NAME
Validation::Class::Engine - Data Validation Engine for Validation::Class
VERSION
version 6.00
SYNOPSIS
package MyApp::User;
use Validation::Class;
# a mixin template
mxn 'basic' => {
required => 1
};
# a validation rule
fld 'login' => {
label => 'User Login',
error => 'Login invalid.',
mixin => 'basic',
validation => sub {
my ($self, $this_field, $all_params) = @_;
return $this_field->{value} eq 'admin' ? 1 : 0;
}
};
# a validation rule
fld 'password' => {
label => 'User Password',
error => 'Password invalid.',
mixin => 'basic',
validation => sub {
my ($self, $this_field, $all_params) = @_;
return $this_field->{value} eq 'pass' ? 1 : 0;
}
};
# a validation profile
pro 'registration' => sub {
my ($self, @args) = @_;
return $self->validate(qw(
+name
+email
-login
+password
))
};
# an auto-validating method
mth 'register' => {
input => 'registration',
using => sub {
my ($self, @args) = shift;
# ... do something
}
};
1;DESCRIPTION
Validation::Class::Engine provides data validation functionality and acts as a role applied to Validation::Class.
ATTRIBUTES
directives
The directives attribute returns a hashref of all defined directives.
my $directives = $self->directives();
...errors
The errors attribute returns an arrayref of all errors set.
my $errors = $self->errors();
...fields
The fields attribute returns a hashref of defined fields, filtered and merged with their parameter counterparts.
my $fields = $self->fields();
...filtering
The filtering attribute (by default set to 'pre') controls when incoming data is filtered. Setting this attribute to 'post' will defer filtering until after validation occurs which allows any errors messages to report errors based on the unaltered data. Alternatively, setting the filtering attribute to '' or undef will bypass all filtering unless explicitly defined at the field-level.
my $filtering = $self->filtering('post');
$self->validate();
...filters
The filters attribute returns a hashref of pre-defined filter definitions.
my $filters = $self->filters();
...hash_inflator
The hash_inflator attribute determines how the hash serializer (inflation/deflation) behaves. The value must be a hashref of "OPTIONS" in Hash::Flatten options. Purely for the sake of consistency, you can use lowercase keys (with underscores) which will be converted to camel-cased keys before passed to the serializer.
my $options = $self->hash_inflator({
hash_delimiter => '/',
array_delimiter => '//'
});
...ignore_failure
The ignore_failure boolean determines whether your application will live or die upon failing to validate a self-validating method defined using the method keyword. This is on (1) by default, method validation failures will set errors and can be determined by checking the error stack using one of the error message methods. If turned off, the application will die and confess on failure.
my $ignoring = $self->ignore_failure(1);
...ignore_unknown
The ignore_unknown boolean determines whether your application will live or die upon encountering unregistered field directives during validation. This is off (0) by default, attempts to validate unknown fields WILL cause the program to die.
my $ignoring = $self->ignore_unknown(1);
...methods
The methods attribute returns a hashref of self-validating method definitions.
my $methods = $self->methods(); # definitions are hashrefs
...mixins
The mixins attribute returns a hashref of defined validation templates.
my $mixins = $self->mixins();
...params
The params attribute gets/sets the parameters to be validated. The assigned value MUST be a hashref but can be flat or complex.
my $params = $self->params();
...plugins
The plugins attribute returns a hashref of loaded plugins.
my $plugins = $self->plugins();
...profiles
The profiles attribute returns a hashref of validation profiles.
my $profiles = $self->profiles();
...queued
The queued attribute returns an arrayref of field names for (auto) validation. It represents a list of field names stored to be used in validation later. If the queued attribute contains a list, you can omit arguments to the validate method.
my $queued = $self->queued([qw/.../]);
...relatives
The relatives attribute returns a hashref of short-name/class-name pairs of loaded child classes.
my $relatives = $self->relatives();
...report_failure
The report_failure boolean determines whether your application will report self-validating method failures as class-level errors. This is off (0) by default, if turned on, an error messages will be generated and set at the class-level specifying the method which failed in addition to the existing messages.
