Commit | Line | Data |
---|---|---|
e425a60b YO |
1 | #!./perl |
2 | # | |
c5de0829 | 3 | # This is a home for regular expression tests that do not fit into |
e425a60b YO |
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 | ||
3fe9640a JH |
7 | BEGIN { |
8 | chdir 't' if -d 't'; | |
3fe9640a | 9 | require './test.pl'; |
624c42e2 | 10 | set_up_inc(qw '../lib .'); |
8c892c9b | 11 | require './charset_tools.pl'; |
624c42e2 | 12 | skip_all_if_miniperl("miniperl can't load Tie::Hash::NamedCapture, need for %+ and %-"); |
5b39eca1 | 13 | } |
3fe9640a | 14 | |
e425a60b YO |
15 | use strict; |
16 | use warnings; | |
17 | use 5.010; | |
fee50582 | 18 | our ($REGMARK, $REGERROR); |
e425a60b | 19 | |
e425a60b YO |
20 | sub run_tests; |
21 | ||
22 | $| = 1; | |
23 | ||
9d45b377 | 24 | run_tests() unless caller; |
e425a60b YO |
25 | |
26 | # | |
27 | # Tests start here. | |
28 | # | |
29 | sub run_tests { | |
30 | ||
e425a60b | 31 | { |
e425a60b YO |
32 | # Japhy -- added 03/03/2001 |
33 | () = (my $str = "abc") =~ /(...)/; | |
34 | $str = "def"; | |
de26e0cc | 35 | is($1, "abc", 'Changing subject does not modify $1'); |
e425a60b YO |
36 | } |
37 | ||
e425a60b YO |
38 | SKIP: |
39 | { | |
40 | # The trick is that in EBCDIC the explicit numeric range should | |
41 | # match (as also in non-EBCDIC) but the explicit alphabetic range | |
42 | # should not match. | |
b7ad6c3e KW |
43 | like "\x8e", qr/[\x89-\x91]/, '"\x8e" =~ /[\x89-\x91]/'; |
44 | like "\xce", qr/[\xc9-\xd1]/, '"\xce" =~ /[\xc9-\xd1]/'; | |
45 | like "\xd0", qr/[\xc9-\xd1]/, '"\xd0" =~ /[\xc9-\xd1]/'; | |
e425a60b YO |
46 | |
47 | skip "Not an EBCDIC platform", 2 unless ord ('i') == 0x89 && | |
48 | ord ('J') == 0xd1; | |
49 | ||
50 | # In most places these tests would succeed since \x8e does not | |
51 | # in most character sets match 'i' or 'j' nor would \xce match | |
52 | # 'I' or 'J', but strictly speaking these tests are here for | |
53 | # the good of EBCDIC, so let's test these only there. | |
b33825c4 NC |
54 | unlike("\x8e", qr/[i-j]/, '"\x8e" !~ /[i-j]/'); |
55 | unlike("\xce", qr/[I-J]/, '"\xce" !~ /[I-J]/'); | |
2ca8589c | 56 | unlike("\xd0", qr/[I-J]/, '"\xd0" !~ /[I-J]/'); |
e425a60b YO |
57 | } |
58 | ||
e425a60b | 59 | { |
b7ad6c3e KW |
60 | like "\x{ab}", qr/\x{ab}/, '"\x{ab}" =~ /\x{ab}/ '; |
61 | like "\x{abcd}", qr/\x{abcd}/, '"\x{abcd}" =~ /\x{abcd}/'; | |
e425a60b YO |
62 | } |
63 | ||
e425a60b | 64 | { |
ee95e30c | 65 | my $message = 'bug id 20001008.001 (#4407)'; |
e425a60b | 66 | |
f36cc39d KW |
67 | my $strasse = "stra" . uni_to_native("\337") . "e"; |
68 | my @x = ("$strasse 138", "$strasse 138"); | |
e425a60b | 69 | for (@x) { |
de946258 NC |
70 | ok(s/(\d+)\s*([\w\-]+)/$1 . uc $2/e, $message); |
71 | ok(my ($latin) = /^(.+)(?:\s+\d)/, $message); | |
f36cc39d KW |
72 | is($latin, $strasse, $message); |
73 | ok($latin =~ s/$strasse/straße/, $message); | |
e425a60b YO |
74 | # |
75 | # Previous code follows, but outcommented - there were no tests. | |
76 | # | |
77 | # $latin =~ s/stra\337e/straße/; # \303\237 after the 2nd a | |
78 | # use utf8; # needed for the raw UTF-8 | |
79 | # $latin =~ s!(s)tr(?:aß|s+e)!$1tr.!; # \303\237 after the a | |
80 | } | |
81 | } | |
82 | ||
e425a60b | 83 | { |
de946258 NC |
84 | my $message = 'Test \x escapes'; |
85 | ok("ba\xd4c" =~ /([a\xd4]+)/ && $1 eq "a\xd4", $message); | |
86 | ok("ba\xd4c" =~ /([a\xd4]+)/ && $1 eq "a\x{d4}", $message); | |
87 | ok("ba\x{d4}c" =~ /([a\xd4]+)/ && $1 eq "a\x{d4}", $message); | |
88 | ok("ba\x{d4}c" =~ /([a\xd4]+)/ && $1 eq "a\xd4", $message); | |
89 | ok("ba\xd4c" =~ /([a\x{d4}]+)/ && $1 eq "a\xd4", $message); | |
90 | ok("ba\xd4c" =~ /([a\x{d4}]+)/ && $1 eq "a\x{d4}", $message); | |
91 | ok("ba\x{d4}c" =~ /([a\x{d4}]+)/ && $1 eq "a\x{d4}", $message); | |
92 | ok("ba\x{d4}c" =~ /([a\x{d4}]+)/ && $1 eq "a\xd4", $message); | |
e425a60b YO |
93 | } |
94 | ||
e425a60b | 95 | { |
de946258 | 96 | my $message = 'Match code points > 255'; |
e425a60b | 97 | $_ = "abc\x{100}\x{200}\x{300}\x{380}\x{400}defg"; |
de946258 NC |
98 | ok(/(.\x{300})./, $message); |
99 | ok($` eq "abc\x{100}" && length ($`) == 4, $message); | |
100 | ok($& eq "\x{200}\x{300}\x{380}" && length ($&) == 3, $message); | |
101 | ok($' eq "\x{400}defg" && length ($') == 5, $message); | |
102 | ok($1 eq "\x{200}\x{300}" && length ($1) == 2, $message); | |
e425a60b YO |
103 | } |
104 | ||
e425a60b YO |
105 | { |
106 | my $x = "\x{10FFFD}"; | |
107 | $x =~ s/(.)/$1/g; | |
108 | ok ord($x) == 0x10FFFD && length($x) == 1, "From Robin Houston"; | |
109 | } | |
110 | ||
e425a60b YO |
111 | { |
112 | my %d = ( | |
113 | "7f" => [0, 0, 0], | |
114 | "80" => [1, 1, 0], | |
115 | "ff" => [1, 1, 0], | |
116 | "100" => [0, 1, 1], | |
117 | ); | |
a2d248f9 | 118 | |
e425a60b | 119 | while (my ($code, $match) = each %d) { |
de946258 | 120 | my $message = "Properties of \\x$code"; |
e425a60b | 121 | my $char = eval qq ["\\x{$code}"]; |
a2d248f9 | 122 | |
de946258 NC |
123 | is(0 + ($char =~ /[\x80-\xff]/), $$match[0], $message); |
124 | is(0 + ($char =~ /[\x80-\x{100}]/), $$match[1], $message); | |
125 | is(0 + ($char =~ /[\x{100}]/), $$match[2], $message); | |
e425a60b YO |
126 | } |
127 | } | |
128 | ||
e425a60b YO |
129 | { |
130 | # From Japhy | |
e4e5d8ba | 131 | foreach (qw(c g o)) { |
4d18b353 NC |
132 | warning_like(sub {'' =~ "(?$_)"}, qr/^Useless \(\?$_\)/); |
133 | warning_like(sub {'' =~ "(?-$_)"}, qr/^Useless \(\?-$_\)/); | |
e4e5d8ba | 134 | } |
e425a60b YO |
135 | |
136 | # Now test multi-error regexes | |
f4554ed5 NC |
137 | foreach (['(?g-o)', qr/^Useless \(\?g\)/, qr/^Useless \(\?-o\)/], |
138 | ['(?g-c)', qr/^Useless \(\?g\)/, qr/^Useless \(\?-c\)/], | |
139 | # (?c) means (?g) error won't be thrown | |
140 | ['(?o-cg)', qr/^Useless \(\?o\)/, qr/^Useless \(\?-c\)/], | |
141 | ['(?ogc)', qr/^Useless \(\?o\)/, qr/^Useless \(\?g\)/, | |
142 | qr/^Useless \(\?c\)/], | |
143 | ) { | |
144 | my ($re, @warnings) = @$_; | |
145 | warnings_like(sub {eval "qr/$re/"}, \@warnings, "qr/$re/ warns"); | |
146 | } | |
e425a60b YO |
147 | } |
148 | ||
e425a60b | 149 | { |
de946258 | 150 | my $message = "/x tests"; |
e425a60b | 151 | $_ = "foo"; |
14358a41 | 152 | foreach my $pat (<<" --", <<" --") { |
e425a60b YO |
153 | /f |
154 | o\r | |
155 | o | |
156 | \$ | |
157 | /x | |
158 | -- | |
e425a60b YO |
159 | /f |
160 | o | |
161 | o | |
162 | \$\r | |
163 | /x | |
164 | -- | |
14358a41 NC |
165 | is(eval $pat, 1, $message); |
166 | is($@, '', $message); | |
167 | } | |
e425a60b YO |
168 | } |
169 | ||
e425a60b | 170 | { |
de946258 | 171 | my $message = "/o feature"; |
e425a60b | 172 | sub test_o {$_ [0] =~ /$_[1]/o; return $1} |
de946258 NC |
173 | is(test_o ('abc', '(.)..'), 'a', $message); |
174 | is(test_o ('abc', '..(.)'), 'a', $message); | |
e425a60b YO |
175 | } |
176 | ||
e425a60b YO |
177 | { |
178 | # Test basic $^N usage outside of a regex | |
de946258 | 179 | my $message = '$^N usage outside of a regex'; |
e425a60b | 180 | my $x = "abcdef"; |
de946258 NC |
181 | ok(($x =~ /cde/ and !defined $^N), $message); |
182 | ok(($x =~ /(cde)/ and $^N eq "cde"), $message); | |
183 | ok(($x =~ /(c)(d)(e)/ and $^N eq "e"), $message); | |
184 | ok(($x =~ /(c(d)e)/ and $^N eq "cde"), $message); | |
185 | ok(($x =~ /(foo)|(c(d)e)/ and $^N eq "cde"), $message); | |
186 | ok(($x =~ /(c(d)e)|(foo)/ and $^N eq "cde"), $message); | |
187 | ok(($x =~ /(c(d)e)|(abc)/ and $^N eq "abc"), $message); | |
188 | ok(($x =~ /(c(d)e)|(abc)x/ and $^N eq "cde"), $message); | |
189 | ok(($x =~ /(c(d)e)(abc)?/ and $^N eq "cde"), $message); | |
190 | ok(($x =~ /(?:c(d)e)/ and $^N eq "d"), $message); | |
191 | ok(($x =~ /(?:c(d)e)(?:f)/ and $^N eq "d"), $message); | |
192 | ok(($x =~ /(?:([abc])|([def]))*/ and $^N eq "f"), $message); | |
193 | ok(($x =~ /(?:([ace])|([bdf]))*/ and $^N eq "f"), $message); | |
194 | ok(($x =~ /(([ace])|([bd]))*/ and $^N eq "e"), $message); | |
195 | {ok(($x =~ /(([ace])|([bdf]))*/ and $^N eq "f"), $message);} | |
e425a60b YO |
196 | ## Test to see if $^N is automatically localized -- it should now |
197 | ## have the value set in the previous test. | |
de26e0cc | 198 | is($^N, "e", '$^N is automatically localized'); |
e425a60b YO |
199 | |
200 | # Now test inside (?{ ... }) | |
de946258 | 201 | $message = '$^N usage inside (?{ ... })'; |
e425a60b | 202 | our ($y, $z); |
de946258 NC |
203 | ok(($x =~ /a([abc])(?{$y=$^N})c/ and $y eq "b"), $message); |
204 | ok(($x =~ /a([abc]+)(?{$y=$^N})d/ and $y eq "bc"), $message); | |
205 | ok(($x =~ /a([abcdefg]+)(?{$y=$^N})d/ and $y eq "bc"), $message); | |
206 | ok(($x =~ /(a([abcdefg]+)(?{$y=$^N})d)(?{$z=$^N})e/ and $y eq "bc" | |
207 | and $z eq "abcd"), $message); | |
208 | ok(($x =~ /(a([abcdefg]+)(?{$y=$^N})de)(?{$z=$^N})/ and $y eq "bc" | |
209 | and $z eq "abcde"), $message); | |
e425a60b YO |
210 | |
211 | } | |
212 | ||
e425a60b YO |
213 | SKIP: |
214 | { | |
215 | ## Should probably put in tests for all the POSIX stuff, | |
216 | ## but not sure how to guarantee a specific locale...... | |
217 | ||
de946258 | 218 | my $message = 'Test [[:cntrl:]]'; |
e425a60b YO |
219 | my $AllBytes = join "" => map {chr} 0 .. 255; |
220 | (my $x = $AllBytes) =~ s/[[:cntrl:]]//g; | |
628d95af KW |
221 | $x = join "", sort { $a cmp $b } |
222 | map { chr utf8::native_to_unicode(ord $_) } split "", $x; | |
de946258 | 223 | is($x, join("", map {chr} 0x20 .. 0x7E, 0x80 .. 0xFF), $message); |
e425a60b YO |
224 | |
225 | ($x = $AllBytes) =~ s/[^[:cntrl:]]//g; | |
628d95af KW |
226 | $x = join "", sort { $a cmp $b } |
227 | map { chr utf8::native_to_unicode(ord $_) } split "", $x; | |
de946258 | 228 | is($x, (join "", map {chr} 0x00 .. 0x1F, 0x7F), $message); |
e425a60b YO |
229 | } |
230 | ||
e425a60b YO |
231 | { |
232 | # With /s modifier UTF8 chars were interpreted as bytes | |
de946258 | 233 | my $message = "UTF-8 chars aren't bytes"; |
e425a60b YO |
234 | my $a = "Hello \x{263A} World"; |
235 | my @a = ($a =~ /./gs); | |
de946258 | 236 | is($#a, 12, $message); |
e425a60b YO |
237 | } |
238 | ||
e425a60b | 239 | { |
e425a60b YO |
240 | no warnings 'digit'; |
241 | # Check that \x## works. 5.6.1 and 5.005_03 fail some of these. | |
242 | my $x; | |
243 | $x = "\x4e" . "E"; | |
b7ad6c3e | 244 | like ($x, qr/^\x4EE$/, "Check only 2 bytes of hex are matched."); |
e425a60b YO |
245 | |
246 | $x = "\x4e" . "i"; | |
b7ad6c3e | 247 | like ($x, qr/^\x4Ei$/, "Check that invalid hex digit stops it (2)"); |
e425a60b YO |
248 | |
249 | $x = "\x4" . "j"; | |
b7ad6c3e | 250 | like ($x, qr/^\x4j$/, "Check that invalid hex digit stops it (1)"); |
e425a60b YO |
251 | |
252 | $x = "\x0" . "k"; | |
b7ad6c3e | 253 | like ($x, qr/^\xk$/, "Check that invalid hex digit stops it (0)"); |
e425a60b YO |
254 | |
255 | $x = "\x0" . "x"; | |
b7ad6c3e | 256 | like ($x, qr/^\xx$/, "\\xx isn't to be treated as \\0"); |
e425a60b YO |
257 | |
258 | $x = "\x0" . "xa"; | |
b7ad6c3e | 259 | like ($x, qr/^\xxa$/, "\\xxa isn't to be treated as \\xa"); |
e425a60b YO |
260 | |
261 | $x = "\x9" . "_b"; | |
b7ad6c3e | 262 | like ($x, qr/^\x9_b$/, "\\x9_b isn't to be treated as \\x9b"); |
e425a60b YO |
263 | |
264 | # and now again in [] ranges | |
265 | ||
266 | $x = "\x4e" . "E"; | |
b7ad6c3e | 267 | like ($x, qr/^[\x4EE]{2}$/, "Check only 2 bytes of hex are matched."); |
e425a60b YO |
268 | |
269 | $x = "\x4e" . "i"; | |
b7ad6c3e | 270 | like ($x, qr/^[\x4Ei]{2}$/, "Check that invalid hex digit stops it (2)"); |
e425a60b YO |
271 | |
272 | $x = "\x4" . "j"; | |
b7ad6c3e | 273 | like ($x, qr/^[\x4j]{2}$/, "Check that invalid hex digit stops it (1)"); |
e425a60b YO |
274 | |
275 | $x = "\x0" . "k"; | |
b7ad6c3e | 276 | like ($x, qr/^[\xk]{2}$/, "Check that invalid hex digit stops it (0)"); |
e425a60b YO |
277 | |
278 | $x = "\x0" . "x"; | |
b7ad6c3e | 279 | like ($x, qr/^[\xx]{2}$/, "\\xx isn't to be treated as \\0"); |
e425a60b YO |
280 | |
281 | $x = "\x0" . "xa"; | |
b7ad6c3e | 282 | like ($x, qr/^[\xxa]{3}$/, "\\xxa isn't to be treated as \\xa"); |
e425a60b YO |
283 | |
284 | $x = "\x9" . "_b"; | |
b7ad6c3e | 285 | like ($x, qr/^[\x9_b]{3}$/, "\\x9_b isn't to be treated as \\x9b"); |
e425a60b YO |
286 | |
287 | # Check that \x{##} works. 5.6.1 fails quite a few of these. | |
288 | ||
289 | $x = "\x9b"; | |
b7ad6c3e | 290 | like ($x, qr/^\x{9_b}$/, "\\x{9_b} is to be treated as \\x9b"); |
e425a60b YO |
291 | |
292 | $x = "\x9b" . "y"; | |
b7ad6c3e | 293 | like ($x, qr/^\x{9_b}y$/, "\\x{9_b} is to be treated as \\x9b (again)"); |
e425a60b YO |
294 | |
295 | $x = "\x9b" . "y"; | |
b7ad6c3e | 296 | like ($x, qr/^\x{9b_}y$/, "\\x{9b_} is to be treated as \\x9b"); |
e425a60b YO |
297 | |
298 | $x = "\x9b" . "y"; | |
b7ad6c3e | 299 | like ($x, qr/^\x{9_bq}y$/, "\\x{9_bc} is to be treated as \\x9b"); |
e425a60b YO |
300 | |
301 | $x = "\x0" . "y"; | |
b7ad6c3e | 302 | like ($x, qr/^\x{x9b}y$/, "\\x{x9b} is to be treated as \\x0"); |
e425a60b YO |
303 | |
304 | $x = "\x0" . "y"; | |
b7ad6c3e | 305 | like ($x, qr/^\x{0x9b}y$/, "\\x{0x9b} is to be treated as \\x0"); |
e425a60b YO |
306 | |
307 | $x = "\x9b" . "y"; | |
b7ad6c3e | 308 | like ($x, qr/^\x{09b}y$/, "\\x{09b} is to be treated as \\x9b"); |
e425a60b YO |
309 | |
310 | $x = "\x9b"; | |
b7ad6c3e | 311 | like ($x, qr/^[\x{9_b}]$/, "\\x{9_b} is to be treated as \\x9b"); |
e425a60b YO |
312 | |
313 | $x = "\x9b" . "y"; | |
b7ad6c3e | 314 | like ($x, qr/^[\x{9_b}y]{2}$/, |
e425a60b YO |
315 | "\\x{9_b} is to be treated as \\x9b (again)"); |
316 | ||
317 | $x = "\x9b" . "y"; | |
b7ad6c3e | 318 | like ($x, qr/^[\x{9b_}y]{2}$/, "\\x{9b_} is to be treated as \\x9b"); |
e425a60b YO |
319 | |
320 | $x = "\x9b" . "y"; | |
b7ad6c3e | 321 | like ($x, qr/^[\x{9_bq}y]{2}$/, "\\x{9_bc} is to be treated as \\x9b"); |
e425a60b YO |
322 | |
323 | $x = "\x0" . "y"; | |
b7ad6c3e | 324 | like ($x, qr/^[\x{x9b}y]{2}$/, "\\x{x9b} is to be treated as \\x0"); |
e425a60b YO |
325 | |
326 | $x = "\x0" . "y"; | |
b7ad6c3e | 327 | like ($x, qr/^[\x{0x9b}y]{2}$/, "\\x{0x9b} is to be treated as \\x0"); |
e425a60b YO |
328 | |
329 | $x = "\x9b" . "y"; | |
b7ad6c3e | 330 | like ($x, qr/^[\x{09b}y]{2}$/, "\\x{09b} is to be treated as \\x9b"); |
e425a60b YO |
331 | |
332 | } | |
333 | ||
e425a60b YO |
334 | { |
335 | # High bit bug -- japhy | |
336 | my $x = "ab\200d"; | |
b7ad6c3e | 337 | like $x, qr/.*?\200/, "High bit fine"; |
e425a60b YO |
338 | } |
339 | ||
e425a60b YO |
340 | { |
341 | # The basic character classes and Unicode | |
b7ad6c3e KW |
342 | like "\x{0100}", qr/\w/, 'LATIN CAPITAL LETTER A WITH MACRON in /\w/'; |
343 | like "\x{0660}", qr/\d/, 'ARABIC-INDIC DIGIT ZERO in /\d/'; | |
