OvO is a Dart-first schema declaration and validation library.
OvO is a Dart-first schema declaration and validation library. We use the technical term "Schema" to define any data type, from simple single data (for example: string/int, etc.) to complex nested Map.
OvO is designed to be as user-friendly and developer-friendly as possible, with the goal of eliminating tedious type checking and object deserialization. It is easy to compose complex data structure validation using simple declaration validation.
several important aspects
- A fun walkthrough of Dart type extensions
- Simple and chained interface calls
- Can be used on any Dart platform (Dart, Web, Flutter)
I am very grateful and encouraged for any level of sponsorship, which will help me continue to develop and maintain this project.
We are more aggressive and use higher versions of Dart stable versions as much as possible.
# Dart project
dart pub add ovo
# Flutter project
flutter pub add ovodependencies:
ovo: latestCreate a simple string schema:
import 'package:ovo/ovo.dart' as ovo;
// Create a schema for string.
final schema = ovo.String();
// Parsing
await schema.parse('Hello World'); // => 'Hello World'
await schema.parse(123); // => throws OvOExceptionCreating an JSON schema:
import 'package:ovo/ovo.dart' as ovo;
final schema = ovo.Object({
'name': ovo.String(),
});
await schema.parse({
'name': 'John',
}); // => {'name': 'John'}OvO provides type validation with dependent type parameters, and also built-in some common types. You can declare a type validation by OvO<T>, where T is a type parameter and can be any type.
Let's try to create a String type validation:
import 'package:ovo/ovo.dart' as ovo;
final schema = ovo.OvO<String>();
await schema.parse('Hello World'); // => 'Hello World'
await schema.parse(123); // => throws OvOExceptionOr, we create a validation of Record type that is not built-in:
import 'package:ovo/ovo.dart' as ovo;
final schema = ovo.OvO<(int, String)>();
await schema.parse((123, 'Hello World')); // => (123, 'Hello World')Of course, you can also use it to validate a custom class:
import 'package:ovo/ovo.dart' as ovo;
class User {
final String name;
final int age;
User(this.name, this.age);
}
final schema = ovo.OvO<User>();
await schema.parse(User('John', 18)); // => User('John', 18)Basic type validation depends on the built-in is keyword in Dart, which is fully capable of most type validation. However, if you need more complex type validation, you can use the constructor of OvO<T> to create a custom type validation.
import 'package:ovo/ovo.dart' as ovo;
class User {
final String name;
final int age;
User(this.name, this.age);
factory User.fromJson(Map<String, dynamic> json) {
return User(json['name'], json['age']);
}
}
class MyOvO implements ovo.OvO<User> {
@override
Future<User> parse(ovo.Context, dynamic value) {
if (value is Map<String, dynamic>) {
return User.fromJson(value);
}
throw ovo.OvOException('Invalid value');
}
}
final schema = ovo.Object({
'data': MyOvO(),
'status': ovo.Boolean(),
});
final data = {
'data': {
'name': 'John',
'age': 18,
},
'status': true,
};
await schema.parse(data); // => {'data': User('John', 18), 'status': true}Any type validation can accept a non-null value of any type. It is an alias of OvO<Object>.
Array type validation can accept an iterable value (for example: List, Set, Iterable, etc.). It is not an alias of OvO<Iterable<T>>, but a specific type validation.
Array accepts a parameter of type OvO<T> to validate each element in the array.
import 'package:ovo/ovo.dart' as ovo;
final schema = ovo.Array(ovo.String());
await schema.parse(['Hello', 'World']); // => ['Hello', 'World']
await schema.parse([123, 456]); // => throws OvOException.min/.max/.size have the same parameters and type signatures, and they all accept a parameter of type int to validate the length of the array.
.minvalidates that the length of the array must be greater than or equal to the specified length..maxvalidates that the length of the array must be less than or equal to the specified length..sizevalidates that the length of the array must be equal to the specified length.
ovo.Array(ovo.String()).min(2); // must contain at least 2 elements
ovo.Array(ovo.String()).max(2); // must contain no more than 2 elements
ovo.Array(ovo.String()).size(2); // must contain exactly 2 elements.unique validates that the elements in the array must be unique, similar to Set in Dart.
ovo.Array(ovo.String()).unique(); // must contain unique elementsBoolean type validation can accept a value of type bool. It is an alias of OvO<bool>.
另外，他还有两个额外的扩展方法：
In addition, it has two additional extension methods:
-
.isTrue- 验证值必须为true。 -
.isFalse- 验证值必须为false。 -
.isTrue- validates that the value must betrue. -
.isFalse- validates that the value must befalse.
ovo.Boolean(); // must be a boolean, `true` or `false`
ovo.Boolean().isTrue(); // must be true
ovo.Boolean().isFalse(); // must be falseOf course, you can use OvO<bool> instead of Boolean type validation, but Boolean type validation is more semantic.
Number type validation can accept a value of type num. It is an alias of OvO<num>.
ovo.Number(); // must be a number, `int` or `double`If you need to validate an integer, you can use the Integer type validation, which is an alias of OvO<int>. To validate a floating-point number, you can use the Double type validation, which is an alias of OvO<double>.
