This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
t/re/reg_mesg.t: Add a couple of tests
[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';
cc7e6304 7 @INC = qw '../lib ../ext/re';
cb79f740 8 require './test.pl';
2b08d1e2 9 skip_all_without_unicode_tables();
cb79f740 10 eval 'require Config'; # assume defaults if this fails
b45f050a
JF
11}
12
cb79f740 13use strict;
c1d900c3 14use open qw(:utf8 :std);
b45f050a
JF
15
16##
17## If the markers used are changed (search for "MARKER1" in regcomp.c),
cb79f740 18## update only these two regexs, and leave the {#} in the @death/@warning
b45f050a
JF
19## arrays below. The {#} is a meta-marker -- it marks where the marker should
20## go.
cb79f740 21##
af01601c
KW
22## Returns empty string if that is what is expected. Otherwise, handles
23## either a scalar, turning it into a single element array; or a ref to an
24## array, adjusting each element. If called in array context, returns an
25## array, otherwise the join of all elements
26
cb79f740 27sub fixup_expect {
af01601c 28 my $expect_ref = shift;
63374c78 29 return "" if $expect_ref eq "";
af01601c
KW
30
31 my @expect;
32 if (ref $expect_ref) {
33 @expect = @$expect_ref;
34 }
35 else {
36 @expect = $expect_ref;
37 }
38
39 foreach my $element (@expect) {
6d24e9d4 40 $element =~ s/{\#}/in regex; marked by <-- HERE in/;
af01601c
KW
41 $element =~ s/{\#}/ <-- HERE /;
42 $element .= " at ";
43 }
44 return wantarray ? @expect : join "", @expect;
cb79f740 45}
b45f050a 46
c1d900c3 47## Because we don't "use utf8" in this file, we need to do some extra legwork
b0e1d434
KW
48## for the utf8 tests: Prepend 'use utf8' to the pattern, and mark the strings
49## to check against as UTF-8, but for this all to work properly, the character
50## 'ネ' (U+30CD) is required in each pattern somewhere as a marker.
4cabb89a
BF
51##
52## This also creates a second variant of the tests to check if the
b0e1d434
KW
53## latin1 error messages are working correctly. Because we don't 'use utf8',
54## we can't tell if something is UTF-8 or Latin1, so you need the suffix
55## '; no latin1' to not have the second variant.
4cabb89a
BF
56my $l1 = "\x{ef}";
57my $utf8 = "\x{30cd}";
58utf8::encode($utf8);
59
c1d900c3
BF
60sub mark_as_utf8 {
61 my @ret;
67cdf558
KW
62 for (my $i = 0; $i < @_; $i += 2) {
63 my $pat = $_[$i];
64 my $msg = $_[$i+1];
4cabb89a
BF
65 my $l1_pat = $pat =~ s/$utf8/$l1/gr;
66 my $l1_msg;
c1d900c3 67 $pat = "use utf8; $pat";
b0e1d434 68
c1d900c3 69 if (ref $msg) {
4cabb89a
BF
70 $l1_msg = [ map { s/$utf8/$l1/gr } @$msg ];
71 @$msg = map { my $c = $_; utf8::decode($c); $c } @$msg;
c1d900c3
BF
72 }
73 else {
4cabb89a 74 $l1_msg = $msg =~ s/$utf8/$l1/gr;
c1d900c3
BF
75 utf8::decode($msg);
76 }
77 push @ret, $pat => $msg;
b0e1d434 78
4cabb89a 79 push @ret, $l1_pat => $l1_msg unless $l1_pat =~ /#no latin1/;
c1d900c3
BF
80 }
81 return @ret;
82}
83
cb79f740
NC
84my $inf_m1 = ($Config::Config{reg_infty} || 32767) - 1;
85my $inf_p1 = $inf_m1 + 2;
b45f050a 86
c877af1b
KW
87my $B_hex = sprintf("\\x%02X", ord "B");
88my $low_mixed_alpha = ('A' lt 'a') ? 'A' : 'a';
89my $high_mixed_alpha = ('A' lt 'a') ? 'a' : 'A';
90my $low_mixed_digit = ('A' lt '0') ? 'A' : '0';
91my $high_mixed_digit = ('A' lt '0') ? '0' : 'A';
92
acdfc3b6
KW
93my $colon_hex = sprintf "%02X", ord(":");
94my $tab_hex = sprintf "%02X", ord("\t");
95
b45f050a
JF
96##
97## Key-value pairs of code/error of code that should have fatal errors.
98##
99my @death =
100(
6d24e9d4 101 '/[[=foo=]]/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/[[=foo=]{#}]/',
46d34d0e
KW
102 '/[[=a]=]]/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/[[=a]=]{#}]/',
103 '/[[.a].]]/' => 'POSIX syntax [. .] is reserved for future extensions {#} m/[[.a].]{#}]/',
b45f050a 104
58e23c8d 105 '/(?<= .*)/' => 'Variable length lookbehind not implemented in regex m/(?<= .*)/',
b45f050a 106
58e23c8d 107 '/(?<= x{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= x{1000})/',
b45f050a 108
6d24e9d4 109 '/(?@)/' => 'Sequence (?@...) not implemented {#} m/(?@{#})/',
b45f050a 110
9da1dd8f 111 '/(?{ 1/' => 'Missing right curly or square bracket',
b45f050a 112
6d24e9d4 113 '/(?(1x))/' => 'Switch condition not recognized {#} m/(?(1x{#}))/',
311cc1ad 114 '/(?(1x(?#)))/'=> 'Switch condition not recognized {#} m/(?(1x{#}(?#)))/',
b45f050a 115
8fb0127d
YO
116 '/(?(1)/' => 'Switch (?(condition)... not terminated {#} m/(?(1){#}/',
117 '/(?(1)x/' => 'Switch (?(condition)... not terminated {#} m/(?(1)x{#}/',
118 '/(?(1)x|y/' => 'Switch (?(condition)... not terminated {#} m/(?(1)x|y{#}/',
6d24e9d4 119 '/(?(1)x|y|z)/' => 'Switch (?(condition)... contains too many branches {#} m/(?(1)x|y|{#}z)/',
b45f050a 120
c1d900c3 121 '/(?(x)y|x)/' => 'Unknown switch condition (?(...)) {#} m/(?(x{#})y|x)/',