my $reporting = $self->report_failure(0);
...report_unknown
The report_unknown boolean determines whether your application will report unregistered fields as class-level errors upon encountering unregistered field directives during validation. This is off (0) by default, attempts to validate unknown fields will NOT be registered as class-level variables.
my $reporting = $self->report_unknown(1);
...METHODS
apply_filters
The apply_filters method (usually called automatically based on the filtering attribute) can be used to run the currently defined parameters through the filters defined in the fields.
my $input = Class->new(filtering => '', params => $params);
if ($input->validate) {
$input->apply_filters; # basically post filtering
}class
The class method returns a new initialize validation class related to the namespace of the calling class, the relative class would've been loaded via the "load" keyword.
Existing parameters and configuration options are passed to the relative class' constructor (including the stash). All attributes can be easily overwritten using the attribute's accessors on the relative class.
Also, you may prevent/override arguments from being copy to the new class object by supplying the them as arguments to this method.
The class method is also quite handy in that it will detect parameters that are prefixed with the name of the class being fetched, and adjust the matching rule (if any) to allow validation to occur.
package Class;
use Validation::Class;
load {
classes => 1 # load child classes e.g. Class::*
};
package main;
my $input = Class->new(params => $params);
my $kid1 = $input->class('Child'); # loads Class::Child;
my $kid2 = $input->class('StepChild'); # loads Class::StepChild;
my $kid3 = $input->class('child'); # loads Class::Child;
my $kid4 = $input->class('step_child'); # loads Class::StepChild;
my $kid5 = $input->class('step-child'); # loads Class::Step::Child;
# intelligently detecting and map params to child class
my $params = {
'my.name' => 'Guy Friday',
'child.name' => 'Guy Friday Jr.'
};
$input->class('child'); # child field *name* mapped to param *child.name*
# without copying params from class
my $kid5 = $input->class('child', params => {}); # .. etc
1;clear_queue
The clear_queue method resets the queue container, see the queue method for more information on queuing fields to be validated. The clear_queue method has yet another useful behavior in that it can assign the values of the queued parameters to the list it is passed, where the values are assigned in the same order queued.
my $input = Class->new(params => $params);
$input->queue(qw(name +email +login +password));
unless ($input->validate) {
return $input->errors_to_string;
}
$input->clear_queue(my($name, $email));
1;clone
The clone method is used to create new fields (rules) from existing fields on-the-fly. This is useful when you have a variable number of parameters being validated that can share existing validation rules. E.g., a web-form on a user's profile page may have dynamically created input boxes for the person's phone numbers allowing the user to add additional parameters to the web-form as needed, in that case as opposed to having multiple validation rules hardcoded for each parameter, you could hardcode one single rule and clone the rule at run-time.
package Class;
use Validation::Class;
field phone => { required => 1 };
package main;
my $input = Class->new(params => $params);
# clone phone rule at run-time to validate dynamically created parameters
$input->clone('phone', 'phone2', { label => 'Other Phone', required => 0 });
$input->clone('phone', 'phone3', { label => 'Third Phone', required => 0 });
$input->clone('phone', 'phone4', { label => 'Forth Phone', required => 0 });
$input->validate(qw/phone phone2 phone3 phone4/);
1;copy_errors
The copy_errors method is used to copy error messages from one class to another. Both classes must be Validation::Class based and/or at-least implement the get_errors and set_errors methods.
$input = Class->new;
$other = $input->class('other');
unless ($other->validate) {
$other->copy_errors($input);
}default_value
The default_value method returns the absolute value (hardcoded, default or parameter specified) for a given field from a hashref of parameters. This method executes specific logic which return the value a field is allowed to have. By default, this method returns undefined.
my $value = $self->default_value('field_name');
# given the following, the $suggested_value will be returned unless the
# field contains rules preventing it from accepting such a value
my $value = $self->default_value('field_name', {
'field_name' => $suggested_value
});error
The error method is used to set and/or retrieve errors encountered during validation. The error method with no parameters returns the error message object which is an arrayref of error messages stored at class-level.
