#!./perl -w
+$|=1; # outherwise things get mixed up in output
+
BEGIN {
chdir 't' if -d 't';
- @INC = qw '../lib ../ext/re';
require './test.pl';
- skip_all_without_unicode_tables();
+ set_up_inc( qw '../lib ../ext/re' );
eval 'require Config'; # assume defaults if this fails
}
+skip_all_without_unicode_tables();
+
use strict;
use open qw(:utf8 :std);
-##
+# Kind of a kludge to mark warnings to be expected only if we are testing
+# under "use re 'strict'"
+my $only_strict_marker = ':expected_only_under_strict';
+
## If the markers used are changed (search for "MARKER1" in regcomp.c),
## update only these two regexs, and leave the {#} in the @death/@warning
## arrays below. The {#} is a meta-marker -- it marks where the marker should
## go.
-##
-## Returns empty string if that is what is expected. Otherwise, handles
-## either a scalar, turning it into a single element array; or a ref to an
-## array, adjusting each element. If called in array context, returns an
-## array, otherwise the join of all elements
-sub fixup_expect {
- my $expect_ref = shift;
+sub fixup_expect ($$) {
+
+ # Fixes up the expected results by inserting the boiler plate text.
+ # Returns empty string if that is what is expected. Otherwise, handles
+ # either a scalar, turning it into a single element array; or a ref to an
+ # array, adjusting each element. If called in array context, returns an
+ # array, otherwise the join of all elements.
+
+ # The string $only_strict_marker will be removed from any expect line it
+ # begins, and if $strict is not true, that expect line will be removed
+ # from the output (hence won't be expected)
+
+ my ($expect_ref, $strict) = @_;
return "" if $expect_ref eq "";
my @expect;
@expect = $expect_ref;
}
+ my @new_expect;
foreach my $element (@expect) {
- $element =~ s/{\#}/in regex; marked by <-- HERE in/;
- $element =~ s/{\#}/ <-- HERE /;
+ $element =~ s/\{\#\}/in regex; marked by <-- HERE in/;
+ $element =~ s/\{\#\}/ <-- HERE /;
$element .= " at ";
+ next if $element =~ s/ ^ $only_strict_marker \s* //x && ! $strict;
+ push @new_expect, $element;
}
- return wantarray ? @expect : join "", @expect;
+ return wantarray ? @new_expect : join "", @new_expect;
+}
+
+sub add_markers {
+ my ($element)= @_;
+ $element =~ s/ at .* line \d+\.?\n$//;
+ $element =~ s/in regex; marked by <-- HERE in/{#}/;
+ $element =~ s/ <-- HERE /{#}/;
+ return $element;
}
## Because we don't "use utf8" in this file, we need to do some extra legwork
my $inf_m1 = ($Config::Config{reg_infty} || 32767) - 1;
my $inf_p1 = $inf_m1 + 2;
-##
-## Key-value pairs of code/error of code that should have fatal errors.
-##
+my $B_hex = sprintf("\\x%02X", ord "B");
+my $low_mixed_alpha = ('A' lt 'a') ? 'A' : 'a';
+my $high_mixed_alpha = ('A' lt 'a') ? 'a' : 'A';
+my $low_mixed_digit = ('A' lt '0') ? 'A' : '0';
+my $high_mixed_digit = ('A' lt '0') ? '0' : 'A';
+
+my $colon_hex = sprintf "%02X", ord(":");
+my $tab_hex = sprintf "%02X", ord("\t");
+
+# Key-value pairs of strings eval'd as patterns => warn/error messages that
+# they should generate. In some cases, the value is an array of multiple
+# messages. Some groups have the message(s) be default on; others, default
+# off. This can be overridden on an individual key basis by preceding the
+# pattern string with either 'default_on' or 'default_off'
+#
+# The first set are those that should be fatal errors.
