README.md

schema-typed

Schema for data modeling & validation

Table of Contents

• Installation
• Usage
  • Getting Started
  • Multiple verification
  • Custom verification
    • Field dependency validation
    • Asynchronous check
  • Validate nested objects
  • Combine
• API
  • SchemaModel
    • static combine(...models)
    • check(data: object)
    • checkAsync(data: object)
    • checkForField(fieldName: string, data: object, options?: { nestedObject?: boolean })
    • checkForFieldAsync(fieldName: string, data: object, options?: { nestedObject?: boolean })
  • MixedType()
    • isRequired(errorMessage?: string, trim: boolean = true)
    • isRequiredOrEmpty(errorMessage?: string, trim: boolean = true)
    • addRule(onValid: Function, errorMessage?: string, priority: boolean)
    • addAsyncRule(onValid: Function, errorMessage?: string, priority: boolean)
    • when(condition: (schemaSpec: SchemaDeclaration<DataType, ErrorMsgType>) => Type)
    • check(value: ValueType, data?: DataType):CheckResult
    • checkAsync(value: ValueType, data?: DataType):Promise<CheckResult>
    • label(label: string)
    • equalTo(fieldName: string, errorMessage?: string)
    • proxy(fieldNames: string[], options?: { checkIfValueExists?: boolean })
  • StringType(errorMessage?: string)
    • isEmail(errorMessage?: string)
    • isURL(errorMessage?: string)
    • isOneOf(items: string[], errorMessage?: string)
    • containsLetter(errorMessage?: string)
    • containsUppercaseLetter(errorMessage?: string)
    • containsLowercaseLetter(errorMessage?: string)
    • containsLetterOnly(errorMessage?: string)
    • containsNumber(errorMessage?: string)
    • pattern(regExp: RegExp, errorMessage?: string)
    • rangeLength(minLength: number, maxLength: number, errorMessage?: string)
    • minLength(minLength: number, errorMessage?: string)
    • maxLength(maxLength: number, errorMessage?: string)
  • NumberType(errorMessage?: string)
    • isInteger(errorMessage?: string)
    • isOneOf(items: number[], errorMessage?: string)
    • pattern(regExp: RegExp, errorMessage?: string)
    • range(minLength: number, maxLength: number, errorMessage?: string)
    • min(min: number, errorMessage?: string)
    • max(max: number, errorMessage?: string)
  • ArrayType(errorMessage?: string)
    • isRequiredOrEmpty(errorMessage?: string)
    • rangeLength(minLength: number, maxLength: number, errorMessage?: string)
    • minLength(minLength: number, errorMessage?: string)
    • maxLength(maxLength: number, errorMessage?: string)
    • unrepeatable(errorMessage?: string)
    • of(type: object)
  • DateType(errorMessage?: string)
    • range(min: Date, max: Date, errorMessage?: string)
    • min(min: Date, errorMessage?: string)
    • max(max: Date, errorMessage?: string)
  • ObjectType(errorMessage?: string)
    • shape(fields: object)
  • BooleanType(errorMessage?: string)
• ⚠️ Notes

Installation

npm install schema-typed --save

Usage

Getting Started

import { SchemaModel, StringType, DateType, NumberType, ObjectType, ArrayType } from 'schema-typed';

const model = SchemaModel({
  username: StringType().isRequired('Username required'),
  email: StringType().isEmail('Email required'),
  age: NumberType('Age should be a number').range(18, 30, 'Over the age limit'),
  tags: ArrayType().of(StringType('The tag should be a string').isRequired()),
  role: ObjectType().shape({
    name: StringType().isRequired('Name required'),
    permissions: ArrayType().isRequired('Permissions required')
  })
});

const checkResult = model.check({
  username: 'foobar',
  email: 'foo@bar.com',
  age: 40,
  tags: ['Sports', 'Games', 10],
  role: { name: 'administrator' }
});

console.log(checkResult);

checkResult return structure is:

{
  username: { hasError: false },
  email: { hasError: false },
  age: { hasError: true, errorMessage: 'Over the age limit' },
  tags: {
    hasError: true,
    array: [
      { hasError: false },
      { hasError: false },
      { hasError: true, errorMessage: 'The tag should be a string' }
    ]
  },
  role: {
    hasError: true,
    object: {
      name: { hasError: false },
      permissions: { hasError: true, errorMessage: 'Permissions required' }
    }
  }
};

Multiple verification

StringType()
  .minLength(6, "Can't be less than 6 characters")
  .maxLength(30, 'Cannot be greater than 30 characters')
  .isRequired('This field required');

Custom verification

Customize a rule with the addRule function.