10380cb3
FC
122 '/(?(??{}))/' => 'Unknown switch condition (?(...)) {#} m/(?(?{#}?{}))/',
123 '/(?(?[]))/' => 'Unknown switch condition (?(...)) {#} m/(?(?{#}[]))/',
b45f050a 124
6d24e9d4 125 '/(?/' => 'Sequence (? incomplete {#} m/(?{#}/',
b45f050a 126
6d24e9d4
KW
127 '/(?;x/' => 'Sequence (?;...) not recognized {#} m/(?;{#}x/',
128 '/(?<;x/' => 'Group name must start with a non-digit word character {#} m/(?<;{#}x/',
129 '/(?\ix/' => 'Sequence (?\...) not recognized {#} m/(?\{#}ix/',
130 '/(?\mx/' => 'Sequence (?\...) not recognized {#} m/(?\{#}mx/',
131 '/(?\:x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}:x/',
132 '/(?\=x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}=x/',
133 '/(?\!x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}!x/',
134 '/(?\<=x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}<=x/',
135 '/(?\<!x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}<!x/',
136 '/(?\>x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}>x/',
137 '/(?^-i:foo)/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i:foo)/',
138 '/(?^-i)foo/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i)foo/',
139 '/(?^d:foo)/' => 'Sequence (?^d...) not recognized {#} m/(?^d{#}:foo)/',
140 '/(?^d)foo/' => 'Sequence (?^d...) not recognized {#} m/(?^d{#})foo/',
141 '/(?^lu:foo)/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/(?^lu{#}:foo)/',
142 '/(?^lu)foo/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/(?^lu{#})foo/',
143'/(?da:foo)/' => 'Regexp modifiers "d" and "a" are mutually exclusive {#} m/(?da{#}:foo)/',
144'/(?lil:foo)/' => 'Regexp modifier "l" may not appear twice {#} m/(?lil{#}:foo)/',
145'/(?aaia:foo)/' => 'Regexp modifier "a" may appear a maximum of twice {#} m/(?aaia{#}:foo)/',
146'/(?i-l:foo)/' => 'Regexp modifier "l" may not appear after the "-" {#} m/(?i-l{#}:foo)/',
cc74c5bd 147
6d24e9d4 148 '/((x)/' => 'Unmatched ( {#} m/({#}(x)/',
b45f050a 149
6d24e9d4 150 "/x{$inf_p1}/" => "Quantifier in {,} bigger than $inf_m1 {#} m/x{{#}$inf_p1}/",
b45f050a 151
b45f050a 152
6d24e9d4 153 '/x**/' => 'Nested quantifiers {#} m/x**{#}/',
b45f050a 154
6d24e9d4 155 '/x[/' => 'Unmatched [ {#} m/x[{#}/',
b45f050a 156
6d24e9d4 157 '/*/', => 'Quantifier follows nothing {#} m/*{#}/',
b45f050a 158
6d24e9d4 159 '/\p{x/' => 'Missing right brace on \p{} {#} m/\p{{#}x/',
b45f050a 160
6d24e9d4 161 '/[\p{x]/' => 'Missing right brace on \p{} {#} m/[\p{{#}x]/',
b45f050a 162
6d24e9d4 163 '/(x)\2/' => 'Reference to nonexistent group {#} m/(x)\2{#}/',
b45f050a 164
779fedd7 165 '/\g/' => 'Unterminated \g... pattern {#} m/\g{#}/',
76cccc4d
KW
166 '/\g{1/' => 'Unterminated \g{...} pattern {#} m/\g{1{#}/',
167
40809656 168 'my $m = "\\\"; $m =~ $m', => 'Trailing \ in regex m/\/',
b45f050a 169
6d24e9d4
KW
170 '/\x{1/' => 'Missing right brace on \x{} {#} m/\x{1{#}/',
171 '/\x{X/' => 'Missing right brace on \x{} {#} m/\x{{#}X/',
172
173 '/[\x{X]/' => 'Missing right brace on \x{} {#} m/[\x{{#}X]/',
174 '/[\x{A]/' => 'Missing right brace on \x{} {#} m/[\x{A{#}]/',
175
176 '/\o{1/' => 'Missing right brace on \o{ {#} m/\o{1{#}/',
177 '/\o{X/' => 'Missing right brace on \o{ {#} m/\o{{#}X/',
178
179 '/[\o{X]/' => 'Missing right brace on \o{ {#} m/[\o{{#}X]/',
180 '/[\o{7]/' => 'Missing right brace on \o{ {#} m/[\o{7{#}]/',
181
182 '/[[:barf:]]/' => 'POSIX class [:barf:] unknown {#} m/[[:barf:]{#}]/',
183
184 '/[[=barf=]]/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/[[=barf=]{#}]/',
185
186 '/[[.barf.]]/' => 'POSIX syntax [. .] is reserved for future extensions {#} m/[[.barf.]{#}]/',
187
188 '/[z-a]/' => 'Invalid [] range "z-a" {#} m/[z-a{#}]/',
189
c30c479a 190 '/\p/' => 'Empty \p {#} m/\p{#}/',
856d91b3
KW
191 '/\P/' => 'Empty \P {#} m/\P{#}/',
192 '/\p{}/' => 'Empty \p{} {#} m/\p{{#}}/',
6d24e9d4 193 '/\P{}/' => 'Empty \P{} {#} m/\P{{#}}/',
64935bc6
KW
194
195'/a\b{cde/' => 'Missing right brace on \b{} {#} m/a\b{{#}cde/',
196'/a\B{cde/' => 'Missing right brace on \B{} {#} m/a\B{{#}cde/',
197
198 '/\b{}/' => 'Empty \b{} {#} m/\b{}{#}/',
199 '/\B{}/' => 'Empty \B{} {#} m/\B{}{#}/',
200
201 '/\b{gc}/' => "'gc' is an unknown bound type {#} m/\\b{gc{#}}/",
202 '/\B{gc}/' => "'gc' is an unknown bound type {#} m/\\B{gc{#}}/",
203
46d34d0e
KW
204 '/(?[[[::]]])/' => "Syntax error in (?[...]) in regex m/(?[[[::]]])/",
205 '/(?[[[:w:]]])/' => "Syntax error in (?[...]) in regex m/(?[[[:w:]]])/",
206 '/(?[[:w:]])/' => "",
6d24e9d4 207 '/(?[a])/' => 'Unexpected character {#} m/(?[a{#}])/',
6d24e9d4
KW
208 '/(?[ + \t ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ +{#} \t ])/',
209 '/(?[ \cK - ( + \t ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ \cK - ( +{#} \t ) ])/',
210 '/(?[ \cK ( \t ) ])/' => 'Unexpected \'(\' with no preceding operator {#} m/(?[ \cK ({#} \t ) ])/',
