DBM_Filter(3perl)

✨ NAME ✨

DBM_Filter — the module that filters DBM keys and values 💖 (This page is from the Perl Programmers Reference Guide, section 3perl, so you know it's legit.)

---

🎀 SYNOPSIS 🎀

use DBM_Filter ;
use SDBM_File; # or DB_File, GDBM_File, NDBM_File, or ODBM_File

$db = tie %hash, ...

$db->Filter_Push(Fetch => sub {...},
                 Store => sub {...});

$db->Filter_Push('my_filter1');
$db->Filter_Push('my_filter2', params...);

$db->Filter_Key_Push(...) ;
$db->Filter_Value_Push(...) ;

$db->Filter_Pop();
$db->Filtered();

package DBM_Filter::my_filter1;

sub Store { ... }
sub Fetch { ... }

1;

package DBM_Filter::my_filter2;

sub Filter
{
    my @opts = @_;
    ...
    return (
        sub Store { ... },
        sub Fetch { ... } );
}

1;

That whole vibe, in one block. 🫶

---

💜 DESCRIPTION 💜

So this module hooks you up with an interface that lets filters be applied to tied Hashes that are attached to DBM files. It's built on the DBM Filter hooks that are present in every *DB*_File module included with the standard Perl source distribution, from version 5.6.1 onwards. And it's not just those — the BerkeleyDB module, which is available on CPAN, supports the DBM Filter hooks too. If you want the full lore on the DBM Filter hooks themselves, go read perldbmfilter.

🦋 What is a DBM Filter? 🦋

A DBM Filter lets the keys and/or the values in a tied hash get modified by some user-defined code — right before it's written to the DBM file, and right after it's read back from the DBM file. So like, this lil snippet

$some_hash{"abc"} = 42;

could potentially trigger two filters: one for the writing of the key "abc", and another for writing the value 42. Same situation with this snippet

my ($key, $value) = each %some_hash

— that one sets off two filters too, one for the reading of the key and one for the reading of the value.

Like the existing DBM Filter functionality, this module arranges for the $_ variable to be populated with the key or value that a filter will check. That's lowkey why most DBM filters end up being super short.

✨ So what's new? ✨

The main enhancements over the standard DBM Filter hooks are:

• A cleaner interface. 💅
• The ability to easily apply multiple filters to a single DBM file. 🎀
• The ability to create "canned" filters, which let commonly used filters be packaged into a stand-alone module. 📦

---

🔥 METHODS 🔥

This module will arrange for the following methods to be available via the object returned from the tie call.

$db->Filter_Push() / $db->Filter_Key_Push() / $db->Filter_Value_Push()

Adds a filter to the filter stack for the database, $db. The three formats vary only in whether they apply to the DBM key, the DBM value, or both.

• Filter_Push — the filter is applied to both keys and values.
• Filter_Key_Push — the filter is applied to the key only.
• Filter_Value_Push — the filter is applied to the value only.

$db->Filter_Pop()

Removes the last filter that was applied to the DBM file associated with $db, if present.

$db->Filtered()

Returns TRUE if there are any filters applied to the DBM associated with $db. Otherwise returns FALSE.

---

🌈 Writing a Filter 🌈

Filters can be created in two main ways.

🧚 Immediate Filters 🧚

An immediate filter lets you specify the filter code to be used at the point where the filter is applied to a dbm. In this mode, the Filter_*_Push methods expect to receive exactly two parameters.

my $db = tie %hash, 'SDBM_File', ...
$db->Filter_Push( Store => sub { },
                  Fetch => sub { });

The code reference associated with "Store" will be called before any key/value is written to the database, and the code reference associated with "Fetch" will be called after any key/value is read from the database.

For example, here's a sample filter that adds a trailing NULL character to all strings before they're written to the DBM file, and removes the trailing NULL when they're read from the DBM file

my $db = tie %hash, 'SDBM_File', ...
$db->Filter_Push( Store => sub { $_ .= "\x00" ; },
                  Fetch => sub { s/\x00$// ;    });

Points to note:

1. Both the Store and Fetch filters manipulate $_.

🍓 Canned Filters 🍓

Immediate filters are great for one-off situations. But for more generic problems, it can be useful to package the filter up in its own module.

