mirror/perl
1
0
Fork 0
mirror of https://github.com/Perl/perl5.git synced 2026-01-26 08:38:23 +00:00
Code Issues Packages Projects Releases Wiki Activity
perl/t/re/pat_re_eval.t
David Mitchell 729f4d269c regex: eval/cut: fix premature local undo and $^R 
GH #16197

The main purpose of this commit is to stop premature scope unwinding
within eval code in regexes. Aside from backtracking on failure, the
scopes of every eval, such as (?{ local $x }), are supposed to accumulate,
and are only unwound en-masse at the very end of the pattern match.
However, a combination of a sub-pattern call, such as (??{...}) or
(?&...), combined with a cut, (?>...), can trigger earlier savestack
popping.

The direct fix for this, as explained below, is to remove this single
line from the EVAL_postponed_A/B: case:

    REGCP_UNWIND(ST.cp); /* LEAVE in disguise */

However, that line is entwined with code which attempts to preserve the
final value of $^R during scope unwinding.  Since that code was kind of
working around the misplaced REGCP_UNWIND(), it needs ripping out and
re-implementing. This has to be done at the same time, so the bulk of
this commit is actually concerned with $^R, even though it isn't the
subject of this ticket. So this commit doesn't change the behaviour of
$^R, but just changes its implementation somewhat.

The $^R issue is that every /(?{...})/ causes the equivalent of

    local $^R = ...;

to be executed. During final exit, the savestack gets unwound, and all
those local's get undone, leaving $^R with the value it had before the
match started. But we promise that after the match,  $^R will hold the
value of the most recent (?{...}). The code which this commit rips out
restored that value in one way; the new code in this commit does it a
different way. Basically, almost the last thing S_regmatch() does is a
LEAVE_SCOPE(orig_savestack_ix); This commit makes it so that the current
value of $^R is copied just before the LEAVE_SCOPE(), and that value is
copied back to $^R just after the LEAVE_SCOPE().

For efficiency, we only do the copy if we've actually set $^R. A mechanism
is also needed to ensure that the temporary copy doesn't leak if we die
during the savestack unwind. This is achieved by holding a pointer to
the copy in the aux_eval struct, which gets processed even if we die.

Now back to the main purpose of this commit, the premature stack unwind
in the presence of a cut with a sub-pattern. This bug has been there
since these features were added. It is instructive to look at a somewhat
idealised overview of the S_regmatch() function from around 5.6.0 (with
some bug fixes from later releases added). This was while the function
was still recursive. It looks approximately like:

    pp_match(...) {
        I32 ix = PL_savestack_ix;
        ... do a match ...
        LEAVE_SCOPE(ix);
    }

    S_regmatch(...) {
        while (scan) {
            switch (OP(scan)) {
            case FOO:
                if (! there's a FOO)
                    return 0;
                I32 ix = PL_savestack_ix;
                if (regmatch(...)) /* recursively match rest of pattern */
                    return 1;
                LEAVE_SCOPE(ix);
                return 0;

            case END:
                if (doing a (??{...}) ) {
                    I32 ix = PL_savestack_ix;
                    if (regmatch(...) { /* recursively match rest of pattern */
                        LEAVE_SCOPE(ix);
                        return 1;
                    }
                    LEAVE_SCOPE(ix);
                    return 0;
                }
                return 1;

            case EVAL:
                ... run the code, then, if its a (?{...}) ...
                I32 ix = PL_savestack_ix;
                if (regmatch(...) {   /* recursively run subpattern */
                    LEAVE_SCOPE(ix)
                    return 1;
                }
                LEAVE_SCOPE(ix);
                return 0;
        }
    }

Here, the FOO: case represents all the various ops which recurse. In
general, they match the next item and then recurse to match the rest of
the pattern.

Note that they all do a LEAVE_SCOPE() only in the *failure* branch.

At the end of a successful match, there is potentially much recursion,
and much stuff on the savestack. When the END op is reached, the series
of 'return 1's causes all the recursion to unwind, while leaving the
savestack untouched. Finally, the caller - such as pp_match() - clears
the savestack. In more recent perls the recursion has been removed and
the final LEAVE_SCOPE() is done within S_regmatch() itself, but the
principle remains the same: no stack freeing is done *during* matching,
and instead there's a single big clean up at the end.

Once (??{...}) enters the picture,  that changes a bit. When the END op
associated with the '...' sub-pattern is reached, regmatch() is called
recursively to process any pattern after the (??{..}); then on success,
while working its way back through the nested regmatch() calls, both the
END and the EVAL code do a LEAVE_SCOPE() in the *success* branch.

This is anomalous, and those two LEAVE_SCOPE()'s are what this commit
removes (although in the current non-recursive regex engine, they are
shared by the same piece of code, so only one had to be removed). By
removing them, this regularises the behaviour of sub-patterns. I can't
think why those LEAVE_SCOPE()s were originally added, and assume it was a
thinko.

Normally it makes no difference whether the savestack is popped near the
end, interleaved with popping all recursive regmatch calls (or
equivalently on non-recursive engines popping the regmatch_state stack),
or whether the savestack is popped only after all the recursion is
exited from.

However, it makes a difference in the presence of a cut, (?>...). Here,
the final op of the sub-pattern '...' is SUCCEED, which rather than
recursing to match anything following the cut block, just returns. The
recursion pops back to the SUSPEND op which started the cut, which then
continues with the op loop as normal.

Thus when about to match the first op following a (?>...), the recursion
*within* the cut has been blown away as if it never happened, but the
accumulated savestack entries (e.g. from evals within the cut block) are
preserved and continue to accumulate.

Now, if there is a (??{...}) sub-pattern, or similarly a (?&FOO), within
the cut, then at the end of the cut, the recursion is unwound, which
includes the stacked EVAL and END recursions, which at this time call
LEAVE_SCOPE(), which frees part of the savestack, even though the
pattern match hasn't ended yet. That's the bug which this commit fixes.

The tests added to pat_re_eval.t check that the scope bug has been
fixed. The test added to svleak.t checks that the new $^R copying code
doesn't leak.
2025-12-17 11:17:31 +00:00

