The Perl Toolchain Summit needs more sponsors. If your company depends on Perl, please support this very important event.

NAME

Web::Util::DBIC::Paging - Easily page, search, and sort DBIx::Class::ResultSets in a web context

VERSION

version 0.001004

SYNOPSIS

 package MyApp::People;

 use Web::Simple;
 use JSON::MaybeXS;
 use Web::Util::ExtPaging;
 use Web::Util::DBIC::Paging;

 sub dispatch_request {
  my $people_rs = get_rs();

  sub (/people) {
    [
       200,
       [ 'Content-type', 'application/json' ],
       [
         encode_json(
            ext_paginate(
               search(
                  page_and_sort($rs)
               )
            )
         ) ],
    ]
  },
  sub () { [ 404, [ 'Content-type', 'text/plain' ], [ 'not found' ] ] }
 }

DESCRIPTION

This module helps you to map various DBIx::Class features to CGI parameters. For the most part that means it will help you search, sort, and paginate with a minimum of effort and thought.

EXPORTED SUBS

All subs take a type, paramish thing, resultset, and optionally a config. All methods return a ResultSet. Subs are exported with Sub::Exporter::Progressive, so should be fast and light for the defaults but upgrade to actually using Sub::Exporter if you need to alias or prefix the subs.

The "paramish thing" is what the type is for and can be any of:

c | ctx | context | catalyst

for the $c argument in a catalyst app

r | req | request

for a Plack::Request object

e | env | psgi_env

for a PSGI Environment hashref.

raw

for a plain hashref.

page_and_sort

 my $result_rs  = page_and_sort(c => $c, $c->model('DB::Foo'));

This is a helper method that will first sort your data and then "paginate" it. Valid configuration parameters are documented for each of those methods.

paginate

 my $result_rs  = paginate(c => $c, $c->model('DB::Foo'));

Paginates the passed in resultset based on the following parameters:

start first row to display
limit amount of rows per page

The sole config param is page_size which will be the page size if there is no limit parameter in the request. The default page_size is 25.

 my $searched_rs = search(c => $c, $c->model('DB::Foo'));

If the $resultset has a controller_search method it will call that method on the passed in resultset with all of the CGI parameters. I like to have this method look something like the following:

 # Base search dispatcher, defined in MyApp::Schema::ResultSet
 sub _build_search {
    my ($rs, $dispatch_table, $q) = @_;

    foreach ( keys %{$q} ) {
       if ( my $fn = $dispatch_table->{$_} and $q->{$_} ) {
          my ( $search, $meta ) = $fn->( $q->{$_} );
          $rs = $rs->search($search, $meta);
       }
    }

    return $rs;
 }

 # search method in specific resultset
 sub controller_search {
    my $self   = shift;
    my $params = shift;
    return $self->_build_search({
       status => sub {
          return { 'repair_order_status' => shift }, {};
       },
       part_id => sub {
          return {
             'lineitems.part_id' => { -like => q{%}.shift( @_ ).q{%} }
          }, { join => 'lineitems' };
       },
    },$params);
 }

If the controller_search method does not exist, this method will call "simple_search" instead.

sort_rs

 my $result_rs  = sort_rs(c => $c, $c->model('DB::Foo'));

Exactly the same as "search", except calls controller_sort or "simple_sort". Here is how I use it:

 # Base sort dispatcher, defined in MyApp::Schema::ResultSet
 sub _build_sort {
    my ($self, $dispatch_table, $default, $q) = @_;

    my %search = ();
    my %meta   = ();

    my $direction = $q->{dir};
    my $sort      = $q->{sort};

    if ( my $fn = $dispatch_table->{$sort} ) {
       my ( $tmp_search, $tmp_meta ) = $fn->( $direction );
       %search = ( %search, %{$tmp_search||{}} );
       %meta   = ( %meta,   %{$tmp_meta||{}} );
    } elsif ( $sort && $direction ) {
       my ( $tmp_search, $tmp_meta ) = $default->( $sort, $direction );
       %search = ( %search, %{$tmp_search||{}} );
       %meta   = ( %meta,   %{$tmp_meta||{}} );
    }

    return $self->search(\%search, \%meta);
 }

 # sort method in specific resultset
 sub controller_sort {
    my $self = shift;
    my $params = shift;
    return $self->_build_sort({
       first_name => sub {
          my $direction = shift;
          return {}, {
             order_by => { "-$direction" => [qw{last_name first_name}] },
          };
       },
    }, sub {
       my $param = shift;
       my $direction = shift;
       return {}, {
          order_by => { "-$direction" => $param },
       };
    },$params);
 }

simple_deletion

 my $deleted_ids = simple_deletion(c => $c, $c->model('DB::Foo'));

Deletes from the passed in resultset based on the sole CGI parameter, to_delete, which must be a list of primary keys.

This is the only method that does not return a ResultSet. Instead it returns an arrayref of the id's that it deleted. If the ResultSet has has a multipk this will expect each tuple of PK's to be separated by commas.

Note that this method uses the $rs->delete method, as opposed to $rs->delete_all

 my $searched_rs = simple_search(c => $c, $c->model('DB::Foo'));

Searches the resultset based on all fields in the request. Searches with $fieldname => { -like => "%$value%" } for char fields, everything else gets basic equality searchs. If there are multiple values for a CGI parameter it will use all values via an or.

The sole configuration value is skip and it is used to skip unsearchable parameters. The default is limit start sort dir _dc rm xaction.

simple_sort

 my $sorted_rs = simple_sort(c => $c, $c->model('DB::Foo'));

Sorts the passed in resultset based on the following CGI parameters:

sort field to sort by, defaults to primarky key =item dir direction to sort

AUTHOR

Arthur Axel "fREW" Schmidt <frioux+cpan@gmail.com>

COPYRIGHT AND LICENSE

This software is copyright (c) 2018 by Arthur Axel "fREW" Schmidt.

This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.