211 '/(?[ \cK \t ])/' => 'Operand with no preceding operator {#} m/(?[ \cK \t{#} ])/',
212 '/(?[ \0004 ])/' => 'Need exactly 3 octal digits {#} m/(?[ \0004 {#}])/',
213 '/(?[ \05 ])/' => 'Need exactly 3 octal digits {#} m/(?[ \05 {#}])/',
214 '/(?[ \o{1038} ])/' => 'Non-octal character {#} m/(?[ \o{1038{#}} ])/',
215 '/(?[ \o{} ])/' => 'Number with no digits {#} m/(?[ \o{}{#} ])/',
216 '/(?[ \x{defg} ])/' => 'Non-hex character {#} m/(?[ \x{defg{#}} ])/',
217 '/(?[ \xabcdef ])/' => 'Use \\x{...} for more than two hex characters {#} m/(?[ \xabc{#}def ])/',
218 '/(?[ \x{} ])/' => 'Number with no digits {#} m/(?[ \x{}{#} ])/',
219 '/(?[ \cK + ) ])/' => 'Unexpected \')\' {#} m/(?[ \cK + ){#} ])/',
220 '/(?[ \cK + ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ \cK + {#}])/',
e7cce976 221 '/(?[ ( ) ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ ( ) {#}])/',
2985caa9 222 '/(?[[0]+()+])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[[0]+()+{#}])/',
29f52644
KW
223 '/(?[ \p{foo} ])/' => 'Can\'t find Unicode property definition "foo" {#} m/(?[ \p{foo}{#} ])/',
224 '/(?[ \p{ foo = bar } ])/' => 'Can\'t find Unicode property definition "foo = bar" {#} m/(?[ \p{ foo = bar }{#} ])/',
6d24e9d4 225 '/(?[ \8 ])/' => 'Unrecognized escape \8 in character class {#} m/(?[ \8{#} ])/',
3f4fde43
KW
226 '/(?[ \t ]/' => 'Syntax error in (?[...]) in regex m/(?[ \t ]/',
227 '/(?[ [ \t ]/' => 'Syntax error in (?[...]) in regex m/(?[ [ \t ]/',
228 '/(?[ \t ] ]/' => 'Syntax error in (?[...]) in regex m/(?[ \t ] ]/',
229 '/(?[ [ ] ]/' => 'Syntax error in (?[...]) in regex m/(?[ [ ] ]/',
230 '/(?[ \t + \e # This was supposed to be a comment ])/' => 'Syntax error in (?[...]) in regex m/(?[ \t + \e # This was supposed to be a comment ])/',
6d24e9d4
KW
231 '/(?[ ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ {#}])/',
232 'm/(?[[a-\d]])/' => 'False [] range "a-\d" {#} m/(?[[a-\d{#}]])/',
233 'm/(?[[\w-x]])/' => 'False [] range "\w-" {#} m/(?[[\w-{#}x]])/',
234 'm/(?[[a-\pM]])/' => 'False [] range "a-\pM" {#} m/(?[[a-\pM{#}]])/',
235 'm/(?[[\pM-x]])/' => 'False [] range "\pM-" {#} m/(?[[\pM-{#}x]])/',
8f0cd35a 236 '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
237 'm/(?[ \p{Digit} & (?(?[ \p{Thai} | \p{Lao} ]))])/' => 'Sequence (?(...) not recognized {#} m/(?[ \p{Digit} & (?({#}?[ \p{Thai} | \p{Lao} ]))])/',
238 'm/(?[ \p{Digit} & (?:(?[ \p{Thai} | \p{Lao} ]))])/' => 'Expecting \'(?flags:(?[...\' {#} m/(?[ \p{Digit} & (?{#}:(?[ \p{Thai} | \p{Lao} ]))])/',
6d24e9d4
KW
239 'm/\o{/' => 'Missing right brace on \o{ {#} m/\o{{#}/',
240 'm/\o/' => 'Missing braces on \o{} {#} m/\o{#}/',
241 'm/\o{}/' => 'Number with no digits {#} m/\o{}{#}/',
242 'm/[\o{]/' => 'Missing right brace on \o{ {#} m/[\o{{#}]/',
243 'm/[\o]/' => 'Missing braces on \o{} {#} m/[\o{#}]/',
244 'm/[\o{}]/' => 'Number with no digits {#} m/[\o{}{#}]/',
245 'm/(?^-i:foo)/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i:foo)/',
f1e1b256
YO
246 'm/\87/' => 'Reference to nonexistent group {#} m/\87{#}/',
247 'm/a\87/' => 'Reference to nonexistent group {#} m/a\87{#}/',
248 'm/a\97/' => 'Reference to nonexistent group {#} m/a\97{#}/',
07ea66ee
FC
249 'm/(*DOOF)/' => 'Unknown verb pattern \'DOOF\' {#} m/(*DOOF){#}/',
250 'm/(?&a/' => 'Sequence (?&... not terminated {#} m/(?&a{#}/',
75839571
FC
251 'm/(?P=/' => 'Sequence ?P=... not terminated {#} m/(?P={#}/',
252 "m/(?'/" => "Sequence (?'... not terminated {#} m/(?'{#}/",
253 "m/(?</" => "Sequence (?<... not terminated {#} m/(?<{#}/",
254 'm/(?&/' => 'Sequence (?&... not terminated {#} m/(?&{#}/',
255 'm/(?(</' => 'Sequence (?(<... not terminated {#} m/(?(<{#}/',
256 "m/(?('/" => "Sequence (?('... not terminated {#} m/(?('{#}/",
257 'm/\g{/' => 'Sequence \g{... not terminated {#} m/\g{{#}/',
258 'm/\k</' => 'Sequence \k<... not terminated {#} m/\k<{#}/',
9b8f4e92 259 'm/\cß/' => "Character following \"\\c\" must be printable ASCII",
cd209d9d
KW
260 '/((?# 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)/',
261 '/((?# 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)/',
5a55ed33 262 '/(?[\ &!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ &!{#}])/', # [perl #126180]
5982ac95 263 '/(?[()-!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[()-!{#}])/', # [perl #126204]
174c9902 264 '/(?[!()])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[!(){#}])/', # [perl #126404]
b45f050a 265);
c1d900c3 266
67cdf558
KW
267# These are messages that are warnings when not strict; death under 'use re
268# "strict". See comment before @warnings as to why some have a \x{100} in
269# them. This array has 3 elements per construct. [0] is the regex to use;
270# [1] is the message under no strict, and [2] is under strict.
