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
return @ret;
}
-my $inf_m1 = ($Config::Config{reg_infty} || 32767) - 1;
+my $inf_m1 = ($Config::Config{reg_infty} || 65535) - 1;
my $inf_p1 = $inf_m1 + 2;
my $B_hex = sprintf("\\x%02X", ord "B");
my $colon_hex = sprintf "%02X", ord(":");
my $tab_hex = sprintf "%02X", ord("\t");
-##
-## Key-value pairs of code/error of code that should have fatal errors.
-##
+# 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 $bug133423 = "(?[(?^:(?[\\\x00]))\\]\x00|2[^^]\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80\x80])R.\\670";
+
my @death =
(
'/[[=foo=]]/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/[[=foo=]{#}]/',
- '/[[=a]=]]/' => 'POSIX syntax [= =] is reserved for future extensions {#} m/[[=a]=]{#}]/',
- '/[[.a].]]/' => 'POSIX syntax [. .] is reserved for future extensions {#} m/[[.a].]{#}]/',
- '/(?<= .*)/' => 'Variable length lookbehind not implemented in regex m/(?<= .*)/',
+ '/(?<= .*)/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= .*)/',
'/(?<= x{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= x{1000})/',
'/(?i-l:foo)/' => 'Regexp modifier "l" may not appear after the "-" {#} m/(?i-l{#}:foo)/',
'/((x)/' => 'Unmatched ( {#} m/({#}(x)/',
+ '/{(}/' => 'Unmatched ( {#} m/{({#}}/', # [perl #127599]
"/x{$inf_p1}/" => "Quantifier in {,} bigger than $inf_m1 {#} m/x{{#}$inf_p1}/",
'/[\x{X]/' => 'Missing right brace on \x{} {#} m/[\x{{#}X]/',
'/[\x{A]/' => 'Missing right brace on \x{} {#} m/[\x{A{#}]/',
- '/\o{1/' => 'Missing right brace on \o{ {#} m/\o{1{#}/',
- '/\o{X/' => 'Missing right brace on \o{ {#} m/\o{{#}X/',
+ '/\o{1/' => 'Missing right brace on \o{} {#} m/\o{1{#}/',
+ '/\o{X/' => 'Missing right brace on \o{} {#} m/\o{{#}X/',
- '/[\o{X]/' => 'Missing right brace on \o{ {#} m/[\o{{#}X]/',
- '/[\o{7]/' => 'Missing right brace on \o{ {#} m/[\o{7{#}]/',
+ '/[\o{X]/' => 'Missing right brace on \o{} {#} m/[\o{{#}X]/',
+ '/[\o{7]/' => 'Missing right brace on \o{} {#} m/[\o{7{#}]/',
'/[[:barf:]]/' => 'POSIX class [:barf:] unknown {#} m/[[:barf:]{#}]/',
'/\b{gc}/' => "'gc' is an unknown bound type {#} m/\\b{gc{#}}/",
'/\B{gc}/' => "'gc' is an unknown bound type {#} m/\\B{gc{#}}/",
- '/(?[[[::]]])/' => "Syntax error in (?[...]) in regex m/(?[[[::]]])/",
- '/(?[[[:w:]]])/' => "Syntax error in (?[...]) in regex m/(?[[[:w:]]])/",
- '/(?[[:w:]])/' => "",
- '/[][[:alpha:]]' => "", # [perl #127581]
+ '/(?[[[::]]])/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[[[::]]{#}])/",
+ '/(?[[[:w:]]])/' => "Unexpected ']' with no following ')' in (?[... {#} m/(?[[[:w:]]{#}])/",
'/(?[a])/' => 'Unexpected character {#} m/(?[a{#}])/',
'/(?[ + \t ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ +{#} \t ])/',
'/(?[ \cK - ( + \t ) ])/' => 'Unexpected binary operator \'+\' with no preceding operand {#} m/(?[ \cK - ( +{#} \t ) ])/',
'/(?[ \0004 ])/' => 'Need exactly 3 octal digits {#} m/(?[ \0004 {#}])/',