If you are validating a string type of data, you can set a regular expression for custom validation by the pattern method.

const model = SchemaModel({
  field1: StringType().addRule((value, data) => {
    return /^[1-9][0-9]{3}\s?[a-zA-Z]{2}$/.test(value);
  }, 'Please enter legal characters'),
  field2: StringType().pattern(/^[1-9][0-9]{3}\s?[a-zA-Z]{2}$/, 'Please enter legal characters')
});

model.check({ field1: '', field2: '' });

/**
{
  field1: {
    hasError: true,
    errorMessage: 'Please enter legal characters'
  },
  field2: {
    hasError: true,
    errorMessage: 'Please enter legal characters'
  }
};
**/

Field dependency validation

1. Use the equalTo method to verify that the values of two fields are equal.

const model = SchemaModel({
  password: StringType().isRequired(),
  confirmPassword: StringType().equalTo('password')
});

2. Use the addRule method to create a custom validation rule.

const model = SchemaModel({
  password: StringType().isRequired(),
  confirmPassword: StringType().addRule(
    (value, data) => value === data.password,
    'Confirm password must be the same as password'
  )
});

3. Use the proxy method to verify that a field passes, and then proxy verification of other fields.

const model = SchemaModel({
  password: StringType().isRequired().proxy(['confirmPassword']),
  confirmPassword: StringType().equalTo('password')
});

Asynchronous check

For example, verify that the mailbox is duplicated

function asyncCheckEmail(email) {
  return new Promise(resolve => {
    setTimeout(() => {
      if (email === 'foo@domain.com') {
        resolve(false);
      } else {
        resolve(true);
      }
    }, 500);
  });
}

const model = SchemaModel({
  email: StringType()
    .isEmail('Please input the correct email address')
    .addAsyncRule((value, data) => {
      return asyncCheckEmail(value);
    }, 'Email address already exists')
    .isRequired('This field cannot be empty')
});

model.checkAsync({ email: 'foo@domain.com' }).then(checkResult => {
  console.log(checkResult);
  /**
  {
    email: {
      hasError: true,
      errorMessage: 'Email address already exists'
    }
  };
  **/
});

Validate nested objects

Validate nested objects, which can be defined using the ObjectType().shape method. E.g:

const model = SchemaModel({
  id: NumberType().isRequired('This field required'),
  name: StringType().isRequired('This field required'),
  info: ObjectType().shape({
    email: StringType().isEmail('Should be an email'),
    age: NumberType().min(18, 'Age should be greater than 18 years old')
  })
});

const user = {
  id: 1,
  name: '',
  info: { email: 'schema-type', age: 17 }
};

model.check(data);

/**
 {
  "id": { "hasError": false },
  "name": { "hasError": true, "errorMessage": "This field required" },
  "info": {
    "hasError": true,
    "object": {
      "email": { "hasError": true, "errorMessage": "Should be an email" },
      "age": { "hasError": true, "errorMessage": "Age should be greater than 18 years old" }
    }
  }
}
*/

Combine

SchemaModel provides a static method combine that can be combined with multiple SchemaModel to return a new SchemaModel.

const model1 = SchemaModel({
  username: StringType().isRequired('This field required'),
  email: StringType().isEmail('Should be an email')
});

const model2 = SchemaModel({
  username: StringType().minLength(7, "Can't be less than 7 characters"),
  age: NumberType().range(18, 30, 'Age should be greater than 18 years old')
});

const model3 = SchemaModel({
  groupId: NumberType().isRequired('This field required')
});

const model4 = SchemaModel.combine(model1, model2, model3);

model4.check({
  username: 'foobar',
  email: 'foo@bar.com',
  age: 40,
  groupId: 1
});

API

SchemaModel

SchemaModel is a JavaScript schema builder for data model creation and validation.

static combine(...models)

A static method for merging multiple models.

const model1 = SchemaModel({
  username: StringType().isRequired('This field required')
});

const model2 = SchemaModel({
  email: StringType().isEmail('Please input the correct email address')
});

const model3 = SchemaModel.combine(model1, model2);

check(data: object)

Check whether the data conforms to the model shape definition. Return a check result.

const model = SchemaModel({
  username: StringType().isRequired('This field required'),
  email: StringType().isEmail('Please input the correct email address')
});

model.check({
  username: 'root',
  email: 'root@email.com'
});

checkAsync(data: object)

Asynchronously check whether the data conforms to the model shape definition. Return a check result.