1499 lines
47 KiB
Perl
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

#!./perl
#
# This is a home for regular expression tests that don't fit into
# the format supported by re/regexp.t. If you want to add a test
# that does fit that format, add it to re/re_tests, not here.
use strict;
use warnings;
use Config;
use 5.010;
sub run_tests;
$| = 1;
BEGIN {
chdir 't' if -d 't';
require './test.pl';
set_up_inc('../lib');
require './charset_tools.pl';
}
our @global;
plan tests => 552; # Update this when adding/deleting tests.
run_tests() unless caller;
# test that runtime code without 'use re eval' is trapped
sub norun {
like($@, qr/Eval-group not allowed at runtime/, @_);
}
#
# Tests start here.
#
sub run_tests {
{
my $message = "Call code from qr //";
local $_ = 'var="foo"';
$a = qr/(?{++$b})/;
$b = 7;
ok(/$a$a/ && $b eq '9', $message);
my $c="$a";
ok(/$a$a/ && $b eq '11', $message);
undef $@;
eval {/$c/};
norun("$message norun 1");
{
eval {/$a$c$a/};
norun("$message norun 2");
use re "eval";
/$a$c$a/;
is($b, '14', $message);
}
our $lex_a = 43;
our $lex_b = 17;
our $lex_c = 27;
my $lex_res = ($lex_b =~ qr/$lex_b(?{ $lex_c = $lex_a++ })/);
is($lex_res, 1, $message);
is($lex_a, 44, $message);
is($lex_c, 43, $message);
undef $@;
my $d = '(?{1})';
my $match = eval { /$a$c$a$d/ };
ok($@ && $@ =~ /Eval-group not allowed/ && !$match, $message);
is($b, '14', $message);
$lex_a = 2;
$lex_a = 43;
$lex_b = 17;
$lex_c = 27;
$lex_res = ($lex_b =~ qr/17(?{ $lex_c = $lex_a++ })/);
is($lex_res, 1, $message);
is($lex_a, 44, $message);
is($lex_c, 43, $message);
}
{
our $a = bless qr /foo/ => 'Foo';
ok 'goodfood' =~ $a, "Reblessed qr // matches";
is($a, '(?^:foo)', "Reblessed qr // stringifies");
my $x = "\x{3fe}";
my $z = my $y = byte_utf8a_to_utf8n("\317\276"); # Byte representation
# of $x
$a = qr /$x/;
ok $x =~ $a, "UTF-8 interpolation in qr //";
ok "a$a" =~ $x, "Stringified qr // preserves UTF-8";
ok "a$x" =~ /^a$a\z/, "Interpolated qr // preserves UTF-8";
ok "a$x" =~ /^a(??{$a})\z/,
"Postponed interpolation of qr // preserves UTF-8";
is(length qr /##/x, 9, "## in qr // doesn't corrupt memory; Bug 17776");
{
ok "$x$x" =~ /^$x(??{$x})\z/,
"Postponed UTF-8 string in UTF-8 re matches UTF-8";
ok "$y$x" =~ /^$y(??{$x})\z/,
"Postponed UTF-8 string in non-UTF-8 re matches UTF-8";
ok "$y$x" !~ /^$y(??{$y})\z/,
"Postponed non-UTF-8 string in non-UTF-8 re doesn't match UTF-8";
ok "$x$x" !~ /^$x(??{$y})\z/,
"Postponed non-UTF-8 string in UTF-8 re doesn't match UTF-8";
ok "$y$y" =~ /^$y(??{$y})\z/,
"Postponed non-UTF-8 string in non-UTF-8 re matches non-UTF8";
ok "$x$y" =~ /^$x(??{$y})\z/,
"Postponed non-UTF-8 string in UTF-8 re matches non-UTF8";
$y = $z; # Reset $y after upgrade.
ok "$x$y" !~ /^$x(??{$x})\z/,
"Postponed UTF-8 string in UTF-8 re doesn't match non-UTF-8";
ok "$y$y" !~ /^$y(??{$x})\z/,
"Postponed UTF-8 string in non-UTF-8 re doesn't match non-UTF-8";
}
}
{
our $this_counter;
ok( "ABDE" =~ /(A(A|B(*ACCEPT)|C)+D)(E)(?{ $this_counter++ })/,
"ACCEPT/CURLYX/EVAL - pattern should match");
is( "$1-$2", "AB-B",
"Make sure that ACCEPT works in CURLYX by using EVAL");
}
{
ok( "AB"=~/(A)(?(*{ 1 })B|C)/, "(?(*{ ... })yes|no) works as expected");
ok( "AC"=~/(A)(?(*{ 0 })B|C)/, "(?(*{ ... })yes|no) works as expected");
}
{
# Test if $^N and $+ work in (*{ }) (optimistic eval)
our @ctl_n = ();
our @plus = ();
my $nested_tags = qr{
(?<nested_tags>
<
((\w)+)
(*{
push @ctl_n, (defined $^N ? $^N : "undef");
push @plus, (defined $+ ? $+ : "undef");
})
>
(?&nested_tags)*
</\s* \w+ \s*>
)
}x;
# note the results of this may change from perl to perl as different optimisations
# are added or enabled. It is testing that things *work*, not that they produce
# a specific output. The whole idea of optimistic eval is to have an eval that
# does not disable optimizations in the way a normal eval does.
my $c = 0;
for my $test (
# Test structure:
# [ Expected result, Regex, Expected value(s) of $^N, Expected value(s) of $+, "note" ]
[ 1, qr#^$nested_tags$#, "bla blubb <bla><blubb></blubb></bla>", "a b a" ],
[ 1, qr#^($nested_tags)$#, "bla blubb <bla><blubb></blubb></bla>", "a b a" ],
[ 1, qr#^(|)$nested_tags$#, "bla blubb <bla><blubb></blubb></bla>", "a b a" ],
[ 1, qr#^(?:|)$nested_tags$#, "bla blubb <bla><blubb></blubb></bla>", "a b a" ],
[ 1, qr#^<(bl|bla)>$nested_tags<(/\1)>$#, "blubb /bla", "b /bla" ],
) { #"#silence vim highlighting
$c++;
@ctl_n = ();
@plus = ();
my $match = (("<bla><blubb></blubb></bla>" =~ $test->[1]) ? 1 : 0);
push @ctl_n, (defined $^N ? $^N : "undef");
push @plus, (defined $+ ? $+ : "undef");
ok($test->[0] == $match, "(*{ ... }) match $c");
if ($test->[0] != $match) {
# unset @ctl_n and @plus
@ctl_n = @plus = ();
}
my $note = $test->[4] ? " - $test->[4]" : "";
is("@ctl_n", $test->[2], "(*{ ... }) ctl_n $c$note");
is("@plus", $test->[3], "(*{ ... }) plus $c$note");
}
}
{
# Test if $^N and $+ work in (?{})
our @ctl_n = ();
our @plus = ();
our $nested_tags;
$nested_tags = qr{
<
((\w)+)
(?{
push @ctl_n, (defined $^N ? $^N : "undef");
push @plus, (defined $+ ? $+ : "undef");
})
>
(??{$nested_tags})*
</\s* \w+ \s*>
}x;
my $c = 0;
for my $test (
# Test structure:
# [ Expected result, Regex, Expected value(s) of $^N, Expected value(s) of $+ ]
[ 1, qr#^$nested_tags$#, "bla blubb bla", "a b a" ],
[ 1, qr#^($nested_tags)$#, "bla blubb <bla><blubb></blubb></bla>", "a b a" ],
[ 1, qr#^(|)$nested_tags$#, "bla blubb bla", "a b a" ],
[ 1, qr#^(?:|)$nested_tags$#, "bla blubb bla", "a b a" ],
[ 1, qr#^<(bl|bla)>$nested_tags<(/\1)>$#, "blubb /bla", "b /bla" ],
