This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
regen/mk_invlists.pl: Add GCB tables
[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=]{#}]/',
b45f050a 102
58e23c8d 103 '/(?<= .*)/' => 'Variable length lookbehind not implemented in regex m/(?<= .*)/',
b45f050a 104
58e23c8d 105 '/(?<= x{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= x{1000})/',
b45f050a 106
6d24e9d4 107 '/(?@)/' => 'Sequence (?@...) not implemented {#} m/(?@{#})/',
b45f050a 108
9da1dd8f 109 '/(?{ 1/' => 'Missing right curly or square bracket',
b45f050a 110
6d24e9d4 111 '/(?(1x))/' => 'Switch condition not recognized {#} m/(?(1x{#}))/',
311cc1ad 112 '/(?(1x(?#)))/'=> 'Switch condition not recognized {#} m/(?(1x{#}(?#)))/',
b45f050a 113
8fb0127d
YO
114 '/(?(1)/' => 'Switch (?(condition)... not terminated {#} m/(?(1){#}/',
115 '/(?(1)x/' => 'Switch (?(condition)... not terminated {#} m/(?(1)x{#}/',
116 '/(?(1)x|y/' => 'Switch (?(condition)... not terminated {#} m/(?(1)x|y{#}/',
6d24e9d4 117 '/(?(1)x|y|z)/' => 'Switch (?(condition)... contains too many branches {#} m/(?(1)x|y|{#}z)/',
b45f050a 118
c1d900c3 119 '/(?(x)y|x)/' => 'Unknown switch condition (?(...)) {#} m/(?(x{#})y|x)/',
10380cb3
FC
120 '/(?(??{}))/' => 'Unknown switch condition (?(...)) {#} m/(?(?{#}?{}))/',
121 '/(?(?[]))/' => 'Unknown switch condition (?(...)) {#} m/(?(?{#}[]))/',
b45f050a 122
6d24e9d4 123 '/(?/' => 'Sequence (? incomplete {#} m/(?{#}/',
b45f050a 124
6d24e9d4
KW
125 '/(?;x/' => 'Sequence (?;...) not recognized {#} m/(?;{#}x/',
126 '/(?<;x/' => 'Group name must start with a non-digit word character {#} m/(?<;{#}x/',
127 '/(?\ix/' => 'Sequence (?\...) not recognized {#} m/(?\{#}ix/',
128 '/(?\mx/' => 'Sequence (?\...) not recognized {#} m/(?\{#}mx/',
129 '/(?\:x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}:x/',
130 '/(?\=x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}=x/',
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 '/(?^-i:foo)/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i:foo)/',
136 '/(?^-i)foo/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i)foo/',
137 '/(?^d:foo)/' => 'Sequence (?^d...) not recognized {#} m/(?^d{#}:foo)/',
138 '/(?^d)foo/' => 'Sequence (?^d...) not recognized {#} m/(?^d{#})foo/',
139 '/(?^lu:foo)/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/(?^lu{#}:foo)/',
140 '/(?^lu)foo/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/(?^lu{#})foo/',
141'/(?da:foo)/' => 'Regexp modifiers "d" and "a" are mutually exclusive {#} m/(?da{#}:foo)/',
142'/(?lil:foo)/' => 'Regexp modifier "l" may not appear twice {#} m/(?lil{#}:foo)/',
143'/(?aaia:foo)/' => 'Regexp modifier "a" may appear a maximum of twice {#} m/(?aaia{#}:foo)/',
144'/(?i-l:foo)/' => 'Regexp modifier "l" may not appear after the "-" {#} m/(?i-l{#}:foo)/',
32d02813
KW
145'/a\b{cde/' => 'Use "\b\{" instead of "\b{" {#} m/a\{#}b{cde/',
146'/a\B{cde/' => 'Use "\B\{" instead of "\B{" {#} m/a\{#}B{cde/',
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
190 '/\p/' => 'Empty \p{} {#} m/\p{#}/',
191
192 '/\P{}/' => 'Empty \P{} {#} m/\P{{#}}/',
193 '/(?[[[:word]]])/' => "Unmatched ':' in POSIX class {#} m/(?[[[:word{#}]]])/",
194 '/(?[[:word]])/' => "Unmatched ':' in POSIX class {#} m/(?[[:word{#}]])/",
195 '/(?[[[:digit: ])/' => "Unmatched '[' in POSIX class {#} m/(?[[[:digit:{#} ])/",
196 '/(?[[:digit: ])/' => "Unmatched '[' in POSIX class {#} m/(?[[:digit:{#} ])/",
197 '/(?[[[::]]])/' => "POSIX class [::] unknown {#} m/(?[[[::]{#}]])/",
198 '/(?[[[:w:]]])/' => "POSIX class [:w:] unknown {#} m/(?[[[:w:]{#}]])/",
199 '/(?[[:w:]])/' => "POSIX class [:w:] unknown {#} m/(?[[:w:]{#}])/",
200 '/(?[a])/' => 'Unexpected character {#} m/(?[a{#}])/',
201 '/(?[\t])/l' => '(?[...]) not valid in locale {#} m/(?[{#}\t])/',
202 '/(?[ + \t ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ +{#} \t ])/',
203 '/(?[ \cK - ( + \t ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ \cK - ( +{#} \t ) ])/',
204 '/(?[ \cK ( \t ) ])/' => 'Unexpected \'(\' with no preceding operator {#} m/(?[ \cK ({#} \t ) ])/',
