Security Advisories (28)
CVE-2018-18312 (2018-12-05)

Perl before 5.26.3 and 5.28.0 before 5.28.1 has a buffer overflow via a crafted regular expression that triggers invalid write operations.

CVE-2016-1238 (2016-08-02)

(1) cpan/Archive-Tar/bin/ptar, (2) cpan/Archive-Tar/bin/ptardiff, (3) cpan/Archive-Tar/bin/ptargrep, (4) cpan/CPAN/scripts/cpan, (5) cpan/Digest-SHA/shasum, (6) cpan/Encode/bin/enc2xs, (7) cpan/Encode/bin/encguess, (8) cpan/Encode/bin/piconv, (9) cpan/Encode/bin/ucmlint, (10) cpan/Encode/bin/unidump, (11) cpan/ExtUtils-MakeMaker/bin/instmodsh, (12) cpan/IO-Compress/bin/zipdetails, (13) cpan/JSON-PP/bin/json_pp, (14) cpan/Test-Harness/bin/prove, (15) dist/ExtUtils-ParseXS/lib/ExtUtils/xsubpp, (16) dist/Module-CoreList/corelist, (17) ext/Pod-Html/bin/pod2html, (18) utils/c2ph.PL, (19) utils/h2ph.PL, (20) utils/h2xs.PL, (21) utils/libnetcfg.PL, (22) utils/perlbug.PL, (23) utils/perldoc.PL, (24) utils/perlivp.PL, and (25) utils/splain.PL in Perl 5.x before 5.22.3-RC2 and 5.24 before 5.24.1-RC2 do not properly remove . (period) characters from the end of the includes directory array, which might allow local users to gain privileges via a Trojan horse module under the current working directory.

CVE-2023-47039 (2023-10-30)

Perl for Windows relies on the system path environment variable to find the shell (cmd.exe). When running an executable which uses Windows Perl interpreter, Perl attempts to find and execute cmd.exe within the operating system. However, due to path search order issues, Perl initially looks for cmd.exe in the current working directory. An attacker with limited privileges can exploit this behavior by placing cmd.exe in locations with weak permissions, such as C:\ProgramData. By doing so, when an administrator attempts to use this executable from these compromised locations, arbitrary code can be executed.

CVE-2023-47100

In Perl before 5.38.2, S_parse_uniprop_string in regcomp.c can write to unallocated space because a property name associated with a \p{...} regular expression construct is mishandled. The earliest affected version is 5.30.0.

CVE-2026-13221 (2026-07-13)

Perl versions through 5.43.9 produce silently incorrect regular expression matches when an alternation of more than 65535 fixed string branches is compiled into a trie in Perl_study_chunk. When such branches are combined into a trie, the delta between the first branch and the shared tail is stored in a 16-bit field. A branch count above 65535 overflows the field, and the trie's match decision table is truncated with no warning or error. A pattern of this shape produces false positive matches (matching strings it should not) and false negative matches (failing to match strings it should). When such a pattern gates an access or filtering decision, the result is wrong.

CVE-2026-15534 (2026-08-09)

Perl versions through 5.45.1 have out-of-bounds heap reads and writes during regular expression matching via an undersized superlinear cache in S_regmatch. The regex engine's superlinear cache holds one bit per subject position for each participating WHILEM node, so the bit count is the subject length plus one times the number of nodes. Nothing checks that product for positive overflow of the signed 32-bit count: a 286331153 byte subject matched against a pattern with 15 participating nodes stores the count as 14, leaving a two byte cache. The cache is then indexed from the real match position and node number, so reads go past the end of the allocation, and on failure CACHEsayNO sets a bit past it. A caller that matches an attacker controlled subject of this size against a pattern of this shape can crash the process or corrupt heap memory.

CVE-2026-4176 (2026-03-29)

Perl versions from 5.9.4 before 5.40.4-RC1, from 5.41.0 before 5.42.2-RC1, from 5.43.0 before 5.43.9 contain a vulnerable version of Compress::Raw::Zlib. Compress::Raw::Zlib is included in the Perl package as a dual-life core module, and is vulnerable to CVE-2026-3381 due to a vendored version of zlib which has several vulnerabilities, including CVE-2026-27171. The bundled Compress::Raw::Zlib was updated to version 2.221 in Perl blead commit c75ae9cc164205e1b6d6dbd57bd2c65c8593fe94.

