1#!./perl -w 2 3$|=1; # outherwise things get mixed up in output 4 5BEGIN { 6 chdir 't' if -d 't'; 7 require './test.pl'; 8 set_up_inc( qw '../lib ../ext/re' ); 9 eval 'require Config'; # assume defaults if this fails 10} 11 12skip_all_without_unicode_tables(); 13 14use strict; 15use open qw(:utf8 :std); 16 17# Show that it works when all warnings are enabled upon invocation. This file 18# includes tests that the default warnings are enabled by default, and the 19# non-default ones aren't. 20use warnings; 21BEGIN { ${^WARNING_BITS} = undef } # Kludge to restore default warnings 22 23# NOTE: 24# This file uses display_rx() from t/test.pl, which escapes 25# non-printable codepoints and percentage symbols as non-standard 26# "%x{01+aa+bc+02}" style sequences. We do this as we may be testing 27# malformed utf8 sequences or other non-printable characters which we 28# want to distinguish from any actual regex pattern escape sequences, 29# while at the same time we don't want to emit any non-printable 30# sequences to a terminal as literals where they may cause problems. 31# Thus "\x{01}" means the string contained six codepoints making up a 32# normal "\x{...}" style escape sequence whereas "%x{01}" means the string 33# contained a single codepoint and "%x{01+02}" means it contained two 34# codepoints, etc. - Yves 35 36# Kind of a kludge to mark warnings to be expected only if we are testing 37# under "use re 'strict'" 38my $only_strict_marker = ':expected_only_under_strict'; 39 40## If the markers used are changed (search for "MARKER1" in regcomp.c), 41## update only these two regexs, and leave the {#} in the @death/@warning 42## arrays below. The {#} is a meta-marker -- it marks where the marker should 43## go. 44 45sub fixup_expect ($$) { 46 47 # Fixes up the expected results by inserting the boiler plate text. 48 # Returns empty string if that is what is expected. Otherwise, handles 49 # either a scalar, turning it into a single element array; or a ref to an 50 # array, adjusting each element. If called in array context, returns an 51 # array, otherwise the join of all elements. 52 53 # The string $only_strict_marker will be removed from any expect line it 54 # begins, and if $strict is not true, that expect line will be removed 55 # from the output (hence won't be expected) 56 57 my ($expect_ref, $strict) = @_; 58 return "" if $expect_ref eq ""; 59 60 my @expect; 61 if (ref $expect_ref) { 62 @expect = @$expect_ref; 63 } 64 else { 65 @expect = $expect_ref; 66 } 67 68 my @new_expect; 69 foreach my $element (@expect) { 70 $element =~ s/\{\#\}/in regex; marked by <-- HERE in/; 71 $element =~ s/\{\#\}/ <-- HERE /; 72 $element .= " at "; 73 next if $element =~ s/ ^ $only_strict_marker \s* //x && ! $strict; 74 push @new_expect, $element; 75 } 76 return wantarray ? @new_expect : join "", @new_expect; 77} 78 79sub add_markers { 80 my ($element)= @_; 81 $element =~ s/ at .* line \d+\.?\n$//; 82 $element =~ s/in regex; marked by <-- HERE in/{#}/; 83 $element =~ s/ <-- HERE /{#}/; 84 return $element; 85} 86 87## Because we don't "use utf8" in this file, we need to do some extra legwork 88## for the utf8 tests: Prepend 'use utf8' to the pattern, and mark the strings 89## to check against as UTF-8, but for this all to work properly, the character 90## 'ネ' (U+30CD) is required in each pattern somewhere as a marker. 91## 92## This also creates a second variant of the tests to check if the 93## latin1 error messages are working correctly. Because we don't 'use utf8', 94## we can't tell if something is UTF-8 or Latin1, so you need the suffix 95## '; no latin1' to not have the second variant. 96my $l1 = "\x{ef}"; 97my $utf8 = "\x{30cd}"; 98utf8::encode($utf8); 99 100sub mark_as_utf8 { 101 my @ret; 102 for (my $i = 0; $i < @_; $i += 2) { 103 my $pat = $_[$i]; 104 my $msg = $_[$i+1]; 105 my $l1_pat = $pat =~ s/$utf8/$l1/gr; 106 my $l1_msg; 107 $pat = "use utf8; $pat"; 108 109 if (ref $msg) { 110 $l1_msg = [ map { s/$utf8/$l1/gr } @$msg ]; 111 @$msg = map { my $c = $_; utf8::decode($c); $c } @$msg; 112 } 113 else { 114 $l1_msg = $msg =~ s/$utf8/$l1/gr; 115 utf8::decode($msg); 116 } 117 push @ret, $pat => $msg; 118 119 push @ret, $l1_pat => $l1_msg unless $l1_pat =~ /#no latin1/; 120 } 121 return @ret; 122} 123 124my $inf_m1 = ($Config::Config{reg_infty} || ((1<<31)-1)) - 1; 125my $inf_p1 = $inf_m1 + 2; 126 127my $B_hex = sprintf("\\x%02X", ord "B"); 128my $low_mixed_alpha = ('A' lt 'a') ? 'A' : 'a'; 129my $high_mixed_alpha = ('A' lt 'a') ? 'a' : 'A'; 130my $low_mixed_digit = ('A' lt '0') ? 'A' : '0'; 131my $high_mixed_digit = ('A' lt '0') ? '0' : 'A'; 132 133my $colon_hex = sprintf "%02X", ord(":"); 134my $tab_hex = sprintf "%02X", ord("\t"); 135 136# Key-value pairs of strings eval'd as patterns => warn/error messages that 137# they should generate. In some cases, the value is an array of multiple 138# messages. Some groups have the message(s) be default on; others, default 139# off. This can be overridden on an individual key basis by preceding the 140# pattern string with either 'default_on' or 'default_off' 141# 142# The first set are those that should be fatal errors. 143 144my $bug133423 = "(?[(?^:(?[\\\x00]))\\]\x00|2[^^]\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80])R.\\670"; 145 146my @death = 147( 148 '/[[=foo=]]/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/[[=foo=]{#}]/', 149 150 '/(?<= .*)/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= .*)/', 151 152 '/(?<= a+)/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= a+)/', 153 '/(?<= a{255})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= a{255})/', 154 '/(?<= a{0,255})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= a{0,255})/', 155 '/(?<= a{200}b{55})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= a{200}b{55})/', 156 157 '/(?<= x{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= x{1000})/', 158 '/(?<= (?&x))(?<x>x+)/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= (?&x))(?<x>x+)/', 159 160 '/(?@)/' => 'Sequence (?@...) not implemented {#} m/(?@{#})/', 161 162 '/(?{ 1/' => 'Missing right curly or square bracket', 163 164 '/(?(1x))/' => 'Switch condition not recognized {#} m/(?(1x{#}))/', 165 '/(?(1x(?#)))/'=> 'Switch condition not recognized {#} m/(?(1x{#}(?#)))/', 166 167 '/(?(1)/' => 'Switch (?(condition)... not terminated {#} m/(?(1){#}/', 168 '/(?(1)x/' => 'Switch (?(condition)... not terminated {#} m/(?(1)x{#}/', 169 '/(?(1)x|y/' => 'Switch (?(condition)... not terminated {#} m/(?