This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
t/re/reg_mesg.t: Add test
[perl5.git] / t / re / reg_mesg.t
CommitLineData
b45f050a
JF
1#!./perl -w
2
21adcf33
KW
3$|=1; # outherwise things get mixed up in output
4
b45f050a
JF
5BEGIN {
6 chdir 't' if -d 't';
cb79f740 7 require './test.pl';
624c42e2 8 set_up_inc( qw '../lib ../ext/re' );
cb79f740 9 eval 'require Config'; # assume defaults if this fails
b45f050a
JF
10}
11
624c42e2
N
12skip_all_without_unicode_tables();
13
cb79f740 14use strict;
c1d900c3 15use open qw(:utf8 :std);
b45f050a 16
b1793c1a
KW
17# Kind of a kludge to mark warnings to be expected only if we are testing
18# under "use re 'strict'"
19my $only_strict_marker = ':expected_only_under_strict';
20
b45f050a 21## If the markers used are changed (search for "MARKER1" in regcomp.c),
cb79f740 22## update only these two regexs, and leave the {#} in the @death/@warning
b45f050a
JF
23## arrays below. The {#} is a meta-marker -- it marks where the marker should
24## go.
af01601c 25
b1793c1a
KW
26sub fixup_expect ($$) {
27
28 # Fixes up the expected results by inserting the boiler plate text.
29 # Returns empty string if that is what is expected. Otherwise, handles
30 # either a scalar, turning it into a single element array; or a ref to an
31 # array, adjusting each element. If called in array context, returns an
32 # array, otherwise the join of all elements.
33
34 # The string $only_strict_marker will be removed from any expect line it
35 # begins, and if $strict is not true, that expect line will be removed
36 # from the output (hence won't be expected)
37
38 my ($expect_ref, $strict) = @_;
63374c78 39 return "" if $expect_ref eq "";
af01601c
KW
40
41 my @expect;
42 if (ref $expect_ref) {
43 @expect = @$expect_ref;
44 }
45 else {
46 @expect = $expect_ref;
47 }
48
b1793c1a 49 my @new_expect;
af01601c 50 foreach my $element (@expect) {
b1793c1a
KW
51 $element =~ s/\{\#\}/in regex; marked by <-- HERE in/;
52 $element =~ s/\{\#\}/ <-- HERE /;
af01601c 53 $element .= " at ";
b1793c1a
KW
54 next if $element =~ s/ ^ $only_strict_marker \s* //x && ! $strict;
55 push @new_expect, $element;
af01601c 56 }
b1793c1a 57 return wantarray ? @new_expect : join "", @new_expect;
cb79f740 58}
b45f050a 59
af05e4c0
YO
60sub add_markers {
61 my ($element)= @_;
62 $element =~ s/ at .* line \d+\.?\n$//;
63 $element =~ s/in regex; marked by <-- HERE in/{#}/;
64 $element =~ s/ <-- HERE /{#}/;
65 return $element;
66}
67
c1d900c3 68## Because we don't "use utf8" in this file, we need to do some extra legwork
b0e1d434
KW
69## for the utf8 tests: Prepend 'use utf8' to the pattern, and mark the strings
70## to check against as UTF-8, but for this all to work properly, the character
71## 'ネ' (U+30CD) is required in each pattern somewhere as a marker.
4cabb89a
BF
72##
73## This also creates a second variant of the tests to check if the
b0e1d434
KW
74## latin1 error messages are working correctly. Because we don't 'use utf8',
75## we can't tell if something is UTF-8 or Latin1, so you need the suffix
76## '; no latin1' to not have the second variant.
4cabb89a
BF
77my $l1 = "\x{ef}";
78my $utf8 = "\x{30cd}";
79utf8::encode($utf8);
80
c1d900c3
BF
81sub mark_as_utf8 {
82 my @ret;
67cdf558
KW
83 for (my $i = 0; $i < @_; $i += 2) {
84 my $pat = $_[$i];
85 my $msg = $_[$i+1];
4cabb89a
BF
86 my $l1_pat = $pat =~ s/$utf8/$l1/gr;
87 my $l1_msg;
c1d900c3 88 $pat = "use utf8; $pat";
b0e1d434 89
c1d900c3 90 if (ref $msg) {
4cabb89a
BF
91 $l1_msg = [ map { s/$utf8/$l1/gr } @$msg ];
92 @$msg = map { my $c = $_; utf8::decode($c); $c } @$msg;
c1d900c3
BF
93 }
94 else {
4cabb89a 95 $l1_msg = $msg =~ s/$utf8/$l1/gr;
c1d900c3
BF
96 utf8::decode($msg);
97 }
98 push @ret, $pat => $msg;
b0e1d434 99
4cabb89a 100 push @ret, $l1_pat => $l1_msg unless $l1_pat =~ /#no latin1/;
c1d900c3
BF
101 }
102 return @ret;
103}
104
4288c5b9 105my $inf_m1 = ($Config::Config{reg_infty} || 65535) - 1;
cb79f740 106my $inf_p1 = $inf_m1 + 2;
b45f050a 107
c877af1b
KW
108my $B_hex = sprintf("\\x%02X", ord "B");
109my $low_mixed_alpha = ('A' lt 'a') ? 'A' : 'a';
110my $high_mixed_alpha = ('A' lt 'a') ? 'a' : 'A';
111my $low_mixed_digit = ('A' lt '0') ? 'A' : '0';
112my $high_mixed_digit = ('A' lt '0') ? '0' : 'A';
113
acdfc3b6
KW
114my $colon_hex = sprintf "%02X", ord(":");
115my $tab_hex = sprintf "%02X", ord("\t");
116
d30277c7
KW
117# Key-value pairs of strings eval'd as patterns => warn/error messages that
118# they should generate. In some cases, the value is an array of multiple
119# messages. Some groups have the message(s) be default on; others, default
120# off. This can be overridden on an individual key basis by preceding the
121# pattern string with either 'default_on' or 'default_off'
122#
123# The first set are those that should be fatal errors.