'/(?[ \05 ])/' => 'Need exactly 3 octal digits {#} m/(?[ \05 {#}])/',
'/(?[ \o{1038} ])/' => 'Non-octal character {#} m/(?[ \o{1038{#}} ])/',
- '/(?[ \o{} ])/' => 'Number with no digits {#} m/(?[ \o{}{#} ])/',
+ '/(?[ \o{} ])/' => 'Empty \o{} {#} m/(?[ \o{}{#} ])/',
'/(?[ \x{defg} ])/' => 'Non-hex character {#} m/(?[ \x{defg{#}} ])/',
'/(?[ \xabcdef ])/' => 'Use \\x{...} for more than two hex characters {#} m/(?[ \xabc{#}def ])/',
- '/(?[ \x{} ])/' => 'Number with no digits {#} m/(?[ \x{}{#} ])/',
+ '/(?[ \x{} ])/' => 'Empty \x{} {#} m/(?[ \x{}{#} ])/',
'/(?[ \cK + ) ])/' => 'Unexpected \')\' {#} m/(?[ \cK + ){#} ])/',
'/(?[ \cK + ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ \cK + {#}])/',
- '/(?[ ( ) ])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[ ( ) {#}])/',
- '/(?[[0]+()+])/' => 'Incomplete expression within \'(?[ ])\' {#} m/(?[[0]+()+{#}])/',
+ '/(?[ ( ) ])/' => '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/(?[[a-\pM]])/' => 'False [] range "a-\pM" {#} m/(?[[a-\pM{#}]])/',
'm/(?[[\pM-x]])/' => 'False [] range "\pM-" {#} m/(?[[\pM-{#}x]])/',
- '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{#}}]])/',
- 'm/(?[ \p{Digit} & (?(?[ \p{Thai} | \p{Lao} ]))])/' => 'Sequence (?(...) not recognized {#} m/(?[ \p{Digit} & (?({#}?[ \p{Thai} | \p{Lao} ]))])/',
- 'm/(?[ \p{Digit} & (?:(?[ \p{Thai} | \p{Lao} ]))])/' => 'Expecting \'(?flags:(?[...\' {#} m/(?[ \p{Digit} & (?{#}:(?[ \p{Thai} | \p{Lao} ]))])/',
- 'm/\o{/' => 'Missing right brace on \o{ {#} m/\o{{#}/',
+ 'm/(?[[^\N{LATIN CAPITAL LETTER A WITH MACRON AND GRAVE}]])/' => '\N{} here is restricted to one character {#} m/(?[[^\N{U+100.300{#}}]])/',
+ 'm/(?[ \p{Digit} & (?^(?[ \p{Thai} | \p{Lao} ]))])/' => 'Sequence (?^(...) not recognized {#} m/(?[ \p{Digit} & (?^({#}?[ \p{Thai} | \p{Lao} ]))])/',
+ 'm/(?[ \p{Digit} & (?(?[ \p{Thai} | \p{Lao} ]))])/' => 'Unexpected character {#} m/(?[ \p{Digit} & (?{#}(?[ \p{Thai} | \p{Lao} ]))])/',
+ 'm/\o{/' => 'Missing right brace on \o{} {#} m/\o{{#}/',
'm/\o/' => 'Missing braces on \o{} {#} m/\o{#}/',
- 'm/\o{}/' => 'Number with no digits {#} m/\o{}{#}/',
- 'm/[\o{]/' => 'Missing right brace on \o{ {#} m/[\o{{#}]/',
+ 'm/\o{}/' => 'Empty \o{} {#} m/\o{}{#}/',
+ 'm/[\o{]/' => 'Missing right brace on \o{} {#} m/[\o{{#}]/',
'm/[\o]/' => 'Missing braces on \o{} {#} m/[\o{#}]/',
- 'm/[\o{}]/' => 'Number with no digits {#} m/[\o{}{#}]/',
+ 'm/[\o{}]/' => 'Empty \o{} {#} m/[\o{}{#}]/',
'm/(?^-i:foo)/' => 'Sequence (?^-...) not recognized {#} m/(?^-{#}i:foo)/',
'm/\87/' => 'Reference to nonexistent group {#} m/\87{#}/',
'm/a\87/' => 'Reference to nonexistent group {#} m/a\87{#}/',
"m/(?('/" => "Sequence (?('... not terminated {#} m/(?('{#}/",
'm/\g{/' => 'Sequence \g{... not terminated {#} m/\g{{#}/',
'm/\k</' => 'Sequence \k<... not terminated {#} m/\k<{#}/',
- '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{#})/',