+
my @death =
(
'/[[=foo=]]/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/[[=foo=]{#}]/',
'/(?lil:foo)/' => 'Regexp modifier "l" may not appear twice {#} m/(?lil{#}:foo)/',
'/(?aaia:foo)/' => 'Regexp modifier "a" may appear a maximum of twice {#} m/(?aaia{#}:foo)/',
'/(?i-l:foo)/' => 'Regexp modifier "l" may not appear after the "-" {#} m/(?i-l{#}:foo)/',
-'/a\b{cde/' => 'Use "\b\{" instead of "\b{" {#} m/a\{#}b{cde/',
-'/a\B{cde/' => 'Use "\B\{" instead of "\B{" {#} m/a\{#}B{cde/',
'/((x)/' => 'Unmatched ( {#} m/({#}(x)/',
+ '/{(}/' => 'Unmatched ( {#} m/{({#}}/', # [perl #127599]
"/x{$inf_p1}/" => "Quantifier in {,} bigger than $inf_m1 {#} m/x{{#}$inf_p1}/",
'/[z-a]/' => 'Invalid [] range "z-a" {#} m/[z-a{#}]/',
- '/\p/' => 'Empty \p{} {#} m/\p{#}/',
-
+ '/\p/' => 'Empty \p {#} m/\p{#}/',
+ '/\P/' => 'Empty \P {#} m/\P{#}/',
+ '/\p{}/' => 'Empty \p{} {#} m/\p{{#}}/',
'/\P{}/' => 'Empty \P{} {#} m/\P{{#}}/',
- '/(?[[[:word]]])/' => "Unmatched ':' in POSIX class {#} m/(?[[[:word{#}]]])/",
- '/(?[[:word]])/' => "Unmatched ':' in POSIX class {#} m/(?[[:word{#}]])/",
- '/(?[[[:digit: ])/' => "Unmatched '[' in POSIX class {#} m/(?[[[:digit:{#} ])/",
- '/(?[[:digit: ])/' => "Unmatched '[' in POSIX class {#} m/(?[[:digit:{#} ])/",
- '/(?[[[::]]])/' => "POSIX class [::] unknown {#} m/(?[[[::]{#}]])/",
- '/(?[[[:w:]]])/' => "POSIX class [:w:] unknown {#} m/(?[[[:w:]{#}]])/",
- '/(?[[:w:]])/' => "POSIX class [:w:] unknown {#} m/(?[[:w:]{#}])/",
+
+'/a\b{cde/' => 'Missing right brace on \b{} {#} m/a\b{{#}cde/',
+'/a\B{cde/' => 'Missing right brace on \B{} {#} m/a\B{{#}cde/',
+
+ '/\b{}/' => 'Empty \b{} {#} m/\b{}{#}/',
+ '/\B{}/' => 'Empty \B{} {#} m/\B{}{#}/',
+
+ '/\b{gc}/' => "'gc' is an unknown bound type {#} m/\\b{gc{#}}/",
+ '/\B{gc}/' => "'gc' is an unknown bound type {#} m/\\B{gc{#}}/",
+
+ '/(?[[[::]]])/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[[[::]]{#}])/",
+ '/(?[[[:w:]]])/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[[[:w:]]{#}])/",
'/(?[a])/' => 'Unexpected character {#} m/(?[a{#}])/',
- '/(?[\t])/l' => '(?[...]) not valid in locale {#} m/(?[{#}\t])/',
'/(?[ + \t ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ +{#} \t ])/',
'/(?[ \cK - ( + \t ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ \cK - ( +{#} \t ) ])/',
'/(?[ \cK ( \t ) ])/' => 'Unexpected \'(\' with no preceding operator {#} m/(?[ \cK ({#} \t ) ])/',
'/(?[ \x{} ])/' => 'Number with no digits {#} m/(?[ \x{}{#} ])/',
'/(?[ \cK + ) ])/' => 'Unexpected \')\' {#} m/(?[ \cK + ){#} ])/',
'/(?[ \cK + ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ \cK + {#}])/',
- '/(?[ \p{foo} ])/' => 'Property \'foo\' is unknown {#} m/(?[ \p{foo}{#} ])/',
- '/(?[ \p{ foo = bar } ])/' => 'Property \'foo = bar\' is unknown {#} m/(?[ \p{ foo = bar }{#} ])/',
+ '/(?[ ( ) ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ ( ){#} ])/',
+ '/(?[[0]+()+])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[[0]+(){#}+])/',
+ '/(?[ \p{foo} ])/' => 'Can\'t find Unicode property definition "foo" {#} m/(?[ \p{foo}{#} ])/',
+ '/(?[ \p{ foo = bar } ])/' => 'Can\'t find Unicode property definition "foo = bar" {#} m/(?[ \p{ foo = bar }{#} ])/',
'/(?[ \8 ])/' => 'Unrecognized escape \8 in character class {#} m/(?[ \8{#} ])/',
- '/(?[ \t ]/' => 'Syntax error in (?[...]) in regex m/(?[ \t ]/',
- '/(?[ [ \t ]/' => 'Syntax error in (?[...]) in regex m/(?[ [ \t ]/',
- '/(?[ \t ] ]/' => 'Syntax error in (?[...]) in regex m/(?[ \t ] ]/',
- '/(?[ [ ] ]/' => 'Syntax error in (?[...]) in regex m/(?[ [ ] ]/',
- '/(?[ \t + \e # This was supposed to be a comment ])/' => 'Syntax error in (?[...]) in regex m/(?[ \t + \e # This was supposed to be a comment ])/',
+ '/(?[ \t ]/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[ \\t ]{#}/",