124
b45f050a
JF
125my @death =
126(
6d24e9d4 127 '/[[=foo=]]/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/[[=foo=]{#}]/',
b45f050a 128
58e23c8d 129 '/(?<= .*)/' => 'Variable length lookbehind not implemented in regex m/(?<= .*)/',
b45f050a 130
58e23c8d 131 '/(?<= x{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= x{1000})/',
b45f050a 132
6d24e9d4 133 '/(?@)/' => 'Sequence (?@...) not implemented {#} m/(?@{#})/',
b45f050a 134
9da1dd8f 135 '/(?{ 1/' => 'Missing right curly or square bracket',
b45f050a 136
6d24e9d4 137 '/(?(1x))/' => 'Switch condition not recognized {#} m/(?(1x{#}))/',
311cc1ad 138 '/(?(1x(?#)))/'=> 'Switch condition not recognized {#} m/(?(1x{#}(?#)))/',
b45f050a 139
8fb0127d
YO
140 '/(?(1)/' => 'Switch (?(condition)... not terminated {#} m/(?(1){#}/',
141 '/(?(1)x/' => 'Switch (?(condition)... not terminated {#} m/(?(1)x{#}/',
142 '/(?(1)x|y/' => 'Switch (?(condition)... not terminated {#} m/(?(1)x|y{#}/',
6d24e9d4 143 '/(?(1)x|y|z)/' => 'Switch (?(condition)... contains too many branches {#} m/(?(1)x|y|{#}z)/',
b45f050a 144
c1d900c3 145 '/(?(x)y|x)/' => 'Unknown switch condition (?(...)) {#} m/(?(x{#})y|x)/',
10380cb3
FC
146 '/(?(??{}))/' => 'Unknown switch condition (?(...)) {#} m/(?(?{#}?{}))/',
147 '/(?(?[]))/' => 'Unknown switch condition (?(...)) {#} m/(?(?{#}[]))/',
b45f050a 148
6d24e9d4 149 '/(?/' => 'Sequence (? incomplete {#} m/(?{#}/',
b45f050a 150
6d24e9d4
KW
151 '/(?;x/' => 'Sequence (?;...) not recognized {#} m/(?;{#}x/',
152 '/(?<;x/' => 'Group name must start with a non-digit word character {#} m/(?<;{#}x/',
153 '/(?\ix/' => 'Sequence (?\...) not recognized {#} m/(?\{#}ix/',
154 '/(?\mx/' => 'Sequence (?\...) not recognized {#} m/(?\{#}mx/',
155 '/(?\:x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}:x/',
156 '/(?\=x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}=x/',
157 '/(?\!x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}!x/',
158 '/(?\<=x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}<=x/',
159 '/(?\<!x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}<!x/',
160 '/(?\>x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}>x/',
161 '/(?^-i:foo)/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i:foo)/',
162 '/(?^-i)foo/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i)foo/',
163 '/(?^d:foo)/' => 'Sequence (?^d...) not recognized {#} m/(?^d{#}:foo)/',
164 '/(?^d)foo/' => 'Sequence (?^d...) not recognized {#} m/(?^d{#})foo/',
165 '/(?^lu:foo)/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/(?^lu{#}:foo)/',
166 '/(?^lu)foo/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/(?^lu{#})foo/',
167'/(?da:foo)/' => 'Regexp modifiers "d" and "a" are mutually exclusive {#} m/(?da{#}:foo)/',
168'/(?lil:foo)/' => 'Regexp modifier "l" may not appear twice {#} m/(?lil{#}:foo)/',
169'/(?aaia:foo)/' => 'Regexp modifier "a" may appear a maximum of twice {#} m/(?aaia{#}:foo)/',
170'/(?i-l:foo)/' => 'Regexp modifier "l" may not appear after the "-" {#} m/(?i-l{#}:foo)/',
cc74c5bd 171
6d24e9d4 172 '/((x)/' => 'Unmatched ( {#} m/({#}(x)/',
77eddfe1 173 '/{(}/' => 'Unmatched ( {#} m/{({#}}/', # [perl #127599]
b45f050a 174
6d24e9d4 175 "/x{$inf_p1}/" => "Quantifier in {,} bigger than $inf_m1 {#} m/x{{#}$inf_p1}/",
b45f050a 176
b45f050a 177
6d24e9d4 178 '/x**/' => 'Nested quantifiers {#} m/x**{#}/',
b45f050a 179
6d24e9d4 180 '/x[/' => 'Unmatched [ {#} m/x[{#}/',
b45f050a 181
6d24e9d4 182 '/*/', => 'Quantifier follows nothing {#} m/*{#}/',
b45f050a 183
6d24e9d4 184 '/\p{x/' => 'Missing right brace on \p{} {#} m/\p{{#}x/',
b45f050a 185
6d24e9d4 186 '/[\p{x]/' => 'Missing right brace on \p{} {#} m/[\p{{#}x]/',
b45f050a 187
6d24e9d4 188 '/(x)\2/' => 'Reference to nonexistent group {#} m/(x)\2{#}/',
b45f050a 189
779fedd7 190 '/\g/' => 'Unterminated \g... pattern {#} m/\g{#}/',
76cccc4d
KW
191 '/\g{1/' => 'Unterminated \g{...} pattern {#} m/\g{1{#}/',
192
40809656 193 'my $m = "\\\"; $m =~ $m', => 'Trailing \ in regex m/\/',
b45f050a 194
6d24e9d4
KW
195 '/\x{1/' => 'Missing right brace on \x{} {#} m/\x{1{#}/',
196 '/\x{X/' => 'Missing right brace on \x{} {#} m/\x{{#}X/',
197
198 '/[\x{X]/' => 'Missing right brace on \x{} {#} m/[\x{{#}X]/',
199 '/[\x{A]/' => 'Missing right brace on \x{} {#} m/[\x{A{#}]/',
200
201 '/\o{1/' => 'Missing right brace on \o{ {#} m/\o{1{#}/',
202 '/\o{X/' => 'Missing right brace on \o{ {#} m/\o{{#}X/',
203
204 '/[\o{X]/' => 'Missing right brace on \o{ {#} m/[\o{{#}X]/',
205 '/[\o{7]/' => 'Missing right brace on \o{ {#} m/[\o{7{#}]/',
206
207 '/[[:barf:]]/' => 'POSIX class [:barf:] unknown {#} m/[[:barf:]{#}]/',
208
209 '/[[=barf=]]/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/[[=barf=]{#}]/',
210
211 '/[[.barf.]]/' => 'POSIX syntax [. .] is reserved for future extensions {#} m/[[.barf.]{#}]/',
212
213 '/[z-a]/' => 'Invalid [] range "z-a" {#} m/[z-a{#}]/',
214
c30c479a 215 '/\p/' => 'Empty \p {#} m/\p{#}/',
856d91b3
KW
216 '/\P/' => 'Empty \P {#} m/\P{#}/',
217 '/\p{}/' => 'Empty \p{} {#} m/\p{{#}}/',
6d24e9d4 218 '/\P{}/' => 'Empty \P{} {#} m/\P{{#}}/',
64935bc6
KW
219
220'/a\b{cde/' => 'Missing right brace on \b{} {#} m/a\b{{#}cde/',
221'/a\B{cde/' => 'Missing right brace on \B{} {#} m/a\B{{#}cde/',
222
223 '/\b{}/' => 'Empty \b{} {#} m/\b{}{#}/',
224 '/\B{}/' => 'Empty \B{} {#} m/\B{}{#}/',
225
226 '/\b{gc}/' => "'gc' is an unknown bound type {#} m/\\b{gc{#}}/",
227 '/\B{gc}/' => "'gc' is an unknown bound type {#} m/\\B{gc{#}}/",
228
19a498a4
YO
229 '/(?[[[::]]])/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[[[::]]{#}])/",
230 '/(?[[[:w:]]])/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[[[:w:]]{#}])/",
6d24e9d4 231 '/(?[a])/' => 'Unexpected character {#} m/(?[a{#}])/',
6d24e9d4
KW
232 '/(?[ + \t ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ +{#} \t ])/',
233 '/(?[ \cK - ( + \t ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ \cK - ( +{#} \t ) ])/',
234 '/(?[ \cK ( \t ) ])/' => 'Unexpected \'(\' with no preceding operator {#} m/(?[ \cK ({#} \t ) ])/',
235 '/(?[ \cK \t ])/' => 'Operand with no preceding operator {#} m/(?[ \cK \t{#} ])/',
236 '/(?[ \0004 ])/' => 'Need exactly 3 octal digits {#} m/(?[ \0004 {#}])/',
237 '/(?[ \05 ])/' => 'Need exactly 3 octal digits {#} m/(?[ \05 {#}])/',
238 '/(?[ \o{1038} ])/' => 'Non-octal character {#} m/(?[ \o{1038{#}} ])/',
239 '/(?[ \o{} ])/' => 'Number with no digits {#} m/(?[ \o{}{#} ])/',
240 '/(?[ \x{defg} ])/' => 'Non-hex character {#} m/(?[ \x{defg{#}} ])/',
241 '/(?[ \xabcdef ])/' => 'Use \\x{...} for more than two hex characters {#} m/(?[ \xabc{#}def ])/',
242 '/(?[ \x{} ])/' => 'Number with no digits {#} m/(?[ \x{}{#} ])/',
