0.3.2 • Published 4 years ago

universal-ddl v0.3.2

Weekly downloads
1
License
CC0-1.0
Repository
github
Last release
4 years ago

Universal DDL — universal-ddl

Build Status Dependencies Status Codacy Badge npm Type definitions GitHub

Parse DDL scripts in a universal format, then generates DDL scripts for several DBMS.

This package provides a CLI and a programmatic API for Node.js.

How to use the Command Line Interface

Example:

npx universal-ddl --autofix --postgresql --sqlite path/to/ddl-file.sql

This command will generate two new files path/to/ddl-file.postgresql.sql and path/to/ddl-file.sqlite.sql.

Available options:

  -h, --help                   Print this help message.
  -a, --autofix                Enable autofix.
  -o, --output-dir directory   The output directory (optional).
  -p, --postgresql             Generate a DDL for Postgresql.
  -s, --sqlite                 Generate a DDL for SQLite.
  -m, --mariadb                Generate a DDL for Mariadb or MySQL.
  -u, --universal-ddl          Generate a DDL using the Universal DDL syntax.
  -d, --generate-drop          Generate drop statements (except for the Universal DDL output).
  -e, --encoding string        Encoding for input and output file(s) (default is utf8).
  -f, --force                  Overwrite output files.
  --src file                   The source file (by default at last position).

How to use the API from Node.js

Install as a dependency:

npm install universal-ddl

Then, use it:

const {
  parseDdl,
  generateDdl,
  createRds,
  parseDdlToRds,
} = require("universal-ddl");

const input = `
  create table t1 (
    a integer not null primary key autoincrement
  );
  `;

// Parse the input DDL and create an AST
const ast = parseDdl(input, {
  autofix: true,
  checkConsistency: true,
  freeze: true,
});

// The AST is a pure JSON format, it can be stringified
console.log(JSON.stringify(ast, undefined, 2));

// How to generate a specific DDL for your DBMS
console.log(generateDdl(ast, "postgresql"));

// Create a RDS (Relational Database Structure). It is a POJO object, higher
// level than an AST. It is recursive, so it can't be stringified.
const rds = createRds(ast, { freeze: true });

// Or, create the same RDS using a shortcut
const rds2 = parseDdlToRds(input, { autofix: true, freeze: true });

// Use the RDS
console.log(rds.tables["t1"].columns["a"].constraints.notNull); // true

Contribute

Install and build

We need a JVM (Java Virtual Machine) to build the parser because we use ANTLR, which is a Java program. So, at first, install a JVM on your system.

In a terminal, open the cloned universal-ddl/ repository. Then:

# Download once the ANTLR JAR file in the project's root directory
wget https://www.antlr.org/download/antlr-4.7.2-complete.jar

# Install once all Node.js dependencies
npm install

# Build
npm run build

# Run tests
npm run test

Development environment

With VS Code, our recommanded plugins are:

  • ANTLR4 grammar syntax support from Mike Lischke (mike-lischke.vscode-antlr4)
  • TSLint from Microsoft (ms-vscode.vscode-typescript-tslint-plugin)