344 | like "\x{1680}", qr/\s/, 'OGHAM SPACE MARK in /\s/'; | |
e425a60b YO |
345 | } |
346 | ||
e425a60b | 347 | { |
de946258 NC |
348 | my $message = "Folding matches and Unicode"; |
349 | like("a\x{100}", qr/A/i, $message); | |
350 | like("A\x{100}", qr/a/i, $message); | |
351 | like("a\x{100}", qr/a/i, $message); | |
352 | like("A\x{100}", qr/A/i, $message); | |
353 | like("\x{101}a", qr/\x{100}/i, $message); | |
354 | like("\x{100}a", qr/\x{100}/i, $message); | |
355 | like("\x{101}a", qr/\x{101}/i, $message); | |
356 | like("\x{100}a", qr/\x{101}/i, $message); | |
357 | like("a\x{100}", qr/A\x{100}/i, $message); | |
358 | like("A\x{100}", qr/a\x{100}/i, $message); | |
359 | like("a\x{100}", qr/a\x{100}/i, $message); | |
360 | like("A\x{100}", qr/A\x{100}/i, $message); | |
361 | like("a\x{100}", qr/[A]/i, $message); | |
362 | like("A\x{100}", qr/[a]/i, $message); | |
363 | like("a\x{100}", qr/[a]/i, $message); | |
364 | like("A\x{100}", qr/[A]/i, $message); | |
365 | like("\x{101}a", qr/[\x{100}]/i, $message); | |
366 | like("\x{100}a", qr/[\x{100}]/i, $message); | |
367 | like("\x{101}a", qr/[\x{101}]/i, $message); | |
368 | like("\x{100}a", qr/[\x{101}]/i, $message); | |
e425a60b YO |
369 | } |
370 | ||
e425a60b YO |
371 | { |
372 | use charnames ':full'; | |
de946258 | 373 | my $message = "Folding 'LATIN LETTER A WITH GRAVE'"; |
e425a60b YO |
374 | |
375 | my $lower = "\N{LATIN SMALL LETTER A WITH GRAVE}"; | |
376 | my $UPPER = "\N{LATIN CAPITAL LETTER A WITH GRAVE}"; | |
0f289c68 | 377 | |
de946258 NC |
378 | like($lower, qr/$UPPER/i, $message); |
379 | like($UPPER, qr/$lower/i, $message); | |
380 | like($lower, qr/[$UPPER]/i, $message); | |
381 | like($UPPER, qr/[$lower]/i, $message); | |
e425a60b | 382 | |
de946258 | 383 | $message = "Folding 'GREEK LETTER ALPHA WITH VRACHY'"; |
e425a60b YO |
384 | |
385 | $lower = "\N{GREEK CAPITAL LETTER ALPHA WITH VRACHY}"; | |
386 | $UPPER = "\N{GREEK SMALL LETTER ALPHA WITH VRACHY}"; | |
387 | ||
de946258 NC |
388 | like($lower, qr/$UPPER/i, $message); |
389 | like($UPPER, qr/$lower/i, $message); | |
390 | like($lower, qr/[$UPPER]/i, $message); | |
391 | like($UPPER, qr/[$lower]/i, $message); | |
e425a60b | 392 | |
de946258 | 393 | $message = "Folding 'LATIN LETTER Y WITH DIAERESIS'"; |
e425a60b YO |
394 | |
395 | $lower = "\N{LATIN SMALL LETTER Y WITH DIAERESIS}"; | |
396 | $UPPER = "\N{LATIN CAPITAL LETTER Y WITH DIAERESIS}"; | |
397 | ||
de946258 NC |
398 | like($lower, qr/$UPPER/i, $message); |
399 | like($UPPER, qr/$lower/i, $message); | |
400 | like($lower, qr/[$UPPER]/i, $message); | |
401 | like($UPPER, qr/[$lower]/i, $message); | |
e425a60b YO |
402 | } |
403 | ||
e425a60b YO |
404 | { |
405 | use charnames ':full'; | |
de946258 | 406 | my $message = "GREEK CAPITAL LETTER SIGMA vs " . |
e425a60b YO |
407 | "COMBINING GREEK PERISPOMENI"; |
408 | ||
409 | my $SIGMA = "\N{GREEK CAPITAL LETTER SIGMA}"; | |
410 | my $char = "\N{COMBINING GREEK PERISPOMENI}"; | |
411 | ||
c11a8df3 NC |
412 | warning_is(sub {unlike("_:$char:_", qr/_:$SIGMA:_/i, $message)}, undef, |
413 | 'Did not warn [change a5961de5f4215b5c]'); | |
e425a60b YO |
414 | } |
415 | ||
e425a60b | 416 | { |
de946258 | 417 | my $message = '\X'; |
e425a60b YO |
418 | use charnames ':full'; |
419 | ||
de946258 NC |
420 | ok("a!" =~ /^(\X)!/ && $1 eq "a", $message); |
421 | ok("\xDF!" =~ /^(\X)!/ && $1 eq "\xDF", $message); | |
422 | ok("\x{100}!" =~ /^(\X)!/ && $1 eq "\x{100}", $message); | |
423 | ok("\x{100}\x{300}!" =~ /^(\X)!/ && $1 eq "\x{100}\x{300}", $message); | |
424 | ok("\N{LATIN CAPITAL LETTER E}!" =~ /^(\X)!/ && | |
425 | $1 eq "\N{LATIN CAPITAL LETTER E}", $message); | |
426 | ok("\N{LATIN CAPITAL LETTER E}\N{COMBINING GRAVE ACCENT}!" | |
e425a60b | 427 | =~ /^(\X)!/ && |
de946258 | 428 | $1 eq "\N{LATIN CAPITAL LETTER E}\N{COMBINING GRAVE ACCENT}", $message); |
e425a60b | 429 | |
e425a60b YO |
430 | } |
431 | ||
e425a60b | 432 | { |
de946258 | 433 | my $message = "Final Sigma"; |
e425a60b YO |
434 | |
435 | my $SIGMA = "\x{03A3}"; # CAPITAL | |
436 | my $Sigma = "\x{03C2}"; # SMALL FINAL | |
437 | my $sigma = "\x{03C3}"; # SMALL | |
438 | ||
de946258 NC |
439 | like($SIGMA, qr/$SIGMA/i, $message); |
440 | like($SIGMA, qr/$Sigma/i, $message); | |
441 | like($SIGMA, qr/$sigma/i, $message); | |
e425a60b | 442 | |
de946258 NC |
443 | like($Sigma, qr/$SIGMA/i, $message); |
444 | like($Sigma, qr/$Sigma/i, $message); | |
445 | like($Sigma, qr/$sigma/i, $message); | |
e425a60b | 446 | |
de946258 NC |
447 | like($sigma, qr/$SIGMA/i, $message); |
448 | like($sigma, qr/$Sigma/i, $message); | |
449 | like($sigma, qr/$sigma/i, $message); | |
0f289c68 | 450 | |
de946258 NC |
451 | like($SIGMA, qr/[$SIGMA]/i, $message); |
452 | like($SIGMA, qr/[$Sigma]/i, $message); | |
453 | like($SIGMA, qr/[$sigma]/i, $message); | |
e425a60b | 454 | |
de946258 NC |
455 | like($Sigma, qr/[$SIGMA]/i, $message); |
456 | like($Sigma, qr/[$Sigma]/i, $message); | |
457 | like($Sigma, qr/[$sigma]/i, $message); | |
e425a60b | 458 | |
de946258 NC |
459 | like($sigma, qr/[$SIGMA]/i, $message); |
460 | like($sigma, qr/[$Sigma]/i, $message); | |
461 | like($sigma, qr/[$sigma]/i, $message); | |
e425a60b | 462 | |
de946258 | 463 | $message = "More final Sigma"; |
e425a60b YO |
464 | |
465 | my $S3 = "$SIGMA$Sigma$sigma"; | |
466 | ||
de946258 NC |
467 | ok(":$S3:" =~ /:(($SIGMA)+):/i && $1 eq $S3 && $2 eq $sigma, $message); |
468 | ok(":$S3:" =~ /:(($Sigma)+):/i && $1 eq $S3 && $2 eq $sigma, $message); | |
469 | ok(":$S3:" =~ /:(($sigma)+):/i && $1 eq $S3 && $2 eq $sigma, $message); | |
e425a60b | 470 | |
de946258 NC |
471 | ok(":$S3:" =~ /:(([$SIGMA])+):/i && $1 eq $S3 && $2 eq $sigma, $message); |
472 | ok(":$S3:" =~ /:(([$Sigma])+):/i && $1 eq $S3 && $2 eq $sigma, $message); | |
473 | ok(":$S3:" =~ /:(([$sigma])+):/i && $1 eq $S3 && $2 eq $sigma, $message); | |
e425a60b YO |
474 | } |
475 | ||
e425a60b YO |
476 | { |
477 | use charnames ':full'; | |
de946258 | 478 | my $message = "Parlez-Vous " . |
e425a60b YO |
479 | "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais?"; |
480 | ||
de946258 NC |
481 | ok("Fran\N{LATIN SMALL LETTER C}ais" =~ /Fran.ais/ && |
482 | $& eq "Francais", $message); | |
483 | ok("Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~ /Fran.ais/ && | |
484 | $& eq "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais", $message); | |
de946258 NC |
485 | ok("Fran\N{LATIN SMALL LETTER C}ais" =~ /Fran\Xais/ && |
486 | $& eq "Francais", $message); | |
487 | ok("Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~ /Fran\Xais/ && | |
488 | $& eq "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais", $message); | |
489 | ok("Franc\N{COMBINING CEDILLA}ais" =~ /Fran\Xais/ && | |
490 | $& eq "Franc\N{COMBINING CEDILLA}ais", $message); | |
491 | ok("Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais" =~ | |
e425a60b | 492 | /Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais/ && |
de946258 NC |
493 | $& eq "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais", $message); |
494 | ok("Franc\N{COMBINING CEDILLA}ais" =~ /Franc\N{COMBINING CEDILLA}ais/ && | |
495 | $& eq "Franc\N{COMBINING CEDILLA}ais", $message); | |
e425a60b YO |
496 | |
497 | my @f = ( | |
498 | ["Fran\N{LATIN SMALL LETTER C}ais", "Francais"], | |
499 | ["Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais", | |
500 | "Fran\N{LATIN SMALL LETTER C WITH CEDILLA}ais"], | |
501 | ["Franc\N{COMBINING CEDILLA}ais", "Franc\N{COMBINING CEDILLA}ais"], | |
502 | ); | |
503 | foreach my $entry (@f) { | |
504 | my ($subject, $match) = @$entry; | |
de946258 | 505 | ok($subject =~ /Fran(?:c\N{COMBINING CEDILLA}?| |
e425a60b | 506 | \N{LATIN SMALL LETTER C WITH CEDILLA})ais/x && |
de946258 | 507 | $& eq $match, $message); |
e425a60b YO |
508 | } |
509 | } | |
510 | ||
e425a60b | 511 | { |
de946258 | 512 | my $message = "Lingering (and useless) UTF8 flag doesn't mess up /i"; |
e425a60b YO |
513 | my $pat = "ABcde"; |
514 | my $str = "abcDE\x{100}"; | |
515 | chop $str; | |
de946258 | 516 | like($str, qr/$pat/i, $message); |
e425a60b YO |
517 | |
518 | $pat = "ABcde\x{100}"; | |
519 | $str = "abcDE"; | |
520 | chop $pat; | |
de946258 | 521 | like($str, qr/$pat/i, $message); |
e425a60b YO |
522 | |
523 | $pat = "ABcde\x{100}"; | |
524 | $str = "abcDE\x{100}"; | |
525 | chop $pat; | |
526 | chop $str; | |
de946258 | 527 | like($str, qr/$pat/i, $message); |
e425a60b YO |
528 | } |
529 | ||
e425a60b YO |
530 | { |
531 | use charnames ':full'; | |
de946258 | 532 | my $message = "LATIN SMALL LETTER SHARP S " . |
e425a60b YO |
533 | "(\N{LATIN SMALL LETTER SHARP S})"; |
534 | ||
de946258 NC |
535 | like("\N{LATIN SMALL LETTER SHARP S}", |
536 | qr/\N{LATIN SMALL LETTER SHARP S}/, $message); | |
537 | like("\N{LATIN SMALL LETTER SHARP S}", | |
945961fd KW |
538 | qr'\N{LATIN SMALL LETTER SHARP S}', $message); |
539 | like("\N{LATIN SMALL LETTER SHARP S}", | |
de946258 NC |
540 | qr/\N{LATIN SMALL LETTER SHARP S}/i, $message); |
541 | like("\N{LATIN SMALL LETTER SHARP S}", | |
945961fd KW |
542 | qr'\N{LATIN SMALL LETTER SHARP S}'i, $message); |
543 | like("\N{LATIN SMALL LETTER SHARP S}", | |
de946258 NC |
544 | qr/[\N{LATIN SMALL LETTER SHARP S}]/, $message); |
545 | like("\N{LATIN SMALL LETTER SHARP S}", | |
945961fd KW |
546 | qr'[\N{LATIN SMALL LETTER SHARP S}]', $message); |
547 | like("\N{LATIN SMALL LETTER SHARP S}", | |
de946258 | 548 | qr/[\N{LATIN SMALL LETTER SHARP S}]/i, $message); |
945961fd KW |
549 | like("\N{LATIN SMALL LETTER SHARP S}", |
550 | qr'[\N{LATIN SMALL LETTER SHARP S}]'i, $message); | |
e425a60b | 551 | |
de946258 | 552 | like("ss", qr /\N{LATIN SMALL LETTER SHARP S}/i, $message); |
945961fd | 553 | like("ss", qr '\N{LATIN SMALL LETTER SHARP S}'i, $message); |
de946258 | 554 | like("SS", qr /\N{LATIN SMALL LETTER SHARP S}/i, $message); |
945961fd | 555 | like("SS", qr '\N{LATIN SMALL LETTER SHARP S}'i, $message); |
de946258 | 556 | like("ss", qr/[\N{LATIN SMALL LETTER SHARP S}]/i, $message); |
945961fd | 557 | like("ss", qr'[\N{LATIN SMALL LETTER SHARP S}]'i, $message); |
de946258 | 558 | like("SS", qr/[\N{LATIN SMALL LETTER SHARP S}]/i, $message); |
945961fd | 559 | like("SS", qr'[\N{LATIN SMALL LETTER SHARP S}]'i, $message); |
e425a60b | 560 | |
de946258 NC |
561 | like("\N{LATIN SMALL LETTER SHARP S}", qr/ss/i, $message); |
562 | like("\N{LATIN SMALL LETTER SHARP S}", qr/SS/i, $message); | |
0f289c68 | 563 | |
de946258 NC |
564 | $message = "Unoptimized named sequence in class"; |
565 | like("ss", qr/[\N{LATIN SMALL LETTER SHARP S}x]/i, $message); | |
945961fd | 566 | like("ss", qr'[\N{LATIN SMALL LETTER SHARP S}x]'i, $message); |
de946258 | 567 | like("SS", qr/[\N{LATIN SMALL LETTER SHARP S}x]/i, $message); |
945961fd | 568 | like("SS", qr'[\N{LATIN SMALL LETTER SHARP S}x]'i, $message); |
de946258 NC |
569 | like("\N{LATIN SMALL LETTER SHARP S}", |
570 | qr/[\N{LATIN SMALL LETTER SHARP S}x]/, $message); | |
571 | like("\N{LATIN SMALL LETTER SHARP S}", | |
945961fd KW |
572 | qr'[\N{LATIN SMALL LETTER SHARP S}x]', $message); |
573 | like("\N{LATIN SMALL LETTER SHARP S}", | |
de946258 | 574 | qr/[\N{LATIN SMALL LETTER SHARP S}x]/i, $message); |
945961fd KW |
575 | like("\N{LATIN SMALL LETTER SHARP S}", |
576 | qr'[\N{LATIN SMALL LETTER SHARP S}x]'i, $message); | |
e425a60b YO |
577 | } |
578 | ||
e425a60b YO |
579 | { |
580 | # More whitespace: U+0085, U+2028, U+2029\n"; | |
581 | ||
582 | # U+0085, U+00A0 need to be forced to be Unicode, the \x{100} does that. | |
b7ad6c3e KW |
583 | like "<\x{100}" . uni_to_native("\x{0085}") . ">", qr/<\x{100}\s>/, '\x{0085} in \s'; |
584 | like "<" . uni_to_native("\x{0085}") . ">", qr/<\v>/, '\x{0085} in \v'; | |
585 | like "<\x{100}" . uni_to_native("\x{00A0}") . ">", qr/<\x{100}\s>/, '\x{00A0} in \s'; | |
586 | like "<" . uni_to_native("\x{00A0}") . ">", qr/<\h>/, '\x{00A0} in \h'; | |
a9c9e371 | 587 | my @h = map {sprintf "%05x" => $_} 0x01680, 0x02000 .. 0x0200A, |
e425a60b YO |
588 | 0x0202F, 0x0205F, 0x03000; |
589 | my @v = map {sprintf "%05x" => $_} 0x02028, 0x02029; | |
590 | ||
591 | my @H = map {sprintf "%05x" => $_} 0x01361, 0x0200B, 0x02408, 0x02420, | |
a9c9e371 | 592 | 0x0303F, 0xE0020, 0x180E; |
e425a60b | 593 | my @V = map {sprintf "%05x" => $_} 0x0008A .. 0x0008D, 0x00348, 0x10100, |
a9c9e371 | 594 | 0xE005F, 0xE007C, 0x180E; |
e425a60b YO |
595 | |
596 | for my $hex (@h) { | |
597 | my $str = eval qq ["<\\x{$hex}>"]; | |
b7ad6c3e KW |
598 | like $str, qr/<\s>/, "\\x{$hex} in \\s"; |
599 | like $str, qr/<\h>/, "\\x{$hex} in \\h"; | |
600 | unlike $str, qr/<\v>/, "\\x{$hex} not in \\v"; | |
e425a60b YO |
601 | } |
602 | ||
603 | for my $hex (@v) { | |
604 | my $str = eval qq ["<\\x{$hex}>"]; | |
b7ad6c3e KW |
605 | like $str, qr/<\s>/, "\\x{$hex} in \\s"; |
606 | like $str, qr/<\v>/, "\\x{$hex} in \\v"; | |
607 | unlike $str, qr/<\h>/, "\\x{$hex} not in \\h"; | |
e425a60b YO |
608 | } |
609 | ||
610 | for my $hex (@H) { | |
611 | my $str = eval qq ["<\\x{$hex}>"]; | |
b7ad6c3e KW |
612 | like $str, qr/<\S>/, "\\x{$hex} in \\S"; |
613 | like $str, qr/<\H>/, "\\x{$hex} in \\H"; | |
e425a60b YO |
614 | } |
615 | ||
616 | for my $hex (@V) { | |
617 | my $str = eval qq ["<\\x{$hex}>"]; | |
b7ad6c3e KW |
618 | like $str, qr/<\S>/, "\\x{$hex} in \\S"; |
619 | like $str, qr/<\V>/, "\\x{$hex} in \\V"; | |
e425a60b YO |
620 | } |
621 | } | |
622 | ||
e425a60b YO |
623 | { |
624 | # . with /s should work on characters, as opposed to bytes | |
de946258 | 625 | my $message = ". with /s works on characters, not bytes"; |
e425a60b YO |
626 | |
627 | my $s = "\x{e4}\x{100}"; | |
628 | # This is not expected to match: the point is that | |
629 | # neither should we get "Malformed UTF-8" warnings. | |
c11a8df3 NC |
630 | warning_is(sub {$s =~ /\G(.+?)\n/gcs}, undef, |
631 | "No 'Malformed UTF-8' warning"); | |
e425a60b YO |
632 | |
633 | my @c; | |
634 | push @c => $1 while $s =~ /\G(.)/gs; | |
635 | ||
636 | local $" = ""; | |
de946258 | 637 | is("@c", $s, $message); |
e425a60b YO |
638 | |
639 | # Test only chars < 256 | |
640 | my $t1 = "Q003\n\n\x{e4}\x{f6}\n\nQ004\n\n\x{e7}"; | |
641 | my $r1 = ""; | |
642 | while ($t1 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) { | |
0f289c68 | 643 | $r1 .= $1 . $2; |
e425a60b YO |
644 | } |
645 | ||
646 | my $t2 = $t1 . "\x{100}"; # Repeat with a larger char | |
647 | my $r2 = ""; | |
648 | while ($t2 =~ / \G ( .+? ) \n\s+ ( .+? ) ( $ | \n\s+ ) /xgcs) { | |
0f289c68 | 649 | $r2 .= $1 . $2; |
e425a60b YO |
650 | } |
651 | $r2 =~ s/\x{100}//; | |
652 | ||
de946258 | 653 | is($r1, $r2, $message); |
e425a60b YO |
654 | } |
655 | ||
e425a60b | 656 | { |