243 '/(?[ \cK + ) ])/' => 'Unexpected \')\' {#} m/(?[ \cK + ){#} ])/',
244 '/(?[ \cK + ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ \cK + {#}])/',
c333712c
KW
245 '/(?[ ( ) ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ ( ){#} ])/',
246 '/(?[[0]+()+])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[[0]+(){#}+])/',
29f52644
KW
247 '/(?[ \p{foo} ])/' => 'Can\'t find Unicode property definition "foo" {#} m/(?[ \p{foo}{#} ])/',
248 '/(?[ \p{ foo = bar } ])/' => 'Can\'t find Unicode property definition "foo = bar" {#} m/(?[ \p{ foo = bar }{#} ])/',
6d24e9d4 249 '/(?[ \8 ])/' => 'Unrecognized escape \8 in character class {#} m/(?[ \8{#} ])/',
19a498a4
YO
250 '/(?[ \t ]/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[ \\t ]{#}/",
251 '/(?[ [ \t ]/' => "Syntax error in (?[...]) {#} m/(?[ [ \\t ]{#}/",
252 '/(?[ \t ] ]/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[ \\t ]{#} ]/",
253 '/(?[ [ ] ]/' => "Syntax error in (?[...]) {#} m/(?[ [ ] ]{#}/",
254 '/(?[ \t + \e # This was supposed to be a comment ])/' =>
255 "Syntax error in (?[...]) {#} m/(?[ \\t + \\e # This was supposed to be a comment ]){#}/",
6d24e9d4
KW
256 '/(?[ ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ {#}])/',
257 'm/(?[[a-\d]])/' => 'False [] range "a-\d" {#} m/(?[[a-\d{#}]])/',
258 'm/(?[[\w-x]])/' => 'False [] range "\w-" {#} m/(?[[\w-{#}x]])/',
259 'm/(?[[a-\pM]])/' => 'False [] range "a-\pM" {#} m/(?[[a-\pM{#}]])/',
260 'm/(?[[\pM-x]])/' => 'False [] range "\pM-" {#} m/(?[[\pM-{#}x]])/',
8f0cd35a 261 'm/(?[[^\N{LATIN CAPITAL LETTER A WITH MACRON AND GRAVE}]])/' => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/(?[[^\N{U+100.300{#}}]])/',
b5864679
KW
262 'm/(?[ \p{Digit} & (?(?[ \p{Thai} | \p{Lao} ]))])/' => 'Sequence (?(...) not recognized {#} m/(?[ \p{Digit} & (?({#}?[ \p{Thai} | \p{Lao} ]))])/',
263 'm/(?[ \p{Digit} & (?:(?[ \p{Thai} | \p{Lao} ]))])/' => 'Expecting \'(?flags:(?[...\' {#} m/(?[ \p{Digit} & (?{#}:(?[ \p{Thai} | \p{Lao} ]))])/',
6d24e9d4
KW
264 'm/\o{/' => 'Missing right brace on \o{ {#} m/\o{{#}/',
265 'm/\o/' => 'Missing braces on \o{} {#} m/\o{#}/',
266 'm/\o{}/' => 'Number with no digits {#} m/\o{}{#}/',
267 'm/[\o{]/' => 'Missing right brace on \o{ {#} m/[\o{{#}]/',
268 'm/[\o]/' => 'Missing braces on \o{} {#} m/[\o{#}]/',
269 'm/[\o{}]/' => 'Number with no digits {#} m/[\o{}{#}]/',
270 'm/(?^-i:foo)/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i:foo)/',
f1e1b256
YO
271 'm/\87/' => 'Reference to nonexistent group {#} m/\87{#}/',
272 'm/a\87/' => 'Reference to nonexistent group {#} m/a\87{#}/',
273 'm/a\97/' => 'Reference to nonexistent group {#} m/a\97{#}/',
07ea66ee
FC
274 'm/(*DOOF)/' => 'Unknown verb pattern \'DOOF\' {#} m/(*DOOF){#}/',
275 'm/(?&a/' => 'Sequence (?&... not terminated {#} m/(?&a{#}/',
75839571
FC
276 'm/(?P=/' => 'Sequence ?P=... not terminated {#} m/(?P={#}/',
277 "m/(?'/" => "Sequence (?'... not terminated {#} m/(?'{#}/",
278 "m/(?</" => "Sequence (?<... not terminated {#} m/(?<{#}/",
279 'm/(?&/' => 'Sequence (?&... not terminated {#} m/(?&{#}/',
280 'm/(?(</' => 'Sequence (?(<... not terminated {#} m/(?(<{#}/',
281 "m/(?('/" => "Sequence (?('... not terminated {#} m/(?('{#}/",
282 'm/\g{/' => 'Sequence \g{... not terminated {#} m/\g{{#}/',
283 'm/\k</' => 'Sequence \k<... not terminated {#} m/\k<{#}/',
9b8f4e92 284 'm/\cß/' => "Character following \"\\c\" must be printable ASCII",
cd209d9d
KW
285 '/((?# This is a comment in the middle of a token)?:foo)/' => 'In \'(?...)\', the \'(\' and \'?\' must be adjacent {#} m/((?# This is a comment in the middle of a token)?{#}:foo)/',
286 '/((?# This is a comment in the middle of a token)*FAIL)/' => 'In \'(*VERB...)\', the \'(\' and \'*\' must be adjacent {#} m/((?# This is a comment in the middle of a token)*{#}FAIL)/',
d9790612
KW
287 '/((?# This is a comment in the middle of a token)*script_run:foo)/' => 'In \'(*...)\', the \'(\' and \'*\' must be adjacent {#} m/((?# This is a comment in the middle of a token)*{#}script_run:foo)/',
288
289 '/(*script_runfoo)/' => 'Unknown \'(*...)\' construct \'script_runfoo\' {#} m/(*script_runfoo){#}/',
290 '/(*srfoo)/' => 'Unknown \'(*...)\' construct \'srfoo\' {#} m/(*srfoo){#}/',
291 '/(*script_run)/' => '\'(*script_run\' requires a terminating \':\' {#} m/(*script_run{#})/',
292 '/(*sr)/' => '\'(*sr\' requires a terminating \':\' {#} m/(*sr{#})/',
e7206367 293 '/(*pla)/' => '\'(*pla\' requires a terminating \':\' {#} m/(*pla{#})/',
d9790612
KW
294 '/(*script_run/' => 'Unterminated \'(*...\' construct {#} m/(*script_run{#}/',
295 '/(*sr/' => 'Unterminated \'(*...\' construct {#} m/(*sr{#}/',
296 '/(*script_run:foo/' => 'Unterminated \'(*...\' argument {#} m/(*script_run:foo{#}/',
297 '/(*sr:foo/' => 'Unterminated \'(*...\' argument {#} m/(*sr:foo{#}/',
5a55ed33 298 '/(?[\ &!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ &!{#}])/', # [perl #126180]
f4b6841f
KW
299 '/(?[\ +!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ +!{#}])/', # [perl #126180]
300 '/(?[\ -!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ -!{#}])/', # [perl #126180]
301 '/(?[\ ^!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ ^!{#}])/', # [perl #126180]
302 '/(?[\ |!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ |!{#}])/', # [perl #126180]
c333712c 303 '/(?[()-!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[(){#}-!])/', # [perl #126204]
174c9902 304 '/(?[!()])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[!(){#}])/', # [perl #126404]
8e84dec2
KW
305 '/\w{/' => 'Unescaped left brace in regex is illegal here {#} m/\w{{#}/',
306 '/\q{/' => 'Unescaped left brace in regex is illegal here {#} m/\q{{#}/',
307 '/\A{/' => 'Unescaped left brace in regex is illegal here {#} m/\A{{#}/',
76416d1a
KW
308 '/.{, 4 }/' => 'Unescaped left brace in regex is illegal here {#} m/.{{#}, 4 }/',
309 '/[x]{, 4}/' => 'Unescaped left brace in regex is illegal here {#} m/[x]{{#}, 4}/',
310 '/\p{Latin}{,4 }/' => 'Unescaped left brace in regex is illegal here {#} m/\p{Latin}{{#},4 }/',
f2e32b2c 311 '/(?<=/' => 'Sequence (?... not terminated {#} m/(?<={#}/', # [perl #128170]
26a0fa3a 312 '/\p{vertical \v tab}/' => 'Can\'t find Unicode property definition "vertical \v tab" {#} m/\\p{vertical \v tab}{#}/', # [perl #132055]
60108b47 313
b45f050a 314);
c1d900c3 315
bd71b4b5
KW
316# These are messages that are death under 'use re "strict"', and may or may
317# not warn otherwise. See comment before @warning as to why some have a
318# \x{100} in them. This array has 3 elements per construct. [0] is the regex
319# to use; [1] is the message under no strict (empty to not warn), and [2] is
320# under strict.
