Skip to content

kena0ki/dddl

Repository files navigation

dddl

codecov

dddl generates test Data from DDL (i.e. create table statements).
How data is generated depends on column types and options but the general idea is simple.
Generator adds 1 to previous data row by row so each column would have sequentially incremented number.
Given that we have the following statement,

 create table a (
   c1 char(5),
   c2 integer,
   c3 float,
   c4 binary(8)
 );

then we would get the following.

   c1      c2  c3    c4
L1 "a0001","1","0.1","b0000001"
L2 "a0002","2","0.2","b0000002"
L3 "a0003","3","0.3","b0000003"
L4 "a0004","4","0.4","b0000004"

This library is not yet stable. Any features or APIs are subject to change even if it's a minor version update.

Getting started

Installation

npm install dddl

Simple usage

import { parseAndGenerate } from 'dddl';
const sql = `
create table A (
    col1 char(3)
);
`;
async function main() {
  try {
      const [rows, errors] = await parseAndGenerate(sql);
      // rows -> generated data
      console.log(rows);
      // errors -> data validation errors
      console.log(errors);
  } catch(error) {
      // error -> parse error or data generation error
      console.log(error);
  }
}
main();

Working demo

Here

Options

See API reference

Supported types

See API reference

Data validation

Currently followings are supported.

  • PRIMARY KEY constraint
  • UNIQUE KEY constraint
  • NOT NULL constraint

SQL parser

SQL syntax quite depends on DBMS products, so the create statement you have may or may not be valid for this library.
However this library basically comforms with the ANSI standards since the parser part of this library has been translated and ported from the Rust project named sqlparser-rs, which aims to comform with the standards, so hopefully the core part of your create statement (i.e. column definition) is valid for this library.

Lisence

Apache License 2.0