Column Search in DataTables using Server-side Processing

Last modified on September 19th, 2018 by Vincy.

DataTables is a feature-packed jQuery plugin used to display results in a tabular form. It provides a rich interface with effective sorting, search, pagination like features to manage a list of records. Using this we can list both static and dynamic results. In a previous article, we have seen how to show dynamic results with DataTables using its server-side processing. As a followup to it, in this article we are going to see how to enable column search in the list view using DataTables. 

https://phppot.com/wp-content/uploads/2018/09/Column-Search-in-DataTables-using-Server-side-Processing

View Demo

In this article, Previously, we have seen DataTable examples with the common search filter. By using this library, it will greatly reduce the developer’s effort to build these features with a custom code. If you want the custom code for search using PHP the linked article will be helpful for you.

DataTable provides the header and footer callback functions to manipulate HTML data with an intuitive user interface. For implementing the server-side processing, I have used the DataTables SSP class to build the data request with the required parameters. The response will be encoded as JSON and loaded into the DataTable UI via AJAX.

Integrating DataTables into HTML

DataTables library is added as the part of the downloadable source code. You can also download the DataTables library from Github. This code shows the HTML to include and initialize DataTables for integrating this library.

A HTML table reference is used as a jQuery selector to initialize DataTables with the required options. For adding jQuery DataTables plugin, we need to go with the following steps.

<!DOCTYPE html>
<head>
    <script src="vendor/jquery/jquery-3.2.1.min.js"></script>

    <link rel="stylesheet"  href="vendor/DataTables/jquery.datatables.min.css">	
    <script src="vendor/DataTables/jquery.dataTables.min.js" type="text/javascript"></script> 
   
    <link href="style.css" rel="stylesheet" type="text/css" />

    <title>Column Search in DataTables using Server-side Processing</title>
    <script>
        $(document).ready(function ()
        {   
            var table = $('#tbl-contact').DataTable({
                	"scrollX": true,
            		"pagingType": "numbers",
                "processing": true,
                "serverSide": true,
                "ajax": "server.php",
                order: [[2, 'asc']],
                columnDefs: [{
                    targets: "_all",
                    orderable: false
                 }]
            });
        });

    </script>
</head>

<body>
    <div class="datatable-container">
        <h2>Column Search in DataTables using Server-side Processing</h2>
        <table name="tbl-contact" id="tbl-contact" class="display" cellspacing="0" width="100%">   

            <thead>
                <tr>
                    <th>Name</th>
                    <th>Last Name</th>
                    <th>Address</th>
                    <th>Phone</th>
                    <th>Date Of Birth</th>
                </tr>
            </thead>
            
        </table>
    </div>
</body>
</html>

DataTable Header Callback to Show Column Search

In this section, it shows the code for the DataTables callback function. In this callback, the table header is changed by appending the search option with each column.

With the reference of the search input field on each column header, the key-up event is mapped to trigger the search request. By typing on the search input field, the DataTables column data will be filtered out.

<script>
    $(document).ready(function ()
    {
        $('#tbl-contact thead th').each(function () {
            var title = $(this).text();
            $(this).html(title+' <input type="text" class="col-search-input" placeholder="Search ' + title + '" />');
        });
        
        table.columns().every(function () {
            var table = this;
            $('input', this.header()).on('keyup change', function () {
                if (table.search() !== this.value) {
                	   table.search(this.value).draw();
                }
            });
        });
    });

</script>

PHP Code to Get Data for DataTables by Server-Side Processing

This is the code that the server.php file contains. This file will be marked as the source option while initializing the DataTables. In this code, the database configuration, table, columns, primary_key specifications are declared and used to request the dynamic data for the DataTables.

The SSP class is used to fetch the database result by using simple/complex server-side processing. The resultant data will be encoded in JSON format and echoed to the AJAX callback.