205 '/(?[ \cK \t ])/' => 'Operand with no preceding operator {#} m/(?[ \cK \t{#} ])/',
206 '/(?[ \0004 ])/' => 'Need exactly 3 octal digits {#} m/(?[ \0004 {#}])/',
207 '/(?[ \05 ])/' => 'Need exactly 3 octal digits {#} m/(?[ \05 {#}])/',
208 '/(?[ \o{1038} ])/' => 'Non-octal character {#} m/(?[ \o{1038{#}} ])/',
209 '/(?[ \o{} ])/' => 'Number with no digits {#} m/(?[ \o{}{#} ])/',
210 '/(?[ \x{defg} ])/' => 'Non-hex character {#} m/(?[ \x{defg{#}} ])/',
211 '/(?[ \xabcdef ])/' => 'Use \\x{...} for more than two hex characters {#} m/(?[ \xabc{#}def ])/',
212 '/(?[ \x{} ])/' => 'Number with no digits {#} m/(?[ \x{}{#} ])/',
213 '/(?[ \cK + ) ])/' => 'Unexpected \')\' {#} m/(?[ \cK + ){#} ])/',
214 '/(?[ \cK + ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ \cK + {#}])/',
215 '/(?[ \p{foo} ])/' => 'Property \'foo\' is unknown {#} m/(?[ \p{foo}{#} ])/',
216 '/(?[ \p{ foo = bar } ])/' => 'Property \'foo = bar\' is unknown {#} m/(?[ \p{ foo = bar }{#} ])/',
217 '/(?[ \8 ])/' => 'Unrecognized escape \8 in character class {#} m/(?[ \8{#} ])/',
3f4fde43
KW
218 '/(?[ \t ]/' => 'Syntax error in (?[...]) in regex m/(?[ \t ]/',
219 '/(?[ [ \t ]/' => 'Syntax error in (?[...]) in regex m/(?[ [ \t ]/',
220 '/(?[ \t ] ]/' => 'Syntax error in (?[...]) in regex m/(?[ \t ] ]/',
221 '/(?[ [ ] ]/' => 'Syntax error in (?[...]) in regex m/(?[ [ ] ]/',
222 '/(?[ \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
223 '/(?[ ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ {#}])/',
224 'm/(?[[a-\d]])/' => 'False [] range "a-\d" {#} m/(?[[a-\d{#}]])/',
225 'm/(?[[\w-x]])/' => 'False [] range "\w-" {#} m/(?[[\w-{#}x]])/',
226 'm/(?[[a-\pM]])/' => 'False [] range "a-\pM" {#} m/(?[[a-\pM{#}]])/',
227 'm/(?[[\pM-x]])/' => 'False [] range "\pM-" {#} m/(?[[\pM-{#}x]])/',
8f0cd35a 228 '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
229 'm/(?[ \p{Digit} & (?(?[ \p{Thai} | \p{Lao} ]))])/' => 'Sequence (?(...) not recognized {#} m/(?[ \p{Digit} & (?({#}?[ \p{Thai} | \p{Lao} ]))])/',
230 'm/(?[ \p{Digit} & (?:(?[ \p{Thai} | \p{Lao} ]))])/' => 'Expecting \'(?flags:(?[...\' {#} m/(?[ \p{Digit} & (?{#}:(?[ \p{Thai} | \p{Lao} ]))])/',
6d24e9d4
KW
231 'm/\o{/' => 'Missing right brace on \o{ {#} m/\o{{#}/',
232 'm/\o/' => 'Missing braces on \o{} {#} m/\o{#}/',
233 'm/\o{}/' => 'Number with no digits {#} m/\o{}{#}/',
234 'm/[\o{]/' => 'Missing right brace on \o{ {#} m/[\o{{#}]/',
235 'm/[\o]/' => 'Missing braces on \o{} {#} m/[\o{#}]/',
236 'm/[\o{}]/' => 'Number with no digits {#} m/[\o{}{#}]/',
237 'm/(?^-i:foo)/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i:foo)/',
f1e1b256
YO
238 'm/\87/' => 'Reference to nonexistent group {#} m/\87{#}/',
239 'm/a\87/' => 'Reference to nonexistent group {#} m/a\87{#}/',
240 'm/a\97/' => 'Reference to nonexistent group {#} m/a\97{#}/',
07ea66ee
FC
241 'm/(*DOOF)/' => 'Unknown verb pattern \'DOOF\' {#} m/(*DOOF){#}/',
242 'm/(?&a/' => 'Sequence (?&... not terminated {#} m/(?&a{#}/',
75839571
FC
243 'm/(?P=/' => 'Sequence ?P=... not terminated {#} m/(?P={#}/',
244 "m/(?'/" => "Sequence (?'... not terminated {#} m/(?'{#}/",
245 "m/(?</" => "Sequence (?<... not terminated {#} m/(?<{#}/",
246 'm/(?&/' => 'Sequence (?&... not terminated {#} m/(?&{#}/',
247 'm/(?(</' => 'Sequence (?(<... not terminated {#} m/(?(<{#}/',
248 "m/(?('/" => "Sequence (?('... not terminated {#} m/(?('{#}/",
249 'm/\g{/' => 'Sequence \g{... not terminated {#} m/\g{{#}/',
250 'm/\k</' => 'Sequence \k<... not terminated {#} m/\k<{#}/',
9b8f4e92 251 'm/\cß/' => "Character following \"\\c\" must be printable ASCII",
cd209d9d
KW
252 '/((?# 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)/',
253 '/((?# 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)/',
b45f050a 254);
c1d900c3 255
67cdf558
KW
256# These are messages that are warnings when not strict; death under 'use re
257# "strict". See comment before @warnings as to why some have a \x{100} in
258# them. This array has 3 elements per construct. [0] is the regex to use;
259# [1] is the message under no strict, and [2] is under strict.