271my @death_only_under_strict = (
272 'm/\xABC/' => "",
273 => 'Use \x{...} for more than two hex characters {#} m/\xABC{#}/',
274 'm/[\xABC]/' => "",
275 => 'Use \x{...} for more than two hex characters {#} m/[\xABC{#}]/',
276
277 # XXX This is a confusing error message. The G isn't ignored; it just
278 # terminates the \x. Also some messages below are missing the <-- HERE,
279 # aren't all category 'regexp'. (Hence we have to turn off 'digit'
280 # messages as well below)
281 'm/\xAG/' => 'Illegal hexadecimal digit \'G\' ignored',
282 => 'Non-hex character {#} m/\xAG{#}/',
283 'm/[\xAG]/' => 'Illegal hexadecimal digit \'G\' ignored',
284 => 'Non-hex character {#} m/[\xAG{#}]/',
285 'm/\o{789}/' => 'Non-octal character \'8\'. Resolved as "\o{7}"',
286 => 'Non-octal character {#} m/\o{78{#}9}/',
287 'm/[\o{789}]/' => 'Non-octal character \'8\'. Resolved as "\o{7}"',
288 => 'Non-octal character {#} m/[\o{78{#}9}]/',
289 'm/\x{}/' => "",
290 => 'Number with no digits {#} m/\x{}{#}/',
291 'm/[\x{}]/' => "",
292 => 'Number with no digits {#} m/[\x{}{#}]/',
293 'm/\x{ABCDEFG}/' => 'Illegal hexadecimal digit \'G\' ignored',
294 => 'Non-hex character {#} m/\x{ABCDEFG{#}}/',
295 'm/[\x{ABCDEFG}]/' => 'Illegal hexadecimal digit \'G\' ignored',
296 => 'Non-hex character {#} m/[\x{ABCDEFG{#}}]/',
67cdf558
KW
297 'm/[\N{}]/' => 'Ignoring zero length \\N{} in character class {#} m/[\\N{}{#}]/',
298 => 'Zero length \\N{} {#} m/[\\N{}]{#}/',
299 "m'[\\y]\\x{100}'" => 'Unrecognized escape \y in character class passed through {#} m/[\y{#}]\x{100}/',
300 => 'Unrecognized escape \y in character class {#} m/[\y{#}]\x{100}/',
301 'm/[a-\d]\x{100}/' => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
302 => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
303 'm/[\w-x]\x{100}/' => 'False [] range "\w-" {#} m/[\w-{#}x]\x{100}/',
304 => 'False [] range "\w-" {#} m/[\w-{#}x]\x{100}/',
305 'm/[a-\pM]\x{100}/' => 'False [] range "a-\pM" {#} m/[a-\pM{#}]\x{100}/',
306 => 'False [] range "a-\pM" {#} m/[a-\pM{#}]\x{100}/',
307 'm/[\pM-x]\x{100}/' => 'False [] range "\pM-" {#} m/[\pM-{#}x]\x{100}/',
308 => 'False [] range "\pM-" {#} m/[\pM-{#}x]\x{100}/',
309 '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}{#}]/',
310 => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[^\N{U+100.300{#}}]/',
311 '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}{#}]/',
312 => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[\x03-\N{U+100.300{#}}]/',
313 '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}]/',
314 => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[\N{U+100.300{#}}-\x{10FFFF}]/',
315 '/[\08]/' => '\'\08\' resolved to \'\o{0}8\' {#} m/[\08{#}]/',
316 => 'Need exactly 3 octal digits {#} m/[\08{#}]/',
317 '/[\018]/' => '\'\018\' resolved to \'\o{1}8\' {#} m/[\018{#}]/',
318 => 'Need exactly 3 octal digits {#} m/[\018{#}]/',
319 '/[\_\0]/' => "",
320 => 'Need exactly 3 octal digits {#} m/[\_\0]{#}/',
321 '/[\07]/' => "",
322 => 'Need exactly 3 octal digits {#} m/[\07]{#}/',
323 '/[\0005]/' => "",
324 => 'Need exactly 3 octal digits {#} m/[\0005]{#}/',
325 '/[\8\9]\x{100}/' => ['Unrecognized escape \8 in character class passed through {#} m/[\8{#}\9]\x{100}/',
326 'Unrecognized escape \9 in character class passed through {#} m/[\8\9{#}]\x{100}/',
327 ],
328 => 'Unrecognized escape \8 in character class {#} m/[\8{#}\9]\x{100}/',
329 '/[a-\d]\x{100}/' => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
330 => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
331 '/[\d-b]\x{100}/' => 'False [] range "\d-" {#} m/[\d-{#}b]\x{100}/',
332 => 'False [] range "\d-" {#} m/[\d-{#}b]\x{100}/',
333 '/[\s-\d]\x{100}/' => 'False [] range "\s-" {#} m/[\s-{#}\d]\x{100}/',
334 => 'False [] range "\s-" {#} m/[\s-{#}\d]\x{100}/',
335 '/[\d-\s]\x{100}/' => 'False [] range "\d-" {#} m/[\d-{#}\s]\x{100}/',
336 => 'False [] range "\d-" {#} m/[\d-{#}\s]\x{100}/',
337 '/[a-[:digit:]]\x{100}/' => 'False [] range "a-[:digit:]" {#} m/[a-[:digit:]{#}]\x{100}/',
338 => 'False [] range "a-[:digit:]" {#} m/[a-[:digit:]{#}]\x{100}/',
339 '/[[:digit:]-b]\x{100}/' => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}b]\x{100}/',
340 => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}b]\x{100}/',
341 '/[[:alpha:]-[:digit:]]\x{100}/' => 'False [] range "[:alpha:]-" {#} m/[[:alpha:]-{#}[:digit:]]\x{100}/',
342 => 'False [] range "[:alpha:]-" {#} m/[[:alpha:]-{#}[:digit:]]\x{100}/',
343 '/[[:digit:]-[:alpha:]]\x{100}/' => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}[:alpha:]]\x{100}/',
344 => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}[:alpha:]]\x{100}/',
345 '/[a\zb]\x{100}/' => 'Unrecognized escape \z in character class passed through {#} m/[a\z{#}b]\x{100}/',
346 => 'Unrecognized escape \z in character class {#} m/[a\z{#}b]\x{100}/',
347);
348
b0e1d434 349# These need the character 'ネ' as a marker for mark_as_utf8()
c1d900c3
BF
350my @death_utf8 = mark_as_utf8(
351 '/ネ[[=ネ=]]ネ/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/ネ[[=ネ=]{#}]ネ/',
352 '/ネ(?<= .*)/' => 'Variable length lookbehind not implemented in regex m/ネ(?<= .*)/',
353
354 '/(?<= ネ{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= ネ{1000})/',
355
356 '/ネ(?ネ)ネ/' => 'Sequence (?ネ...) not recognized {#} m/ネ(?ネ{#})ネ/',
357
476afc4b 358 '/ネ(?(1ネ))ネ/' => 'Switch condition not recognized {#} m/ネ(?(1ネ{#}))ネ/',
