projekt kivitendo

A search filter will usually search for things in relations of the actual

search target. A search for sales orders may be filtered by the name of the

  query => [

    'customer.name'          => 'John Doe',

    'orddate'                => [ lt    => DateTime->today ],

  ]

parser will convert them into nested structures like this:

  $::form = bless {

  [% L.select_tag('filter.salesman.id', ...) %]

  [% L.input_tag('filter.department.description:substr::ilike', ...) %]

=head1 LAUNDERING

Unfortunately Template cannot parse the postfixes if you want to

default laundered into underscores, so you can use them like this:

  [% L.input_tag('filter.price:number::lt', filter.price_number__lt) %]

All of your original entries will stay intact. If you don't want this to

happen pass C<< no_launder => 1 >> as a parameter.  Additionally you can pass a

different target for the laundered values with the C<launder_to>  parameter. It

if you have a filter that looks like this:

  $filter = {

    ]

  ]

=head1 FILTERS (leading with :)

  L.input_tag('customer.name:substr::ilike', ...)

  L.input_tag('invoice.customer.name:substr::ilike', ...)

both inputs. This is because tables are aliased by their name and not by their

position in with_objects.

=item *

=back

=head1 NAME

=head1 SYNOPSIS

DBUtils relies heavily on two parameters which have to be passed to almost every function: $form and $dbh.

  - $form is used for error handling only. It can be omitted in theory, but should not.

Every function here should accomplish the follwing things:

        0,

        0]) unless $info_ok;

      # and no empty headers.

      if ($info_ok && $self->header) {

        my @header = @{ $self->header };

  # people insist that case sensitivity doesn't exist and try to enter all

  # sorts of stuff. at this point we've got a profile (with keys that represent

  # valid methods), and a header full of strings. if two of them match, the user

  if ($self->case_insensitive_header) {

    die 'case_insensitive_header is only possible with profile' unless $self->profile;

    if ($header) {

sub _check_multiplex_datatype_position {

  my ($self) = @_;

  my @positions = map { firstidx { 'datatype' eq lc($_) } @{ $_ } } @{ $self->header };

  my $first_pos = $positions[0];

See Synopsis.

so this model just skips ahead and gives you objects.

Its basic assumptions are:

=item You do know what you expect to be in that csv file.

date format, csv parameters such as quoting and separation characters. You also

know what content will be in that csv and what L<Rose::DB> is responsible for

it. You provide valid header columns and their mapping to the objects.

In this case C<listprice_as_number> will be used to read in values from the

C<listprice> column.

  [ {profile => { customer => 'customer.name' }} ]