const model = SchemaModel({
  username: StringType()
    .isRequired('This field required')
    .addRule(value => {
      return new Promise(resolve => {
        // Asynchronous processing logic
      });
    }, 'Username already exists'),
  email: StringType().isEmail('Please input the correct email address')
});

model
  .checkAsync({
    username: 'root',
    email: 'root@email.com'
  })
  .then(result => {
    // Data verification result
  });

checkForField(fieldName: string, data: object, options?: { nestedObject?: boolean })

Check whether a field in the data conforms to the model shape definition. Return a check result.

const model = SchemaModel({
  username: StringType().isRequired('This field required'),
  email: StringType().isEmail('Please input the correct email address')
});

const data = {
  username: 'root'
};

model.checkForField('username', data);

checkForFieldAsync(fieldName: string, data: object, options?: { nestedObject?: boolean })

Asynchronously check whether a field in the data conforms to the model shape definition. Return a check result.

const model = SchemaModel({
  username: StringType()
    .isRequired('This field required')
    .addAsyncRule(value => {
      return new Promise(resolve => {
        // Asynchronous processing logic
      });
    }, 'Username already exists'),
  email: StringType().isEmail('Please input the correct email address')
});

const data = {
  username: 'root'
};

model.checkForFieldAsync('username', data).then(result => {
  // Data verification result
});

MixedType()

Creates a type that matches all types. All types inherit from this base type.

isRequired(errorMessage?: string, trim: boolean = true)

MixedType().isRequired('This field required');

isRequiredOrEmpty(errorMessage?: string, trim: boolean = true)

MixedType().isRequiredOrEmpty('This field required');

addRule(onValid: Function, errorMessage?: string, priority: boolean)

MixedType().addRule((value, data) => {
  return /^[1-9][0-9]{3}\s?[a-zA-Z]{2}$/.test(value);
}, 'Please enter a legal character.');

addAsyncRule(onValid: Function, errorMessage?: string, priority: boolean)

MixedType().addAsyncRule((value, data) => {
  return new Promise(resolve => {
    // Asynchronous processing logic
  });
}, 'Please enter a legal character.');

when(condition: (schemaSpec: SchemaDeclaration<DataType, ErrorMsgType>) => Type)

Conditional validation, the return value is a new type.

const model = SchemaModel({
  option: StringType().isOneOf(['a', 'b', 'other']),
  other: StringType().when(schema => {
    const { value } = schema.option;
    return value === 'other' ? StringType().isRequired('Other required') : StringType();
  })
});

/**
{
  option: { hasError: false },
  other: { hasError: false }
}
*/
model.check({ option: 'a', other: '' });

/*
{
  option: { hasError: false },
  other: { hasError: true, errorMessage: 'Other required' }
}
*/
model.check({ option: 'other', other: '' });

Check whether a field passes the validation to determine the validation rules of another field.

const model = SchemaModel({
  password: StringType().isRequired('Password required'),
  confirmPassword: StringType().when(schema => {
    const { hasError } = schema.password.check();
    return hasError
      ? StringType()
      : StringType().addRule(
          value => value === schema.password.value,
          'The passwords are inconsistent twice'
        );
  })
});

check(value: ValueType, data?: DataType):CheckResult

const type = MixedType().addRule(v => {
  if (typeof v === 'number') {
    return true;
  }
  return false;
}, 'Please enter a valid number');

type.check('1'); //  { hasError: true, errorMessage: 'Please enter a valid number' }
type.check(1); //  { hasError: false }

checkAsync(value: ValueType, data?: DataType):Promise<CheckResult>

const type = MixedType().addRule(v => {
  return new Promise(resolve => {
    setTimeout(() => {
      if (typeof v === 'number') {
        resolve(true);
      } else {
        resolve(false);
      }
    }, 500);
  });
}, 'Please enter a valid number');

type.checkAsync('1').then(checkResult => {
  //  { hasError: true, errorMessage: 'Please enter a valid number' }
});
type.checkAsync(1).then(checkResult => {
  //  { hasError: false }
});

label(label: string)

Overrides the key name in error messages.

MixedType().label('Username');

Eg:

SchemaModel({
  first_name: StringType().label('First name'),
  age: NumberType().label('Age')
});

equalTo(fieldName: string, errorMessage?: string)

Check if the value is equal to the value of another field.

SchemaModel({
  password: StringType().isRequired(),
  confirmPassword: StringType().equalTo('password')
});

proxy(fieldNames: string[], options?: { checkIfValueExists?: boolean })

After the field verification passes, proxy verification of other fields.