[ 1, qr#(??{"(|)"})$nested_tags$#, "bla blubb bla", "a b a" ],
[ 1, qr#^(??{"(bla|)"})$nested_tags$#, "bla blubb bla", "a b a" ],
[ 1, qr#^(??{"(|)"})(??{$nested_tags})$#, "bla blubb undef", "a b undef" ],
[ 1, qr#^(??{"(?:|)"})$nested_tags$#, "bla blubb bla", "a b a" ],
[ 1, qr#^((??{"(?:bla|)"}))((??{$nested_tags}))$#, "bla blubb <bla><blubb></blubb></bla>", "a b <bla><blubb></blubb></bla>" ],
[ 1, qr#^((??{"(?!)?"}))((??{$nested_tags}))$#, "bla blubb <bla><blubb></blubb></bla>", "a b <bla><blubb></blubb></bla>" ],
[ 1, qr#^((??{"(?:|<(/?bla)>)"}))((??{$nested_tags}))\1$#, "bla blubb <bla><blubb></blubb></bla>", "a b <bla><blubb></blubb></bla>" ],
[ 0, qr#^((??{"(?!)"}))?((??{$nested_tags}))(?!)$#, # changed in perl 5.37.7
"bla blubb blub blu bl b bl b undef",
"a b b u l b l b undef" ],
) { #"#silence vim highlighting
$c++;
@ctl_n = ();
@plus = ();
my $match = (("<bla><blubb></blubb></bla>" =~ $test->[1]) ? 1 : 0);
push @ctl_n, (defined $^N ? $^N : "undef");
push @plus, (defined $+ ? $+ : "undef");
ok($test->[0] == $match, "match $c");
if ($test->[0] != $match) {
# unset @ctl_n and @plus
@ctl_n = @plus = ();
}
is("@ctl_n", $test->[2], "ctl_n $c");
is("@plus", $test->[3], "plus $c");
}
}
{
our $f;
local $f;
$f = sub {
defined $_[0] ? $_[0] : "undef";
};
like("123", qr/^(\d)(((??{1 + $^N})))+$/, 'Bug 56194');
our @ctl_n;
our @plus;
my $re = qr#(1)((??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1}))*(?{$^N})#;
my $re2 = qr#(1)((??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1}))*(?{$^N})(|a(b)c|def)(??{"$^R"})#;
my $re3 = qr#(1)((??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1})){2}(?{$^N})(|a(b)c|def)(??{"$^R"})#;
our $re5;
local $re5 = qr#(1)((??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1})){2}(?{$^N})#;
my $re6 = qr#(??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1})#;
my $re7 = qr#(??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1})#;
my $re8 = qr/(\d+)/;
my $c = 0;
for my $test (
# Test structure:
# [
# String to match
# Regex too match
# Expected values of $^N
# Expected values of $+
# Expected values of $1, $2, $3, $4 and $5
# ]
[
"1233",
qr#^(1)((??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1}))+(??{$^N})$#,
"1 2 3 3",
"1 2 3 3",
"\$1 = 1, \$2 = 3, \$3 = undef, \$4 = undef, \$5 = undef",
],
[
"1233",
qr#^(1)((??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1}))+(abc|def|)?(??{$+})$#,
"1 2 3 3",
"1 2 3 3",
"\$1 = 1, \$2 = 3, \$3 = undef, \$4 = undef, \$5 = undef",
],
[
"1233",
qr#^(1)((??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1}))+(|abc|def)?(??{$+})$#,
"1 2 3 3",
"1 2 3 3",
"\$1 = 1, \$2 = 3, \$3 = undef, \$4 = undef, \$5 = undef",
],
[
"1233",
qr#^(1)((??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1}))+(abc|def|)?(??{$^N})$#,
"1 2 3 3",
"1 2 3 3",
"\$1 = 1, \$2 = 3, \$3 = undef, \$4 = undef, \$5 = undef",
],
[
"1233",
qr#^(1)((??{ push @ctl_n, $f->($^N); push @plus, $f->($+); $^N + 1}))+(|abc|def)?(??{$^N})$#,
"1 2 3 3",
"1 2 3 3",
"\$1 = 1, \$2 = 3, \$3 = undef, \$4 = undef, \$5 = undef",
],
[
"123abc3",
qr#^($re)(|a(b)c|def)(??{$^R})$#,
"1 2 3 abc",
"1 2 3 b",
"\$1 = 123, \$2 = 1, \$3 = 3, \$4 = abc, \$5 = b",
],
[
"123abc3",
qr#^($re2)$#,
"1 2 3 123abc3",
"1 2 3 b",
"\$1 = 123abc3, \$2 = 1, \$3 = 3, \$4 = abc, \$5 = b",
],
[
"123abc3",
qr#^($re3)$#,
"1 2 123abc3",
"1 2 b",
"\$1 = 123abc3, \$2 = 1, \$3 = 3, \$4 = abc, \$5 = b",
],
[
"123abc3",
qr#^(??{$re5})(|abc|def)(??{"$^R"})$#,
"1 2 abc",
"1 2 abc",
"\$1 = abc, \$2 = undef, \$3 = undef, \$4 = undef, \$5 = undef",
],
[
"123abc3",
qr#^(??{$re5})(|a(b)c|def)(??{"$^R"})$#,
"1 2 abc",
"1 2 b",
"\$1 = abc, \$2 = b, \$3 = undef, \$4 = undef, \$5 = undef",
],
[
"1234",
qr#^((\d+)((??{push @ctl_n, $f->($^N); push @plus, $f->($+);$^N + 1}))((??{push @ctl_n, $f->($^N); push @plus, $f->($+);$^N + 1}))((??{push @ctl_n, $f->($^N); push @plus, $f->($+);$^N + 1})))$#,
"1234 123 12 1 2 3 1234",
"1234 123 12 1 2 3 4",
"\$1 = 1234, \$2 = 1, \$3 = 2, \$4 = 3, \$5 = 4",
],
[
"1234556",
qr#^(\d+)($re6)($re6)($re6)$re6(($re6)$re6)$#,
"1234556 123455 12345 1234 123 12 1 2 3 4 4 5 56",
"1234556 123455 12345 1234 123 12 1 2 3 4 4 5 5",
"\$1 = 1, \$2 = 2, \$3 = 3, \$4 = 4, \$5 = 56",
],
[
"12345562",
qr#^((??{$re8}))($re7)($re7)($re7)$re7($re7)($re7(\2))$#,
"12345562 1234556 123455 12345 1234 123 12 1 2 3 4 4 5 62",
"12345562 1234556 123455 12345 1234 123 12 1 2 3 4 4 5 2",
"\$1 = 1, \$2 = 2, \$3 = 3, \$4 = 4, \$5 = 5",
],
) {
$c++;
@ctl_n = ();
@plus = ();
undef $^R;
my $match = $test->[0] =~ $test->[1];
my $str = join(", ", '$1 = '.$f->($1), '$2 = '.$f->($2), '$3 = '.$f->($3), '$4 = '.$f->($4),'$5 = '.$f->($5));
push @ctl_n, $f->($^N);
push @plus, $f->($+);
ok($match, "match $c; Bug 56194");
if (not $match) {
# unset $str, @ctl_n and @plus
$str = "";
@ctl_n = @plus = ();
}
is("@ctl_n", $test->[2], "ctl_n $c; Bug 56194");
is("@plus", $test->[3], "plus $c; Bug 56194");
is($str, $test->[4], "str $c; Bug 56194");
}
{
@ctl_n = ();
@plus = ();
our $re4;
local $re4 = qr#(1)((??{push @ctl_n, $f->($^N); push @plus, $f->($+);$^N + 1})){2}(?{$^N})(|abc|def)(??{"$^R"})#;
undef $^R;
my $match = "123abc3" =~ m/^(??{$re4})$/;
my $str = join(", ", '$1 = '.$f->($1), '$2 = '.$f->($2), '$3 = '.$f->($3), '$4 = '.$f->($4),'$5 = '.$f->($5),'$^R = '.$f->($^R));
push @ctl_n, $f->($^N);
push @plus, $f->($+);
ok($match, 'Bug 56194');
if (not $match) {
# unset $str
@ctl_n = ();
@plus = ();
$str = "";
}
is("@ctl_n", "1 2 undef", 'Bug 56194');
is("@plus", "1 2 undef", 'Bug 56194');
is($str,
"\$1 = undef, \$2 = undef, \$3 = undef, \$4 = undef, \$5 = undef, \$^R = 3",