260my @death_only_under_strict = (
261 'm/\xABC/' => "",
262 => 'Use \x{...} for more than two hex characters {#} m/\xABC{#}/',
263 'm/[\xABC]/' => "",
264 => 'Use \x{...} for more than two hex characters {#} m/[\xABC{#}]/',
265
266 # XXX This is a confusing error message. The G isn't ignored; it just
267 # terminates the \x. Also some messages below are missing the <-- HERE,
268 # aren't all category 'regexp'. (Hence we have to turn off 'digit'
269 # messages as well below)
270 'm/\xAG/' => 'Illegal hexadecimal digit \'G\' ignored',
271 => 'Non-hex character {#} m/\xAG{#}/',
272 'm/[\xAG]/' => 'Illegal hexadecimal digit \'G\' ignored',
273 => 'Non-hex character {#} m/[\xAG{#}]/',
274 'm/\o{789}/' => 'Non-octal character \'8\'. Resolved as "\o{7}"',
275 => 'Non-octal character {#} m/\o{78{#}9}/',
276 'm/[\o{789}]/' => 'Non-octal character \'8\'. Resolved as "\o{7}"',
277 => 'Non-octal character {#} m/[\o{78{#}9}]/',
278 'm/\x{}/' => "",
279 => 'Number with no digits {#} m/\x{}{#}/',
280 'm/[\x{}]/' => "",
281 => 'Number with no digits {#} m/[\x{}{#}]/',
282 'm/\x{ABCDEFG}/' => 'Illegal hexadecimal digit \'G\' ignored',
283 => 'Non-hex character {#} m/\x{ABCDEFG{#}}/',
284 'm/[\x{ABCDEFG}]/' => 'Illegal hexadecimal digit \'G\' ignored',
285 => 'Non-hex character {#} m/[\x{ABCDEFG{#}}]/',
286 'm/[[:ascii]]/' => "",
287 => 'Unmatched \':\' in POSIX class {#} m/[[:ascii{#}]]/',
288 'm/[\N{}]/' => 'Ignoring zero length \\N{} in character class {#} m/[\\N{}{#}]/',
289 => 'Zero length \\N{} {#} m/[\\N{}]{#}/',
290 "m'[\\y]\\x{100}'" => 'Unrecognized escape \y in character class passed through {#} m/[\y{#}]\x{100}/',
291 => 'Unrecognized escape \y in character class {#} m/[\y{#}]\x{100}/',
292 'm/[a-\d]\x{100}/' => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
293 => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
294 'm/[\w-x]\x{100}/' => 'False [] range "\w-" {#} m/[\w-{#}x]\x{100}/',
295 => 'False [] range "\w-" {#} m/[\w-{#}x]\x{100}/',
296 'm/[a-\pM]\x{100}/' => 'False [] range "a-\pM" {#} m/[a-\pM{#}]\x{100}/',
297 => 'False [] range "a-\pM" {#} m/[a-\pM{#}]\x{100}/',
298 'm/[\pM-x]\x{100}/' => 'False [] range "\pM-" {#} m/[\pM-{#}x]\x{100}/',
299 => 'False [] range "\pM-" {#} m/[\pM-{#}x]\x{100}/',
300 '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}{#}]/',
301 => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[^\N{U+100.300{#}}]/',
302 '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}{#}]/',
303 => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[\x03-\N{U+100.300{#}}]/',
304 '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}]/',
305 => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[\N{U+100.300{#}}-\x{10FFFF}]/',
306 '/[\08]/' => '\'\08\' resolved to \'\o{0}8\' {#} m/[\08{#}]/',
307 => 'Need exactly 3 octal digits {#} m/[\08{#}]/',
308 '/[\018]/' => '\'\018\' resolved to \'\o{1}8\' {#} m/[\018{#}]/',
309 => 'Need exactly 3 octal digits {#} m/[\018{#}]/',
310 '/[\_\0]/' => "",
311 => 'Need exactly 3 octal digits {#} m/[\_\0]{#}/',
312 '/[\07]/' => "",
313 => 'Need exactly 3 octal digits {#} m/[\07]{#}/',
314 '/[\0005]/' => "",
315 => 'Need exactly 3 octal digits {#} m/[\0005]{#}/',
316 '/[\8\9]\x{100}/' => ['Unrecognized escape \8 in character class passed through {#} m/[\8{#}\9]\x{100}/',
317 'Unrecognized escape \9 in character class passed through {#} m/[\8\9{#}]\x{100}/',
318 ],
319 => 'Unrecognized escape \8 in character class {#} m/[\8{#}\9]\x{100}/',
320 '/[a-\d]\x{100}/' => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
321 => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
322 '/[\d-b]\x{100}/' => 'False [] range "\d-" {#} m/[\d-{#}b]\x{100}/',
323 => 'False [] range "\d-" {#} m/[\d-{#}b]\x{100}/',
324 '/[\s-\d]\x{100}/' => 'False [] range "\s-" {#} m/[\s-{#}\d]\x{100}/',
325 => 'False [] range "\s-" {#} m/[\s-{#}\d]\x{100}/',
326 '/[\d-\s]\x{100}/' => 'False [] range "\d-" {#} m/[\d-{#}\s]\x{100}/',
327 => 'False [] range "\d-" {#} m/[\d-{#}\s]\x{100}/',
