To install simply move htdocs to your public html directory, then set BOILER_LOCATION on line 3 of index to point to the framework folder.
Controllers are the start point of your page.
Testing and Building
Building and testing (optional) is done with a build.xml make file for "ant". You may also consider using the following to help test your code.
The following tools have a config shipped with ant targets, and are fully compatible with Hudson/Jenkins CI
pear channel-discover pear.phpmd.org
pear channel-discover pear.pdepend.org
pear install --alldeps phpmd/PHP_PMD
pear install PHP_CodeSniffer
- PHP Copy and Paste detector
pear channel-discover components.ez.no
pear install phpunit/phpcpd
pear channel-discover pear.phpunit.de
pear channel-discover pear.symfony-project.com
pear install phpunit/PHPUnit
pear install phpunit/PHPUnit_SkeletonGenerator
CONTROLLERS
-Namespacing follows folder structure to allow PHP lazy loading (please see php.net for Namespacing and __autoload)
-All classes must be namespaced with Controller. This is to allow you to have a controller and model, for example, with the same name.
-Functions being called must be public
-First letter only of class is capitalized
Examples:
http://localhost/class/function/args1/args2/args3
Would execute:
$c = new \Controller\Class();
$c->function(args1, args2, args3);
http://localhost/ns1/ns2/ns3/class/function
Would Execute:
$c = new \Controller\ns1\ns2\ns3\Class();
$c->function();
MODELS
-Models should extend DBObject normally for MySQL objects
-Models must be namespaced
-DBObject class (found in application/model/DBObject.php) must have login details entered for PHP
<?php
namespace Model;
class MySQLTable extends DBObject {
public static function getTable($read=true) {
return "mysql_table_name";
}
public static function getPrimaryKey() {
return "mysql_primary_key";
OR
return array("concat", "key");
}
}
//That's it! No SQL! Your MySQL settings are retrieved from a config.php file, built by ant. This stops you committing your MySQL details to git - pretty irritating!
?>
More advanced functions:
N.B. This is a demo which came from when the framework was not namespaced.
<?php
ini_set('display_errors', "On");
include "Linq.php";
include "DBObject.php";
/*
Employee
*/
class Employee extends DBObject {
public static function getPrimaryKey() {
//Use an array for concatinated keys
return "id";
}
public static function getTable($read=true) {
//$read variable gives the class a "heads up" about what is going to be done. Sometimes I have created a
//MySQL view for reading (which of course is read only) and therefore have had to specify a table as well
//for write operations
return "employee";
}
public static function getDB() {
return LinqDB::getDB("mysql.bcslichfield.com", "star241_6", "devpasswd123=", "bcslichfield_dev");
}
//Some nice utility functions
public function getJobs() {
return EmployeeJob::getJobsByEmployee($this);
}
public function giveJob(Job $j) {
EmployeeJob::giveJob($this, $j);
}
}
/*
LINK TABLE to stop many-to-many between Employee and Job
*/
class EmployeeJob extends DBObject {
public static function getPrimaryKey() {
//Use an array for concatinated keys
return array("employee", "job");
}
public static function getTable($read=true) {
//$read variable gives the class a "heads up" about what is going to be done. Sometimes I have created a
//MySQL view for reading (which of course is read only) and therefore have had to specify a table as well
//for write operations
return "employee_job";
}
public static function getDB() {
return LinqDB::getDB("mysql.bcslichfield.com", "star241_6", "devpasswd123=", "bcslichfield_dev");
}
//Some nice function we'll just add in because they are nice to use in a link table
public static function getJobsByEmployee(Employee $e) {
return self::getByAttribute("employee", $e->id);
}
public static function getEmployeesByJob(Job $j) {
return self::getByAttribute("job", $j->id);
}
public static function giveJob(Employee $e, Job $j) {
self::Create(array('employee'=>$e->id, 'job'=>$j->id));
}
}
/*
Jobs table
*/
class Job extends DBObject {
public static function getPrimaryKey() {
//Use an array for concatinated keys
return array("id");
}
public static function getTable($read=true) {
//$read variable gives the class a "heads up" about what is going to be done. Sometimes I have created a
//MySQL view for reading (which of course is read only) and therefore have had to specify a table as well
//for write operations
return "job";
}
public static function getDB() {
return LinqDB::getDB("mysql.bcslichfield.com", "star241_6", "devpasswd123=", "bcslichfield_dev");
}
//A nice little function to aid us
public function getEmployees() {
return EmployeeJob::getEmployeesByJob($this);
}
}
// class LinqDB extends mysqli, so feel free to use it as mysqli too!