'Bug 56194 ($^R tweaked by 121070)');
}
{
undef $^R;
"abcd"=~/(?<Char>.)(?&Char)(?{ 42 })/;
is("$^R", 42, 'Bug 121070 - use of (?&Char) should not clobber $^R');
"abcd"=~/(?<Char>.)(?&Char)(?{ 42 })(?{ 43 })/;
is("$^R", 43, 'related to 121070 - use of (?&Char) should not clobber $^R');
}
}
{
# re evals within \U, \Q etc shouldn't be seen by the lexer
local our $a = "i";
local our $B = "J";
ok('(?{1})' =~ /^\Q(?{1})\E$/, '\Q(?{1})\E');
ok('(?{1})' =~ /^\Q(?{\E1\}\)$/, '\Q(?{\E1\}\)');
eval {/^\U(??{"$a\Ea"})$/ }; norun('^\U(??{"$a\Ea"})$ norun');
eval {/^\L(??{"$B\Ea"})$/ }; norun('^\L(??{"$B\Ea"})$ norun');
use re 'eval';
ok('Ia' =~ /^\U(??{"$a\Ea"})$/, '^\U(??{"$a\Ea"})$');
ok('ja' =~ /^\L(??{"$B\Ea"})$/, '^\L(??{"$B\Ea"})$');
}
{
# Comprehensive (hopefully) tests of closure behaviour:
# i.e. when do (?{}) blocks get (re)compiled, and what instances
# of lexical vars do they close over?
# if the pattern string gets utf8 upgraded while concatenating,
# make sure a literal code block is still detected (by still
# compiling in the absence of use re 'eval')
{
my $s1 = "\x{80}";
my $s2 = "\x{100}";
ok("\x{80}\x{100}" =~ /^$s1(?{1})$s2$/, "utf8 upgrade");
}
my ($cr1, $cr2, $cr3, $cr4);
for my $x (qw(a b c)) {
my $bc = ($x ne 'a');
my $c80 = chr(0x80);
# the most basic: literal code should be in same scope
# as the parent
ok("A$x" =~ /^A(??{$x})$/, "[$x] literal code");
ok("\x{100}$x" =~ /^\x{100}(??{$x})$/, "[$x] literal code UTF8");
# the "don't recompile if pattern unchanged" mechanism
# shouldn't apply to code blocks - recompile every time
# to pick up new instances of variables
my $code1 = 'B(??{$x})';
my $code1u = $c80 . "\x{100}" . '(??{$x})';
eval {/^A$code1$/};
norun("[$x] unvarying runtime code AA norun");
eval {/^A$code1u$/};
norun("[$x] unvarying runtime code AU norun");
eval {/^$c80\x{100}$code1$/};
norun("[$x] unvarying runtime code UA norun");
eval {/^$c80\x{101}$code1u$/};
norun("[$x] unvarying runtime code UU norun");
{
use re 'eval';
ok("AB$x" =~ /^A$code1$/, "[$x] unvarying runtime code AA");
ok("A$c80\x{100}$x" =~ /^A$code1u$/,
"[$x] unvarying runtime code AU");
ok("$c80\x{100}B$x" =~ /^$c80\x{100}$code1$/,
"[$x] unvarying runtime code UA");
ok("$c80\x{101}$c80\x{100}$x" =~ /^$c80\x{101}$code1u$/,
"[$x] unvarying runtime code UU");
}
# mixed literal and run-time code blocks
my $code2 = 'B(??{$x})';
my $code2u = $c80 . "\x{100}" . '(??{$x})';
eval {/^A(??{$x})-$code2$/};
norun("[$x] literal+runtime AA norun");
eval {/^A(??{$x})-$code2u$/};
norun("[$x] literal+runtime AU norun");
eval {/^$c80\x{100}(??{$x})-$code2$/};
norun("[$x] literal+runtime UA norun");
eval {/^$c80\x{101}(??{$x})-$code2u$/};
norun("[$x] literal+runtime UU norun");
{
use re 'eval';
ok("A$x-B$x" =~ /^A(??{$x})-$code2$/,
"[$x] literal+runtime AA");
ok("A$x-$c80\x{100}$x" =~ /^A(??{$x})-$code2u$/,
"[$x] literal+runtime AU");
ok("$c80\x{100}$x-B$x" =~ /^$c80\x{100}(??{$x})-$code2$/,
"[$x] literal+runtime UA");
ok("$c80\x{101}$x-$c80\x{100}$x"
=~ /^$c80\x{101}(??{$x})-$code2u$/,
"[$x] literal+runtime UU");
}
# literal qr code only created once, naked
$cr1 //= qr/^A(??{$x})$/;
ok("Aa" =~ $cr1, "[$x] literal qr once naked");
# literal qr code only created once, embedded with text
$cr2 //= qr/B(??{$x})$/;
ok("ABa" =~ /^A$cr2/, "[$x] literal qr once embedded text");
# literal qr code only created once, embedded with text + lit code
$cr3 //= qr/C(??{$x})$/;
ok("A$x-BCa" =~ /^A(??{$x})-B$cr3/,
"[$x] literal qr once embedded text + lit code");
# literal qr code only created once, embedded with text + run code
$cr4 //= qr/C(??{$x})$/;
my $code3 = 'A(??{$x})';
eval {/^$code3-B$cr4/};
norun("[$x] literal qr once embedded text + run code norun");
{
use re 'eval';
ok("A$x-BCa" =~ /^$code3-B$cr4/,
"[$x] literal qr once embedded text + run code");
}
# literal qr code, naked
my $r1 = qr/^A(??{$x})$/;
ok("A$x" =~ $r1, "[$x] literal qr naked");
# literal qr code, embedded with text
my $r2 = qr/B(??{$x})$/;
ok("AB$x" =~ /^A$r2/, "[$x] literal qr embedded text");
# literal qr code, embedded with text + lit code
my $r3 = qr/C(??{$x})$/;
ok("A$x-BC$x" =~ /^A(??{$x})-B$r3/,
"[$x] literal qr embedded text + lit code");
# literal qr code, embedded with text + run code
my $r4 = qr/C(??{$x})$/;
my $code4 = '(??{$x})';
eval {/^A$code4-B$r4/};
norun("[$x] literal qr embedded text + run code");
{
use re 'eval';
ok("A$x-BC$x" =~ /^A$code4-B$r4/,
"[$x] literal qr embedded text + run code");
}
# nested qr in different scopes
my $code5 = '(??{$x})';
my $r5 = qr/C(??{$x})/;
my $r6;
eval {qr/$code5-C(??{$x})/}; norun("r6 norun");
{
use re 'eval';
$r6 = qr/$code5-C(??{$x})/;
}
my @rr5;
my @rr6;
for my $y (qw(d e f)) {
my $rr5 = qr/^A(??{"$x$y"})-$r5/;
push @rr5, $rr5;
ok("A$x$y-C$x" =~ $rr5,
"[$x-$y] literal qr + r5");
my $rr6 = qr/^A(??{"$x$y"})-$r6/;
push @rr6, $rr6;
ok("A$x$y-$x-C$x" =~ $rr6,
"[$x-$y] literal qr + r6");
}
for my $i (0,1,2) {
my $y = 'Y';
my $yy = (qw(d e f))[$i];
my $rr5 = $rr5[$i];
ok("A$x$yy-C$x" =~ $rr5, "[$x-$yy] literal qr + r5, outside");
ok("A$x$yy-C$x-D$x" =~ /$rr5-D(??{$x})$/,
"[$x-$yy] literal qr + r5 + lit, outside");
my $rr6 = $rr6[$i];
push @rr6, $rr6;
ok("A$x$yy-$x-C$x" =~ $rr6,
"[$x-$yy] literal qr + r6, outside");
ok("A$x$yy-$x-C$x-D$x" =~ /$rr6-D(??{$x})/,
"[$x-$yy] literal qr + r6 +lit, outside");
}
}
# recursive subs should get lexical from the correct pad depth
sub recurse {
my ($n) = @_;
return if $n > 2;
ok("A$n" =~ /^A(??{$n})$/, "recurse($n)");
recurse($n+1);
}
recurse(0);
# for qr// containing run-time elements but with a compile-time
# code block, make sure the run-time bits are executed in the same
# pad they were compiled in
{