# set errors at the class-level
return $self->error('this isnt cool', 'unknown somethingorother');
# set an error at the field-level, using the field ref (not field name)
$self->error($field_object, "i am your error message");
# return all errors encountered/set as an arrayref
my $all_errors = $self->error();
# return all error for a specific field, ... see the get_errors() method
my @errors = $self->get_errors('field_name');error_count
The error_count method returns the total number of error encountered from the last validation call.
return $self->error_count();
unless ($self->validate) {
print "Found ". $self->error_count ." Errors";
}error_fields
The error_fields method returns a hashref of fields whose value is an arrayref of error messages.
unless ($self->validate) {
my $bad_fields = $self->error_fields();
}
my $bad_fields = $self->error_fields('login', 'password');errors_to_string
The errors_to_string method stringifies the error arrayref object using the specified delimiter or ', ' by default.
return $self->errors_to_string("<br/>\n");
return $self->errors_to_string("<br/>\n", sub{ uc shift });
unless ($self->validate) {
return $self->errors_to_string;
}get_classes
The get_classes method returns the list of instantiated child class objects based on the list of class names specified.
my ($user, $pref) = $self->get_classes('user', 'preference');
if ($user->validate) {
if ($pref->validate) {
...
}
}get_errors
The get_errors method returns the list of class-level error set on the current class or a list of errors from the specified fields.
my @errors = $self->get_errors();
my @lp_errors = $self->get_errors('login', 'password');get_fields
The get_fields method returns the list of references to the specified fields. Returns undef if no arguments are passed. This method is likely to be used more internally than externally.
my ($login, $password) = ($self->get_fields('login', 'password'));get_params
The get_params method returns the values (in list form) of the parameters specified.
if ($self->validate) {
my $name_a = $self->get_params('name');
my ($name_b, $email, $login, $password) =
$self->get_params(qw/name email login password/);
# you should note that if the params dont exist they will return undef
# ... meaning you should check that it exists before checking its value
# e.g.
if (defined $name) {
if ($name eq '') {
print 'name parameter was passed but was empty';
}
}
else {
print 'name parameter was never submitted';
}
}get_params_hash
If your fields and parameters are designed with complex hash structures, the get_params_hash method returns the deserialized hashref of specified parameters based on the the default or custom configuration of the hash serializer Hash::Flatten.
my $params = {
'user.login' => 'member',
'user.password' => 'abc123456'
};
if ($self->validate(keys %$params)) {
my $params = $self->get_params_hash;
print $params->{user}->{login};
}normalize
The normalize method executes a set of routines that reset the parameter environment filtering any parameters present. This method is executed automatically at instantiation and again just before each validation event.
$self->normalize();param
The param method gets/sets a single parameter by name.
my $pass = $self->param('password');
$self->param('password', '******');queue
The queue method is a convenience method used specifically to append the stashed attribute allowing you to *queue* field to be validated. This method also allows you to set fields that must always be validated.
# conditional validation flow WITHOUT the queue method
# imagine a user profile update action
my $input = MyApp::Validation->new(params => $params);
my @fields = qw/name login/;
push @fields, 'email_confirm' if $input->param('chg_email');
push @fields, 'password_confirm' if $input->param('chg_pass');
... if $input->validate(@fields);
# conditional validation WITH the queue method
my $input = MyApp::Validation->new(params => $params);
$input->queue(qw/name login/);
$input->queue(qw/email_confirm/) if $input->param('chg_email');
$input->queue(qw/password_confirm/) if $input->param('chg_pass');
... if $input->validate();
# set fields that must ALWAYS be validated
# imagine a simple REST server
my $input = MyApp::Validation->new(params => $params);
$input->queue(qw/login password/);
if ($request eq '/resource/:id') {
if ($input->validate('id')) {
# validated login, password and id
...
}
}reset
The reset method clears all errors, fields and stashed field names, both at the class and individual field levels.
$self->reset();reset_errors
The reset_errors method clears all errors, both at the class and individual field levels. This method is called automatically every time the validate() method is triggered.
$self->reset_errors();reset_fields
The reset_fields method clears all errors and field values, both at the class and individual field levels. This method is executed automatically at instantiation.