67cdf558
KW
321my @death_only_under_strict = (
322 'm/\xABC/' => "",
323 => 'Use \x{...} for more than two hex characters {#} m/\xABC{#}/',
324 'm/[\xABC]/' => "",
325 => 'Use \x{...} for more than two hex characters {#} m/[\xABC{#}]/',
326
327 # XXX This is a confusing error message. The G isn't ignored; it just
328 # terminates the \x. Also some messages below are missing the <-- HERE,
329 # aren't all category 'regexp'. (Hence we have to turn off 'digit'
330 # messages as well below)
331 'm/\xAG/' => 'Illegal hexadecimal digit \'G\' ignored',
332 => 'Non-hex character {#} m/\xAG{#}/',
333 'm/[\xAG]/' => 'Illegal hexadecimal digit \'G\' ignored',
334 => 'Non-hex character {#} m/[\xAG{#}]/',
335 'm/\o{789}/' => 'Non-octal character \'8\'. Resolved as "\o{7}"',
336 => 'Non-octal character {#} m/\o{78{#}9}/',
337 'm/[\o{789}]/' => 'Non-octal character \'8\'. Resolved as "\o{7}"',
338 => 'Non-octal character {#} m/[\o{78{#}9}]/',
339 'm/\x{}/' => "",
340 => 'Number with no digits {#} m/\x{}{#}/',
341 'm/[\x{}]/' => "",
342 => 'Number with no digits {#} m/[\x{}{#}]/',
343 'm/\x{ABCDEFG}/' => 'Illegal hexadecimal digit \'G\' ignored',
344 => 'Non-hex character {#} m/\x{ABCDEFG{#}}/',
345 'm/[\x{ABCDEFG}]/' => 'Illegal hexadecimal digit \'G\' ignored',
346 => 'Non-hex character {#} m/[\x{ABCDEFG{#}}]/',
67cdf558
KW
347 "m'[\\y]\\x{100}'" => 'Unrecognized escape \y in character class passed through {#} m/[\y{#}]\x{100}/',
348 => 'Unrecognized escape \y in character class {#} m/[\y{#}]\x{100}/',
349 'm/[a-\d]\x{100}/' => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
350 => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
351 'm/[\w-x]\x{100}/' => 'False [] range "\w-" {#} m/[\w-{#}x]\x{100}/',
352 => 'False [] range "\w-" {#} m/[\w-{#}x]\x{100}/',
353 'm/[a-\pM]\x{100}/' => 'False [] range "a-\pM" {#} m/[a-\pM{#}]\x{100}/',
354 => 'False [] range "a-\pM" {#} m/[a-\pM{#}]\x{100}/',
355 'm/[\pM-x]\x{100}/' => 'False [] range "\pM-" {#} m/[\pM-{#}x]\x{100}/',
356 => 'False [] range "\pM-" {#} m/[\pM-{#}x]\x{100}/',
357 'm/[^\N{LATIN CAPITAL LETTER A WITH MACRON AND GRAVE}]/' => 'Using just the first character returned by \N{} in character class {#} m/[^\N{U+100.300}{#}]/',
358 => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[^\N{U+100.300{#}}]/',
359 'm/[\x03-\N{LATIN CAPITAL LETTER A WITH MACRON AND GRAVE}]/' => 'Using just the first character returned by \N{} in character class {#} m/[\x03-\N{U+100.300}{#}]/',
360 => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[\x03-\N{U+100.300{#}}]/',
361 'm/[\N{LATIN CAPITAL LETTER A WITH MACRON AND GRAVE}-\x{10FFFF}]/' => 'Using just the first character returned by \N{} in character class {#} m/[\N{U+100.300}{#}-\x{10FFFF}]/',
362 => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[\N{U+100.300{#}}-\x{10FFFF}]/',
363 '/[\08]/' => '\'\08\' resolved to \'\o{0}8\' {#} m/[\08{#}]/',
364 => 'Need exactly 3 octal digits {#} m/[\08{#}]/',
365 '/[\018]/' => '\'\018\' resolved to \'\o{1}8\' {#} m/[\018{#}]/',
366 => 'Need exactly 3 octal digits {#} m/[\018{#}]/',
367 '/[\_\0]/' => "",
368 => 'Need exactly 3 octal digits {#} m/[\_\0]{#}/',
369 '/[\07]/' => "",
370 => 'Need exactly 3 octal digits {#} m/[\07]{#}/',
371 '/[\0005]/' => "",
372 => 'Need exactly 3 octal digits {#} m/[\0005]{#}/',
373 '/[\8\9]\x{100}/' => ['Unrecognized escape \8 in character class passed through {#} m/[\8{#}\9]\x{100}/',
374 'Unrecognized escape \9 in character class passed through {#} m/[\8\9{#}]\x{100}/',
375 ],
376 => 'Unrecognized escape \8 in character class {#} m/[\8{#}\9]\x{100}/',
377 '/[a-\d]\x{100}/' => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
378 => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
379 '/[\d-b]\x{100}/' => 'False [] range "\d-" {#} m/[\d-{#}b]\x{100}/',
380 => 'False [] range "\d-" {#} m/[\d-{#}b]\x{100}/',
381 '/[\s-\d]\x{100}/' => 'False [] range "\s-" {#} m/[\s-{#}\d]\x{100}/',
382 => 'False [] range "\s-" {#} m/[\s-{#}\d]\x{100}/',
383 '/[\d-\s]\x{100}/' => 'False [] range "\d-" {#} m/[\d-{#}\s]\x{100}/',
384 => 'False [] range "\d-" {#} m/[\d-{#}\s]\x{100}/',
385 '/[a-[:digit:]]\x{100}/' => 'False [] range "a-[:digit:]" {#} m/[a-[:digit:]{#}]\x{100}/',
386 => 'False [] range "a-[:digit:]" {#} m/[a-[:digit:]{#}]\x{100}/',
387 '/[[:digit:]-b]\x{100}/' => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}b]\x{100}/',
388 => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}b]\x{100}/',
389 '/[[:alpha:]-[:digit:]]\x{100}/' => 'False [] range "[:alpha:]-" {#} m/[[:alpha:]-{#}[:digit:]]\x{100}/',
390 => 'False [] range "[:alpha:]-" {#} m/[[:alpha:]-{#}[:digit:]]\x{100}/',
391 '/[[:digit:]-[:alpha:]]\x{100}/' => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}[:alpha:]]\x{100}/',
392 => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}[:alpha:]]\x{100}/',
393 '/[a\zb]\x{100}/' => 'Unrecognized escape \z in character class passed through {#} m/[a\z{#}b]\x{100}/',
394 => 'Unrecognized escape \z in character class {#} m/[a\z{#}b]\x{100}/',
2a6971a9
KW
395 '/[a\fb]/' => "",
396 => 'Literal vertical space in [] is illegal except under /x {#} m/[a\f{#}b]/',
76416d1a
KW
397 '/:{4,a}/' => 'Unescaped left brace in regex is passed through {#} m/:{{#}4,a}/',
398 => 'Unescaped left brace in regex is illegal here {#} m/:{{#}4,a}/',
399 '/xa{3\,4}y/' => 'Unescaped left brace in regex is passed through {#} m/xa{{#}3\,4}y/',
400 => 'Unescaped left brace in regex is illegal here {#} m/xa{{#}3\,4}y/',
401 '/\\${[^\\}]*}/' => 'Unescaped left brace in regex is passed through {#} m/\\${{#}[^\\}]*}/',
402 => 'Unescaped left brace in regex is illegal here {#} m/\\${{#}[^\\}]*}/',
403 '/.{/' => 'Unescaped left brace in regex is passed through {#} m/.{{#}/',
404 => 'Unescaped left brace in regex is illegal here {#} m/.{{#}/',
405 '/[x]{/' => 'Unescaped left brace in regex is passed through {#} m/[x]{{#}/',
406 => 'Unescaped left brace in regex is illegal here {#} m/[x]{{#}/',
407 '/\p{Latin}{/' => 'Unescaped left brace in regex is passed through {#} m/\p{Latin}{{#}/',