de946258 | 657 | my $message = "Unicode lookbehind"; |
b7ad6c3e | 658 | like("A\x{100}B", qr/(?<=A.)B/, $message); |
de946258 | 659 | like("A\x{200}\x{300}B", qr/(?<=A..)B/, $message); |
b7ad6c3e KW |
660 | like("\x{400}AB", qr/(?<=\x{400}.)B/, $message); |
661 | like("\x{500}\x{600}B", qr/(?<=\x{500}.)B/, $message); | |
e425a60b YO |
662 | |
663 | # Original code also contained: | |
664 | # ok "\x{500\x{600}}B" =~ /(?<=\x{500}.)B/; | |
665 | # but that looks like a typo. | |
666 | } | |
667 | ||
e425a60b | 668 | { |
de946258 | 669 | my $message = 'UTF-8 hash keys and /$/'; |
e425a60b YO |
670 | # http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters |
671 | # /2002-01/msg01327.html | |
672 | ||
673 | my $u = "a\x{100}"; | |
674 | my $v = substr ($u, 0, 1); | |
675 | my $w = substr ($u, 1, 1); | |
676 | my %u = ($u => $u, $v => $v, $w => $w); | |
677 | for (keys %u) { | |
678 | my $m1 = /^\w*$/ ? 1 : 0; | |
679 | my $m2 = $u {$_} =~ /^\w*$/ ? 1 : 0; | |
de946258 | 680 | is($m1, $m2, $message); |
e425a60b YO |
681 | } |
682 | } | |
683 | ||
e425a60b | 684 | { |
de946258 | 685 | my $message = "No SEGV in s/// and UTF-8"; |
e425a60b | 686 | my $s = "s#\x{100}" x 4; |
de946258 | 687 | ok($s =~ s/[^\w]/ /g, $message); |
2e84be61 | 688 | if ( 1 or $ENV{PERL_TEST_LEGACY_POSIX_CC} ) { |
de946258 | 689 | is($s, "s \x{100}" x 4, $message); |
e425a60b YO |
690 | } |
691 | else { | |
de946258 | 692 | is($s, "s " x 4, $message); |
e425a60b YO |
693 | } |
694 | } | |
695 | ||
e425a60b | 696 | { |
de946258 | 697 | my $message = "UTF-8 bug (maybe already known?)"; |
e425a60b YO |
698 | my $u = "foo"; |
699 | $u =~ s/./\x{100}/g; | |
de946258 | 700 | is($u, "\x{100}\x{100}\x{100}", $message); |
e425a60b YO |
701 | |
702 | $u = "foobar"; | |
703 | $u =~ s/[ao]/\x{100}/g; | |
de946258 | 704 | is($u, "f\x{100}\x{100}b\x{100}r", $message); |
e425a60b YO |
705 | |
706 | $u =~ s/\x{100}/e/g; | |
de946258 | 707 | is($u, "feeber", $message); |
e425a60b YO |
708 | } |
709 | ||
e425a60b | 710 | { |
de946258 | 711 | my $message = "UTF-8 bug with s///"; |
e425a60b YO |
712 | # check utf8/non-utf8 mixtures |
713 | # try to force all float/anchored check combinations | |
714 | ||
715 | my $c = "\x{100}"; | |
716 | my $subst; | |
717 | for my $re ("xx.*$c", "x.*$c$c", "$c.*xx", "$c$c.*x", | |
718 | "xx.*(?=$c)", "(?=$c).*xx",) { | |
de946258 NC |
719 | unlike("xxx", qr/$re/, $message); |
720 | ok(+($subst = "xxx") !~ s/$re//, $message); | |
e425a60b YO |
721 | } |
722 | for my $re ("xx.*$c*", "$c*.*xx") { | |
de946258 NC |
723 | like("xxx", qr/$re/, $message); |
724 | ok(+($subst = "xxx") =~ s/$re//, $message); | |
725 | is($subst, "", $message); | |
e425a60b YO |
726 | } |
727 | for my $re ("xxy*", "y*xx") { | |
de946258 NC |
728 | like("xx$c", qr/$re/, $message); |
729 | ok(+($subst = "xx$c") =~ s/$re//, $message); | |
730 | is($subst, $c, $message); | |
731 | unlike("xy$c", qr/$re/, $message); | |
732 | ok(+($subst = "xy$c") !~ s/$re//, $message); | |
e425a60b YO |
733 | } |
734 | for my $re ("xy$c*z", "x$c*yz") { | |
de946258 NC |
735 | like("xyz", qr/$re/, $message); |
736 | ok(+($subst = "xyz") =~ s/$re//, $message); | |
737 | is($subst, "", $message); | |
e425a60b YO |
738 | } |
739 | } | |
740 | ||
e425a60b | 741 | { |
c72077c4 AC |
742 | # The second half of RT #114808 |
743 | warning_is(sub {'aa' =~ /.+\x{100}/}, undef, | |
744 | 'utf8-only floating substr, non-utf8 target, no warning'); | |
745 | } | |
746 | ||
747 | { | |
de946258 | 748 | my $message = "qr /.../x"; |
e425a60b | 749 | my $R = qr / A B C # D E/x; |
de946258 NC |
750 | ok("ABCDE" =~ $R && $& eq "ABC", $message); |
751 | ok("ABCDE" =~ /$R/ && $& eq "ABC", $message); | |
752 | ok("ABCDE" =~ m/$R/ && $& eq "ABC", $message); | |
753 | ok("ABCDE" =~ /($R)/ && $1 eq "ABC", $message); | |
754 | ok("ABCDE" =~ m/($R)/ && $1 eq "ABC", $message); | |
e425a60b YO |
755 | } |
756 | ||
e425a60b YO |
757 | { |
758 | local $\; | |
759 | $_ = 'aaaaaaaaaa'; | |
760 | utf8::upgrade($_); chop $_; $\="\n"; | |
761 | ok /[^\s]+/, 'm/[^\s]/ utf8'; | |
762 | ok /[^\d]+/, 'm/[^\d]/ utf8'; | |
763 | ok +($a = $_, $_ =~ s/[^\s]+/./g), 's/[^\s]/ utf8'; | |
764 | ok +($a = $_, $a =~ s/[^\d]+/./g), 's/[^\s]/ utf8'; | |
765 | } | |
766 | ||
e425a60b YO |
767 | { |
768 | # Subject: Odd regexp behavior | |
769 | # From: Markus Kuhn <Markus.Kuhn@cl.cam.ac.uk> | |
770 | # Date: Wed, 26 Feb 2003 16:53:12 +0000 | |
771 | # Message-Id: <E18o4nw-0008Ly-00@wisbech.cl.cam.ac.uk> | |
772 | # To: perl-unicode@perl.org | |
773 | ||
de946258 | 774 | my $message = 'Markus Kuhn 2003-02-26'; |
0f289c68 | 775 | |
e425a60b | 776 | my $x = "\x{2019}\nk"; |
de946258 NC |
777 | ok($x =~ s/(\S)\n(\S)/$1 $2/sg, $message); |
778 | is($x, "\x{2019} k", $message); | |
e425a60b YO |
779 | |
780 | $x = "b\nk"; | |
de946258 NC |
781 | ok($x =~ s/(\S)\n(\S)/$1 $2/sg, $message); |
782 | is($x, "b k", $message); | |
e425a60b | 783 | |
de946258 | 784 | like("\x{2019}", qr/\S/, $message); |
e425a60b YO |
785 | } |
786 | ||
e425a60b | 787 | { |
b7ad6c3e | 788 | like "\x{100}\n", qr/\x{100}\n$/, "UTF-8 length cache and fbm_compile"; |
e425a60b YO |
789 | } |
790 | ||
e425a60b YO |
791 | { |
792 | package Str; | |
793 | use overload q /""/ => sub {${$_ [0]};}; | |
794 | sub new {my ($c, $v) = @_; bless \$v, $c;} | |
795 | ||
796 | package main; | |
797 | $_ = Str -> new ("a\x{100}/\x{100}b"); | |
798 | ok join (":", /\b(.)\x{100}/g) eq "a:/", "re_intuit_start and PL_bostr"; | |
799 | } | |
800 | ||
e425a60b YO |
801 | { |
802 | my $re = qq /^([^X]*)X/; | |
803 | utf8::upgrade ($re); | |
8f46dbea | 804 | like "\x{100}X", qr/$re/, "S_cl_and ANYOF_UNICODE & ANYOF_INVERTED"; |
0ab1ff8e KW |
805 | my $loc_re = qq /(?l:^([^X]*)X)/; |
806 | utf8::upgrade ($loc_re); | |
613abc6d | 807 | no warnings 'locale'; |
8f46dbea | 808 | like "\x{100}X", qr/$loc_re/, "locale, S_cl_and ANYOF_UNICODE & ANYOF_INVERTED"; |
e425a60b YO |
809 | } |
810 | ||
e425a60b | 811 | { |
b7ad6c3e | 812 | like "123\x{100}", qr/^.*1.*23\x{100}$/, |
e425a60b YO |
813 | 'UTF-8 + multiple floating substr'; |
814 | } | |
815 | ||
e425a60b | 816 | { |
de946258 | 817 | my $message = '<20030808193656.5109.1@llama.ni-s.u-net.com>'; |
e425a60b YO |
818 | |
819 | # LATIN SMALL/CAPITAL LETTER A WITH MACRON | |
de946258 | 820 | like(" \x{101}", qr/\x{100}/i, $message); |
e425a60b YO |
821 | |
822 | # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW | |
de946258 | 823 | like(" \x{1E01}", qr/\x{1E00}/i, $message); |
e425a60b YO |
824 | |
825 | # DESERET SMALL/CAPITAL LETTER LONG I | |
de946258 | 826 | like(" \x{10428}", qr/\x{10400}/i, $message); |
e425a60b YO |
827 | |
828 | # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X' | |
de946258 | 829 | like(" \x{1E01}x", qr/\x{1E00}X/i, $message); |
e425a60b YO |
830 | } |
831 | ||
e425a60b | 832 | { |
26faadbd | 833 | for (120 .. 130, 240 .. 260) { |
e425a60b | 834 | my $head = 'x' x $_; |
de946258 | 835 | my $message = q [Don't misparse \x{...} in regexp ] . |
26faadbd | 836 | q [near EXACT char count limit]; |
e425a60b | 837 | for my $tail ('\x{0061}', '\x{1234}', '\x61') { |
14358a41 NC |
838 | eval qq{like("$head$tail", qr/$head$tail/, \$message)}; |
839 | is($@, '', $message); | |
e425a60b | 840 | } |
de946258 | 841 | $message = q [Don't misparse \N{...} in regexp ] . |
26faadbd | 842 | q [near EXACT char count limit]; |
e425a60b | 843 | for my $tail ('\N{SNOWFLAKE}') { |
14358a41 NC |
844 | eval qq {use charnames ':full'; |
845 | like("$head$tail", qr/$head$tail/, \$message)}; | |
945961fd KW |
846 | eval qq {use charnames ':full'; |
847 | like("$head$tail", qr'$head$tail', \$message)}; | |
14358a41 | 848 | is($@, '', $message); |
e425a60b YO |
849 | } |
850 | } | |
851 | } | |
852 | ||
e425a60b YO |
853 | { # TRIE related |
854 | our @got = (); | |
855 | "words" =~ /(word|word|word)(?{push @got, $1})s$/; | |
de26e0cc | 856 | is(@got, 1, "TRIE optimisation"); |
e425a60b YO |
857 | |
858 | @got = (); | |
859 | "words" =~ /(word|word|word)(?{push @got,$1})s$/i; | |
de26e0cc | 860 | is(@got, 1,"TRIEF optimisation"); |
e425a60b YO |
861 | |
862 | my @nums = map {int rand 1000} 1 .. 100; | |
863 | my $re = "(" . (join "|", @nums) . ")"; | |
864 | $re = qr/\b$re\b/; | |
865 | ||
866 | foreach (@nums) { | |
b7ad6c3e | 867 | like $_, qr/$re/, "Trie nums"; |
e425a60b YO |
868 | } |
869 | ||
870 | $_ = join " ", @nums; | |
871 | @got = (); | |
872 | push @got, $1 while /$re/g; | |
873 | ||
874 | my %count; | |
875 | $count {$_} ++ for @got; | |
876 | my $ok = 1; | |
877 | for (@nums) { | |
878 | $ok = 0 if --$count {$_} < 0; | |
879 | } | |
880 | ok $ok, "Trie min count matches"; | |
881 | } | |
882 | ||
e425a60b YO |
883 | { |
884 | # TRIE related | |
885 | # LATIN SMALL/CAPITAL LETTER A WITH MACRON | |
886 | ok "foba \x{101}foo" =~ qr/(foo|\x{100}foo|bar)/i && | |
887 | $1 eq "\x{101}foo", | |
888 | "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH MACRON"; | |
889 | ||
890 | # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW | |
891 | ok "foba \x{1E01}foo" =~ qr/(foo|\x{1E00}foo|bar)/i && | |
892 | $1 eq "\x{1E01}foo", | |
893 | "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH RING BELOW"; | |
894 | ||
895 | # DESERET SMALL/CAPITAL LETTER LONG I | |
896 | ok "foba \x{10428}foo" =~ qr/(foo|\x{10400}foo|bar)/i && | |
897 | $1 eq "\x{10428}foo", | |
898 | "TRIEF + DESERET SMALL/CAPITAL LETTER LONG I"; | |
899 | ||
900 | # LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X' | |
901 | ok "foba \x{1E01}xfoo" =~ qr/(foo|\x{1E00}Xfoo|bar)/i && | |
902 | $1 eq "\x{1E01}xfoo", | |
903 | "TRIEF + LATIN SMALL/CAPITAL LETTER A WITH RING BELOW + 'X'"; | |
904 | ||
905 | use charnames ':full'; | |
906 | ||
907 | my $s = "\N{LATIN SMALL LETTER SHARP S}"; | |
908 | ok "foba ba$s" =~ qr/(foo|Ba$s|bar)/i && $1 eq "ba$s", | |
909 | "TRIEF + LATIN SMALL LETTER SHARP S =~ ss"; | |
910 | ok "foba ba$s" =~ qr/(Ba$s|foo|bar)/i && $1 eq "ba$s", | |
911 | "TRIEF + LATIN SMALL LETTER SHARP S =~ ss"; | |
912 | ok "foba ba$s" =~ qr/(foo|bar|Ba$s)/i && $1 eq "ba$s", | |
913 | "TRIEF + LATIN SMALL LETTER SHARP S =~ ss"; | |
914 | ||
915 | ok "foba ba$s" =~ qr/(foo|Bass|bar)/i && $1 eq "ba$s", | |
916 | "TRIEF + LATIN SMALL LETTER SHARP S =~ ss"; | |
917 | ||
918 | ok "foba ba$s" =~ qr/(foo|BaSS|bar)/i && $1 eq "ba$s", | |
919 | "TRIEF + LATIN SMALL LETTER SHARP S =~ SS"; | |
920 | ||
921 | ok "foba ba${s}pxySS$s$s" =~ qr/(b(?:a${s}t|a${s}f|a${s}p)[xy]+$s*)/i | |
922 | && $1 eq "ba${s}pxySS$s$s", | |
923 | "COMMON PREFIX TRIEF + LATIN SMALL LETTER SHARP S"; | |
924 | } | |
925 | ||
e425a60b | 926 | { |
3fe9640a JH |
927 | BEGIN { |
928 | unshift @INC, 'lib'; | |
929 | } | |
b99b8a84 KW |
930 | use Cname; # Our custom charname plugin, currently found in |
931 | # t/lib/Cname.pm | |
0f289c68 | 932 | |
b7ad6c3e | 933 | like 'fooB', qr/\N{foo}[\N{B}\N{b}]/, "Passthrough charname"; |
94ca1619 KW |
934 | my $name = "foo\xDF"; |
935 | my $result = eval "'A${name}B' =~ /^A\\N{$name}B\$/"; | |
936 | ok !$@ && $result, "Passthrough charname of non-ASCII, Latin1"; | |
673c254b KW |
937 | eval "qr/\\p{name=foo}/"; |
938 | like($@, qr/Can't find Unicode property definition "name=foo"/, | |
939 | '\p{name=} doesn\'t see a cumstom charnames translator'); | |
e425a60b YO |
940 | # |
941 | # Why doesn't must_warn work here? | |
942 | # | |
943 | my $w; | |
944 | local $SIG {__WARN__} = sub {$w .= "@_"}; | |
a78e2a97 | 945 | $result = eval 'q(WARN) =~ /^[\N{WARN}]$/'; |
8f0cd35a | 946 | ok !$@ && $result && ! $w, '\N{} returning multi-char works'; |
e425a60b YO |
947 | |
948 | undef $w; | |
b7ad6c3e | 949 | eval q [unlike "\0", qr/[\N{EMPTY-STR}XY]/, |
5895685f | 950 | "Zerolength charname in charclass doesn't match \\\\0"]; |
ff3f963a KW |
951 | ok $w && $w =~ /Ignoring zero length/, |
952 | 'Ignoring zero length \N{} in character class warning'; | |
67048c1b | 953 | undef $w; |
b7ad6c3e | 954 | eval q [like 'xy', qr/x[\N{EMPTY-STR} y]/x, |
67048c1b KW |
955 | 'Empty string charname in [] is ignored; finds a following character']; |
956 | ok $w && $w =~ /Ignoring zero length/, | |
957 | 'Ignoring zero length \N{} in character class warning'; | |
958 | undef $w; | |
b7ad6c3e | 959 | eval q [like 'x ', qr/x[\N{EMPTY-STR} y]/, |
67048c1b | 960 | 'Empty string charname in [] is ignored; finds a following blank under /x']; |
b7ad6c3e | 961 | like $w, qr/Ignoring zero length/, |
67048c1b | 962 | 'Ignoring zero length \N{} in character class warning'; |
e425a60b | 963 | |
b99b8a84 KW |
964 | # EVIL keeps track of its calls, and appends a new character each |
965 | # time: A AB ABC ABCD ... | |
e425a60b | 966 | ok 'AB' =~ /(\N{EVIL})/ && $1 eq 'A', 'Charname caching $1'; |
b7ad6c3e | 967 | like 'ABC', qr/(\N{EVIL})/, 'Charname caching $1'; |
945961fd KW |
968 | ok 'ABCD' =~ m'(\N{EVIL})' && $1 eq 'ABC', 'Charname caching $1'; |
969 | ok 'ABCDE' =~ m'(\N{EVIL})', 'Charname caching $1'; | |
b7ad6c3e | 970 | like 'xy', qr/x\N{EMPTY-STR}y/, |
e425a60b | 971 | 'Empty string charname produces NOTHING node'; |
945961fd KW |
972 | ok 'xy' =~ 'x\N{EMPTY-STR}y', |
973 | 'Empty string charname produces NOTHING node'; | |
b7ad6c3e | 974 | like '', qr/\N{EMPTY-STR}/, |
e425a60b | 975 | 'Empty string charname produces NOTHING node'; |
b7ad6c3e KW |
976 | like "\N{LONG-STR}", qr/^\N{LONG-STR}$/, 'Verify that long string works'; |
977 | like "\N{LONG-STR}", qr/^\N{LONG-STR}$/i, 'Verify under folding that long string works'; | |
1e629c22 KW |
978 | |
979 | # perlhacktips points out that these work on both ASCII and EBCDIC | |
980 | like "\xfc", qr/\N{EMPTY-STR}\xdc/i, 'Empty \N{} should change /d to /u'; | |
945961fd | 981 | like "\xfc", qr'\N{EMPTY-STR}\xdc'i, 'Empty \N{} should change /d to /u'; |
ff3f963a | 982 | |
902994e4 | 983 | eval '/(?[[\N{EMPTY-STR}]])/'; |
b7ad6c3e | 984 | like $@, qr/Zero length \\N\{\}/, 'Verify zero-length return from \N{} correctly fails'; |
945961fd KW |
985 | ok "\N{LONG-STR}" =~ /^\N{LONG-STR}$/, 'Verify that long string works'; |
986 | ok "\N{LONG-STR}" =~ '^\N{LONG-STR}$', 'Verify that long string works'; | |
987 | ok "\N{LONG-STR}" =~ /^\N{LONG-STR}$/i, 'Verify under folding that long string works'; | |
988 | ok "\N{LONG-STR}" =~ m'^\N{LONG-STR}$'i, 'Verify under folding that long string works'; | |
902994e4 | 989 | |
bd299e29 | 990 | undef $w; |
bd299e29 | 991 | { |
3036c853 | 992 | () = eval q ["\N{TOO MANY SPACES}"]; |
2d8eb851 | 993 | like ($@, qr/charnames alias definitions may not contain a sequence of multiple spaces/, "Multiple spaces in a row in a charnames alias is fatal"); |
3036c853 | 994 | eval q [use utf8; () = "\N{TOO MANY SPACES}"]; |
