The Perl Toolchain Summit needs more sponsors. If your company depends on Perl, please support this very important event.

NAME

Validation::Class::Cookbook - Recipes for Validation::Class

VERSION

version 7.900054

GUIDED TOUR

The instructions contained in this documentation are also relevant for configuring any class derived from Validation::Class. The validation logic that follows is not specific to a particular use-case.

Parameter Handling

There are three ways to declare parameters you wish to have validated. The first and most common approach is to supply the target parameters to the validation class constructor:

    use Validation::Class::Simple;

    my $rules = Validation::Class::Simple->new(params => $params);

All input parameters are wrapped by the Validation::Class::Params container which provides generic functionality for managing hashes. Additionally you can declare parameters by using the params object directly:

    use Validation::Class::Simple;

    my $rules = Validation::Class::Simple->new;

    $rules->params->clear;

    $rules->params->add(user => 'admin', pass => 's3cret');

    printf "%s parameters were submitted", $rules->params->count;

Finally, any parameter which has corresponding validation rules that has been declared in a validation class derived from Validation::Class will have an accessor which can be used directly or as an argument to the constructor:

    package MyApp::Person;

    use Validation::Class;

    field 'name' => {
        required => 1
    };

    package main;

    my $rules = MyApp::Person->new(name => 'Egon Spangler');

    $rules->name('Egon Spengler');

Validation Rules

Validation::Class comes with a complete standard set of validation rules which allows you to easily describe the constraints and operations that need to be performed per parameter.

Validation rules are referred to as fields, fields are named after the parameters they expect to be matched against. A field is also a hashref whose keys are called directives which correspond with the names of classes in the directives namespace, and whose values are arguments which control how directives carry-out their operations.

    use Validation::Class::Simple;

    my $rules = Validation::Class::Simple->new;

    $rules->fields->clear;

    $rules->fields->add(name => { required => 1, max_length => 255 });

Fields can be specified as an argument to the class constructor, or managed directly using the Validation::Class::Fields container. Every field is wrapped by the Validation::Class::Field container which provides accessors for all core directives. Directives can be found under the directives namespace, e.g. the required directive refers to Validation::Class::Directive::Required. Please see Validation::Class::Directives for a list of all core directives.

Flow Control

A good data validation tool is not simply checking input against constraints, its also providing a means to easily handle different and often complex data input scenarios.

The queue method allows you to designate and defer fields to be validated. It also allows you to set fields that must be validated regardless of what has been passed to the validate method. Additionally it allows you to conditionally specify constraints:

    use Validation::Class::Simple;

    my $rules = Validation::Class::Simple->new;

    $rules->queue('name'); # always validate the name parameter

    $rules->queue('email', 'email2') if $rules->param('change_email');
    $rules->queue('login', 'login2') if $rules->param('change_login');

    # validate name
    # validate email and email confirmation if change_email is true
    # validate login and login confirmation if change_login is true

    $rules->validate('password'); # additionally, validate password
    $rules->clear_queue;          # reset the queue when finished

Akin to the queue method is the stash method. At-times it is necessary to break out of the box in order to design constraints that fit your particular use-case. The stash method allows you to share arbitrary objects with routines used by validation classes.

    use Validation::Class::Simple;

    my $rules = Validation::Class::Simple->new;

    $rules->fields->add(
        email => {
            # email validation relies on a stashed object
            validation => sub {
                my ($self, $field, $params) = @_;
                return 0 if ! my $dbo = $self->stash('dbo');
                return 0 if ! $dbo->email_exists($field->value);
                return 1;
            }
        }
    );

    # elsewhere in the program
    $rules->stash(dbo => $database_object); # stash the database object

Error Handling

When validation fails, and it will, you need to be able to report what failed and why. Validation::Class give you complete control over error handling and messages. Errors can exist at the field-level and class-level (errors not specific to a particular field). All errors are wrapped in a Validation::Class::Errors container.

    use Validation::Class::Simple;

    my $rules = Validation::Class::Simple->new;

    # print a comma separated list of class and field errors
    print $rules->errors_to_string unless $rules->validate;

    # print a newline separated list of class and field errors
    print $rules->errors_to_string("\n") unless $rules->validate;

    # print a comma separated list of class and upper-cased field errors
    print $rules->errors_to_string(undef, sub{ ucfirst lc shift })

    # print total number of errors at the class and field levels
    print "Found %s errors", $rules->error_count;

    # return a hashref of fields with errors
    my $errors = $rules->error_fields;

    # get errors for specific fields only
    my @errors = $rules->get_errors('email', 'login');

Input Filtering

Filtering data is one fringe benefits of a good data validation framework. The process is also known as scrubbing or sanitizing data. The process ensures that the data being passed to the business logic will be clean and consistent.