CVE-2020-12723 (2020-06-05)

regcomp.c in Perl before 5.30.3 allows a buffer overflow via a crafted regular expression because of recursive S_study_chunk calls.

CVE-2015-8853 (2016-05-25)

The (1) S_reghop3, (2) S_reghop4, and (3) S_reghopmaybe3 functions in regexec.c in Perl before 5.24.0 allow context-dependent attackers to cause a denial of service (infinite loop) via crafted utf-8 data, as demonstrated by "a\x80."

CVE-2009-3626 (2009-10-29)

Perl 5.10.1 allows context-dependent attackers to cause a denial of service (application crash) via a UTF-8 character with a large, invalid codepoint, which is not properly handled during a regular-expression match.

CVE-2016-2381 (2016-04-08)

Perl might allow context-dependent attackers to bypass the taint protection mechanism in a child process via duplicate environment variables in envp.

CVE-2013-7422 (2015-08-16)

Integer underflow in regcomp.c in Perl before 5.20, as used in Apple OS X before 10.10.5 and other products, allows context-dependent attackers to execute arbitrary code or cause a denial of service (application crash) via a long digit string associated with an invalid backreference within a regular expression.

CVE-2026-57432 (2026-07-13)

Perl versions through 5.43.10 have an integer overflow in S_measure_struct leading to an out-of-bounds heap read in pack and unpack. S_measure_struct adds each item's size times its repeat count to a running total with no overflow check, so a large repeat count in a pack or unpack template wraps the signed SSize_t total negative. The @, X, and x position codes then guard their moves with a signed length comparison that passes when the length is negative, advancing the buffer pointer out of bounds. A template derived from untrusted input can read heap memory past the buffer and return it to the caller.

CVE-2026-19487 (2026-08-13)

Perl versions from 5.9.4 before 5.41.9 produce incorrect regular expression match results when a stale failure flag ends the Aho-Corasick prescan early in S_find_byclass. The prescan walks the subject for positions where the full pattern could match, and the engine tries it from the leftmost one recorded. A failing transition sets the failed flag, and a later successful transition does not clear it, so the prescan reads the stale flag as a failure and stops before it can record a candidate that starts earlier. It takes a subject where one candidate is recorded and a later character then forces a fallback through a fail link that succeeds. Example: "ABCDE" =~ m/ABCF|BCDE|C/; # matches C at offset 2, not BCDE "ABCDE" =~ m/ABCF|BCDE|C(G)/; # no match, BCDE missed An alternation like this can miss input it should match, or match it on the wrong branch, so an access or filtering decision made from the result can be wrong.

CVE-2026-8376 (2026-05-25)

Perl versions through 5.43.10 have a heap buffer overflow when compiling regular expressions with a repeated fixed string on 32-bit builds. Perl_study_chunk in regcomp_study.c checked the size of the joined substring buffer in characters rather than bytes. For a quantified fixed substring with a large minimum count, the byte length mincount * l could overflow SSize_t, producing an undersized SvGROW allocation; the subsequent copy writes past the end of the buffer. A caller that compiles an attacker-controlled regular expression on a 32-bit perl build triggers a heap buffer overflow at compile time.

CVE-2015-8608 (2017-02-07)

The VDir::MapPathA and VDir::MapPathW functions in Perl 5.22 allow remote attackers to cause a denial of service (out-of-bounds read) and possibly execute arbitrary code via a crafted (1) drive letter or (2) pInName argument.

CVE-2011-2728 (2012-12-21)

The bsd_glob function in the File::Glob module for Perl before 5.14.2 allows context-dependent attackers to cause a denial of service (crash) via a glob expression with the GLOB_ALTDIRFUNC flag, which triggers an uninitialized pointer dereference.