//Lets use it to setup some temp tables
$db = LinqDB::getDB("mysql.bcslichfield.com", "star241_6", "devpasswd123=", "bcslichfield_dev");
$db->query("CREATE TEMPORARY TABLE IF NOT EXISTS employee (
id INT AUTO_INCREMENT,
name varchar(50) NOT NULL,
male tinyint(1) NOT NULL DEFAULT 1,
dob DATETIME,
PRIMARY KEY (id)
) ENGINE=InnoDB");
$db->query("CREATE TEMPORARY TABLE IF NOT EXISTS job (
id INT AUTO_INCREMENT,
name varchar(50) NOT NULL,
PRIMARY KEY (id)
) ENGINE=InnoDB");
$db->query("CREATE TEMPORARY TABLE IF NOT EXISTS employee_job (
employee INT NOT NULL,
job INT NOT NULL,
PRIMARY KEY (employee, job)
) ENGINE=InnoDB");
/*
For debug (not applicable on temporary tables):
FOREIGN KEY (employee) REFERENCES employee (id) ON DELETE CASCADE ON UPDATE CASCADE,
FOREIGN KEY (job) REFERENCES job (id) ON DELETE CASCADE ON UPDATE CASCADE
*/
/*
Once upon a time, in a factory far away there were three workers!
*/
$me = Employee::Create(array("name"=>"Will Tinsdeall"));
$bob = Employee::Create(array("name"=>"Bob Bloggs"));
$joe = Employee::Create(array("name"=>"Joe Bloggs"));
/*
And my AUTO_INCREMENT ID was:
*/
echo "TEST 1: ".$me->id."\r\n";
/*
In the factory there were many jobs
*/
$joba = Job::Create(array('name'=>'A very hard job'));
$jobb = Job::Create(array('name'=>'An average job'));
$jobc = Job::Create(array('name'=>'An easy job'));
/*
And we all had jobs
*/
$me->giveJob($joba);
$me->giveJob($jobb);
$bob->giveJob($joba);
$bob->giveJob($jobc);
$joe->giveJob($jobb);
$joe->giveJob($jobc);
/*
First of all the boss wanted to know all the jobs that were being done
*/
foreach (Job::getAll() as $j) {
echo "Job {$j->id}: {$j->name}"."\r\n";
}
/*
Next, he decided to quickly call me
*/
$me = new Employee("1");
/*
And asked me to check to see who was working well.
*/
$db->Select(Employee);
/*
This consisted of people
EITHER:
*/
$db->getOrFilter();
/*
Those doing a medium job only
*/
$jobs = $db->Select(EmployeeJob);
$jobs->addCount("jobs");
$jobs->addField("employee");
$jobs->addField("job");
$jobs->setGroup("employee");
/SELECT COUNT() as jobs, employee FROM employee_job GROUP BY employee */
$j = $jobs->Select();
$jFilter = $db->getAndFilter();
$jFilter->eq("job","2")->eq("jobs", "1");
$j->setFilter($jFilter);
//SELECT FROM (SELECT COUNT() as jobs, id FROM employee_job GROUP BY employee) WHERE jobs=1 AND job=2
/*
Or those doing an easy job and a very hard job
*/
$jobsB = $db->Select(EmployeeJob);
$jobsB->addField("employee");
$jBFilter = $db->getAndFilter();
$jBFilter->eq('job', 1)->eq("job", 3);
$jobsB->setFilter($jBFilter);
//XXX this would work if the tables weren't temporary!
/*
$u = $db->Union();
$u->addSelect($jobsB)->addSelect($jobs);
echo $u->getSQL();
*/
$out = array();
foreach (array_merge($jobsB->Exec(), $jobs->Exec()) as $a) {
var_dump($a);
$out[$a->employee] = new Employee($a->employee);
}
var_dump($out);
/*
Lastly he wanted a to assess everyone at their jobs (simple JOIN)
*/
$qEmployee = $db->Select(Employee);
$qEmployeeJob = $db->Select(EmployeeJob);
$qJob = $db->Select(Job);
$qEmployeeJob->joinLeft("employee", $qEmployee, "id");
$qEmployeeJob->joinLeft("job", $qJob, "id");
var_dump($qEmployeeJob->Exec());
?>
Download .zip
Documentation
build.sh
If you know an application of this package, send a message to the author to add a link here.