328 '/[a-[:digit:]]\x{100}/' => 'False [] range "a-[:digit:]" {#} m/[a-[:digit:]{#}]\x{100}/',
329 => 'False [] range "a-[:digit:]" {#} m/[a-[:digit:]{#}]\x{100}/',
330 '/[[:digit:]-b]\x{100}/' => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}b]\x{100}/',
331 => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}b]\x{100}/',
332 '/[[:alpha:]-[:digit:]]\x{100}/' => 'False [] range "[:alpha:]-" {#} m/[[:alpha:]-{#}[:digit:]]\x{100}/',
333 => 'False [] range "[:alpha:]-" {#} m/[[:alpha:]-{#}[:digit:]]\x{100}/',
334 '/[[:digit:]-[:alpha:]]\x{100}/' => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}[:alpha:]]\x{100}/',
335 => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}[:alpha:]]\x{100}/',
336 '/[a\zb]\x{100}/' => 'Unrecognized escape \z in character class passed through {#} m/[a\z{#}b]\x{100}/',
337 => 'Unrecognized escape \z in character class {#} m/[a\z{#}b]\x{100}/',
338);
339
b0e1d434 340# These need the character 'ネ' as a marker for mark_as_utf8()
c1d900c3
BF
341my @death_utf8 = mark_as_utf8(
342 '/ネ[[=ネ=]]ネ/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/ネ[[=ネ=]{#}]ネ/',
343 '/ネ(?<= .*)/' => 'Variable length lookbehind not implemented in regex m/ネ(?<= .*)/',
344
345 '/(?<= ネ{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= ネ{1000})/',
346
347 '/ネ(?ネ)ネ/' => 'Sequence (?ネ...) not recognized {#} m/ネ(?ネ{#})ネ/',
348
476afc4b 349 '/ネ(?(1ネ))ネ/' => 'Switch condition not recognized {#} m/ネ(?(1ネ{#}))ネ/',
c1d900c3
BF
350
351 '/(?(1)ネ|y|ヌ)/' => 'Switch (?(condition)... contains too many branches {#} m/(?(1)ネ|y|{#}ヌ)/',
352
353 '/(?(ネ)y|ネ)/' => 'Unknown switch condition (?(...)) {#} m/(?(ネ{#})y|ネ)/',
354
355 '/ネ(?/' => 'Sequence (? incomplete {#} m/ネ(?{#}/',
356
357 '/ネ(?;ネ/' => 'Sequence (?;...) not recognized {#} m/ネ(?;{#}ネ/',
358 '/ネ(?<;ネ/' => 'Group name must start with a non-digit word character {#} m/ネ(?<;{#}ネ/',
359 '/ネ(?\ixネ/' => 'Sequence (?\...) not recognized {#} m/ネ(?\{#}ixネ/',
360 '/ネ(?^lu:ネ)/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/ネ(?^lu{#}:ネ)/',
361'/ネ(?lil:ネ)/' => 'Regexp modifier "l" may not appear twice {#} m/ネ(?lil{#}:ネ)/',
362'/ネ(?aaia:ネ)/' => 'Regexp modifier "a" may appear a maximum of twice {#} m/ネ(?aaia{#}:ネ)/',
363'/ネ(?i-l:ネ)/' => 'Regexp modifier "l" may not appear after the "-" {#} m/ネ(?i-l{#}:ネ)/',
364
365 '/ネ((ネ)/' => 'Unmatched ( {#} m/ネ({#}(ネ)/',
366
367 "/ネ{$inf_p1}ネ/" => "Quantifier in {,} bigger than $inf_m1 {#} m/ネ{{#}$inf_p1}ネ/",
368
369
370 '/ネ**ネ/' => 'Nested quantifiers {#} m/ネ**{#}ネ/',
371
372 '/ネ[ネ/' => 'Unmatched [ {#} m/ネ[{#}ネ/',
373
374 '/*ネ/', => 'Quantifier follows nothing {#} m/*{#}ネ/',
375
376 '/ネ\p{ネ/' => 'Missing right brace on \p{} {#} m/ネ\p{{#}ネ/',
377
378 '/(ネ)\2ネ/' => 'Reference to nonexistent group {#} m/(ネ)\2{#}ネ/',
379
4cabb89a 380 '/\g{ネ/; #no latin1' => 'Sequence \g{... not terminated {#} m/\g{ネ{#}/',
c1d900c3
BF
381
382 'my $m = "ネ\\\"; $m =~ $m', => 'Trailing \ in regex m/ネ\/',
383
384 '/\x{ネ/' => 'Missing right brace on \x{} {#} m/\x{{#}ネ/',
385 '/ネ[\x{ネ]ネ/' => 'Missing right brace on \x{} {#} m/ネ[\x{{#}ネ]ネ/',
386 '/ネ[\x{ネ]/' => 'Missing right brace on \x{} {#} m/ネ[\x{{#}ネ]/',
387
388 '/ネ\o{ネ/' => 'Missing right brace on \o{ {#} m/ネ\o{{#}ネ/',
389 '/ネ[[:ネ:]]ネ/' => 'POSIX class [:ネ:] unknown {#} m/ネ[[:ネ:]{#}]ネ/',
390
391 '/ネ[[=ネ=]]ネ/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/ネ[[=ネ=]{#}]ネ/',
392
393 '/ネ[[.ネ.]]ネ/' => 'POSIX syntax [. .] is reserved for future extensions {#} m/ネ[[.ネ.]{#}]ネ/',
394
395 '/[ネ-a]ネ/' => 'Invalid [] range "ネ-a" {#} m/[ネ-a{#}]ネ/',
396
397 '/ネ\p{}ネ/' => 'Empty \p{} {#} m/ネ\p{{#}}ネ/',
398
399 '/ネ(?[[[:ネ]]])ネ/' => "Unmatched ':' in POSIX class {#} m/ネ(?[[[:ネ{#}]]])ネ/",
400 '/ネ(?[[[:ネ: ])ネ/' => "Unmatched '[' in POSIX class {#} m/ネ(?[[[:ネ:{#} ])ネ/",
401 '/ネ(?[[[::]]])ネ/' => "POSIX class [::] unknown {#} m/ネ(?[[[::]{#}]])ネ/",
402 '/ネ(?[[[:ネ:]]])ネ/' => "POSIX class [:ネ:] unknown {#} m/ネ(?[[[:ネ:]{#}]])ネ/",
403 '/ネ(?[[:ネ:]])ネ/' => "POSIX class [:ネ:] unknown {#} m/ネ(?[[:ネ:]{#}])ネ/",