2d8eb851 | 995 | like ($@, qr/charnames alias definitions may not contain a sequence of multiple spaces/, "... same under utf8"); |
94ec3a20 | 996 | } |
bd299e29 KW |
997 | |
998 | undef $w; | |
7fc82458 | 999 | my $Cedilla_Latin1 = "GAR" |
d1cef54a | 1000 | . uni_to_native("\xC7") |
7fc82458 KW |
1001 | . "ON"; |
1002 | my $Cedilla_utf8 = $Cedilla_Latin1; | |
1003 | utf8::upgrade($Cedilla_utf8); | |
1004 | eval qq[is("\\N{$Cedilla_Latin1}", "$Cedilla_Latin1", "A cedilla in character name works")]; | |
1005 | undef $w; | |
1006 | { | |
1007 | use feature 'unicode_eval'; | |
1008 | eval qq[use utf8; is("\\N{$Cedilla_utf8}", "$Cedilla_utf8", "... same under 'use utf8': they work")]; | |
1009 | } | |
1010 | ||
1011 | undef $w; | |
df758df2 | 1012 | my $NBSP_Latin1 = "NBSP" |
d1cef54a | 1013 | . uni_to_native("\xA0") |
df758df2 | 1014 | . "SEPARATED" |
d1cef54a | 1015 | . uni_to_native("\xA0") |
df758df2 KW |
1016 | . "SPACE"; |
1017 | my $NBSP_utf8 = $NBSP_Latin1; | |
1018 | utf8::upgrade($NBSP_utf8); | |
61463a3e KW |
1019 | () = eval qq[is("\\N{$NBSP_Latin1}", "$NBSP_Latin1"]; |
1020 | like ($@, qr/Invalid character in \\N\{...}/, "A NO-BREAK SPACE in a charnames alias is fatal"); | |
df758df2 KW |
1021 | undef $w; |
1022 | { | |
1023 | use feature 'unicode_eval'; | |
61463a3e KW |
1024 | eval qq[use utf8; is("\\N{$NBSP_utf8}"]; |
1025 | like ($@, qr/Invalid character in \\N\{...}/, "A NO-BREAK SPACE in a charnames alias is fatal"); | |
df758df2 KW |
1026 | } |
1027 | ||
6a642c21 FC |
1028 | { |
1029 | BEGIN { no strict; *CnameTest:: = *{"_charnames\0A::" } } | |
1030 | package CnameTest { sub translator { pop } } | |
1031 | BEGIN { $^H{charnames} = \&CnameTest::translator } | |
1032 | undef $w; | |
1033 | () = eval q ["\N{TOO MANY SPACES}"]; | |
2d8eb851 | 1034 | like ($@, qr/charnames alias definitions may not contain a sequence of multiple spaces/, |
6a642c21 FC |
1035 | 'translators in _charnames\0* packages get validated'); |
1036 | } | |
1037 | ||
ff3f963a KW |
1038 | # If remove the limitation in regcomp code these should work |
1039 | # differently | |
1040 | undef $w; | |
b7ad6c3e | 1041 | eval q [like "\N{TOO-LONG-STR}" =~ /^\N{TOO-LONG-STR}$/, 'Verify that what once was too long a string works']; |
cb233ae3 | 1042 | eval 'q() =~ /\N{4F}/'; |
4d7cd482 | 1043 | ok $@ && $@ =~ /Invalid character/, 'Verify that leading digit in name gives error'; |
cb233ae3 | 1044 | eval 'q() =~ /\N{COM,MA}/'; |
4d7cd482 | 1045 | ok $@ && $@ =~ /Invalid character/, 'Verify that comma in name gives error'; |
f36cc39d | 1046 | $name = "A" . uni_to_native("\x{D7}") . "O"; |
cb233ae3 | 1047 | eval "q(W) =~ /\\N{$name}/"; |
4d7cd482 | 1048 | ok $@ && $@ =~ /Invalid character/, 'Verify that latin1 symbol in name gives error'; |
bde9e88d KW |
1049 | my $utf8_name = "7 CITIES OF GOLD"; |
1050 | utf8::upgrade($utf8_name); | |
1051 | eval "use utf8; q(W) =~ /\\N{$utf8_name}/"; | |
1052 | ok $@ && $@ =~ /Invalid character/, 'Verify that leading digit in utf8 name gives error'; | |
1053 | $utf8_name = "SHARP #"; | |
1054 | utf8::upgrade($utf8_name); | |
1055 | eval "use utf8; q(W) =~ /\\N{$utf8_name}/"; | |
1056 | ok $@ && $@ =~ /Invalid character/, 'Verify that ASCII symbol in utf8 name gives error'; | |
f36cc39d | 1057 | $utf8_name = "A HOUSE " . uni_to_native("\xF7") . " AGAINST ITSELF"; |
bde9e88d KW |
1058 | utf8::upgrade($utf8_name); |
1059 | eval "use utf8; q(W) =~ /\\N{$utf8_name}/"; | |
1060 | ok $@ && $@ =~ /Invalid character/, 'Verify that latin1 symbol in utf8 name gives error'; | |
1061 | $utf8_name = "\x{664} HORSEMEN}"; | |
1062 | eval "use utf8; q(W) =~ /\\N{$utf8_name}/"; | |
1063 | ok $@ && $@ =~ /Invalid character/, 'Verify that leading above Latin1 digit in utf8 name gives error'; | |
1064 | $utf8_name = "A \x{1F4A9} WOULD SMELL AS SWEET}"; | |
1065 | eval "use utf8; q(W) =~ /\\N{$utf8_name}/"; | |
1066 | ok $@ && $@ =~ /Invalid character/, 'Verify that above Latin1 symbol in utf8 name gives error'; | |
1067 | ||
cb233ae3 | 1068 | undef $w; |
f36cc39d | 1069 | $name = "A" . uni_to_native("\x{D1}") . "O"; |
cb233ae3 KW |
1070 | eval "q(W) =~ /\\N{$name}/"; |
1071 | ok ! $w, 'Verify that latin1 letter in name doesnt give warning'; | |
0f289c68 | 1072 | |
ba7b73c5 NC |
1073 | # This tests the code path that restarts the parse when the recursive |
1074 | # call to S_reg() from within S_grok_bslash_N() discovers that the | |
1075 | # pattern needs to be recalculated as UTF-8. use eval to avoid | |
1076 | # needing literal Unicode in this source file: | |
1077 | my $r = eval "qr/\\N{\x{100}\x{100}}/"; | |
1078 | isnt $r, undef, "Generated regex for multi-char UTF-8 charname" | |
1079 | or diag($@); | |
b7ad6c3e | 1080 | like "\x{100}\x{100}", $r, "which matches"; |
e425a60b YO |
1081 | } |
1082 | ||
e425a60b YO |
1083 | { |
1084 | use charnames ':full'; | |
1085 | ||
b7ad6c3e KW |
1086 | unlike 'aabc', qr/a\N{PLUS SIGN}b/, '/a\N{PLUS SIGN}b/ against aabc'; |
1087 | like 'a+bc', qr/a\N{PLUS SIGN}b/, '/a\N{PLUS SIGN}b/ against a+bc'; | |
e425a60b | 1088 | |
b7ad6c3e | 1089 | like ' A B', qr/\N{SPACE}\N{U+0041}\N{SPACE}\N{U+0042}/, |
e425a60b | 1090 | 'Intermixed named and unicode escapes'; |
b7ad6c3e KW |
1091 | like "\N{SPACE}\N{U+0041}\N{SPACE}\N{U+0042}", |
1092 | qr/\N{SPACE}\N{U+0041}\N{SPACE}\N{U+0042}/, | |
e425a60b | 1093 | 'Intermixed named and unicode escapes'; |
b7ad6c3e KW |
1094 | like "\N{SPACE}\N{U+0041}\N{SPACE}\N{U+0042}", |
1095 | qr/[\N{SPACE}\N{U+0041}][\N{SPACE}\N{U+0042}]/, | |
0f289c68 | 1096 | 'Intermixed named and unicode escapes'; |
b7ad6c3e | 1097 | like "\0", qr/^\N{NULL}$/, 'Verify that \N{NULL} works; is not confused with an error'; |
e425a60b YO |
1098 | } |
1099 | ||
e425a60b YO |
1100 | { |
1101 | our $brackets; | |
1102 | $brackets = qr{ | |
1103 | { (?> [^{}]+ | (??{ $brackets }) )* } | |
1104 | }x; | |
1105 | ||
b7ad6c3e | 1106 | unlike "{b{c}d", qr/^((??{ $brackets }))/, "Bracket mismatch"; |
e425a60b YO |
1107 | |
1108 | SKIP: { | |
1109 | our @stack = (); | |
1110 | my @expect = qw( | |
1111 | stuff1 | |
1112 | stuff2 | |
1113 | <stuff1>and<stuff2> | |
1114 | right | |
1115 | <right> | |
1116 | <<right>> | |
1117 | <<<right>>> | |
1118 | <<stuff1>and<stuff2>><<<<right>>>> | |
1119 | ); | |
1120 | ||
1121 | local $_ = '<<<stuff1>and<stuff2>><<<<right>>>>>'; | |
1122 | ok /^(<((?:(?>[^<>]+)|(?1))*)>(?{push @stack, $2 }))$/, | |
1123 | "Recursion matches"; | |
de26e0cc | 1124 | is(@stack, @expect, "Right amount of matches") |
e425a60b YO |
1125 | or skip "Won't test individual results as count isn't equal", |
1126 | 0 + @expect; | |
1127 | my $idx = 0; | |
1128 | foreach my $expect (@expect) { | |
de26e0cc NC |
1129 | is($stack [$idx], $expect, |
1130 | "Expecting '$expect' at stack pos #$idx"); | |
e425a60b YO |
1131 | $idx ++; |
1132 | } | |
1133 | } | |
1134 | } | |
1135 | ||
e425a60b YO |
1136 | { |
1137 | my $s = '123453456'; | |
1138 | $s =~ s/(?<digits>\d+)\k<digits>/$+{digits}/; | |
1139 | ok $s eq '123456', 'Named capture (angle brackets) s///'; | |
1140 | $s = '123453456'; | |
1141 | $s =~ s/(?'digits'\d+)\k'digits'/$+{digits}/; | |
0f289c68 | 1142 | ok $s eq '123456', 'Named capture (single quotes) s///'; |
e425a60b YO |
1143 | } |
1144 | ||
e425a60b YO |
1145 | { |
1146 | my @ary = ( | |
726f0605 KW |
1147 | pack('U', utf8::unicode_to_native(0x00F1)), # n-tilde |
1148 | '_'.pack('U', utf8::unicode_to_native(0x00F1)), # _ + n-tilde | |
e425a60b | 1149 | 'c'.pack('U', 0x0327), # c + cedilla |
726f0605 | 1150 | pack('U*', utf8::unicode_to_native(0x00F1), 0x0327),# n-tilde + cedilla |
e425a60b YO |
1151 | pack('U', 0x0391), # ALPHA |
1152 | pack('U', 0x0391).'2', # ALPHA + 2 | |
1153 | pack('U', 0x0391).'_', # ALPHA + _ | |
1154 | ); | |
1155 | ||
1156 | for my $uni (@ary) { | |
1157 | my ($r1, $c1, $r2, $c2) = eval qq { | |
1158 | use utf8; | |
1159 | scalar ("..foo foo.." =~ /(?'${uni}'foo) \\k'${uni}'/), | |
1160 | \$+{${uni}}, | |
1161 | scalar ("..bar bar.." =~ /(?<${uni}>bar) \\k<${uni}>/), | |
1162 | \$+{${uni}}; | |
1163 | }; | |
1164 | ok $r1, "Named capture UTF (?'')"; | |
1165 | ok defined $c1 && $c1 eq 'foo', "Named capture UTF \%+"; | |
1166 | ok $r2, "Named capture UTF (?<>)"; | |
1167 | ok defined $c2 && $c2 eq 'bar', "Named capture UTF \%+"; | |
1168 | } | |
1169 | } | |
1170 | ||
e425a60b YO |
1171 | { |
1172 | my $s = 'foo bar baz'; | |
1173 | my @res; | |
1174 | if ('1234' =~ /(?<A>1)(?<B>2)(?<A>3)(?<B>4)/) { | |
1175 | foreach my $name (sort keys(%-)) { | |
1176 | my $ary = $- {$name}; | |
1177 | foreach my $idx (0 .. $#$ary) { | |
1178 | push @res, "$name:$idx:$ary->[$idx]"; | |
1179 | } | |
1180 | } | |
1181 | } | |
1182 | my @expect = qw (A:0:1 A:1:3 B:0:2 B:1:4); | |
de26e0cc | 1183 | is("@res", "@expect", "Check %-"); |
e425a60b YO |
1184 | eval' |
1185 | no warnings "uninitialized"; | |
1186 | print for $- {this_key_doesnt_exist}; | |
1187 | '; | |
1188 | ok !$@,'lvalue $- {...} should not throw an exception'; | |
1189 | } | |
1190 | ||
e425a60b | 1191 | { |
e425a60b | 1192 | # \c\ followed by _ |
b7ad6c3e KW |
1193 | unlike "x\c_y", qr/x\c\_y/, '\_ in a pattern'; |
1194 | like "x\c\_y", qr/x\c\_y/, '\_ in a pattern'; | |
e425a60b YO |
1195 | |
1196 | # \c\ followed by other characters | |
1197 | for my $c ("z", "\0", "!", chr(254), chr(256)) { | |
f36cc39d | 1198 | my $targ = "a" . uni_to_native("\034") . "$c"; |
e425a60b YO |
1199 | my $reg = "a\\c\\$c"; |
1200 | ok eval ("qq/$targ/ =~ /$reg/"), "\\c\\ in pattern"; | |
1201 | } | |
1202 | } | |
1203 | ||
e425a60b YO |
1204 | { # Test the (*PRUNE) pattern |
1205 | our $count = 0; | |
1206 | 'aaab' =~ /a+b?(?{$count++})(*FAIL)/; | |
de26e0cc | 1207 | is($count, 9, "Expect 9 for no (*PRUNE)"); |
e425a60b YO |
1208 | $count = 0; |
1209 | 'aaab' =~ /a+b?(*PRUNE)(?{$count++})(*FAIL)/; | |
de26e0cc | 1210 | is($count, 3, "Expect 3 with (*PRUNE)"); |
e425a60b YO |
1211 | local $_ = 'aaab'; |
1212 | $count = 0; | |
1213 | 1 while /.(*PRUNE)(?{$count++})(*FAIL)/g; | |
de26e0cc | 1214 | is($count, 4, "/.(*PRUNE)/"); |
e425a60b YO |
1215 | $count = 0; |
1216 | 'aaab' =~ /a+b?(??{'(*PRUNE)'})(?{$count++})(*FAIL)/; | |
de26e0cc | 1217 | is($count, 3, "Expect 3 with (*PRUNE)"); |
e425a60b YO |
1218 | local $_ = 'aaab'; |
1219 | $count = 0; | |
1220 | 1 while /.(??{'(*PRUNE)'})(?{$count++})(*FAIL)/g; | |
de26e0cc | 1221 | is($count, 4, "/.(*PRUNE)/"); |
e425a60b YO |
1222 | } |
1223 | ||
e425a60b YO |
1224 | { # Test the (*SKIP) pattern |
1225 | our $count = 0; | |
1226 | 'aaab' =~ /a+b?(*SKIP)(?{$count++})(*FAIL)/; | |
de26e0cc | 1227 | is($count, 1, "Expect 1 with (*SKIP)"); |
e425a60b YO |
1228 | local $_ = 'aaab'; |
1229 | $count = 0; | |
1230 | 1 while /.(*SKIP)(?{$count++})(*FAIL)/g; | |
de26e0cc | 1231 | is($count, 4, "/.(*SKIP)/"); |
e425a60b YO |
1232 | $_ = 'aaabaaab'; |
1233 | $count = 0; | |
1234 | our @res = (); | |
1235 | 1 while /(a+b?)(*SKIP)(?{$count++; push @res,$1})(*FAIL)/g; | |
de26e0cc NC |
1236 | is($count, 2, "Expect 2 with (*SKIP)"); |
1237 | is("@res", "aaab aaab", "Adjacent (*SKIP) works as expected"); | |
b9f8f219 YO |
1238 | |
1239 | $_ = "dir/file.mp3"; | |
1240 | my $got = m{ ( ([^/]+) (?: / (*SKIP)(*FAIL) | \z ) ) }x ? $1 : undef; | |
1241 | is($got, "file.mp3", "(*SKIP) and find_byclass() work together"); | |
e425a60b YO |
1242 | } |
1243 | ||
e425a60b YO |
1244 | { # Test the (*SKIP) pattern |
1245 | our $count = 0; | |
1246 | 'aaab' =~ /a+b?(*MARK:foo)(*SKIP)(?{$count++})(*FAIL)/; | |
de26e0cc | 1247 | is($count, 1, "Expect 1 with (*SKIP)"); |
e425a60b YO |
1248 | local $_ = 'aaab'; |
1249 | $count = 0; | |
1250 | 1 while /.(*MARK:foo)(*SKIP)(?{$count++})(*FAIL)/g; | |
de26e0cc | 1251 | is($count, 4, "/.(*SKIP)/"); |
e425a60b YO |
1252 | $_ = 'aaabaaab'; |
1253 | $count = 0; | |
1254 | our @res = (); | |
1255 | 1 while /(a+b?)(*MARK:foo)(*SKIP)(?{$count++; push @res,$1})(*FAIL)/g; | |
de26e0cc NC |
1256 | is($count, 2, "Expect 2 with (*SKIP)"); |
1257 | is("@res", "aaab aaab", "Adjacent (*SKIP) works as expected"); | |
e425a60b YO |
1258 | } |
1259 | ||
e425a60b YO |
1260 | { # Test the (*SKIP) pattern |
1261 | our $count = 0; | |
1262 | 'aaab' =~ /a*(*MARK:a)b?(*MARK:b)(*SKIP:a)(?{$count++})(*FAIL)/; | |
de26e0cc | 1263 | is($count, 3, "Expect 3 with *MARK:a)b?(*MARK:b)(*SKIP:a)"); |
e425a60b YO |
1264 | local $_ = 'aaabaaab'; |
1265 | $count = 0; | |
1266 | our @res = (); | |
1267 | 1 while | |
1268 | /(a*(*MARK:a)b?)(*MARK:x)(*SKIP:a)(?{$count++; push @res,$1})(*FAIL)/g; | |
de26e0cc NC |
1269 | is($count, 5, "Expect 5 with (*MARK:a)b?)(*MARK:x)(*SKIP:a)"); |
1270 | is("@res", "aaab b aaab b ", | |
1271 | "Adjacent (*MARK:a)b?)(*MARK:x)(*SKIP:a) works as expected"); | |
e425a60b YO |
1272 | } |
1273 | ||
e425a60b YO |
1274 | { # Test the (*COMMIT) pattern |
1275 | our $count = 0; | |
1276 | 'aaabaaab' =~ /a+b?(*COMMIT)(?{$count++})(*FAIL)/; | |
de26e0cc | 1277 | is($count, 1, "Expect 1 with (*COMMIT)"); |
e425a60b YO |
1278 | local $_ = 'aaab'; |
1279 | $count = 0; | |
1280 | 1 while /.(*COMMIT)(?{$count++})(*FAIL)/g; | |
de26e0cc | 1281 | is($count, 1, "/.(*COMMIT)/"); |
e425a60b YO |
1282 | $_ = 'aaabaaab'; |
1283 | $count = 0; | |
1284 | our @res = (); | |
1285 | 1 while /(a+b?)(*COMMIT)(?{$count++; push @res,$1})(*FAIL)/g; | |
de26e0cc NC |
1286 | is($count, 1, "Expect 1 with (*COMMIT)"); |
1287 | is("@res", "aaab", "Adjacent (*COMMIT) works as expected"); | |
3542935d | 1288 | |
b7ad6c3e | 1289 | unlike("1\n2a\n", qr/^\d+(*COMMIT)\w+/m, "COMMIT and anchors"); |
e425a60b YO |
1290 | } |
1291 | ||
e425a60b YO |
1292 | { |
1293 | # Test named commits and the $REGERROR var | |
fee50582 | 1294 | local $REGERROR; |
e425a60b YO |
1295 | for my $name ('', ':foo') { |
1296 | for my $pat ("(*PRUNE$name)", | |
1297 | ($name ? "(*MARK$name)" : "") . "(*SKIP$name)", | |
0f289c68 | 1298 | "(*COMMIT$name)") { |
e425a60b YO |
1299 | for my $suffix ('(*FAIL)', '') { |
1300 | 'aaaab' =~ /a+b$pat$suffix/; | |
de26e0cc | 1301 | is($REGERROR, |
e425a60b | 1302 | ($suffix ? ($name ? 'foo' : "1") : ""), |
de26e0cc | 1303 | "Test $pat and \$REGERROR $suffix"); |
e425a60b YO |
1304 | } |
1305 | } | |
1306 | } | |
1307 | } | |
1308 | ||
e425a60b YO |
1309 | { |
1310 | # Test named commits and the $REGERROR var | |
1311 | package Fnorble; | |
1312 | our $REGERROR; | |
fee50582 | 1313 | local $REGERROR; |
e425a60b YO |
1314 | for my $name ('', ':foo') { |
1315 | for my $pat ("(*PRUNE$name)", | |
1316 | ($name ? "(*MARK$name)" : "") . "(*SKIP$name)", | |
0f289c68 | 1317 | "(*COMMIT$name)") { |
e425a60b YO |
1318 | for my $suffix ('(*FAIL)','') { |
1319 | 'aaaab' =~ /a+b$pat$suffix/; | |
