Laravel 4 DataTables library to handle the server-side calls.

0.1.3 2014-04-15 23:25 UTC

This package is not auto-updated.

Last update: 2024-03-25 23:30:18 UTC


README

Build Status

This library is for the server side processing of the client requests. It aims to be different than existing DataTable packages, taking a different approach that allows greater flexibility.

Composer

"require": {
    "samvaughton/ldt": "0.1.*"
}

Then run composer update.

Simple Example

$customers = \Customer::select('id', 'name', 'email', 'phone', 'date_registered');

$dth = new DataTable(
    new LaravelBuilder($customers),
    new Request(\Input::all()),
    array(
        new Column('id'),
        new Column('name', array('searchable' => true)),
        new Column('email', array('searchable' => true)),
        new Column('phone'),
        new Column('date_registered', array(
            'rowProcessor' => new \DateColumnProcessor()
        )),
        new Column('actions', array(
            'type' => Column::TYPE_STATIC,
            'rowProcessor' => function($value, $row, $originalRow) {
                return sprintf(
                    '<a href="/customer/edit/%s">Edit</a>',
                    $row['id']
                );
            }
        )
    )
);

return $dth->make();

I have imported the necessary namespace paths via use for DataTable, LaravelBuilder, Request and Column. DateColumnProcessor is a custom class, you can look at an example in the Column namespace.

Quite a lot is going on here, but it is very readable. The DataTable class accepts three parameters. A class that implements BuilderInterface (there is one already built for Laravel), a Request class which handles the parsing of the client side request and thirdly an array of columns which are visible to the user.

The first two parameters will be kept exactly the same 99% of the time.

Column

The column class accepts two parameters, the first one is required. It can either be a string or an array, this depends on the complexity of the column.

Each column can either be dynamic or static. A dynamic column is one that originates from the data source. Whereas a static column is appended onto the results after being fetched. Such as an actions column that contains buttons for edit, delete etc..

Aliases

If you are using aliases in your query and performing searches as well, then you will need to use an array to define the SQL column separately. An example will clear this up.

$query = DB::table('customers')->select('customers.name AS customerName');
$dth = new DataTable(
    new LaravelBuilder($customers),
    new Request(\Input::all()),
    array(
        new Column(array('customerName', 'customers.name'), array(
            'searchable' => true
        ))
    )
);

The select statement is using an alias, since MySQL cannot utilise aliases within the WHERE clause, we have to use its original column name customers.name. Otherwise the generated SQL for WHERE would look something like:

WHERE `customerName` LIKE '%john doe%'

Which is illegal.

Options

The column class has default options which are listed below and explained, these can be set via the second parameter like the example above.

'type'                 => self::TYPE_DYNAMIC
'sortable'             => true
'searchable'           => false
'rowProcessor'         => false
'filterTermProcessor'  => false
'filterQueryProcessor' => false
  • type can be TYPE_DYNAMIC or TYPE_STATIC.
  • sortable and searchable are booleans (true/false).
  • rowProcessor is a callback / class that implements the RowProcessorInterface.
  • filterTermProcessor is a callback / class that implements the FilterTermProcessorInterface.
  • filterQueryProcessor is a callback / class that implements the FilterQueryProcessorInterface. This is a bit more advanced and lets you define your own SQL to filter against.

Processor's

The rowProcessor options allows you to run a function against each column's data, this is for scenarios where you need to append some action buttons or convert a unix timestamp to a more readable date.

new Column('actions', array(
    'type' => Column::TYPE_STATIC,
    'rowProcessor' => function($value, $row, $originalRow) {
        return sprintf(
            '<a href="/customer/edit/%s">Edit</a>',
            $row['id']
        );
    }
)

This is a static column that appends an edit button onto every row, it utilises the customers id from the $row array. You may be wondering what the $originalRow is for, this is an untouched row that contains every column from your select statement. Using the above code, if we were to modify the id column and set the value to null then this processor would return:

<a href="/customer/edit/">Edit</a>

Due to the id column being modified before this one. Here we could utilise $originalRow['id'] to get the unaltered value.

If you are trying to use a column that isn't defined within the column array then you have to use $originalRow as these columns are not contained within $row.

Instead of passing a callback, you can also pass a class that implements the RowProcessorInterface.

<?php

namespace \Samvaughton\Ldt;

class ExampleColumnProcessor implements RowProcessorInterface
{

    /**
     * This will simply append the date to the column.
     */
    public function run($value, $row, $originalRow)
    {
        return sprintf("%s - %s", $value, date("Y-m-d"));
    }

}

The column instantiation may look something like this:

new Column('date', array(
    'type' => Column::TYPE_STATIC,
    'rowProcessor' => new \Samvaughton\Ldt\ExampleColumnProcessor
)

The filterTermProcessor option is similar to the rowProcessor except that it modifies the search term before the query is executed to fetch the results. Say you only wanted users to search in lowercase you could provide a callback/class that performs this.

new Column('name', array(
    'filterTermProcessor' => function($term) {
        return strtolower(trim($term));
    }
)

You can find examples of these processors within the Samvaughton\Ldt\Column namespace.

Contributing

I would love for people to help contribute to this library, send a pull request and I'll check it out!

Todo:

  • More Tests
  • Better Documentation