404 '/ネ(?[ネ])ネ/' => 'Unexpected character {#} m/ネ(?[ネ{#}])ネ/',
405 '/ネ(?[ネ])/l' => '(?[...]) not valid in locale {#} m/ネ(?[{#}ネ])/',
406 '/ネ(?[ + [ネ] ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/ネ(?[ +{#} [ネ] ])/',
407 '/ネ(?[ \cK - ( + [ネ] ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/ネ(?[ \cK - ( +{#} [ネ] ) ])/',
408 '/ネ(?[ \cK ( [ネ] ) ])/' => 'Unexpected \'(\' with no preceding operator {#} m/ネ(?[ \cK ({#} [ネ] ) ])/',
409 '/ネ(?[ \cK [ネ] ])ネ/' => 'Operand with no preceding operator {#} m/ネ(?[ \cK [ネ{#}] ])ネ/',
410 '/ネ(?[ \0004 ])ネ/' => 'Need exactly 3 octal digits {#} m/ネ(?[ \0004 {#}])ネ/',
411 '/(?[ \o{ネ} ])ネ/' => 'Non-octal character {#} m/(?[ \o{ネ{#}} ])ネ/',
412 '/ネ(?[ \o{} ])ネ/' => 'Number with no digits {#} m/ネ(?[ \o{}{#} ])ネ/',
413 '/(?[ \x{ネ} ])ネ/' => 'Non-hex character {#} m/(?[ \x{ネ{#}} ])ネ/',
414 '/(?[ \p{ネ} ])/' => 'Property \'ネ\' is unknown {#} m/(?[ \p{ネ}{#} ])/',
415 '/(?[ \p{ ネ = bar } ])/' => 'Property \'ネ = bar\' is unknown {#} m/(?[ \p{ ネ = bar }{#} ])/',
416 '/ネ(?[ \t ]/' => 'Syntax error in (?[...]) in regex m/ネ(?[ \t ]/',
417 '/(?[ \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
418 'm/(*ネ)ネ/' => q<Unknown verb pattern 'ネ' {#} m/(*ネ){#}ネ/>,
419 '/\cネ/' => "Character following \"\\c\" must be printable ASCII",
c1d900c3
BF
420);
421push @death, @death_utf8;
422
67cdf558
KW
423my @death_utf8_only_under_strict = (
424 "m'ネ[\\y]ネ'" => 'Unrecognized escape \y in character class passed through {#} m/ネ[\y{#}]ネ/',
425 => 'Unrecognized escape \y in character class {#} m/ネ[\y{#}]ネ/',
426 'm/ネ[ネ-\d]ネ/' => 'False [] range "ネ-\d" {#} m/ネ[ネ-\d{#}]ネ/',
427 => 'False [] range "ネ-\d" {#} m/ネ[ネ-\d{#}]ネ/',
428 'm/ネ[\w-ネ]ネ/' => 'False [] range "\w-" {#} m/ネ[\w-{#}ネ]ネ/',
429 => 'False [] range "\w-" {#} m/ネ[\w-{#}ネ]ネ/',
430 'm/ネ[ネ-\pM]ネ/' => 'False [] range "ネ-\pM" {#} m/ネ[ネ-\pM{#}]ネ/',
431 => 'False [] range "ネ-\pM" {#} m/ネ[ネ-\pM{#}]ネ/',
432 '/ネ[ネ-[:digit:]]ネ/' => 'False [] range "ネ-[:digit:]" {#} m/ネ[ネ-[:digit:]{#}]ネ/',
433 => 'False [] range "ネ-[:digit:]" {#} m/ネ[ネ-[:digit:]{#}]ネ/',
434 '/ネ[\d-\s]ネ/' => 'False [] range "\d-" {#} m/ネ[\d-{#}\s]ネ/',
435 => 'False [] range "\d-" {#} m/ネ[\d-{#}\s]ネ/',
436 '/ネ[a\zb]ネ/' => 'Unrecognized escape \z in character class passed through {#} m/ネ[a\z{#}b]ネ/',
437 => 'Unrecognized escape \z in character class {#} m/ネ[a\z{#}b]ネ/',
438);
902994e4 439# Tests involving a user-defined charnames translator are in pat_advanced.t
b45f050a 440
af01601c
KW
441# In the following arrays of warnings, the value can be an array of things to
442# expect. If the empty string, it means no warning should be raised.
443
6d24e9d4 444
499333dc
KW
445# Key-value pairs of code/error of code that should have non-fatal regexp
446# warnings. Most currently have \x{100} appended to them to force them to be
447# upgraded to UTF-8, and the first pass restarted. Previously this would
448# cause some warnings to be output twice. This tests that that behavior has
449# been fixed.
6d24e9d4 450
499333dc
KW
451my @warning = (
452 'm/\b*\x{100}/' => '\b* matches null string many times {#} m/\b*{#}\x{100}/',
453 'm/[:blank:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:blank:]{#}\x{100}/',
7dbd4c44 454 'm/[[:cntrl:]][:^ascii:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[[:cntrl:]][:^ascii:]{#}\x{100}/',
499333dc 455 "m'\\y\\x{100}'" => 'Unrecognized escape \y passed through {#} m/\y{#}\x{100}/',
6d24e9d4
KW
456 '/x{3,1}/' => 'Quantifier {n,m} with n > m can\'t match {#} m/x{3,1}{#}/',
457 '/\08/' => '\'\08\' resolved to \'\o{0}8\' {#} m/\08{#}/',
458 '/\018/' => '\'\018\' resolved to \'\o{1}8\' {#} m/\018{#}/',
6d24e9d4
KW
459 '/(?=a)*/' => '(?=a)* matches null string many times {#} m/(?=a)*{#}/',
460 'my $x = \'\m\'; qr/a$x/' => 'Unrecognized escape \m passed through {#} m/a\m{#}/',
461 '/\q/' => 'Unrecognized escape \q passed through {#} m/\q{#}/',
499333dc 462
e1729dc6
FC
463 # These two tests do not include the marker, because regcomp.c no
464 # longer knows where it goes by the time this warning is emitted.
