view release on metacpan or  search on metacpan

lib/App/ElasticSearch/Utilities/Aggregations.pm  view on Meta::CPAN

    type      => 'bucket',
    composite => 1,
};


$Aggregations{significant_terms} = {
    params => sub { $_[0] =~ /^\d+$/ ? { size => $_[0] } : {} },
    type   => 'bucket',
};


$Aggregations{rare_terms} = {
    params => sub { $_[0] =~ /^\d+$/ ? { max_doc_count => $_[0] } : {} },
    type   => 'bucket',
};


$Aggregations{histogram} = {
    params => sub {
        return unless defined $_[0];
        return unless $_[0] > 0;
        return { interval => $_[0] };
    },
    type      => 'bucket',
    composite => 1,
};


$Aggregations{date_histogram} = {
    params    => sub { { calendar_interval => $_[0] || '1h' } },
    type      => 'bucket',
    composite => 1,
};


$Aggregations{geohash_grid} = {
    params    => sub { $_[0] =~ /^\d+$/ ? { precision => $_[0] } : {} },
    type      => 'bucket',
    composite => 1,
};


$Aggregations{missing} = { type => 'bucket' };


$Aggregations{avg} = { single_stat => 1, type => 'metric' };
$Aggregations{max} = { single_stat => 1, type => 'metric' };
$Aggregations{min} = { single_stat => 1, type => 'metric' };
$Aggregations{sum} = { single_stat => 1, type => 'metric' };


$Aggregations{cardinality} = { single_stat => 1, type => 'metric' };


$Aggregations{stats} = { type => 'metric' };


$Aggregations{extended_stats} = { type => 'metric' };


$Aggregations{percentiles} = {
    params => sub {
        my @pcts = $_[0] ? split /,/, $_[0] : qw(25 50 75 90);
        return { percents => \@pcts };
    },
};


$Aggregations{geo_centroid} = { type => 'metric' };




sub is_single_stat {
    my ($agg) = @_;
    return unless $agg;
    return unless exists $Aggregations{$agg};
    return unless exists $Aggregations{$agg}->{single_stat};
    return $Aggregations{$agg}->{single_stat};
}


sub expand_aggregate_string {
    my ($token) = @_;

    my %aggs = ();
    foreach my $def ( split /\+/, $token ) {
        my $alias = $def =~ s/^(\w+)=// ? $1 : undef;
        my @parts = split /:/, $def, 3;
        if( @parts == 1 ) {
            $alias ||= $def;
            $aggs{$alias} = { terms => { field => $def, size => 20 } };
            next;
        }
        my ($agg, $field);
        if( exists $Aggregations{$parts[0]} ) {
            $agg     = shift @parts;
            $field   = shift @parts;
        }
        else {
            $agg = 'terms';
            $field = shift @parts;
        }
        my $params  = {};
        my $paramStr = shift @parts;

        if( $paramStr && $paramStr =~ /\w+=/ ) {
            # split on commas using a positive lookahead for a "word="
            foreach my $token (split /,(?=\w+=)/, $paramStr) {
                my ($k,$v) = split /=/, $token, 2;
                next unless $k and $v;
                $params->{$k} = $v =~ /,/ ? [ split /,/, $v ] : $v;
            }
        }
        elsif( exists $Aggregations{$agg}->{params} ) {
            # Process parameters
            $params = $Aggregations{$agg}->{params}->($paramStr);
        }
        $alias ||= join ".", $agg eq 'terms' ? ($field) : ($agg, $field);
        $aggs{$alias} = { $agg => { field => $field, %{ $params || {} } } };
    }
    return \%aggs;
}

lib/App/ElasticSearch/Utilities/Aggregations.pm  view on Meta::CPAN

Results in

    {
        "sum.field_names": {
            "sum": {
                "field": "field_name"
            }
        }
    }

=item B<cardinality>

Computes the unique count of terms in a field.

    cardinality:field_name

Results in

    {
        "cardinality.field_names": {
            "cardinality": {
                "field": "field_name"
            }
        }
    }

=item B<stats>

Runs the stats aggregation that returns min, max, avg, sum, and count.

    stats:field_name

Results in

    {
        "stats.field_names": {
            "stats": {
                "field": "field_name"
            }
        }
    }

=item B<extended_stats>

Runs the stats aggregation that returns the same data as the C<sum> aggregation
plus variance, sum of squares, and standard deviation.

    extended_stats:field_name

Results in

    {
        "extended_stats.field_names": {
            "extended_stats": {
                "field": "field_name"
            }
        }
    }

=item B<percentiles>

Computes percentiles for the enclosing bucket. The positional parameter is
interpretted at the percents computed.  If ommitted, the percentiles computed
will be: 25, 50, 75, 90.

    percentiles:field_name:75,95,99

Results in

    {
        "percentiles.field_names": {
            "percentiles": {
                "field": "field_name",
                "percents": [ 75, 95, 99 ]
            }
        }
    }

=item B<geo_centroid>

Computes center of a group of geo points. No positional parameters supported.

    geo_centroid:field_name

Results in

    {
        "geo_centroid.field_names": {
            "geo_centroid": {
                "field": "field_name"
            }
        }
    }

=back

=for Pod::Coverage es_flatten_aggs

=head1 AUTHOR

Brad Lhotsky <brad@divisionbyzero.net>

=head1 COPYRIGHT AND LICENSE

This software is Copyright (c) 2026 by Brad Lhotsky.

This is free software, licensed under:

  The (three-clause) BSD License

=cut