8000
Skip to content

codegen

Directory actions

More options

Directory actions

More options

Latest commit

 

History

History

codegen

README.md

oxc-codegen

Fast, synchronous code generation for JavaScript and TypeScript ASTs.

oxc-codegen turns an ESTree or TS-ESTree AST into formatted source code. It supports JavaScript, JSX, TypeScript, and TSX.

The printer is a port of Oxc's Rust oxc_codegen crate. With the default options, both printers produce byte-identical output: tab indentation, double-quoted strings, and no comments.

Installation

npm install oxc-codegen

oxc-codegen is ESM-only and requires Node.js ^20.19.0 or >=22.12.0.

Quick start

Pair it with oxc-parser to parse and print source code:

import { printSync } from "oxc-codegen";
import { parseSync } from "oxc-parser";

const { program } = parseSync("input.js", "const answer=6*7");
const { code } = printSync(program);

console.log(code);
// const answer = 6 * 7;

You can also print a manually constructed AST:

const program = {
  type: "Program",
  sourceType: "script",
  body: [
    {
      type: "ExpressionStatement",
      expression: {
        type: "CallExpression",
        callee: {
          type: "MemberExpression",
          object: { type: "Identifier", name: "console" },
          property: { type: "Identifier", name: "log" },
          computed: false,
          optional: false,
        },
        arguments: [{ type: "Literal", value: "Hello!" }],
        optional: false,
      },
    },
  ],
};

console.log(printSync(program).code);
// console.log("Hello!");

TypeScript and TSX

Set ts when the AST can contain TypeScript nodes. For TSX, set both ts and jsx:

const { program } = parseSync("component.tsx", "const Box = <T,>(value: T) => <div>{value}</div>");

const { code } = printSync(program, {
  ts: true,
  jsx: true,
});

API

printSync(node, options?)

function printSync(
  node: ESTree.Program | ESTree.Statement,
  options?: Options,
): {
  code: string;
  map: SourceMap | null;
};

Prints a complete Program or a single statement and returns the generated source code, and (when requested) a standard Source Map v3 object.

import { printSync } from "oxc-codegen";
import { parseSync } from "oxc-parser";

const sourceText = "const answer=6*7";
const { program } = parseSync("input.js", sourceText);
const { code, map } = printSync(program, {
  sourcemap: true,
  sourceFilename: "input.js",
  sourceText,
});

Source-map mappings require sourceText and nodes with valid Oxc start / end offsets. A manually constructed AST without offsets can still be printed, but its source map has an empty mappings string.

Options

Option Type Default Description
indent string "\t" Non-empty string of spaces and/or tabs used for one indent level
startingIndentLevel number 0 Starting indent level, from 0 to 1000
jsx boolean false Enable TSX-safe printing for ambiguous TypeScript syntax
ts boolean false Select the printer that supports TypeScript nodes
sourcemap boolean false Return a Source Map v3 object in map
sourceFilename string "" Original source filename recorded in the source map
sourceText string - Original text required for source-map mappings and content

Why pure JavaScript?

Most Oxc packages use native bindings. This package deliberately does not: when an AST already lives in JavaScript, passing the entire object graph across a JS/native boundary can cost more than printing it in place. oxc-codegen avoids that serialization and uses specialized printer builds for JavaScript and TypeScript workloads.

See DESIGN.md for the implementation details and performance constraints.

Current limitations

  • Comments are not printed.
  • Minified output is not supported.

Benchmarks

Representative time per printSync call:

Fixture Bytes Time
tiny.js 26 0.0001 ms
RadixUIAdoptionSection.jsx 2,518 0.0033 ms
react.development.js 72,141 0.1138 ms
binder.ts 193,077 0.2472 ms
App.tsx 415,340 0.7490 ms
lodash.js 544,096 0.4995 ms
kitchen-sink.tsx 732,222 2.5682 ms
antd.js 6,683,633 11.3914 ms

These figures come from one machine and are illustrative, not a regression baseline. Results—most noticeably for large fixtures such as antd.js—vary between runs.
0