+ '/(?[ [ \t ]/' => "Syntax error in (?[...]) {#} m/(?[ [ \\t ]{#}/",
+ '/(?[ \t ] ]/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[ \\t ]{#} ]/",
+ '/(?[ [ ] ]/' => "Syntax error in (?[...]) {#} m/(?[ [ ] ]{#}/",
+ '/(?[ \t + \e # This was supposed to be a comment ])/' =>
+ "Syntax error in (?[...]) {#} m/(?[ \\t + \\e # This was supposed to be a comment ]){#}/",
'/(?[ ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ {#}])/',
'm/(?[[a-\d]])/' => 'False [] range "a-\d" {#} m/(?[[a-\d{#}]])/',
'm/(?[[\w-x]])/' => 'False [] range "\w-" {#} m/(?[[\w-{#}x]])/',
'm/\cß/' => "Character following \"\\c\" must be printable ASCII",
'/((?# 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)/',
'/((?# 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)/',
+ '/((?# 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)/',
+
+ '/(*script_runfoo)/' => 'Unknown \'(*...)\' construct \'script_runfoo\' {#} m/(*script_runfoo){#}/',
+ '/(*srfoo)/' => 'Unknown \'(*...)\' construct \'srfoo\' {#} m/(*srfoo){#}/',
+ '/(*script_run)/' => '\'(*script_run\' requires a terminating \':\' {#} m/(*script_run{#})/',
+ '/(*sr)/' => '\'(*sr\' requires a terminating \':\' {#} m/(*sr{#})/',
+ '/(*script_run/' => 'Unterminated \'(*...\' construct {#} m/(*script_run{#}/',
+ '/(*sr/' => 'Unterminated \'(*...\' construct {#} m/(*sr{#}/',
+ '/(*script_run:foo/' => 'Unterminated \'(*...\' argument {#} m/(*script_run:foo{#}/',
+ '/(*sr:foo/' => 'Unterminated \'(*...\' argument {#} m/(*sr:foo{#}/',
+ '/(?[\ &!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ &!{#}])/', # [perl #126180]
+ '/(?[\ +!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ +!{#}])/', # [perl #126180]
+ '/(?[\ -!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ -!{#}])/', # [perl #126180]
+ '/(?[\ ^!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ ^!{#}])/', # [perl #126180]
+ '/(?[\ |!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[\ |!{#}])/', # [perl #126180]
+ '/(?[()-!])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[(){#}-!])/', # [perl #126204]
+ '/(?[!()])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[!(){#}])/', # [perl #126404]
+ '/\w{/' => 'Unescaped left brace in regex is illegal here {#} m/\w{{#}/',
+ '/\q{/' => 'Unescaped left brace in regex is illegal here {#} m/\q{{#}/',
+ '/\A{/' => 'Unescaped left brace in regex is illegal here {#} m/\A{{#}/',
+ '/(?<=/' => 'Sequence (?... not terminated {#} m/(?<={#}/', # [perl #128170]
+
);
-# These are messages that are warnings when not strict; death under 'use re
-# "strict". See comment before @warnings as to why some have a \x{100} in
-# them. This array has 3 elements per construct. [0] is the regex to use;
-# [1] is the message under no strict, and [2] is under strict.
+# These are messages that are death under 'use re "strict"', and may or may
+# not warn otherwise. See comment before @warning as to why some have a
+# \x{100} in them. This array has 3 elements per construct. [0] is the regex
+# to use; [1] is the message under no strict (empty to not warn), and [2] is
+# under strict.
my @death_only_under_strict = (
'm/\xABC/' => "",
=> 'Use \x{...} for more than two hex characters {#} m/\xABC{#}/',
=> 'Non-hex character {#} m/\x{ABCDEFG{#}}/',
'm/[\x{ABCDEFG}]/' => 'Illegal hexadecimal digit \'G\' ignored',
=> 'Non-hex character {#} m/[\x{ABCDEFG{#}}]/',
- 'm/[[:ascii]]/' => "",
- => 'Unmatched \':\' in POSIX class {#} m/[[:ascii{#}]]/',
- 'm/[\N{}]/' => 'Ignoring zero length \\N{} in character class {#} m/[\\N{}{#}]/',
- => 'Zero length \\N{} {#} m/[\\N{}]{#}/',
"m'[\\y]\\x{100}'" => 'Unrecognized escape \y in character class passed through {#} m/[\y{#}]\x{100}/',