(1)x|y{#}/', 170 '/(?(1)x|y|z)/' => 'Switch (?(condition)... contains too many branches {#} m/(?(1)x|y|{#}z)/', 171 172 '/(?(x)y|x)/' => 'Unknown switch condition (?(...)) {#} m/(?(x{#})y|x)/', 173 '/(?(??{}))/' => 'Unknown switch condition (?(...)) {#} m/(?(?{#}?{}))/', 174 '/(?(?[]))/' => 'Unknown switch condition (?(...)) {#} m/(?(?{#}[]))/', 175 176 '/(?/' => 'Sequence (? incomplete {#} m/(?{#}/', 177 178 '/(?;x/' => 'Sequence (?;...) not recognized {#} m/(?;{#}x/', 179 '/(?<;x/' => 'Group name must start with a non-digit word character {#} m/(?<;{#}x/', 180 '/(?\ix/' => 'Sequence (?\...) not recognized {#} m/(?\{#}ix/', 181 '/(?\mx/' => 'Sequence (?\...) not recognized {#} m/(?\{#}mx/', 182 '/(?\:x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}:x/', 183 '/(?\=x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}=x/', 184 '/(?\!x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}!x/', 185 '/(?\<=x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}<=x/', 186 '/(?\<!x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}<!x/', 187 '/(?\>x/' => 'Sequence (?\...) not recognized {#} m/(?\{#}>x/', 188 '/(?^-i:foo)/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i:foo)/', 189 '/(?^-i)foo/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i)foo/', 190 '/(?^d:foo)/' => 'Sequence (?^d...) not recognized {#} m/(?^d{#}:foo)/', 191 '/(?^d)foo/' => 'Sequence (?^d...) not recognized {#} m/(?^d{#})foo/', 192 '/(?^lu:foo)/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/(?^lu{#}:foo)/', 193 '/(?^lu)foo/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/(?^lu{#})foo/', 194'/(?da:foo)/' => 'Regexp modifiers "d" and "a" are mutually exclusive {#} m/(?da{#}:foo)/', 195'/(?lil:foo)/' => 'Regexp modifier "l" may not appear twice {#} m/(?lil{#}:foo)/', 196'/(?aaia:foo)/' => 'Regexp modifier "a" may appear a maximum of twice {#} m/(?aaia{#}:foo)/', 197'/(?i-l:foo)/' => 'Regexp modifier "l" may not appear after the "-" {#} m/(?i-l{#}:foo)/', 198 199 '/((x)/' => 'Unmatched ( {#} m/({#}(x)/', 200 '/{(}/' => 'Unmatched ( {#} m/{({#}}/', # [perl #127599] 201 202 "/x{$inf_p1}/" => "Quantifier in {,} bigger than $inf_m1 {#} m/x{$inf_p1\{#}}/", 203 "/x{$inf_p1,}/" => "Quantifier in {,} bigger than $inf_m1 {#} m/x{$inf_p1\{#},}/", 204 "/x{01,2}/" => "Invalid quantifier in {,} {#} m/x{01{#},2}/", 205 "/x{1,02}/" => "Invalid quantifier in {,} {#} m/x{1,02{#}}/", 206 207 208 '/x**/' => 'Nested quantifiers {#} m/x**{#}/', 209 210 '/x[/' => 'Unmatched [ {#} m/x[{#}/', 211 212 '/*/', => 'Quantifier follows nothing {#} m/*{#}/', 213 214 '/\p{x/' => 'Missing right brace on \p{} {#} m/\p{{#}x/', 215 216 '/[\p{x]/' => 'Missing right brace on \p{} {#} m/[\p{{#}x]/', 217 218 '/(x)\2/' => 'Reference to nonexistent group {#} m/(x)\2{#}/', 219 220 '/\g/' => 'Unterminated \g... pattern {#} m/\g{#}/', 221 '/\g{1/' => 'Unterminated \g{...} pattern {#} m/\g{1{#}/', 222 '/\g{-abc}/' => 'Group name must start with a non-digit word character {#} m/\g{-{#}abc}/', 223 '/(?<;x/' => 'Group name must start with a non-digit word character {#} m/(?<;{#}x/', 224 225 'my $m = "\\\"; $m =~ $m', => 'Trailing \ in regex m/\/', 226 227 '/\x{ 1 /' => 'Missing right brace on \x{} {#} m/\x{ 1{#} /', 228 '/\x{X/' => 'Missing right brace on \x{} {#} m/\x{{#}X/', 229 230 '/[\x{X]/' => 'Missing right brace on \x{} {#} m/[\x{{#}X]/', 231 '/[\x{ A ]/' => 'Missing right brace on \x{} {#} m/[\x{ A{#} ]/', 232 233 '/\o{ 1 /' => 'Missing right brace on \o{} {#} m/\o{ 1{#} /', 234 '/\o{X/' => 'Missing right brace on \o{} {#} m/\o{{#}X/', 235 236 '/[\o{X]/' => 'Missing right brace on \o{} {#} m/[\o{{#}X]/', 237 '/[\o{ 7 ]/' => 'Missing right brace on \o{} {#} m/[\o{ 7{#} ]/', 238 239 '/[[:barf:]]/' => 'POSIX class [:barf:] unknown {#} m/[[:barf:]{#}]/', 240 241 '/[[=barf=]]/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/[[=barf=]{#}]/', 242 243 '/[[.barf.]]/' => 'POSIX syntax [. .] is reserved for future extensions {#} m/[[.barf.]{#}]/', 244 245 '/[z-a]/' => 'Invalid [] range "z-a" {#} m/[z-a{#}]/', 246 247 '/\p/' => 'Empty \p {#} m/\p{#}/', 248 '/\P/' => 'Empty \P {#} m/\P{#}/', 249 '/\p{}/' => 'Empty \p{} {#} m/\p{{#}}/', 250 '/\P{}/' => 'Empty \P{} {#} m/\P{{#}}/', 251 252'/a\b{cde/' => 'Missing right brace on \b{} {#} m/a\b{{#}cde/', 253'/a\B{cde/' => 'Missing right brace on \B{} {#} m/a\B{{#}cde/', 254 255 '/\b{}/' => 'Empty \b{} {#} m/\b{}{#}/', 256 '/\B{}/' => 'Empty \B{} {#} m/\B{}{#}/', 257 258 '/\b{gc}/' => "'gc' is an unknown bound type {#} m/\\b{gc{#}}/", 259 '/\B{gc}/' => "'gc' is an unknown bound type {#} m/\\B{gc{#}}/", 260 261 '/(?[[[::]]])/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[[[::]]{#}])/", 262 '/(?[[[:w:]]])/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[[[:w:]]{#}])/", 263 '/(?[a])/' => 'Unexpected character {#} m/(?[a{#}])/', 264 '/(?[ + \t ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ +{#} \t ])/', 265 '/(?[ \cK - ( + \t ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ \cK - ( +{#} \t ) ])/', 266 '/(?[ \cK ( \t ) ])/' => 'Unexpected \'(\' with no preceding operator {#} m/(?[ \cK ({#} \t ) ])/', 267 '/(?[ \cK \t ])/' => 'Operand with no preceding operator {#} m/(?[ \cK \t{#} ])/', 268 '/(?[ \0004 ])/' => 'Need exactly 3 octal digits {#} m/(?[ \0004 {#}])/', 269 '/(?[ \05 ])/' => 'Need exactly 3 octal digits {#} m/(?[ \05 {#}])/', 270 '/(?[ \o{1038} ])/' => 'Non-octal character {#} m/(?[ \o{1038{#}} ])/', 271 '/(?[ \o{} ])/' => 'Empty \o{} {#} m/(?[ \o{}{#} ])/', 272 '/(?[ \x{defg} ])/' => 'Non-hex character {#} m/(?[ \x{defg{#}} ])/', 273 '/(?[ \xabcdef ])/' => 'Use \\x{...} for more than two hex characters {#} m/(?[ \xabc{#}def ])/', 274 '/(?[ \x{} ])/' => 'Empty \x{} {#} m/(?[ \x{}{#} ])/', 275 '/(?[ \cK + ) ])/' => 'Unexpected \')\' {#} m/(?[ \cK + ){#} ])/', 276 '/(?[ \cK + ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ \cK + {#}])/', 277 '/(?[ ( ) ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ ( ){#} ])/', 278 '/(?[[0]+()+])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[[0]+(){#}+])/', 279 '/(?[ \p{foo} ])/' => 'Can\'t find Unicode property definition "foo" {#} m/(?[ \p{foo}{#} ])/', 280 '/(?[ \p{ foo = bar } ])/' => 'Can\'t find Unicode property definition "foo = bar" {#} m/(?[ \p{ foo = bar }{#} ])/', 281 '/(?[ \8 ])/' => 'Unrecognized escape \8 in character class {#} m/(?[ \8{#} ])/', 282 '/(?[ \t ]/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[ \\t ]{#}/", 283 '/(?[ [ \t ]/' => "Syntax error in (?[...]) {#} m/(?[ [ \\t ]{#}/", 284 '/(?