App-TrimHistories

 view release on metacpan or  search on metacpan

README  view on Meta::CPAN

    See the included script trim-histories.

FUNCTIONS
  trim_histories
    Usage:

     trim_histories(%args) -> [status, msg, payload, meta]

    Keep only a certain number of sets of file histories, delete the rest.

    This script can be used to delete old backup or log files. The files
    must be named with timestamps, e.g. "mydb-2017-06-14.sql.gz". By
    default, it keeps only 7 daily, 4 weekly, and 6 monthly histories. The
    rest will be deleted.

    This function is not exported.

    This function supports dry-run operation.

    Arguments ('*' denotes required arguments):

    *   discard_old_histories => *bool*

    *   discard_young_histories => *bool*

    *   files* => *array[filename]*

        Each file name must be unique and contain date, e.g.
        `backup-2017-06-14.tar.gz`.

    *   sets => *array[str]* (default: ["daily",7,"weekly",4,"monthly",6])

        History sets to keep.

        Expressed as a list of (period, num-to-keep) pairs. Period can be
        number of seconds or either "hourly", "daily", "weekly", "monthly",
        "yearly". The default is:

         ['daily', 7, 'weekly', 4, 'monthly', 6]

lib/App/TrimHistories.pm  view on Meta::CPAN

use Log::ger;

our %SPEC;

$SPEC{trim_histories} = {
    v => 1.1,
    summary => 'Keep only a certain number of sets of file histories, '.
        'delete the rest',
    description => <<'_',

This script can be used to delete old backup or log files. The files must be
named with timestamps, e.g. `mydb-2017-06-14.sql.gz`. By default, it keeps only
7 daily, 4 weekly, and 6 monthly histories. The rest will be deleted.

_
    args => {
        files => {
            'x.name.is_plural' => 1,
            'x.name.singular' => 'file',
            schema => ['array*', of=>'filename*'],
            summary => 'Each file name must be unique and contain date, '.
                'e.g. `backup-2017-06-14.tar.gz`',
            req => 1,
            pos => 0,
            greedy => 1,
        },
        sets => {
            summary => 'History sets to keep',
            schema => ['array*', of=>'str*', min_len=>2, 'x.perl.coerce_rules' => ['From_str::comma_sep']],
            default => [daily => 7, weekly => 4, monthly => 6],
            description => <<'_',

lib/App/TrimHistories.pm  view on Meta::CPAN

=head2 trim_histories

Usage:

 trim_histories(%args) -> [status, msg, payload, meta]

Keep only a certain number of sets of file histories, delete the rest.

This script can be used to delete old backup or log files. The files must be
named with timestamps, e.g. C<mydb-2017-06-14.sql.gz>. By default, it keeps only
7 daily, 4 weekly, and 6 monthly histories. The rest will be deleted.

This function is not exported.

This function supports dry-run operation.


Arguments ('*' denotes required arguments):

=over 4

=item * B<discard_old_histories> => I<bool>

=item * B<discard_young_histories> => I<bool>

=item * B<files>* => I<array[filename]>

Each file name must be unique and contain date, e.g. `backup-2017-06-14.tar.gz`.

=item * B<sets> => I<array[str]> (default: ["daily",7,"weekly",4,"monthly",6])

History sets to keep.

Expressed as a list of (period, num-to-keep) pairs. Period can be number of
seconds or either C<hourly>, C<daily>, C<weekly>, C<monthly>, C<yearly>. The default
is:

 ['daily', 7, 'weekly', 4, 'monthly', 6]

script/trim-histories  view on Meta::CPAN

Like previous, but older and younger files are deleted:

 % trim-histories --discard-old --discard-young *

Only keep 5 daily, 2 weekly histories:

 % trim-histories --sets daily,5,weekly,2 *

=head1 DESCRIPTION

This script can be used to delete old backup or log files. The files must be
named with timestamps, e.g. C<mydb-2017-06-14.sql.gz>. By default, it keeps only
7 daily, 4 weekly, and 6 monthly histories. The rest will be deleted.

=head1 OPTIONS

C<*> marks required options.

=head2 Main options

=over

=item B<--discard-old-histories>

=item B<--discard-young-histories>

=item B<--file>=I<s@>*

Each file name must be unique and contain date, e.g. `backup-2017-06-14.tar.gz`.

Can be specified multiple times.

=item B<--files-json>=I<s>

Each file name must be unique and contain date, e.g. `backup-2017-06-14.tar.gz` (JSON-encoded).

See C<--file>.

=item B<--sets-json>=I<s>

History sets to keep (JSON-encoded).

See C<--sets>.

=item B<--sets>=I<s@>