=> 'Unrecognized escape \y in character class {#} m/[\y{#}]\x{100}/',
'm/[a-\d]\x{100}/' => 'False [] range "a-\d" {#} m/[a-\d{#}]\x{100}/',
=> 'False [] range "[:digit:]-" {#} m/[[:digit:]-{#}[:alpha:]]\x{100}/',
'/[a\zb]\x{100}/' => 'Unrecognized escape \z in character class passed through {#} m/[a\z{#}b]\x{100}/',
=> 'Unrecognized escape \z in character class {#} m/[a\z{#}b]\x{100}/',
+ 'default_on/:{4,a}/' => 'Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through {#} m/:{{#}4,a}/',
+ => 'Unescaped left brace in regex is illegal here {#} m/:{{#}4,a}/',
+ 'default_on/xa{3\,4}y/' => 'Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through {#} m/xa{{#}3\,4}y/',
+ => 'Unescaped left brace in regex is illegal here {#} m/xa{{#}3\,4}y/',
+ 'default_on/\\${[^\\}]*}/' => 'Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through {#} m/\\${{#}[^\\}]*}/',
+ => 'Unescaped left brace in regex is illegal here {#} m/\\${{#}[^\\}]*}/',
+ '/[a\fb]/' => "",
+ => 'Literal vertical space in [] is illegal except under /x {#} m/[a\f{#}b]/',
);
# These need the character 'ネ' as a marker for mark_as_utf8()
my @death_utf8 = mark_as_utf8(
- '/ネ[[=ネ=]]ネ/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/ネ[[=ネ=]{#}]ネ/',
'/ネ(?<= .*)/' => 'Variable length lookbehind not implemented in regex m/ネ(?<= .*)/',
'/(?<= ネ{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= ネ{1000})/',
'/ネ[\x{ネ]/' => 'Missing right brace on \x{} {#} m/ネ[\x{{#}ネ]/',
'/ネ\o{ネ/' => 'Missing right brace on \o{ {#} m/ネ\o{{#}ネ/',
- '/ネ[[:ネ:]]ネ/' => 'POSIX class [:ネ:] unknown {#} m/ネ[[:ネ:]{#}]ネ/',
-
- '/ネ[[=ネ=]]ネ/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/ネ[[=ネ=]{#}]ネ/',
-
- '/ネ[[.ネ.]]ネ/' => 'POSIX syntax [. .] is reserved for future extensions {#} m/ネ[[.ネ.]{#}]ネ/',
'/[ネ-a]ネ/' => 'Invalid [] range "ネ-a" {#} m/[ネ-a{#}]ネ/',
'/ネ\p{}ネ/' => 'Empty \p{} {#} m/ネ\p{{#}}ネ/',
- '/ネ(?[[[:ネ]]])ネ/' => "Unmatched ':' in POSIX class {#} m/ネ(?[[[:ネ{#}]]])ネ/",
- '/ネ(?[[[:ネ: ])ネ/' => "Unmatched '[' in POSIX class {#} m/ネ(?[[[:ネ:{#} ])ネ/",
- '/ネ(?[[[::]]])ネ/' => "POSIX class [::] unknown {#} m/ネ(?[[[::]{#}]])ネ/",
- '/ネ(?[[[:ネ:]]])ネ/' => "POSIX class [:ネ:] unknown {#} m/ネ(?[[[:ネ:]{#}]])ネ/",
- '/ネ(?[[:ネ:]])ネ/' => "POSIX class [:ネ:] unknown {#} m/ネ(?[[:ネ:]{#}])ネ/",
+ '/ネ(?[[[:ネ]]])ネ/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[[[:ネ]]{#}])ネ/",
+ '/ネ(?[[[:ネ: ])ネ/' => "Syntax error in (?[...]) {#} m/ネ(?[[[:ネ: ])ネ{#}/",
+ '/ネ(?[[[::]]])ネ/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[[[::]]{#}])ネ/",
+ '/ネ(?[[[:ネ:]]])ネ/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[[[:ネ:]]{#}])ネ/",
'/ネ(?[ネ])ネ/' => 'Unexpected character {#} m/ネ(?[ネ{#}])ネ/',
- '/ネ(?[ネ])/l' => '(?[...]) not valid in locale {#} m/ネ(?[{#}ネ])/',
'/ネ(?[ + [ネ] ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/ネ(?[ +{#} [ネ] ])/',
'/ネ(?[ \cK - ( + [ネ] ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/ネ(?[ \cK - ( +{#} [ネ] ) ])/',
'/ネ(?[ \cK ( [ネ] ) ])/' => 'Unexpected \'(\' with no preceding operator {#} m/ネ(?[ \cK ({#} [ネ] ) ])/',
'/(?[ \o{ネ} ])ネ/' => 'Non-octal character {#} m/(?[ \o{ネ{#}} ])ネ/',
'/ネ(?[ \o{} ])ネ/' => 'Number with no digits {#} m/ネ(?[ \o{}{#} ])ネ/',
'/(?[ \x{ネ} ])ネ/' => 'Non-hex character {#} m/(?[ \x{ネ{#}} ])ネ/',
- '/(?[ \p{ネ} ])/' => 'Property \'ネ\' is unknown {#} m/(?[ \p{ネ}{#} ])/',
- '/(?[ \p{ ネ = bar } ])/' => 'Property \'ネ = bar\' is unknown {#} m/(?[ \p{ ネ = bar }{#} ])/',
- '/ネ(?[ \t ]/' => 'Syntax error in (?[...]) in regex m/ネ(?[ \t ]/',
- '/(?[ \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 ])/',
- 'm/(*ネ)ネ/' => q<Unknown verb pattern 'ネ' {#} m/(*ネ){#}ネ/>,
+ '/(?[ \p{ネ} ])/' => 'Can\'t find Unicode property definition "ネ" {#} m/(?[ \p{ネ}{#} ])/',
+ '/(?[ \p{ ネ = bar } ])/' => 'Can\'t find Unicode property definition "ネ = bar" {#} m/(?[ \p{ ネ = bar }{#} ])/',