[ \t ] ]/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[ \\t ]{#} ]/", 285 '/(?[ [ ] ]/' => "Syntax error in (?[...]) {#} m/(?[ [ ] ]{#}/", 286 '/(?[ \t + \e # This was supposed to be a comment ])/' => 287 "Syntax error in (?[...]) {#} m/(?[ \\t + \\e # This was supposed to be a comment ]){#}/", 288 '/(?[ ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ {#}])/', 289 'm/(?[[a-\d]])/' => 'False [] range "a-\d" {#} m/(?[[a-\d{#}]])/', 290 'm/(?[[\w-x]])/' => 'False [] range "\w-" {#} m/(?[[\w-{#}x]])/', 291 'm/(?[[a-\pM]])/' => 'False [] range "a-\pM" {#} m/(?[[a-\pM{#}]])/', 292 'm/(?[[\pM-x]])/' => 'False [] range "\pM-" {#} m/(?[[\pM-{#}x]])/', 293 'm/(?[[^\N{LATIN CAPITAL LETTER A WITH MACRON AND GRAVE}]])/' => '\N{} here is restricted to one character {#} m/(?[[^\N{U+100.300{#}}]])/', 294 'm/(?[ \p{Digit} & (?^(?[ \p{Thai} | \p{Lao} ]))])/' => 'Sequence (?^(...) not recognized {#} m/(?[ \p{Digit} & (?^({#}?[ \p{Thai} | \p{Lao} ]))])/', 295 'm/(?[ \p{Digit} & (?(?[ \p{Thai} | \p{Lao} ]))])/' => 'Unexpected character {#} m/(?[ \p{Digit} & (?{#}(?[ \p{Thai} | \p{Lao} ]))])/', 296 'm/\p{Is_Is_Any}/' => 'Unknown user-defined property name \p{main::Is_Is_Any}', 297 'm/\o{/' => 'Missing right brace on \o{} {#} m/\o{{#}/', 298 'm/\o/' => 'Missing braces on \o{} {#} m/\o{#}/', 299 'm/\o{}/' => 'Empty \o{} {#} m/\o{}{#}/', 300 'm/[\o{]/' => 'Missing right brace on \o{} {#} m/[\o{{#}]/', 301 'm/[\o]/' => 'Missing braces on \o{} {#} m/[\o{#}]/', 302 'm/[\o{}]/' => 'Empty \o{} {#} m/[\o{}{#}]/', 303 'm/(?^-i:foo)/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i:foo)/', 304 'm/\87/' => 'Reference to nonexistent group {#} m/\87{#}/', 305 'm/a\87/' => 'Reference to nonexistent group {#} m/a\87{#}/', 306 'm/a\97/' => 'Reference to nonexistent group {#} m/a\97{#}/', 307 'm/(*DOOF)/' => 'Unknown verb pattern \'DOOF\' {#} m/(*DOOF){#}/', 308 'm/(?&a/' => 'Sequence (?&... not terminated {#} m/(?&a{#}/', 309 'm/(?P=/' => 'Sequence ?P=... not terminated {#} m/(?P={#}/', 310 "m/(?'/" => "Sequence (?'... not terminated {#} m/(?'{#}/", 311 "m/(?</" => "Sequence (?<... not terminated {#} m/(?<{#}/", 312 'm/(?&/' => 'Sequence (?&... not terminated {#} m/(?&{#}/', 313 'm/(?(</' => 'Sequence (?(<... not terminated {#} m/(?(<{#}/', 314 "m/(?('/" => "Sequence (?('... not terminated {#} m/(?('{#}/", 315 'm/\g{/' => 'Sequence \g{... not terminated {#} m/\g{{#}/', 316 'm/\k</' => 'Sequence \k<... not terminated {#} m/\k<{#}/', 317 '/((?# 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)/', 318 '/((?# 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)/', 319 '/((?# 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)/', 320 321 '/(*script_runfoo)/' => 'Unknown \'(*...)\' construct \'script_runfoo\' {#} m/(*script_runfoo){#}/', 322 '/(*srfoo)/' => 'Unknown \'(*...)\' construct \'srfoo\' {#} m/(*srfoo){#}/', 323 '/(*script_run)/' => '\'(*script_run\' requires a terminating \':\' {#} m/(*script_run{#})/', 324 '/(*sr)/' => '\'(*sr\' requires a terminating \':\' {#} m/(*sr{#})/', 325 '/(*pla)/' => '\'(*pla\' requires a terminating \':\' {#} m/(*pla{#})/', 326 '/(*script_run/' => 'Unterminated \'(*...\' construct {#} m/(*script_run{#}/', 327 '/(*sr/' => 'Unterminated \'(*...\' construct {#} m/(*sr{#}/', 328 '/(*script_run:foo/' => 'Unterminated \'(*...\' argument {#} m/(*script_run:foo{#}/', 329 '/(*sr:foo/' => 'Unterminated \'(*...\' argument {#} m/(*sr:foo{#}/', 330 '/(?[\ &!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ &!{#}])/', # [perl #126180] 331 '/(?[\ +!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ +!{#}])/', # [perl #126180] 332 '/(?[\ -!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ -!{#}])/', # [perl #126180] 333 '/(?[\ ^!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ ^!{#}])/', # [perl #126180] 334 '/(?[\ |!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ |!{#}])/', # [perl #126180] 335 '/(?[()-!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[(){#}-!])/', # [perl #126204] 336 '/(?[!()])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[!(){#}])/', # [perl #126404] 337 '/\w{/' => 'Unescaped left brace in regex is illegal here {#} m/\w{{#}/', 338 '/\q{/' => 'Unescaped left brace in regex is illegal here {#} m/\q{{#}/', 339 '/\A{/' => 'Unescaped left brace in regex is illegal here {#} m/\A{{#}/', 340 '/(?<=/' => 'Sequence (?<=... not terminated {#} m/(?<={#}/', # [perl #128170] 341 '/(?<!/' => 'Sequence (?<!... not terminated {#} m/(?<!{#}/', 342 '/(?!/' => 'Sequence (?!... not terminated {#} m/(?!{#}/', 343 '/(?=/' => 'Sequence (?=... not terminated {#} m/(?={#}/', 344 '/\p{vertical tab}/' => 'Can\'t find Unicode property definition "vertical tab" {#} m/\\p{vertical tab}{#}/', # [perl #132055] 345 "/$bug133423/" => "Unexpected ']' with no following ')' in (?[... {#} m/(?[(?^:(?[\\�]))\\]{#}�|2[^^]\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80])R.\\670/", 346 '/[^/' => 'Unmatched [ {#} m/[{#}^/', # [perl #133767] 347 '/\p{Is_Other_Alphabetic=F}/ ' => 'Can\'t find Unicode property definition "Is_Other_Alphabetic=F" {#} m/\p{Is_Other_Alphabetic=F}{#}/', 348 '/\p{Is_Other_Alphabetic=F}/ ' => 'Can\'t find Unicode property definition "Is_Other_Alphabetic=F" {#} m/\p{Is_Other_Alphabetic=F}{#}/', 349 '/\x{100}(?(/' => 'Unknown switch condition (?(...)) {#} m/\\x{100}(?({#}/', # [perl #133896] 350 '/(?[\N{KEYCAP DIGIT NINE}/' => '\N{} here is restricted to one character {#} m/(?[\\N{U+39.FE0F.20E3{#}}/', # [perl #133988] 351 '/0000000000000000[\N{U+0.00}0000/' => 'Unmatched [ {#} m/0000000000000000[{#}\N{U+0.00}0000/', # [perl #134059] 352 '/\p{nv=\b5\b}/' => 'Can\'t find Unicode property definition "nv=\\b5\\b" {#} m/\\p{nv=\\b5\\b}{#}/', 353 '/\p{nv=:(?g)10:}/' => 'Use of modifier \'g\' is not allowed in Unicode property wildcard subpatterns {#} m/(?g{#})10/', 354 '/\p{gc=:L*:}/' => 'Use of quantifier \'*\' is not allowed in Unicode property wildcard subpatterns {#} m/L*{#}/', 355 '/\p{gc=:L\G:}/' => 'Use of \'\G\' is not allowed in Unicode property wildcard subpatterns {#} m/L\G{#}/', 356 '/\p{gc=:(?a)L:}/' => 'Use of modifier \'a\' is not allowed in Unicode property wildcard subpatterns {#} m/(?a){#}L/', 357 '/\p{gc=:(?u)L:}/' => 'Use of modifier \'u\' is not allowed in Unicode property wildcard subpatterns {#} m/(?u){#}L/', 358 '/\p{gc=:(?d)L:}/' => 'Use of modifier \'d\' is not allowed in Unicode property wildcard subpatterns {#} m/(?d){#}L/', 359 '/\p{gc=:(?l)L:}/' => 'Use of modifier \'l\' is not allowed in Unicode property wildcard subpatterns {#} m/(?l){#}L/', 360 '/\p{gc=:(?