Filtering data is not as simple and straight-forward as it may seem which is why it is necessary to think-through your applications interactions before implementation.

Filtering is the process of applying transformations to the incoming data. The problem with filtering is that it permanently alters the data input and in the event of a failure could report inconsistent error messages:

    use Validation::Class::Simple;

    my $rules = Validation::Class::Simple->new;

    $rules->fields->add(
        # even if the input is submitted as lowercase it will fail
        # the filter is run as a pre-process by default
        username => {
            filters => ['uppercase'],
            validation => sub {
                return 0 if $_[1]->value =~ /[A-Z]/;
                return 1;
            }
        }
    );

When designing a system to filter data, it is always necessary to differentiate pre-processing filters from post-processing filters. Validation::Class provides a filtering directive which designates certain fields to run filters in post-processing:

    $rules->fields->add(
        # if the input is submitted as lowercase it will pass
        username => {
            filters => ['uppercase'],
            filtering => 'post',
            validation => sub {
                return 0 if $_[1]->value =~ /[A-Z]/;
                return 1;
            }
        }
    );

Handling Failures

A data validation framework exists to handle failures, it is its main function and purpose, in-fact, the difference between a validation framework and a type-constraint system is how it responds to errors.

When a type-constraint system finds an error it raises an exception. Exception handling is the process of responding to the occurrence, during computation, of exceptions (anomalous or exceptional situations).

Typically the errors reported when an exception is raised includes a dump of the program's state up until the point of the exception which is apropos as exceptions are unexpected.

A data validation framework can also be thought-of as a type system but one that is specifically designed to expect input errors and report user-friendly error messages.

Validation::Class may encounter exceptions as programmers defined validation rules which remain mutable. Validation::Class provides attributes for determining how the validation engine reacts to exceptions and validation failures:

    use Validation::Class::Simple;

    my $rules = Validation::Class::Simple->new(
        ignore_failure => 1, # do not throw errors if validation fails
        ignore_unknown => 0, # throw errors if unknown directives are found
        report_failure => 0, # register errors if "method validations" fail
        report_unknown => 0, # register errors if "unknown directives" are found
    );

Data Validation

Once your fields are defined and you have your parameter rules configured as desired you will like use the validate method to perform all required operations. The validation operations occur in the following order:

    normalization   (resetting fields, clearing existing errors, etc)
    pre-processing  (applying filters, etc)
    validation      (processing directives, etc)
    post-processing (applying filters, etc)

What gets validated is determined by the state and arguments passed to the validate method. The validate method determines what to validate in the following order:

    checks the validation queue for fields
    checks arguments for regular expression objects and adds matching fields
    validates fields with matching parameters if no fields are specified
    validates all fields if no parameters are specified

It is also important to under what it means to declare a field as being required. A field is a data validation rule matching a specific parameter, A required field simply means that if-and-when a parameter is submitted, it is required to have a value. It does not mean that a field is always required to be validated.

Occasionally you may need to temporarily set a field as required or not-required for a specific validation operation. This requirement is referred to as the toggle function. The toggle function is enacted by prefixing a field name with a plus or minus sign (+|-) when passed to the validate method:

    use Validation::Class::Simple;

    my $rules = Validation::Class::Simple->new(fields => {...});

    # meaning, email is always required to have a value
    # however password and password2 can be submitted as empty strings
    # but if password and password2 have values they will be validated
    $rules->validate('+email', '-password', '-password2');