+ '/ネ(?[ \t ]/' => "Unexpected ']' with no following ')' in (?[... {#} m/ネ(?[ \\t ]{#}/",
+ '/(?[ \t + \e # ネ This was supposed to be a comment ])/' =>
+ "Syntax error in (?[...]) {#} m/(?[ \\t + \\e # ネ This was supposed to be a comment ]){#}/",
+ 'm/(*ネ)ネ/' => q<Unknown '(*...)' construct 'ネ' {#} m/(*ネ){#}ネ/>,
'/\cネ/' => "Character following \"\\c\" must be printable ASCII",
+ '/\b{ネ}/' => "'ネ' is an unknown bound type {#} m/\\b{ネ{#}}/",
+ '/\B{ネ}/' => "'ネ' is an unknown bound type {#} m/\\B{ネ{#}}/",
);
push @death, @death_utf8;
my @warning = (
'm/\b*\x{100}/' => '\b* matches null string many times {#} m/\b*{#}\x{100}/',
+ '/\b{g}/a' => "Using /u for '\\b{g}' instead of /a {#} m/\\b{g}{#}/",
+ '/\B{gcb}/a' => "Using /u for '\\B{gcb}' instead of /a {#} m/\\B{gcb}{#}/",
'm/[:blank:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:blank:]{#}\x{100}/',
+ 'm/[[:cntrl:]][:^ascii:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[[:cntrl:]][:^ascii:]{#}\x{100}/',
+ 'm/[[:ascii]]\x{100}/' => "Assuming NOT a POSIX class since there is no terminating ':' {#} m/[[:ascii{#}]]\\x{100}/",
+ 'm/(?[[:word]])\x{100}/' => "Assuming NOT a POSIX class since there is no terminating ':' {#} m/(?[[:word{#}]])\\x{100}/",
"m'\\y\\x{100}'" => 'Unrecognized escape \y passed through {#} m/\y{#}\x{100}/',
'/x{3,1}/' => 'Quantifier {n,m} with n > m can\'t match {#} m/x{3,1}{#}/',
'/\08/' => '\'\08\' resolved to \'\o{0}8\' {#} m/\08{#}/',
'/\_/' => "",
'/[\006]/' => "",
'/[:alpha:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:alpha:]{#}\x{100}/',
- '/[:zog:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:zog:]{#}\x{100}/',
- '/[.zog.]\x{100}/' => 'POSIX syntax [. .] belongs inside character classes {#} m/[.zog.]{#}\x{100}/',
+ '/[:zog:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/[:zog:]{#}\x{100}/',
+ '/[.zog.]\x{100}/' => 'POSIX syntax [. .] belongs inside character classes (but this one isn\'t implemented) {#} m/[.zog.]{#}\x{100}/',
'/[a-b]/' => "",
'/(?c)\x{100}/' => 'Useless (?c) - use /gc modifier {#} m/(?c{#})\x{100}/',
'/(?-c)\x{100}/' => 'Useless (?-c) - don\'t use /gc modifier {#} m/(?-c{#})\x{100}/',
'Useless (?c) - use /gc modifier {#} m/(?ogc{#})\x{100}/',
],
'/a{1,1}?\x{100}/' => 'Useless use of greediness modifier \'?\' {#} m/a{1,1}?{#}\x{100}/',
- '/b{3} +\x{100}/x' => 'Useless use of greediness modifier \'+\' {#} m/b{3} +{#}\x{100}/',
+ "/(?[ [ % - % ] ])/" => "",
+ "/(?[ [ : - \\x$colon_hex ] ])\\x{100}/" => "\": - \\x$colon_hex \" is more clearly written simply as \":\" {#} m/(?[ [ : - \\x$colon_hex {#}] ])\\x{100}/",
+ "/(?[ [ \\x$colon_hex - : ] ])\\x{100}/" => "\"\\x$colon_hex\ - : \" is more clearly written simply as \":\" {#} m/(?[ [ \\x$colon_hex - : {#}] ])\\x{100}/",
+ "/(?[ [ \\t - \\x$tab_hex ] ])\\x{100}/" => "\"\\t - \\x$tab_hex \" is more clearly written simply as \"\\t\" {#} m/(?[ [ \\t - \\x$tab_hex {#}] ])\\x{100}/",
+ "/(?[ [ \\x$tab_hex - \\t ] ])\\x{100}/" => "\"\\x$tab_hex\ - \\t \" is more clearly written simply as \"\\t\" {#} m/(?[ [ \\x$tab_hex - \\t {#}] ])\\x{100}/",
+ "/(?[ [ $B_hex - C ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ $B_hex - C {#}] ])/",
+ "/(?[ [ A - $B_hex ] ])/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/(?[ [ A - $B_hex {#}] ])/",
+ "/(?[ [ $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 {#}] ])/",
+ "/(?[ [ $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 {#}] ])/",
+ "/[alnum]/" => "",
+ "/[^alnum]/" => "",
+ '/[:blank]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/[:blank{#}]\x{100}/',
+ '/[[:digit]]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:digit{#}]]\x{100}/', # [perl # 8904]
+ '/[[:digit:foo]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:digit:{#}foo]\x{100}/',