<?php
 
/*
 * DataTables example server-side processing script.
 *
 * Please note that this script is intentionally extremely simply to show how
 * server-side processing can be implemented, and probably shouldn't be used as
 * the basis for a large complex system. It is suitable for simple use cases as
 * for learning.
 *
 * See http://datatables.net/usage/server-side for full details on the server-
 * side processing requirements of DataTables.
 *
 * @license MIT - http://datatables.net/license_mit
 */
 
/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
 * Easy set variables
 */
 
// DB table to use
$table = 'tbl_contact';
 
// Table's primary key
$primaryKey = 'id';
 
// Array of database columns which should be read and sent back to DataTables.
// The `db` parameter represents the column name in the database, while the `dt`
// parameter represents the DataTables column identifier. In this case simple
// indexes
$columns = array(
    array( 'db' => 'first_name', 'dt' => 0 ),
    array( 'db' => 'last_name',  'dt' => 1 ),
    array( 'db' => 'address',   'dt' => 2 ),
    array( 'db' => 'phone', 'dt' => 3,),
    array( 'db' => 'date_of_birth','dt' => 4,
        'formatter' => function( $d, $row ) {
            return date( 'd-m-Y', strtotime($d));
        }
    )
   
);
 
// SQL server connection information
$sql_details = array(
    'user' => 'root',
    'pass' => 'test',
    'db'   => 'blog_samples',
    'host' => 'localhost'
);
 
 
/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
 * If you just want to use the basic configuration for DataTables with PHP
 * server-side, there is no need to edit below this line.
 */
 
require( 'vendor/DataTables/server-side/scripts/ssp.class.php' );
 
echo json_encode(
    SSP::simple( $_GET, $sql_details, $table, $primaryKey, $columns )
);

DataTables SSP Class

This is the PHP class used for the DataTables server-side processing. It handles the Database connection establishment and query handling for data filtering, ordering and many purposes.

<?php

/*
 * Helper functions for building a DataTables server-side processing SQL query
 *
 * The static functions in this class are just helper functions to help build
 * the SQL used in the DataTables demo server-side processing scripts. These
 * functions obviously do not represent all that can be done with server-side
 * processing, they are intentionally simple to show how it works. More complex
 * server-side processing operations will likely require a custom script.
 *
 * See http://datatables.net/usage/server-side for full details on the server-
 * side processing requirements of DataTables.
 *
 * @license MIT - http://datatables.net/license_mit
 */


// REMOVE THIS BLOCK - used for DataTables test environment only!
$file = $_SERVER['DOCUMENT_ROOT'].'/datatables/pdo.php';
if ( is_file( $file ) ) {
	include( $file );
}


class SSP {
	/**
	 * Create the data output array for the DataTables rows
	 *
	 *  @param  array $columns Column information array
	 *  @param  array $data    Data from the SQL get
	 *  @return array          Formatted data in a row based format
	 */
	static function data_output ( $columns, $data )
	{
		$out = array();

		for ( $i=0, $ien=count($data) ; $i<$ien ; $i++ ) {
			$row = array();

			for ( $j=0, $jen=count($columns) ; $j<$jen ; $j++ ) {
				$column = $columns[$j];

				// Is there a formatter?
				if ( isset( $column['formatter'] ) ) {
					$row[ $column['dt'] ] = $column['formatter']( $data[$i][ $column['db'] ], $data[$i] );
				}
				else {
					$row[ $column['dt'] ] = $data[$i][ $columns[$j]['db'] ];
				}
			}

			$out[] = $row;
		}

		return $out;
	}


	/**
	 * Database connection
	 *
	 * Obtain an PHP PDO connection from a connection details array
	 *
	 *  @param  array $conn SQL connection details. The array should have
	 *    the following properties
	 *     * host - host name
	 *     * db   - database name
	 *     * user - user name
	 *     * pass - user password
	 *  @return resource PDO connection
	 */
	static function db ( $conn )
	{
		if ( is_array( $conn ) ) {
			return self::sql_connect( $conn );
		}

		return $conn;
	}


	/**
	 * Paging
	 *
	 * Construct the LIMIT clause for server-side processing SQL query
	 *
	 *  @param  array $request Data sent to server by DataTables
	 *  @param  array $columns Column information array
	 *  @return string SQL limit clause
	 */
	static function limit ( $request, $columns )
	{
		$limit = '';