c1d900c3
BF
359
360 '/(?(1)ネ|y|ヌ)/' => 'Switch (?(condition)... contains too many branches {#} m/(?(1)ネ|y|{#}ヌ)/',
361
362 '/(?(ネ)y|ネ)/' => 'Unknown switch condition (?(...)) {#} m/(?(ネ{#})y|ネ)/',
363
364 '/ネ(?/' => 'Sequence (? incomplete {#} m/ネ(?{#}/',
365
366 '/ネ(?;ネ/' => 'Sequence (?;...) not recognized {#} m/ネ(?;{#}ネ/',
367 '/ネ(?<;ネ/' => 'Group name must start with a non-digit word character {#} m/ネ(?<;{#}ネ/',
368 '/ネ(?\ixネ/' => 'Sequence (?\...) not recognized {#} m/ネ(?\{#}ixネ/',
369 '/ネ(?^lu:ネ)/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/ネ(?^lu{#}:ネ)/',
370'/ネ(?lil:ネ)/' => 'Regexp modifier "l" may not appear twice {#} m/ネ(?lil{#}:ネ)/',
371'/ネ(?aaia:ネ)/' => 'Regexp modifier "a" may appear a maximum of twice {#} m/ネ(?aaia{#}:ネ)/',
372'/ネ(?i-l:ネ)/' => 'Regexp modifier "l" may not appear after the "-" {#} m/ネ(?i-l{#}:ネ)/',
373
374 '/ネ((ネ)/' => 'Unmatched ( {#} m/ネ({#}(ネ)/',
375
376 "/ネ{$inf_p1}ネ/" => "Quantifier in {,} bigger than $inf_m1 {#} m/ネ{{#}$inf_p1}ネ/",
377
378
379 '/ネ**ネ/' => 'Nested quantifiers {#} m/ネ**{#}ネ/',
380
381 '/ネ[ネ/' => 'Unmatched [ {#} m/ネ[{#}ネ/',
382
383 '/*ネ/', => 'Quantifier follows nothing {#} m/*{#}ネ/',
384
385 '/ネ\p{ネ/' => 'Missing right brace on \p{} {#} m/ネ\p{{#}ネ/',
386
387 '/(ネ)\2ネ/' => 'Reference to nonexistent group {#} m/(ネ)\2{#}ネ/',
388
4cabb89a 389 '/\g{ネ/; #no latin1' => 'Sequence \g{... not terminated {#} m/\g{ネ{#}/',
c1d900c3
BF
390
391 'my $m = "ネ\\\"; $m =~ $m', => 'Trailing \ in regex m/ネ\/',
392
393 '/\x{ネ/' => 'Missing right brace on \x{} {#} m/\x{{#}ネ/',
394 '/ネ[\x{ネ]ネ/' => 'Missing right brace on \x{} {#} m/ネ[\x{{#}ネ]ネ/',
395 '/ネ[\x{ネ]/' => 'Missing right brace on \x{} {#} m/ネ[\x{{#}ネ]/',
396
397 '/ネ\o{ネ/' => 'Missing right brace on \o{ {#} m/ネ\o{{#}ネ/',
46d34d0e 398 '/ネ[[:ネ:]]ネ/' => "",
c1d900c3
BF
399
400 '/ネ[[=ネ=]]ネ/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/ネ[[=ネ=]{#}]ネ/',
401
402 '/ネ[[.ネ.]]ネ/' => 'POSIX syntax [. .] is reserved for future extensions {#} m/ネ[[.ネ.]{#}]ネ/',
403
404 '/[ネ-a]ネ/' => 'Invalid [] range "ネ-a" {#} m/[ネ-a{#}]ネ/',
405
406 '/ネ\p{}ネ/' => 'Empty \p{} {#} m/ネ\p{{#}}ネ/',
407
46d34d0e
KW
408 '/ネ(?[[[:ネ]]])ネ/' => "Syntax error in (?[...]) in regex m/ネ(?[[[:ネ]]])ネ/",
409 '/ネ(?[[[:ネ: ])ネ/' => "Syntax error in (?[...]) in regex m/ネ(?[[[:ネ: ])ネ/",
410 '/ネ(?[[[::]]])ネ/' => "Syntax error in (?[...]) in regex m/ネ(?[[[::]]])ネ/",
411 '/ネ(?[[[:ネ:]]])ネ/' => "Syntax error in (?[...]) in regex m/ネ(?[[[:ネ:]]])ネ/",
412 '/ネ(?[[:ネ:]])ネ/' => "",
c1d900c3 413 '/ネ(?[ネ])ネ/' => 'Unexpected character {#} m/ネ(?[ネ{#}])ネ/',
c1d900c3
BF
414 '/ネ(?[ + [ネ] ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/ネ(?[ +{#} [ネ] ])/',
415 '/ネ(?[ \cK - ( + [ネ] ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/ネ(?[ \cK - ( +{#} [ネ] ) ])/',
416 '/ネ(?[ \cK ( [ネ] ) ])/' => 'Unexpected \'(\' with no preceding operator {#} m/ネ(?[ \cK ({#} [ネ] ) ])/',
417 '/ネ(?[ \cK [ネ] ])ネ/' => 'Operand with no preceding operator {#} m/ネ(?[ \cK [ネ{#}] ])ネ/',
418 '/ネ(?[ \0004 ])ネ/' => 'Need exactly 3 octal digits {#} m/ネ(?[ \0004 {#}])ネ/',
419 '/(?[ \o{ネ} ])ネ/' => 'Non-octal character {#} m/(?[ \o{ネ{#}} ])ネ/',
420 '/ネ(?[ \o{} ])ネ/' => 'Number with no digits {#} m/ネ(?[ \o{}{#} ])ネ/',
421 '/(?[ \x{ネ} ])ネ/' => 'Non-hex character {#} m/(?[ \x{ネ{#}} ])ネ/',
29f52644
KW
422 '/(?[ \p{ネ} ])/' => 'Can\'t find Unicode property definition "ネ" {#} m/(?[ \p{ネ}{#} ])/',
423 '/(?[ \p{ ネ = bar } ])/' => 'Can\'t find Unicode property definition "ネ = bar" {#} m/(?[ \p{ ネ = bar }{#} ])/',
c1d900c3
BF
424 '/ネ(?[ \t ]/' => 'Syntax error in (?[...]) in regex m/ネ(?[ \t ]/',
425 '/(?[ \t + \e # ネ This was supposed to be a comment ])/' => 'Syntax error in (?[...]) in regex m/(?[ \t + \e # ネ This was supposed to be a comment ])/',
9b8f4e92
KW
426 'm/(*ネ)ネ/' => q<Unknown verb pattern 'ネ' {#} m/(*ネ){#}ネ/>,
427 '/\cネ/' => "Character following \"\\c\" must be printable ASCII",
64935bc6
KW
428 '/\b{ネ}/' => "'ネ' is an unknown bound type {#} m/\\b{ネ{#}}/",
429 '/\B{ネ}/' => "'ネ' is an unknown bound type {#} m/\\B{ネ{#}}/",
c1d900c3
BF
430);
431push @death, @death_utf8;
432
67cdf558
KW
433my @death_utf8_only_under_strict = (
434 "m'ネ[\\y]ネ'" => 'Unrecognized escape \y in character class passed through {#} m/ネ[\y{#}]ネ/',
435 => 'Unrecognized escape \y in character class {#} m/ネ[\y{#}]ネ/',
436 'm/ネ[ネ-\d]ネ/' => 'False [] range "ネ-\d" {#} m/ネ[ネ-\d{#}]ネ/',
437 => 'False [] range "ネ-\d" {#} m/ネ[ネ-\d{#}]ネ/',
438 'm/ネ[\w-ネ]ネ/' => 'False [] range "\w-" {#} m/ネ[\w-{#}ネ]ネ/',
439 => 'False [] range "\w-" {#} m/ネ[\w-{#}ネ]ネ/',
440 'm/ネ[ネ-\pM]ネ/' => 'False [] range "ネ-\pM" {#} m/ネ[ネ-\pM{#}]ネ/',
441 => 'False [] range "ネ-\pM" {#} m/ネ[ネ-\pM{#}]ネ/',
442 '/ネ[ネ-[:digit:]]ネ/' => 'False [] range "ネ-[:digit:]" {#} m/ネ[ネ-[:digit:]{#}]ネ/',
443 => 'False [] range "ネ-[:digit:]" {#} m/ネ[ネ-[:digit:]{#}]ネ/',
444 '/ネ[\d-\s]ネ/' => 'False [] range "\d-" {#} m/ネ[\d-{#}\s]ネ/',
445 => 'False [] range "\d-" {#} m/ネ[\d-{#}\s]ネ/',
446 '/ネ[a\zb]ネ/' => 'Unrecognized escape \z in character class passed through {#} m/ネ[a\z{#}b]ネ/',
447 => 'Unrecognized escape \z in character class {#} m/ネ[a\z{#}b]ネ/',
448);
902994e4 449# Tests involving a user-defined charnames translator are in pat_advanced.t
b45f050a 450
af01601c
KW
451# In the following arrays of warnings, the value can be an array of things to
452# expect. If the empty string, it means no warning should be raised.
