Skip to main content

@dbml/core

NPM

This is a core package that is responsible for parsing and converting between different formats:

  • Parse DBML and SQL to Database object
  • Export SQL and DBML from Database object
  • Convert DBML to SQL and SQL to DBML
  • Generate DBML from DatabaseSchema object

Installation​

npm install @dbml/core

APIs​

importer​

const { importer } = require('@dbml/core');

importer.import(str, format)​

  • Arguments:

    • {string} str
    • {'mysql'|'mysqlLegacy'|'postgres'|'postgresLegacy'|'dbml'|'schemarb'|'mssql'|'mssqlLegacy'|'snowflake'|'json'} format

  • Returns:

    • {string} DBML

  • Usage: Generate DBML from SQL.

note

The postgresLegacy, mysqlLegacy and mssqlLegacy options import PostgreSQL/MySQL/MSSQL to dbml using the old parsers. It's quicker but less accurate.

const fs = require('fs');
const { importer } = require('@dbml/core');

// read PostgreSQL file script
const postgreSQL = fs.readFileSync('./schema.sql', 'utf-8');

// generate DBML from PostgreSQL script
const dbml = importer.import(postgreSQL, 'postgres');

importer.generateDbml(schemaJson)​

  • Arguments:

    • {DatabaseSchema} schemaJson

  • Returns:

    • {string} DBML

  • Usage: Generate DBML from a DatabaseSchema object.

The following example use the @dbml/connector to get the DatabaseSchema object.

const { importer } = require('@dbml/core');
const { connector } = require('@dbml/connector');

// Use the dbml connector to get the DatabaseSchema object
const connection = 'postgresql://dbml_user:dbml_pass@localhost:5432/dbname?schemas=public';
const databaseType = 'postgres';

const schemaJson = await connector.fetchSchemaJson(connection, databaseType);

// Generate DBML from the DatabaseSchema object
const dbml = importer.generateDbml(schemaJson);

exporter​

const { exporter } = require('@dbml/core');

exporter.export(str, format)​

  • Arguments:

    • {string} str
    • {'mysql'|'postgres'|'oracle'|'dbml'|'schemarb'|'mssql'|'json'} format

  • Returns:

    • {string} SQL

  • Usage: Export DBML to SQL

const fs = require('fs');
const { exporter } = require('@dbml/core');

// get DBML file content
const dbml = fs.readFileSync('./schema.dbml', 'utf-8');

// generate MySQL from DBML
const mysql = exporter.export(dbml, 'mysql');

Parser​

const { Parser } = require('@dbml/core');

const parser = new Parser();

parser.parse(str, format)​

  • Arguments:

    • {string} str
    • {'mysql'|'mysqlLegacy'|'postgres'|'postgresLegacy'|'dbml'|'schemarb'|'mssql'|'mssqlLegacy'|'snowflake'|'json'|'dbmlv2'} format

  • Returns: Database object

  • Usage: Parse specified format to Database object

note

  • The postgresLegacy, mysqlLegacy and mssqlLegacy options import PostgreSQL/MySQL/MSSQL to dbml using the old parsers. It's quicker but less accurate.

  • The dbmlv2 option parse dbml using the new parser. It's quicker and more robust to errors/more user-friendly error messages.

const fs = require('fs');
const { Parser } = require('@dbml/core');

const parser = new Parser();

// get DBML file content
const dbml = fs.readFileSync('./schema.dbml', 'utf-8');

// parse DBML to Database object
const database = parser.parse(dbml, 'dbml');

ModelExporter​

const { ModelExporter } = require('@dbml/core');

ModelExporter.export(model, format, isNormalized)​

  • Arguments:

    • {model} Database
    • {'mysql'|'postgres'|'oracle'|'dbml'|'schemarb'|'mssql'|'json'} format
    • {boolean} isNormalized

  • Returns: specified format string

  • Usage: Export Database object to specified format

const { ModelExporter, Parser } = require('@dbml/core');

// get DBML file content
const dbml = fs.readFileSync('./schema.dbml', 'utf-8');

// parse DBML to Database object
const parser = new Parser();
const database = parser.parse(dbml, 'dbml');

// Export Database object to PostgreSQL
const postgreSQL = ModelExporter.export(database, 'postgres', false);
// or
const postgreSQL = ModelExporter.export(database.normalize(), 'postgres');