465 # See [perl #122680] regcomp warning gives wrong position of
499333dc 466 # problem.
e1729dc6
FC
467 '/(?=a){1,3}\x{100}/' => 'Quantifier unexpected on zero-length expression in regex m/(?=a){1,3}\x{100}/',
468 '/(a|b)(?=a){3}\x{100}/' => 'Quantifier unexpected on zero-length expression in regex m/(a|b)(?=a){3}\x{100}/',
499333dc 469
63fbd1cb 470 '/\_/' => "",
63fbd1cb 471 '/[\006]/' => "",
499333dc
KW
472 '/[:alpha:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:alpha:]{#}\x{100}/',
473 '/[:zog:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:zog:]{#}\x{100}/',
474 '/[.zog.]\x{100}/' => 'POSIX syntax [. .] belongs inside character classes {#} m/[.zog.]{#}\x{100}/',
63fbd1cb 475 '/[a-b]/' => "",
499333dc
KW
476 '/(?c)\x{100}/' => 'Useless (?c) - use /gc modifier {#} m/(?c{#})\x{100}/',
477 '/(?-c)\x{100}/' => 'Useless (?-c) - don\'t use /gc modifier {#} m/(?-c{#})\x{100}/',
478 '/(?g)\x{100}/' => 'Useless (?g) - use /g modifier {#} m/(?g{#})\x{100}/',
479 '/(?-g)\x{100}/' => 'Useless (?-g) - don\'t use /g modifier {#} m/(?-g{#})\x{100}/',
480 '/(?o)\x{100}/' => 'Useless (?o) - use /o modifier {#} m/(?o{#})\x{100}/',
481 '/(?-o)\x{100}/' => 'Useless (?-o) - don\'t use /o modifier {#} m/(?-o{#})\x{100}/',
482 '/(?g-o)\x{100}/' => [ 'Useless (?g) - use /g modifier {#} m/(?g{#}-o)\x{100}/',
483 'Useless (?-o) - don\'t use /o modifier {#} m/(?g-o{#})\x{100}/',
63fbd1cb 484 ],
499333dc
KW
485 '/(?g-c)\x{100}/' => [ 'Useless (?g) - use /g modifier {#} m/(?g{#}-c)\x{100}/',
486 'Useless (?-c) - don\'t use /gc modifier {#} m/(?g-c{#})\x{100}/',
63fbd1cb
KW
487 ],
488 # (?c) means (?g) error won't be thrown
499333dc
KW
489 '/(?o-cg)\x{100}/' => [ 'Useless (?o) - use /o modifier {#} m/(?o{#}-cg)\x{100}/',
490 'Useless (?-c) - don\'t use /gc modifier {#} m/(?o-c{#}g)\x{100}/',
63fbd1cb 491 ],
499333dc
KW
492 '/(?ogc)\x{100}/' => [ 'Useless (?o) - use /o modifier {#} m/(?o{#}gc)\x{100}/',
493 'Useless (?g) - use /g modifier {#} m/(?og{#}c)\x{100}/',
494 'Useless (?c) - use /gc modifier {#} m/(?ogc{#})\x{100}/',
63fbd1cb 495 ],
499333dc
KW
496 '/a{1,1}?\x{100}/' => 'Useless use of greediness modifier \'?\' {#} m/a{1,1}?{#}\x{100}/',
497 '/b{3} +\x{100}/x' => 'Useless use of greediness modifier \'+\' {#} m/b{3} +{#}\x{100}/',
c877af1b 498 "/(?[ [ % - % ] ])/" => "",
acdfc3b6
KW
499 "/(?[ [ : - \\x$colon_hex ] ])\\x{100}/" => "\": - \\x$colon_hex \" is more clearly written simply as \":\" {#} m/(?[ [ : - \\x$colon_hex {#}] ])\\x{100}/",
500 "/(?[ [ \\x$colon_hex - : ] ])\\x{100}/" => "\"\\x$colon_hex\ - : \" is more clearly written simply as \":\" {#} m/(?[ [ \\x$colon_hex - : {#}] ])\\x{100}/",
501 "/(?[ [ \\t - \\x$tab_hex ] ])\\x{100}/" => "\"\\t - \\x$tab_hex \" is more clearly written simply as \"\\t\" {#} m/(?[ [ \\t - \\x$tab_hex {#}] ])\\x{100}/",
502 "/(?[ [ \\x$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
503 "/(?[ [ $B_hex - C ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ $B_hex - C {#}] ])/",
504 "/(?[ [ A - $B_hex ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ A - $B_hex {#}] ])/",
505 "/(?[ [ $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 {#}] ])/",
506 "/(?[ [ $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 {#}] ])/",
499333dc 507); # See comments before this for why '\x{100}' is generally needed
6fe2934b 508
b0e1d434 509# These need the character 'ネ' as a marker for mark_as_utf8()
c1d900c3
BF
510my @warnings_utf8 = mark_as_utf8(
511 'm/ネ\b*ネ/' => '\b* matches null string many times {#} m/ネ\b*{#}ネ/',
512 '/(?=ネ)*/' => '(?=ネ)* matches null string many times {#} m/(?=ネ)*{#}/',
513 'm/ネ[:foo:]ネ/' => 'POSIX syntax [: :] belongs inside character classes {#} m/ネ[:foo:]{#}ネ/',
b0e1d434 514 '/ネ(?c)ネ/' => 'Useless (?c) - use /gc modifier {#} m/ネ(?c{#})ネ/',
c1d900c3
BF
515 '/utf8 ネ (?ogc) ネ/' => [
516 'Useless (?o) - use /o modifier {#} m/utf8 ネ (?o{#}gc) ネ/',
517 'Useless (?g) - use /g modifier {#} m/utf8 ネ (?og{#}c) ネ/',