+ '/[[:di#it:foo]\x{100}/x' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:di#it:{#}foo]\x{100}/',
+ '/[[:dgit]]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:dgit{#}]]\x{100}/',
+ '/[[:dgit:foo]\x{100}/' => 'Assuming NOT a POSIX class since there is no terminating \']\' {#} m/[[:dgit:{#}foo]\x{100}/',
+ '/[[:dgt]]\x{100}/' => "", # Far enough away from a real class to not be recognized as one
+ '/[[:dgt:foo]\x{100}/' => "",
+ '/[[:DIGIT]]\x{100}/' => [ 'Assuming NOT a POSIX class since the name must be all lowercase letters {#} m/[[:DIGIT{#}]]\x{100}/',
+ 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[:DIGIT{#}]]\x{100}/',
+ ],
+ '/[[digit]\x{100}/' => [ 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[{#}digit]\x{100}/',
+ 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[digit{#}]\x{100}/',
+ ],
+ '/[[alpha]]\x{100}/' => [ 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[{#}alpha]]\x{100}/',
+ 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[alpha{#}]]\x{100}/',
+ ],
+ '/[[^word]\x{100}/' => [ 'Assuming NOT a POSIX class since the \'^\' must come after the colon {#} m/[[^{#}word]\x{100}/',
+ 'Assuming NOT a POSIX class since there must be a starting \':\' {#} m/[[^{#}word]\x{100}/',
+ 'Assuming NOT a POSIX class since there is no terminating \':\' {#} m/[[^word{#}]\x{100}/',
+ ],
+ '/[[ ^ : 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}/',
+ 'Assuming NOT a POSIX class since the \'^\' must come after the colon {#} m/[[ ^{#} : 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}/',
+ 'Assuming NOT a POSIX class since no blanks are allowed in one {#} m/[[ ^ : {#}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}/',
+ $only_strict_marker . 'Unescaped literal \']\' {#} m/[[ ^ : x d i g i t : ] ]{#}\x{100}/',
+ ],
+ '/[foo:lower:]]\x{100}/' => 'Assuming NOT a POSIX class since it doesn\'t start with a \'[\' {#} m/[foo{#}:lower:]]\x{100}/',
+ '/[[;upper;]]\x{100}/' => [ 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[[;{#}upper;]]\x{100}/',
+ 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[[;upper;]{#}]\x{100}/',
+ ],
+ '/[foo;punct;]]\x{100}/' => [ 'Assuming NOT a POSIX class since it doesn\'t start with a \'[\' {#} m/[foo{#};punct;]]\x{100}/',
+ 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[foo;{#}punct;]]\x{100}/',
+ 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[foo;punct;]{#}]\x{100}/',
+ ],
+ '/[][[:alpha:]]/' => "", # [perl #127581]
+ '/[][[:alpha:]\\@\\\\^_?]/' => "", # [perl #131522]
+ '/(?[[:w:]])/' => "",
+ '/([.].*)[.]/' => "", # [perl #127582]
+ '/[.].*[.]/' => "", # [perl #127604]
+ '/abc/xix' => "",
+ '/(?xmsixp:abc)/' => "",
+ '/(?xmsixp)abc/' => "",
+ '/(?xxxx:abc)/' => "",
+
); # See comments before this for why '\x{100}' is generally needed
# These need the character 'ネ' as a marker for mark_as_utf8()
my @warnings_utf8 = mark_as_utf8(
'm/ネ\b*ネ/' => '\b* matches null string many times {#} m/ネ\b*{#}ネ/',
'/(?=ネ)*/' => '(?=ネ)* matches null string many times {#} m/(?=ネ)*{#}/',
- 'm/ネ[:foo:]ネ/' => 'POSIX syntax [: :] belongs inside character classes {#} m/ネ[:foo:]{#}ネ/',
+ 'm/ネ[:foo:]ネ/' => 'POSIX syntax [: :] belongs inside character classes (but this one isn\'t fully valid) {#} m/ネ[:foo:]{#}ネ/',
'/ネ(?c)ネ/' => 'Useless (?c) - use /gc modifier {#} m/ネ(?c{#})ネ/',
'/utf8 ネ (?ogc) ネ/' => [
'Useless (?o) - use /o modifier {#} m/utf8 ネ (?o{#}gc) ネ/',
'Useless (?g) - use /g modifier {#} m/utf8 ネ (?og{#}c) ネ/',
'Useless (?c) - use /gc modifier {#} m/utf8 ネ (?ogc{#}) ネ/',
],
+ '/ネ[[:ネ:]]ネ/' => "",
+ '/ネ(?[[:ネ:]])ネ/' => "",
);
'/[\o{0}-\N{U+01}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\o{0}-\N{U+01}{#}]\x{100}/',