$self->reset_fields();set_errors
The set_errors method pushes its arguments (error messages) onto the class-level error stack of the current class.
my $count = $self->set_errors('Oops', 'OMG', 'WTF');set_method
The set_method method conveniently creates a method on the calling class, this method is primarily intended to be used during instantiation of a plugin during instantiation of the validation class.
my $sub = $self->set_method(__PACKAGE__ => sub { ... });
my $sub = $self->set_method(do_something => sub { ... });Additionally, method names are flattened, e.g. ThisPackage will be converted to this_package for convenience and consistency.
set_params_hash
Depending on how parameters are being input into your application, if your input parameters are already complex hash structures, The set_params_hash method will set and return the serialized version of your hashref based on the the default or custom configuration of the hash serializer Hash::Flatten.
my $params = {
user => {
login => 'member',
password => 'abc123456'
}
};
my $serialized_params = $self->set_params_hash($params);stash
The stash method provides a container for context/instance specific information. The stash is particularly useful when custom validation routines require insight into context/instance specific operations.
package MyApp::Validation;
use Validation::Class;
fld 'email' => {
validation => sub {
my $db = shift->stash->{database};
my $this = shift;
return $db->find('email' => $this->{value}) ? 0 : 1 ; # email exists
}
};
package main;
$self->stash( { database => $dbix_object } );
$self->stash( ftp => $net_ftp, database => $dbix_object );
...validate
The validate method returns true/false depending on whether all specified fields passed validation checks.
use MyApp::Validation;
my $input = MyApp::Validation->new(params => $params);
# validate specific fields
unless ($input->validate('field1','field2')){
return $input->errors_to_string;
}
# validate fields based on a regex pattern
unless ($input->validate(qr/^field(\d+)?/)){
return $input->errors_to_string;
}
# validate existing parameters, if no parameters exist,
# validate all fields ... which will return true unless field(s) exist
# with a required directive
unless ($input->validate()){
return $input->errors_to_string;
}
# validate all fields period, obviously
unless ($input->validate(keys %{$input->fields})){
return $input->errors_to_string;
}
# validate specific parameters (by name) after mapping them to other fields
my $parameter_map = {
user => 'hey_im_not_named_login',
pass => 'password_is_that_really_you'
};
unless ($input->validate($parameter_map)){
return $input->errors_to_string;
}Another cool trick the validate() method can perform is the ability to temporarily alter whether a field is required or not during run-time. This functionality is often referred to as the *toggle* function.
This method is important when you define a field (or two or three) as required or non and want to change that per validation. This is done by calling the validate() method with a list of fields to be validated and prefixing the target fields with a plus or minus as follows:
use MyApp::Validation;
my $input = MyApp::Validation->new(params => $params);
# validate specific fields, force name, email and phone to be required
# regardless of the field definitions directives ... and force the age, sex
# and birthday to be optional
my @spec = qw(+name +email +phone -age -sex -birthday);
unless ($input->validate(@spec)){
return $input->errors_to_string;
}validate_profile
The validate_profile method executes a stored validation profile, it requires a profile name and can be passed additional parameters which get forwarded into the profile routine in the order received.
unless ($self->validate_profile('password_change')) {
die $self->errors_to_string;
}
unless ($self->validate_profile('email_change', $dbi_handle)) {
die $self->errors_to_string;
}DIRECTIVES
package MyApp::Validation;
use Validation::Class;
# a validation template
mixin '...' => {
# mixin directives here
...
};
# a validation rule
field '...' => {
# field directives here
...
};
1;When building a validation class, the first encountered and arguably two most important keyword functions are field() and mixin(), which are used to declare their respective properties. A mixin() declares a validation template where its properties are intended to be copied within field() declarations which declares validation rules, filters and other properties.
Both the field() and mixin() declarations/functions require two parameters, the first being a name, used to identify the declaration and to be matched against incoming input parameters and the second being a hashref of key/value pairs. The key(s) within a declaration are commonly referred to as directives.
The following is a list of default directives which can be used in field/mixin declarations:
alias
The alias directive is useful when many different parameters with different names can be validated using a single rule. E.g. The paging parameters in a webapp may take on different names but require the same validation.
# the alias directive
field 'pager' => {
alias => ['page_user_list', 'page_other_list']
...