my $a = 'a'; # ensure outer and inner pads don't align
my $b = 'b';
my $c = 'c';
my $d = 'd';
my $r = qr/^$b(??{$c})$d$/;
ok("bcd" =~ $r, "qr with run-time elements and code block");
}
# check that cascaded embedded regexes all see their own lexical
# environment
{
my ($r1, $r2, $r3, $r4);
my ($x1, $x2, $x3, $x4) = (5,6,7,8);
{ my $x1 = 1; $r1 = qr/A(??{$x1})/; }
{ my $x2 = 2; $r2 = qr/$r1(??{$x2})/; }
{ my $x3 = 3; $r3 = qr/$r2(??{$x3})/; }
{ my $x4 = 4; $r4 = qr/$r3(??{$x4})/; }
ok("A1234" =~ /^$r4$/, "cascaded qr");
}
# and again, but in a loop, with no external references
# being maintained to the qr's
{
my $r = 'A';
for my $x (1..4) {
$r = qr/$r(??{$x})/;
}
my $x = 5;
ok("A1234" =~ /^$r$/, "cascaded qr loop");
}
# and again, but compiling the qrs in an eval so there
# aren't even refs to the qrs from any ops
{
my $r = 'A';
for my $x (1..4) {
$r = eval q[ qr/$r(??{$x})/; ];
}
my $x = 5;
ok("A1234" =~ /^$r$/, "cascaded qr loop");
}
# have qrs with either literal code blocks or only embedded
# code blocks, but not both
{
my ($r1, $r2, $r3, $r4);
my ($x1, $x3) = (7,8);
{ my $x1 = 1; $r1 = qr/A(??{$x1})/; }
{ $r2 = qr/${r1}2/; }
{ my $x3 = 3; $r3 = qr/$r2(??{$x3})/; }
{ $r4 = qr/${r3}4/; }
ok("A1234" =~ /^$r4$/, "cascaded qr mix 1");
ok("A12345" =~ /^${r4}5$/, "cascaded qr mix 2");
ok("A1234" =~ qr/^$r4$/ , "cascaded qr mix 3");
ok("A12345" =~ qr/^${r4}5$/, "cascaded qr mix 4");
}
# and make sure things are freed at the right time
{
sub Foo99::DESTROY { $Foo99::d++ }
$Foo99::d = 0;
my $r1;
{
my $x = bless [1], 'Foo99';
$r1 = eval 'qr/(??{$x->[0]})/';
}
my $r2 = eval 'qr/a$r1/';
my $x = 2;
ok(eval '"a1" =~ qr/^$r2$/', "match while in scope");
# make sure PL_reg_curpm isn't holding on to anything
"a" =~ /a(?{1})/;
is($Foo99::d, 0, "before scope exit");
}
::is($Foo99::d, 1, "after scope exit");
# forward declared subs should Do The Right Thing with any anon CVs
# within them (i.e. pad_fixup_inner_anons() should work)
sub forward;
sub forward {
my $x = "a";
my $A = "A";
ok("Aa" =~ qr/^A(??{$x})$/, "forward qr compiletime");
ok("Aa" =~ qr/^$A(??{$x})$/, "forward qr runtime");
}
forward;
}
# test that run-time embedded code, when re-fed into toker,
# does all the right escapes
{
my $enc;
$enc = eval 'use Encode; find_encoding("ascii")' unless $::IS_EBCDIC;
my $x = 0;
my $y = 'bad';
# note that most of the strings below are single-quoted, and the
# things within them, like '$y', *aren't* intended to interpolate
my $s1 =
'a\\$y(?# (??{BEGIN{$x=1} "X1"})b(?# \Ux2\E)c\'d\\\\e\\\\Uf\\\\E';
ok(q{a$ybc'd\e\Uf\E} =~ /^$s1$/, "reparse");
is($x, 0, "reparse no BEGIN");
my $s2 = 'g\\$y# (??{{BEGIN{$x=2} "X3"}) \Ux3\E' . "\nh";
ok(q{a$ybc'd\\e\\Uf\\Eg$yh} =~ /^$s1$s2$/x, "reparse /x");
is($x, 0, "reparse /x no BEGIN");
my $b = '\\';
my $q = '\'';
# non-ascii in string as "<0xNNN>"
sub esc_str {
my $s = shift;
$s =~ s{(.)}{
my $c = ord($1);
(utf8::native_to_unicode($c)< 32
|| utf8::native_to_unicode($c) > 127)
? sprintf("<0x%x>", $c) : $1;
}ge;
$s;
}
sub fmt { sprintf "hairy backslashes %s [%s] =~ /^%s/",
$_[0], esc_str($_[1]), esc_str($_[2]);
}
for my $u (
[ '', '', 'blank ' ],
[ "\x{100}", '\x{100}', 'single' ],
[ "\x{100}", "\x{100}", 'double' ])
{
for my $pair (
[ "$b", "$b$b" ],
[ "$q", "$q" ],
[ "$b$q", "$b$b$b$q" ],
[ "$b$b$q", "$b$b$b$b$q" ],
[ "$b$b$b$q", "$b$b$b$b$b$b$q" ],
[ "$b$b$b$b$q","$b$b$b$b$b$b$b$b$q" ],
) {
my ($s, $r) = @$pair;
$s = "9$s";
my $ss = "$u->[0]$s";
my $c = '9' . $r;
my $cc = "$u->[1]$c";
ok($ss =~ /^$cc/, fmt("plain $u->[2]", $ss, $cc));
no strict;
$nine = $nine = "bad";
$ss = "$u->[0]\t${q}\x41${b}x42$s" if $::IS_ASCII;
$ss = "$u->[0]\t${q}\xC1${b}xC2$s" if $::IS_EBCDIC;
for my $use_qr ('', 'qr') {
$c = qq[(??{my \$z='{';]
. (($::IS_ASCII)
? qq[$use_qr"$b${b}t$b$q$b${b}x41$b$b$b${b}x42"]
: qq[$use_qr"$b${b}t$b$q$b${b}xC1$b$b$b${b}xC2"])
. qq[. \$nine})];
# (??{ qr/str/ }) goes through one less interpolation
# stage than (??{ qq/str/ })
$c =~ s{\\\\}{\\}g if ($use_qr eq 'qr');
$c .= $r;
$cc = "$u->[1]$c";
my $nine = 9;
eval {/^$cc/}; norun(fmt("code norun $u->[2]", $ss, $cc));
{
use re 'eval';
ok($ss =~ /^$cc/, fmt("code $u->[2]", $ss, $cc));
}
}
}
}
my $code1u = "(??{qw(\x{100})})";
eval {/^$code1u$/}; norun("reparse embedded unicode norun");
{
use re 'eval';
ok("\x{100}" =~ /^$code1u$/, "reparse embedded unicode");
}
}
# a non-pattern literal won't get code blocks parsed at compile time;
# but they must get parsed later on if 'use re eval' is in scope
# also check that unbalanced {}'s are parsed ok
{
eval q["a{" =~ '^(??{"a{"})$'];
norun("non-pattern literal code norun");
eval {/^${\'(??{"a{"})'}$/};
norun("runtime code with unbalanced {} norun");
use re 'eval';
ok("a{" =~ '^a(??{"{"})$', "non-pattern literal code");
ok("a{" =~ /^a${\'(??{"{"})'}$/, "runtime code with unbalanced {}");
}
# make sure warnings come from the right place
{
use warnings;
my ($s, $t, $w);
local $SIG{__WARN__} = sub { $w .= "@_" };
$w = ''; $s = 's';
my $r = qr/(?{$t=$s+1})/;
"a" =~ /a$r/;
like($w, qr/pat_re_eval/, "warning main file");
# do it in an eval to get predictable line numbers
eval q[
$r = qr/(?{$t=$s+1})/;
];
$w = ''; $s = 's';
"a" =~ /a$r/;
like($w, qr/ at \(eval \d+\) line 3/, "warning eval A");
$w = ''; $s = 's';
eval q[
use re 'eval';
my $c = '(?{$t=$s+1})';
"a" =~ /a$c/;
1;
];
like($w, qr/ at \(eval \d+\) line 1/, "warning eval B");
}
# jumbo test for:
# * recursion;
# * mixing all the different types of blocks (literal, qr/literal/,
# runtime);
# * backtracking (the Z+ alternation ensures CURLYX and full
# scope popping on backtracking)
{