		if ( isset($request['start']) && $request['length'] != -1 ) {
			$limit = "LIMIT ".intval($request['start']).", ".intval($request['length']);
		}

		return $limit;
	}


	/**
	 * Ordering
	 *
	 * Construct the ORDER BY clause for server-side processing SQL query
	 *
	 *  @param  array $request Data sent to server by DataTables
	 *  @param  array $columns Column information array
	 *  @return string SQL order by clause
	 */
	static function order ( $request, $columns )
	{
		$order = '';

		if ( isset($request['order']) && count($request['order']) ) {
			$orderBy = array();
			$dtColumns = self::pluck( $columns, 'dt' );

			for ( $i=0, $ien=count($request['order']) ; $i<$ien ; $i++ ) {
				// Convert the column index into the column data property
				$columnIdx = intval($request['order'][$i]['column']);
				$requestColumn = $request['columns'][$columnIdx];

				$columnIdx = array_search( $requestColumn['data'], $dtColumns );
				$column = $columns[ $columnIdx ];

				if ( $requestColumn['orderable'] == 'true' ) {
					$dir = $request['order'][$i]['dir'] === 'asc' ?
						'ASC' :
						'DESC';

					$orderBy[] = '`'.$column['db'].'` '.$dir;
				}
			}

			if ( count( $orderBy ) ) {
				$order = 'ORDER BY '.implode(', ', $orderBy);
			}
		}

		return $order;
	}


	/**
	 * Searching / Filtering
	 *
	 * Construct the WHERE clause for server-side processing SQL query.
	 *
	 * NOTE this does not match the built-in DataTables filtering which does it
	 * word by word on any field. It's possible to do here performance on large
	 * databases would be very poor
	 *
	 *  @param  array $request Data sent to server by DataTables
	 *  @param  array $columns Column information array
	 *  @param  array $bindings Array of values for PDO bindings, used in the
	 *    sql_exec() function
	 *  @return string SQL where clause
	 */
	static function filter ( $request, $columns, &$bindings )
	{
		$globalSearch = array();
		$columnSearch = array();
		$dtColumns = self::pluck( $columns, 'dt' );

		if ( isset($request['search']) && $request['search']['value'] != '' ) {
			$str = $request['search']['value'];

			for ( $i=0, $ien=count($request['columns']) ; $i<$ien ; $i++ ) {
				$requestColumn = $request['columns'][$i];
				$columnIdx = array_search( $requestColumn['data'], $dtColumns );
				$column = $columns[ $columnIdx ];

				if ( $requestColumn['searchable'] == 'true' ) {
					$binding = self::bind( $bindings, '%'.$str.'%', PDO::PARAM_STR );
					$globalSearch[] = "`".$column['db']."` LIKE ".$binding;
				}
			}
		}

		// Individual column filtering
		if ( isset( $request['columns'] ) ) {
			for ( $i=0, $ien=count($request['columns']) ; $i<$ien ; $i++ ) {
				$requestColumn = $request['columns'][$i];
				$columnIdx = array_search( $requestColumn['data'], $dtColumns );
				$column = $columns[ $columnIdx ];

				$str = $requestColumn['search']['value'];

				if ( $requestColumn['searchable'] == 'true' &&
				 $str != '' ) {
					$binding = self::bind( $bindings, '%'.$str.'%', PDO::PARAM_STR );
					$columnSearch[] = "`".$column['db']."` LIKE ".$binding;
				}
			}
		}