408 => 'Unescaped left brace in regex is illegal here {#} m/\p{Latin}{{#}/',
67cdf558
KW
409);
410
b0e1d434 411# These need the character 'ネ' as a marker for mark_as_utf8()
c1d900c3 412my @death_utf8 = mark_as_utf8(
c1d900c3
BF
413 '/ネ(?<= .*)/' => 'Variable length lookbehind not implemented in regex m/ネ(?<= .*)/',
414
415 '/(?<= ネ{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= ネ{1000})/',
416
417 '/ネ(?ネ)ネ/' => 'Sequence (?ネ...) not recognized {#} m/ネ(?ネ{#})ネ/',
418
476afc4b 419 '/ネ(?(1ネ))ネ/' => 'Switch condition not recognized {#} m/ネ(?(1ネ{#}))ネ/',
c1d900c3
BF
420
421 '/(?(1)ネ|y|ヌ)/' => 'Switch (?(condition)... contains too many branches {#} m/(?(1)ネ|y|{#}ヌ)/',
422
423 '/(?(ネ)y|ネ)/' => 'Unknown switch condition (?(...)) {#} m/(?(ネ{#})y|ネ)/',
424
425 '/ネ(?/' => 'Sequence (? incomplete {#} m/ネ(?{#}/',
426
427 '/ネ(?;ネ/' => 'Sequence (?;...) not recognized {#} m/ネ(?;{#}ネ/',
428 '/ネ(?<;ネ/' => 'Group name must start with a non-digit word character {#} m/ネ(?<;{#}ネ/',
429 '/ネ(?\ixネ/' => 'Sequence (?\...) not recognized {#} m/ネ(?\{#}ixネ/',
430 '/ネ(?^lu:ネ)/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/ネ(?^lu{#}:ネ)/',
431'/ネ(?lil:ネ)/' => 'Regexp modifier "l" may not appear twice {#} m/ネ(?lil{#}:ネ)/',
432'/ネ(?aaia:ネ)/' => 'Regexp modifier "a" may appear a maximum of twice {#} m/ネ(?aaia{#}:ネ)/',
433'/ネ(?i-l:ネ)/' => 'Regexp modifier "l" may not appear after the "-" {#} m/ネ(?i-l{#}:ネ)/',
434
435 '/ネ((ネ)/' => 'Unmatched ( {#} m/ネ({#}(ネ)/',
436
437 "/ネ{$inf_p1}ネ/" => "Quantifier in {,} bigger than $inf_m1 {#} m/ネ{{#}$inf_p1}ネ/",
438
439
440 '/ネ**ネ/' => 'Nested quantifiers {#} m/ネ**{#}ネ/',
441
442 '/ネ[ネ/' => 'Unmatched [ {#} m/ネ[{#}ネ/',
443
444 '/*ネ/', => 'Quantifier follows nothing {#} m/*{#}ネ/',
445
446 '/ネ\p{ネ/' => 'Missing right brace on \p{} {#} m/ネ\p{{#}ネ/',
447
448 '/(ネ)\2ネ/' => 'Reference to nonexistent group {#} m/(ネ)\2{#}ネ/',
449
4cabb89a 450 '/\g{ネ/; #no latin1' => 'Sequence \g{... not terminated {#} m/\g{ネ{#}/',
c1d900c3
BF
451
452 'my $m = "ネ\\\"; $m =~ $m', => 'Trailing \ in regex m/ネ\/',
453
454 '/\x{ネ/' => 'Missing right brace on \x{} {#} m/\x{{#}ネ/',
455 '/ネ[\x{ネ]ネ/' => 'Missing right brace on \x{} {#} m/ネ[\x{{#}ネ]ネ/',
456 '/ネ[\x{ネ]/' => 'Missing right brace on \x{} {#} m/ネ[\x{{#}ネ]/',
457
458 '/ネ\o{ネ/' => 'Missing right brace on \o{ {#} m/ネ\o{{#}ネ/',
c1d900c3 459
c1d900c3
BF
460 '/[ネ-a]ネ/' => 'Invalid [] range "ネ-a" {#} m/[ネ-a{#}]ネ/',
461
462 '/ネ\p{}ネ/' => 'Empty \p{} {#} m/ネ\p{{#}}ネ/',
463
19a498a4
YO
464 '/ネ(?[[[:ネ]]])ネ/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[[[:ネ]]{#}])ネ/",
465 '/ネ(?[[[:ネ: ])ネ/' => "Syntax error in (?[...]) {#} m/ネ(?[[[:ネ: ])ネ{#}/",
466 '/ネ(?[[[::]]])ネ/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[[[::]]{#}])ネ/",
467 '/ネ(?[[[:ネ:]]])ネ/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[[[:ネ:]]{#}])ネ/",
c1d900c3 468 '/ネ(?[ネ])ネ/' => 'Unexpected character {#} m/ネ(?[ネ{#}])ネ/',
c1d900c3
BF
469 '/ネ(?[ + [ネ] ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/ネ(?[ +{#} [ネ] ])/',
470 '/ネ(?[ \cK - ( + [ネ] ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/ネ(?[ \cK - ( +{#} [ネ] ) ])/',
471 '/ネ(?[ \cK ( [ネ] ) ])/' => 'Unexpected \'(\' with no preceding operator {#} m/ネ(?[ \cK ({#} [ネ] ) ])/',
472 '/ネ(?[ \cK [ネ] ])ネ/' => 'Operand with no preceding operator {#} m/ネ(?[ \cK [ネ{#}] ])ネ/',
473 '/ネ(?[ \0004 ])ネ/' => 'Need exactly 3 octal digits {#} m/ネ(?[ \0004 {#}])ネ/',
474 '/(?[ \o{ネ} ])ネ/' => 'Non-octal character {#} m/(?[ \o{ネ{#}} ])ネ/',
475 '/ネ(?[ \o{} ])ネ/' => 'Number with no digits {#} m/ネ(?[ \o{}{#} ])ネ/',
476 '/(?[ \x{ネ} ])ネ/' => 'Non-hex character {#} m/(?[ \x{ネ{#}} ])ネ/',
29f52644
KW
477 '/(?[ \p{ネ} ])/' => 'Can\'t find Unicode property definition "ネ" {#} m/(?[ \p{ネ}{#} ])/',
478 '/(?[ \p{ ネ = bar } ])/' => 'Can\'t find Unicode property definition "ネ = bar" {#} m/(?[ \p{ ネ = bar }{#} ])/',
19a498a4
YO
479 '/ネ(?[ \t ]/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[ \\t ]{#}/",
480 '/(?[ \t + \e # ネ This was supposed to be a comment ])/' =>
481 "Syntax error in (?[...]) {#} m/(?[ \\t + \\e # ネ This was supposed to be a comment ]){#}/",
d9790612 482 'm/(*ネ)ネ/' => q<Unknown '(*...)' construct 'ネ' {#} m/(*ネ){#}ネ/>,
9b8f4e92 483 '/\cネ/' => "Character following \"\\c\" must be printable ASCII",
64935bc6
KW
484 '/\b{ネ}/' => "'ネ' is an unknown bound type {#} m/\\b{ネ{#}}/",
485 '/\B{ネ}/' => "'ネ' is an unknown bound type {#} m/\\B{ネ{#}}/",
c1d900c3
BF
486);
487push @death, @death_utf8;
488
67cdf558
KW
489my @death_utf8_only_under_strict = (
490 "m'ネ[\\y]ネ'" => 'Unrecognized escape \y in character class passed through {#} m/ネ[\y{#}]ネ/',
491 => 'Unrecognized escape \y in character class {#} m/ネ[\y{#}]ネ/',
492 'm/ネ[ネ-\d]ネ/' => 'False [] range "ネ-\d" {#} m/ネ[ネ-\d{#}]ネ/',
493 => 'False [] range "ネ-\d" {#} m/ネ[ネ-\d{#}]ネ/',
494 'm/ネ[\w-ネ]ネ/' => 'False [] range "\w-" {#} m/ネ[\w-{#}ネ]ネ/',
495 => 'False [] range "\w-" {#} m/ネ[\w-{#}ネ]ネ/',
496 'm/ネ[ネ-\pM]ネ/' => 'False [] range "ネ-\pM" {#} m/ネ[ネ-\pM{#}]ネ/',
497 => 'False [] range "ネ-\pM" {#} m/ネ[ネ-\pM{#}]ネ/',
498 '/ネ[ネ-[:digit:]]ネ/' => 'False [] range "ネ-[:digit:]" {#} m/ネ[ネ-[:digit:]{#}]ネ/',
499 => 'False [] range "ネ-[:digit:]" {#} m/ネ[ネ-[:digit:]{#}]ネ/',
500 '/ネ[\d-\s]ネ/' => 'False [] range "\d-" {#} m/ネ[\d-{#}\s]ネ/',
501 => 'False [] range "\d-" {#} m/ネ[\d-{#}\s]ネ/',
502 '/ネ[a\zb]ネ/' => 'Unrecognized escape \z in character class passed through {#} m/ネ[a\z{#}b]ネ/',
503 => 'Unrecognized escape \z in character class {#} m/ネ[a\z{#}b]ネ/',
504);
902994e4 505# Tests involving a user-defined charnames translator are in pat_advanced.t
b45f050a 506
af01601c
KW
507# In the following arrays of warnings, the value can be an array of things to
508# expect. If the empty string, it means no warning should be raised.
509
6d24e9d4 510
499333dc
KW
511# Key-value pairs of code/error of code that should have non-fatal regexp
512# warnings. Most currently have \x{100} appended to them to force them to be
513# upgraded to UTF-8, and the first pass restarted. Previously this would
514# cause some warnings to be output twice. This tests that that behavior has
515# been fixed.