de26e0cc | 1320 | ::is($REGERROR, |
e425a60b | 1321 | ($suffix ? ($name ? 'foo' : "1") : ""), |
de26e0cc | 1322 | "Test $pat and \$REGERROR $suffix"); |
e425a60b YO |
1323 | } |
1324 | } | |
0f289c68 YO |
1325 | } |
1326 | } | |
e425a60b | 1327 | |
e425a60b YO |
1328 | { |
1329 | # Test named commits and the $REGERROR var | |
de946258 | 1330 | my $message = '$REGERROR'; |
fee50582 | 1331 | local $REGERROR; |
e425a60b YO |
1332 | for my $word (qw (bar baz bop)) { |
1333 | $REGERROR = ""; | |
1334 | "aaaaa$word" =~ | |
1335 | /a+(?:bar(*COMMIT:bar)|baz(*COMMIT:baz)|bop(*COMMIT:bop))(*FAIL)/; | |
de946258 | 1336 | is($REGERROR, $word, $message); |
0f289c68 | 1337 | } |
e425a60b YO |
1338 | } |
1339 | ||
e425a60b YO |
1340 | { |
1341 | #Mindnumbingly simple test of (*THEN) | |
1342 | for ("ABC","BAX") { | |
1343 | ok /A (*THEN) X | B (*THEN) C/x, "Simple (*THEN) test"; | |
1344 | } | |
1345 | } | |
1346 | ||
e425a60b | 1347 | { |
de946258 | 1348 | my $message = "Relative Recursion"; |
e425a60b YO |
1349 | my $parens = qr/(\((?:[^()]++|(?-1))*+\))/; |
1350 | local $_ = 'foo((2*3)+4-3) + bar(2*(3+4)-1*(2-3))'; | |
1351 | my ($all, $one, $two) = ('', '', ''); | |
de946258 NC |
1352 | ok(m/foo $parens \s* \+ \s* bar $parens/x, $message); |
1353 | is($1, '((2*3)+4-3)', $message); | |
1354 | is($2, '(2*(3+4)-1*(2-3))', $message); | |
1355 | is($&, 'foo((2*3)+4-3) + bar(2*(3+4)-1*(2-3))', $message); | |
1356 | is($&, $_, $message); | |
e425a60b YO |
1357 | } |
1358 | ||
1359 | { | |
1360 | my $spaces=" "; | |
1361 | local $_ = join 'bar', $spaces, $spaces; | |
1362 | our $count = 0; | |
1363 | s/(?>\s+bar)(?{$count++})//g; | |
de26e0cc NC |
1364 | is($_, $spaces, "SUSPEND final string"); |
1365 | is($count, 1, "Optimiser should have prevented more than one match"); | |
e425a60b YO |
1366 | } |
1367 | ||
e425a60b YO |
1368 | { |
1369 | # From Message-ID: <877ixs6oa6.fsf@k75.linux.bogus> | |
1370 | my $dow_name = "nada"; | |
1371 | my $parser = "(\$dow_name) = \$time_string =~ /(D\x{e9}\\ " . | |
1372 | "C\x{e9}adaoin|D\x{e9}\\ Sathairn|\\w+|\x{100})/"; | |
1373 | my $time_string = "D\x{e9} C\x{e9}adaoin"; | |
1374 | eval $parser; | |
1375 | ok !$@, "Test Eval worked"; | |
de26e0cc | 1376 | is($dow_name, $time_string, "UTF-8 trie common prefix extraction"); |
e425a60b YO |
1377 | } |
1378 | ||
e425a60b YO |
1379 | { |
1380 | my $v; | |
1381 | ($v = 'bar') =~ /(\w+)/g; | |
1382 | $v = 'foo'; | |
de26e0cc NC |
1383 | is("$1", 'bar', |
1384 | '$1 is safe after /g - may fail due to specialized config in pp_hot.c'); | |
e425a60b YO |
1385 | } |
1386 | ||
e425a60b | 1387 | { |
de946258 | 1388 | my $message = "http://nntp.perl.org/group/perl.perl5.porters/118663"; |
e425a60b | 1389 | my $qr_barR1 = qr/(bar)\g-1/; |
de946258 NC |
1390 | like("foobarbarxyz", $qr_barR1, $message); |
1391 | like("foobarbarxyz", qr/foo${qr_barR1}xyz/, $message); | |
1392 | like("foobarbarxyz", qr/(foo)${qr_barR1}xyz/, $message); | |
1393 | like("foobarbarxyz", qr/(foo)(bar)\g{-1}xyz/, $message); | |
1394 | like("foobarbarxyz", qr/(foo${qr_barR1})xyz/, $message); | |
1395 | like("foobarbarxyz", qr/(foo(bar)\g{-1})xyz/, $message); | |
0f289c68 | 1396 | } |
e425a60b | 1397 | |
e425a60b | 1398 | { |
de946258 | 1399 | my $message = '$REGMARK'; |
e425a60b | 1400 | our @r = (); |
fee50582 YO |
1401 | local $REGMARK; |
1402 | local $REGERROR; | |
de946258 NC |
1403 | like('foofoo', qr/foo (*MARK:foo) (?{push @r,$REGMARK}) /x, $message); |
1404 | is("@r","foo", $message); | |
1405 | is($REGMARK, "foo", $message); | |
1406 | unlike('foofoo', qr/foo (*MARK:foo) (*FAIL) /x, $message); | |
1407 | is($REGMARK, '', $message); | |
1408 | is($REGERROR, 'foo', $message); | |
e425a60b YO |
1409 | } |
1410 | ||
e425a60b | 1411 | { |
de946258 | 1412 | my $message = '\K test'; |
e425a60b YO |
1413 | my $x; |
1414 | $x = "abc.def.ghi.jkl"; | |
1415 | $x =~ s/.*\K\..*//; | |
de946258 | 1416 | is($x, "abc.def.ghi", $message); |
0f289c68 | 1417 | |
e425a60b YO |
1418 | $x = "one two three four"; |
1419 | $x =~ s/o+ \Kthree//g; | |
de946258 | 1420 | is($x, "one two four", $message); |
0f289c68 | 1421 | |
e425a60b YO |
1422 | $x = "abcde"; |
1423 | $x =~ s/(.)\K/$1/g; | |
de946258 | 1424 | is($x, "aabbccddee", $message); |
e425a60b YO |
1425 | } |
1426 | ||
e425a60b YO |
1427 | { |
1428 | sub kt { | |
1429 | return '4' if $_[0] eq '09028623'; | |
1430 | } | |
1431 | # Nested EVAL using PL_curpm (via $1 or friends) | |
1432 | my $re; | |
1433 | our $grabit = qr/ ([0-6][0-9]{7}) (??{ kt $1 }) [890] /x; | |
1434 | $re = qr/^ ( (??{ $grabit }) ) $ /x; | |
1435 | my @res = '0902862349' =~ $re; | |
de26e0cc NC |
1436 | is(join ("-", @res), "0902862349", |
1437 | 'PL_curpm is set properly on nested eval'); | |
e425a60b YO |
1438 | |
1439 | our $qr = qr/ (o) (??{ $1 }) /x; | |
1440 | ok 'boob'=~/( b (??{ $qr }) b )/x && 1, "PL_curpm, nested eval"; | |
1441 | } | |
1442 | ||
e425a60b YO |
1443 | { |
1444 | use charnames ":full"; | |
b7ad6c3e KW |
1445 | like "\N{ROMAN NUMERAL ONE}", qr/\p{Alphabetic}/, "I =~ Alphabetic"; |
1446 | like "\N{ROMAN NUMERAL ONE}", qr/\p{Uppercase}/, "I =~ Uppercase"; | |
1447 | unlike "\N{ROMAN NUMERAL ONE}", qr/\p{Lowercase}/, "I !~ Lowercase"; | |
1448 | like "\N{ROMAN NUMERAL ONE}", qr/\p{IDStart}/, "I =~ ID_Start"; | |
1449 | like "\N{ROMAN NUMERAL ONE}", qr/\p{IDContinue}/, "I =~ ID_Continue"; | |
1450 | like "\N{SMALL ROMAN NUMERAL ONE}", qr/\p{Alphabetic}/, "i =~ Alphabetic"; | |
1451 | unlike "\N{SMALL ROMAN NUMERAL ONE}", qr/\p{Uppercase}/, "i !~ Uppercase"; | |
1452 | like "\N{SMALL ROMAN NUMERAL ONE}", qr/\p{Uppercase}/i, "i =~ Uppercase under /i"; | |
1453 | unlike "\N{SMALL ROMAN NUMERAL ONE}", qr/\p{Titlecase}/, "i !~ Titlecase"; | |
1454 | like "\N{SMALL ROMAN NUMERAL ONE}", qr/\p{Titlecase}/i, "i =~ Titlecase under /i"; | |
1455 | like "\N{ROMAN NUMERAL ONE}", qr/\p{Lowercase}/i, "I =~ Lowercase under /i"; | |
1456 | ||
1457 | like "\N{SMALL ROMAN NUMERAL ONE}", qr/\p{Lowercase}/, "i =~ Lowercase"; | |
1458 | like "\N{SMALL ROMAN NUMERAL ONE}", qr/\p{IDStart}/, "i =~ ID_Start"; | |
1459 | like "\N{SMALL ROMAN NUMERAL ONE}", qr/\p{IDContinue}/, "i =~ ID_Continue" | |
e425a60b YO |
1460 | } |
1461 | ||
0ab1ff8e KW |
1462 | { # More checking that /i works on the few properties that it makes a |
1463 | # difference. Uppercase, Lowercase, and Titlecase were done in the | |
1464 | # block above | |
b7ad6c3e KW |
1465 | like "A", qr/\p{PosixUpper}/, "A =~ PosixUpper"; |
1466 | like "A", qr/\p{PosixUpper}/i, "A =~ PosixUpper under /i"; | |
1467 | unlike "A", qr/\p{PosixLower}/, "A !~ PosixLower"; | |
1468 | like "A", qr/\p{PosixLower}/i, "A =~ PosixLower under /i"; | |
1469 | unlike "a", qr/\p{PosixUpper}/, "a !~ PosixUpper"; | |
1470 | like "a", qr/\p{PosixUpper}/i, "a =~ PosixUpper under /i"; | |
1471 | like "a", qr/\p{PosixLower}/, "a =~ PosixLower"; | |
1472 | like "a", qr/\p{PosixLower}/i, "a =~ PosixLower under /i"; | |
1473 | ||
1474 | like uni_to_native("\xC0"), qr/\p{XPosixUpper}/, "\\xC0 =~ XPosixUpper"; | |
1475 | like uni_to_native("\xC0"), qr/\p{XPosixUpper}/i, "\\xC0 =~ XPosixUpper under /i"; | |
1476 | unlike uni_to_native("\xC0"), qr/\p{XPosixLower}/, "\\xC0 !~ XPosixLower"; | |
1477 | like uni_to_native("\xC0"), qr/\p{XPosixLower}/i, "\\xC0 =~ XPosixLower under /i"; | |
1478 | unlike uni_to_native("\xE0"), qr/\p{XPosixUpper}/, "\\xE0 !~ XPosixUpper"; | |
1479 | like uni_to_native("\xE0"), qr/\p{XPosixUpper}/i, "\\xE0 =~ XPosixUpper under /i"; | |
1480 | like uni_to_native("\xE0"), qr/\p{XPosixLower}/, "\\xE0 =~ XPosixLower"; | |
1481 | like uni_to_native("\xE0"), qr/\p{XPosixLower}/i, "\\xE0 =~ XPosixLower under /i"; | |
1482 | ||
1483 | like uni_to_native("\xC0"), qr/\p{UppercaseLetter}/, "\\xC0 =~ UppercaseLetter"; | |
1484 | like uni_to_native("\xC0"), qr/\p{UppercaseLetter}/i, "\\xC0 =~ UppercaseLetter under /i"; | |
1485 | unlike uni_to_native("\xC0"), qr/\p{LowercaseLetter}/, "\\xC0 !~ LowercaseLetter"; | |
1486 | like uni_to_native("\xC0"), qr/\p{LowercaseLetter}/i, "\\xC0 =~ LowercaseLetter under /i"; | |
1487 | unlike uni_to_native("\xC0"), qr/\p{TitlecaseLetter}/, "\\xC0 !~ TitlecaseLetter"; | |
1488 | like uni_to_native("\xC0"), qr/\p{TitlecaseLetter}/i, "\\xC0 =~ TitlecaseLetter under /i"; | |
1489 | unlike uni_to_native("\xE0"), qr/\p{UppercaseLetter}/, "\\xE0 !~ UppercaseLetter"; | |
1490 | like uni_to_native("\xE0"), qr/\p{UppercaseLetter}/i, "\\xE0 =~ UppercaseLetter under /i"; | |
1491 | like uni_to_native("\xE0"), qr/\p{LowercaseLetter}/, "\\xE0 =~ LowercaseLetter"; | |
1492 | like uni_to_native("\xE0"), qr/\p{LowercaseLetter}/i, "\\xE0 =~ LowercaseLetter under /i"; | |
1493 | unlike uni_to_native("\xE0"), qr/\p{TitlecaseLetter}/, "\\xE0 !~ TitlecaseLetter"; | |
1494 | like uni_to_native("\xE0"), qr/\p{TitlecaseLetter}/i, "\\xE0 =~ TitlecaseLetter under /i"; | |
1495 | unlike "\x{1C5}", qr/\p{UppercaseLetter}/, "\\x{1C5} !~ UppercaseLetter"; | |
1496 | like "\x{1C5}", qr/\p{UppercaseLetter}/i, "\\x{1C5} =~ UppercaseLetter under /i"; | |
1497 | unlike "\x{1C5}", qr/\p{LowercaseLetter}/, "\\x{1C5} !~ LowercaseLetter"; | |
1498 | like "\x{1C5}", qr/\p{LowercaseLetter}/i, "\\x{1C5} =~ LowercaseLetter under /i"; | |
1499 | like "\x{1C5}", qr/\p{TitlecaseLetter}/, "\\x{1C5} =~ TitlecaseLetter"; | |
1500 | like "\x{1C5}", qr/\p{TitlecaseLetter}/i, "\\x{1C5} =~ TitlecaseLetter under /i"; | |
0ab1ff8e KW |
1501 | } |
1502 | ||
e425a60b YO |
1503 | { |
1504 | # requirement of Unicode Technical Standard #18, 1.7 Code Points | |
1505 | # cf. http://www.unicode.org/reports/tr18/#Supplementary_Characters | |
1506 | for my $u (0x7FF, 0x800, 0xFFFF, 0x10000) { | |
1507 | no warnings 'utf8'; # oops | |
1508 | my $c = chr $u; | |
1509 | my $x = sprintf '%04X', $u; | |
b7ad6c3e | 1510 | like "A${c}B", qr/A[\0-\x{10000}]B/, "Unicode range - $x"; |
e425a60b YO |
1511 | } |
1512 | } | |
1513 | ||
e425a60b | 1514 | { |
fe5492d9 YO |
1515 | no warnings 'uninitialized'; |
1516 | my $res = ""; | |
e425a60b YO |
1517 | |
1518 | if ('1' =~ /(?|(?<digit>1)|(?<digit>2))/) { | |
1519 | $res = "@{$- {digit}}"; | |
1520 | } | |
fe5492d9 YO |
1521 | is($res, "1 ", |
1522 | "Check that repeated named captures in branch reset (?|...) work as expected"); | |
1523 | if ('2' =~ /(?|(?<digit>1)|(?<digit>2))/) { | |
1524 | $res = "@{$- {digit}}"; | |
1525 | } | |
1526 | is($res, " 2", | |
1527 | "Check that repeated named captures in branch reset (?|...) work as expected"); | |
0f289c68 | 1528 | |
e425a60b YO |
1529 | $res = ""; |
1530 | if ('11' =~ /(?|(?<digit>1)|(?<digit>2))(?&digit)/) { | |
1531 | $res = "@{$- {digit}}"; | |
1532 | } | |
fe5492d9 | 1533 | is($res, "1 ", |
de26e0cc | 1534 | "Check that (?&..) to a buffer inside a (?|...) goes to the leftmost"); |
e425a60b YO |
1535 | } |
1536 | ||
e425a60b YO |
1537 | { |
1538 | use warnings; | |
de946258 | 1539 | my $message = "ASCII pattern that really is UTF-8"; |
e425a60b YO |
1540 | my @w; |
1541 | local $SIG {__WARN__} = sub {push @w, "@_"}; | |
0f289c68 | 1542 | my $c = qq (\x{DF}); |
de946258 NC |
1543 | like($c, qr/${c}|\x{100}/, $message); |
1544 | is("@w", '', $message); | |
0f289c68 | 1545 | } |
e425a60b | 1546 | |
e425a60b | 1547 | { |
de946258 | 1548 | my $message = "Corruption of match results of qr// across scopes"; |
e425a60b YO |
1549 | my $qr = qr/(fo+)(ba+r)/; |
1550 | 'foobar' =~ /$qr/; | |
de946258 | 1551 | is("$1$2", "foobar", $message); |
e425a60b YO |
1552 | { |
1553 | 'foooooobaaaaar' =~ /$qr/; | |
de946258 | 1554 | is("$1$2", 'foooooobaaaaar', $message); |
e425a60b | 1555 | } |
de946258 | 1556 | is("$1$2", "foobar", $message); |
e425a60b YO |
1557 | } |
1558 | ||
e425a60b | 1559 | { |
de946258 | 1560 | my $message = "HORIZWS"; |
e425a60b YO |
1561 | local $_ = "\t \r\n \n \t".chr(11)."\n"; |
1562 | s/\H/H/g; | |
1563 | s/\h/h/g; | |
de946258 | 1564 | is($_, "hhHHhHhhHH", $message); |
e425a60b YO |
1565 | $_ = "\t \r\n \n \t" . chr (11) . "\n"; |
1566 | utf8::upgrade ($_); | |
1567 | s/\H/H/g; | |
1568 | s/\h/h/g; | |
de946258 | 1569 | is($_, "hhHHhHhhHH", $message); |
0f289c68 | 1570 | } |
e425a60b | 1571 | |
e425a60b | 1572 | { |
de946258 | 1573 | # Various whitespace special patterns |
f36cc39d KW |
1574 | my @h = map {chr utf8::unicode_to_native($_) } |
1575 | 0x09, 0x20, 0xa0, 0x1680, 0x2000, | |
e425a60b YO |
1576 | 0x2001, 0x2002, 0x2003, 0x2004, 0x2005, 0x2006, |
1577 | 0x2007, 0x2008, 0x2009, 0x200a, 0x202f, 0x205f, | |
1578 | 0x3000; | |
f36cc39d KW |
1579 | my @v = map {chr utf8::unicode_to_native($_) } |
1580 | 0x0a, 0x0b, 0x0c, 0x0d, 0x85, 0x2028, | |
e425a60b | 1581 | 0x2029; |
f36cc39d KW |
1582 | my @lb = (uni_to_native("\x0D\x0A"), |
1583 | map {chr utf8::unicode_to_native($_) } | |
1584 | 0x0A .. 0x0D, 0x85, 0x2028, 0x2029); | |
e425a60b YO |
1585 | foreach my $t ([\@h, qr/\h/, qr/\h+/], |
1586 | [\@v, qr/\v/, qr/\v+/], | |
1587 | [\@lb, qr/\R/, qr/\R+/],) { | |
1588 | my $ary = shift @$t; | |
1589 | foreach my $pat (@$t) { | |
1590 | foreach my $str (@$ary) { | |
df6841b6 KW |
1591 | my $temp_str = $str; |
1592 | $temp_str = display($temp_str); | |
1593 | ok $str =~ /($pat)/, $temp_str . " =~ /($pat)"; | |
1594 | my $temp_1 = $1; | |
1595 | is($1, $str, "\$1='" . display($temp_1) . "' eq '" . $temp_str . "' after ($pat)"); | |
e425a60b | 1596 | utf8::upgrade ($str); |
df6841b6 KW |
1597 | ok $str =~ /($pat)/, "Upgraded " . $temp_str . " =~ /($pat)/"; |
1598 | is($1, $str, "\$1='" . display($temp_1) . "' eq '" . $temp_str . "'(upgraded) after ($pat)"); | |
e425a60b YO |
1599 | } |
1600 | } | |
1601 | } | |
1602 | } | |
1603 | ||
e425a60b | 1604 | { |
de946258 | 1605 | # Check that \\xDF match properly in its various forms |
e425a60b YO |
1606 | # Test that \xDF matches properly. this is pretty hacky stuff, |
1607 | # but its actually needed. The malarky with '-' is to prevent | |
1608 | # compilation caching from playing any role in the test. | |
f36cc39d | 1609 | my @df = (chr utf8::unicode_to_native(0xDF), '-', chr utf8::unicode_to_native(0xDF)); |
e425a60b | 1610 | utf8::upgrade ($df [2]); |
f36cc39d | 1611 | my @strs = ('ss', 'sS', 'Ss', 'SS', chr utf8::unicode_to_native(0xDF)); |
e425a60b YO |
1612 | my @ss = map {("$_", "$_")} @strs; |
1613 | utf8::upgrade ($ss [$_ * 2 + 1]) for 0 .. $#strs; | |
1614 | ||
1615 | for my $ssi (0 .. $#ss) { | |
1616 | for my $dfi (0 .. $#df) { | |
1617 | my $pat = $df [$dfi]; | |
1618 | my $str = $ss [$ssi]; | |
1619 | my $utf_df = ($dfi > 1) ? 'utf8' : ''; | |
1620 | my $utf_ss = ($ssi % 2) ? 'utf8' : ''; | |
f36cc39d KW |
1621 | my $sstr; # We hard-code the ebcdic value below to avoid |
1622 | # perturbing the test | |
1623 | ($sstr = $str) =~ s/\xDF/\\xDF/ if $::IS_ASCII; | |