+ '/(*pla)/' => '\'(*pla\' requires a terminating \':\' {#} m/(*pla{#})/',
+ '/(*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 #126204]
+ '/(?[\ +!])/' => '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{{#}/',
+ '/.{, 4 }/' => 'Unescaped left brace in regex is illegal here {#} m/.{{#}, 4 }/',
+ '/[x]{, 4}/' => 'Unescaped left brace in regex is illegal here {#} m/[x]{{#}, 4}/',
+ '/\p{Latin}{,4 }/' => 'Unescaped left brace in regex is illegal here {#} m/\p{Latin}{{#},4 }/',
+ '/(?<=/' => 'Sequence (?... not terminated {#} m/(?<={#}/', # [perl #128170]
+ '/\p{vertical \v tab}/' => 'Can\'t find Unicode property definition "vertical \v tab" {#} m/\\p{vertical \v tab}{#}/', # [perl #132055]
+ "/$bug133423/" => "Operand with no preceding operator {#} m/(?[(?^:(?[\\\0]))\\{#}]\0|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/",
+ '/[^/' => 'Unmatched [ {#} m/[{#}^/', # [perl #133767]
+ '/\p{Is_Other_Alphabetic=F}/ ' => 'Can\'t find Unicode property definition "Is_Other_Alphabetic=F" {#} m/\p{Is_Other_Alphabetic=F}{#}/',
+ '/\p{Is_Other_Alphabetic=F}/ ' => 'Can\'t find Unicode property definition "Is_Other_Alphabetic=F" {#} m/\p{Is_Other_Alphabetic=F}{#}/',
+ '/\x{100}(?(/' => 'Unknown switch condition (?(...)) {#} m/\\x{100}(?({#}/', # [perl #133896]
+ '/(?[\N{KEYCAP DIGIT NINE}/' => '\N{} here is restricted to one character {#} m/(?[\\N{U+39.FE0F.20E3{#}}/', # [perl #133988]
+ '/0000000000000000[\N{U+0.00}0000/' => 'Unmatched [ {#} m/0000000000000000[{#}\N{U+0.00}0000/', # [perl #134059]
+ '/\p{nv=\b5\b}/' => 'Can\'t find Unicode property definition "nv=\\b5\\b" {#} m/\\p{nv=\\b5\\b}{#}/',
);
-# 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{#}/',
'm/[\xABC]/' => "",
=> 'Use \x{...} for more than two hex characters {#} m/[\xABC{#}]/',
- # XXX This is a confusing error message. The G isn't ignored; it just
- # terminates the \x. Also some messages below are missing the <-- HERE,
- # aren't all category 'regexp'. (Hence we have to turn off 'digit'
- # messages as well below)
- 'm/\xAG/' => 'Illegal hexadecimal digit \'G\' ignored',
+ # some messages below aren't all category 'regexp'. (Hence we have to
+ # turn off 'digit' messages as well below)
+ 'm/\xAG/' => 'Non-hex character \'G\' terminates \x early. Resolved as "\x0AG" {#} m/\xA{#}G/',
=> 'Non-hex character {#} m/\xAG{#}/',
- 'm/[\xAG]/' => 'Illegal hexadecimal digit \'G\' ignored',
+ 'm/[\xAG]/' => 'Non-hex character \'G\' terminates \x early. Resolved as "\x0AG" {#} m/[\xA{#}G]/',
=> 'Non-hex character {#} m/[\xAG{#}]/',
- 'm/\o{789}/' => 'Non-octal character \'8\'. Resolved as "\o{7}"',
+ 'm/\o{789}/' => 'Non-octal character \'8\' terminates \o early. Resolved as "\o{007}" {#} m/\o{789}{#}/',
=> 'Non-octal character {#} m/\o{78{#}9}/',
- 'm/[\o{789}]/' => 'Non-octal character \'8\'. Resolved as "\o{7}"',
+ 'm/[\o{789}]/' => 'Non-octal character \'8\' terminates \o early. Resolved as "\o{007}" {#} m/[\o{789}{#}]/',