-m)L:}/' => 'Use of modifier \'-m\' is not allowed in Unicode property wildcard subpatterns {#} m/(?-m{#})L/', 361 '/\p{gc=:\pS:}/' => 'Use of \'\\pS\' is not allowed in Unicode property wildcard subpatterns {#} m/\\pS{#}/', 362 '/\p{gc=:\PS:}/' => 'Use of \'\\PS\' is not allowed in Unicode property wildcard subpatterns {#} m/\\PS{#}/', 363 '/\p{gc=:[\pS]:}/' => 'Use of \'\\pS\' is not allowed in Unicode property wildcard subpatterns {#} m/[\\pS{#}]/', 364 '/\p{gc=:[\PS]:}/' => 'Use of \'\\PS\' is not allowed in Unicode property wildcard subpatterns {#} m/[\\PS{#}]/', 365 '/(?[\p{name=KATAKANA LETTER AINU P}])/' => 'Unicode string properties are not implemented in (?[...]) {#} m/(?[\p{name=KATAKANA LETTER AINU P}{#}])/', 366 '/(?[ (?^x:(?[ ])) ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ (?^x:(?[ {#}])) ])/', 367 '/(?[ (?x:(?[ ])) ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ (?x:(?[ {#}])) ])/', # GH #16779 368); 369 370# These are messages that are death under 'use re "strict"', and may or may 371# not warn otherwise. See comment before @warning as to why some have a 372# \x{100} in them. This array has 3 elements per construct. [0] is the regex 373# to use; [1] is the message under no strict (empty to not warn), and [2] is 374# under strict. 375my @death_only_under_strict = ( 376 'm/\xABC/' => "", 377 => 'Use \x{...} for more than two hex characters {#} m/\xABC{#}/', 378 'm/[\xABC]/' => "", 379 => 'Use \x{...} for more than two hex characters {#} m/[\xABC{#}]/', 380 381 # some messages below aren't all category 'regexp'. (Hence we have to 382 # turn off 'digit' messages as well below) 383 'm/\xAG/' => 'Non-hex character \'G\' terminates \x early. Resolved as "\x0AG" {#} m/\xA{#}G/', 384 => 'Non-hex character {#} m/\xAG{#}/', 385 'm/[\xAG]/' => 'Non-hex character \'G\' terminates \x early. Resolved as "\x0AG" {#} m/[\xA{#}G]/', 386 => 'Non-hex character {#} m/[\xAG{#}]/', 387 'm/\o{789}/' => 'Non-octal character \'8\' terminates \o early. Resolved as "\o{007}" {#} m/\o{789}{#}/', 388 => 'Non-octal character {#} m/\o{78{#}9}/', 389 'm/[\o{789}]/' => 'Non-octal character \'8\' terminates \o early. Resolved as "\o{007}" {#} m/[\o{789}{#}]/', 390 => 'Non-octal character {#} m/[\o{78{#}9}]/', 391 'm/\x{}/' => "", 392 => 'Empty \x{} {#} m/\x{}{#}/', 393 'm/[\x{}]/' => "", 394 => 'Empty \x{} {#} m/[\x{}{#}]/', 395 'm/\x{ABCDEFG}/' => 'Non-hex character \'G\' terminates \x early. Resolved as "\x{ABCDEF}" {#} m/\x{ABCDEFG}{#}/', 396 => 'Non-hex character {#} m/\x{ABCDEFG{#}}/', 397 'm/[\x{ABCDEFG}]/' => 'Non-hex character \'G\' terminates \x early. Resolved as "\x{ABCDEF}" {#} m/[\x{ABCDEFG}{#}]/', 398 => 'Non-hex character {#} m/[\x{ABCDEFG{#}}]/', 399 "m'[\\y]\\x{100}'" => 'Unrecognized escape \y in character class passed through {#} m/[\y{#}]\x{100}/', 400 => 'Unrecognized escape \y in character class {#} m/[\y{#}]\x{100}/', 401 'm/[a-\d]\x{100}/' => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/', 402 => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/', 403 'm/[\w-x]\x{100}/' => 'False [] range "\w-" {#} m/[\w-{#}x]\x{100}/', 404 => 'False [] range "\w-" {#} m/[\w-{#}x]\x{100}/', 405 'm/[a-\pM]\x{100}/' => 'False [] range "a-\pM" {#} m/[a-\pM{#}]\x{100}/', 406 => 'False [] range "a-\pM" {#} m/[a-\pM{#}]\x{100}/', 407 'm/[\pM-x]\x{100}/' => 'False [] range "\pM-" {#} m/[\pM-{#}x]\x{100}/', 408 => 'False [] range "\pM-" {#} m/[\pM-{#}x]\x{100}/', 409 '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}{#}]/', 410 => '\N{} here is restricted to one character {#} m/[^\N{U+100.300{#}}]/', 411 '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}{#}]/', 412 => '\N{} here is restricted to one character {#} m/[\x03-\N{U+100.300{#}}]/', 413 '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}]/', 414 => '\N{} here is restricted to one character {#} m/[\N{U+100.300{#}}-\x{10FFFF}]/', 415 '/[\08]/' => 'Non-octal character \'8\' terminates \0 early. Resolved as "\0008" {#} m/[\08{#}]/', 416 => 'Need exactly 3 octal digits {#} m/[\08{#}]/', 417 '/[\018]/' => 'Non-octal character \'8\' terminates \0 early. Resolved as "\0018" {#} m/[\018{#}]/', 418 => 'Need exactly 3 octal digits {#} m/[\018{#}]/', 419 '/[\_\0]/' => "", 420 => 'Need exactly 3 octal digits {#} m/[\_\0]{#}/', 421 '/[\07]/' => "", 422 => 'Need exactly 3 octal digits {#} m/[\07]{#}/', 423 '/[\0005]/' => "", 424 => 'Need exactly 3 octal digits {#} m/[\0005]{#}/', 425 '/[\8\9]\x{100}/' => ['Unrecognized escape \8 in character class passed through {#} m/[\8{#}\9]\x{100}/', 426 'Unrecognized escape \9 in character class passed through {#} m/[\8\9{#}]\x{100}/', 427 ], 428 => 'Unrecognized escape \8 in character class {#} m/[\8{#}\9]\x{100}/', 429 '/[a-\d]\x{100}/' => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/', 430 => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/', 431 '/[\d-b]\x{100}/' => 'False [] range "\d-" {#} m/[\d-{#}b]\x{100}/', 432 => 'False [] range "\d-" {#} m/[\d-{#}b]\x{100}/', 433 '/[\s-\d]\x{100}/' => 'False [] range "\s-" {#} m/[\s-{#}\d]\x{100}/', 434 => 'False [] range "\s-" {#} m/[\s-{#}\d]\x{100}/', 435 '/[\d-\s]\x{100}/' => 'False [] range "\d-" {#} m/[\d-{#}\s]\x{100}/', 436 => 'False [] range "\d-" {#} m/[\d-{#}\s]\x{100}/', 437 '/[a-[:digit:]]\x{100}/' => 'False [] range "a-[:digit:]" {#} m/[a-[:digit:]{#}]\x{100}/', 438 => 'False [] range "a-[:digit:]" {#} m/[a-[:digit:]{#}]\x{100}/', 439 '/[[:digit:]-b]\x{100}/' => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}b]\x{100}/', 440 => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}b]\x{100}/', 441 '/[[:alpha:]-[:digit:]]\x{100}/' => 'False [] range "[:alpha:]-" {#} m/[[:alpha:]-{#}[:digit:]]\x{100}/', 442 => 'False [] range "[:alpha:]-" {#} m/[[:alpha:]-{#}[:digit:]]\x{100}/', 443 '/[[:digit:]-[:alpha:]]\x{100}/' => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}[:alpha:]]\x{100}/', 444 => 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}[:alpha:]]\x{100}/', 445 '/[a\zb]\x{100}/' => 'Unrecognized escape \z in character class passed through {#} m/[a\z{#}b]\x{100}/', 446 => 'Unrecognized escape \z in character class {#} m/[a\z{#}b]\x{100}/', 447 '/[a b]/' => "", 448 => 'Literal vertical space in [] is illegal except under /x {#} m/[a {#}b]/', 449 '/:{4,a}/' => 'Unescaped left brace in regex is passed through {#} m/:{{#}4,a}/', 450 => 'Unescaped left brace in regex is illegal here {#} m/:{{#}4,a}/', 451 '/xa{3\,4}y/' => 'Unescaped left brace in regex is passed through {#} m/xa{{#}3\,4}y/', 452 => 'Unescaped left brace in regex is illegal here {#} m/xa{{#}3\,4}y/', 453 '/\\${[^\\}]*}/' => 'Unescaped left brace in regex is passed through {#} m/\\${{#}[^\\}]*}/', 454 => 'Unescaped left brace in regex is illegal here {#} m/\\${{#}[^\\}]*}/', 455 '/.