Question

I have 2 subs in my perl code that are all but identical. I’m curious as to how SO would refactor them to make one sub.

The only real difference in them is a regex, and a query via prepared statement. The prepared statements take different parameters as well.

Thoughts?

sub showcaseViewsSubData {
    my ($api_call, $stat_section, $idsite, $prev_date, $last_of_month, $params, $subtable) = @_;

    return unless ($subtable);

    my %sub_params = %{ clone ($params) };
    $sub_params{'idSubtable'} = $subtable->{'idsubdatatable'};

    # $data contains views for each primary showcase page
    my $data = &fetchStatsData($api_call, \%sub_params);

    foreach my $visit_group (@$data) {

        # ignore product pages
        next if ($visit_group->{'url'} && $visit_group->{'url'} =~ /\/products?\//);

        # if ($visit_group->{'url'} && $visit_group->{'url'} =~ /inthenews|pressreleases|downloads/) {
        if ($visit_group->{'idsubdatatable'}) {
            &showcaseViewsSubData($api_call, $stat_section, $idsite, $prev_date, $last_of_month, $params, $visit_group);
            next;
        }

        my $division_name;

        if ($visit_group->{'url'}) {
            my $showcase_id = $visit_group->{'url'};
            $showcase_id =~ s/^.*?\/(\d+)\/.*$/$1/;

            $division_by_showcase_id_sth->execute($showcase_id);
            ($division_name) = $division_by_showcase_id_sth->fetchrow_array();

        } else {
            $visit_group->{'orig_label'} = $visit_group->{'label'};
            $visit_group->{'label'} =~ s/-/%/g;
            $visit_group->{'label'} =~ s|^/||g;
            $visit_group->{'label'} .= '%';
            $division_sth->execute($visit_group->{'label'});
            ($division_name) = $division_sth->fetchrow_array();
        }

        if ($division_name) {

            ## no idea why this is nb_hits, and not nb_actions, like every other method
            my @data_value = ( { 'nb_actions' => ($visit_group->{'nb_hits'} || $visit_group->{'nb_visits'}), 'label' => $division_name } );
            &updateCompanyStats($idsite, 'showcase', $prev_date, \@data_value);
        }
    }
    return 1;
}

and

sub researchViewsSubData {
    my ($api_call, $stat_section, $idsite, $prev_date, $last_of_month, $params, $subtable) = @_;

    return unless ($subtable);

    my %sub_params = %{ clone ($params) };
    $sub_params{'idSubtable'} = $subtable->{'idsubdatatable'};

    # $data contains views for each primary showcase page
    my $data = &fetchStatsData($api_call, \%sub_params);

    if (ref $data ne 'ARRAY') {
        carp "$api_call returned something not an array!";
        return 0;
    }

    foreach my $visit_group (@$data) {
        # if ($visit_group->{'url'} && $visit_group->{'url'} =~ /inthenews|pressreleases|downloads/) {
        if ($visit_group->{'idsubdatatable'}) {
            &researchViewsSubData($api_call, $stat_section, $idsite, $prev_date, $last_of_month, $params, $visit_group);
            next;
        }

        my $division_name;

        if ($visit_group->{'url'}) {
            my $tag_id = $visit_group->{'url'};
            $tag_id =~ s/^.*?\/(\d+)\/.*$/$1/;

            next if ($tag_id !~ /^\d+$/);

            $division_by_tag_id_sth->execute(int($tag_id), int($idsite));
            ($division_name) = $division_by_tag_id_sth->fetchrow_array();

        } else {
            carp Dumper($visit_group);
        }

        if ($division_name) {
            ## no idea why this is nb_hits, and not nb_actions, like every other method
            my @data_value = ( { 'nb_actions' => ($visit_group->{'nb_hits'} || $visit_group->{'nb_visits'}), 'label' => $division_name } );

            &updateCompanyStats($idsite, 'research', $prev_date, \@data_value);
        }
    }
    return 1;
}

Answer 1

As what Sinan said, you should start from scratch and design a new class or module (or set of them) for dealing with the functionality you need.

Below is something that I produced based on your code. It is meant just as an outline to give you a direction.

package MyCompanyStats;

sub process_stats {
    my $params = shift;
    my $data = fetchStatsData($params);
    foreach my $visit_group (@$data) {
        process_stats_group($visit_group);
    }
}

sub process_stats_group {
    my $group = shift;
    if ($group->{'url'}) {
        my $showcase_id = get_showcase_id($group);
        save_by_showcase_id($showcase_id, $group);
    } else {
        my $group_label = get_group_label($group);
        save_by_label($group_label, $group);
    }
}

sub get_group_label {
    my $group = shift;
    my $label = $group->{'label'};
    for ($label) {
        s/-/%/g;
        s|^/||g;
    }
    $label .= '%';
    return $label;
}

sub get_showcase_id {
    my $group = shift;
    my $showcase_id = $visit_group->{'url'};
    $showcase_id =~ s/^.*?\/(\d+)\/.*$/$1/;
    return $showcase_id;
}

1;