=> 'Non-octal character {#} m/[\o{78{#}9}]/',
'm/\x{}/' => "",
- => 'Number with no digits {#} m/\x{}{#}/',
+ => 'Empty \x{} {#} m/\x{}{#}/',
'm/[\x{}]/' => "",
- => 'Number with no digits {#} m/[\x{}{#}]/',
- 'm/\x{ABCDEFG}/' => 'Illegal hexadecimal digit \'G\' ignored',
+ => 'Empty \x{} {#} m/[\x{}{#}]/',
+ 'm/\x{ABCDEFG}/' => 'Non-hex character \'G\' terminates \x early. Resolved as "\x{ABCDEF}" {#} m/\x{ABCDEFG}{#}/',
=> 'Non-hex character {#} m/\x{ABCDEFG{#}}/',
- 'm/[\x{ABCDEFG}]/' => 'Illegal hexadecimal digit \'G\' ignored',
+ 'm/[\x{ABCDEFG}]/' => 'Non-hex character \'G\' terminates \x early. Resolved as "\x{ABCDEF}" {#} m/[\x{ABCDEFG}{#}]/',
=> 'Non-hex character {#} m/[\x{ABCDEFG{#}}]/',
- 'm/[\N{}]/' => 'Ignoring zero length \\N{} in character class {#} m/[\\N{}{#}]/',
- => 'Zero length \\N{} {#} m/[\\N{}{#}]/',
- '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}/',
'm/[\pM-x]\x{100}/' => 'False [] range "\pM-" {#} m/[\pM-{#}x]\x{100}/',
=> 'False [] range "\pM-" {#} m/[\pM-{#}x]\x{100}/',
'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}{#}]/',
- => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[^\N{U+100.300{#}}]/',
+ => '\N{} here is restricted to one character {#} m/[^\N{U+100.300{#}}]/',
'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}{#}]/',
- => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[\x03-\N{U+100.300{#}}]/',
+ => '\N{} here is restricted to one character {#} m/[\x03-\N{U+100.300{#}}]/',
'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}]/',
- => '\N{} in inverted character class or as a range end-point is restricted to one character {#} m/[\N{U+100.300{#}}-\x{10FFFF}]/',
- '/[\08]/' => '\'\08\' resolved to \'\o{0}8\' {#} m/[\08{#}]/',
+ => '\N{} here is restricted to one character {#} m/[\N{U+100.300{#}}-\x{10FFFF}]/',
+ '/[\08]/' => 'Non-octal character \'8\' terminates \0 early. Resolved as "\0008" {#} m/[\08{#}]/',
=> 'Need exactly 3 octal digits {#} m/[\08{#}]/',
- '/[\018]/' => '\'\018\' resolved to \'\o{1}8\' {#} m/[\018{#}]/',
+ '/[\018]/' => 'Non-octal character \'8\' terminates \0 early. Resolved as "\0018" {#} m/[\018{#}]/',
=> 'Need exactly 3 octal digits {#} m/[\018{#}]/',
'/[\_\0]/' => "",
=> 'Need exactly 3 octal digits {#} m/[\_\0]{#}/',
=> '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}/',
+ '/[a\fb]/' => "",
+ => 'Literal vertical space in [] is illegal except under /x {#} m/[a\f{#}b]/',
+ '/:{4,a}/' => 'Unescaped left brace in regex is passed through {#} m/:{{#}4,a}/',
+ => 'Unescaped left brace in regex is illegal here {#} m/:{{#}4,a}/',
+ '/xa{3\,4}y/' => 'Unescaped left brace in regex is passed through {#} m/xa{{#}3\,4}y/',
+ => 'Unescaped left brace in regex is illegal here {#} m/xa{{#}3\,4}y/',
+ '/\\${[^\\}]*}/' => 'Unescaped left brace in regex is passed through {#} m/\\${{#}[^\\}]*}/',
+ => 'Unescaped left brace in regex is illegal here {#} m/\\${{#}[^\\}]*}/',