{/' => 'Unescaped left brace in regex is passed through {#} m/.{{#}/', 456 => 'Unescaped left brace in regex is illegal here {#} m/.{{#}/', 457 '/[x]{/' => 'Unescaped left brace in regex is passed through {#} m/[x]{{#}/', 458 => 'Unescaped left brace in regex is illegal here {#} m/[x]{{#}/', 459 '/\p{Latin}{/' => 'Unescaped left brace in regex is passed through {#} m/\p{Latin}{{#}/', 460 => 'Unescaped left brace in regex is illegal here {#} m/\p{Latin}{{#}/', 461 '/\x{100}\x/' => "", 462 => "Empty \\x {#} m/\\x{100}\\x{#}/", 463 '/\o{ 1 20 }/' => 'Non-octal character \' \' terminates \o early. Resolved as "\o{001}" {#} m/\o{ 1 20 }{#}/', 464 => 'Non-octal character {#} m/\\o{ 1 {#}20 }/', 465 '/\x{ 5 0 }/' => 'Non-hex character \' \' terminates \x early. Resolved as "\x{05}" {#} m/\x{ 5 0 }{#}/', 466 => 'Non-hex character {#} m/\\x{ 5 {#}0 }/', 467); 468 469# These need the character 'ネ' as a marker for mark_as_utf8() 470my @death_utf8 = mark_as_utf8( 471 '/ネ(?<= .*)/' => 'Lookbehind longer than 255 not implemented in regex m/ネ(?<= .*)/', 472 473 '/(?<= ネ{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= ネ{1000})/', 474 475 '/ネ(?ネ)ネ/' => 'Sequence (?ネ...) not recognized {#} m/ネ(?ネ{#})ネ/', 476 477 '/ネ(?(1ネ))ネ/' => 'Switch condition not recognized {#} m/ネ(?(1ネ{#}))ネ/', 478 479 '/(?(1)ネ|y|ヌ)/' => 'Switch (?(condition)... contains too many branches {#} m/(?(1)ネ|y|{#}ヌ)/', 480 481 '/(?(ネ)y|ネ)/' => 'Unknown switch condition (?(...)) {#} m/(?(ネ{#})y|ネ)/', 482 483 '/ネ(?/' => 'Sequence (? incomplete {#} m/ネ(?{#}/', 484 485 '/ネ(?;ネ/' => 'Sequence (?;...) not recognized {#} m/ネ(?;{#}ネ/', 486 '/ネ(?<;ネ/' => 'Group name must start with a non-digit word character {#} m/ネ(?<;{#}ネ/', 487 '/ネ(?\ixネ/' => 'Sequence (?\...) not recognized {#} m/ネ(?\{#}ixネ/', 488 '/ネ(?^lu:ネ)/' => 'Regexp modifiers "l" and "u" are mutually exclusive {#} m/ネ(?^lu{#}:ネ)/', 489'/ネ(?lil:ネ)/' => 'Regexp modifier "l" may not appear twice {#} m/ネ(?lil{#}:ネ)/', 490'/ネ(?aaia:ネ)/' => 'Regexp modifier "a" may appear a maximum of twice {#} m/ネ(?aaia{#}:ネ)/', 491'/ネ(?i-l:ネ)/' => 'Regexp modifier "l" may not appear after the "-" {#} m/ネ(?i-l{#}:ネ)/', 492 493 '/ネ((ネ)/' => 'Unmatched ( {#} m/ネ({#}(ネ)/', 494 495 "/ネ{$inf_p1}ネ/" => "Quantifier in {,} bigger than $inf_m1 {#} m/ネ{$inf_p1\{#}}ネ/", 496 "/ネ{$inf_p1,}ネ/" => "Quantifier in {,} bigger than $inf_m1 {#} m/ネ{$inf_p1\{#},}ネ/", 497 "/ネ{01}ネ/" => "Invalid quantifier in {,} {#} m/ネ{01{#}}ネ/", 498 "/ネ{1,02}ネ/" => "Invalid quantifier in {,} {#} m/ネ{1,02{#}}ネ/", 499 500 501 '/ネ**ネ/' => 'Nested quantifiers {#} m/ネ**{#}ネ/', 502 503 '/ネ[ネ/' => 'Unmatched [ {#} m/ネ[{#}ネ/', 504 505 '/*ネ/', => 'Quantifier follows nothing {#} m/*{#}ネ/', 506 507 '/ネ\p{ネ/' => 'Missing right brace on \p{} {#} m/ネ\p{{#}ネ/', 508 509 '/(ネ)\2ネ/' => 'Reference to nonexistent group {#} m/(ネ)\2{#}ネ/', 510 511 '/\g{ネ/; #no latin1' => 'Sequence \g{... not terminated {#} m/\g{ネ{#}/', 512 513 'my $m = "ネ\\\"; $m =~ $m', => 'Trailing \ in regex m/ネ\/', 514 515 '/\x{ネ/' => 'Missing right brace on \x{} {#} m/\x{{#}ネ/', 516 '/ネ[\x{ネ]ネ/' => 'Missing right brace on \x{} {#} m/ネ[\x{{#}ネ]ネ/', 517 '/ネ[\x{ネ]/' => 'Missing right brace on \x{} {#} m/ネ[\x{{#}ネ]/', 518 519 '/ネ\o{ネ/' => 'Missing right brace on \o{} {#} m/ネ\o{{#}ネ/', 520 521 '/[ネ-a]ネ/' => 'Invalid [] range "ネ-a" {#} m/[ネ-a{#}]ネ/', 522 523 '/ネ\p{}ネ/' => 'Empty \p{} {#} m/ネ\p{{#}}ネ/', 524 525 '/ネ(?[[[:ネ]]])ネ/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[[[:ネ]]{#}])ネ/", 526 '/ネ(?[[[:ネ: ])ネ/' => "Syntax error in (?[...]) {#} m/ネ(?[[[:ネ: ])ネ{#}/", 527 '/ネ(?[[[::]]])ネ/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[[[::]]{#}])ネ/", 528 '/ネ(?[[[:ネ:]]])ネ/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[[[:ネ:]]{#}])ネ/", 529 '/ネ(?[ネ])ネ/' => 'Unexpected character {#} m/ネ(?[ネ{#}])ネ/', 530 '/ネ(?[ + [ネ] ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/ネ(?[ +{#} [ネ] ])/', 531 '/ネ(?[ \cK - ( + [ネ] ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/ネ(?[ \cK - ( +{#} [ネ] ) ])/', 532 '/ネ(?[ \cK ( [ネ] ) ])/' => 'Unexpected \'(\' with no preceding operator {#} m/ネ(?[ \cK ({#} [ネ] ) ])/', 533 '/ネ(?[ \cK [ネ] ])ネ/' => 'Operand with no preceding operator {#} m/ネ(?[ \cK [ネ{#}] ])ネ/', 534 '/ネ(?[ \0004 ])ネ/' => 'Need exactly 3 octal digits {#} m/ネ(?[ \0004 {#}])ネ/', 535 '/(?[ \o{ネ} ])ネ/' => 'Non-octal character {#} m/(?[ \o{ネ{#}} ])ネ/', 536 '/ネ(?[ \o{} ])ネ/' => 'Empty \o{} {#} m/ネ(?[ \o{}{#} ])ネ/', 537 '/(?[ \x{ネ} ])ネ/' => 'Non-hex character {#} m/(?[ \x{ネ{#}} ])ネ/', 538 '/(?[ \p{ネ} ])/' => 'Can\'t find Unicode property definition "ネ" {#} m/(?[ \p{ネ}{#} ])/', 539 '/(?[ \p{ ネ = bar } ])/' => 'Can\'t find Unicode property definition "ネ = bar" {#} m/(?[ \p{ ネ = bar }{#} ])/', 540 '/ネ(?[ \t ]/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[ \\t ]{#}/", 541 '/(?[ \t + \e # ネ This was supposed to be a comment ])/' => 542 "Syntax error in (?[...]) {#} m/(?[ \\t + \\e # ネ This was supposed to be a comment ]){#}/", 543 'm/(*ネ)ネ/' => q<Unknown '(*...)' construct 'ネ' {#} m/(*ネ){#}ネ/>, 544 '/\cネ/' => "Character following \"\\c\" must be printable ASCII {#} m/\\cネ{#}/", 545 '/[\cネ]/' => "Character following \"\\c\" must be printable ASCII {#} m/[\\cネ{#}]/", 546 '/\b{ネ}/' => "'ネ' is an unknown bound type {#} m/\\b{ネ{#}}/", 547 '/\B{ネ}/' => "'ネ' is an unknown bound type {#} m/\\B{ネ{#}}/", 548); 549push @death, @death_utf8; 550 551my @death_utf8_only_under_strict = ( 552 "m'ネ[\\y]ネ'" => 'Unrecognized escape \y in character class passed through {#} m/ネ[\y{#}]ネ/', 553 => 'Unrecognized escape \y in character class {#} m/ネ[\y{#}]ネ/', 554 'm/ネ[ネ-\d]ネ/' => 'False [] range "ネ-\d" {#} m/ネ[ネ-\d{#}]ネ/', 555 => 'False [] range "ネ-\d" {#} m/ネ[ネ-\d{#}]ネ/', 556 'm/ネ[\w-ネ]ネ/' => 'False [] range "\w-" {#} m/ネ[\w-{#}ネ]ネ/', 557 => 'False [] range "\w-" {#} m/ネ[\w-{#}ネ]ネ/', 558 'm/ネ[ネ-\pM]ネ/' => 'False [] range "ネ-\pM" {#} m/ネ[ネ-\pM{#}]ネ/', 559 => 'False [] range "ネ-\pM" {#} m/ネ[ネ-\pM{#}]ネ/', 560 '/ネ[ネ-[:digit:]]ネ/' => 'False [] range "ネ-[:digit:]" {#} m/ネ[ネ-[:digit:]{#}]ネ/', 561 => 'False [] range "ネ-[:digit:]" {#} m/ネ[ネ-[:digit:]{#}]ネ/', 562 '/ネ[\d-\s]ネ/' => 'False [] range "\d-" {#} m/ネ[\d-{#}\s]ネ/', 563 => 'False [] range "\d-" {#} m/ネ[\d-{#}\s]ネ/', 564 '/ネ[a\zb]ネ/' => 'Unrecognized escape \z in character class passed through {#} m/ネ[a\z{#}b]ネ/', 565 => 'Unrecognized escape \z in character class {#} m/ネ[a\z{#}b]ネ/', 566); 567# Tests involving a user-defined charnames translator are in pat_advanced.t 568 569# In the following arrays of warnings, the value can be an array of things to 570# expect. If the empty string, it means no warning should be raised. 