Here are a few examples and explanations of using the validate method:

    use Validation::Class::Simple;

    my $rules = Validation::Class::Simple->new(fields => {...});

    unless ($rules->validate) {
        # validate all fields with matching parameters
    }

    unless ($rules->validate) {
        # validate all fields because no parameters were submitted
    }

    unless ($rules->validate(qr/^email/)) {
        # validate all fields whose name being with email
        # e.g. email, email2, email_update
    }

    unless ($rules->validate('login', 'password')) {
        # validate the login and password specifically
        # regardless of what parameters have been set
    }

    unless ($rules->validate({ user => 'login', pass => 'password' })) {
        # map user and pass parameters to the appropriate fields as aliases
        # and validate login and password fields using the aliases
    }

BUILDING CLASSES

This recipe displays the usage of keywords to configure a validation class.

Problem

You want to know how to use the Validation::Class keywords to define a validation class.

Solution

Use the keywords exported by Validation::Class to register validation rules, templates, profiles, methods and filters.

Discussion

Your validation class can be thought of as your data-model/input-firewall. The benefits this approach provides might require you to change your perspective on parameter handling and workflow. Typically when designing an application we tend to name parameters arbitrarily and validate the same data at various stages during a program's execution in various places in the application stack. This approach is inefficient and prone to bugs and security problems.

To get the most out of Validation::Class you should consider each parameter hitting your application (individually) as a transmission fitting a very specific criteria, yes, like a field in a data model.

Your validation rules will act as filters which will reject or accept and format the transmission for use within your application, yes, almost exactly like a firewall.

A validation class is defined as follows:

    package MyApp::Person;

    use Validation::Class;

    # a validation rule template

    mixin 'basic'  => {
        required   => 1,
        min_length => 1,
        max_length => 255,
        filters    => ['lowercase', 'alphanumeric']
    };

    # a validation rule

    field 'login'  => {
        mixin      => 'basic',
        label      => 'user login',
        error      => 'login invalid',
        validation => sub {

            my ($self, $field, $params) = @_;

            return $field->value eq 'admin' ? 1 : 0;

        }
    };

    # a validation rule

    field 'password'  => {
        mixin         => 'basic',
        label         => 'user password',
        error         => 'password invalid',
        validation    => sub {

            my ($self, $field, $params) = @_;

            return $field->value eq 'pass' ? 1 : 0;

        }
    };

    # a validation profile

    profile 'registration'  => sub {

        my ($self, @args) = @_;

        return $self->validate(qw(login password));

    };

    # an auto-validating method

    method 'registers'  => {

        input => 'registration',
        using => sub {

            my ($self, @args) = shift;

            # ... do something

        }

    };

    1;

The fields defined will be used to validate the specified input parameters. You specify the input parameters at/after instantiation, parameters should take the form of a hashref of key/value pairs passed to the params attribute, or attribute/value pairs. The following is an example on using your validate class to validate input in various scenarios:

    # web app
    package MyApp;

    use MyApp::User;
    use Misc::WebAppFramework;

    get '/auth' => sub {

        # get user input parameters
        my $params = shift;

        # initialize validation class and set input parameters
        my $user = MyApp::User->new(params => $params);

        unless ($user->registers) {

            # print errors to browser unless validation is successful
            return $user->errors_to_string;

        }

        return 'you have authenticated';

    };

A field can have aliases, parameter names that if detected will be mapped to the parameter name matching the field definition. Multiple fields cannot have the same alias defined, such a configuration would result in a runtime error.

    use MyApp::User;

    my $user = MyApp::User->new(params => $params);

    unless ($user->validate) {

        return $input->errors_to_string;

    }

    package MyApp::User;

    field 'email' => {
        ...,
        alias => [
            'emails',
            'email_address',
            'email_addresses'
        ]

    };

    package main;

    use MyApp::User;

    my  $user = MyApp::User->new(params => { email_address => '...' });

    unless ($user->validate('email'){

        return $user->errors_to_string;

    }

    # valid because email_address is an alias on the email field

INTEGRATING CLASSES AND FRAMEWORKS

This recipe displays methods of configuring your validation class to cooperate with your pre-existing classes and object-system.

Problem

You want to know how to configure Validation::Class to cooperate with pre-existing classes or object systems like Mo, Moo, Mouse, and Moose.