+ '/.{/' => 'Unescaped left brace in regex is passed through {#} m/.{{#}/',
+ => 'Unescaped left brace in regex is illegal here {#} m/.{{#}/',
+ '/[x]{/' => 'Unescaped left brace in regex is passed through {#} m/[x]{{#}/',
+ => 'Unescaped left brace in regex is illegal here {#} m/[x]{{#}/',
+ '/\p{Latin}{/' => 'Unescaped left brace in regex is passed through {#} m/\p{Latin}{{#}/',
+ => 'Unescaped left brace in regex is illegal here {#} m/\p{Latin}{{#}/',
+ '/\x{100}\x/' => "",
+ => "Empty \\x {#} m/\\x{100}\\x{#}/",
);
# 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(?<= .*)/' => 'Lookbehind longer than 255 not implemented in regex m/ネ(?<= .*)/',
'/(?<= ネ{1000})/' => 'Lookbehind longer than 255 not implemented in regex m/(?<= ネ{1000})/',
'/ネ[\x{ネ]ネ/' => 'Missing right brace on \x{} {#} m/ネ[\x{{#}ネ]ネ/',
'/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/' => "",
-
- '/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\o{ネ/' => 'Missing right brace on \o{} {#} m/ネ\o{{#}ネ/',
'/[ネ-a]ネ/' => 'Invalid [] range "ネ-a" {#} m/[ネ-a{#}]ネ/',
'/ネ\p{}ネ/' => 'Empty \p{} {#} m/ネ\p{{#}}ネ/',
- '/ネ(?[[[:ネ]]])ネ/' => "Syntax error in (?[...]) in regex m/ネ(?[[[:ネ]]])ネ/",
- '/ネ(?[[[:ネ: ])ネ/' => "Syntax error in (?[...]) in regex m/ネ(?[[[:ネ: ])ネ/",
- '/ネ(?[[[::]]])ネ/' => "Syntax error in (?[...]) in regex m/ネ(?[[[::]]])ネ/",
- '/ネ(?[[[:ネ:]]])ネ/' => "Syntax error in (?[...]) in regex m/ネ(?[[[:ネ:]]])ネ/",
- '/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] ])/' => '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] ])ネ/' => 'Operand with no preceding operator {#} m/ネ(?[ \cK [ネ{#}] ])ネ/',
'/ネ(?[ \0004 ])ネ/' => 'Need exactly 3 octal digits {#} m/ネ(?[ \0004 {#}])ネ/',
'/(?[ \o{ネ} ])ネ/' => 'Non-octal character {#} m/(?[ \o{ネ{#}} ])ネ/',
- '/ネ(?[ \o{} ])ネ/' => 'Number with no digits {#} m/ネ(?[ \o{}{#} ])ネ/',
+ '/ネ(?[ \o{} ])ネ/' => 'Empty \o{} {#} m/ネ(?[ \o{}{#} ])ネ/',
'/(?[ \x{ネ} ])ネ/' => 'Non-hex character {#} m/(?[ \x{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 ]/' => '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/>,
- '/\cネ/' => "Character following \"\\c\" must be printable ASCII",
+ '/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 {#} m/\\cネ{#}/",
+ '/[\cネ]/' => "Character following \"\\c\" must be printable ASCII {#} m/[\\cネ{#}]/",
'/\b{ネ}/' => "'ネ' is an unknown bound type {#} m/\\b{ネ{#}}/",
'/\B{ネ}/' => "'ネ' is an unknown bound type {#} m/\\B{ネ{#}}/",
);
'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{#}/',
- '/\018/' => '\'\018\' resolved to \'\o{1}8\' {#} m/\018{#}/',
+ '/\08/' => 'Non-octal character \'8\' terminates \0 early. Resolved as "\0008" {#} m/\08{#}/',
+
+ '/\018/' => 'Non-octal character \'8\' terminates \0 early. Resolved as "\0018" {#} m/\018{#}/',
'/(?=a)*/' => '(?=a)* matches null string many times {#} m/(?=a)*{#}/',
'my $x = \'\m\'; qr/a$x/' => 'Unrecognized escape \m passed through {#} m/a\m{#}/',