1624 | ($sstr = $str) =~ s/\x59/\\x59/ if $::IS_EBCDIC; | |
e425a60b YO |
1625 | |
1626 | if ($utf_df || $utf_ss || length ($ss [$ssi]) == 1) { | |
1627 | my $ret = $str =~ /$pat/i; | |
1628 | next if $pat eq '-'; | |
f36cc39d KW |
1629 | if ($::IS_ASCII) { |
1630 | ok $ret, "\"$sstr\" =~ /\\xDF/i " . | |
1631 | "(str is @{[$utf_ss||'latin']}, pat is " . | |
1632 | "@{[$utf_df||'latin']})"; | |
1633 | } | |
1634 | else { | |
1635 | ok $ret, "\"$sstr\" =~ /\\x59/i " . | |
e425a60b YO |
1636 | "(str is @{[$utf_ss||'latin']}, pat is " . |
1637 | "@{[$utf_df||'latin']})"; | |
f36cc39d | 1638 | } |
e425a60b YO |
1639 | } |
1640 | else { | |
1641 | my $ret = $str !~ /$pat/i; | |
1642 | next if $pat eq '-'; | |
f36cc39d KW |
1643 | if ($::IS_EBCDIC) { |
1644 | ok $ret, "\"$sstr\" !~ /\\x59/i " . | |
e425a60b YO |
1645 | "(str is @{[$utf_ss||'latin']}, pat is " . |
1646 | "@{[$utf_df||'latin']})"; | |
f36cc39d KW |
1647 | } |
1648 | else { | |
1649 | ok $ret, "\"$sstr\" !~ /\\xDF/i " . | |
1650 | "(str is @{[$utf_ss||'latin']}, pat is " . | |
1651 | "@{[$utf_df||'latin']})"; | |
1652 | } | |
e425a60b YO |
1653 | } |
1654 | } | |
1655 | } | |
1656 | } | |
1657 | ||
e425a60b | 1658 | { |
de946258 | 1659 | my $message = "BBC(Bleadperl Breaks CPAN) Today: String::Multibyte"; |
e425a60b YO |
1660 | my $re = qr/(?:[\x00-\xFF]{4})/; |
1661 | my $hyp = "\0\0\0-"; | |
1662 | my $esc = "\0\0\0\\"; | |
1663 | ||
1664 | my $str = "$esc$hyp$hyp$esc$esc"; | |
1665 | my @a = ($str =~ /\G(?:\Q$esc$esc\E|\Q$esc$hyp\E|$re)/g); | |
1666 | ||
de946258 | 1667 | is(@a,3, $message); |
e425a60b | 1668 | local $" = "="; |
de946258 | 1669 | is("@a","$esc$hyp=$hyp=$esc$esc", $message); |
e425a60b YO |
1670 | } |
1671 | ||
e425a60b YO |
1672 | { |
1673 | # Test for keys in %+ and %- | |
de946258 | 1674 | my $message = 'Test keys in %+ and %-'; |
7fba2966 FC |
1675 | no warnings 'uninitialized'; |
1676 | local $_ = "abcdef"; | |
e425a60b | 1677 | /(?<foo>a)|(?<foo>b)/; |
de946258 NC |
1678 | is((join ",", sort keys %+), "foo", $message); |
1679 | is((join ",", sort keys %-), "foo", $message); | |
1680 | is((join ",", sort values %+), "a", $message); | |
1681 | is((join ",", sort map "@$_", values %-), "a ", $message); | |
e425a60b | 1682 | /(?<bar>a)(?<bar>b)(?<quux>.)/; |
de946258 NC |
1683 | is((join ",", sort keys %+), "bar,quux", $message); |
1684 | is((join ",", sort keys %-), "bar,quux", $message); | |
1685 | is((join ",", sort values %+), "a,c", $message); # leftmost | |
1686 | is((join ",", sort map "@$_", values %-), "a b,c", $message); | |
e425a60b | 1687 | /(?<un>a)(?<deux>c)?/; # second buffer won't capture |
de946258 NC |
1688 | is((join ",", sort keys %+), "un", $message); |
1689 | is((join ",", sort keys %-), "deux,un", $message); | |
1690 | is((join ",", sort values %+), "a", $message); | |
1691 | is((join ",", sort map "@$_", values %-), ",a", $message); | |
e425a60b YO |
1692 | } |
1693 | ||
e425a60b YO |
1694 | { |
1695 | # length() on captures, the numbered ones end up in Perl_magic_len | |
7fba2966 | 1696 | local $_ = "aoeu " . uni_to_native("\xe6") . "var ook"; |
e425a60b YO |
1697 | /^ \w+ \s (?<eek>\S+)/x; |
1698 | ||
de26e0cc NC |
1699 | is(length $`, 0, q[length $`]); |
1700 | is(length $', 4, q[length $']); | |
1701 | is(length $&, 9, q[length $&]); | |
1702 | is(length $1, 4, q[length $1]); | |
1703 | is(length $+{eek}, 4, q[length $+{eek} == length $1]); | |
e425a60b YO |
1704 | } |
1705 | ||
e425a60b YO |
1706 | { |
1707 | my $ok = -1; | |
1708 | ||
1709 | $ok = exists ($-{x}) ? 1 : 0 if 'bar' =~ /(?<x>foo)|bar/; | |
de26e0cc NC |
1710 | is($ok, 1, '$-{x} exists after "bar"=~/(?<x>foo)|bar/'); |
1711 | is(scalar (%+), 0, 'scalar %+ == 0 after "bar"=~/(?<x>foo)|bar/'); | |
1712 | is(scalar (%-), 1, 'scalar %- == 1 after "bar"=~/(?<x>foo)|bar/'); | |
e425a60b YO |
1713 | |
1714 | $ok = -1; | |
1715 | $ok = exists ($+{x}) ? 1 : 0 if 'bar' =~ /(?<x>foo)|bar/; | |
de26e0cc NC |
1716 | is($ok, 0, '$+{x} not exists after "bar"=~/(?<x>foo)|bar/'); |
1717 | is(scalar (%+), 0, 'scalar %+ == 0 after "bar"=~/(?<x>foo)|bar/'); | |
1718 | is(scalar (%-), 1, 'scalar %- == 1 after "bar"=~/(?<x>foo)|bar/'); | |
e425a60b YO |
1719 | |
1720 | $ok = -1; | |
1721 | $ok = exists ($-{x}) ? 1 : 0 if 'foo' =~ /(?<x>foo)|bar/; | |
de26e0cc NC |
1722 | is($ok, 1, '$-{x} exists after "foo"=~/(?<x>foo)|bar/'); |
1723 | is(scalar (%+), 1, 'scalar %+ == 1 after "foo"=~/(?<x>foo)|bar/'); | |
1724 | is(scalar (%-), 1, 'scalar %- == 1 after "foo"=~/(?<x>foo)|bar/'); | |
e425a60b YO |
1725 | |
1726 | $ok = -1; | |
1727 | $ok = exists ($+{x}) ? 1 : 0 if 'foo'=~/(?<x>foo)|bar/; | |
de26e0cc | 1728 | is($ok, 1, '$+{x} exists after "foo"=~/(?<x>foo)|bar/'); |
e425a60b YO |
1729 | } |
1730 | ||
e425a60b YO |
1731 | { |
1732 | local $_; | |
1733 | ($_ = 'abc') =~ /(abc)/g; | |
0f289c68 | 1734 | $_ = '123'; |
de26e0cc | 1735 | is("$1", 'abc', "/g leads to unsafe match vars: $1"); |
1b474ee3 NC |
1736 | |
1737 | fresh_perl_is(<<'EOP', ">abc<\n", {}, 'mention $&'); | |
1738 | $&; | |
1739 | my $x; | |
1740 | ($x='abc')=~/(abc)/g; | |
1741 | $x='123'; | |
1742 | print ">$1<\n"; | |
1743 | EOP | |
1744 | ||
1b474ee3 NC |
1745 | fresh_perl_is(<<'EOP', ">abc<\n", {}, 'no mention of $&'); |
1746 | my $x; | |
1747 | ($x='abc')=~/(abc)/g; | |
1748 | $x='123'; | |
1749 | print ">$1<\n"; | |
1750 | EOP | |
e425a60b YO |
1751 | } |
1752 | ||
e425a60b | 1753 | { |
de946258 | 1754 | # Message-ID: <20070818091501.7eff4831@r2d2> |
e425a60b YO |
1755 | my $str = ""; |
1756 | for (0 .. 5) { | |
1757 | my @x; | |
1758 | $str .= "@x"; # this should ALWAYS be the empty string | |
1759 | 'a' =~ /(a|)/; | |
1760 | push @x, 1; | |
1761 | } | |
de26e0cc | 1762 | is(length $str, 0, "Trie scope error, string should be empty"); |
e425a60b YO |
1763 | $str = ""; |
1764 | my @foo = ('a') x 5; | |
1765 | for (@foo) { | |
1766 | my @bar; | |
1767 | $str .= "@bar"; | |
1768 | s/a|/push @bar, 1/e; | |
1769 | } | |
de26e0cc | 1770 | is(length $str, 0, "Trie scope error, string should be empty"); |
e425a60b YO |
1771 | } |
1772 | ||
e425a60b | 1773 | { |
e425a60b | 1774 | # more TRIE/AHOCORASICK problems with mixed utf8 / latin-1 and case folding |
f36cc39d KW |
1775 | for my $ord (160 .. 255) { |
1776 | my $chr = utf8::unicode_to_native($ord); | |
0f289c68 YO |
1777 | my $chr_byte = chr($chr); |
1778 | my $chr_utf8 = chr($chr); utf8::upgrade($chr_utf8); | |
1779 | my $rx = qr{$chr_byte|X}i; | |
b7ad6c3e | 1780 | like($chr_utf8, $rx, "utf8/latin, codepoint $chr"); |
0f289c68 | 1781 | } |
e425a60b YO |
1782 | } |
1783 | ||
1784 | { | |
e425a60b YO |
1785 | our $a = 3; "" =~ /(??{ $a })/; |
1786 | our $b = $a; | |
de26e0cc | 1787 | is($b, $a, "Copy of scalar used for postponed subexpression"); |
e425a60b YO |
1788 | } |
1789 | ||
e425a60b | 1790 | { |
e425a60b YO |
1791 | our @ctl_n = (); |
1792 | our @plus = (); | |
1793 | our $nested_tags; | |
1794 | $nested_tags = qr{ | |
1795 | < | |
1796 | (\w+) | |
1797 | (?{ | |
1798 | push @ctl_n,$^N; | |
1799 | push @plus,$+; | |
1800 | }) | |
1801 | > | |
1802 | (??{$nested_tags})* | |
1803 | </\s* \w+ \s*> | |
1804 | }x; | |
1805 | ||
1806 | my $match = '<bla><blubb></blubb></bla>' =~ m/^$nested_tags$/; | |
1807 | ok $match, 'nested construct matches'; | |
de26e0cc NC |
1808 | is("@ctl_n", "bla blubb", '$^N inside of (?{}) works as expected'); |
1809 | is("@plus", "bla blubb", '$+ inside of (?{}) works as expected'); | |
e425a60b YO |
1810 | } |
1811 | ||
e425a60b YO |
1812 | SKIP: { |
1813 | # XXX: This set of tests is essentially broken, POSIX character classes | |
0f289c68 | 1814 | # should not have differing definitions under Unicode. |
e425a60b | 1815 | # There are property names for that. |
ef237063 | 1816 | skip "Tests assume ASCII", 4 unless $::IS_ASCII; |
e425a60b YO |
1817 | |
1818 | my @notIsPunct = grep {/[[:punct:]]/ and not /\p{IsPunct}/} | |
1819 | map {chr} 0x20 .. 0x7f; | |
1a3f15c1 | 1820 | is(join ('', @notIsPunct), '$+<=>^`|~', |
de26e0cc | 1821 | '[:punct:] disagrees with IsPunct on Symbols'); |
e425a60b YO |
1822 | |
1823 | my @isPrint = grep {not /[[:print:]]/ and /\p{IsPrint}/} | |
1824 | map {chr} 0 .. 0x1f, 0x7f .. 0x9f; | |
de26e0cc NC |
1825 | is(join ('', @isPrint), "", |
1826 | 'IsPrint agrees with [:print:] on control characters'); | |
e425a60b YO |
1827 | |
1828 | my @isPunct = grep {/[[:punct:]]/ != /\p{IsPunct}/} | |
1829 | map {chr} 0x80 .. 0xff; | |
7620cb10 | 1830 | is(join ('', @isPunct), "\xa1\xa7\xab\xb6\xb7\xbb\xbf", # ¡ « · » ¿ |
de26e0cc | 1831 | 'IsPunct disagrees with [:punct:] outside ASCII'); |
e425a60b YO |
1832 | |
1833 | my @isPunctLatin1 = eval q { | |
3fc5553b | 1834 | grep {/[[:punct:]]/u != /\p{IsPunct}/} map {chr} 0x80 .. 0xff; |
e425a60b YO |
1835 | }; |
1836 | skip "Eval failed ($@)", 1 if $@; | |
1837 | skip "PERL_LEGACY_UNICODE_CHARCLASS_MAPPINGS set to 0", 1 | |
1838 | if !$ENV{PERL_TEST_LEGACY_POSIX_CC}; | |
de26e0cc NC |
1839 | is(join ('', @isPunctLatin1), '', |
1840 | 'IsPunct agrees with [:punct:] with explicit Latin1'); | |
0f289c68 | 1841 | } |
d1c771f5 | 1842 | |
d1c771f5 B |
1843 | { |
1844 | # Tests for [#perl 71942] | |
1845 | our $count_a; | |
1846 | our $count_b; | |
1847 | ||
1848 | my $c = 0; | |
1849 | for my $re ( | |
1850 | # [ | |
1851 | # should match?, | |
1852 | # input string, | |
1853 | # re 1, | |
1854 | # re 2, | |
1855 | # expected values of count_a and count_b, | |
1856 | # ] | |
1857 | [ | |
1858 | 0, | |
1859 | "xababz", | |
1860 | qr/a+(?{$count_a++})b?(*COMMIT)(*FAIL)/, | |
1861 | qr/a+(?{$count_b++})b?(*COMMIT)z/, | |
1862 | 1, | |
1863 | ], | |
1864 | [ | |
1865 | 0, | |
1866 | "xababz", | |
1867 | qr/a+(?{$count_a++})b?(*COMMIT)\s*(*FAIL)/, | |
1868 | qr/a+(?{$count_b++})b?(*COMMIT)\s*z/, | |
1869 | 1, | |
1870 | ], | |
1871 | [ | |
1872 | 0, | |
1873 | "xababz", | |
1874 | qr/a+(?{$count_a++})(?:b|)?(*COMMIT)(*FAIL)/, | |
1875 | qr/a+(?{$count_b++})(?:b|)?(*COMMIT)z/, | |
1876 | 1, | |
1877 | ], | |
1878 | [ | |
1879 | 0, | |
1880 | "xababz", | |
1881 | qr/a+(?{$count_a++})b{0,6}(*COMMIT)(*FAIL)/, | |
1882 | qr/a+(?{$count_b++})b{0,6}(*COMMIT)z/, | |
1883 | 1, | |
1884 | ], | |
1885 | [ | |
1886 | 0, | |
1887 | "xabcabcz", | |
1888 | qr/a+(?{$count_a++})(bc){0,6}(*COMMIT)(*FAIL)/, | |
1889 | qr/a+(?{$count_b++})(bc){0,6}(*COMMIT)z/, | |
1890 | 1, | |
1891 | ], | |
1892 | [ | |
1893 | 0, | |
1894 | "xabcabcz", | |
1895 | qr/a+(?{$count_a++})(bc*){0,6}(*COMMIT)(*FAIL)/, | |
1896 | qr/a+(?{$count_b++})(bc*){0,6}(*COMMIT)z/, | |
1897 | 1, | |
1898 | ], | |
1899 | ||
1900 | ||
1901 | [ | |
1902 | 0, | |
1903 | "aaaabtz", | |
1904 | qr/a+(?{$count_a++})b?(*PRUNE)(*FAIL)/, | |
1905 | qr/a+(?{$count_b++})b?(*PRUNE)z/, | |
1906 | 4, | |
1907 | ], | |
1908 | [ | |
1909 | 0, | |
1910 | "aaaabtz", | |
1911 | qr/a+(?{$count_a++})b?(*PRUNE)\s*(*FAIL)/, | |
1912 | qr/a+(?{$count_b++})b?(*PRUNE)\s*z/, | |
1913 | 4, | |
1914 | ], | |
1915 | [ | |
1916 | 0, | |
1917 | "aaaabtz", | |
1918 | qr/a+(?{$count_a++})(?:b|)(*PRUNE)(*FAIL)/, | |
1919 | qr/a+(?{$count_b++})(?:b|)(*PRUNE)z/, | |
1920 | 4, | |
1921 | ], | |
1922 | [ | |
1923 | 0, | |
1924 | "aaaabtz", | |
1925 | qr/a+(?{$count_a++})b{0,6}(*PRUNE)(*FAIL)/, | |
1926 | qr/a+(?{$count_b++})b{0,6}(*PRUNE)z/, | |
1927 | 4, | |
1928 | ], | |
1929 | [ | |
1930 | 0, | |
1931 | "aaaabctz", | |
1932 | qr/a+(?{$count_a++})(bc){0,6}(*PRUNE)(*FAIL)/, | |
1933 | qr/a+(?{$count_b++})(bc){0,6}(*PRUNE)z/, | |
1934 | 4, | |
1935 | ], | |
1936 | [ | |
1937 | 0, | |
1938 | "aaaabctz", | |
1939 | qr/a+(?{$count_a++})(bc*){0,6}(*PRUNE)(*FAIL)/, | |
1940 | qr/a+(?{$count_b++})(bc*){0,6}(*PRUNE)z/, | |
1941 | 4, | |
1942 | ], | |
1943 | ||
1944 | [ | |
1945 | 0, | |
1946 | "aaabaaab", | |
1947 | qr/a+(?{$count_a++;})b?(*SKIP)(*FAIL)/, | |
1948 | qr/a+(?{$count_b++;})b?(*SKIP)z/, | |
1949 | 2, | |
1950 | ], | |
1951 | [ | |
1952 | 0, | |
1953 | "aaabaaab", | |
1954 | qr/a+(?{$count_a++;})b?(*SKIP)\s*(*FAIL)/, | |
1955 | qr/a+(?{$count_b++;})b?(*SKIP)\s*z/, | |
1956 | 2, | |
1957 | ], | |
1958 | [ | |
1959 | 0, | |
1960 | "aaabaaab", | |
1961 | qr/a+(?{$count_a++;})(?:b|)(*SKIP)(*FAIL)/, | |
1962 | qr/a+(?{$count_b++;})(?:b|)(*SKIP)z/, | |
1963 | 2, | |
1964 | ], | |
1965 | [ | |
1966 | 0, | |
1967 | "aaabaaab", | |
1968 | qr/a+(?{$count_a++;})b{0,6}(*SKIP)(*FAIL)/, | |
1969 | qr/a+(?{$count_b++;})b{0,6}(*SKIP)z/, | |
1970 | 2, | |
1971 | ], | |
1972 | [ | |
1973 | 0, | |
1974 | "aaabcaaabc", | |
1975 | qr/a+(?{$count_a++;})(bc){0,6}(*SKIP)(*FAIL)/, | |
1976 | qr/a+(?{$count_b++;})(bc){0,6}(*SKIP)z/, | |
1977 | 2, | |
1978 | ], | |
1979 | [ | |
1980 | 0, | |
1981 | "aaabcaaabc", | |
1982 | qr/a+(?{$count_a++;})(bc*){0,6}(*SKIP)(*FAIL)/, | |
1983 | qr/a+(?{$count_b++;})(bc*){0,6}(*SKIP)z/, | |
1984 | 2, | |
1985 | ], | |
1986 | ||
1987 | ||
1988 | [ | |
1989 | 0, | |
1990 | "aaddbdaabyzc", | |
1991 | qr/a (?{$count_a++;}) (*MARK:T1) (a*) .*? b? (*SKIP:T1) (*FAIL) \s* c \1 /x, | |
1992 | qr/a (?{$count_b++;}) (*MARK:T1) (a*) .*? b? (*SKIP:T1) z \s* c \1 /x, | |
1993 | 4, | |
1994 | ], | |
1995 | [ | |
1996 | 0, | |
1997 | "aaddbdaabyzc", | |
1998 | qr/a (?{$count_a++;}) (*MARK:T1) (a*) .*? b? (*SKIP:T1) \s* (*FAIL) \s* c \1 /x, | |
1999 | qr/a (?{$count_b++;}) (*MARK:T1) (a*) .*? b? (*SKIP:T1) \s* z \s* c \1 /x, | |
2000 | 4, | |
2001 | ], | |
2002 | [ | |
2003 | 0, | |
2004 | "aaddbdaabyzc", | |
2005 | qr/a (?{$count_a++;}) (*MARK:T1) (a*) .*? (?:b|) (*SKIP:T1) (*FAIL) \s* c \1 /x, | |
2006 | qr/a (?{$count_b++;}) (*MARK:T1) (a*) .*? (?:b|) (*SKIP:T1) z \s* c \1 /x, | |
2007 | 4, | |
2008 | ], | |
2009 | [ | |
2010 | 0, | |
2011 | "aaddbdaabyzc", | |
2012 | qr/a (?{$count_a++;}) (*MARK:T1) (a*) .*? b{0,6} (*SKIP:T1) (*FAIL) \s* c \1 /x, | |
2013 | qr/a (?{$count_b++;}) (*MARK:T1) (a*) .*? b{0,6} (*SKIP:T1) z \s* c \1 /x, | |
2014 | 4, | |
2015 | ], | |
2016 | [ | |
2017 | 0, | |
2018 | "aaddbcdaabcyzc", | |
2019 | qr/a (?{$count_a++;}) (*MARK:T1) (a*) .*? (bc){0,6} (*SKIP:T1) (*FAIL) \s* c \1 /x, | |
2020 | qr/a (?{$count_b++;}) (*MARK:T1) (a*) .*? (bc){0,6} (*SKIP:T1) z \s* c \1 /x, | |
2021 | 4, | |
2022 | ], | |
2023 | [ | |
2024 | 0, | |
2025 | "aaddbcdaabcyzc", | |
2026 | qr/a (?{$count_a++;}) (*MARK:T1) (a*) .*? (bc*){0,6} (*SKIP:T1) (*FAIL) \s* c \1 /x, | |
2027 | qr/a (?{$count_b++;}) (*MARK:T1) (a*) .*? (bc*){0,6} (*SKIP:T1) z \s* c \1 /x, | |
2028 | 4, | |
2029 | ], | |
2030 | ||
2031 | ||
2032 | [ | |
2033 | 0, | |
2034 | "aaaaddbdaabyzc", | |
2035 | qr/a (?{$count_a++;}) (a?) (*MARK:T1) (a*) .*? b? (*MARK:T1) (*SKIP:T1) (*FAIL) \s* c \1 /x, | |
2036 | qr/a (?{$count_b++;}) (a?) (*MARK:T1) (a*) .*? b? (*MARK:T1) (*SKIP:T1) z \s* c \1 /x, | |