Solution

Use a combination of techniques such as excluding keywords exported by Validation::Class and utilizing the initialize_validator method.

Discussion

Validation::Class will atuomatically inject a method name `initialize_validator` if a pre-existing `new` method is dicovered which allows you to execute certain validation class normalization routines. When, the initialize_validator method is called is not important, it is only important that it is called before your object is used as a validation class object.

A validation class using Moose as an object system could be configured as follows:

    package MyApp::Person;

    use Moose;
    use Validation::Class qw(fld mxn);

    # the order in which these frameworks are used is important
    # loading Moose first ensures that the Moose::Object constructor
    # has precendence

    sub BUILD {

        my ($self, $params) = @_;

        $self->initialize_validator($params);

    }

    mxn 'basic'  => {
        required   => 1,
        min_length => 1,
        max_length => 255,
        filters    => ['lowercase', 'alphanumeric']
    };

    fld 'login'  => {
        mixin => 'basic',
        label => 'user login',
        error => 'login invalid'
    };

    fld 'password'  => {
        mixin => 'basic',
        label => 'user password',
        error => 'password invalid'
    };

    has 'profile' => (
        is  => 'rw',
        isa => 'MyApp::Person::Profile'
    );

    1;

FILTERING DATA

This recipe describes how to define filtering in your validation class rules.

Problem

You want to know how to define filters to sanatize and transform your data although some transformations may need to occur after a successful validation.

Solution

Data validation rules can be configured to apply filtering as both pre-and-post processing operations.

Discussion

Validation::Class supports pre/post filtering but is configured to pre-filter incoming data by default. This means that based upon the filtering options supplied within the individual fields, filtering will happen before validation (technically at instantiation and again just before validation). As expected, this is configurable via the filtering attribute.

A WORD OF CAUTION: Validation::Class is configured to pre-filter incoming data which boosts application security and is best used with passive filtering (e.g. converting character case - filtering which only alters the input in predictable ways), versus aggressive filtering (e.g. formatting a telephone number) which completely and permanently changes the incoming data ... so much so that if the validation still fails ... errors that are reported may not match the data that was submitted.

If you're sure you'd rather employ aggressive filtering, I suggest setting the filtering attribute to 'post' for post-filtering or setting it to null and applying the filters manually by calling the apply_filters() method.

DELEGATING VALIDATION

This recipe describes how to separate validation logic between multiple related classes.

Problem

You want to know how to define multiple validation classes and pass input data and input parameters between them.

Solution

Use classes as validation domains, as a space to logically group related validation rules, then use built-in methods to have multiple validation classes validate in-concert.

Discussion

For larger applications where a single validation class might become cluttered and inefficient, Validation::Class comes equipped to help you separate your validation rules into separate classes.

The idea is that you'll end up with a main validation class (most likely empty) that will simply serve as your point of entry into your relative (child) classes. The following is an example of this:

    package MyApp::User;

    use Validation::Class;

    field name      => { ... };
    field email     => { ... };
    field login     => { ... };
    field password  => { ... };

    package MyApp::Profile;

    use Validation::Class;

    field age       => { ... };
    field sex       => { ... };
    field birthday  => { ... };

    package MyApp;

    use Validation::Class;

    set classes => 1;

    package main;

    my $input = MyApp->new(params => $params);

    my $user = $input->class('user');

    my $profile = $input->class('profile');

    1;

INTROSPECT AND EXTEND

This recipe describes how to peek under the curtain and leverage the framework for other purposes.

Problem

You want to know how to use your data validation classes to perform other tasks programatically (e.g. generate documentation, etc).

Solution

By using the prototype class associated with your validation class you can introspect it's configuration and perform additional tasks programatically.

Discussion

Most users will never venture beyond the public API, but powerful abilities await the more adventureous developer and this section was written specifically for you. To assist you on along your journey, let me explain exactly what happens when you define and instantiate a validation class.

Classes are defined using keywords (field, mixin, filter, etc) which register rule definitions on a cached class profile (of-sorts) associated with the class which is being constructed. On instantiation, the cached class profile is cloned then merged with any arguments provided to the constructor, this means that even in a persistent environment the original class profile is never altered.