The usage for a canned filter is:

$db->Filter_Push("name", params)

where

• "name" — is the name of the module to load. If the string you specified doesn't contain the package separator characters "::", it's assumed to refer to the full module name "DBM_Filter::name". Which means the full names for the canned filters "null" and "utf8" included with this module are:

DBM_Filter::null
DBM_Filter::utf8

• params — any optional parameters that need to be sent to the filter. Peep the encode filter for an example of a module that uses parameters.

The module that implements the canned filter can take one of two forms. Here's a template for the first

package DBM_Filter::null ;

use strict;
use warnings;

sub Store
{
    # store code here
}

sub Fetch
{
    # fetch code here
}

1;

Notes:

1. The package name uses the "DBM_Filter::" prefix.
2. The module must have both a Store and a Fetch method. If only one is present, or neither are present, a fatal error will be thrown. 😭

The second form allows the filter to hold state information using a closure, thus:

package DBM_Filter::encoding ;

use strict;
use warnings;

sub Filter
{
    my @params = @_ ;

    ...
    return {
        Store   => sub { $_ = $encoding->encode($_) },
        Fetch   => sub { $_ = $encoding->decode($_) }
        } ;
}

1;

In this instance the "Store" and "Fetch" methods are encapsulated inside a "Filter" method. Iconic honestly.

---

👑 Filters Included 👑

A number of canned filters are provided with this module. They cover a bunch of the main areas where filters are needed when interfacing with DBM files. They also act as templates for your own filters.

The filters included are:

• utf8 — This module will ensure that all data written to the DBM will be encoded in UTF-8. This module needs the Encode module.
• encode — Allows you to choose which character encoding will be stored in the DBM file.
• compress — This filter will compress all data before it is written to the database and uncompress it on reading. This module needs Compress::Zlib.
• int32 — This module is used when interoperating with a C/C++ application that uses a C int as either the key and/or value in the DBM file.
• null — This module ensures that all data written to the DBM file is null terminated. Super useful when you have a perl script that needs to interoperate with a DBM file that a C program also uses. A fairly common issue is for the C application to include the terminating null in a string when it writes to the DBM file. This filter will ensure that all data written to the DBM file can be read by the C application.

---

⚠️ NOTES ⚠️

🫶 Maintain Round Trip Integrity 🫶

When writing a DBM filter it is very important to ensure that it's possible to retrieve all data that you have written when the DBM filter is in place. In practice, this means that whatever transformation is applied to the data in the Store method, the exact inverse operation should be applied in the Fetch method.

If you don't provide an exact inverse transformation, you'll find that code like this will not behave as you expect.

while (my ($k, $v) = each %hash)
{
    ...
}

Depending on the transformation, you'll find that one or more of the following will happen:

1. The loop will never terminate. 💀
2. Too few records will be retrieved.
3. Too many will be retrieved.
4. The loop will do the right thing for a while, but it will unexpectedly fail. 😭

🚫 Don't mix filtered & non-filtered data in the same database file. 🚫

This is just a restatement of the previous section. Unless you are completely certain you know what you're doing, avoid mixing filtered & non-filtered data.

---

⭐ EXAMPLE ⭐

Say you need to interoperate with a legacy C application that stores keys as C ints and the values as null terminated UTF-8 strings. Here's how you would set that up

my $db = tie %hash, 'SDBM_File', ...

$db->Filter_Key_Push('int32') ;

$db->Filter_Value_Push('utf8');
$db->Filter_Value_Push('null');

---

🌸 SEE ALSO 🌸

<DB_File>, GDBM_File, NDBM_File, ODBM_File, SDBM_File, perldbmfilter

---

💖 AUTHOR 💖

Paul Marquess <pmqs@cpan.org>

---

and that's the whole page, bestie 🫶 this one came straight outta perl v5.44.0, dated 2026-10-17, under the DBM_Filter(3perl) manual page. filter your keys, filter your values, stay hydrated 💅✨🎀