		// Combine the filters into a single string
		$where = '';

		if ( count( $globalSearch ) ) {
			$where = '('.implode(' OR ', $globalSearch).')';
		}

		if ( count( $columnSearch ) ) {
			$where = $where === '' ?
				implode(' AND ', $columnSearch) :
				$where .' AND '. implode(' AND ', $columnSearch);
		}

		if ( $where !== '' ) {
			$where = 'WHERE '.$where;
		}

		return $where;
	}


	/**
	 * Perform the SQL queries needed for an server-side processing requested,
	 * utilising the helper functions of this class, limit(), order() and
	 * filter() among others. The returned array is ready to be encoded as JSON
	 * in response to an SSP request, or can be modified if needed before
	 * sending back to the client.
	 *
	 *  @param  array $request Data sent to server by DataTables
	 *  @param  array|PDO $conn PDO connection resource or connection parameters array
	 *  @param  string $table SQL table to query
	 *  @param  string $primaryKey Primary key of the table
	 *  @param  array $columns Column information array
	 *  @return array          Server-side processing response array
	 */
	static function simple ( $request, $conn, $table, $primaryKey, $columns )
	{
		$bindings = array();
		$db = self::db( $conn );

		// Build the SQL query string from the request
		$limit = self::limit( $request, $columns );
		$order = self::order( $request, $columns );
		$where = self::filter( $request, $columns, $bindings );

		// Main query to actually get the data
		$data = self::sql_exec( $db, $bindings,
			"SELECT `".implode("`, `", self::pluck($columns, 'db'))."`
			 FROM `$table`
			 $where
			 $order
			 $limit"
		);

		// Data set length after filtering
		$resFilterLength = self::sql_exec( $db, $bindings,
			"SELECT COUNT(`{$primaryKey}`)
			 FROM   `$table`
			 $where"
		);
		$recordsFiltered = $resFilterLength[0][0];

		// Total data set length
		$resTotalLength = self::sql_exec( $db,
			"SELECT COUNT(`{$primaryKey}`)
			 FROM   `$table`"
		);
		$recordsTotal = $resTotalLength[0][0];

		/*
		 * Output
		 */
		return array(
			"draw"            => isset ( $request['draw'] ) ?
				intval( $request['draw'] ) :
				0,
			"recordsTotal"    => intval( $recordsTotal ),
			"recordsFiltered" => intval( $recordsFiltered ),
			"data"            => self::data_output( $columns, $data )
		);
	}


	/**
	 * The difference between this method and the `simple` one, is that you can
	 * apply additional `where` conditions to the SQL queries. These can be in
	 * one of two forms:
	 *
	 * * 'Result condition' - This is applied to the result set, but not the
	 *   overall paging information query - i.e. it will not effect the number
	 *   of records that a user sees they can have access to. This should be
	 *   used when you want apply a filtering condition that the user has sent.
	 * * 'All condition' - This is applied to all queries that are made and
	 *   reduces the number of records that the user can access. This should be
	 *   used in conditions where you don't want the user to ever have access to
	 *   particular records (for example, restricting by a login id).
	 *
	 *  @param  array $request Data sent to server by DataTables
	 *  @param  array|PDO $conn PDO connection resource or connection parameters array
	 *  @param  string $table SQL table to query
	 *  @param  string $primaryKey Primary key of the table
	 *  @param  array $columns Column information array
	 *  @param  string $whereResult WHERE condition to apply to the result set
	 *  @param  string $whereAll WHERE condition to apply to all queries
	 *  @return array          Server-side processing response array
	 */
	static function complex ( $request, $conn, $table, $primaryKey, $columns, $whereResult=null, $whereAll=null )
	{
		$bindings = array();
		$db = self::db( $conn );
		$localWhereResult = array();
		$localWhereAll = array();
		$whereAllSql = '';

		// Build the SQL query string from the request
		$limit = self::limit( $request, $columns );
		$order = self::order( $request, $columns );
		$where = self::filter( $request, $columns, $bindings );

		$whereResult = self::_flatten( $whereResult );
		$whereAll = self::_flatten( $whereAll );

		if ( $whereResult ) {
			$where = $where ?
				$where .' AND '.$whereResult :
				'WHERE '.$whereResult;
		}

		if ( $whereAll ) {
			$where = $where ?
				$where .' AND '.$whereAll :
				'WHERE '.$whereAll;

