NAME
   JSON::Validator - Validate data against a JSON schema

VERSION
   0.92

SYNOPSIS
     use JSON::Validator;
     my $validator = JSON::Validator->new;

     # Define a schema - http://json-schema.org/examples.html
     # You can also load schema from disk or web
     $validator->schema(
       {
         type       => "object",
         required   => ["firstName", "lastName"],
         properties => {
           firstName => {type => "string"},
           lastName  => {type => "string"},
           age       => {type => "integer", minimum => 0, description => "Age in years"}
         }
       }
     );

     # Validate your data
     @errors = $validator->validate({firstName => "Jan Henning", lastName => "Thorsen", age => -42});

     # Do something if any errors was found
     die "@errors" if @errors;

DESCRIPTION
   JSON::Validator is a class for validating data against JSON schemas. You
   might want to use this instead of JSON::Schema if you need to validate
   data against draft 4
   <https://github.com/json-schema/json-schema/tree/master/draft-04> of the
   specification.

   This module can be used standalone, but if you want to define a
   specification for your webserver's API, then have a look at
   Mojolicious::Plugin::OpenAPI, which will replace
   Mojolicious::Plugin::Swagger2.

 Supported schema formats
   JSON::Validator can load JSON schemas in multiple formats: Plain perl
   data structured (as shown in "SYNOPSIS") or files on disk/web in the
   JSON/YAML format. The JSON parsing is done using Mojo::JSON, while the
   YAML parsing is done with an optional modules which need to be installed
   manually. JSON::Validator will look for the YAML modules in this order:
   YAML::XS, YAML::Syck. The order is set by which module that performs the
   best, so it might change in the future.

 Resources
   Here are some resources that are related to JSON schemas and validation:

   *   <http://json-schema.org/documentation.html>

   *   <http://spacetelescope.github.io/understanding-json-schema/index.htm
       l>

   *   <https://github.com/json-schema/json-schema/>

   *   Swagger2

 Bundled specifications
   This module comes with some JSON specifications bundled, so your
   application don't have to fetch those from the web. These specifications
   should be up to date, but please submit an issue if they are not.

   Note that the bundled specifications can be ignored, by customizing
   "cache_paths".

   * JSON schema, draft 4

     Web page: <http://json-schema.org>

     $ref: <http://json-schema.org/draft-04/schema#>

   * JSON schema for JSONPatch files

     Web page: <http://jsonpatch.com>

     $ref: <http://json.schemastore.org/json-patch#>

   * Swagger / OpenAPI specification, version 2

     Web page: <https://openapis.org>

     $ref: <http://swagger.io/v2/schema.json#>

   * Custom error document

     There is a custom schema used by Mojolicious::Plugin::OpenAPI as a
     default error document. This document might be extended later, but it
     will always be backward compatible.

     Specification:
     <https://github.com/jhthorsen/json-validator/blob/master/lib/JSON/Vali
     dator/cache/630949337805585c8e52deea27d11419>

     $ref: <http://git.io/vcKD4#>.

   * Swagger Petstore

     This is used for unit tests, and should probably not be relied on by
     external users.

ERROR OBJECT
 Overview
   The method "validate" and the function "validate_json" returns error
   objects when the input data violates the "schema". Each of the objects
   looks like this:

     bless {
       message => "Some description",
       path => "/json/path/to/node",
     }, "JSON::Validator::Error"

   See also JSON::Validator::Error.

 Operators
   The error object overloads the following operators:

   *   bool

       Returns a true value.

   *   string

       Returns the "path" and "message" part as a string: "$path:
       $message".

 Special cases
   Have a look at the test suite
   <https://github.com/jhthorsen/json-validator/tree/master/t> for
   documented examples of the error cases. Especially look at "jv-allof.t",
   "jv-anyof.t" and "jv-oneof.t".

   The special cases for "allOf", "anyOf" and "oneOf" will contain the
   error messages from all the failing rules below. It can be a bit hard to
   read, so if the error message is long, then you might want to run a
   smaller test with "JSON_VALIDATOR_DEBUG=1".

   Example error object:

     bless {
       message => "(String is too long: 8/5. String is too short: 8/12)",
       path => "/json/path/to/node",
     }, "JSON::Validator::Error"

   Note that these error messages are subject for change. Any suggestions
   are most welcome!

FUNCTIONS
 validate_json
     use JSON::Validator "validate_json";
     @errors = validate_json $data, $schema;

   This can be useful in web applications:

     @errors = validate_json $c->req->json, "data://main/spec.json";

   See also "validate" and "ERROR OBJECT" for more details.