};default
The default directive is used as a default value for a field to be used when a matching parameter is not present. Optionally, the default directive can be a coderef allowing the default value to be set on request.
# the default directive
field 'quantity' => {
default => 1,
...
};
# the default directive as a coderef
field 'quantity' => {
default => sub {
return DB::Settings->default_quantity
}
};error/errors
The error/errors directive is used to replace the system generated error messages when a particular field doesn't validate. If a field fails multiple directives, multiple errors will be generated for the same field. This may not be desirable, the error directive overrides this behavior and only the specified error is registered and displayed.
# the error(s) directive
field 'foobar' => {
errors => 'Foobar failed processing, Wtf?',
...
};filtering
The filtering directive is used to control when field filters are applied. The default recognized values are pre/post. A value of 'pre' instructs the validation class to apply the field's filters at instantiation and before validation whereas a value of 'post' instructs the validation class to apply the field's filters after validation. Alternatively, a value of undef or '' will bypass filtering altogether.
# the filtering directive
field 'foobar' => {
filtering => 'post',
...
};label
The label directive is used as a user-friendly reference when the field name is a serialized hash key or just plain ugly.
# the label directive
field 'hashref.foo.bar' => {
label => 'Foo Bar',
...
};mixin
The mixin directive is used to create a template of directives to be applied to other fields.
mixin 'ID' => {
required => 1,
min_length => 1,
max_length => 11
};
# the mixin directive
field 'user.id' => {
mixin => 'ID',
...
};mixin_field
The mixin directive is used to copy all directives from an existing field except for the name, label, and validation directives.
# the mixin_field directive
field 'foobar' => {
label => 'Foo Bar',
required => 1
};
field 'barbaz' => {
mixin_field => 'foobar',
label => 'Bar Baz',
...
};name
The name directive is used *internally* and cannot be changed.
# the name directive
field 'thename' => {
...
};readonly
The readonly directive is used to symbolize a field whose parameter value should not be honored and if encountered, deleted. Unlike the read-only attribute options in other object systems, setting this will not cause you program to die and in-fact, an experience programmer can selectively bypass this constraint.
# the readonly directive
field 'thename' => {
readonly => 1,
...
};value
The value directive is used internally to store the field's matching parameter's value. This value can be set in the definition but SHOULD NOT BE used as a default value unless you're sure no parameter will overwrite it during run-time. If you need to set a default value, see the default directive.
# the value directive
field 'quantity' => {
value => 1,
...
};filters
The filters directive is used to correct, alter and/or format the values of the matching input parameter. Note: Filtering is applied before validation. The filter directive can have multiple filters (even a coderef) in the form of an arrayref of values.
# the filter(s) directive
field 'text' => {
filters => [qw/trim strip/ => sub {
$_[0] =~ s/\D//g;
}],
...
};The following is a list of default filters that may be used with the filter directive:
alpha
The alpha filter removes all non-Alphabetic characters from the field's value.
field 'foobar' => {
filters => 'alpha',
};alphanumeric
The alpha filter removes all non-Alphabetic and non-Numeric characters from the field's value.
field 'foobar' => {
filters => 'alphanumeric',
};capitalize
The capitalize filter attempts to capitalize the first word in each sentence, where sentences are separated by a period and space, within the field's value.
field 'foobar' => {
filters => 'capitalize',
};decimal
The decimal filter removes all non-decimal-based characters from the field's value. Allows only: decimal, comma, and numbers.
field 'foobar' => {
filters => 'decimal',
};lowercase
The lowercase filter converts the field's value to lowercase.
field 'foobar' => {
filters => 'lowercase',
};numeric
The numeric filter removes all non-numeric characters from the field's value.
field 'foobar' => {
filters => 'numeric',
};strip
As with the trim filter the strip filter removes leading and trailing whitespaces from the field's value and additionally removes multiple whitespaces from between the values characters.
field 'foobar' => {
filters => 'strip',
};titlecase
The titlecase filter converts the field's value to titlecase by capitalizing the first letter of each word.
field 'foobar' => {
filters => 'titlecase',
};trim
The trim filter removes leading and trailing whitespace from the field's value.
field 'foobar' => {
filters => 'trim',
};uppercase
The uppercase filter converts the field's value to uppercase.
field 'foobar' => {
filters => 'uppercase',
};validators
package MyApp::Validation;
use Validation::Class;
# a validation rule with validator directives
field 'telephone_number' => {
length => 14,
pattern => '(###) ###-####',
...
