Code Generation

The codegen options control how Oxc minifier prints the output.

Whitespace Stripping

Oxc minifier removes whitespace by default. Set codegen.removeWhitespace to false to keep the output formatted.

Comments are removed by default. Use codegen.legalComments to control whether legal comments are preserved, moved to the end of the file, or extracted.

ASCII Escaping

Set codegen.asciiOnly to true to escape non-ASCII characters in string literals, untagged template literals, regular expression literals, and identifier names. The option defaults to false and works with or without whitespace stripping.

Node.js

import { minifySync } from "oxc-minify";

const result = minifySync("input.js", "export let π = '☕';", {
  codegen: {
    asciiOnly: true,
    removeWhitespace: false,
  },
});

console.log(result.code);

Output:

export let \u03C0 = "\u2615";

The async minify function accepts the same option.

Rust

Set ascii_only on oxc_codegen::CodegenOptions when generating code:

use oxc_codegen::{Codegen, CodegenOptions};

let output = Codegen::new()
    .with_options(CodegenOptions {
        ascii_only: true,
        ..CodegenOptions::default()
    })
    .build(&program);

Escapes and Limitations

Characters up to U+FFFF use \uXXXX escapes. Higher code points use \u{...} escapes, which require ES2015 or later. This option does not provide ES5-compatible output.

Regular expressions use escaped UTF-16 surrogate pairs for higher code points instead. Escaping a regular expression changes its observable RegExp.prototype.source value.

Some output can still contain non-ASCII characters:

• Tagged template literal text is preserved because tag functions such as String.raw can observe its raw contents.
• JSX names, JSX text, and JSX attribute strings are preserved.
• Hashbangs and preserved comments, including legal comments, are preserved.

JavaScript expressions inside tagged templates and JSX are escaped normally.