Integer and Double are both subtypes of Number:
ovo.Integer(); // must be an integer, `int`
ovo.Double(); // must be a double, `double`OvO<num> also contains some additional methods:
ovo.Number().gt(10); // must be greater than 10
ovo.Number().gte(10); // must be greater than or equal to 10
ovo.Number().lt(10); // must be less than 10
ovo.Number().lte(10); // must be less than or equal to 10
ovo.Number().finite(); // must be finite
ovo.Number().negative(); // must be negativeString type validation can accept a value of type String. It is an alias of OvO<String>.
ovo.String(); // must be a string, `String`OvO<String> also contains some additional methods:
// Validations
ovo.String().min(5); // must be at least 5 characters long
ovo.String().max(5); // must be no more than 5 characters long
ovo.String().length(5); // must be exactly 5 characters long
ovo.String().regex(RegExp(r'^[a-z]+$')); // must match the regular expression
ovo.String().contains('abc'); // must contain the substring
ovo.String().isNotEmpty(); // must not be empty
ovo.String().startsWith('abc'); // must start with the substring
ovo.String().endsWith('abc'); // must end with the substring
ovo.String().equals('abc'); // must be equal to the string
// Transformations
ovo.String().trim(); // trim whitespace
ovo.String().toLowerCase(); // convert to lowercase
ovo.String().toUpperCase(); // convert to uppercaseObject type validation can accept a value of type Map. It is an implementation of OvO<Map<String, T>>.
import 'package:ovo/ovo.dart' as ovo;
final user = ovo.Object({
'name': ovo.String(),
'age': ovo.Number(),
});
await user.parse({
'name': 'John',
'age': 18,
}); // => {'name': 'John', 'age': 18}
await user.parse({
'name': 'John',
'age': '18',
}); // => throws OvOExceptionOf course, if you just want to simply validate a value of type Map<K, T>, you can use OvO<Map<K, T>> instead of Object type validation.
import 'package:ovo/ovo.dart' as ovo;
final user = ovo.OvO<Map<String, dynamic>>();
await user.parse({
'name': 'John',
'age': 18,
}); // => {'name': 'John', 'age': 18}
await user.parse({
'name': 'John',
'age': '18',
}); // => {'name': 'John', 'age': '18'}The .nullable method can convert a type validation to a type validation that accepts null.
ovo.String().nullable(); // must be a string or null.refine is a method that allows you to customize the validation. It accepts a validation function of FutureOr<bool> Function(T data) to facilitate validation according to the actual situation.
ovo.String().refine(
(value) => value.length > 5,
message: 'must be greater than 5',
); // must be a string and length greater than 5It is worth noting that many of the built-in extension methods are implemented based on the
.refinemethod.
.transform is a method that allows you to customize the method of converting data types. It works on the principle of Onion Model.
Pre-transformation allows you to pre-process the raw data to be parsed, and then hand it over to the next converter, and finally hand it over to the type validator for verification:
final schema = ovo.String().transform(
(ovo.Context context, dynamic data, Future<T> Function(dynamic data) next) {
// data convert to string
return next(data.toString());
}
);
await schema.parse(123); // => '123'
await schema.parse(#symbol); // => '#symbol'Using the next parameter in the callback, you can perform post-transformation of the data after the type validator is verified successfully:
final schema = ovo.String().transform(
(ovo.Context context, dynamic data, Future<T> Function(dynamic data) next) async {
final value = await next(data);
// value convert to int
return int.parse(value);
}
);
await schema.parse('123'); // => 123.withDefault allows a nullable value T? to be replaced with a default value T when the value is null.
final schema = ovo.String().withDefault('Hello World'); // must be a string or null, default value is 'Hello World'
await schema.parse(null); // => 'Hello World'
await schema.parse('Hello'); // => 'Hello'As you can see, the .withDefault method will automatically attach the .nullable method, so you don't need to call the .nullable method manually.
AnyOf type validation can accept any of the multiple type validations.
ovo.AnyOf([
ovo.String(),
ovo.Integer(),
]); // must be a string or an integerAllOf type validation can accept all types in multiple type validations.
ovo.AllOf([
ovo.Double(),
ovo.Integer(),
]); // must be a double and an integerOneOf type validation can accept one type in multiple type validations. If multiple validations match, an exception is thrown.
final schema = ovo.OneOf([
ovo.String().size(5),
ovo.String().size(10),
]);
await schema.parse('12345'); // => '12345'
await schema.parse('1234567890'); // => '1234567890'
await schema.parse('123456'); // => throws OvOExceptionNot type validation can accept any type validation, but if the specified type validation is matched, an exception is thrown.
final schema = ovo.Not(ovo.String());
await schema.parse(123); // => 123
await schema.parse({'name': 'Seven'}) // => {'name': 'Seven'}
await schema.parse('123'); // => throws OvOExceptionConst type validation can accept a constant value.
ovo.Const(123); // must be 123Using Const we can implement string literals in JSON:
final schema = ovo.OneOf([
ovo.Const('mobile'),
ovo.Const('web'),
]);
await schema.parse('mobile'); // => 'mobile'
await schema.parse('web'); // => 'web'Used in conjunction with conversion, we can implement Enum type validation:
enum MyEnum {
mobile,
web,
}
final schema = ovo.OneOf(
MyEnum.values.map((e) => ovo.Const(e.name)),
).transform<MyEnum>(
(ovo.Context context, dynamic data, Future<T> Function(dynamic data) next) async {
final value = await next(data);
return MyEnum.values.firstWhere((e) => e.name == value);
}
);
await schema.parse('mobile'); // => MyEnum.mobile
await schema.parse('web'); // => MyEnum.web
await schema.parse('desktop'); // => throws OvOException
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent
