 # second member forming the pattern. Thus each fold member gets tested as
463 # the string, and the pattern with every other member in the opposite role.
464 while (my $pair = shift @pairs) {
465 my ($target, $pattern) = @$pair;
467 # When testing a char that doesn't fold, we can get the same
468 # permutation twice; so skip all but the first.
469 next if $previous_target
470 && $previous_target == $target
471 && $previous_pattern == $pattern;
472 ($previous_target, $previous_pattern) = ($target, $pattern);
474 # Each side may be either a single char or a string. Extract each into an
475 # array (perhaps of length 1)
476 my @target, my @pattern;
477 @target = (ref $target) ? @$target : $target;
478 @pattern = (ref $pattern) ? @$pattern : $pattern;
480 # We are testing just folds to/from a single character. If our pairs
481 # happens to generate multi/multi, skip.
482 next if @target > 1 && @pattern > 1;
485 my @x_target = map { sprintf "\\x{%04X}", $_ } @target;
486 my @x_pattern = map { sprintf "\\x{%04X}", $_ } @pattern;
488 my $target_above_latin1 = grep { $_ > 255 } @target;
489 my $pattern_above_latin1 = grep { $_ > 255 } @pattern;
490 my $target_has_ascii = grep { utf8::native_to_unicode($_) < 128 } @target;
491 my $pattern_has_ascii = grep { utf8::native_to_unicode($_) < 128 } @pattern;
492 my $target_only_ascii = ! grep { utf8::native_to_unicode($_) > 127 } @target;
493 my $pattern_only_ascii = ! grep { utf8::native_to_unicode($_) > 127 } @pattern;
494 my $target_has_latin1 = grep { $_ < 256 } @target;
495 my $target_has_upper_latin1
496 = grep { $_ < 256 && utf8::native_to_unicode($_) > 127 } @target;
497 my $pattern_has_upper_latin1
498 = grep { $_ < 256 && utf8::native_to_unicode($_) > 127 } @pattern;
499 my $pattern_has_latin1 = grep { $_ < 256 } @pattern;
500 my $is_self = @target == 1 && @pattern == 1 && $target[0] == $pattern[0];
502 # We don't test multi-char folding into other multi-chars. We are testing
503 # a code point that folds to or from other characters. Find the single
504 # code point for diagnostic purposes. (If both are single, choose the
506 my $ord = @target == 1 ? $target[0] : $pattern[0];
507 my $progress = sprintf "%04X: \"%s\" and /%s/",
510 join("", @x_pattern);
513 # Now grind out tests, using various combinations.
514 foreach my $charset (@charsets) {
515 my $charset_mod = lc $charset;
516 my $current_locale = "";
517 if ($charset_mod eq 'l') {
518 $current_locale = POSIX::setlocale(&POSIX::LC_CTYPE,
522 $current_locale = 'C locale' if $current_locale eq 'C';
527 # To cut down somewhat on the enormous quantity of tests this currently
528 # runs, skip some for some of the character sets whose results aren't
529 # likely to differ from others. But run all tests on the code points
530 # that don't fold, plus one other set in each range group.
533 # /aa should only affect things with folds in the ASCII range. But, try
534 # it on one set in the other ranges just to make sure it doesn't break
536 if ($charset eq 'aa') {
538 # It may be that this $pair of code points to test are both
539 # non-ascii, but if either of them actually fold to ascii, that is
540 # suspect and should be tested. So for /aa, use whether their folds
542 my $target_has_ascii = $target_has_ascii;
543 my $pattern_has_ascii = $pattern_has_ascii;
544 if (! $target_has_ascii) {
545 foreach my $cp (@target) {
546 if (exists $folds{$cp}
547 && grep { utf8::native_to_unicode($_) < 128 } @{$folds{$cp}} )
549 $target_has_ascii = 1;
554 if (! $pattern_has_ascii) {
555 foreach my $cp (@pattern) {
556 if (exists $folds{$cp}
557 && grep { utf8::native_to_unicode($_) < 128 } @{$folds{$cp}} )
559 $pattern_has_ascii = 1;
565 if (! $target_has_ascii && ! $pattern_has_ascii) {
566 if ($target_above_latin1 || $pattern_above_latin1) {
567 next if defined $has_tested_aa_above_latin1
568 && $has_tested_aa_above_latin1 != $test;
569 $has_tested_aa_above_latin1 = $test;
571 next if defined $has_tested_latin1_aa
572 && $has_tested_latin1_aa != $test;
573 $has_tested_latin1_aa = $test;
575 elsif ($target_only_ascii && $pattern_only_ascii) {
577 # And, except for one set just to make sure, skip tests
578 # where both elements in the pair are ASCII. If one works for
579 # aa, the others are likely too. This skips tests where the
580 # fold is from non-ASCII to ASCII, but this part of the test
581 # is just about the ASCII components.
582 next if defined $has_tested_ascii_l
583 && $has_tested_ascii_l != $test;
584 $has_tested_ascii_l = $test;
587 elsif ($charset eq 'l') {
589 # For l, don't need to test beyond one set those things that are
590 # all above latin1, because unlikely to have different successes
591 # than /u. But, for the same reason as described in the /aa above,
592 # it is suspect and should be tested, if either of the folds are to
594 my $target_has_latin1 = $target_has_latin1;
595 my $pattern_has_latin1 = $pattern_has_latin1;
596 if (! $target_has_latin1) {
597 foreach my $cp (@target) {
598 if (exists $folds{$cp}
599 && grep { $_ < 256 } @{$folds{$cp}} )
601 $target_has_latin1 = 1;
606 if (! $pattern_has_latin1) {
607 foreach my $cp (@pattern) {
608 if (exists $folds{$cp}
609 && grep { $_ < 256 } @{$folds{$cp}} )
611 $pattern_has_latin1 = 1;
616 if (! $target_has_latin1 && ! $pattern_has_latin1) {
617 next if defined $has_tested_above_latin1_l
618 && $has_tested_above_latin1_l != $test;
619 $has_tested_above_latin1_l = $test;
621 elsif ($target_only_ascii && $pattern_only_ascii) {
623 # And, except for one set just to make sure, skip tests
624 # where both elements in the pair are ASCII. This is
625 # essentially the same reasoning as above for /aa.
626 next if defined $has_tested_ascii_l
627 && $has_tested_ascii_l != $test;
628 $has_tested_ascii_l = $test;
631 elsif ($charset eq 'd') {
632 # Similarly for d. Beyond one test (besides self) each, we don't
633 # test pairs that are both ascii; or both above latin1, or are
634 # combinations of ascii and above latin1.
635 if (! $target_has_upper_latin1 && ! $pattern_has_upper_latin1) {
636 if ($target_has_ascii && $pattern_has_ascii) {
637 next if defined $has_tested_ascii_d
638 && $has_tested_ascii_d != $test;
639 $has_tested_ascii_d = $test
641 elsif (! $target_has_latin1 && ! $pattern_has_latin1) {
642 next if defined $has_tested_above_latin1_d
643 && $has_tested_above_latin1_d != $test;
644 $has_tested_above_latin1_d = $test;
647 next if defined $has_tested_non_latin1_d
648 && $has_tested_non_latin1_d != $test;
649 $has_tested_non_latin1_d = $test;
653 elsif ($charset eq 'a') {
654 # Similarly for a. This should match identically to /u, so wasn't
655 # tested at all until a bug was found that was thereby missed.
656 # As a compromise, beyond one test (besides self) each, we don't
657 # test pairs that are both ascii; or both above latin1, or are
658 # combinations of ascii and above latin1.
659 if (! $target_has_upper_latin1 && ! $pattern_has_upper_latin1) {
660 if ($target_has_ascii && $pattern_has_ascii) {
661 next if defined $has_tested_ascii_a
662 && $has_tested_ascii_a != $test;
663 $has_tested_ascii_a = $test
665 elsif (! $target_has_latin1 && ! $pattern_has_latin1) {
666 next if defined $has_tested_above_latin1_a
667 && $has_tested_above_latin1_a != $test;
668 $has_tested_above_latin1_a = $test;
671 next if defined $has_tested_non_latin1_a
672 && $has_tested_non_latin1_a != $test;
673 $has_tested_non_latin1_a = $test;
679 foreach my $utf8_target (0, 1) { # Both utf8 and not, for
681 my $upgrade_target = "";
683 # These must already be in utf8 because the string to match has
684 # something above latin1. So impossible to test if to not to be in
685 # utf8; and otherwise, no upgrade is needed.
686 next if $target_above_latin1 && ! $utf8_target;
687 $upgrade_target = ' utf8::upgrade($c);' if ! $target_above_latin1 && $utf8_target;
689 foreach my $utf8_pattern (0, 1) {
690 next if $pattern_above_latin1 && ! $utf8_pattern;
692 # Our testing of 'l' uses the POSIX locale, which is ASCII-only
693 my $uni_semantics = $charset ne 'l' && ($utf8_target || $charset eq 'u' || $charset eq 'L' || ($charset eq 'd' && $utf8_pattern) || $charset =~ /a/);
694 my $upgrade_pattern = "";
695 $upgrade_pattern = ' utf8::upgrade($p);' if ! $pattern_above_latin1 && $utf8_pattern;
697 my $lhs = join "", @x_target;
698 my $lhs_str = eval qq{"$lhs"}; fail($@) if $@;
699 my @rhs = @x_pattern;
700 my $rhs = join "", @rhs;
701 my $should_fail = (! $uni_semantics && $ord < 256 && ! $is_self && utf8::native_to_unicode($ord) >= 128)
702 || ($charset eq 'aa' && $target_has_ascii != $pattern_has_ascii)
703 || ($charset eq 'l' && $target_has_latin1 != $pattern_has_latin1);
705 # Do simple tests of referencing capture buffers, named and
708 $op = '!~' if $should_fail;
710 my $todo = 0; # No longer any todo's
711 my $eval = "my \$c = \"$lhs$rhs\"; my \$p = qr/(?$charset_mod:^($rhs)\\1\$)/i;$upgrade_target$upgrade_pattern \$c $op \$p";
712 run_test($eval, $todo, ($charset_mod eq 'l'), "");
714 $eval = "my \$c = \"$lhs$rhs\"; my \$p = qr/(?$charset_mod:^(?<grind>$rhs)\\k<grind>\$)/i;$upgrade_target$upgrade_pattern \$c $op \$p";
715 run_test($eval, $todo, ($charset_mod eq 'l'), "");
718 $eval = "my \$c = \"$rhs$lhs\"; my \$p = qr/(?$charset_mod:^($rhs)\\1\$)/i;$upgrade_target$upgrade_pattern \$c $op \$p";
719 run_test($eval, "", ($charset_mod eq 'l'), "");
721 $eval = "my \$c = \"$rhs$lhs\"; my \$p = qr/(?$charset_mod:^(?<grind>$rhs)\\k<grind>\$)/i;$upgrade_target$upgrade_pattern \$c $op \$p";
722 run_test($eval, "", ($charset_mod eq 'l'), "");
725 # See if works on what could be a simple trie.
728 # Keep the alternate | branch the same length as the tested one so
729 # that it's length doesn't influence things
730 my $evaled = eval "\"$rhs\""; # Convert e.g. \x{foo} into its
733 $alternate = 'q' x length $evaled;
735 $eval = "my \$c = \"$lhs\"; my \$p = qr/$rhs|$alternate/i$charset_mod;$upgrade_target$upgrade_pattern \$c $op \$p";
736 run_test($eval, "", ($charset_mod eq 'l'), "");
738 # Check that works when the folded character follows something that
739 # is quantified. This test knows the regex code internals to the
740 # extent that it knows this is a potential problem, and that there
741 # are three different types of quantifiers generated: 1) The thing
742 # being quantified matches a single character; 2) it matches more
743 # than one character, but is fixed width; 3) it can match a variable
744 # number of characters. (It doesn't know that case 3 shouldn't
745 # matter, since it doesn't do anything special for the character
746 # following the quantifier; nor that some of the different
747 # quantifiers execute the same underlying code, as these tests are
748 # quick, and this insulates these tests from changes in the
750 for my $quantifier ('?', '??', '*', '*?', '+', '+?', '{1,2}', '{1,2}?') {
751 $eval = "my \$c = \"_$lhs\"; my \$p = qr/(?$charset_mod:.$quantifier$rhs)/i;$upgrade_target$upgrade_pattern \$c $op \$p";
752 run_test($eval, "", ($charset_mod eq 'l'), "");
753 $eval = "my \$c = \"__$lhs\"; my \$p = qr/(?$charset_mod:(?:..)$quantifier$rhs)/i;$upgrade_target$upgrade_pattern \$c $op \$p";
754 run_test($eval, "", ($charset_mod eq 'l'), "");
755 $eval = "my \$c = \"__$lhs\"; my \$p = qr/(?$charset_mod:(?:.|\\R)$quantifier$rhs)/i;$upgrade_target$upgrade_pattern \$c $op \$p";
756 run_test($eval, "", ($charset_mod eq 'l'), "");
759 foreach my $bracketed (0, 1) { # Put rhs in [...], or not
760 next if $bracketed && @pattern != 1; # bracketed makes these
761 # or's instead of a sequence
762 foreach my $optimize_bracketed (0, 1) {
763 next if $optimize_bracketed && ! $bracketed;
764 foreach my $inverted (0,1) {
765 next if $inverted && ! $bracketed; # inversion only valid
767 next if $inverted && @target != 1; # [perl #89750] multi-char
768 # not valid in [^...]
770 # In some cases, add an extra character that doesn't fold, and
771 # looks ok in the output.
772 my $extra_char = "_";
773 foreach my $prepend ("", $extra_char) {
774 foreach my $append ("", $extra_char) {
776 # Assemble the rhs. Put each character in a separate
777 # bracketed if using charclasses. This creates a stress on
778 # the code to span a match across multiple elements
780 foreach my $rhs_char (@rhs) {
781 $rhs .= '[' if $bracketed;
782 $rhs .= '^' if $inverted;
785 # Add a character to the class, so class doesn't get
786 # optimized out, unless we are testing that optimization
787 $rhs .= '_' if $optimize_bracketed;
788 $rhs .= ']' if $bracketed;
791 # Add one of: no capturing parens
794 # Use quantifiers and extra variable width matches inside
795 # them to keep some optimizations from happening
796 foreach my $parend (0, 1, 2) {
797 my $interior = (! $parend)
802 foreach my $quantifier ("", '?', '*', '+', '{1,3}') {
804 # Perhaps should be TODOs, as are unimplemented, but
805 # maybe will never be implemented
806 next if @pattern != 1 && $quantifier;
808 # A ? or * quantifier normally causes the thing to be
809 # able to match a null string
810 my $quantifier_can_match_null = $quantifier eq '?'
811 || $quantifier eq '*';
813 # But since we only quantify the last character in a
814 # multiple fold, the other characters will have width,
815 # except if we are quantifying the whole rhs
816 my $can_match_null = $quantifier_can_match_null
817 && (@rhs == 1 || $parend);
819 foreach my $l_anchor ("", '^') { # '\A' didn't change
821 foreach my $r_anchor ("", '$') { # '\Z', '\z' didn't
823 # The folded part can match the null string if it
824 # isn't required to have width, and there's not
825 # something on one or both sides that force it to.
826 my $both_sides = ($l_anchor && $r_anchor)
827 || ($l_anchor && $append)
828 || ($r_anchor && $prepend)
829 || ($prepend && $append);
830 my $must_match = ! $can_match_null || $both_sides;
831 # for performance, but doing this missed many failures
832 #next unless $must_match;
833 my $quantified = "(?$charset_mod:$l_anchor$prepend$interior${quantifier}$append$r_anchor)";
835 if ($must_match && $should_fail) {
840 $op = ! $op if $must_match && $inverted;
842 if ($inverted && @target > 1) {
843 # When doing an inverted match against a
844 # multi-char target, and there is not something on
845 # the left to anchor the match, if it shouldn't
846 # succeed, skip, as what will happen (when working
847 # correctly) is that it will match the first
848 # position correctly, and then be inverted to not
849 # match; then it will go to the second position
850 # where it won't match, but get inverted to match,
851 # and hence succeeding.
852 next if ! ($l_anchor || $prepend) && ! $op;
854 # Can't ever match for latin1 code points non-uni
855 # semantics that have a inverted multi-char fold
856 # when there is something on both sides and the
857 # quantifier isn't such as to span the required
858 # width, which is 2 or 3.
859 $op = 0 if $ord < 255
862 && ( ! $quantifier || $quantifier eq '?')
865 # Similarly can't ever match when inverting a
866 # multi-char fold for /aa and the quantifier
867 # isn't sufficient to allow it to span to both
869 $op = 0 if $target_has_ascii
872 && ( ! $quantifier || $quantifier eq '?')
876 $op = 0 if $target_has_latin1 && $charset eq 'l'
878 && ( ! $quantifier || $quantifier eq '?')
884 if ($charset_mod eq 'l') {
885 $desc .= 'setlocale(LC_CTYPE, "'
886 . POSIX::setlocale(&POSIX::LC_CTYPE)
889 $desc .= "my \$c = \"$prepend$lhs$append\"; "
890 . "my \$p = qr/$quantified/i;"
891 . "$upgrade_target$upgrade_pattern "
892 . "\$c " . ($op ? "=~" : "!~") . " \$p; ";
895 "; uni_semantics=$uni_semantics, "
896 . "should_fail=$should_fail, "
897 . "bracketed=$bracketed, "
898 . "prepend=$prepend, "
901 . "quantifier=$quantifier, "
902 . "l_anchor=$l_anchor, "
903 . "r_anchor=$r_anchor; "
904 . "pattern_above_latin1=$pattern_above_latin1; "
905 . "utf8_pattern=$utf8_pattern"
909 my $c = "$prepend$lhs_str$append";
910 my $p = qr/$quantified/i;
911 utf8::upgrade($c) if length($upgrade_target);
912 utf8::upgrade($p) if length($upgrade_pattern);
913 my $res = $op ? ($c =~ $p): ($c !~ $p);
915 if (!$res || $list_all_tests) {
916 # Failed or debug; output the result
918 ok($res, "test $count - $desc");
920 # Just count the test as passed
935 unless($list_all_tests) {
937 is $okays, $this_iteration, "$okays subtests ok for"
939 . (($charset_mod eq 'l') ? " ($current_locale)" : "")
940 . ', target="' . join("", @x_target) . '",'
941 . ' pat="' . join("", @x_pattern) . '"';