571 572 573# Key-value pairs of code/error of code that should have non-fatal regexp 574# warnings. Most currently have \x{100} appended to them to force them to be 575# upgraded to UTF-8, and the first pass restarted. Previously this would 576# cause some warnings to be output twice. This tests that that behavior has 577# been fixed. 578 579my @warning = ( 580 'm/\b*\x{100}/' => '\b* matches null string many times {#} m/\b*{#}\x{100}/', 581 '/\b{g}/a' => "Using /u for '\\b{g}' instead of /a {#} m/\\b{g}{#}/", 582 '/\B{gcb}/a' => "Using /u for '\\B{gcb}' instead of /a {#} m/\\B{gcb}{#}/", 583 'm/[:blank:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:blank:]{#}\x{100}/', 584 'm/[[:cntrl:]][:^ascii:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[[:cntrl:]][:^ascii:]{#}\x{100}/', 585 'm/[[:ascii]]\x{100}/' => "Assuming NOT a POSIX class since there is no terminating ':' {#} m/[[:ascii{#}]]\\x{100}/", 586 'm/(?[[:word]])\x{100}/' => "Assuming NOT a POSIX class since there is no terminating ':' {#} m/(?[[:word{#}]])\\x{100}/", 587 "m'\\y\\x{100}'" => 'Unrecognized escape \y passed through {#} m/\y{#}\x{100}/', 588 '/x{3,1}/' => 'Quantifier {n,m} with n > m can\'t match {#} m/x{3,1}{#}/', 589 '/\08/' => 'Non-octal character \'8\' terminates \0 early. Resolved as "\0008" {#} m/\08{#}/', 590 591 '/\018/' => 'Non-octal character \'8\' terminates \0 early. Resolved as "\0018" {#} m/\018{#}/', 592 '/(?=a)*/' => '(?=a)* matches null string many times {#} m/(?=a)*{#}/', 593 'my $x = \'\m\'; qr/a$x/' => 'Unrecognized escape \m passed through {#} m/a\m{#}/', 594 '/\q/' => 'Unrecognized escape \q passed through {#} m/\q{#}/', 595 '/\q\p{Any}/' => 'Unrecognized escape \q passed through {#} m/\q{#}\p{Any}/', 596 597 # These two tests do not include the marker, because regcomp.c no 598 # longer knows where it goes by the time this warning is emitted. 599 # See [perl #122680] regcomp warning gives wrong position of 600 # problem. 601 '/(?=a){1,3}\x{100}/' => 'Quantifier unexpected on zero-length expression in regex m/(?=a){1,3}\x{100}/', 602 '/(a|b)(?=a){3}\x{100}/' => 'Quantifier unexpected on zero-length expression in regex m/(a|b)(?=a){3}\x{100}/', 603 604 '/\_/' => "", 605 '/[\006]/' => "", 606 '/[:alpha:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:alpha:]{#}\x{100}/', 607 '/[:zog:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/[:zog:]{#}\x{100}/', 608 '/[.zog.]\x{100}/' => 'POSIX syntax [. .] belongs inside character classes (but this one isn\'t implemented) {#} m/[.zog.]{#}\x{100}/', 609 '/[a-b]/' => "", 610 '/(?c)\x{100}/' => 'Useless (?c) - use /gc modifier {#} m/(?c{#})\x{100}/', 611 '/(?-c)\x{100}/' => 'Useless (?-c) - don\'t use /gc modifier {#} m/(?-c{#})\x{100}/', 612 '/(?g)\x{100}/' => 'Useless (?g) - use /g modifier {#} m/(?g{#})\x{100}/', 613 '/(?-g)\x{100}/' => 'Useless (?-g) - don\'t use /g modifier {#} m/(?-g{#})\x{100}/', 614 '/(?o)\x{100}/' => 'Useless (?o) - use /o modifier {#} m/(?o{#})\x{100}/', 615 '/(?-o)\x{100}/' => 'Useless (?-o) - don\'t use /o modifier {#} m/(?-o{#})\x{100}/', 616 '/(?g-o)\x{100}/' => [ 'Useless (?g) - use /g modifier {#} m/(?g{#}-o)\x{100}/', 617 'Useless (?-o) - don\'t use /o modifier {#} m/(?g-o{#})\x{100}/', 618 ], 619 '/(?g-c)\x{100}/' => [ 'Useless (?g) - use /g modifier {#} m/(?g{#}-c)\x{100}/', 620 'Useless (?-c) - don\'t use /gc modifier {#} m/(?g-c{#})\x{100}/', 621 ], 622 # (?c) means (?g) error won't be thrown 623 '/(?o-cg)\x{100}/' => [ 'Useless (?o) - use /o modifier {#} m/(?o{#}-cg)\x{100}/', 624 'Useless (?-c) - don\'t use /gc modifier {#} m/(?o-c{#}g)\x{100}/', 625 ], 626 '/(?ogc)\x{100}/' => [ 'Useless (?o) - use /o modifier {#} m/(?o{#}gc)\x{100}/', 627 'Useless (?g) - use /g modifier {#} m/(?og{#}c)\x{100}/', 628 'Useless (?c) - use /gc modifier {#} m/(?ogc{#})\x{100}/', 629 ], 630 '/a{1,1}?\x{100}/' => 'Useless use of greediness modifier \'?\' {#} m/a{1,1}?{#}\x{100}/', 631 "/(?[ [ % - % ] ])/" => "", 632 "/(?[ [ : - \\x$colon_hex ] ])\\x{100}/" => "\": - \\x$colon_hex \" is more clearly written simply as \":\" {#} m/(?[ [ : - \\x$colon_hex {#}] ])\\x{100}/", 633 "/(?[ [ \\x$colon_hex - : ] ])\\x{100}/" => "\"\\x$colon_hex\ - : \" is more clearly written simply as \":\" {#} m/(?[ [ \\x$colon_hex - : {#}] ])\\x{100}/", 634 "/(?[ [ \\t - \\x$tab_hex ] ])\\x{100}/" => "\"\\t - \\x$tab_hex \" is more clearly written simply as \"\\t\" {#} m/(?[ [ \\t - \\x$tab_hex {#}] ])\\x{100}/", 635 "/(?[ [ \\x$tab_hex - \\t ] ])\\x{100}/" => "\"\\x$tab_hex\ - \\t \" is more clearly written simply as \"\\t\" {#} m/(?[ [ \\x$tab_hex - \\t {#}] ])\\x{100}/", 636 "/(?[ [ $B_hex - C ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ $B_hex - C {#}] ])/", 637 "/(?[ [ A - $B_hex ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ A - $B_hex {#}] ])/", 638 "/(?[ [ $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 {#}] ])/", 639 "/(?[ [ $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 {#}] ])/", 640 "/[alnum]/" => "", 641 "/[^alnum]/" => "", 642 '/[:blank]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/[:blank{#}]\x{100}/', 643 '/[[:digit]]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:digit{#}]]\x{100}/', # [perl # 8904] 644 '/[[:digit:foo]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:digit:{#}foo]\x{100}/', 645 '/[[:di#it:foo]\x{100}/x' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:di#it:{#}foo]\x{100}/', 646 '/[[:dgit]]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:dgit{#}]]\x{100}/', 647 '/[[:dgit:foo]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:dgit:{#}foo]\x{100}/', 648 '/[[:dgt]]\x{100}/' => "", # Far enough away from a real class to not be recognized as one 649 '/[[:dgt:foo]\x{100}/' => "", 650 '/[[:DIGIT]]\x{100}/' => [ 'Assuming NOT a POSIX class since the name must be all lowercase letters {#} m/[[:DIGIT{#}]]\x{100}/', 651 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:DIGIT{#}]]\x{100}/', 652 ], 653 '/[[digit]\x{100}/' => [ 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[{#}digit]\x{100}/', 654 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[digit{#}]\x{100}/', 655 ], 656 '/[[alpha]]\x{100}/' => [ 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[{#}alpha]]\x{100}/', 657 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[alpha{#}]]\x{100}/', 658 ], 659 '/[[^word]\x{100}/' => [ 'Assuming NOT a POSIX class since the \'^\' must come after the colon {#} m/[[^{#}word]\x{100}/', 660 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[^{#}word]\x{100}/', 661 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[^word{#}]\x{100}/', 662 ], 663 '/[[ ^ : 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}/', 664 'Assuming NOT a POSIX class since the \'^\' must come after the colon {#} m/[[ ^{#} : x d i g i t : ] ]\x{100}/', 665 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ ^ {#}: x d i g i t : ] ]\x{100}/', 666 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ ^ : {#}x d i g i t : ] ]\x{100}/', 667 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ ^ : x d i g i t : ]{#} ]\x{100}/', 668 $only_strict_marker . 