Skip to content

alexborisov/foxfire

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Introducing FoxFire

FoxFire is a PHP => NoSQL => SQL object-relational mapping library. It connects PHP to NoSQL data stores like Redis and Memcached, SQL datastores like Postgres and MySQL, and adds thread-safe multilevel caching with full transaction support.

There are plenty of other ORM's out there, but all of them have at least one of the following problems:

  • They don't let the developer manipulate the data as an object
  • They don't handle transactions
  • They don't handle caching
  • They try to handle every possible SQL operation, adding massive complexity

FoxFire lets developers work with tables in the database as if they were objects, and allows developers to run queries on them in a structured, object-oriented manner. It has ACID compliant transaction support, and handles transaction coupling between the SQL servers and cache engines. That means if you run a transaction against a cached table, and the transaction fails, the transaction gets rolled-back in the database and the cache.

FoxFire does not support every possible query you can throw at an SQL server. Its designed to handle the types of operations typically seen in large web apps, and to do so quickly and securely. Properly used, FoxFire makes web apps immune to SQL Injection, and eliminates vexing problems like casting between PHP and SQL data types.

FoxFire is also WordPress aware, and will automatically load if placed in the /wp-content/plugins/ folder.

Jump Points

FoxFire has over 540 unit tests providing 100% coverage for every engine, on every platform. Our Razor Test Platform runs on Linux, Windows, and OSX, with zero configuration, "it just works" ...and it can be safely stored to a version control system.

FoxFire is heavily documented in-code (example). For executive-level documentation covering key concepts and design rationale, along with examples and illustrations, see the docs folder.

FoxFire powers Radient, a plugin we're writing that more-or-less implements Flickr + YouTube for WordPress sites with 10K-500K users.

Quick Examples

Create a Table:

class foo extends FOX_db_base {

	var $id;	    // The foo's id
	var $name;	    // Human readable name of the foo
	var $text;	    // Foo text block
	var $priv;	    // Privacy level of the foo
	var $files;	    // Number of files owned by foo instance
	var $space;	    // Disk space used by foo

	// ============================================================================================================ //

	public static $struct_primary = array(

		"table" => "test_foo",
		"engine" => "InnoDB",
		"cache_namespace" => "test_foo",
		"cache_strategy" => "monolithic",
		"cache_engine" => array("memcached", "redis", "apc"),
		"columns" => array(
		    "id" =>	array(	"php"=>"int",	    "sql"=>"smallint",	"format"=>"%d", "width"=>6,	"flags"=>"NOT NULL", "auto_inc"=>true,  "default"=>null,  "index"=>"PRIMARY"),
		    "name" =>	array(	"php"=>"string",    "sql"=>"varchar",	"format"=>"%s", "width"=>250,	"flags"=>"NOT NULL", "auto_inc"=>false, "default"=>null,  "index"=>"UNIQUE"),
		    "text" =>	array(	"php"=>"string",    "sql"=>"varchar",	"format"=>"%s", "width"=>250,	"flags"=>null,	     "auto_inc"=>false, "default"=>null,  "index"=>false),
		    "priv" =>	array(	"php"=>"int",	    "sql"=>"tinyint",	"format"=>"%d", "width"=>2,	"flags"=>"NOT NULL", "auto_inc"=>false, "default"=>0,	"index"=>true),
		    "files" =>	array(	"php"=>"int",	    "sql"=>"mediumint",	"format"=>"%d", "width"=>7,	"flags"=>"NOT NULL", "auto_inc"=>false, "default"=>0,   "index"=>false),
		    "space" =>	array(	"php"=>"float",	    "sql"=>"bigint",	"format"=>"%d", "width"=>null,	"flags"=>"NOT NULL", "auto_inc"=>false, "default"=>0,   "index"=>false)
		 )
	);

	public static function _struct() {
		return self::$struct;
	}
}

$cls = new foo();
$cls->install();

Create Another Table:

class bar extends FOX_db_base {

	var $id;	    // The bar's id
	var $size;	    // Size of the bar
	var $color;	    // Color of the bar
	var $weight;	    // Weight of the bar in grams

