Overview
TypeConverter is a static utility class used to convert between JSON Schema types and various database types. It is the base dependency of all exporters.
Core functions
- ✅ JSON Schema ↔ MongoDB BSON type conversion
- ✅ JSON Schema ↔ MySQL type conversion
- ✅ JSON Schema ↔ PostgreSQL type conversion
- ✅ Attribute name normalization (camel case ↔ underline)
- ✅ Format validation regular expression
- ✅ Schema merge and constraint extraction
quick start
import { TypeConverter } from 'schema-dsl/pure';
// Convert JSON Schema type to MongoDB type
const mongoType = TypeConverter.toMongoDBType('integer');
console.log(mongoType); // 'int'
// Convert JSON Schema type to MySQL type
const mysqlType = TypeConverter.toMySQLType('string', { maxLength: 100 });
console.log(mysqlType); // 'VARCHAR(100)'
// Convert JSON Schema type to PostgreSQL type
const pgType = TypeConverter.toPostgreSQLType('string', { format: 'date-time' });
console.log(pgType); // 'TIMESTAMP'
API reference
toJSONSchemaType(nativeType)
Convert the type identifier to a JSON Schema type string.
TypeConverter.toJSONSchemaType('string');
// 'string'
TypeConverter.toJSONSchemaType('integer');
// 'integer'
TypeConverter.toJSONSchemaType('email');
// 'string'
toMongoDBType(jsonSchemaType)
Convert JSON Schema type to MongoDB BSON type.
TypeConverter.toMongoDBType('string'); // 'string'
TypeConverter.toMongoDBType('number'); // 'double'
TypeConverter.toMongoDBType('integer'); // 'int'
TypeConverter.toMongoDBType('boolean'); // 'bool'
toMySQLType(jsonSchemaType, schemaFragment)
Convert JSON Schema type to MySQL data type.
//Basic conversion
TypeConverter.toMySQLType('string');
// 'VARCHAR(255)'
//With length constraints
TypeConverter.toMySQLType('string', { maxLength: 50 });
// 'VARCHAR(50)'
// long text
TypeConverter.toMySQLType('string', { maxLength: 500 });
// 'TEXT'
// Email format
TypeConverter.toMySQLType('string', { format: 'email' });
// 'VARCHAR(255)'
// Integer range (minimum + maximum hits the TINYINT branch)
TypeConverter.toMySQLType('integer', { minimum: 0, maximum: 100 });
// 'TINYINT'
toPostgreSQLType(jsonSchemaType, schemaFragment)
Convert JSON Schema type to PostgreSQL data type.
// UUID format
TypeConverter.toPostgreSQLType('string', { format: 'uuid' });
// 'VARCHAR(255)' // The current implementation does not automatically switch to the UUID column type due to format=uuid
// date time
TypeConverter.toPostgreSQLType('string', { format: 'date-time' });
// 'TIMESTAMP'
// JSON object
TypeConverter.toPostgreSQLType('object');
// 'JSONB'
normalizePropertyName(name)
Normalize attribute names: remove leading and trailing whitespace, replace illegal characters with underscores, and compress consecutive underscores.
TypeConverter.normalizePropertyName('user name');
// 'user_name'
TypeConverter.normalizePropertyName('123created-at');
// '123created_at'
Get the RegExp instance corresponding to the format; unknown format returns null.
const emailRegex = TypeConverter.formatToRegex('email');
emailRegex?.test('user@example.com');
// true
const uuidRegex = TypeConverter.formatToRegex('uuid');
uuidRegex?.test('123e4567-e89b-12d3-a456-426614174000');
// true
TypeConverter.formatToRegex('unknown');
// null
mergeSchemas(base, override)
Merge two JSON Schema objects.
const base = {
type: 'object',
properties: { name: { type: 'string' } },
required: ['name']
};
const override = {
properties: { email: { type: 'string' } },
required: ['email']
};
const merged = TypeConverter.mergeSchemas(base, override);
// {
// type: 'object',
// properties: { name: {...}, email: {...} },
// required: ['name', 'email']
// }
Extract constraints from the Schema.
const schema = {
type: 'string',
minLength: 3,
maxLength: 32,
pattern: '^[a-z]+$',
format: 'email'
};
const constraints = TypeConverter.extractConstraints(schema);
// {
// minLength: 3,
// maxLength: 32,
// pattern: '^[a-z]+$',
// format: 'email'
// }
type mapping table
JSON Schema → MongoDB
JSON Schema → MySQL
JSON Schema → PostgreSQL
Practical examples
Batch type conversion
import { TypeConverter } from 'schema-dsl/pure';
const fields = ['string', 'number', 'integer', 'boolean', 'object', 'array'];
console.log('=== type mapping comparison ===');
fields.forEach(type => {
console.log(`${type}:`);
console.log(` MongoDB: ${TypeConverter.toMongoDBType(type)}`);
console.log(` MySQL: ${TypeConverter.toMySQLType(type)}`);
console.log(` PostgreSQL: ${TypeConverter.toPostgreSQLType(type)}`);
});
const emailRegex = TypeConverter.formatToRegex('email');
const regex = new RegExp(emailRegex);
console.log(regex.test('user@example.com')); // true
console.log(regex.test('invalid-email')); // false
Corresponding sample file
Example entry: type-converter.ts
Description: Covers type mapping, enumeration to MySQL ENUM(...), PostgreSQL actual UUID mapping, property name normalization, regex fetching, Schema merging and constraint extraction.