'/[\000-\N{U+01}]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\000-\N{U+01}{#}]\x{100}/',
'/[\N{DEL}-\377]\x{100}/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+7F}-\377{#}]\x{100}/',
- '/[\N{U+00}-A]\x{100}/' => "",
- '/[a-\N{U+FF}]\x{100}/' => "",
+ '/[\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}/',
+ '/[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}/',
'/[\N{U+00}-\a]\x{100}/' => "",
'/[\a-\N{U+FF}]\x{100}/' => "",
'/[\N{U+FF}-\x{100}]/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+FF}-\x{100}{#}]/',
'/[\N{U+100}-\x{101}]/' => "",
+ "/[%-%]/" => "",
+ "/[:-\\x$colon_hex]\\x{100}/" => "\":-\\x$colon_hex\" is more clearly written simply as \":\" {#} m/[:-\\x$colon_hex\{#}]\\x{100}/",
+ "/[\\x$colon_hex-:]\\x{100}/" => "\"\\x$colon_hex-:\" is more clearly written simply as \":\" {#} m/[\\x$colon_hex\-:{#}]\\x{100}/",
+ "/[\\t-\\x$tab_hex]\\x{100}/" => "\"\\t-\\x$tab_hex\" is more clearly written simply as \"\\t\" {#} m/[\\t-\\x$tab_hex\{#}]\\x{100}/",
+ "/[\\x$tab_hex-\\t]\\x{100}/" => "\"\\x$tab_hex-\\t\" is more clearly written simply as \"\\t\" {#} m/[\\x$tab_hex\-\\t{#}]\\x{100}/",
+ "/[$B_hex-C]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[$B_hex-C{#}]/",
+ "/[A-$B_hex]/" => "Ranges of ASCII printables should be some subset of \"0-9\", \"A-Z\", or \"a-z\" {#} m/[A-$B_hex\{#}]/",
+ "/[$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-$high_mixed_alpha\{#}]/",
+ "/[$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\{#}]/",
+ '/\b<GCB}/' => 'Unescaped literal \'}\' {#} m/\b<GCB}{#}/',
+ '/[ ]def]/' => 'Unescaped literal \']\' {#} m/[ ]def]{#}/',
+);
+
+my @warning_utf8_only_under_strict = mark_as_utf8(
+ '/ネ[᪉-᪐]/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ[᪉-᪐{#}]/",
+ '/ネ(?[ [ ᪉ - ᪐ ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ ᪉ - ᪐ {#}] ])/",
+ '/ネ[᧙-᧚]/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ[᧙-᧚{#}]/",
+ '/ネ(?[ [ ᧙ - ᧚ ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ ᧙ - ᧚ {#}] ])/",
+ '/ネ(?[ [ 𝟘 - 𝟡 ] ])/; #no latin1' => "",
+ '/ネ(?[ [ 𝟧 - 𝟱 ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ 𝟧 - 𝟱 {#}] ])/",
+ '/ネ(?[ [ 𝟧 - 𝟰 ] ])/; #no latin1' => "Ranges of digits should be from the same group of 10 {#} m/ネ(?[ [ 𝟧 - 𝟰 {#}] ])/",
);
+push @warning_only_under_strict, @warning_utf8_only_under_strict;
+
my @experimental_regex_sets = (
'/(?[ \t ])/' => 'The regex_sets feature is experimental {#} m/(?[{#} \t ])/',
'use utf8; /utf8 ネ (?[ [\tネ] ])/' => do { use utf8; 'The regex_sets feature is experimental {#} m/utf8 ネ (?[{#} [\tネ] ])/' },
'/noutf8 ネ (?[ [\tネ] ])/' => 'The regex_sets feature is experimental {#} m/noutf8 ネ (?[{#} [\tネ] ])/',
);
+my @experimental_script_run = (
+ '/(*script_run:paypal.com)/' => 'The script_run feature is experimental {#} m/(*script_run:{#}paypal.com)/',
+ 'use utf8; /utf8 ネ (*script_run:ネ)/' => do { use utf8; 'The script_run feature is experimental {#} m/utf8 ネ (*script_run:{#}ネ)/' },
+ '/noutf8 ネ (*script_run:ネ)/' => 'The script_run feature is experimental {#} m/noutf8 ネ (*script_run:{#}ネ)/',
+);
+
my @deprecated = (
- '/\w{/' => 'Unescaped left brace in regex is deprecated, passed through {#} m/\w{{#}/',
- '/\q{/' => [
- 'Unrecognized escape \q{ passed through {#} m/\q{{#}/',
- 'Unescaped left brace in regex is deprecated, passed through {#} m/\q{{#}/'
- ],
- '/:{4,a}/' => 'Unescaped left brace in regex is deprecated, passed through {#} m/:{{#}4,a}/',
- '/abc/xix' => 'Having more than one /x regexp modifier is deprecated',
- '/(?xmsixp:abc)/' => 'Having more than one /x regexp modifier is deprecated',
- '/(?xmsixp)abc/' => 'Having more than one /x regexp modifier is deprecated',
- '/(?xxxx:abc)/' => 'Having more than one /x regexp modifier is deprecated',
+ '/^{/' => "",