6d24e9d4 516
499333dc
KW
517my @warning = (
518 'm/\b*\x{100}/' => '\b* matches null string many times {#} m/\b*{#}\x{100}/',
64935bc6
KW
519 '/\b{g}/a' => "Using /u for '\\b{g}' instead of /a {#} m/\\b{g}{#}/",
520 '/\B{gcb}/a' => "Using /u for '\\B{gcb}' instead of /a {#} m/\\B{gcb}{#}/",
499333dc 521 'm/[:blank:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:blank:]{#}\x{100}/',
7dbd4c44 522 'm/[[:cntrl:]][:^ascii:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[[:cntrl:]][:^ascii:]{#}\x{100}/',
46d34d0e
KW
523 'm/[[:ascii]]\x{100}/' => "Assuming NOT a POSIX class since there is no terminating ':' {#} m/[[:ascii{#}]]\\x{100}/",
524 'm/(?[[:word]])\x{100}/' => "Assuming NOT a POSIX class since there is no terminating ':' {#} m/(?[[:word{#}]])\\x{100}/",
499333dc 525 "m'\\y\\x{100}'" => 'Unrecognized escape \y passed through {#} m/\y{#}\x{100}/',
6d24e9d4
KW
526 '/x{3,1}/' => 'Quantifier {n,m} with n > m can\'t match {#} m/x{3,1}{#}/',
527 '/\08/' => '\'\08\' resolved to \'\o{0}8\' {#} m/\08{#}/',
528 '/\018/' => '\'\018\' resolved to \'\o{1}8\' {#} m/\018{#}/',
6d24e9d4
KW
529 '/(?=a)*/' => '(?=a)* matches null string many times {#} m/(?=a)*{#}/',
530 'my $x = \'\m\'; qr/a$x/' => 'Unrecognized escape \m passed through {#} m/a\m{#}/',
531 '/\q/' => 'Unrecognized escape \q passed through {#} m/\q{#}/',
282634ed 532 '/\q\p{Any}/' => 'Unrecognized escape \q passed through {#} m/\q{#}\p{Any}/',
499333dc 533
e1729dc6
FC
534 # These two tests do not include the marker, because regcomp.c no
535 # longer knows where it goes by the time this warning is emitted.
536 # See [perl #122680] regcomp warning gives wrong position of
499333dc 537 # problem.
e1729dc6
FC
538 '/(?=a){1,3}\x{100}/' => 'Quantifier unexpected on zero-length expression in regex m/(?=a){1,3}\x{100}/',
539 '/(a|b)(?=a){3}\x{100}/' => 'Quantifier unexpected on zero-length expression in regex m/(a|b)(?=a){3}\x{100}/',
499333dc 540
63fbd1cb 541 '/\_/' => "",
63fbd1cb 542 '/[\006]/' => "",
499333dc 543 '/[:alpha:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:alpha:]{#}\x{100}/',
46d34d0e
KW
544 '/[:zog:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/[:zog:]{#}\x{100}/',
545 '/[.zog.]\x{100}/' => 'POSIX syntax [. .] belongs inside character classes (but this one isn\'t implemented) {#} m/[.zog.]{#}\x{100}/',
63fbd1cb 546 '/[a-b]/' => "",
499333dc
KW
547 '/(?c)\x{100}/' => 'Useless (?c) - use /gc modifier {#} m/(?c{#})\x{100}/',
548 '/(?-c)\x{100}/' => 'Useless (?-c) - don\'t use /gc modifier {#} m/(?-c{#})\x{100}/',
549 '/(?g)\x{100}/' => 'Useless (?g) - use /g modifier {#} m/(?g{#})\x{100}/',
550 '/(?-g)\x{100}/' => 'Useless (?-g) - don\'t use /g modifier {#} m/(?-g{#})\x{100}/',
551 '/(?o)\x{100}/' => 'Useless (?o) - use /o modifier {#} m/(?o{#})\x{100}/',
552 '/(?-o)\x{100}/' => 'Useless (?-o) - don\'t use /o modifier {#} m/(?-o{#})\x{100}/',
553 '/(?g-o)\x{100}/' => [ 'Useless (?g) - use /g modifier {#} m/(?g{#}-o)\x{100}/',
554 'Useless (?-o) - don\'t use /o modifier {#} m/(?g-o{#})\x{100}/',
63fbd1cb 555 ],
499333dc
KW
556 '/(?g-c)\x{100}/' => [ 'Useless (?g) - use /g modifier {#} m/(?g{#}-c)\x{100}/',
557 'Useless (?-c) - don\'t use /gc modifier {#} m/(?g-c{#})\x{100}/',
63fbd1cb
KW
558 ],
559 # (?c) means (?g) error won't be thrown
499333dc
KW
560 '/(?o-cg)\x{100}/' => [ 'Useless (?o) - use /o modifier {#} m/(?o{#}-cg)\x{100}/',
561 'Useless (?-c) - don\'t use /gc modifier {#} m/(?o-c{#}g)\x{100}/',
63fbd1cb 562 ],
499333dc
KW
563 '/(?ogc)\x{100}/' => [ 'Useless (?o) - use /o modifier {#} m/(?o{#}gc)\x{100}/',
564 'Useless (?g) - use /g modifier {#} m/(?og{#}c)\x{100}/',
565 'Useless (?c) - use /gc modifier {#} m/(?ogc{#})\x{100}/',
63fbd1cb 566 ],
499333dc 567 '/a{1,1}?\x{100}/' => 'Useless use of greediness modifier \'?\' {#} m/a{1,1}?{#}\x{100}/',
c877af1b 568 "/(?[ [ % - % ] ])/" => "",
d4360efa
S
569 "/(?[ [ : - \\x$colon_hex ] ])\\x{100}/" => "\": - \\x$colon_hex \" is more clearly written simply as \":\" {#} m/(?[ [ : - \\x$colon_hex {#}] ])\\x{100}/",
570 "/(?[ [ \\x$colon_hex - : ] ])\\x{100}/" => "\"\\x$colon_hex\ - : \" is more clearly written simply as \":\" {#} m/(?[ [ \\x$colon_hex - : {#}] ])\\x{100}/",
571 "/(?[ [ \\t - \\x$tab_hex ] ])\\x{100}/" => "\"\\t - \\x$tab_hex \" is more clearly written simply as \"\\t\" {#} m/(?[ [ \\t - \\x$tab_hex {#}] ])\\x{100}/",
572 "/(?[ [ \\x$tab_hex - \\t ] ])\\x{100}/" => "\"\\x$tab_hex\ - \\t \" is more clearly written simply as \"\\t\" {#} m/(?[ [ \\x$tab_hex - \\t {#}] ])\\x{100}/",
c877af1b
KW
573 "/(?[ [ $B_hex - C ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ $B_hex - C {#}] ])/",
574 "/(?[ [ A - $B_hex ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ A - $B_hex {#}] ])/",
575 "/(?[ [ $low_mixed_alpha - $high_mixed_alpha ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ $low_mixed_alpha - $high_mixed_alpha {#}] ])/",
576 "/(?[ [ $low_mixed_digit - $high_mixed_digit ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ $low_mixed_digit - $high_mixed_digit {#}] ])/",
46d34d0e
KW
577 "/[alnum]/" => "",
578 "/[^alnum]/" => "",
579 '/[:blank]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/[:blank{#}]\x{100}/',
580 '/[[:digit]]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:digit{#}]]\x{100}/', # [perl # 8904]
581 '/[[:digit:foo]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:digit:{#}foo]\x{100}/',
582 '/[[:di#it:foo]\x{100}/x' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:di#it:{#}foo]\x{100}/',
583 '/[[:dgit]]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:dgit{#}]]\x{100}/',
584 '/[[:dgit:foo]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:dgit:{#}foo]\x{100}/',
585 '/[[:dgt]]\x{100}/' => "", # Far enough away from a real class to not be recognized as one
586 '/[[:dgt:foo]\x{100}/' => "",
587 '/[[:DIGIT]]\x{100}/' => [ 'Assuming NOT a POSIX class since the name must be all lowercase letters {#} m/[[:DIGIT{#}]]\x{100}/',
588 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:DIGIT{#}]]\x{100}/',
589 ],
590 '/[[digit]\x{100}/' => [ 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[{#}digit]\x{100}/',
591 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[digit{#}]\x{100}/',
592 ],
593 '/[[alpha]]\x{100}/' => [ 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[{#}alpha]]\x{100}/',
594 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[alpha{#}]]\x{100}/',
595 ],
596 '/[[^word]\x{100}/' => [ 'Assuming NOT a POSIX class since the \'^\' must come after the colon {#} m/[[^{#}word]\x{100}/',
597 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[^{#}word]\x{100}/',
598 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[^word{#}]\x{100}/',
599 ],
600 '/[[ ^ : x d i g i t : ] ]\x{100}/' => [ 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ {#}^ : x d i g i t : ] ]\x{100}/',
601 'Assuming NOT a POSIX class since the \'^\' must come after the colon {#} m/[[ ^{#} : x d i g i t : ] ]\x{100}/',