CVE-2020-10878 (2020-06-05)

Perl before 5.30.3 has an integer overflow related to mishandling of a "PL_regkind[OP(n)] == NOTHING" situation. A crafted regular expression could lead to malformed bytecode with a possibility of instruction injection.

CVE-2020-10543 (2020-06-05)

Perl before 5.30.3 on 32-bit platforms allows a heap-based buffer overflow because nested regular expression quantifiers have an integer overflow.

CVE-2018-6913 (2018-04-17)

Heap-based buffer overflow in the pack function in Perl before 5.26.2 allows context-dependent attackers to execute arbitrary code via a large item count.

CVE-2018-18314 (2018-12-07)

Perl before 5.26.3 has a buffer overflow via a crafted regular expression that triggers invalid write operations.

CVE-2018-18313 (2018-12-07)

Perl before 5.26.3 has a buffer over-read via a crafted regular expression that triggers disclosure of sensitive information from process memory.

CVE-2018-18311 (2018-12-07)

Perl before 5.26.3 and 5.28.x before 5.28.1 has a buffer overflow via a crafted regular expression that triggers invalid write operations.

CVE-2013-1667 (2013-03-14)

The rehash mechanism in Perl 5.8.2 through 5.16.x allows context-dependent attackers to cause a denial of service (memory consumption and crash) via a crafted hash key.

CVE-2011-0761 (2011-05-13)

Perl 5.10.x allows context-dependent attackers to cause a denial of service (NULL pointer dereference and application crash) by leveraging an ability to inject arguments into a (1) getpeername, (2) readdir, (3) closedir, (4) getsockname, (5) rewinddir, (6) tell, or (7) telldir function call.

CVE-2010-4777 (2014-02-10)

The Perl_reg_numbered_buff_fetch function in Perl 5.10.0, 5.12.0, 5.14.0, and other versions, when running with debugging enabled, allows context-dependent attackers to cause a denial of service (assertion failure and application exit) via crafted input that is not properly handled when using certain regular expressions, as demonstrated by causing SpamAssassin and OCSInventory to crash.

CVE-2012-5195 (2012-12-18)

Heap-based buffer overflow in the Perl_repeatcpy function in util.c in Perl 5.12.x before 5.12.5, 5.14.x before 5.14.3, and 5.15.x before 15.15.5 allows context-dependent attackers to cause a denial of service (memory consumption and crash) or possibly execute arbitrary code via the 'x' string repeat operator.

CVE-2011-1487 (2011-04-11)

The (1) lc, (2) lcfirst, (3) uc, and (4) ucfirst functions in Perl 5.10.x, 5.11.x, and 5.12.x through 5.12.3, and 5.13.x through 5.13.11, do not apply the taint attribute to the return value upon processing tainted input, which might allow context-dependent attackers to bypass the taint protection mechanism via a crafted string.

NAME

autodie::hints - Provide hints about user subroutines to autodie

SYNOPSIS

package Your::Module;

our %DOES = ( 'autodie::hints::provider' => 1 );

sub AUTODIE_HINTS {
    return {
        foo => { scalar => HINTS, list => SOME_HINTS },
        bar => { scalar => HINTS, list => MORE_HINTS },
    }
}

# Later, in your main program...

use Your::Module qw(foo bar);
use autodie      qw(:default foo bar);

foo();         # succeeds or dies based on scalar hints

# Alternatively, hints can be set on subroutines we've
# imported.

use autodie::hints;
use Some::Module qw(think_positive);

BEGIN {
    autodie::hints->set_hints_for(
        \&think_positive,
        {
            fail => sub { $_[0] <= 0 }
        }
    )
}
use autodie qw(think_positive);

think_positive(...);    # Returns positive or dies.

DESCRIPTION

Introduction

The autodie pragma is very smart when it comes to working with Perl's built-in functions. The behaviour for these functions are fixed, and autodie knows exactly how they try to signal failure.