+ '/foo|{/' => "",
+ '/foo|^{/' => "",
+ '/foo({bar)/' => 'Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.32), passed through {#} m/foo({{#}bar)/',
+ '/foo(:?{bar)/' => "",
+ '/\s*{/' => "",
+ '/a{3,4}{/' => "",
+ '/.{/' => 'Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through {#} m/.{{#}/',
+ '/[x]{/' => 'Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through {#} m/[x]{{#}/',
+ '/\p{Latin}{/' => 'Unescaped left brace in regex is deprecated here (and will be fatal in Perl 5.30), passed through {#} m/\p{Latin}{{#}/',
);
for my $strict ("", "use re 'strict';") {
}
}
for (my $i = 0; $i < @death; $i += 2) {
- my $regex = $death[$i];
- my $expect = fixup_expect($death[$i+1]);
- no warnings 'experimental::regex_sets';
- no warnings 'experimental::re_strict';
+ my $regex = $death[$i] =~ s/ default_ (on | off) //rx;
+ my $expect = fixup_expect($death[$i+1], $strict);
+ if ($expect eq "") {
+ fail("$0: Internal error: '$death[$i]' should have an error message");
+ }
+ else {
+ no warnings 'experimental::regex_sets';
+ no warnings 'experimental::script_run';
+ no warnings 'experimental::re_strict';
- warning_is(sub {
+ warning_is(sub {
+ my $meaning_of_life;
my $eval_string = "$strict $regex";
$_ = "x";
- eval $eval_string;
- like($@, qr/\Q$expect/, $eval_string);
- }, undef, "... and died without any other warnings");
+ eval "$eval_string; \$meaning_of_life = 42";
+ ok (! defined $meaning_of_life, "$eval_string died");
+ my $error= $@;
+ if ($error =~ qr/\Q$expect/) {
+ ok(1, "... and gave expected message");
+ } else {
+ ok(0,$eval_string);
+ diag("Have: " . _q(add_markers($error)));
+ diag("Want: " . _q($death[$i+1]));
+ }
+ }, undef, "... and no other warnings");
+ }
}
}
}
else {
for (my $i = 0; $i < @warning_only_under_strict; $i += 2) {
+
+ # (?[ ]) are always under strict
if ($warning_only_under_strict[$i] =~ /\Q(?[/) {
push @warning_tests, $warning_only_under_strict[$i], # The regex
$warning_only_under_strict[$i+1];
}
}
- foreach my $ref (\@warning_tests, \@experimental_regex_sets, \@deprecated) {
+ foreach my $ref (\@warning_tests,
+ \@experimental_regex_sets,
+ \@experimental_script_run,
+ \@deprecated)
+ {
my $warning_type;
my $turn_off_warnings = "";
my $default_on;
if ($ref == \@warning_tests) {
$warning_type = 'regexp, digit';
- $turn_off_warnings = "no warnings 'experimental::regex_sets';";
+ $turn_off_warnings = "no warnings 'experimental::regex_sets';"
+ . "no warnings 'experimental::script_run';";
$default_on = $strict;
}
elsif ($ref == \@deprecated) {
$warning_type = 'regexp, deprecated';
$default_on = 1;
}
- else {
+ elsif ($ref == \@experimental_regex_sets) {
$warning_type = 'experimental::regex_sets';
$default_on = 1;
}
+ elsif ($ref == \@experimental_script_run) {
+ $warning_type = 'experimental::script_run';
+ $default_on = 1;
+ }
+ else {
+ fail("$0: Internal error: Unexpected loop variable");
+ }
+
for (my $i = 0; $i < @$ref; $i += 2) {
+ my $this_default_on = $default_on;
my $regex = $ref->[$i];
- my @expect = fixup_expect($ref->[$i+1]);
+ if ($regex =~ s/ default_ (on | off) //x) {
+ $this_default_on = $1 eq 'on';
+ }
+ my @expect = fixup_expect($ref->[$i+1], $strict);
# A length-1 array with an empty warning means no warning gets
# generated at all.
{
$_ = "x";
+ #use feature 'unicode_eval';
+ #print STDERR __LINE__, ": ", "eval '$strict no warnings; $regex'", "\n";
eval "$strict no warnings; $regex";
}
if (is($@, "", "$strict $regex did not die")) {
eval "$strict $regex" });
# Warning should be on as well if is testing
# '(?[...])' which turns on strict
- if ($default_on || grep { $_ =~ /\Q(?[/ } @expect ) {
+ if ($this_default_on || grep { $_ =~ /\Q(?[/ } @expect ) {
ok @warns > 0, "... and the warning is on by default";
}
else {