'/\q/' => 'Unrecognized escape \q passed through {#} m/\q{#}/',
+ '/\q\p{Any}/' => 'Unrecognized escape \q passed through {#} m/\q{#}\p{Any}/',
# These two tests do not include the marker, because regcomp.c no
# longer knows where it goes by the time this warning is emitted.
'/[:alpha:]\x{100}/' => 'POSIX syntax [: :] belongs inside character classes {#} m/[:alpha:]{#}\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}/',
- '/[.z#g.]\x{100}/x' => "", # Runs into a comment
- '/[.z\#g.]\x{100}/x' => 'POSIX syntax [. .] belongs inside character classes (but this one isn\'t implemented) {#} m/[.z\#g.]{#}\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}/',
'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}/',
+ $only_strict_marker . 'Unescaped literal \']\' {#} m/[[ ^ : x d i g i t : ] ]{#}\x{100}/',
],
'/[foo:lower:]]\x{100}/' => 'Assuming NOT a POSIX class since it doesn\'t start with a \'[\' {#} m/[foo{#}:lower:]]\x{100}/',
'/[[;upper;]]\x{100}/' => [ 'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[[;{#}upper;]]\x{100}/',
'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[foo;{#}punct;]]\x{100}/',
'Assuming NOT a POSIX class since a semi-colon was found instead of a colon {#} m/[foo;punct;]{#}]\x{100}/',
],
+ '/[][[:alpha:]]/' => "", # [perl #127581]
+ '/[][[:alpha:]\\@\\\\^_?]/' => "", # [perl #131522]
+ '/(?[[:w:]])/' => "",
+ '/([.].*)[.]/' => "", # [perl #127582]
+ '/[.].*[.]/' => "", # [perl #127604]
+ '/abc/xix' => "",
+ '/(?xmsixp:abc)/' => "",
+ '/(?xmsixp)abc/' => "",
+ '/(?xxxx:abc)/' => "",
); # See comments before this for why '\x{100}' is generally needed
'Useless (?g) - use /g modifier {#} m/utf8 ネ (?og{#}c) ネ/',
'Useless (?c) - use /gc modifier {#} m/utf8 ネ (?ogc{#}) ネ/',
],
+ '/ネ[[:ネ:]]ネ/' => "",
+ '/ネ(?[[:ネ:]])ネ/' => "",
);
'/[a-\N{U+FF}]\x{100}/' => 'Ranges of ASCII printables should be some subset of "0-9", "A-Z", or "a-z" {#} m/[a-\N{U+FF}{#}]\x{100}/',
'/[\N{U+00}-\a]\x{100}/' => "",
'/[\a-\N{U+FF}]\x{100}/' => "",
- '/[\N{U+FF}-\x{100}]/' => 'Both or neither range ends should be Unicode {#} m/[\N{U+FF}-\x{100}{#}]/',
'/[\N{U+100}-\x{101}]/' => "",
"/[%-%]/" => "",
"/[:-\\x$colon_hex]\\x{100}/" => "\":-\\x$colon_hex\" is more clearly written simply as \":\" {#} m/[:-\\x$colon_hex\{#}]\\x{100}/",
"/[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]{#}/',
+ '/(?)/' => 'Empty (?) without any modifiers {#} m/(?){#}/', # [perl #132851]
);
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[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 @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)/' => "",
+ '/\s*{/' => "",
+ '/a{3,4}{/' => "",
);
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::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,
+ \@deprecated)
+ {
my $warning_type;
my $turn_off_warnings = "";
my $default_on;
$warning_type = 'regexp, deprecated';
$default_on = 1;
}
- else {
+ elsif ($ref == \@experimental_regex_sets) {
$warning_type = 'experimental::regex_sets';
$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 {
https://perl5.git.perl.org / perl5.git / blobdiff