518 'Useless (?c) - use /gc modifier {#} m/utf8 ネ (?ogc{#}) ネ/',
519 ],
520
521);
522
523push @warning, @warnings_utf8;
524
63374c78 525my @warning_only_under_strict = (
b927b7e9
KW
526 '/[\N{U+00}-\x01]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+00}-\x01{#}]\x{100}/',
527 '/[\x00-\N{SOH}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\x00-\N{U+01}{#}]\x{100}/',
528 '/[\N{DEL}-\o{377}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+7F}-\o{377}{#}]\x{100}/',
529 '/[\o{0}-\N{U+01}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\o{0}-\N{U+01}{#}]\x{100}/',
530 '/[\000-\N{U+01}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\000-\N{U+01}{#}]\x{100}/',
531 '/[\N{DEL}-\377]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+7F}-\377{#}]\x{100}/',
c877af1b
KW
532 '/[\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}/',
533 '/[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
534 '/[\N{U+00}-\a]\x{100}/' => "",
535 '/[\a-\N{U+FF}]\x{100}/' => "",
536 '/[\N{U+FF}-\x{100}]/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+FF}-\x{100}{#}]/',
537 '/[\N{U+100}-\x{101}]/' => "",
c877af1b 538 "/[%-%]/" => "",
acdfc3b6
KW
539 "/[:-\\x$colon_hex]\\x{100}/" => "\":-\\x$colon_hex\" is more clearly written simply as \":\" {#} m/[:-\\x$colon_hex\{#}]\\x{100}/",
540 "/[\\x$colon_hex-:]\\x{100}/" => "\"\\x$colon_hex-:\" is more clearly written simply as \":\" {#} m/[\\x$colon_hex\-:{#}]\\x{100}/",
541 "/[\\t-\\x$tab_hex]\\x{100}/" => "\"\\t-\\x$tab_hex\" is more clearly written simply as \"\\t\" {#} m/[\\t-\\x$tab_hex\{#}]\\x{100}/",
542 "/[\\x$tab_hex-\\t]\\x{100}/" => "\"\\x$tab_hex-\\t\" is more clearly written simply as \"\\t\" {#} m/[\\x$tab_hex\-\\t{#}]\\x{100}/",
c877af1b
KW
543 "/[$B_hex-C]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[$B_hex-C{#}]/",
544 "/[A-$B_hex]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[A-$B_hex\{#}]/",
545 "/[$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-$high_mixed_alpha\{#}]/",
546 "/[$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\{#}]/",
547);
548
549my @warning_utf8_only_under_strict = mark_as_utf8(
550 '/ネ[᪉-᪐]/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ[᪉-᪐{#}]/",
551 '/ネ(?[ [ ᪉ - ᪐ ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ ᪉ - ᪐ {#}] ])/",
552 '/ネ[᧙-᧚]/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ[᧙-᧚{#}]/",
553 '/ネ(?[ [ ᧙ - ᧚ ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ ᧙ - ᧚ {#}] ])/",
63374c78
KW
554);
555
c877af1b
KW
556push @warning_only_under_strict, @warning_utf8_only_under_strict;
557
6fe2934b 558my @experimental_regex_sets = (
6d24e9d4 559 '/(?[ \t ])/' => 'The regex_sets feature is experimental {#} m/(?[{#} \t ])/',
c1d900c3
BF
560 'use utf8; /utf8 ネ (?[ [\tネ] ])/' => do { use utf8; 'The regex_sets feature is experimental {#} m/utf8 ネ (?[{#} [\tネ] ])/' },
561 '/noutf8 ネ (?[ [\tネ] ])/' => 'The regex_sets feature is experimental {#} m/noutf8 ネ (?[{#} [\tネ] ])/',
b45f050a
JF
562);
563
63fbd1cb 564my @deprecated = (
412f55bb
KW
565 '/\w{/' => 'Unescaped left brace in regex is deprecated, passed through {#} m/\w{{#}/',
566 '/\q{/' => [
567 'Unrecognized escape \q{ passed through {#} m/\q{{#}/',
568 'Unescaped left brace in regex is deprecated, passed through {#} m/\q{{#}/'
569 ],
570 '/:{4,a}/' => 'Unescaped left brace in regex is deprecated, passed through {#} m/:{{#}4,a}/',
cc4d09e1
KW
571 '/abc/xix' => 'Having more than one /x regexp modifier is deprecated',
572 '/(?xmsixp:abc)/' => 'Having more than one /x regexp modifier is deprecated',
573 '/(?xmsixp)abc/' => 'Having more than one /x regexp modifier is deprecated',
574 '/(?xxxx:abc)/' => 'Having more than one /x regexp modifier is deprecated',
63fbd1cb
KW
575);
576
67cdf558
KW
577for my $strict ("", "use re 'strict';") {
578
579 # First time just use @death; but under strict we add the things that fail
580 # there. Doing it this way makes sure that 'strict' doesnt change the
581 # things that are already fatal when not under strict.