453
6d24e9d4 454
499333dc
KW
455# Key-value pairs of code/error of code that should have non-fatal regexp
456# warnings. Most currently have \x{100} appended to them to force them to be
457# upgraded to UTF-8, and the first pass restarted. Previously this would
458# cause some warnings to be output twice. This tests that that behavior has
459# been fixed.
6d24e9d4 460
499333dc
KW
461my @warning = (
462 'm/\b*\x{100}/' => '\b* matches null string many times {#} m/\b*{#}\x{100}/',
64935bc6
KW
463 '/\b{g}/a' => "Using /u for '\\b{g}' instead of /a {#} m/\\b{g}{#}/",
464 '/\B{gcb}/a' => "Using /u for '\\B{gcb}' instead of /a {#} m/\\B{gcb}{#}/",
499333dc 465 'm/[:blank:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:blank:]{#}\x{100}/',
7dbd4c44 466 'm/[[:cntrl:]][:^ascii:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[[:cntrl:]][:^ascii:]{#}\x{100}/',
46d34d0e
KW
467 'm/[[:ascii]]\x{100}/' => "Assuming NOT a POSIX class since there is no terminating ':' {#} m/[[:ascii{#}]]\\x{100}/",
468 'm/(?[[:word]])\x{100}/' => "Assuming NOT a POSIX class since there is no terminating ':' {#} m/(?[[:word{#}]])\\x{100}/",
499333dc 469 "m'\\y\\x{100}'" => 'Unrecognized escape \y passed through {#} m/\y{#}\x{100}/',
6d24e9d4
KW
470 '/x{3,1}/' => 'Quantifier {n,m} with n > m can\'t match {#} m/x{3,1}{#}/',
471 '/\08/' => '\'\08\' resolved to \'\o{0}8\' {#} m/\08{#}/',
472 '/\018/' => '\'\018\' resolved to \'\o{1}8\' {#} m/\018{#}/',
6d24e9d4
KW
473 '/(?=a)*/' => '(?=a)* matches null string many times {#} m/(?=a)*{#}/',
474 'my $x = \'\m\'; qr/a$x/' => 'Unrecognized escape \m passed through {#} m/a\m{#}/',
475 '/\q/' => 'Unrecognized escape \q passed through {#} m/\q{#}/',
499333dc 476
e1729dc6
FC
477 # These two tests do not include the marker, because regcomp.c no
478 # longer knows where it goes by the time this warning is emitted.
479 # See [perl #122680] regcomp warning gives wrong position of
499333dc 480 # problem.
e1729dc6
FC
481 '/(?=a){1,3}\x{100}/' => 'Quantifier unexpected on zero-length expression in regex m/(?=a){1,3}\x{100}/',
482 '/(a|b)(?=a){3}\x{100}/' => 'Quantifier unexpected on zero-length expression in regex m/(a|b)(?=a){3}\x{100}/',
499333dc 483
63fbd1cb 484 '/\_/' => "",
63fbd1cb 485 '/[\006]/' => "",
499333dc 486 '/[:alpha:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:alpha:]{#}\x{100}/',
46d34d0e
KW
487 '/[:zog:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/[:zog:]{#}\x{100}/',
488 '/[.zog.]\x{100}/' => 'POSIX syntax [. .] belongs inside character classes (but this one isn\'t implemented) {#} m/[.zog.]{#}\x{100}/',
489 '/[.z#g.]\x{100}/x' => "", # Runs into a comment
490 '/[.z\#g.]\x{100}/x' => 'POSIX syntax [. .] belongs inside character classes (but this one isn\'t implemented) {#} m/[.z\#g.]{#}\x{100}/',
63fbd1cb 491 '/[a-b]/' => "",
499333dc
KW
492 '/(?c)\x{100}/' => 'Useless (?c) - use /gc modifier {#} m/(?c{#})\x{100}/',
493 '/(?-c)\x{100}/' => 'Useless (?-c) - don\'t use /gc modifier {#} m/(?-c{#})\x{100}/',
494 '/(?g)\x{100}/' => 'Useless (?g) - use /g modifier {#} m/(?g{#})\x{100}/',
495 '/(?-g)\x{100}/' => 'Useless (?-g) - don\'t use /g modifier {#} m/(?-g{#})\x{100}/',
496 '/(?o)\x{100}/' => 'Useless (?o) - use /o modifier {#} m/(?o{#})\x{100}/',
497 '/(?-o)\x{100}/' => 'Useless (?-o) - don\'t use /o modifier {#} m/(?-o{#})\x{100}/',
498 '/(?g-o)\x{100}/' => [ 'Useless (?g) - use /g modifier {#} m/(?g{#}-o)\x{100}/',
499 'Useless (?-o) - don\'t use /o modifier {#} m/(?g-o{#})\x{100}/',
63fbd1cb 500 ],
499333dc
KW
501 '/(?g-c)\x{100}/' => [ 'Useless (?g) - use /g modifier {#} m/(?g{#}-c)\x{100}/',
502 'Useless (?-c) - don\'t use /gc modifier {#} m/(?g-c{#})\x{100}/',
63fbd1cb
KW
503 ],
504 # (?c) means (?g) error won't be thrown
499333dc
KW
505 '/(?o-cg)\x{100}/' => [ 'Useless (?o) - use /o modifier {#} m/(?o{#}-cg)\x{100}/',
506 'Useless (?-c) - don\'t use /gc modifier {#} m/(?o-c{#}g)\x{100}/',
63fbd1cb 507 ],
499333dc
KW
508 '/(?ogc)\x{100}/' => [ 'Useless (?o) - use /o modifier {#} m/(?o{#}gc)\x{100}/',
509 'Useless (?g) - use /g modifier {#} m/(?og{#}c)\x{100}/',
510 'Useless (?c) - use /gc modifier {#} m/(?ogc{#})\x{100}/',
63fbd1cb 511 ],
499333dc 512 '/a{1,1}?\x{100}/' => 'Useless use of greediness modifier \'?\' {#} m/a{1,1}?{#}\x{100}/',
c877af1b 513 "/(?[ [ % - % ] ])/" => "",
acdfc3b6
KW
514 "/(?[ [ : - \\x$colon_hex ] ])\\x{100}/" => "\": - \\x$colon_hex \" is more clearly written simply as \":\" {#} m/(?[ [ : - \\x$colon_hex {#}] ])\\x{100}/",
515 "/(?[ [ \\x$colon_hex - : ] ])\\x{100}/" => "\"\\x$colon_hex\ - : \" is more clearly written simply as \":\" {#} m/(?[ [ \\x$colon_hex - : {#}] ])\\x{100}/",
516 "/(?[ [ \\t - \\x$tab_hex ] ])\\x{100}/" => "\"\\t - \\x$tab_hex \" is more clearly written simply as \"\\t\" {#} m/(?[ [ \\t - \\x$tab_hex {#}] ])\\x{100}/",
517 "/(?[ [ \\x$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
518 "/(?[ [ $B_hex - C ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ $B_hex - C {#}] ])/",
519 "/(?[ [ A - $B_hex ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ A - $B_hex {#}] ])/",
520 "/(?[ [ $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 {#}] ])/",
521 "/(?[ [ $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
522 "/[alnum]/" => "",
523 "/[^alnum]/" => "",
524 '/[:blank]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/[:blank{#}]\x{100}/',
525 '/[[:digit]]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:digit{#}]]\x{100}/', # [perl # 8904]
526 '/[[:digit:foo]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:digit:{#}foo]\x{100}/',
527 '/[[:di#it:foo]\x{100}/x' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:di#it:{#}foo]\x{100}/',
528 '/[[:dgit]]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:dgit{#}]]\x{100}/',
529 '/[[:dgit:foo]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:dgit:{#}foo]\x{100}/',
530 '/[[:dgt]]\x{100}/' => "", # Far enough away from a real class to not be recognized as one
531 '/[[:dgt:foo]\x{100}/' => "",
532 '/[[:DIGIT]]\x{100}/' => [ 'Assuming NOT a POSIX class since the name must be all lowercase letters {#} m/[[:DIGIT{#}]]\x{100}/',
533 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:DIGIT{#}]]\x{100}/',
534 ],
535 '/[[digit]\x{100}/' => [ 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[{#}digit]\x{100}/',
536 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[digit{#}]\x{100}/',