	// ============================================================================================================ //

	static $struct_join = array(

		"table" => "test_bar",
		"engine" => "InnoDB",
		"cache_namespace" => "test_bar",
		"cache_strategy" => "monolithic",
		"cache_engine" => array("memcached", "redis", "apc"),
		"columns" => array(
		    "id" =>	array(	"php"=>"int",	    "sql"=>"smallint",	"format"=>"%d", "width"=>6,	"flags"=>"NOT NULL", "auto_inc"=>true,  "default"=>null,  "index"=>"PRIMARY"),
		    "size" =>	array(	"php"=>"string",    "sql"=>"varchar",	"format"=>"%s", "width"=>250,	"flags"=>"NOT NULL", "auto_inc"=>false, "default"=>null,  "index"=>false),
		    "color" =>	array(	"php"=>"string",    "sql"=>"varchar",	"format"=>"%s", "width"=>250,	"flags"=>null,	     "auto_inc"=>false, "default"=>null,  "index"=>false),
		    "weight" =>	array(	"php"=>"float",	    "sql"=>"bigint",	"format"=>"%d", "width"=>10,	"flags"=>"NOT NULL", "auto_inc"=>false, "default"=>0,	  "index"=>true),
		 )
	);

	public static function _struct() {
		return self::$struct;
	}
}

$cls = new bar();
$cls->install();

Load Tables With Data, Using Arrays:

$foo_data = array(
		array( "name"=>"data_01", "text"=>"Test Text String 01", "priv"=>1, "files"=>222, "space"=>3333),
		array( "name"=>"data_02", "text"=>"Test Text String 02", "priv"=>1, "files"=>222, "space"=>4444),
		array( "name"=>"data_03", "text"=>"Test Text String 03", "priv"=>1, "files"=>555, "space"=>6666),
		array( "name"=>"data_04", "text"=>"Test Text String 04", "priv"=>3, "files"=>555, "space"=>6666)
);

$bar_data = array(
		array( "size"=>"data_01", "color"=>"red", "weight"=>11111),
		array( "size"=>"data_02", "color"=>"green", "weight"=>22222)
);

$db = new FOX_db();

$db->runInsertQueryMulti(foo::_struct(), $foo_data, $columns=null, $check=true);
$db->runInsertQueryMulti(bar::_struct(), $bar_data, $columns=null, $check=true);

Load Tables With Data, Using Objects:

$db = new FOX_db();
$bar = new bar();

$bar->size = "data_03";
$bar->color = "blue";
$bar->weight = 33333;

$db->runInsertQuery($bar::_struct(), $bar, $columns=null, $check=true);

Run a Simple Query:

$db = new FOX_db;

$columns = array("mode"=>"exclude", "col"=>"id");
$ctrl=array("format"=>"array_array");

$result = $db->runSelectQuery(foo::_struct(), $args=null, $columns, $ctrl, $check = true);

var_dump($result);

//  $result = array(
//		array( "name"=>"data_01", "text"=>"Test Text String 01", "priv"=>1, "files"=>222, "space"=>3333),
//		array( "name"=>"data_02", "text"=>"Test Text String 02", "priv"=>1, "files"=>222, "space"=>4444),
//		array( "name"=>"data_03", "text"=>"Test Text String 03", "priv"=>1, "files"=>555, "space"=>6666),
//		array( "name"=>"data_04", "text"=>"Test Text String 04", "priv"=>3, "files"=>555, "space"=>6666)
//  );

Run an Advanced Query:

$db = new FOX_db;

$primary = array( "class"=>foo::_struct(), "args"=>array( "col"=>"priv", "op"=>"=", "val"=>1) );

$join = array(
		array(
		    "class"=>bar::_struct(),
		    "on"=>array("pri"=>"name", "op"=>"=", "sec"=>"size"),
		    "args"=>array( "col"=>"weight", "op"=>"=", "val"=>22222 )
		)
);

$columns = array("mode"=>"exclude", "col"=>"id");

$result = $db->runSelectQueryJoin($primary, $join, $columns, $ctrl=array("format"=>"array_array"), $check=true);

var_dump($result);

//  $result = array(
//		array( "name"=>"data_02", "text"=>"Test Text String 02", "priv"=>1, "files"=>222, "space"=>4444)
//  );

About

A fault-tolerant PHP->NoSQL->SQL object relational mapping library

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published