582 if ($strict) {
583 for (my $i = 0; $i < @death_only_under_strict; $i += 3) {
584 push @death, $death_only_under_strict[$i], # The regex
585 $death_only_under_strict[$i+2]; # The fatal msg
586 }
587 for (my $i = 0; $i < @death_utf8_only_under_strict; $i += 3) {
588
589 # Same with the utf8 versions
590 push @death, mark_as_utf8($death_utf8_only_under_strict[$i],
591 $death_utf8_only_under_strict[$i+2]);
592 }
593 }
d6609144
KW
594 for (my $i = 0; $i < @death; $i += 2) {
595 my $regex = $death[$i];
596 my $expect = fixup_expect($death[$i+1]);
597 no warnings 'experimental::regex_sets';
598 no warnings 'experimental::re_strict';
d6609144
KW
599
600 warning_is(sub {
601 my $eval_string = "$strict $regex";
602 $_ = "x";
603 eval $eval_string;
604 like($@, qr/\Q$expect/, $eval_string);
605 }, undef, "... and died without any other warnings");
606 }
67cdf558 607}
b45f050a 608
63374c78
KW
609for my $strict ("", "no warnings 'experimental::re_strict'; use re 'strict';") {
610 my @warning_tests = @warning;
67cdf558 611
63374c78
KW
612 # Build the tests for @warning. Use the strict/non-strict versions
613 # appropriately.
614 if ($strict) {
615 push @warning_tests, @warning_only_under_strict;
616 }
617 else {
618 for (my $i = 0; $i < @warning_only_under_strict; $i += 2) {
619 if ($warning_only_under_strict[$i] =~ /\Q(?[/) {
620 push @warning_tests, $warning_only_under_strict[$i], # The regex
621 $warning_only_under_strict[$i+1];
622 }
623 else {
624 push @warning_tests, $warning_only_under_strict[$i], # The regex
625 ""; # No warning because not strict
626 }
627 }
67cdf558 628 for (my $i = 0; $i < @death_only_under_strict; $i += 3) {
63374c78
KW
629 push @warning_tests, $death_only_under_strict[$i], # The regex
630 $death_only_under_strict[$i+1]; # The warning
67cdf558
KW
631 }
632 for (my $i = 0; $i < @death_utf8_only_under_strict; $i += 3) {
63374c78 633 push @warning_tests, mark_as_utf8($death_utf8_only_under_strict[$i],
d6609144 634 $death_utf8_only_under_strict[$i+1]);
67cdf558
KW
635 }
636 }
d6609144 637
63374c78 638 foreach my $ref (\@warning_tests, \@experimental_regex_sets, \@deprecated) {
d6609144 639 my $warning_type;
63374c78 640 my $turn_off_warnings = "";
d6609144 641 my $default_on;
63374c78 642 if ($ref == \@warning_tests) {
d6609144 643 $warning_type = 'regexp, digit';
63374c78 644 $turn_off_warnings = "no warnings 'experimental::regex_sets';";
d6609144 645 $default_on = $strict;
af01601c 646 }
d6609144
KW
647 elsif ($ref == \@deprecated) {
648 $warning_type = 'regexp, deprecated';
649 $default_on = 1;
650 }
651 else {
652 $warning_type = 'experimental::regex_sets';
653 $default_on = 1;
654 }
655 for (my $i = 0; $i < @$ref; $i += 2) {
656 my $regex = $ref->[$i];
657 my @expect = fixup_expect($ref->[$i+1]);
63374c78
KW
658
659 # A length-1 array with an empty warning means no warning gets
660 # generated at all.
661 undef @expect if @expect == 1 && $expect[0] eq "";
662
d6609144
KW
663 {
664 $_ = "x";
665 eval "$strict no warnings; $regex";
af01601c 666 }
d6609144
KW
667 if (is($@, "", "$strict $regex did not die")) {
668 my @got = capture_warnings(sub {
669 $_ = "x";
63374c78 670 eval "$strict $turn_off_warnings $regex" });
d6609144
KW
671 my $count = @expect;
672 if (! is(scalar @got, scalar @expect,
673 "... and gave expected number ($count) of warnings"))
674 {
675 if (@got < @expect) {
676 $count = @got;
677 note "Expected warnings not gotten:\n\t" . join "\n\t",
678 @expect[$count .. $#expect];
679 }
680 else {
681 note "Unexpected warnings gotten:\n\t" . join("\n\t",
682 @got[$count .. $#got]);
683 }
b0b90d25 684 }
d6609144
KW
685 foreach my $i (0 .. $count - 1) {
686 if (! like($got[$i], qr/\Q$expect[$i]/,
687 "... and gave expected warning"))
688 {
689 chomp($got[$i]);
690 chomp($expect[$i]);
691 diag("GOT\n'$got[$i]'\nEXPECT\n'$expect[$i]'");
67cdf558
KW
692 }
693 else {
d6609144
KW
694 ok (0 == capture_warnings(sub {
695 $_ = "x";
696 eval "$strict no warnings '$warning_type'; $regex;" }
697 ),
698 "... and turning off '$warning_type' warnings suppressed it");
699
700 # Test that whether the warning is on by default is
701 # correct. This test relies on the fact that we
702 # are outside the scope of any ‘use warnings’.
703 local $^W;
704 my @warns = capture_warnings(sub { $_ = "x";
705 eval "$strict $regex" });
63374c78
KW
706 # Warning should be on as well if is testing
707 # '(?[...])' which turns on strict
708 if ($default_on || grep { $_ =~ /\Q(?[/ } @expect ) {
d6609144
KW
709 ok @warns > 0, "... and the warning is on by default";
710 }
711 else {
712 ok @warns == 0, "... and the warning is off by default";
713 }
67cdf558 714 }
af01601c
KW
715 }
716 }
b42857f3 717 }
6fe2934b
KW
718 }
719}
b45f050a 720
cb79f740 721done_testing();