Skip to content

A filtering implementation for verifying correct data and performing typical modifications to data.

License

Notifications You must be signed in to change notification settings

dominionenterprises/filter-php

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

#filter-php A filtering implementation for verifying correct data and performing typical modifications to data.

Build Status Scrutinizer Code Quality Code Coverage

Latest Stable Version Total Downloads License

##Features

  • Compact, readable specification
  • Filter with any php callable such as
  • Anonymous function
  • Class function
  • Built-in function
  • Optional/Required support, field and global level
  • Default support
  • Chaining filters
  • Optionally returns unknown fields
  • Filter alias support

##Example

class AppendFilter
{
    public function filter($value, $extraArg)
    {
        return $value . $extraArg;
    }
}
$appendFilter = new AppendFilter();

$trimFunc = function($val) { return trim($val); };

list($status, $result, $error, $unknowns) = DominionEnterprises\Filterer::filter(
    [
        'field one' => [[$trimFunc], ['substr', 0, 3], [[$appendFilter, 'filter'], 'boo']],
        'field two' => ['required' => true, ['floatval']],
        'field three' => ['required' => false, ['float']],
        'field four' => ['required' => true, 'default' => 1, ['uint']],
    ],
    ['field one' => ' abcd', 'field two' => '3.14']
);

var_dump($status);
var_dump($result);
var_dump($error);
var_dump($unknowns);

prints

bool(true)
array(3) {
  'field one' =>
  string(6) "abcboo"
  'field two' =>
  double(3.14)
  'field four' =>
  int(1)
}
NULL
array(0) {
}

##Composer To add the library as a local, per-project dependency use Composer! Simply add a dependency on dominionenterprises/filter to your project's composer.json file such as:

{
    "require": {
        "dominionenterprises/filter": "~1.0"
    }
}

##Documentation Found in the source itself, take a look!

Filterer

At the core of this library is a Filterer class that can validate the structure of an array and map the data through filters. This behavior is defined by a specification of the different filters to apply and some additional options.

Specification

The specification is an array of key => filter specification pairs.

The keys define the known fields in the array. Any fields in the array that are not in the specification are treated as "unknown" fields and may cause validation to fail, depending on the value of the allowUnknowns option.

The filter specification for a single field is also an array. It can contain two special keys:

  • required defines whether this field is a required element of the array. This value overrides the global filter specification's defaultRequired option.
  • default defines what the default value of this field is if none is given. A field with a default value will be guaranteed to be in the result. The required value does not affect default behavior.

The rest of the specification for the field are the filters to apply.

The first element in the filter is the filter to run. This can either be something that passes is_callable (e.g., 'trim' or [$object, 'method']) or it can be one of our predefined aliases (e.g., 'float').

The rest of the elements in the filter are the extra arguments to the filter (the value being filtered is always the first argument).

A filter specification can contain any number of filters and the result of each filter is piped in as the input to the next filter. The result of the final filter is set in the result array.

The example above should help clarify all this.

Included Filters

Of course, any function can potentially be used as a filter, but we include some useful filters with aliases for common circumstances.

Arrays::in

Aliased in the filterer as in, this filter is a wrapper around in_array including support for strict equality testing.

The following does a strict check for $value against the 3 accepted values.

\DominionEnterprises\Filter\Arrays::in($value, ['a', 'b', 'c']);

Arrays::filter

Aliased in the filterer as array, this filter verifies that the argument is an array and checks the length of the array against bounds. The default bounds are 1+, so an empty array fails by default.

The following checks that the $value is an array with exactly 3 elements.

\DominionEnterprises\Filter\Arrays::filter($value, 3, 3);

Arrays::ofScalars

Aliased in the filterer as ofScalars, this filter verifies that the argument is an array (possibly empty) of scalar items that each pass the given filters (given in the same format as used by Filterer::filter.

The following checks that $value is an array of unsigned integers.

$value = \DominionEnterprises\Filter\Arrays::ofScalars($value, [['uint']]);

Arrays::ofArrays

Aliased in the filterer as ofArrays, this filter verifies that the argument is an array (possibly empty) of arrays that each pass the given filters (given in the same format as used by Filterer::filter.

The following checks that $value is an array of items that each have an id key with a numeric value. No other keys would be allowed. For example, the following is valid input: [['id' => '1'], ['id' => '2']].