ATTRIBUTES
 cache_dir
   Deprecated in favor of "cache_paths".

 cache_paths
     $self = $self->cache_paths(\@paths);
     $array_ref = $self->cache_paths;

   Search paths to where cached specifications are stored. Defaults to
   "JSON_VALIDATOR_CACHE_DIR" and the bundled spec files that are shipped
   with this distribution.

     JSON_VALIDATOR_CACHE_DIR=/cache/dir:/some/other/location perl script.pl

   To download a file and add it to the cache, do this:

     $ curl http://swagger.io/v2/schema.json > /cache/dir/$(md5 -qs http://swagger.io/v2/schema.json)

   Files referenced to an URL will automatically be cached if the first
   path in "cache_paths" is writable.

 formats
     $hash_ref = $self->formats;
     $self = $self->formats(\%hash);

   Holds a hash-ref, where the keys are supported JSON type "formats", and
   the values holds a code block which can validate a given format.

   Note! The modules mentioned below are optional.

   *   date-time

       An RFC3339 timestamp in UTC time. This is formatted as
       "YYYY-MM-DDThh:mm:ss.fffZ". The milliseconds portion (".fff") is
       optional

   *   email

       Validated against the RFC5322 spec.

   *   hostname

       Will be validated using Data::Validate::Domain if installed.

   *   ipv4

       Will be validated using Data::Validate::IP if installed or fall back
       to a plain IPv4 IP regex.

   *   ipv6

       Will be validated using Data::Validate::IP if installed.

   *   regex

       EXPERIMENTAL. Will check if the string is a regex, using "qr{...}".

   *   uri

       Validated against the RFC3986 spec.

 resolver
     $code = $self->resolver;
     $self = $self->resolver(sub { my ($self, $namespace, $refs) = @_; });

   Set this to a sub without any logic if you want to skip resolving
   references, like this:

     $self->resolver(sub {});

   This attribute is EXPERIMENTAL.

 ua
     $ua = $self->ua;
     $self = $self->ua(Mojo::UserAgent->new);

   Holds a Mojo::UserAgent object, used by "schema" to load a JSON schema
   from remote location.

   Note that the default Mojo::UserAgent will detect proxy settings and
   have "max_redirects" in Mojo::UserAgent set to 3. (These settings are
   EXPERIMENTAL and might change without a warning)

METHODS
 coerce
     $self = $self->coerce(booleans => 1, numbers => 1, strings => 1);
     $self = $self->coerce({booleans => 1, numbers => 1, strings => 1});
     $self = $self->coerce(1) # enable all
     $hash = $self->coerce;

   Set the given type to coerce. Before enabling coercion this module is
   very strict when it comes to validating types. Example: The string "1"
   is not the same as the number 1, unless you have coercion enabled.

   WARNING! Enabling coercion might hide bugs in your api, which would have
   been detected if you were strict. For example JavaScript is very picky
   on a number being an actual number. This module tries it best to convert
   the data on the fly into the proper value, but this means that you unit
   tests might be ok, but the client side libraries (that care about types)
   might break.

   Loading a YAML document will enable "booleans" automatically. This
   feature is experimental, but was added since YAML has no real concept of
   booleans, such as Mojo::JSON or other JSON parsers.

   The coercion rules are EXPERIMENTAL and will be tighten/loosen if bugs
   are reported. See <https://github.com/jhthorsen/json-validator/issues/8>
   for more details.

 schema
     $self = $self->schema(\%schema);
     $self = $self->schema($url);
     $schema = $self->schema;

   Used to set a schema from either a data structure or a URL.

   $schema will be a Mojo::JSON::Pointer object when loaded, and "undef" by
   default.

   The $url can take many forms, but needs to point to a text file in the
   JSON or YAML format.

   *   http://... or https://...

       A web resource will be fetched using the Mojo::UserAgent, stored in
       "ua".

   *   data://Some::Module/file.name

       This version will use "data_section" in Mojo::Loader to load
       "file.name" from the module "Some::Module".

   *   /path/to/file

       An URL (without a recognized scheme) will be loaded from disk.

 singleton
     $self = $class->singleton;

   Returns the JSON::Validator object used by "validate_json".

 validate
     @errors = $self->validate($data);
     @errors = $self->validate($data, $schema);

   Validates $data against a given JSON "schema". @errors will contain
   validation error objects or be an empty list on success.

   See "ERROR OBJECT" for details.

   $schema is optional, but when specified, it will override schema stored
   in "schema". Example:

     $self->validate({hero => "superwoman"}, {type => "object"});

COPYRIGHT AND LICENSE
   Copyright (C) 2014-2015, Jan Henning Thorsen

   This program is free software, you can redistribute it and/or modify it
   under the terms of the Artistic License version 2.0.

AUTHOR
   Jan Henning Thorsen - "[email protected]"

   Daniel Böhmer - "[email protected]"

   Martin Renvoize - "[email protected]"

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.