To begin introspection, simply look into the attributes attached to the class prototype, e.g. fields, mixins, filters, etc., the following examples will give you an idea of how to use introspection to extend your application code using Validation::Class.

Please keep in mind that Validation::Class is likely to already have most of the functionalty you would need to introspect your codebase. The following is an introspection design template that will work in most cases:

    package MyApp::Introspect;

    use Validation::Class;

    load classes => 'MyApp'; # load MyApp and all child classes

    sub per_class {

        my ($self, $code) = @_;

        my %relatives = %{$self->proto->settings->{relatives}};

        while (my($parent, $children) =  each(%relatives)) {

            while (my($nickname, $namespace) = each(%{$children})) {

                # do something with each class
                $code->($namespace);

            }

        }

    }

    sub per_field_per_class {

        my ($self, $code) = @_;

        $self->per_class(sub{

            my $namespace = shift;

            my $class = $namespace->new;

            foreach my $field ($class->fields->values) {

                # do something with each field in each class
                $code->($class, $class->fields->{$field});

            }

        });

    }

CLIENT-SIDE VALIDATION

This recipe describes how to generate JSON objects which can be used to validate user input in the web-browser (client-side).

Problem

You want to know how to make the most out of your data validation rules by making your configuration available as JSON objects in the browser.

Solution

Using introspection, you can leverage the prototype class associated with your validation class to generate JSON objects based on your validation class configuration.

Discussion

In the context of a web-application, it is often best to perform the initial input validation on the client (web-browser) before submitting data to the server for further validation and processing. In the following code we will generate javascript objects that match our Validation::Class data models which we will then use with some js library to validate form data, etc.

... example validation class

    package MyApp::Model;

    use Validation::Class;
    use Validation::Class::Plugin::JavascriptObjects;

    mxn scrub => {
        filters => ['trim', 'strip']
    };

    fld login => {
        mixin    => 'scrub'
        email    => 1,
        required => 1,
        alias    => 'user',
    };

    fld password    => {
        mixin       => 'scrub',
        required    => 1,
        alias       => 'pass',
        min_length  => 5,
        min_symbols => 1,
        min_alpha   => 1,
        min_digits  => 1
    };

... in your webapp controller

    get '/js/model'   => sub {

        my $model     = MyApp::Model->new;

        # generate the JS object
        my $data = $model->plugin('javascript_objects')->render(
            namespace => 'validate.model',
            fields    => [qw/email password/],
            include   => [qw/required email minlength maxlength/]
        )

        return print $data;

    };

The output of the /js/model route should generate a javascript object which looks similar to the following:

    var validate = {
        "model" : {
            "email" : {
               "minlength" : 3,
               "required" : 1,
               "maxlength" : 255
            },
            "password" : {
               "minlength" : 5,
               "required" : 1,
               "maxlength" : 255
            }
        }
    };

If its not obvious yet, we can now easily use this generated javascript API with jQuery (or other client-side library) to validate form data, etc.

    <!DOCTYPE html>
    <html>
        <head>
            <title>AUTH REQUIRED</title>
            <script type="text/javascript" src="/js/jquery.js"></script>
            <script type="text/javascript" src="/js/jquery.validate.js"></script>
            <script type="text/javascript" src="/js/model"></script>
            <script type="text/javascript">
                $(document).ready(function() {
                    $("#form").validate({rules:validate.model});
                });
            </script>
        </head>
        <body>
            <div>[% input.errors_to_string %]</div>
            <form id="form" autocomplete="off" method="post" action="/">
            <fieldset>
                <legend><h2><strong>Halt</strong>, who goes there?</h2></legend>
                <label for="email">Email</label><br/>
                <input id="email" name="email" value="" /><br/>
                <label for="password">Password</label><br/>
                <input id="password" name="password" type="password" /><br/>
                <br/><input type="submit" value="Submit" />
            </fieldset>
            </form>
        </body>
    </html>

AUTHOR

Al Newkirk <anewkirk@ana.io>

COPYRIGHT AND LICENSE

This software is copyright (c) 2011 by Al Newkirk.

This is free software; you can redistribute it and/or modify it under the same terms as the Perl 5 programming language system itself.