2037 | 2, | |
2038 | ], | |
2039 | [ | |
2040 | 0, | |
2041 | "aaaaddbdaabyzc", | |
2042 | qr/a (?{$count_a++;}) (a?) (*MARK:T1) (a*) .*? b? (*MARK:T1) (*SKIP:T1) \s* (*FAIL) \s* c \1 /x, | |
2043 | qr/a (?{$count_b++;}) (a?) (*MARK:T1) (a*) .*? b? (*MARK:T1) (*SKIP:T1) \s* z \s* c \1 /x, | |
2044 | 2, | |
2045 | ], | |
2046 | [ | |
2047 | 0, | |
2048 | "aaaaddbdaabyzc", | |
2049 | qr/a (?{$count_a++;}) (a?) (*MARK:T1) (a*) .*? (?:b|) (*MARK:T1) (*SKIP:T1) (*FAIL) \s* c \1 /x, | |
2050 | qr/a (?{$count_b++;}) (a?) (*MARK:T1) (a*) .*? (?:b|) (*MARK:T1) (*SKIP:T1) z \s* c \1 /x, | |
2051 | 2, | |
2052 | ], | |
2053 | [ | |
2054 | 0, | |
2055 | "aaaaddbdaabyzc", | |
2056 | qr/a (?{$count_a++;}) (a?) (*MARK:T1) (a*) .*? b{0,6} (*MARK:T1) (*SKIP:T1) (*FAIL) \s* c \1 /x, | |
2057 | qr/a (?{$count_b++;}) (a?) (*MARK:T1) (a*) .*? b{0,6} (*MARK:T1) (*SKIP:T1) z \s* c \1 /x, | |
2058 | 2, | |
2059 | ], | |
2060 | [ | |
2061 | 0, | |
2062 | "aaaaddbcdaabcyzc", | |
2063 | qr/a (?{$count_a++;}) (a?) (*MARK:T1) (a*) .*? (bc){0,6} (*MARK:T1) (*SKIP:T1) (*FAIL) \s* c \1 /x, | |
2064 | qr/a (?{$count_b++;}) (a?) (*MARK:T1) (a*) .*? (bc){0,6} (*MARK:T1) (*SKIP:T1) z \s* c \1 /x, | |
2065 | 2, | |
2066 | ], | |
2067 | [ | |
2068 | 0, | |
2069 | "aaaaddbcdaabcyzc", | |
2070 | qr/a (?{$count_a++;}) (a?) (*MARK:T1) (a*) .*? (bc*){0,6} (*MARK:T1) (*SKIP:T1) (*FAIL) \s* c \1 /x, | |
2071 | qr/a (?{$count_b++;}) (a?) (*MARK:T1) (a*) .*? (bc*){0,6} (*MARK:T1) (*SKIP:T1) z \s* c \1 /x, | |
2072 | 2, | |
2073 | ], | |
2074 | ||
2075 | ||
2076 | [ | |
2077 | 0, | |
2078 | "AbcdCBefgBhiBqz", | |
2079 | qr/(A (.*) (?{ $count_a++ }) C? (*THEN) | A D) (*FAIL)/x, | |
2080 | qr/(A (.*) (?{ $count_b++ }) C? (*THEN) | A D) z/x, | |
2081 | 1, | |
2082 | ], | |
2083 | [ | |
2084 | 0, | |
2085 | "AbcdCBefgBhiBqz", | |
2086 | qr/(A (.*) (?{ $count_a++ }) C? (*THEN) | A D) \s* (*FAIL)/x, | |
2087 | qr/(A (.*) (?{ $count_b++ }) C? (*THEN) | A D) \s* z/x, | |
2088 | 1, | |
2089 | ], | |
2090 | [ | |
2091 | 0, | |
2092 | "AbcdCBefgBhiBqz", | |
2093 | qr/(A (.*) (?{ $count_a++ }) (?:C|) (*THEN) | A D) (*FAIL)/x, | |
2094 | qr/(A (.*) (?{ $count_b++ }) (?:C|) (*THEN) | A D) z/x, | |
2095 | 1, | |
2096 | ], | |
2097 | [ | |
2098 | 0, | |
2099 | "AbcdCBefgBhiBqz", | |
2100 | qr/(A (.*) (?{ $count_a++ }) C{0,6} (*THEN) | A D) (*FAIL)/x, | |
2101 | qr/(A (.*) (?{ $count_b++ }) C{0,6} (*THEN) | A D) z/x, | |
2102 | 1, | |
2103 | ], | |
2104 | [ | |
2105 | 0, | |
2106 | "AbcdCEBefgBhiBqz", | |
2107 | qr/(A (.*) (?{ $count_a++ }) (CE){0,6} (*THEN) | A D) (*FAIL)/x, | |
2108 | qr/(A (.*) (?{ $count_b++ }) (CE){0,6} (*THEN) | A D) z/x, | |
2109 | 1, | |
2110 | ], | |
2111 | [ | |
2112 | 0, | |
2113 | "AbcdCBefgBhiBqz", | |
2114 | qr/(A (.*) (?{ $count_a++ }) (CE*){0,6} (*THEN) | A D) (*FAIL)/x, | |
2115 | qr/(A (.*) (?{ $count_b++ }) (CE*){0,6} (*THEN) | A D) z/x, | |
2116 | 1, | |
2117 | ], | |
2118 | ) { | |
2119 | $c++; | |
2120 | $count_a = 0; | |
2121 | $count_b = 0; | |
2122 | ||
2123 | my $match_a = ($re->[1] =~ $re->[2]) || 0; | |
2124 | my $match_b = ($re->[1] =~ $re->[3]) || 0; | |
2125 | ||
de26e0cc NC |
2126 | is($match_a, $re->[0], "match a " . ($re->[0] ? "succeeded" : "failed") . " ($c)"); |
2127 | is($match_b, $re->[0], "match b " . ($re->[0] ? "succeeded" : "failed") . " ($c)"); | |
2128 | is($count_a, $re->[4], "count a ($c)"); | |
2129 | is($count_b, $re->[4], "count b ($c)"); | |
d1c771f5 B |
2130 | } |
2131 | } | |
e425a60b | 2132 | |
b57e4118 KW |
2133 | { # Bleadperl v5.13.8-292-gf56b639 breaks NEZUMI/Unicode-LineBreak-1.011 |
2134 | # \xdf in lookbehind failed to compile as is multi-char fold | |
14358a41 NC |
2135 | my $message = "Lookbehind with \\xdf matchable compiles"; |
2136 | my $r = eval 'qr{ | |
b57e4118 KW |
2137 | (?u: (?<=^url:) | |
2138 | (?<=[/]) (?=[^/]) | | |
2139 | (?<=[^-.]) (?=[-~.,_?\#%=&]) | | |
2140 | (?<=[=&]) (?=.) | |
14358a41 NC |
2141 | )}iox'; |
2142 | is($@, '', $message); | |
bbce3ca6 | 2143 | object_ok($r, 'Regexp', $message); |
b57e4118 KW |
2144 | } |
2145 | ||
84193928 | 2146 | # RT #82610 |
8f46dbea | 2147 | like 'foo/file.fob', qr,^(?=[^\.])[^/]*/(?=[^\.])[^/]*\.fo[^/]$,; |
84193928 | 2148 | |
295c2f7d | 2149 | { # This was failing unless an explicit /d was added |
f36cc39d | 2150 | my $E0 = uni_to_native("\xE0"); |
b6990aaa | 2151 | utf8::upgrade($E0); |
f36cc39d | 2152 | my $p = qr/[_$E0]/i; |
f36cc39d | 2153 | like(uni_to_native("\xC0"), qr/$p/, "Verify \"\\xC0\" =~ /[\\xE0_]/i; pattern in utf8"); |
295c2f7d KW |
2154 | } |
2155 | ||
b7ad6c3e | 2156 | like "x", qr/\A(?>(?:(?:)A|B|C?x))\z/, |
a40630bf | 2157 | "Check TRIE does not overwrite EXACT following NOTHING at start - RT #111842"; |
c221a219 | 2158 | |
26faadbd | 2159 | { |
9e3a0e9b | 2160 | my $single = "z"; |
26faadbd KW |
2161 | my $upper = "\x{390}"; # Fold is 3 chars. |
2162 | my $multi = CORE::fc($upper); | |
2163 | ||
2164 | my $failed = 0; | |
2165 | ||
2166 | # Try forcing a node to be split, with a multi-char fold at the | |
2167 | # boundary | |
2168 | for my $repeat (1 .. 300) { | |
2169 | my $string = $single x $repeat; | |
2170 | my $lhs = $string . $upper; | |
2171 | if ($lhs !~ m/$string$multi/i) { | |
2172 | $failed = $repeat; | |
2173 | last; | |
2174 | } | |
2175 | } | |
2176 | ok(! $failed, "Matched multi-char fold across EXACTFish node boundaries; if failed, was at count $failed"); | |
2177 | ||
b9c48b5b KW |
2178 | $failed = 0; |
2179 | for my $repeat (1 .. 300) { | |
2180 | my $string = $single x $repeat; | |
2181 | my $lhs = $string . "\N{LATIN SMALL LIGATURE FFI}"; | |
2182 | if ($lhs !~ m/${string}ff\N{LATIN SMALL LETTER I}/i) { | |
2183 | $failed = $repeat; | |
2184 | last; | |
2185 | } | |
2186 | } | |
2187 | ok(! $failed, "Matched multi-char fold across EXACTFish node boundaries; if failed, was at count $failed"); | |
2188 | ||
2189 | $failed = 0; | |
2190 | for my $repeat (1 .. 300) { | |
2191 | my $string = $single x $repeat; | |
2192 | my $lhs = $string . "\N{LATIN SMALL LIGATURE FFL}"; | |
2193 | if ($lhs !~ m/${string}ff\N{U+6c}/i) { | |
2194 | $failed = $repeat; | |
2195 | last; | |
2196 | } | |
2197 | } | |
2198 | ok(! $failed, "Matched multi-char fold across EXACTFish node boundaries; if failed, was at count $failed"); | |
88b3a463 KW |
2199 | |
2200 | # This tests that under /d matching that an 'ss' split across two | |
2201 | # parts of a node doesn't end up turning into something that matches | |
2202 | # \xDF unless it is in utf8. | |
2203 | $failed = 0; | |
2204 | $single = 'a'; # Is non-terminal multi-char fold char | |
2205 | for my $repeat (1 .. 300) { | |
2206 | my $string = $single x $repeat; | |
2207 | my $lhs = "$string\N{LATIN SMALL LETTER SHARP S}"; | |
2208 | utf8::downgrade($lhs); | |
2209 | $string .= "s"; | |
2210 | if ($lhs =~ m/${string}s/di) { | |
2211 | $failed = $repeat; | |
2212 | last; | |
2213 | } | |
2214 | } | |
2215 | ok(! $failed, "Matched multi-char fold 'ss' across EXACTF node boundaries; if failed, was at count $failed"); | |
73976260 KW |
2216 | |
2217 | for my $non_finals ("t", "ft", "ift", "sift") { | |
2218 | my $base_pat = $non_finals . "enKalt"; # (The tail is taken from | |
2219 | # the trouble ticket, is | |
2220 | # arbitrary) | |
2221 | for my $utf8 ("non-UTF-8", "UTF-8") { | |
2222 | ||
2223 | # Try at different lengths to be sure to get a node boundary | |
2224 | for my $repeat (120 .. 270) { # [perl #133756] | |
2225 | my $head = ("b" x $repeat) . "\xDC"; | |
2226 | my $pat = $base_pat; | |
2227 | utf8::upgrade($pat) if $utf8 eq "UTF-8"; | |
2228 | $pat = $head . $pat; | |
2229 | my $text = $head . $base_pat; | |
2230 | ||
2231 | if ($text !~ /$pat/i) { | |
2232 | $failed = $repeat; | |
2233 | last; | |
2234 | } | |
2235 | } | |
2236 | ||
2237 | ok(! $failed, "A non-final fold character " | |
2238 | . (length($non_finals) - 1) | |
2239 | . " characters from the end of an EXACTFish" | |
2240 | . " $utf8 pattern works; if failed, was at count $failed"); | |
2241 | } | |
2242 | } | |
26faadbd KW |
2243 | } |
2244 | ||
2b233a8f | 2245 | { |
2b233a8f KW |
2246 | fresh_perl_is('print eval "\"\x{101}\" =~ /[[:lower:]]/", "\n"; print eval "\"\x{100}\" =~ /[[:lower:]]/i", "\n";', |
2247 | "1\n1", # Both re's should match | |
20e5bab4 | 2248 | {}, |
2b233a8f KW |
2249 | "get [:lower:] swash in first eval; test under /i in second"); |
2250 | } | |
2251 | ||
c5de0829 | 2252 | { |
3553f4fa DM |
2253 | fresh_perl_is(<<'EOF', |
2254 | my $s = "\x{41c}"; | |
2255 | $s =~ /(.*)/ or die; | |
2256 | $ls = lc $1; | |
2257 | print $ls eq lc $s ? "good\n" : "bad: [$ls]\n"; | |
2258 | EOF | |
2259 | "good\n", | |
2260 | {}, | |
2261 | "swash triggered by lc() doesn't corrupt \$1" | |
2262 | ); | |
2263 | } | |
2264 | ||
2265 | { | |
c5de0829 | 2266 | #' RT #119075 |
fd2268bc | 2267 | no warnings 'regexp'; # Silence "has useless greediness modifier" |
c5de0829 JK |
2268 | local $@; |
2269 | eval { /a{0}?/; }; | |
2270 | ok(! $@, | |
2271 | "PCRE regression test: No 'Quantifier follows nothing in regex' warning"); | |
2272 | ||
2273 | } | |
2274 | ||
709d08c2 KW |
2275 | { |
2276 | unlike("\xB5", qr/^_?\p{IsMyRuntimeProperty}\z/, "yadayada"); | |
2277 | like("\xB6", qr/^_?\p{IsMyRuntimeProperty}\z/, "yadayada"); | |
2278 | unlike("\xB7", qr/^_?\p{IsMyRuntimeProperty}\z/, "yadayada"); | |
2279 | like("\xB5", qr/^_?\P{IsMyRuntimeProperty}\z/, "yadayada"); | |
2280 | unlike("\xB6", qr/^_?\P{IsMyRuntimeProperty}\z/, "yadayada"); | |
2281 | like("\xB7", qr/^_?\P{IsMyRuntimeProperty}\z/, "yadayada"); | |
2282 | ||
2283 | unlike("_\xB5", qr/^_?\p{IsMyRuntimeProperty}\z/, "yadayada"); | |
2284 | like("_\xB6", qr/^_?\p{IsMyRuntimeProperty}\z/, "yadayada"); | |
2285 | unlike("_\xB7", qr/^_?\p{IsMyRuntimeProperty}\z/, "yadayada"); | |
2286 | like("_\xB5", qr/^_?\P{IsMyRuntimeProperty}\z/, "yadayada"); | |
2287 | unlike("_\xB6", qr/^_?\P{IsMyRuntimeProperty}\z/, "yadayada"); | |
2288 | like("_\xB7", qr/^_?\P{IsMyRuntimeProperty}\z/, "yadayada"); | |
2289 | } | |
2290 | ||
2291 | # These are defined later, so won't be known at regex compile time above | |
2292 | sub IsMyRuntimeProperty { | |
2293 | return "B6\n"; | |
2294 | } | |
2295 | ||
2296 | sub IsntMyRuntimeProperty { | |
2297 | return "!B6\n"; | |
2298 | } | |
2299 | ||
80836a6e KW |
2300 | { # [perl 121777] |
2301 | my $regex; | |
2302 | { package Some; | |
2303 | # define a Unicode propertyIs_q | |
2304 | sub Is_q | |
2305 | { | |
2306 | sprintf '%x', ord 'q' | |
2307 | } | |
2308 | $regex = qr/\p{Is_q}/; | |
2309 | ||
2310 | # If we uncomment the following line, prior to the patch that | |
de68ac66 KW |
2311 | # fixed this, everything would work because we would have expanded |
2312 | # the property by the time the regex in the 'like' below got | |
2313 | # compiled. | |
80836a6e KW |
2314 | #'q' =~ $regex; |
2315 | } | |
2316 | ||
2317 | like('q', $regex, 'User-defined property matches outside package'); | |
c0611711 KW |
2318 | |
2319 | package Some { | |
2320 | main::like('abcq', qr/abc$regex/, 'Run-time compiled in-package user-defined property matches'); | |
2321 | } | |
80836a6e KW |
2322 | } |
2323 | ||
bc031a7d KW |
2324 | { # From Lingua::Stem::UniNE; no ticket filed but related to #121778 |
2325 | use utf8; | |
2326 | my $word = 'рабта'; | |
2327 | $word =~ s{ (?: | |
2328 | ия # definite articles for nouns: | |
2329 | | ът # ∙ masculine | |
2330 | | та # ∙ feminine | |
2331 | | то # ∙ neutral | |
2332 | | те # ∙ plural | |
2333 | ) $ }{}x; | |
2334 | is($word, 'раб', "Handles UTF8 trie correctly"); | |
2335 | } | |
2336 | ||
842a9d21 KW |
2337 | { # [perl #122460] |
2338 | my $a = "rdvark"; | |
2339 | $a =~ /(?{})(?=[A-Za-z0-9_])a*?/g; | |
2340 | is (pos $a, 0, "optimizer correctly thinks (?=...) is 0-length"); | |
2341 | } | |
2342 | ||
4cbd7e22 FC |
2343 | { # [perl #123417] multi-char \N{...} tripping roundly |
2344 | use Cname; | |
2345 | my $qr = qr$(\N{foo})$; | |
2346 | "afoot" =~ eval "qr/$qr/"; | |
2347 | is "$1" || $@, "foo", 'multichar \N{...} stringified and retoked'; | |
2348 | } | |
2349 | ||
7195e5da KW |
2350 | is (scalar split(/\b{sb}/, "Don't think twice. It's all right."), |
2351 | 2, '\b{wb} splits sentences correctly'); | |
2352 | ||
39ce401c KW |
2353 | ok "my/dir/audio_07.mp3" =~ |
2354 | qr/(.*)\/(.*)\/(.*)\.(?<=(?=(?:\.(?!\d+\b)\w{1,4}$)$)\.)(.*)$()/, | |
2355 | "[perl #133948]"; | |
2356 | ||
4cbd7e22 | 2357 | |
de085464 KW |
2358 | # !!! NOTE! Keep the following tests last -- they may crash perl |
2359 | ||
e425a60b | 2360 | print "# Tests that follow may crash perl\n"; |
e425a60b YO |
2361 | { |
2362 | eval '/\k/'; | |
b7ad6c3e | 2363 | like $@, qr/\QSequence \k... not terminated in regex;\E/, |
e425a60b YO |
2364 | 'Lone \k not allowed'; |
2365 | } | |
2366 | ||
e425a60b | 2367 | { |
de946258 | 2368 | my $message = "Substitution with lookahead (possible segv)"; |
e425a60b YO |
2369 | $_ = "ns1ns1ns1"; |
2370 | s/ns(?=\d)/ns_/g; | |
de946258 | 2371 | is($_, "ns_1ns_1ns_1", $message); |
e425a60b YO |
2372 | $_ = "ns1"; |
2373 | s/ns(?=\d)/ns_/; | |
de946258 | 2374 | is($_, "ns_1", $message); |
e425a60b YO |
2375 | $_ = "123"; |
2376 | s/(?=\d+)|(?<=\d)/!Bang!/g; | |
de946258 | 2377 | is($_, "!Bang!1!Bang!2!Bang!3!Bang!", $message); |
e425a60b YO |
2378 | } |
2379 | ||
6182169b KW |
2380 | { |
2381 | # Earlier versions of Perl said this was fatal. | |
de946258 | 2382 | my $message = "U+0FFFF shouldn't crash the regex engine"; |
6182169b KW |
2383 | no warnings 'utf8'; |
2384 | my $a = eval "chr(65535)"; | |
2385 | use warnings; | |
2386 | my $warning_message; | |
2387 | local $SIG{__WARN__} = sub { $warning_message = $_[0] }; | |
2388 | eval $a =~ /[a-z]/; | |
de946258 | 2389 | ok(1, $message); # If it didn't crash, it worked. |
6182169b | 2390 | } |
b57e4118 | 2391 | |
91d110e0 | 2392 | { # Was looping |
ec8902ce | 2393 | watchdog(10); |
6438af90 | 2394 | like("\x{00DF}", qr/[\x{1E9E}_]*/i, "\"\\x{00DF}\" =~ /[\\x{1E9E}_]*/i was looping"); |
d44172c6 | 2395 | watchdog(0); |
6438af90 KW |
2396 | } |
2397 | ||
f9126265 KW |
2398 | { # Bug #90536, caused failed assertion |
2399 | unlike("s\N{U+DF}", qr/^\x{00DF}/i, "\"s\\N{U+DF}\", qr/^\\x{00DF}/i"); | |
2400 | } | |
2401 | ||
361ee0fe | 2402 | # User-defined Unicode properties to match above-Unicode code points |
56a85032 | 2403 | sub Is_31_Bit_Super { return "110000\t7FFFFFFF\n" } |
361ee0fe | 2404 | sub Is_Portable_Super { return '!utf8::Any' } # Matches beyond 32 bits |
45d91b83 KW |
2405 | |
2406 | { # Assertion was failing on on 64-bit platforms; just didn't work on 32. | |
2407 | no warnings qw(non_unicode portable); | |
2408 | use Config; | |