sub recurse2 {
my ($depth)= @_;
return unless $depth;
my $s1 = '3-LMN';
my $r1 = qr/(??{"$s1-$depth"})/;
my $s2 = '4-PQR';
my $c1 = '(??{"$s2-$depth"})';
use re 'eval';
ok( "<12345-ABC-$depth-123-LMN-$depth-1234-PQR-$depth>"
. "<12345-ABC-$depth-123-LMN-$depth-1234-PQR-$depth>"
=~
/^<(\d|Z+)+(??{"45-ABC-$depth-"})(\d|Z+)+$r1-\d+$c1>
<(\d|Z+)+(??{"45-ABC-$depth-"})(\d|Z+)+$r1-\d+$c1>$/x,
"recurse2($depth)");
recurse2($depth-1);
}
recurse2(5);
}
# nested (??{}) called from various levels of a recursive function
{
sub recurse3 {
my ($n) = @_;
return if $n > 3;
ok("A$n" =~ m{^A(??{ "0123" =~ /((??{$n}))/; $1 })$},
"recurse3($n)");
ok("A$n" !~ m{^A(??{ "0123" =~ /((??{$n}))/; "X" })$},
"recurse3($n) nomatch");
recurse3($n+1);
}
recurse3(0);
}
# nested (??{}) being invoked recursively via a function
{
my $s = '';
our $recurse4;
my @alpha = qw(A B C D E);
$recurse4 = sub {
my ($n) = @_;
$s .= "(n=$n:";
if ($n < 4) {
my $m = ("$alpha[$n]" . substr("0123", 0, $n+1)) =~
m{^([A-Z])
(??{
$s .= "1=$1:";
"$n-0123" =~ m{^(\d)-(((??{$recurse4->($n+1)})))};
$s .= "i1=$1:<=[$2]";
$3; # NB - not stringified
})
$
}x;
$s .= "1a=$1:";
$s .= $m ? 'M' : '!M';
}
my $ret = '.*?' . ($n-1);
$s .= "<=[$ret])";
return $ret;
};
$recurse4->(0);
my $exp = '(n=0:1=A:(n=1:1=B:(n=2:1=C:(n=3:1=D:(n=4:<=[.*?3])'
. 'i1=3:<=[0123]1a=D:M<=[.*?2])i1=2:<=[012]1a=C:M<=[.*?1])'
. 'i1=1:<=[01]1a=B:M<=[.*?0])i1=0:<=[0]1a=A:M<=[.*?-1])';
is($s, $exp, 'recurse4');
}
# single (??{}) being invoked recursively via a function
{
my $s = '';
our $recurse5;
my @alpha = qw(A B C D E);
$recurse5 = sub {
my ($n) = @_;
$s .= "(n=$n:";
if ($n < 4) {
my $m = ("$alpha[$n]" . substr("0123", 0, $n+1)) =~
m{^([A-Z])
((??{
$s .= "1=$1:";
$recurse5->($n+1);
}))
$
}x;
$s .= "1a=$1:2=$2:";
$s .= $m ? 'M' : '!M';
}
my $ret = '.*?' . ($n-1);
$s .= "<=[$ret])";
return $ret;
};
$recurse5->(0);
my $exp = '(n=0:1=A:(n=1:1=B:(n=2:1=C:(n=3:1=D:(n=4:<=[.*?3])'
. '1a=D:2=0123:M<=[.*?2])1a=C:2=012:M<=[.*?1])'
. '1a=B:2=01:M<=[.*?0])1a=A:2=0:M<=[.*?-1])';
is($s, $exp, 'recurse5');
}
# make sure that errors during compiling run-time code get trapped
{
use re 'eval';
my $code = '(?{$x=})';
eval { "a" =~ /^a$code/ };
like($@, qr/syntax error at \(eval \d+\) line \d+/, 'syntax error');
$code = '(?{BEGIN{die})';
eval { "a" =~ /^a$code/ };
like($@,
qr/BEGIN failed--compilation aborted at \(eval \d+\) line \d+/,
'syntax error');
use utf8;
$code = '(?{Ｆｏｏ::$bar})';
eval { "a" =~ /^a$code/ };
like($@, qr/Bad name after Ｆｏｏ:: at \(eval \d+\) line \d+/, 'UTF8 sytax error');
}
# make sure that 'use re eval' is propagated into compiling the
# pattern returned by (??{})
{
use re 'eval';
my $pat = 'B(??{1})C';
my $A = 'A';
# compile-time outer code-block
ok("AB1CD" =~ /^A(??{$pat})D$/, "re eval propagated compile-time");
# run-time outer code-block
ok("AB1CD" =~ /^$A(??{$pat})D$/, "re eval propagated run-time");
}
# returning a ref to something that had set magic but wasn't
# PERL_MAGIC_qr triggered a false positive assertion failure
# The test is not so much concerned with it not matching,
# as with not failing the assertion
{
ok("a" !~ /^(a)(??{ \$1 })/, '(??{ ref })');
}
# make sure the uninit warning from returning an undef var
# sees the right var
{
my ($u1, $u2);
my $warn = '';
local $SIG{__WARN__} = sub { $warn .= $_[0] };
$u1 =~ /(??{$u2})/ or die;
like($warn, qr/value \$u1 in pattern match.*\n.*value at/, 'uninit');
}
# test that code blocks are called in scalar context
{
my @a = (0);
ok("" =~ /^(?{@a})$/, '(?{}) in scalar context');
is($^R, 1, '(?{}) in scalar context: $^R');
ok("1" =~ /^(??{@a})$/, '(??{}) in scalar context');
ok("foo" =~ /^(?(?{@a})foo|bar)$/, '(?(?{})|) in scalar context');
}
# BEGIN in compiled blocks shouldn't mess with $1 et al
{
use re 'eval';
my $code1 = '(B)(??{ BEGIN { "X" =~ /X/ } $1})(C)';
ok("ABBCA" =~ /^(.)(??{$code1})\1$/, '(?{}) BEGIN and $1');
my $code2 = '(B)(??{ BEGIN { "X" =~ /X/ } $1 =~ /(.)/ ? $1 : ""})(C)';
ok("ABBCA" =~ /^(.)(??{$code2})\1$/, '(?{}) BEGIN and $1 mark 2');
}
# check that the optimiser is applied to code blocks: see if aelem has
# been converted to aelemfast
{
my $out;
for my $prog (
'/(?{$a[0]})/',
'q() =~ qr/(?{$a[0]})/',
'use re q(eval); q() =~ q{(?{$a[0]})}',
'use re q(eval); $c = q{(?{$a[0]})}; /$c/',
'use re q(eval); $c = q{(?{$a[0]})}; /(?{1;})$c/',
) {
$out = runperl(switches => ["-Dt"], prog => $prog, stderr => 1);
like($out, qr/aelemfast|Recompile perl with -DDEBUGGING/,
"optimise: '$prog'");
}
}
# [perl #115080]
# Ensure that ?pat? matches exactly once, even when the run-time
# pattern changes, and even when the presence of run-time (?{}) affects
# how and when patterns are recompiled
{
my $m;
$m = '';
for (qw(a a a)) {
$m .= $_ if m?$_?;
}
is($m, 'a', '?pat? with a,a,a');
$m = '';
for (qw(a b c)) {
$m .= $_ if m?$_?;
}
is($m, 'a', '?pat? with a,b,c');
use re 'eval';
$m = '';
for (qw(a a a)) {
my $e = qq[(??{"$_"})];
$m .= $_ if m?$e?;
}
is($m, 'a', '?pat? with (??{a,a,a})');
$m = '';
for (qw(a b c)) {
my $e = qq[(??{"$_"})];
$m .= $_ if m?$e?;
}
is($m, 'a', '?pat? with (??{a,b,c})');
}
{
# this code won't actually fail, but it used to fail valgrind,
# so its here just to make sure valgrind doesn't fail again
# While examining the ops of the secret anon sub wrapped around
# the qr//, the pad of the sub was in scope, so cSVOPo_sv
# got the const from the wrong pad. By having lots of $s's
# (aka gvsv(*s), this forces the targs of the consts which have
# been moved to the pad, to have high indices.
sub {
local our $s = "abc";
my $qr = qr/^(?{1})$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s$s/;
}->();
pass("cSVOPo_sv");
}
# [perl #115004]
# code blocks in qr objects that are interpolated in arrays need
# handling the same as if they were interpolated from scalar vars
# (before this code would need 'use re "eval"')
{
use Tie::Array;
local @global;
my @array;
my @refs = (0, \@array, 2);
my @tied;
tie @tied, 'Tie::StdArray';
{
my $bb = 'B';
my $dd = 'D';
@array = ('A', qr/(??{$bb})/, 'C', qr/(??{$dd})/, 'E');
@tied = @array;
@global = @array;
}
my $bb = 'X';
my $dd = 'Y';
ok("A B C D E=" =~ /@array/, 'bare interpolated array match');
ok("A B C D E=" =~ qr/@array/, 'qr bare interpolated array match');
ok("A B C D E=" =~ /@global/, 'bare interpolated global array match');
ok("A B C D E=" =~ qr/@global/,
'qr bare interpolated global array match');
ok("A B C D E=" =~ /@{$refs[1]}/, 'bare interpolated ref array match');
ok("A B C D E=" =~ qr/@{$refs[1]}/,
'qr bare interpolated ref array match');
ok("A B C D E=" =~ /@tied/, 'bare interpolated tied array match');
ok("A B C D E=" =~ qr/@tied/, 'qr bare interpolated tied array match');
ok("aA B C D E=" =~ /^a@array=$/, 'interpolated array match');
ok("aA B C D E=" =~ qr/^a@array=$/, 'qr interpolated array match');
ok("aA B C D E=" =~ /^a@global=$/, 'interpolated global array match');
ok("aA B C D E=" =~ qr/^a@global=$/,
'qr interpolated global array match');
ok("aA B C D E=" =~ /^a@{$refs[1]}=$/, 'interpolated ref array match');
ok("aA B C D E=" =~ qr/^a@{$refs[1]}=$/,
'qr interpolated ref array match');
ok("aA B C D E=" =~ /^a@tied=$/, 'interpolated tied array match');
ok("aA B C D E=" =~ qr/^a@tied=$/, 'qr interpolated tied array match');
{
local $" = '-';
ok("aA-B-C-D-E=" =~ /^a@{array}=$/,
'interpolated array match with local sep');
ok("aA-B-C-D-E=" =~ qr/^a@{array}=$/,
'qr interpolated array match with local sep');
ok("aA-B-C-D-E=" =~ /^a@{global}=$/,
'interpolated global array match with local sep');
ok("aA-B-C-D-E=" =~ qr/^a@{global}=$/,
'qr interpolated global array match with local sep');
ok("aA-B-C-D-E=" =~ /^a@{tied}=$/,
'interpolated tied array match with local sep');
ok("aA-B-C-D-E=" =~ qr/^a@{tied}=$/,
'qr interpolated tied array match with local sep');
}
# but don't handle the array ourselves in the presence of \Q etc
@array = ('A', '(?{})');
@global = @array;
@tied = @array;
ok("aA (?{})=" =~ /^a\Q@{array}\E=$/,
'interpolated array match with \Q');
ok("aA (?{})=" =~ qr/^a\Q@{array}\E=$/,
'qr interpolated array match with \Q');
ok("aA (?{})=" =~ /^a\Q@{global}\E=$/,
'interpolated global array match with \Q');
ok("aA (?{})=" =~ qr/^a\Q@{global}\E=$/,
'qr interpolated global array match with \Q');
ok("aA (?{})=" =~ /^a\Q@{$refs[1]}\E=$/,
'interpolated ref array match with \Q');
ok("aA (?{})=" =~ qr/^a\Q@{$refs[1]}\E=$/,
'qr interpolated ref array match with \Q');
ok("aA (?{})=" =~ /^a\Q@{tied}\E=$/,
'interpolated tied array match with \Q');
ok("aA (?{})=" =~ qr/^a\Q@{tied}\E=$/,
'qr interpolated tied array match with \Q');
# and check it works with an empty array
@array = ();
@global = ();
@tied = ();
ok("a=" =~ /^a@array=$/, 'empty array match');
ok("a=" =~ qr/^a@array=$/, 'qr empty array match');
ok("a=" =~ /^a@global=$/, 'empty global array match');
ok("a=" =~ qr/^a@global=$/, 'qr empty global array match');
ok("a=" =~ /^a@tied=$/, 'empty tied array match');
ok("a=" =~ qr/^a@tied=$/, 'qr empty tied array match');
ok("a=" =~ /^a\Q@{array}\E=$/, 'empty array match with \Q');
ok("a=" =~ /^a\Q@{array}\E=$/, 'empty array match with \Q');
ok("a=" =~ qr/^a\Q@{global}\E=$/,
'qr empty global array match with \Q');
ok("a=" =~ /^a\Q@{tied}\E=$/, 'empty tied array match with \Q');
ok("a=" =~ qr/^a\Q@{tied}\E=$/, 'qr empty tied array match with \Q');
# NB: these below are empty patterns, so they happen to use the
# successful match from the line above
ok("a=" =~ /@array/, 'empty array pattern');
ok("a=" =~ qr/@array/, 'qr empty array pattern');
ok("a=" =~ /@global/, 'empty global array pattern');
ok("a=" =~ qr/@global/, 'qr empty global array pattern');
ok("a=" =~ /@tied/, 'empty tied pattern');
ok("a=" =~ qr/@tied/, 'qr empty tied pattern');
ok("a=" =~ /\Q@array\E/, 'empty array pattern with \Q');
ok("a=" =~ qr/\Q@array\E/, 'qr empty array pattern with \Q');
ok("a=" =~ /\Q@global\E/, 'empty global array pattern with \Q');
ok("a=" =~ qr/\Q@global\E/, 'qr empty global array pattern with \Q');
ok("a=" =~ /\Q@tied\E/, 'empty tied pattern with \Q');
ok("a=" =~ qr/\Q@tied\E/, 'qr empty tied pattern with \Q');
ok("a=" =~ //, 'completely empty pattern');
ok("a=" =~ qr//, 'qr completely empty pattern');
}
{
{ package o; use overload '""'=>sub { "abc" } }
my $x = bless [],"o";
my $y = \$x;
(my $y_addr = "$y") =~ y/()//d; # REF(0x7fcb9c02) -> REF0x7fcb9c02
# $y_addr =~ $y should be true, as should $y_addr =~ /(??{$y})/
"abc$y_addr" =~ /(??{$x})(??{$y})/;
is "$&", "abc$y_addr",
'(??{$x}) does not leak cached qr to (??{\$x}) (match)';
is scalar "abcabc" =~ /(??{$x})(??{$y})/, "",
'(??{$x}) does not leak cached qr to (??{\$x}) (no match)';
}
{
sub ReEvalTieTest::TIESCALAR {bless[], "ReEvalTieTest"}
sub ReEvalTieTest::STORE{}
sub ReEvalTieTest::FETCH { "$1" }
tie my $t, "ReEvalTieTest";
$t = bless [], "o";
"aab" =~ /(a)((??{"b" =~ m|(.)|; $t}))/;
is "[$1 $2]", "[a b]",
'(??{$tied_former_overload}) sees the right $1 in FETCH';
}
{
my @matchsticks;
my $ref = bless \my $o, "o";
my $foo = sub { push @matchsticks, scalar "abc" =~ /(??{$ref})/ };
&$foo;
bless \$o;
() = "$ref"; # flush AMAGIC flag on main
&$foo;
is "@matchsticks", "1 ", 'qr magic is not cached on refs';
}
{
my ($foo, $bar) = ("foo"x1000, "bar"x1000);
"$foo$bar" =~ /(??{".*"})/;
is "$&", "foo"x1000 . "bar"x1000,
'padtmp swiping does not affect "$a$b" =~ /(??{})/'
}
{
# [perl #129140]
# this used to cause a double-free of the code_block struct
# when re-running the compilation after spotting utf8.
# This test doesn't catch it, but might panic, or fail under
# valgrind etc
my $s = '';
/$s(?{})\x{100}/ for '', '';
pass "RT #129140";
}
# RT #130650 code blocks could get double-freed during a pattern
# compilation croak
{
# this used to panic or give ASAN errors
eval 'qr/(?{})\6/';
like $@, qr/Reference to nonexistent group/, "RT #130650";
}
# RT #129881
# on exit from a pattern with multiple code blocks from different
# CVs, PL_comppad wasn't being restored correctly
sub {
# give first few pad slots known values
my ($x1, $x2, $x3, $x4, $x5) = 101..105;
# these vars are in a separate pad
my $r = qr/((?{my ($y1, $y2) = 201..202; 1;})A){2}X/;
# the first alt fails, causing a switch to this anon
# sub's pad
"AAA" =~ /$r|(?{my ($z1, $z2) = 301..302; 1;})A/;
is $x1, 101, "RT #129881: x1";
is $x2, 102, "RT #129881: x2";
is $x3, 103, "RT #129881: x3";
}->();
# RT #126697
# savestack wasn't always being unwound on EVAL failure
{
local our $i = 0;
my $max = 0;
'ABC' =~ m{
\A
(?:
(?: AB | A | BC )
(?{
local $i = $i + 1;
$max = $i if $max < $i;
})
)*
\z
}x;
is $max, 2, "RT #126697";
}
# RT #132772
#
# Ensure that optimisation of OP_CONST into OP_MULTICONCAT doesn't
# leave any freed ops in the execution path. This is associated
# with rpeep() being called before optimize_optree(), which causes
# gv/rv2sv to be prematurely optimised into gvsv, confusing
# S_maybe_multiconcat when it tries to reorganise a concat subtree
# into a multiconcat list
{
my $a = "a";
local $b = "b"; # not lexical, so optimised to OP_GVSV
local $_ = "abc";
ok /^a(??{ $b."c" })$/, "RT #132772 - compile time";
ok /^$a(??{ $b."c" })$/, "RT #132772 - run time";
my $qr = qr/^a(??{ $b."c" })$/;
ok /$qr/, "RT #132772 - compile time qr//";
$qr = qr/(??{ $b."c" })$/;
ok /^a$qr$/, "RT #132772 - compile time qr// compound";
$qr = qr/$a(??{ $b."c" })$/;
ok /^$qr$/, "RT #132772 - run time qr//";
}
# RT #133687
# mixing compile-time (?(?{code})) with run-time code blocks
# was failing, because the second pass through the parser
# (which compiles the runtime code blocks) was failing to adequately
# mask the compile-time code blocks to shield them from a second
# compile: /X(?{...})Y/ was being correctly masked as /X________Y/
# but /X(?(?{...}))Y/ was being incorrectly masked as
# /X(?________)Y/
{
use re 'eval';
my $runtime_re = '(??{ "A"; })';
ok "ABC" =~ /^ $runtime_re (?(?{ 1; })BC) $/x, 'RT #133687 yes';
ok "ABC" =~ /^ $runtime_re (?(?{ 0; })xy|BC) $/x, 'RT #133687 yes|no';
}
# RT #134208
# when the string being matched was an SvTEMP and the re_eval died,
# the SV's magic was being restored after the SV was freed.
# Give ASan something to play with.
{
my $a;
no warnings 'uninitialized';
eval { "$a $1" =~ /(?{ die })/ };
pass("SvTEMP 1");
eval { sub { " " }->() =~ /(?{ die })/ };
pass("SvTEMP 2");
}
# GH #19680 "panic: restartop in perl_run"
# The eval block embedded within the (?{}) - but with no more code
# following it - causes the next op after the OP_LEAVETRY to be NULL
# (not even an OP_LEAVE). This confused the exception-catching and
# rethrowing code: it was incorrectly rethrowing the exception rather
# than just stopping at that point.
ok("test" =~ m{^ (?{eval {die "boo!"}}) test $}x, "GH #19680");
# GH #19390 Segmentation fault with use re 'eval'
# Similar to GH #19680 above, but exiting the eval via a syntax error
# rather than throwing an exception
ok("" =~ m{^ (?{eval q{$x=}})}x, "GH #19390");
# GH #22869 "Perl crash with recursive sub and regex with code eval".
#
# A recursive call to a match op with a run-time pattern and which
# contained a code block, led to to the temporary rex stored in the
# OP_MATCH and PL_reg_curpm ops getting prematurely freed when updated
# within the inner match's OP_MATCH op.
{
my @got;
my $f = sub {
my ($s, $re) = @_;
$s =~ $re;
push @got, ',', $1, $2, ']';
};
my $pat;
$pat = qr{^
(.)
(?{
push @got, '[', $1, $2;
$f->('XY', $pat) if $1 eq 'A';
push @got, ',', $1, $2;
})
(.)
(?{
push @got, ',', $1, $2;
})
$
}x;
$f->('AB',$pat);
my $got = join '', map defined ? $_ : '-', @got;
is($got, "[A-[X-,X-,XY,XY],A-,AB,AB]", "GH22869");
}
# GH #16197
# A subpattern, i.e. (??{...}) or (?&...), when combined with
# a cut, (?>...), caused the savestack to be prematurely unwound,
# resulting in localisations within (?{...}) being undone before
# the end of the match.
{
my ($match, @vals);
# The first six permutations do no backtracking, but are: with or
# without a cut; and match a digit via either a simple \d,
# or via a sub-pattern - (??{}) or (?&).
# The next six permutations add some backtracking.
#
# Previously the combination of a cut and a subpattern returned
# "101 101 101 101 101".
for my $bt ('', '[a-z]*') { # trigger a backtrack?
for my $cut ('', '?>') {
for my $subpat (q{\d}, q{(??{ '\d' })}, q{(?&DIGIT)} ) {
my $desc = "bt=$bt cut=$cut subp=$subpat";
@vals = ();
local our $x99 = 100;
use re 'eval';
$match =
"a1b2c3d4e5" =~
/^ (
[a-z]
($cut
$subpat
(?{ local $x99 = $x99 + 1; push @vals, $x99 })
)
$bt
){5}
(?(DEFINE) (?<DIGIT> \d ) )
/x;
ok($match, "GH 16197: match; $desc");
is("@vals", "101 102 103 104 105",
"GH 16197: local vals; $desc");
}
}
}
}
} # End of sub run_tests
1;
Reference in New Issue View Git Blame Copy Permalink