But what about user-defined subroutines from modules? If you use autodie on a user-defined subroutine then it assumes the following behaviour to demonstrate failure:

  • A false value, in scalar context

  • An empty list, in list context

  • A list containing a single undef, in list context

All other return values (including the list of the single zero, and the list containing a single empty string) are considered successful. However, real-world code isn't always that easy. Perhaps the code you're working with returns a string containing the word "FAIL" upon failure, or a two element list containing (undef, "human error message"). To make autodie work with these sorts of subroutines, we have the hinting interface.

The hinting interface allows hints to be provided to autodie on how it should detect failure from user-defined subroutines. While these can be provided by the end-user of autodie, they are ideally written into the module itself, or into a helper module or sub-class of autodie itself.

What are hints?

A hint is a subroutine or value that is checked against the return value of an autodying subroutine. If the match returns true, autodie considers the subroutine to have failed.

If the hint provided is a subroutine, then autodie will pass the complete return value to that subroutine. If the hint is any other value, then autodie will smart-match against the value provided. In Perl 5.8.x there is no smart-match operator, and as such only subroutine hints are supported in these versions.

Hints can be provided for both scalar and list contexts. Note that an autodying subroutine will never see a void context, as autodie always needs to capture the return value for examination. Autodying subroutines called in void context act as if they're called in a scalar context, but their return value is discarded after it has been checked.

Example hints

Hints may consist of scalars, array references, regular expressions and subroutine references. You can specify different hints for how failure should be identified in scalar and list contexts.

These examples apply for use in the AUTODIE_HINTS subroutine and when calling autodie::hints-set_hints_for()>.

The most common context-specific hints are:

# Scalar failures always return undef:
    {  scalar => undef  }

# Scalar failures return any false value [default expectation]:
    {  scalar => sub { ! $_[0] }  }

# Scalar failures always return zero explicitly:
    {  scalar => '0'  }

# List failures always return an empty list:
    {  list => []  }

# List failures return () or (undef) [default expectation]:
    {  list => sub { ! @_ || @_ == 1 && !defined $_[0] }  }

# List failures return () or a single false value:
    {  list => sub { ! @_ || @_ == 1 && !$_[0] }  }

# List failures return (undef, "some string")
    {  list => sub { @_ == 2 && !defined $_[0] }  }

# Unsuccessful foo() returns 'FAIL' or '_FAIL' in scalar context,
#                    returns (-1) in list context...
autodie::hints->set_hints_for(
    \&foo,
    {
        scalar => qr/^ _? FAIL $/xms,
        list   => [-1],
    }
);

# Unsuccessful foo() returns 0 in all contexts...
autodie::hints->set_hints_for(
    \&foo,
    {
        scalar => 0,
        list   => [0],
    }
);

This "in all contexts" construction is very common, and can be abbreviated, using the 'fail' key. This sets both the scalar and list hints to the same value:

        # Unsuccessful foo() returns 0 in all contexts...
        autodie::hints->set_hints_for(
            \&foo,
            {
                fail => sub { @_ == 1 and defined $_[0] and $_[0] == 0 }
            }
	);

        # Unsuccessful think_positive() returns negative number on failure...
        autodie::hints->set_hints_for(
            \&think_positive,
            {
                fail => sub { $_[0] < 0 }
            }
	);

        # Unsuccessful my_system() returns non-zero on failure...
        autodie::hints->set_hints_for(
            \&my_system,
            {
                fail => sub { $_[0] != 0 }
            }
	);

Manually setting hints from within your program

If you are using a module which returns something special on failure, then you can manually create hints for each of the desired subroutines. Once the hints are specified, they are available for all files and modules loaded thereafter, thus you can move this work into a module and it will still work.

	use Some::Module qw(foo bar);
	use autodie::hints;

	autodie::hints->set_hints_for(
		\&foo,
		{
			scalar => SCALAR_HINT,
			list   => LIST_HINT,
		}
	);
	autodie::hints->set_hints_for(
		\&bar,
                { fail => SOME_HINT, }
	);

It is possible to pass either a subroutine reference (recommended) or a fully qualified subroutine name as the first argument. This means you can set hints on modules that might get loaded:

use autodie::hints;
autodie::hints->set_hints_for(
	'Some::Module:bar', { fail => SCALAR_HINT, }
);

This technique is most useful when you have a project that uses a lot of third-party modules. You can define all your possible hints in one-place. This can even be in a sub-class of autodie. For example:

package my::autodie;

use parent qw(autodie);
use autodie::hints;

autodie::hints->set_hints_for(...);

1;

You can now use my::autodie, which will work just like the standard autodie, but is now aware of any hints that you've set.