$value = \DominionEnterprises\Filter\Arrays::ofArrays($value, ['id' => [['uint']]]);

Arrays::ofArray

Aliased in the filterer as ofArray, this filter verifies that the argument is an array that passes the given specification. This is essentially a flipped version of Filterer::filter that allows for testing nested associative arrays.

Arrays::flatten

Aliased in the filterer as flatten, this filter flattens a multi-dimensional array to a single dimension. The order of values will be maintained, but the keys themselves will not. For example:

$value = \DominionEnterprises\Filter\Arrays::flatten([[1, 2], [3, [4, 5]]]);
assert($value === [1, 2, 3, 4, 5]);

Booleans::filter

Aliased in the filterer as bool, this filter verifies that the argument is a boolean value or a string that maps to one. The second parameter can be set to true to allow null values through without an error (they will stay null and not get converted to false). The last parameters are lists of strings for true values and false values. By default, the strings "true" and "false" map to their boolean counterparts.

The following example converts $value to a boolean allowing the strings "on" and "of".

$enabled = \DominionEnterprises\Filter\Booleans::filter($value, false, ['on'], ['off']);

Booleans::convert

Aliased in the filterer as bool-convert, this filter will convert a given boolean value into the provided true or false conditions. By default the return values are the strings 'true' and 'false'

The following converts the boolean $value to either 'yes' or 'no'

$answer = \DominionEnterprises\Filter\Booleans::convert($value, 'yes', 'no');

Floats/Ints/UnsignedInt::filter

Aliased in the filterer as float, int, and uint, respectively, these filters verify that the arguments are of the proper numeric type and allow for bounds checking. The second parameter to each of them can be set to true to allow null values through without an error (they will stay null and not get converted to false). The next two parameters are the min and max bounds and can be used to limit the domain of allowed numbers.

Non-numeric strings will fail validation, and numeric strings will be cast.

The float parameter has an additional parameter that can be set to true to cast integers to floats. Without this, integers will fail validation.

The following checks that $value is an integer between 1 and 100 inclusive, and returns the integer (after casting it if it was a string).

$value = \DominionEnterprises\Filter\UnsignedInt::filter($value, false, 1, 100);

Strings::filter

Aliased in the filterer as string, this filter verifies that the argument is a string. The second parameter can be set to true to allow null values through without an error (they will stay null and not get converted to false). The last parameters specify the length bounds of the string. The default bounds are 1+, so an empty string fails by default.

The following checks that $value is a non-empty string.

\DominionEnterprises\Filter\Strings::filter($value);

Strings::explode

Aliased in the filterer as explode, this filter is essentially a wrapper around the built-in explode method with the value first in order to work with the Filterer. It also defaults to using , as a delimiter. For example:

$value = \DominionEnterprises\Filter\Strings::explode('abc,def,ghi');
assert($value === ['abc', 'def', 'ghi']);

Url::filter

Aliased in the filterer as url, this filter verifies that the argument is a URL string according to RFC2396. The second parameter can be set to true to allow null values through without an error (they will stay null and not get converted to false).

The following checks that $value is a URL.

\DominionEnterprises\Filter\Url::filter($value);

Email::filter

Aliased in the filterer as email, this filter verifies that the argument is an email.

The following checks that $value is an email.

\DominionEnterprises\Filter\Email::filter($value);

DateTime::filter

Aliased in the filterer as date, this will filter the value as a \DateTime object. The value can be any string that conforms to PHP's valid date/time formats

The following checks that $value is a date/time.

$dateTime = \DominionEnterprises\Filter\DateTime::filter('2014-02-04T11:55:00-0500');

DateTime::format

Aliased in the filterer as date-format, this will filter a given `\DateTime' value to a string based on the given format.

The following returns formatted string for a given \DateTime $value

$formatted = \DominionEnterprises\Filter\DateTime::format($value, 'Y-m-d H:i:s');

DateTimeZone::filter

Aliased in the filterer as date, this will filter the value as a \DateTimeZone object. The value can be any supported timezone name

The following checks that $value is a timezone

$timezone = \DominionEnterprises\Filter\DateTimeZone::filter('America/New_York');

##Contact Developers may be contacted at:

##Project Build With a checkout of the code get Composer in your PATH and run:

./build.php

There is also a docker-based fig configuration that will execute the build inside a docker container. This is an easy way to build the application:

fig run build

For more information on our build process, read through out our Contribution Guidelines.

About

A filtering implementation for verifying correct data and performing typical modifications to data.

Resources

License

Stars

Watchers

Forks

Packages

No packages published

Languages

  • PHP 100.0%