'Unescaped literal \']\' {#} m/[[ ^ : x d i g i t : ] ]{#}\x{100}/', 669 ], 670 '/[foo:lower:]]\x{100}/' => 'Assuming NOT a POSIX class since it doesn\'t start with a \'[\' {#} m/[foo{#}:lower:]]\x{100}/', 671 '/[[;upper;]]\x{100}/' => [ 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[[;{#}upper;]]\x{100}/', 672 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[[;upper;]{#}]\x{100}/', 673 ], 674 '/[foo;punct;]]\x{100}/' => [ 'Assuming NOT a POSIX class since it doesn\'t start with a \'[\' {#} m/[foo{#};punct;]]\x{100}/', 675 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[foo;{#}punct;]]\x{100}/', 676 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[foo;punct;]{#}]\x{100}/', 677 ], 678 '/[][[:alpha:]]/' => "", # [perl #127581] 679 '/[][[:alpha:]\\@\\\\^_?]/' => "", # [perl #131522] 680 '/(?[[:w:]])/' => "", 681 '/([.].*)[.]/' => "", # [perl #127582] 682 '/[.].*[.]/' => "", # [perl #127604] 683 '/abc/xix' => "", 684 '/(?xmsixp:abc)/' => "", 685 '/(?xmsixp)abc/' => "", 686 '/(?xxxx:abc)/' => "", 687 688); # See comments before this for why '\x{100}' is generally needed 689 690# These need the character 'ネ' as a marker for mark_as_utf8() 691my @warnings_utf8 = mark_as_utf8( 692 'm/ネ\b*ネ/' => '\b* matches null string many times {#} m/ネ\b*{#}ネ/', 693 '/(?=ネ)*/' => '(?=ネ)* matches null string many times {#} m/(?=ネ)*{#}/', 694 'm/ネ[:foo:]ネ/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/ネ[:foo:]{#}ネ/', 695 '/ネ(?c)ネ/' => 'Useless (?c) - use /gc modifier {#} m/ネ(?c{#})ネ/', 696 '/utf8 ネ (?ogc) ネ/' => [ 697 'Useless (?o) - use /o modifier {#} m/utf8 ネ (?o{#}gc) ネ/', 698 'Useless (?g) - use /g modifier {#} m/utf8 ネ (?og{#}c) ネ/', 699 'Useless (?c) - use /gc modifier {#} m/utf8 ネ (?ogc{#}) ネ/', 700 ], 701 '/ネ[[:ネ:]]ネ/' => "", 702 '/ネ(?[[:ネ:]])ネ/' => "", 703 704); 705 706push @warning, @warnings_utf8; 707 708my @warning_only_under_strict = ( 709 '/[\N{U+00}-\x01]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+00}-\x01{#}]\x{100}/', 710 '/[\x00-\N{SOH}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\x00-\N{U+01}{#}]\x{100}/', 711 '/[\N{DEL}-\o{377}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+7F}-\o{377}{#}]\x{100}/', 712 '/[\o{0}-\N{U+01}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\o{0}-\N{U+01}{#}]\x{100}/', 713 '/[\000-\N{U+01}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\000-\N{U+01}{#}]\x{100}/', 714 '/[\N{DEL}-\377]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+7F}-\377{#}]\x{100}/', 715 '/[\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}/', 716 '/[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}/', 717 '/[\N{U+00}-\a]\x{100}/' => "", 718 '/[\a-\N{U+FF}]\x{100}/' => "", 719 '/[\N{U+100}-\x{101}]/' => "", 720 "/[%-%]/" => "", 721 "/[:-\\x$colon_hex]\\x{100}/" => "\":-\\x$colon_hex\" is more clearly written simply as \":\" {#} m/[:-\\x$colon_hex\{#}]\\x{100}/", 722 "/[\\x$colon_hex-:]\\x{100}/" => "\"\\x$colon_hex-:\" is more clearly written simply as \":\" {#} m/[\\x$colon_hex\-:{#}]\\x{100}/", 723 "/[\\t-\\x$tab_hex]\\x{100}/" => "\"\\t-\\x$tab_hex\" is more clearly written simply as \"\\t\" {#} m/[\\t-\\x$tab_hex\{#}]\\x{100}/", 724 "/[\\x$tab_hex-\\t]\\x{100}/" => "\"\\x$tab_hex-\\t\" is more clearly written simply as \"\\t\" {#} m/[\\x$tab_hex\-\\t{#}]\\x{100}/", 725 "/[$B_hex-C]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[$B_hex-C{#}]/", 726 "/[A-$B_hex]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[A-$B_hex\{#}]/", 727 "/[$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-$high_mixed_alpha\{#}]/", 728 "/[$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\{#}]/", 729 '/\b<GCB}/' => 'Unescaped literal \'}\' {#} m/\b<GCB}{#}/', 730 '/[ ]def]/' => 'Unescaped literal \']\' {#} m/[ ]def]{#}/', 731 '/(?)/' => 'Empty (?) without any modifiers {#} m/(?){#}/', # [perl #132851] 732); 733 734my @warning_utf8_only_under_strict = mark_as_utf8( 735 '/ネ[᪉-᪐]/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ[᪉-᪐{#}]/", 736 '/ネ(?[ [ ᪉ - ᪐ ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ ᪉ - ᪐ {#}] ])/", 737 '/ネ[᧙-᧚]/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ[᧙-᧚{#}]/", 738 '/ネ(?[ [ ᧙ - ᧚ ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ ᧙ - ᧚ {#}] ])/", 739 '/ネ(?[ [ �� - �� ] ])/; #no latin1' => "", 740 '/ネ(?[ [ �� - �� ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ �� - �� {#}] ])/", 741 '/ネ(?[ [ �� - �� ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ �� - �� {#}] ])/", 742); 743 744push @warning_only_under_strict, @warning_utf8_only_under_strict; 745 746my @experimental_vlb = ( 747 '/(?<=(p|qq|rrr))/' => 'Variable length positive lookbehind with capturing' . 748 ' is experimental {#} m/(?<=(p|qq|rrr)){#}/', 749 '/(?<!(p|qq|rrr))/' => 'Variable length negative lookbehind with capturing' . 750 ' is experimental {#} m/(?<!(p|qq|rrr)){#}/', 751 '/(?| (?=(foo)) | (?<=(foo)|p) )/' 752 => 'Variable length positive lookbehind with capturing' . 753 ' is experimental {#} m/(?| (?=(foo)) | (?<=(foo)|p) ){#}/', 754 '/(?| (?=(foo)) | (?<=(foo)|p) )/x' 755 => 'Variable length positive lookbehind with capturing' . 756 ' is experimental {#} m/(?| (?=(foo)) | (?<=(foo)|p) ){#}/', 757 '/(?| (?=(foo)) | (?<!(foo)|p) )/' 758 => 'Variable length negative lookbehind with capturing' . 759 ' is experimental {#} m/(?| (?=(foo)) | (?<!(foo)|p) ){#}/', 760 '/(?| (?=(foo)) | (?<!(foo)|p) )/x' 761 => 'Variable length negative lookbehind with capturing' . 762 ' is experimental {#} m/(?| (?=(foo)) | (?<!(foo)|p) ){#}/', 763 '/(?<!(foo|bop(*ACCEPT)|bar)baz)/' 764 => 'Variable length negative lookbehind with capturing' . 