537 ],
538 '/[[alpha]]\x{100}/' => [ 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[{#}alpha]]\x{100}/',
539 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[alpha{#}]]\x{100}/',
540 ],
541 '/[[^word]\x{100}/' => [ 'Assuming NOT a POSIX class since the \'^\' must come after the colon {#} m/[[^{#}word]\x{100}/',
542 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[^{#}word]\x{100}/',
543 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[^word{#}]\x{100}/',
544 ],
545 '/[[ ^ : 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}/',
546 'Assuming NOT a POSIX class since the \'^\' must come after the colon {#} m/[[ ^{#} : x d i g i t : ] ]\x{100}/',
547 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ ^ {#}: x d i g i t : ] ]\x{100}/',
548 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ ^ : {#}x d i g i t : ] ]\x{100}/',
549 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ ^ : x d i g i t : ]{#} ]\x{100}/',
550 ],
551 '/[foo:lower:]]\x{100}/' => 'Assuming NOT a POSIX class since it doesn\'t start with a \'[\' {#} m/[foo{#}:lower:]]\x{100}/',
552 '/[[;upper;]]\x{100}/' => [ 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[[;{#}upper;]]\x{100}/',
553 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[[;upper;]{#}]\x{100}/',
554 ],
555 '/[foo;punct;]]\x{100}/' => [ 'Assuming NOT a POSIX class since it doesn\'t start with a \'[\' {#} m/[foo{#};punct;]]\x{100}/',
556 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[foo;{#}punct;]]\x{100}/',
557 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[foo;punct;]{#}]\x{100}/',
558 ],
559
499333dc 560); # See comments before this for why '\x{100}' is generally needed
6fe2934b 561
b0e1d434 562# These need the character 'ネ' as a marker for mark_as_utf8()
c1d900c3
BF
563my @warnings_utf8 = mark_as_utf8(
564 'm/ネ\b*ネ/' => '\b* matches null string many times {#} m/ネ\b*{#}ネ/',
565 '/(?=ネ)*/' => '(?=ネ)* matches null string many times {#} m/(?=ネ)*{#}/',
46d34d0e 566 'm/ネ[:foo:]ネ/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/ネ[:foo:]{#}ネ/',
b0e1d434 567 '/ネ(?c)ネ/' => 'Useless (?c) - use /gc modifier {#} m/ネ(?c{#})ネ/',
c1d900c3
BF
568 '/utf8 ネ (?ogc) ネ/' => [
569 'Useless (?o) - use /o modifier {#} m/utf8 ネ (?o{#}gc) ネ/',
570 'Useless (?g) - use /g modifier {#} m/utf8 ネ (?og{#}c) ネ/',
571 'Useless (?c) - use /gc modifier {#} m/utf8 ネ (?ogc{#}) ネ/',
572 ],
573
574);
575
576push @warning, @warnings_utf8;
577
63374c78 578my @warning_only_under_strict = (
b927b7e9
KW
579 '/[\N{U+00}-\x01]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+00}-\x01{#}]\x{100}/',
580 '/[\x00-\N{SOH}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\x00-\N{U+01}{#}]\x{100}/',
581 '/[\N{DEL}-\o{377}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+7F}-\o{377}{#}]\x{100}/',
582 '/[\o{0}-\N{U+01}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\o{0}-\N{U+01}{#}]\x{100}/',
583 '/[\000-\N{U+01}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\000-\N{U+01}{#}]\x{100}/',
584 '/[\N{DEL}-\377]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+7F}-\377{#}]\x{100}/',
c877af1b
KW
585 '/[\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}/',
586 '/[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
587 '/[\N{U+00}-\a]\x{100}/' => "",
588 '/[\a-\N{U+FF}]\x{100}/' => "",
589 '/[\N{U+FF}-\x{100}]/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+FF}-\x{100}{#}]/',
590 '/[\N{U+100}-\x{101}]/' => "",
c877af1b 591 "/[%-%]/" => "",
acdfc3b6
KW
592 "/[:-\\x$colon_hex]\\x{100}/" => "\":-\\x$colon_hex\" is more clearly written simply as \":\" {#} m/[:-\\x$colon_hex\{#}]\\x{100}/",
593 "/[\\x$colon_hex-:]\\x{100}/" => "\"\\x$colon_hex-:\" is more clearly written simply as \":\" {#} m/[\\x$colon_hex\-:{#}]\\x{100}/",
594 "/[\\t-\\x$tab_hex]\\x{100}/" => "\"\\t-\\x$tab_hex\" is more clearly written simply as \"\\t\" {#} m/[\\t-\\x$tab_hex\{#}]\\x{100}/",
595 "/[\\x$tab_hex-\\t]\\x{100}/" => "\"\\x$tab_hex-\\t\" is more clearly written simply as \"\\t\" {#} m/[\\x$tab_hex\-\\t{#}]\\x{100}/",
c877af1b
KW
596 "/[$B_hex-C]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[$B_hex-C{#}]/",
597 "/[A-$B_hex]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[A-$B_hex\{#}]/",
598 "/[$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-$high_mixed_alpha\{#}]/",
599 "/[$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\{#}]/",
600);
601
602my @warning_utf8_only_under_strict = mark_as_utf8(
603 '/ネ[᪉-᪐]/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ[᪉-᪐{#}]/",
604 '/ネ(?[ [ ᪉ - ᪐ ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ ᪉ - ᪐ {#}] ])/",
605 '/ネ[᧙-᧚]/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ[᧙-᧚{#}]/",
606 '/ネ(?[ [ ᧙ - ᧚ ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ ᧙ - ᧚ {#}] ])/",
63374c78
KW
607);
608
c877af1b
KW
609push @warning_only_under_strict, @warning_utf8_only_under_strict;
610
6fe2934b 611my @experimental_regex_sets = (
6d24e9d4 612 '/(?[ \t ])/' => 'The regex_sets feature is experimental {#} m/(?[{#} \t ])/',
c1d900c3
BF
613 'use utf8; /utf8 ネ (?[ [\tネ] ])/' => do { use utf8; 'The regex_sets feature is experimental {#} m/utf8 ネ (?[{#} [\tネ] ])/' },
614 '/noutf8 ネ (?[ [\tネ] ])/' => 'The regex_sets feature is experimental {#} m/noutf8 ネ (?[{#} [\tネ] ])/',
b45f050a
JF
615);
616
63fbd1cb 617my @deprecated = (
412f55bb
KW
618 '/\w{/' => 'Unescaped left brace in regex is deprecated, passed through {#} m/\w{{#}/',
619 '/\q{/' => [
620 'Unrecognized escape \q{ passed through {#} m/\q{{#}/',
621 'Unescaped left brace in regex is deprecated, passed through {#} m/\q{{#}/'
622 ],
623 '/:{4,a}/' => 'Unescaped left brace in regex is deprecated, passed through {#} m/:{{#}4,a}/',
cc4d09e1
KW
624 '/abc/xix' => 'Having more than one /x regexp modifier is deprecated',
625 '/(?xmsixp:abc)/' => 'Having more than one /x regexp modifier is deprecated',
626 '/(?xmsixp)abc/' => 'Having more than one /x regexp modifier is deprecated',
627 '/(?xxxx:abc)/' => 'Having more than one /x regexp modifier is deprecated',
63fbd1cb
KW
628);
629
67cdf558
KW
630for my $strict ("", "use re 'strict';") {
631
632 # First time just use @death; but under strict we add the things that fail
633 # there. Doing it this way makes sure that 'strict' doesnt change the
634 # things that are already fatal when not under strict.