602 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ ^ {#}: x d i g i t : ] ]\x{100}/',
603 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ ^ : {#}x d i g i t : ] ]\x{100}/',
604 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ ^ : x d i g i t : ]{#} ]\x{100}/',
a4368cc3 605 $only_strict_marker . 'Unescaped literal \']\' {#} m/[[ ^ : x d i g i t : ] ]{#}\x{100}/',
46d34d0e
KW
606 ],
607 '/[foo:lower:]]\x{100}/' => 'Assuming NOT a POSIX class since it doesn\'t start with a \'[\' {#} m/[foo{#}:lower:]]\x{100}/',
608 '/[[;upper;]]\x{100}/' => [ 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[[;{#}upper;]]\x{100}/',
609 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[[;upper;]{#}]\x{100}/',
610 ],
611 '/[foo;punct;]]\x{100}/' => [ 'Assuming NOT a POSIX class since it doesn\'t start with a \'[\' {#} m/[foo{#};punct;]]\x{100}/',
612 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[foo;{#}punct;]]\x{100}/',
613 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[foo;punct;]{#}]\x{100}/',
614 ],
d730a801
YO
615 '/[][[:alpha:]]/' => "", # [perl #127581]
616 '/[][[:alpha:]\\@\\\\^_?]/' => "", # [perl #131522]
611d6243
KW
617 '/(?[[:w:]])/' => "",
618 '/([.].*)[.]/' => "", # [perl #127582]
619 '/[.].*[.]/' => "", # [perl #127604]
620 '/abc/xix' => "",
621 '/(?xmsixp:abc)/' => "",
622 '/(?xmsixp)abc/' => "",
623 '/(?xxxx:abc)/' => "",
624
499333dc 625); # See comments before this for why '\x{100}' is generally needed
6fe2934b 626
b0e1d434 627# These need the character 'ネ' as a marker for mark_as_utf8()
c1d900c3
BF
628my @warnings_utf8 = mark_as_utf8(
629 'm/ネ\b*ネ/' => '\b* matches null string many times {#} m/ネ\b*{#}ネ/',
630 '/(?=ネ)*/' => '(?=ネ)* matches null string many times {#} m/(?=ネ)*{#}/',
46d34d0e 631 'm/ネ[:foo:]ネ/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/ネ[:foo:]{#}ネ/',
b0e1d434 632 '/ネ(?c)ネ/' => 'Useless (?c) - use /gc modifier {#} m/ネ(?c{#})ネ/',
c1d900c3
BF
633 '/utf8 ネ (?ogc) ネ/' => [
634 'Useless (?o) - use /o modifier {#} m/utf8 ネ (?o{#}gc) ネ/',
635 'Useless (?g) - use /g modifier {#} m/utf8 ネ (?og{#}c) ネ/',
636 'Useless (?c) - use /gc modifier {#} m/utf8 ネ (?ogc{#}) ネ/',
637 ],
611d6243
KW
638 '/ネ[[:ネ:]]ネ/' => "",
639 '/ネ(?[[:ネ:]])ネ/' => "",
c1d900c3
BF
640
641);
642
643push @warning, @warnings_utf8;
644
63374c78 645my @warning_only_under_strict = (
b927b7e9
KW
646 '/[\N{U+00}-\x01]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+00}-\x01{#}]\x{100}/',
647 '/[\x00-\N{SOH}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\x00-\N{U+01}{#}]\x{100}/',
648 '/[\N{DEL}-\o{377}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+7F}-\o{377}{#}]\x{100}/',
649 '/[\o{0}-\N{U+01}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\o{0}-\N{U+01}{#}]\x{100}/',
650 '/[\000-\N{U+01}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\000-\N{U+01}{#}]\x{100}/',
651 '/[\N{DEL}-\377]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+7F}-\377{#}]\x{100}/',
c877af1b
KW
652 '/[\N{U+00}-A]\x{100}/' => 'Ranges of ASCII printables should be some subset of "0-9", "A-Z", or "a-z" {#} m/[\N{U+00}-A{#}]\x{100}/',
653 '/[a-\N{U+FF}]\x{100}/' => 'Ranges of ASCII printables should be some subset of "0-9", "A-Z", or "a-z" {#} m/[a-\N{U+FF}{#}]\x{100}/',
b927b7e9
KW
654 '/[\N{U+00}-\a]\x{100}/' => "",
655 '/[\a-\N{U+FF}]\x{100}/' => "",
656 '/[\N{U+FF}-\x{100}]/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+FF}-\x{100}{#}]/',
657 '/[\N{U+100}-\x{101}]/' => "",
c877af1b 658 "/[%-%]/" => "",
d4360efa
S
659 "/[:-\\x$colon_hex]\\x{100}/" => "\":-\\x$colon_hex\" is more clearly written simply as \":\" {#} m/[:-\\x$colon_hex\{#}]\\x{100}/",
660 "/[\\x$colon_hex-:]\\x{100}/" => "\"\\x$colon_hex-:\" is more clearly written simply as \":\" {#} m/[\\x$colon_hex\-:{#}]\\x{100}/",
661 "/[\\t-\\x$tab_hex]\\x{100}/" => "\"\\t-\\x$tab_hex\" is more clearly written simply as \"\\t\" {#} m/[\\t-\\x$tab_hex\{#}]\\x{100}/",
662 "/[\\x$tab_hex-\\t]\\x{100}/" => "\"\\x$tab_hex-\\t\" is more clearly written simply as \"\\t\" {#} m/[\\x$tab_hex\-\\t{#}]\\x{100}/",
c877af1b
KW
663 "/[$B_hex-C]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[$B_hex-C{#}]/",
664 "/[A-$B_hex]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[A-$B_hex\{#}]/",
665 "/[$low_mixed_alpha-$high_mixed_alpha]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[$low_mixed_alpha-$high_mixed_alpha\{#}]/",
666 "/[$low_mixed_digit-$high_mixed_digit]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[$low_mixed_digit-$high_mixed_digit\{#}]/",
a4368cc3
KW
667 '/\b<GCB}/' => 'Unescaped literal \'}\' {#} m/\b<GCB}{#}/',
668 '/[ ]def]/' => 'Unescaped literal \']\' {#} m/[ ]def]{#}/',
c877af1b
KW
669);
670
671my @warning_utf8_only_under_strict = mark_as_utf8(
672 '/ネ[᪉-᪐]/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ[᪉-᪐{#}]/",
673 '/ネ(?[ [ ᪉ - ᪐ ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ ᪉ - ᪐ {#}] ])/",
674 '/ネ[᧙-᧚]/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ[᧙-᧚{#}]/",
675 '/ネ(?[ [ ᧙ - ᧚ ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ ᧙ - ᧚ {#}] ])/",
8bafd282
KW
676 '/ネ(?[ [ 𝟘 - 𝟡 ] ])/; #no latin1' => "",
677 '/ネ(?[ [ 𝟧 - 𝟱 ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ 𝟧 - 𝟱 {#}] ])/",
678 '/ネ(?[ [ 𝟧 - 𝟰 ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ 𝟧 - 𝟰 {#}] ])/",
63374c78
KW
679);
680
c877af1b
KW
681push @warning_only_under_strict, @warning_utf8_only_under_strict;
682
6fe2934b 683my @experimental_regex_sets = (
6d24e9d4 684 '/(?[ \t ])/' => 'The regex_sets feature is experimental {#} m/(?[{#} \t ])/',
c1d900c3
BF
685 'use utf8; /utf8 ネ (?[ [\tネ] ])/' => do { use utf8; 'The regex_sets feature is experimental {#} m/utf8 ネ (?[{#} [\tネ] ])/' },
686 '/noutf8 ネ (?[ [\tネ] ])/' => 'The regex_sets feature is experimental {#} m/noutf8 ネ (?[{#} [\tネ] ])/',
b45f050a
JF
687);
688
034602eb 689my @experimental_script_run = (
d9790612
KW
690 '/(*script_run:paypal.com)/' => 'The script_run feature is experimental {#} m/(*script_run:{#}paypal.com)/',
691 'use utf8; /utf8 ネ (*script_run:ネ)/' => do { use utf8; 'The script_run feature is experimental {#} m/utf8 ネ (*script_run:{#}ネ)/' },
692 '/noutf8 ネ (*script_run:ネ)/' => 'The script_run feature is experimental {#} m/noutf8 ネ (*script_run:{#}ネ)/',
034602eb
KW
693);
694
63fbd1cb 695my @deprecated = (
8e84dec2
KW
696 '/^{/' => "",
697 '/foo|{/' => "",
698 '/foo|^{/' => "",
0367231c 699 '/foo({bar)/' => 'Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.32), passed through {#} m/foo({{#}bar)/',
8e84dec2
KW
700 '/foo(:?{bar)/' => "",
701 '/\s*{/' => "",
702 '/a{3,4}{/' => "",
63fbd1cb
KW
703);
704
67cdf558
KW
705for my $strict ("", "use re 'strict';") {
706
707 # First time just use @death; but under strict we add the things that fail
708 # there. Doing it this way makes sure that 'strict' doesnt change the
709 # things that are already fatal when not under strict.