765 ' is experimental {#} m/(?<!(foo|bop(*ACCEPT)|bar)baz){#}/', 766 '/(?<=(foo|bop(*ACCEPT)|bar)baz)/' 767 => 'Variable length positive lookbehind with capturing' . 768 ' is experimental {#} m/(?<=(foo|bop(*ACCEPT)|bar)baz){#}/', 769); 770 771my @wildcard = ( 772 'm!(?[\p{name=/KATAKANA/}])$!' => 773 [ 774 'The Unicode property wildcards feature is experimental', 775 'Using just the single character results returned by \p{} in (?[...]) {#} m/(?[\p{name=/KATAKANA/}{#}])$/' 776 ], # [GH #17732] Null pointer deref 777); 778 779my @deprecated = ( 780 '/^{/' => "", 781 '/foo|{/' => "", 782 '/foo|^{/' => "", 783 '/foo(:?{bar)/' => "", 784 '/\s*{/' => "", 785 '/a{3,4}{/' => "", 786); 787 788for my $strict ("", "use re 'strict';") { 789 790 # First time just use @death; but under strict we add the things that fail 791 # there. Doing it this way makes sure that 'strict' doesnt change the 792 # things that are already fatal when not under strict. 793 if ($strict) { 794 for (my $i = 0; $i < @death_only_under_strict; $i += 3) { 795 push @death, $death_only_under_strict[$i], # The regex 796 $death_only_under_strict[$i+2]; # The fatal msg 797 } 798 for (my $i = 0; $i < @death_utf8_only_under_strict; $i += 3) { 799 800 # Same with the utf8 versions 801 push @death, mark_as_utf8($death_utf8_only_under_strict[$i], 802 $death_utf8_only_under_strict[$i+2]); 803 } 804 } 805 for (my $i = 0; $i < @death; $i += 2) { 806 my $regex = $death[$i] =~ s/ default_ (on | off) //rx; 807 my $expect = fixup_expect($death[$i+1], $strict); 808 if ($expect eq "") { 809 fail("$0: Internal error: '$death[$i]' should have an error message"); 810 } 811 else { 812 no warnings 'experimental::re_strict'; 813 no warnings 'experimental::uniprop_wildcards'; 814 815 warning_is(sub { 816 my $meaning_of_life; 817 my $eval_string = "$strict $regex"; 818 $_ = "x"; 819 eval "$eval_string; \$meaning_of_life = 42"; 820 ok(! defined $meaning_of_life, display_rx("$eval_string died")); 821 my $error= $@; 822 if ($error =~ qr/\Q$expect/) { 823 ok(1, "... and gave expected message"); 824 } else { 825 ok(0, display_rx($eval_string)); 826 diag("Have: " . display_rx(add_markers($error))); 827 diag("Want: " . display_rx($death[$i+1])); 828 } 829 }, undef, "... and no other warnings"); 830 } 831 } 832} 833 834for my $strict ("", "no warnings 'experimental::re_strict'; use re 'strict';") { 835 my @warning_tests = @warning; 836 837 # Build the tests for @warning. Use the strict/non-strict versions 838 # appropriately. 839 if ($strict) { 840 push @warning_tests, @warning_only_under_strict; 841 } 842 else { 843 for (my $i = 0; $i < @warning_only_under_strict; $i += 2) { 844 845 # (?[ ]) are always under strict 846 if ($warning_only_under_strict[$i] =~ /\Q(?[/) { 847 push @warning_tests, $warning_only_under_strict[$i], # The regex 848 $warning_only_under_strict[$i+1]; 849 } 850 else { 851 push @warning_tests, $warning_only_under_strict[$i], # The regex 852 ""; # No warning because not strict 853 } 854 } 855 for (my $i = 0; $i < @death_only_under_strict; $i += 3) { 856 push @warning_tests, $death_only_under_strict[$i], # The regex 857 $death_only_under_strict[$i+1]; # The warning 858 } 859 for (my $i = 0; $i < @death_utf8_only_under_strict; $i += 3) { 860 push @warning_tests, mark_as_utf8($death_utf8_only_under_strict[$i], 861 $death_utf8_only_under_strict[$i+1]); 862 } 863 } 864 865 foreach my $ref ( 866 \@warning_tests, 867 \@wildcard, 868 \@deprecated, 869 \@experimental_vlb, 870 ){ 871 my $warning_type; 872 my $default_on; 873 if ($ref == \@warning_tests) { 874 $warning_type = 'regexp, digit'; 875 $default_on = $strict; 876 } 877 elsif ($ref == \@deprecated) { 878 $warning_type = 'regexp, deprecated'; 879 $default_on = 1; 880 } 881 elsif ($ref == \@wildcard) { 882 $warning_type = 'experimental::uniprop_wildcards'; 883 $default_on = 1; 884 } 885 elsif ($ref == \@experimental_vlb) { 886 $warning_type = 'experimental::vlb'; 887 $default_on = 1; 888 } 889 else { 890 fail("$0: Internal error: Unexpected loop variable"); 891 } 892 893 for (my $i = 0; $i < @$ref; $i += 2) { 894 my $this_default_on = $default_on; 895 my $regex = $ref->[$i]; 896 if ($regex =~ s/ default_ (on | off) //x) { 897 $this_default_on = $1 eq 'on'; 898 } 899 my @expect = fixup_expect($ref->[$i+1], $strict); 900 901 # A length-1 array with an empty warning means no warning gets 902 # generated at all. 903 undef @expect if @expect == 1 && $expect[0] eq ""; 904 905 { 906 $_ = "x"; 907 #use feature 'unicode_eval'; 908 #print STDERR __LINE__, ": ", "eval '$strict no warnings; $regex'", "\n"; 909 eval "$strict no warnings; $regex"; 910 } 911 my $display = display_rx($regex); # used in the test names. 912 if (is($@, "", "$strict $display did not die")) { 913 my @got = capture_warnings(sub { 914 $_ = "x"; 915 eval "$strict $regex" }); 916 my $count = @expect; 917 if (! is(scalar @got, scalar @expect, 918 "... and gave expected number ($count) of warnings")) 919 { 920 if (@got < @expect) { 921 $count = @got; 922 note display_rx("Expected warnings not gotten:\n\t" 923 . join("\n\t", @expect[$count .. $#expect])); 924 } 925 else { 926 note display_rx("Unexpected warnings gotten:\n\t" 927 . join("\n\t", @got[$count .. $#got])); 928 } 929 } 930 foreach my $i (0 .. $count - 1) { 931 if (! like($got[$i], qr/\Q$expect[$i]/, 932 "... and gave expected warning")) 933 { 934 chomp($got[$i]); 935 chomp($expect[$i]); 936 diag(display_rx("GOT\n'$got[$i]'\nEXPECT\n'$expect[$i]'")); 937 } 938 else { 939 # Turning off this type of warning should make the 940 # count go down by at least 1. 941 ok ($count - 1 >= capture_warnings(sub { 942 $_ = "x"; 943 eval "$strict no warnings '$warning_type'; $regex;" } 944 ), 945 "... and turning off '$warning_type' warnings suppressed it"); 946 947 # Test that whether the warning is on by default is 948 # correct. This test relies on the fact that we 949 # are outside the scope of any ‘use warnings’. 950 local $^W; 951 my @warns = capture_warnings(sub { 952 $_ = "x"; 953 eval "$strict $regex" 954 }); 955 # Warning should be on as well if is testing 956 # '(?[...])' which turns on strict 957 if ( $this_default_on 958 || ( grep { $_ =~ /\Q(?[/ } @expect 959 and $ref != \@warning_tests)) 960 { 961 ok @warns > 0, "... and the warning is on by default"; 962 } 963 elsif (! (ok @warns == 0, 964 "... and the warning is off by default")) 965 { 966 diag(display_rx("GOT\n" . join "\n", @warns)); 967 } 968 } 969 } 970 } 971 } 972 } 973} 974 975done_testing(); 976