635 if ($strict) {
636 for (my $i = 0; $i < @death_only_under_strict; $i += 3) {
637 push @death, $death_only_under_strict[$i], # The regex
638 $death_only_under_strict[$i+2]; # The fatal msg
639 }
640 for (my $i = 0; $i < @death_utf8_only_under_strict; $i += 3) {
641
642 # Same with the utf8 versions
643 push @death, mark_as_utf8($death_utf8_only_under_strict[$i],
644 $death_utf8_only_under_strict[$i+2]);
645 }
646 }
d6609144
KW
647 for (my $i = 0; $i < @death; $i += 2) {
648 my $regex = $death[$i];
649 my $expect = fixup_expect($death[$i+1]);
650 no warnings 'experimental::regex_sets';
651 no warnings 'experimental::re_strict';
d6609144
KW
652
653 warning_is(sub {
654 my $eval_string = "$strict $regex";
655 $_ = "x";
656 eval $eval_string;
657 like($@, qr/\Q$expect/, $eval_string);
658 }, undef, "... and died without any other warnings");
659 }
67cdf558 660}
b45f050a 661
63374c78
KW
662for my $strict ("", "no warnings 'experimental::re_strict'; use re 'strict';") {
663 my @warning_tests = @warning;
67cdf558 664
63374c78
KW
665 # Build the tests for @warning. Use the strict/non-strict versions
666 # appropriately.
667 if ($strict) {
668 push @warning_tests, @warning_only_under_strict;
669 }
670 else {
671 for (my $i = 0; $i < @warning_only_under_strict; $i += 2) {
672 if ($warning_only_under_strict[$i] =~ /\Q(?[/) {
673 push @warning_tests, $warning_only_under_strict[$i], # The regex
674 $warning_only_under_strict[$i+1];
675 }
676 else {
677 push @warning_tests, $warning_only_under_strict[$i], # The regex
678 ""; # No warning because not strict
679 }
680 }
67cdf558 681 for (my $i = 0; $i < @death_only_under_strict; $i += 3) {
63374c78
KW
682 push @warning_tests, $death_only_under_strict[$i], # The regex
683 $death_only_under_strict[$i+1]; # The warning
67cdf558
KW
684 }
685 for (my $i = 0; $i < @death_utf8_only_under_strict; $i += 3) {
63374c78 686 push @warning_tests, mark_as_utf8($death_utf8_only_under_strict[$i],
d6609144 687 $death_utf8_only_under_strict[$i+1]);
67cdf558
KW
688 }
689 }
d6609144 690
63374c78 691 foreach my $ref (\@warning_tests, \@experimental_regex_sets, \@deprecated) {
d6609144 692 my $warning_type;
63374c78 693 my $turn_off_warnings = "";
d6609144 694 my $default_on;
63374c78 695 if ($ref == \@warning_tests) {
d6609144 696 $warning_type = 'regexp, digit';
63374c78 697 $turn_off_warnings = "no warnings 'experimental::regex_sets';";
d6609144 698 $default_on = $strict;
af01601c 699 }
d6609144
KW
700 elsif ($ref == \@deprecated) {
701 $warning_type = 'regexp, deprecated';
702 $default_on = 1;
703 }
704 else {
705 $warning_type = 'experimental::regex_sets';
706 $default_on = 1;
707 }
708 for (my $i = 0; $i < @$ref; $i += 2) {
709 my $regex = $ref->[$i];
710 my @expect = fixup_expect($ref->[$i+1]);
63374c78
KW
711
712 # A length-1 array with an empty warning means no warning gets
713 # generated at all.
714 undef @expect if @expect == 1 && $expect[0] eq "";
715
d6609144
KW
716 {
717 $_ = "x";
718 eval "$strict no warnings; $regex";
af01601c 719 }
d6609144
KW
720 if (is($@, "", "$strict $regex did not die")) {
721 my @got = capture_warnings(sub {
722 $_ = "x";
63374c78 723 eval "$strict $turn_off_warnings $regex" });
d6609144
KW
724 my $count = @expect;
725 if (! is(scalar @got, scalar @expect,
726 "... and gave expected number ($count) of warnings"))
727 {
728 if (@got < @expect) {
729 $count = @got;
730 note "Expected warnings not gotten:\n\t" . join "\n\t",
731 @expect[$count .. $#expect];
732 }
733 else {
734 note "Unexpected warnings gotten:\n\t" . join("\n\t",
735 @got[$count .. $#got]);
736 }
b0b90d25 737 }
d6609144
KW
738 foreach my $i (0 .. $count - 1) {
739 if (! like($got[$i], qr/\Q$expect[$i]/,
740 "... and gave expected warning"))
741 {
742 chomp($got[$i]);
743 chomp($expect[$i]);
744 diag("GOT\n'$got[$i]'\nEXPECT\n'$expect[$i]'");
67cdf558
KW
745 }
746 else {
d6609144
KW
747 ok (0 == capture_warnings(sub {
748 $_ = "x";
749 eval "$strict no warnings '$warning_type'; $regex;" }
750 ),
751 "... and turning off '$warning_type' warnings suppressed it");
752
753 # Test that whether the warning is on by default is
754 # correct. This test relies on the fact that we
755 # are outside the scope of any ‘use warnings’.
756 local $^W;
757 my @warns = capture_warnings(sub { $_ = "x";
758 eval "$strict $regex" });
63374c78
KW
759 # Warning should be on as well if is testing
760 # '(?[...])' which turns on strict
761 if ($default_on || grep { $_ =~ /\Q(?[/ } @expect ) {
d6609144
KW
762 ok @warns > 0, "... and the warning is on by default";
763 }
764 else {
765 ok @warns == 0, "... and the warning is off by default";
766 }
67cdf558 767 }
af01601c
KW
768 }
769 }
b42857f3 770 }
6fe2934b
KW
771 }
772}
b45f050a 773
cb79f740 774done_testing();