2409 | ||
2410 | # We use 'ok' instead of 'like' because the warnings are lexically | |
2411 | # scoped, and want to turn them off, so have to do the match in this | |
c0236afe | 2412 | # scope. |
76513bdc | 2413 | if ($Config{uvsize} < 8) { |
8f46dbea | 2414 | like(chr(0x7FFF_FFFE), qr/\p{Is_31_Bit_Super}/, |
56a85032 | 2415 | "chr(0x7FFF_FFFE) can match a Unicode property"); |
8f46dbea | 2416 | like(chr(0x7FFF_FFFF), qr/\p{Is_31_Bit_Super}/, |
56a85032 KW |
2417 | "chr(0x7FFF_FFFF) can match a Unicode property"); |
2418 | my $p = qr/^[\x{7FFF_FFFF}]$/; | |
8f46dbea | 2419 | like(chr(0x7FFF_FFFF), qr/$p/, |
56a85032 | 2420 | "chr(0x7FFF_FFFF) can match itself in a [class]"); |
8f46dbea | 2421 | like(chr(0x7FFF_FFFF), qr/$p/, # Tests any caching |
56a85032 | 2422 | "chr(0x7FFF_FFFF) can match itself in a [class] subsequently"); |
84ea5ef6 | 2423 | } |
76513bdc KW |
2424 | else { |
2425 | no warnings 'overflow'; | |
8f46dbea | 2426 | like(chr(0x7FFF_FFFF_FFFF_FFFE), qr/\p{Is_Portable_Super}/, |
56a85032 | 2427 | "chr(0x7FFF_FFFF_FFFF_FFFE) can match a Unicode property"); |
8f46dbea | 2428 | like(chr(0x7FFF_FFFF_FFFF_FFFF), qr/^\p{Is_Portable_Super}$/, |
56a85032 | 2429 | "chr(0x7FFF_FFFF_FFFF_FFFF) can match a Unicode property"); |
76513bdc | 2430 | |
1a9f5e2a | 2431 | my $p = eval 'qr/^\x{7FFF_FFFF_FFFF_FFFF}$/'; |
8f46dbea | 2432 | like(chr(0x7FFF_FFFF_FFFF_FFFF), qr/$p/, |
56a85032 | 2433 | "chr(0x7FFF_FFFF_FFFF_FFFF) can match itself in a [class]"); |
8f46dbea | 2434 | like(chr(0x7FFF_FFFF_FFFF_FFFF), qr/$p/, # Tests any caching |
56a85032 | 2435 | "chr(0x7FFF_FFFF_FFFF_FFFF) can match itself in a [class] subsequently"); |
76513bdc KW |
2436 | |
2437 | # This test is because something was declared as 32 bits, but | |
2438 | # should have been cast to 64; only a problem where | |
2439 | # sizeof(STRLEN) != sizeof(UV) | |
8f46dbea KW |
2440 | unlike(chr(0x7FFF_FFFF_FFFF_FFFE), qr/\p{Is_31_Bit_Super}/, |
2441 | "chr(0x7FFF_FFFF_FFFF_FFFE) shouldn't match a range ending in 0x7FFF_FFFF"); | |
76513bdc | 2442 | } |
45d91b83 KW |
2443 | } |
2444 | ||
9d501133 KW |
2445 | { # [perl #112530], the code below caused a panic |
2446 | sub InFoo { "a\tb\n9\ta\n" } | |
f36cc39d | 2447 | like(chr(0xA), qr/\p{InFoo}/, |
9d501133 KW |
2448 | "Overlapping ranges in user-defined properties"); |
2449 | } | |
2450 | ||
512c0f5a KW |
2451 | { # [perl #125990], the final 2 tests below each caused a panic. |
2452 | # The \0's are not necessary; it could be a printable character | |
2453 | # instead, but were in the ticket, so using them. | |
2454 | my $sharp_s = chr utf8::unicode_to_native(0xdf); | |
2455 | my $string = ("\0" x 8) | |
2456 | . ($sharp_s x 3) | |
2457 | . ("\0" x 42) | |
2458 | . "ý"; | |
2459 | my $folded_string = ("\0" x 8) | |
2460 | . ("ss" x 3) | |
2461 | . ("\0" x 42) | |
2462 | . "ý"; | |
2463 | utf8::downgrade($string); | |
2464 | utf8::downgrade($folded_string); | |
2465 | ||
37ee5518 | 2466 | use Cname; |
512c0f5a KW |
2467 | like($string, qr/$string/i, "LATIN SMALL SHARP S matches itself under /id"); |
2468 | unlike($folded_string, qr/$string/i, "LATIN SMALL SHARP S doesn't match 'ss' under /di"); | |
37ee5518 KW |
2469 | like($folded_string, qr/\N{EMPTY-STR}$string/i, "\\N{} earlier than LATIN SMALL SHARP S transforms /di into /ui, matches 'ss'"); |
2470 | like($folded_string, qr/$string\N{EMPTY-STR}/i, "\\N{} after LATIN SMALL SHARP S transforms /di into /ui, matches 'ss'"); | |
512c0f5a KW |
2471 | } |
2472 | ||
90c17285 | 2473 | { # [perl #126606 crashed the interpreter |
37ee5518 KW |
2474 | use Cname; |
2475 | like("sS", qr/\N{EMPTY-STR}Ss|/i, '\N{} with empty branch alternation works'); | |
945961fd | 2476 | like("sS", qr'\N{EMPTY-STR}Ss|'i, '\N{} with empty branch alternation works'); |
90c17285 KW |
2477 | } |
2478 | ||
2e3a23da KW |
2479 | { # Regexp:Grammars was broken: |
2480 | # http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/2013-06/msg01290.html | |
2481 | fresh_perl_like('use warnings; "abc" =~ qr{(?&foo){0}abc(?<foo>)}', | |
aaa63dae | 2482 | qr/Quantifier unexpected on zero-length expression/, |
20e5bab4 | 2483 | {}, |
2e3a23da KW |
2484 | 'No segfault on qr{(?&foo){0}abc(?<foo>)}'); |
2485 | } | |
2486 | ||
668fcfea TC |
2487 | SKIP: |
2488 | { # [perl #125826] buffer overflow in TRIE_STORE_REVCHAR | |
2489 | # (during compilation, so use a fresh perl) | |
2490 | $Config{uvsize} == 8 | |
2491 | or skip("need large code-points for this test", 1); | |
760c7c2f | 2492 | |
d22ec717 | 2493 | fresh_perl_is('/\x{E000000000}|/ and print qq(ok\n)', "ok\n", {}, |
668fcfea TC |
2494 | "buffer overflow in TRIE_STORE_REVCHAR"); |
2495 | } | |
2496 | ||
8571c4b0 KW |
2497 | { |
2498 | fresh_perl_like('use warnings; s\00(?(?!00000000000000000000000000·000000)\500000000\00000000000000000000000000000000000000000000000000000·00000000000000000000000000000000\00', | |
2499 | qr/Switch \(\?\(condition\)\.\.\. not terminated/, | |
2500 | {}, | |
2501 | 'No segfault [perl #126886]'); | |
2502 | } | |
2503 | ||
553fa53c JK |
2504 | { |
2505 | # [perl 130010] Downstream application texinfo started to report panics | |
2506 | # as of commit a5540cf. | |
2507 | ||
2508 | runperl( prog => 'A::xx(); package A; sub InFullwidth{ return qq|\n| } sub xx { split /[^\s\p{InFullwidth}]/, q|x| }' ); | |
2509 | ok(! $?, "User-defined pattern did not cause panic [perl 130010]"); | |
2510 | } | |
2511 | ||
b92b2705 KW |
2512 | { # [perl #133999] Previously assertion failure |
2513 | fresh_perl_like('0 =~ /\p{nv:(\B(*COMMIT)C+)}/', | |
2514 | qr/No Unicode property value wildcard matches/, | |
2515 | {}, | |
2516 | "Assertion failure with *COMMIT and wildcard property"); | |
2517 | } | |
2518 | ||
11b48e76 KW |
2519 | { # [perl #134029] Previously assertion failure |
2520 | fresh_perl_like('qr/\p{upper:]}|\337(?|ss)|)(?0/', | |
2521 | qr/Unicode property wildcard not terminated/, | |
2522 | {}, | |
2523 | "Assertion failure with single character wildcard"); | |
2524 | } | |
2525 | ||
8c9c2723 KW |
2526 | { # [perl #134034] Previously assertion failure |
2527 | fresh_perl_is('use utf8; q!Ȧिम한글💣ყაოსაა!=~/(?li)\b{wb}\B(*COMMIT)0/;', | |
2528 | "", {}, "*COMMIT caused positioning beyond EOS"); | |
2529 | } | |
2530 | ||
d013a11f KW |
2531 | { # [GH #17486] Previously assertion failure |
2532 | fresh_perl_is('0=~/(?iaa)ss\337(?0)|/', | |
2533 | "", {}, "EXACTFUP node isn't changed into something else"); | |
2534 | } | |
2535 | ||
daa74010 KW |
2536 | { # [GH #17593] |
2537 | fresh_perl_is('qr/((?+2147483647))/', | |
2538 | "Invalid reference to group in regex; marked by <--" | |
2539 | . " HERE in m/((?+2147483647) <-- HERE )/ at - line 1.", | |
2540 | {}, "integer overflow, undefined behavior in ASAN"); | |
99d2ace5 KW |
2541 | fresh_perl_is('qr/((?-2147483647))/', |
2542 | "Reference to nonexistent group in regex; marked by <--" | |
2543 | . " HERE in m/((?-2147483647) <-- HERE )/ at - line 1.", | |
2544 | {}, "Large negative relative capture group"); | |
5c1c42cb KW |
2545 | fresh_perl_is('qr/((?+18446744073709551615))/', |
2546 | "Invalid reference to group in regex; marked by <--" | |
2547 | . " HERE in m/((?+18446744073709551615 <-- HERE ))/ at -" | |
2548 | . " line 1.", | |
2549 | {}, "Too large relative group number"); | |
2550 | fresh_perl_is('qr/((?-18446744073709551615))/', | |
2551 | "Invalid reference to group in regex; marked by <--" | |
2552 | . " HERE in m/((?-18446744073709551615 <-- HERE ))/ at -" | |
2553 | . " line 1.", | |
2554 | {}, "Too large negative relative group number"); | |
99d2ace5 KW |
2555 | } |
2556 | ||
07b23240 KW |
2557 | { # GH #17734, ASAN use after free |
2558 | fresh_perl_like('no warnings "experimental::uniprop_wildcards"; | |
2559 | my $re = q<[[\p{name=/[Y-]+Z/}]]>; | |
2560 | eval { "\N{BYZANTINE MUSICAL SYMBOL PSILI}" | |
2561 | =~ /$re/ }; print $@ if $@; print "Done\n";', | |
2562 | qr/Done/, | |
2563 | {}, "GH #17734"); | |
2564 | } | |
2565 | ||
bb58640a KW |
2566 | { # GH $17278 assertion fails |
2567 | fresh_perl_is('use locale; | |
2568 | my $A_grave = "\N{LATIN CAPITAL LETTER A WITH GRAVE}"; | |
bb58640a | 2569 | my $a_grave = "\N{LATIN SMALL LETTER A WITH GRAVE}"; |
bb58640a KW |
2570 | |
2571 | my $z="q!$a_grave! =~ m!(?^i)[$A_grave]!"; | |
bb58640a KW |
2572 | print eval $z, "\n";', |
2573 | 1, | |
2574 | {}, "GH #17278"); | |
2575 | } | |
33fc7377 YO |
2576 | |
2577 | for my $try ( 1 .. 10 ) { | |
2578 | # GH $19350 assertion fails - run 10 times as this bug is a heisenbug | |
2579 | # and does not always fail, but should fail at least once in 10 tries. | |
2580 | fresh_perl_is('use re Debug=>"ALL";qr{(?{a})(?<b>\g{c}})', | |
2581 | <<'EOF_DEBUG_OUT', | |
2582 | Assembling pattern from 2 elements | |
2583 | Compiling REx "(?{a})(?<b>\g{c}" | |
2584 | Starting parse and generation | |
2585 | <(?{a})(?<b>>...| 1| reg | |
2586 | | | brnc | |
2587 | | | piec | |
2588 | | | atom | |
2589 | <?{a})(?<b>\>...| | reg | |
2590 | <(?<b>\g{c}> | 4| piec | |
2591 | | | atom | |
2592 | <?<b>\g{c}> | | reg | |
2593 | | | Setting open paren #1 to 4 | |
2594 | <\g{c}> | 6| brnc | |
2595 | | | piec | |
2596 | | | atom | |
d78630f1 YO |
2597 | <> | 9| tail~ OPEN1 'b' (4) -> REFN |
2598 | | | Setting close paren #1 to 9 | |
2599 | | 11| lsbr~ tying lastbr REFN <1> (6) to ender CLOSE1 'b' (9) offset 3 | |
2600 | | | tail~ REFN <1> (6) -> CLOSE | |
33fc7377 YO |
2601 | Unmatched ( in regex; marked by <-- HERE in m/(?{a})( <-- HERE ?<b>\g{c}/ at - line 1. |
2602 | Freeing REx: "(?{a})(?<b>\g{c}" | |
2603 | EOF_DEBUG_OUT | |
fe5492d9 YO |
2604 | {rtrim_result=>1}, |
2605 | "Github Issue #19350, assert fail in " | |
33fc7377 YO |
2606 | . "Debug => 'ALL' from malformed qr// (heisenbug try $try)"); |
2607 | } | |
2608 | { # Related to GH $19350 but segfaults instead of asserts, and does so reliably, not randomly. | |
2609 | # use re Debug => "PARSE" is similar to "ALL", but does not include the optimize info, so we | |
2610 | # do not need to deal with normlazing memory addresses in the output. | |
2611 | fresh_perl_is( | |
2612 | 'use re Debug=>"PARSE";qr{(?<b>\g{c})(?<c>x)(?&b)}', | |
2613 | <<'EOF_DEBUG_OUT', | |
2614 | Assembling pattern from 1 elements | |
2615 | Compiling REx "(?<b>\g{c})(?<c>x)(?&b)" | |
2616 | Starting parse and generation | |
2617 | <(?<b>\g{c})>...| 1| reg | |
2618 | | | brnc | |
2619 | | | piec | |
2620 | | | atom | |
2621 | <?<b>\g{c})(>...| | reg | |
2622 | <\g{c})(?<c>>...| 3| brnc | |
2623 | | | piec | |
2624 | | | atom | |
d78630f1 YO |
2625 | <)(?<c>x)(?&b)> | 6| tail~ OPEN1 'b' (1) -> REFN |
2626 | | 8| lsbr~ tying lastbr REFN <1> (3) to ender CLOSE1 'b' (6) offset 3 | |
2627 | | | tail~ REFN <1> (3) -> CLOSE | |
33fc7377 YO |
2628 | <(?<c>x)(?&b)> | | piec |
2629 | | | atom | |
2630 | <?<c>x)(?&b)> | | reg | |
d78630f1 | 2631 | <x)(?&b)> | 10| brnc |
33fc7377 YO |
2632 | | | piec |
2633 | | | atom | |
d78630f1 YO |
2634 | <)(?&b)> | 12| tail~ OPEN2 'c' (8) -> EXACT |
2635 | | 14| lsbr~ tying lastbr EXACT <x> (10) to ender CLOSE2 'c' (12) offset 2 | |
2636 | | | tail~ EXACT <x> (10) -> CLOSE | |
33fc7377 | 2637 | <(?&b)> | | tail~ OPEN1 'b' (1) |
d78630f1 YO |
2638 | | | ~ REFN <1> (3) |
2639 | | | ~ CLOSE1 'b' (6) -> OPEN | |
33fc7377 YO |
2640 | | | piec |
2641 | | | atom | |
2642 | <?&b)> | | reg | |
d78630f1 YO |
2643 | <> | 17| tail~ OPEN2 'c' (8) |
2644 | | | ~ EXACT <x> (10) | |
2645 | | | ~ CLOSE2 'c' (12) -> GOSUB | |
2646 | | 18| lsbr~ tying lastbr OPEN1 'b' (1) to ender END (17) offset 16 | |
33fc7377 | 2647 | | | tail~ OPEN1 'b' (1) |
d78630f1 YO |
2648 | | | ~ REFN <1> (3) |
2649 | | | ~ CLOSE1 'b' (6) | |
2650 | | | ~ OPEN2 'c' (8) | |
2651 | | | ~ EXACT <x> (10) | |
2652 | | | ~ CLOSE2 'c' (12) | |
2653 | | | ~ GOSUB1[+0:14] 'b' (14) -> END | |
33fc7377 YO |
2654 | Need to redo parse |
2655 | Freeing REx: "(?<b>\g{c})(?<c>x)(?&b)" | |
2656 | Starting parse and generation | |
2657 | <(?<b>\g{c})>...| 1| reg | |
2658 | | | brnc | |
2659 | | | piec | |
2660 | | | atom | |
2661 | <?<b>\g{c})(>...| | reg | |
2662 | <\g{c})(?<c>>...| 3| brnc | |
2663 | | | piec | |
2664 | | | atom | |
d78630f1 YO |
2665 | <)(?<c>x)(?&b)> | 6| tail~ OPEN1 'b' (1) -> REFN |
2666 | | 8| lsbr~ tying lastbr REFN2 'c' <1> (3) to ender CLOSE1 'b' (6) offset 3 | |
2667 | | | tail~ REFN2 'c' <1> (3) -> CLOSE | |
33fc7377 YO |
2668 | <(?<c>x)(?&b)> | | piec |
2669 | | | atom | |
2670 | <?<c>x)(?&b)> | | reg | |
d78630f1 | 2671 | <x)(?&b)> | 10| brnc |
33fc7377 YO |
2672 | | | piec |
2673 | | | atom | |
d78630f1 YO |
2674 | <)(?&b)> | 12| tail~ OPEN2 'c' (8) -> EXACT |
2675 | | 14| lsbr~ tying lastbr EXACT <x> (10) to ender CLOSE2 'c' (12) offset 2 | |
2676 | | | tail~ EXACT <x> (10) -> CLOSE | |
33fc7377 | 2677 | <(?&b)> | | tail~ OPEN1 'b' (1) |
d78630f1 YO |
2678 | | | ~ REFN2 'c' <1> (3) |
2679 | | | ~ CLOSE1 'b' (6) -> OPEN | |
33fc7377 YO |
2680 | | | piec |
2681 | | | atom | |
2682 | <?&b)> | | reg | |
d78630f1 YO |
2683 | <> | 17| tail~ OPEN2 'c' (8) |
2684 | | | ~ EXACT <x> (10) | |
2685 | | | ~ CLOSE2 'c' (12) -> GOSUB | |
2686 | | 18| lsbr~ tying lastbr OPEN1 'b' (1) to ender END (17) offset 16 | |
33fc7377 | 2687 | | | tail~ OPEN1 'b' (1) |
d78630f1 YO |
2688 | | | ~ REFN2 'c' <1> (3) |
2689 | | | ~ CLOSE1 'b' (6) | |
2690 | | | ~ OPEN2 'c' (8) | |
2691 | | | ~ EXACT <x> (10) | |
2692 | | | ~ CLOSE2 'c' (12) | |
2693 | | | ~ GOSUB1[+0:14] 'b' (14) -> END | |
2694 | Required size 17 nodes | |
33fc7377 YO |
2695 | first at 3 |
2696 | Freeing REx: "(?<b>\g{c})(?<c>x)(?&b)" | |
2697 | EOF_DEBUG_OUT | |
fe5492d9 YO |
2698 | {rtrim_result=>1}, |
2699 | "Related to Github Issue #19350, forward \\g{x} pattern segv under use re Debug => 'PARSE'"); | |
33fc7377 | 2700 | } |
bb58640a | 2701 | |
ff1f9f59 KW |
2702 | { # perl-security#140, read/write past buffer end |
2703 | fresh_perl_like('qr/\p{utf8::perl x}/', | |
2704 | qr/Illegal user-defined property name "utf8::perl x" in regex/, | |
2705 | {}, "perl-security#140"); | |
2706 | fresh_perl_is('qr/\p{utf8::_perl_surrogate}/', "", | |
2707 | {}, "perl-security#140"); | |
2708 | } | |
2709 | ||
8fbe3f6f KW |
2710 | { # GH 20009 |
2711 | my $x = "awesome quotes"; | |
2712 | utf8::upgrade($x); | |
2713 | $x =~ s/^[\x{0301}\x{030C}]+//; | |
2714 | } | |
2715 | ||
553fa53c | 2716 | |
b57e4118 | 2717 | # !!! NOTE that tests that aren't at all likely to crash perl should go |
de085464 KW |
2718 | # a ways above, above these last ones. There's a comment there that, like |
2719 | # this comment, contains the word 'NOTE' | |
f4554ed5 NC |
2720 | |
2721 | done_testing(); | |
e425a60b YO |
2722 | } # End of sub run_tests |
2723 | ||
2724 | 1; |