			$whereAllSql = 'WHERE '.$whereAll;
		}

		// Main query to actually get the data
		$data = self::sql_exec( $db, $bindings,
			"SELECT `".implode("`, `", self::pluck($columns, 'db'))."`
			 FROM `$table`
			 $where
			 $order
			 $limit"
		);

		// Data set length after filtering
		$resFilterLength = self::sql_exec( $db, $bindings,
			"SELECT COUNT(`{$primaryKey}`)
			 FROM   `$table`
			 $where"
		);
		$recordsFiltered = $resFilterLength[0][0];

		// Total data set length
		$resTotalLength = self::sql_exec( $db, $bindings,
			"SELECT COUNT(`{$primaryKey}`)
			 FROM   `$table` ".
			$whereAllSql
		);
		$recordsTotal = $resTotalLength[0][0];

		/*
		 * Output
		 */
		return array(
			"draw"            => isset ( $request['draw'] ) ?
				intval( $request['draw'] ) :
				0,
			"recordsTotal"    => intval( $recordsTotal ),
			"recordsFiltered" => intval( $recordsFiltered ),
			"data"            => self::data_output( $columns, $data )
		);
	}


	/**
	 * Connect to the database
	 *
	 * @param  array $sql_details SQL server connection details array, with the
	 *   properties:
	 *     * host - host name
	 *     * db   - database name
	 *     * user - user name
	 *     * pass - user password
	 * @return resource Database connection handle
	 */
	static function sql_connect ( $sql_details )
	{
		try {
			$db = @new PDO(
				"mysql:host={$sql_details['host']};dbname={$sql_details['db']}",
				$sql_details['user'],
				$sql_details['pass'],
				array( PDO::ATTR_ERRMODE => PDO::ERRMODE_EXCEPTION )
			);
		}
		catch (PDOException $e) {
			self::fatal(
				"An error occurred while connecting to the database. ".
				"The error reported by the server was: ".$e->getMessage()
			);
		}

		return $db;
	}


	/**
	 * Execute an SQL query on the database
	 *
	 * @param  resource $db  Database handler
	 * @param  array    $bindings Array of PDO binding values from bind() to be
	 *   used for safely escaping strings. Note that this can be given as the
	 *   SQL query string if no bindings are required.
	 * @param  string   $sql SQL query to execute.
	 * @return array         Result from the query (all rows)
	 */
	static function sql_exec ( $db, $bindings, $sql=null )
	{
		// Argument shifting
		if ( $sql === null ) {
			$sql = $bindings;
		}

		$stmt = $db->prepare( $sql );
		//echo $sql;

		// Bind parameters
		if ( is_array( $bindings ) ) {
			for ( $i=0, $ien=count($bindings) ; $i<$ien ; $i++ ) {
				$binding = $bindings[$i];
				$stmt->bindValue( $binding['key'], $binding['val'], $binding['type'] );
			}
		}

		// Execute
		try {
			$stmt->execute();
		}
		catch (PDOException $e) {
			self::fatal( "An SQL error occurred: ".$e->getMessage() );
		}

		// Return all
		return $stmt->fetchAll( PDO::FETCH_BOTH );
	}


	/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
	 * Internal methods
	 */

	/**
	 * Throw a fatal error.
	 *
	 * This writes out an error message in a JSON string which DataTables will
	 * see and show to the user in the browser.
	 *
	 * @param  string $msg Message to send to the client
	 */
	static function fatal ( $msg )
	{
		echo json_encode( array( 
			"error" => $msg
		) );

		exit(0);
	}

	/**
	 * Create a PDO binding key which can be used for escaping variables safely
	 * when executing a query with sql_exec()
	 *
	 * @param  array &$a    Array of bindings
	 * @param  *      $val  Value to bind
	 * @param  int    $type PDO field type
	 * @return string       Bound key to be used in the SQL where this parameter
	 *   would be used.
	 */
	static function bind ( &$a, $val, $type )
	{
		$key = ':binding_'.count( $a );