• fieldNames: The field name to be proxied.
• options.checkIfValueExists: When the value of other fields exists, the verification is performed (default: false)

SchemaModel({
  password: StringType().isRequired().proxy(['confirmPassword']),
  confirmPassword: StringType().equalTo('password')
});

StringType(errorMessage?: string)

Define a string type. Supports all the same methods as MixedType.

isEmail(errorMessage?: string)

StringType().isEmail('Please input the correct email address');

isURL(errorMessage?: string)

StringType().isURL('Please enter the correct URL address');

isOneOf(items: string[], errorMessage?: string)

StringType().isOneOf(['Javascript', 'CSS'], 'Can only type `Javascript` and `CSS`');

containsLetter(errorMessage?: string)

StringType().containsLetter('Must contain English characters');

containsUppercaseLetter(errorMessage?: string)

StringType().containsUppercaseLetter('Must contain uppercase English characters');

containsLowercaseLetter(errorMessage?: string)

StringType().containsLowercaseLetter('Must contain lowercase English characters');

containsLetterOnly(errorMessage?: string)

StringType().containsLetterOnly('English characters that can only be included');

containsNumber(errorMessage?: string)

StringType().containsNumber('Must contain numbers');

pattern(regExp: RegExp, errorMessage?: string)

StringType().pattern(/^[1-9][0-9]{3}\s?[a-zA-Z]{2}$/, 'Please enter legal characters');

rangeLength(minLength: number, maxLength: number, errorMessage?: string)

StringType().rangeLength(6, 30, 'The number of characters can only be between 6 and 30');

minLength(minLength: number, errorMessage?: string)

StringType().minLength(6, 'Minimum 6 characters required');

maxLength(maxLength: number, errorMessage?: string)

StringType().maxLength(30, 'The maximum is only 30 characters.');

NumberType(errorMessage?: string)

Define a number type. Supports all the same methods as MixedType.

isInteger(errorMessage?: string)

NumberType().isInteger('It can only be an integer');

isOneOf(items: number[], errorMessage?: string)

NumberType().isOneOf([5, 10, 15], 'Can only be `5`, `10`, `15`');

pattern(regExp: RegExp, errorMessage?: string)

NumberType().pattern(/^[1-9][0-9]{3}$/, 'Please enter a legal character.');

range(minLength: number, maxLength: number, errorMessage?: string)

NumberType().range(18, 40, 'Please enter a number between 18 - 40');

min(min: number, errorMessage?: string)

NumberType().min(18, 'Minimum 18');

max(max: number, errorMessage?: string)

NumberType().max(40, 'Maximum 40');

ArrayType(errorMessage?: string)

Define a array type. Supports all the same methods as MixedType.

isRequiredOrEmpty(errorMessage?: string)

ArrayType().isRequiredOrEmpty('This field required');

rangeLength(minLength: number, maxLength: number, errorMessage?: string)

ArrayType().rangeLength(1, 3, 'Choose at least one, but no more than three');

minLength(minLength: number, errorMessage?: string)

ArrayType().minLength(1, 'Choose at least one');

maxLength(maxLength: number, errorMessage?: string)

ArrayType().maxLength(3, "Can't exceed three");

unrepeatable(errorMessage?: string)

ArrayType().unrepeatable('Duplicate options cannot appear');

of(type: object)

ArrayType().of(StringType('The tag should be a string').isRequired());

DateType(errorMessage?: string)

Define a date type. Supports all the same methods as MixedType.

range(min: Date, max: Date, errorMessage?: string)

DateType().range(
  new Date('08/01/2017'),
  new Date('08/30/2017'),
  'Date should be between 08/01/2017 - 08/30/2017'
);

min(min: Date, errorMessage?: string)

DateType().min(new Date('08/01/2017'), 'Minimum date 08/01/2017');

max(max: Date, errorMessage?: string)

DateType().max(new Date('08/30/2017'), 'Maximum date 08/30/2017');

ObjectType(errorMessage?: string)

Define a object type. Supports all the same methods as MixedType.

shape(fields: object)

ObjectType().shape({
  email: StringType().isEmail('Should be an email'),
  age: NumberType().min(18, 'Age should be greater than 18 years old')
});

BooleanType(errorMessage?: string)

Define a boolean type. Supports all the same methods as MixedType.

⚠️ Notes

Default check priority:

• 1.isRequired
• 2.All other checks are executed in sequence

If the third argument to addRule is true, the priority of the check is as follows:

• 1.addRule
• 2.isRequired
• 3.Predefined rules (if there is no isRequired, value is empty, the rule is not executed)