Adding hints to your module

autodie provides a passive interface to allow you to declare hints for your module. These hints will be found and used by autodie if it is loaded, but otherwise have no effect (or dependencies) without autodie. To set these, your module needs to declare that it does the autodie::hints::provider role. This can be done by writing your own DOES method, using a system such as Class::DOES to handle the heavy-lifting for you, or declaring a %DOES package variable with a autodie::hints::provider key and a corresponding true value.

Note that checking for a %DOES hash is an autodie-only short-cut. Other modules do not use this mechanism for checking roles, although you can use the Class::DOES module from the CPAN to allow it.

In addition, you must define a AUTODIE_HINTS subroutine that returns a hash-reference containing the hints for your subroutines:

        package Your::Module;

        # We can use the Class::DOES from the CPAN to declare adherence
        # to a role.

        use Class::DOES 'autodie::hints::provider' => 1;

        # Alternatively, we can declare the role in %DOES.  Note that
        # this is an autodie specific optimisation, although Class::DOES
        # can be used to promote this to a true role declaration.

        our %DOES = ( 'autodie::hints::provider' => 1 );

        # Finally, we must define the hints themselves.

	sub AUTODIE_HINTS {
	    return {
	        foo => { scalar => HINTS, list => SOME_HINTS },
	        bar => { scalar => HINTS, list => MORE_HINTS },
	        baz => { fail => HINTS },
	    }
	}

This allows your code to set hints without relying on autodie and autodie::hints being loaded, or even installed. In this way your code can do the right thing when autodie is installed, but does not need to depend upon it to function.

Insisting on hints

When a user-defined subroutine is wrapped by autodie, it will use hints if they are available, and otherwise reverts to the default behaviour described in the introduction of this document. This can be problematic if we expect a hint to exist, but (for whatever reason) it has not been loaded.

We can ask autodie to insist that a hint be used by prefixing an exclamation mark to the start of the subroutine name. A lone exclamation mark indicates that all subroutines after it must have hints declared.

	# foo() and bar() must have their hints defined
	use autodie qw( !foo !bar baz );

	# Everything must have hints (recommended).
	use autodie qw( ! foo bar baz );

	# bar() and baz() must have their hints defined
	use autodie qw( foo ! bar baz );

        # Enable autodie for all of Perl's supported built-ins,
        # as well as for foo(), bar() and baz().  Everything must
        # have hints.
        use autodie qw( ! :all foo bar baz );

If hints are not available for the specified subroutines, this will cause a compile-time error. Insisting on hints for Perl's built-in functions (eg, open and close) is always successful.

Insisting on hints is strongly recommended.

Diagnostics

Attempts to set_hints_for unidentifiable subroutine

You've called autodie::hints->set_hints_for() using a subroutine reference, but that reference could not be resolved back to a subroutine name. It may be an anonymous subroutine (which can't be made autodying), or may lack a name for other reasons.

If you receive this error with a subroutine that has a real name, then you may have found a bug in autodie. See "BUGS" in autodie for how to report this.

fail hints cannot be provided with either scalar or list hints for %s

When defining hints, you can either supply both list and scalar keywords, or you can provide a single fail keyword. You can't mix and match them.

%s hint missing for %s

You've provided either a scalar hint without supplying a list hint, or vice-versa. You must supply both scalar and list hints, or a single fail hint.

ACKNOWLEDGEMENTS

  • Dr Damian Conway for suggesting the hinting interface and providing the example usage.

  • Jacinta Richardson for translating much of my ideas into this documentation.

AUTHOR

Copyright 2009, Paul Fenwick <pjf@perltraining.com.au>

LICENSE

This module is free software. You may distribute it under the same terms as Perl itself.

SEE ALSO

autodie, Class::DOES