710 if ($strict) {
711 for (my $i = 0; $i < @death_only_under_strict; $i += 3) {
712 push @death, $death_only_under_strict[$i], # The regex
713 $death_only_under_strict[$i+2]; # The fatal msg
714 }
715 for (my $i = 0; $i < @death_utf8_only_under_strict; $i += 3) {
716
717 # Same with the utf8 versions
718 push @death, mark_as_utf8($death_utf8_only_under_strict[$i],
719 $death_utf8_only_under_strict[$i+2]);
720 }
721 }
d6609144 722 for (my $i = 0; $i < @death; $i += 2) {
d30277c7 723 my $regex = $death[$i] =~ s/ default_ (on | off) //rx;
b1793c1a 724 my $expect = fixup_expect($death[$i+1], $strict);
1765a25a
KW
725 if ($expect eq "") {
726 fail("$0: Internal error: '$death[$i]' should have an error message");
727 }
728 else {
d4abdeed 729 no warnings 'experimental::regex_sets';
034602eb 730 no warnings 'experimental::script_run';
d4abdeed 731 no warnings 'experimental::re_strict';
cc38b95d 732 no warnings 'experimental::alpha_assertions';
d6609144 733
d4abdeed 734 warning_is(sub {
1765a25a 735 my $meaning_of_life;
d6609144
KW
736 my $eval_string = "$strict $regex";
737 $_ = "x";
1765a25a
KW
738 eval "$eval_string; \$meaning_of_life = 42";
739 ok (! defined $meaning_of_life, "$eval_string died");
af05e4c0
YO
740 my $error= $@;
741 if ($error =~ qr/\Q$expect/) {
1765a25a 742 ok(1, "... and gave expected message");
af05e4c0
YO
743 } else {
744 ok(0,$eval_string);
3fb37286
YO
745 diag("Have: " . _q(add_markers($error)));
746 diag("Want: " . _q($death[$i+1]));
af05e4c0 747 }
1765a25a
KW
748 }, undef, "... and no other warnings");
749 }
d6609144 750 }
67cdf558 751}
b45f050a 752
63374c78
KW
753for my $strict ("", "no warnings 'experimental::re_strict'; use re 'strict';") {
754 my @warning_tests = @warning;
67cdf558 755
63374c78
KW
756 # Build the tests for @warning. Use the strict/non-strict versions
757 # appropriately.
758 if ($strict) {
759 push @warning_tests, @warning_only_under_strict;
760 }
761 else {
762 for (my $i = 0; $i < @warning_only_under_strict; $i += 2) {
a4368cc3
KW
763
764 # (?[ ]) are always under strict
63374c78
KW
765 if ($warning_only_under_strict[$i] =~ /\Q(?[/) {
766 push @warning_tests, $warning_only_under_strict[$i], # The regex
767 $warning_only_under_strict[$i+1];
768 }
769 else {
770 push @warning_tests, $warning_only_under_strict[$i], # The regex
771 ""; # No warning because not strict
772 }
773 }
67cdf558 774 for (my $i = 0; $i < @death_only_under_strict; $i += 3) {
63374c78
KW
775 push @warning_tests, $death_only_under_strict[$i], # The regex
776 $death_only_under_strict[$i+1]; # The warning
67cdf558
KW
777 }
778 for (my $i = 0; $i < @death_utf8_only_under_strict; $i += 3) {
63374c78 779 push @warning_tests, mark_as_utf8($death_utf8_only_under_strict[$i],
d6609144 780 $death_utf8_only_under_strict[$i+1]);
67cdf558
KW
781 }
782 }
d6609144 783
034602eb
KW
784 foreach my $ref (\@warning_tests,
785 \@experimental_regex_sets,
786 \@experimental_script_run,
787 \@deprecated)
788 {
d6609144 789 my $warning_type;
63374c78 790 my $turn_off_warnings = "";
d6609144 791 my $default_on;
63374c78 792 if ($ref == \@warning_tests) {
d6609144 793 $warning_type = 'regexp, digit';
034602eb
KW
794 $turn_off_warnings = "no warnings 'experimental::regex_sets';"
795 . "no warnings 'experimental::script_run';";
d6609144 796 $default_on = $strict;
af01601c 797 }
d6609144
KW
798 elsif ($ref == \@deprecated) {
799 $warning_type = 'regexp, deprecated';
800 $default_on = 1;
801 }
034602eb 802 elsif ($ref == \@experimental_regex_sets) {
d6609144
KW
803 $warning_type = 'experimental::regex_sets';
804 $default_on = 1;
805 }
034602eb
KW
806 elsif ($ref == \@experimental_script_run) {
807 $warning_type = 'experimental::script_run';
808 $default_on = 1;
809 }
810 else {
811 fail("$0: Internal error: Unexpected loop variable");
812 }
813
d6609144 814 for (my $i = 0; $i < @$ref; $i += 2) {
d30277c7 815 my $this_default_on = $default_on;
d6609144 816 my $regex = $ref->[$i];
d30277c7
KW
817 if ($regex =~ s/ default_ (on | off) //x) {
818 $this_default_on = $1 eq 'on';
819 }
b1793c1a 820 my @expect = fixup_expect($ref->[$i+1], $strict);
63374c78
KW
821
822 # A length-1 array with an empty warning means no warning gets
823 # generated at all.
824 undef @expect if @expect == 1 && $expect[0] eq "";
825
d6609144
KW
826 {
827 $_ = "x";
d8d26cac
KW
828 #use feature 'unicode_eval';
829 #print STDERR __LINE__, ": ", "eval '$strict no warnings; $regex'", "\n";
d6609144 830 eval "$strict no warnings; $regex";
af01601c 831 }
d6609144
KW
832 if (is($@, "", "$strict $regex did not die")) {
833 my @got = capture_warnings(sub {
834 $_ = "x";
63374c78 835 eval "$strict $turn_off_warnings $regex" });
d6609144
KW
836 my $count = @expect;
837 if (! is(scalar @got, scalar @expect,
838 "... and gave expected number ($count) of warnings"))
839 {
840 if (@got < @expect) {
841 $count = @got;
842 note "Expected warnings not gotten:\n\t" . join "\n\t",
843 @expect[$count .. $#expect];
844 }
845 else {
846 note "Unexpected warnings gotten:\n\t" . join("\n\t",
847 @got[$count .. $#got]);
848 }
b0b90d25 849 }
d6609144
KW
850 foreach my $i (0 .. $count - 1) {
851 if (! like($got[$i], qr/\Q$expect[$i]/,
852 "... and gave expected warning"))
853 {
854 chomp($got[$i]);
855 chomp($expect[$i]);
856 diag("GOT\n'$got[$i]'\nEXPECT\n'$expect[$i]'");
67cdf558
KW
857 }
858 else {
d6609144
KW
859 ok (0 == capture_warnings(sub {
860 $_ = "x";
861 eval "$strict no warnings '$warning_type'; $regex;" }
862 ),
863 "... and turning off '$warning_type' warnings suppressed it");
864
865 # Test that whether the warning is on by default is
866 # correct. This test relies on the fact that we
867 # are outside the scope of any ‘use warnings’.
868 local $^W;
869 my @warns = capture_warnings(sub { $_ = "x";
870 eval "$strict $regex" });
63374c78
KW
871 # Warning should be on as well if is testing
872 # '(?[...])' which turns on strict
d30277c7 873 if ($this_default_on || grep { $_ =~ /\Q(?[/ } @expect ) {
d6609144
KW
874 ok @warns > 0, "... and the warning is on by default";
875 }
876 else {
877 ok @warns == 0, "... and the warning is off by default";
878 }
67cdf558 879 }
af01601c
KW
880 }
881 }
b42857f3 882 }
6fe2934b
KW
883 }
884}
b45f050a 885
cb79f740 886done_testing();