README

Provides php tools used by other grithin/php* projects.

Tools

Arrays

# get arbitrarily deep item using a path (like lodash.get)
$user_input = [
	'name'=> [
		'first'=>'bob',
		'last'=>'bobl',
	]
];

$x = Arrays::get($user_input, 'name.first');
#> 'bob'


#+ flatten array picking one value {
$user_input = [
	'name'=> [
		'first'=>'bob',
		'last'=>'bobl',
	]
];

$x = Arrays::flatten_values($user_input);
ppe($x);
#>  {"name": "bob"}
#+ }



#+ flatten structured array {
$user_input = [
	'name'=> [
		'first'=>'bob',
		'last'=>'bobl',
	]
];

$x = Arrays::flatten($user_input);
ppe($x);
/*
{"name_first": "bob",
    "name_last": "bobl"}
*/
#+ }



#+ pick and ensure
$user_input = [
	'first_name'=>'bob',
	'last_name'=>'bobl',
	'unwanted'=>'blah'
];
$pick = ['first_name', 'last_name', 'middle_name'];

$x = Arrays::pick_default($user_input, $pick, 'N/A');
/*
{"first_name": "bob",
    "last_name": "bobl",
    "middle_name": "N\/A"}
*/
#+ }




#+ rekey and exclude {

$user_input = [
	'first_name'=>'bob',
	'last_name'=>'bobl',
	'unwanted'=>'blah'
];
$map = ['first_name'=>'FirstName', 'last_name'=>'LastName'];

$x = Arrays::map_only($user_input, $map);
ppe($x);
/*
{"FirstName": "bob",
    "LastName": "bobl"}
*/
#+ }

Files

The primary functions are file inclusion functions inc, req, incOnce, reqOnce. They allow inclusion tracking, context isolation, variable injection, variable extraction, and global injection.

Example file bob.php

$bill = [$bob]
$bill[] = 'monkey'
return 'blue'

Use

Files::inc('bob.php')
#< 'blue'

# Using variable injection and variable extraction
Files::inc('bob.php',['bob'=>'sue'], ['extract'=>['bill']])
#< ['sue', 'monkey']

Bench

Simple benching utility

$bench = new \Grithin\Bench();
for($i=0; $i<10000; $i++){
	mt_rand(0,100000) + mt_rand(0,100000);
}
$bench->mark();
for($i=0; $i<10000; $i++){
	mt_rand(0,100000) + mt_rand(0,100000);
}
$bench->end_out();

...
   "value": {
        "intervals": [
            {
                "time": 0.0035231113433838,
                "mem.change": 808
            },
            {
                "time": 0.0028860569000244,
                "mem.change": 776
            }
        ],
        "summary": {
            "mem": {
                "start": 403168,
                "end": 404752,
                "diff": 1584
            },
            "time": 0.0064091682434082
        }
    }
...

Memoized

The Memoized trait adds magic methods that interpret the method call, and if the call is prefixed with either memoize_ or memoized_, Memoized will handle memoizing the method.

class Test{
	use \Grithin\Traits\Memoized;

	public function random(){
		return microtime()
	}
}
$Test = new Test;
$x = $Test->memoized_random();
$y = $Test->memoized_random();
$x == $y; # > true

If a memoized function needs to be re-memoized, you can prefix it with memoize_

# ...
$x = $Test->memoized_random();
$y = $Test->memoize_random();
$x == $y; # > false

Memoized will make a key out of the function parameters, so if the function parameters are different, they will generate different memoize cache - and it does this independently of which parameters are used.

# ...
$x = $Test->memoized_random(1);
$y = $Test->memoized_random(2);
$x == $y; # > false

Memoized can be used with static methods

class Test{
	use \Grithin\Traits\Memoized;

	static function random(){
		return microtime();
	}
}
$x = Test::memoized_random();
$y = Test::memoized_random();
$x == $y; # > true

In some cases, it is desirable to know whether a method is currently being memoized. Let's say we have a user_get function that returns data about a user, and within that function there is another location_get function, that returns data about location linked to the user. Let's say there are two scenarios where we ant output from user_get:

code that needs the most up-to-date user_get data
code that can expect user_get calls to be the same data during its run

Memoized provides a way to check whether the current function is within a memoize chain. For instance level memoizing $this->memoizing(); for static memoizing, the Memoized::static_memoizing().

We could have a user_get that looks like

class User{
	use \Grithin\Traits\Memoized;
	public user_get($id){
		$user_data = Db::get('...');
		if($this->memoizing()){ # function has been called with a memoize_ or memoized_ prefix
			$location = Location::memoized_get($user_data['location']);
		}else{
			$location = Location::get($user_data['location']);
		}
	}
}

Finally, a function can check if it was directly called to be memoized by using ->caller_requested_memoized() and ::static_caller_requested_memoized().

Debug

Handling errors and print output

Set up for error handling

# optionally configure
Debug::configure([
	'log_file'=>__DIR__.'/log/'.date('Ymd').'.log',
	'err_file'=>__DIR__.'/log/'.date('Ymd').'.err',	]);

set_error_handler(['\Grithin\Debug','handleError'], E_ALL & ~E_DEPRECATED & ~E_STRICT & ~E_NOTICE);
set_exception_handler(['\Grithin\Debug','handleException']);

Output

Printing out any variable

$bob = ['bob'];
$sue = ['sue'];
$bob[] = &$sue;
$sue[] = &$bob;
$sue[] = 'moe';

Debug::out($sue);

echo 'bob';

Outputs:

{
    "file": "\/test.php",
    "line": 37,
    "i": 1,
    "value": []
}bob

If run within a web server, it will enclose with <pre>

Debug::quit(); will do ::out() then exit.

grithin / phpbase

Maintainers

Details