};
1;Validator directives are special directives with associated validation code that is used to validate common use cases such as "checking the length of a parameter", etc.
Please note that a directive's value can be either a scalar or arrayref, most directives take a scalar value though some have the ability to take an arrayref or parse a delimited list. There is no official documentation as to which directives can accept lists of values although it will typically be evident.
E.g. (a directive with a list of values)
{
directive => [qw(1 2 3)],
directive => ['1', '2', '3'],
directive => '1, 2, 3',
directive => '1-2-3',
directive => '1,2,3',
}The following is a list of the default validators which can be used in field/mixin declarations:
between
# the between directive
field 'foobar' => {
between => '1-5',
...
};depends_on
# the depends_on directive
field 'change_password' => {
depends_on => ['password', 'password_confirm'],
...
};length
# the length directive
field 'foobar' => {
length => 20,
...
};matches
# the matches directive
field 'this_field' => {
matches => 'another_field',
...
};max_alpha
# the max_alpha directive
field 'password' => {
max_alpha => 30,
...
};max_digits
# the max_digits directive
field 'password' => {
max_digits => 5,
...
};max_length
# the max_length directive
field 'foobar' => {
max_length => '...',
...
};max_sum
# the max_sum directive
field 'vacation_days' => {
max_sum => 5,
...
};max_symbols
# the max_symbols directive
field 'password' => {
max_symbols => 1,
...
};min_alpha
# the min_alpha directive
field 'password' => {
min_alpha => 2,
...
};min_digits
# the min_digits directive
field 'password' => {
min_digits => 1,
...
};min_length
# the min_length directive
field 'foobar' => {
min_length => '...',
...
};min_sum
# the min_sum directive
field 'vacation_days' => {
min_sum => 0,
...
};min_symbols
# the min_symbols directive
field 'password' => {
min_symbols => 0,
...
};options
# the options directive
field 'status' => {
options => 'Active, Inactive',
...
};pattern
# the pattern directive
field 'telephone' => {
# simple pattern
pattern => '### ###-####',
...
};
field 'country_code' => {
# simple pattern
pattern => 'XX',
filter => 'uppercase'
...
};
field 'complex' => {
# regex pattern
pattern => qr/[0-9]+\,\s\.\.\./,
...
};required
The required directive is an important directive but can be misunderstood. The required directive is used to ensure the submitted parameter exists and has a value. If the parameter is never submitted, the directive is effectively skipped. This directive can be thought of as the "must-have-a-value-if-exists" directive.
# the required directive
field 'foobar' => {
required => 1,
...
};
# pass
my $rules = MyApp::Validation->new(params => { });
$rules->validate(); #validate everything
# fail
my $rules = MyApp::Validation->new(params => { foobar => '' });
$rules->validate(); #validate everything
# fail
my $rules = MyApp::Validation->new(params => { });
$rules->validate('foobar');
# fail
my $rules = MyApp::Validation->new(params => { foobar => '' });
$rules->validate('foobar');
# pass
my $rules = MyApp::Validation->new(params => { foobar => 'Nice' });
$rules->validate('foobar');See the toggle functionality within the validate() method. This method allows you to temporarily alter whether a field is required or not.
validation
The validation directive is a coderef used add additional custom validation to the field. The coderef must return true (to pass) or false (to fail). Custom error messages can be set from within the coderef so make sure they are set based on appropriate logic as the registration of error message are not contingent on the success or failure of the routine.
# the validation directive
field 'login' => {
validation => sub {
my ($self, $this_field, $all_params) = @_;
return 0 unless $this_field->{value};
return $this_field->{value} eq 'admin' ? 1 : 0;
},
...
};AUTHOR
Al Newkirk <awncorp@cpan.org>
COPYRIGHT AND LICENSE
This software is copyright (c) 2011 by awncorp.
This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.