		$a[] = array(
			'key' => $key,
			'val' => $val,
			'type' => $type
		);

		return $key;
	}


	/**
	 * Pull a particular property from each assoc. array in a numeric array, 
	 * returning and array of the property values from each item.
	 *
	 *  @param  array  $a    Array to get data from
	 *  @param  string $prop Property to read
	 *  @return array        Array of property values
	 */
	static function pluck ( $a, $prop )
	{
		$out = array();

		for ( $i=0, $len=count($a) ; $i<$len ; $i++ ) {
			$out[] = $a[$i][$prop];
		}

		return $out;
	}


	/**
	 * Return a string from an array or a string
	 *
	 * @param  array|string $a Array to join
	 * @param  string $join Glue for the concatenation
	 * @return string Joined string
	 */
	static function _flatten ( $a, $join = ' AND ' )
	{
		if ( ! $a ) {
			return '';
		}
		else if ( $a && is_array($a) ) {
			return implode( $join, $a );
		}
		return $a;
	}
}

Database Script

This is the SQL script containing the database table tbl_contact structure and the data dump. Import this script while setting this example in your local.

--
-- Table structure for table `tbl_contact`
--

CREATE TABLE `tbl_contact` (
  `id` int(11) NOT NULL,
  `first_name` varchar(255) NOT NULL,
  `last_name` varchar(255) NOT NULL,
  `address` varchar(255) NOT NULL,
  `email` varchar(255) NOT NULL,
  `phone` varchar(20) NOT NULL,
  `date_of_birth` date NOT NULL
) ENGINE=InnoDB DEFAULT CHARSET=latin1;

--
-- Dumping data for table `tbl_contact`
--

INSERT INTO `tbl_contact` (`id`, `first_name`, `last_name`, `address`, `email`, `phone`, `date_of_birth`) VALUES
(1, 'Peter', 'Parker', '970 Princee St.\r\n\r\nPiqua, OH 45356', 'peter@parker.com', '4455664455', '1990-04-10'),
(2, 'Barry', 'Allen', '976 Livingston Lane \r\nSaint Peterssburg, FL 33702', 'barry@allen.com', '2211335566', '1983-02-02'),
(3, 'Bruce', 'Banner', '69 Bridge Lane \r\nBrooklyn, NY 11201', 'bruce@banner.com', '7788995566', '1987-04-14'),
(4, 'Bruce', 'Wayne', '896 East Smith Store Dr. \r\nLake Jackson, TX 77566', 'bruce@wayne.com', '8877887744', '1991-11-15'),
(5, 'Harvy', 'Dent', '35 Wakehurst Avenue \r\nNoblesville, IN 46060', 'harvy@dent.com', '9988774445', '1990-10-01'),
(6, 'Tony', 'Stark', '31 Edgewater Court \r\nMalden, MA 02148', 'tony@stark.com', '8899886655', '1984-10-05'),
(7, 'Nick', 'Fury', '70 WakePrin St.\r\n\r\nPiqua, OL 356444', 'nick@fury.com', '9966554488', '1980-01-25'),
(8, 'John', 'Mclane', '76 Kevins Lane \r\n\r\nSt. Petersburg, FN 33711', 'john@maclay.com', '7744114411', '2000-11-15'),
(9, 'Howard', 'Roark', '88 Golden Lane \r\n\r\nBrooklyn, LS 11204', 'howard@roark.com', '8745554413', '2011-11-15'),
(10, 'Peter', 'Keating', '86 Smith Road\r\n\r\nLake Jackson, TQ 77566', 'peter@keating.com', '9089094445', '2013-11-15');


--
-- Indexes for table `tbl_contact`
--
ALTER TABLE `tbl_contact`
  ADD PRIMARY KEY (`id`);
COMMIT;

Column Search with DataTable Server Side Processing Output

This output screenshot shows the dynamically fetched results by using DataTables Server-side scripts. In this output, the column-wise search option is shown to filter the column data by keyword.

column-search-in-datatables-using-server-side-processing-output

View DemoDownload

This PHP code tutorial was published on September 18, 2018.

↑ Back to Top

Share this Article