code

Code processing utilities, providing a JavaScript keyword set and a function to strip comments and strings, used for static code analysis and keyword detection scenarios.

Import Methods

// Submodule import (recommended)
import { JS_KEYWORDS, stripCommentsAndStrings } from '@meng-xi/vite-plugin/common/code'

// Barrel import
import { JS_KEYWORDS, stripCommentsAndStrings } from '@meng-xi/vite-plugin/common'

---

JS_KEYWORDS

A set of JavaScript keywords and global built-in objects, used for keyword detection, lexical analysis, and similar scenarios.

const JS_KEYWORDS: Set<string>

Description

• Includes ECMAScript keywords (break, case, class, const, let, etc.)
• Includes reserved words (await, async, yield, of, etc.)
• Includes literals (true, false, null, undefined, NaN, Infinity)
• Includes global objects (console, window, document, globalThis, process, etc.)
• Includes built-in constructors and types (Object, Array, String, Number, Promise, Map, Set, RegExp, Date, etc.)
• Includes error types (Error, TypeError, RangeError, SyntaxError, ReferenceError)
• Includes typed arrays (Int8Array, Uint8Array, Float32Array, DataView, ArrayBuffer, etc.)

Examples

JS_KEYWORDS.has('class')    // true
JS_KEYWORDS.has('myVar')    // false
JS_KEYWORDS.has('Promise')  // true

// Used to filter out non-business identifiers
const identifiers = ['myFunc', 'console', 'myVar', 'class']
const businessIds = identifiers.filter(id => !JS_KEYWORDS.has(id))
// ['myFunc', 'myVar']

---

stripCommentsAndStrings

Removes comments and string contents from JavaScript/TypeScript code while preserving the code structure, for subsequent static analysis (such as identifier extraction, keyword detection).

function stripCommentsAndStrings(code: string): string

Parameters

Parameter	Type	Description
code	string	Source code string

Returns

string - Code with comments and string contents removed (preserves newlines to keep line numbers aligned)

Processing Rules

Syntax Element	Processing
Single-line comment //	Content replaced with spaces, newline preserved
Multi-line comment /* */	Content replaced with spaces, newlines preserved (to keep line numbers aligned)
Single-quoted string '...'	Content replaced with spaces, escape sequences preserve length
Double-quoted string "..."	Content replaced with spaces, escape sequences preserve length
Template literal `...`	Content replaced with spaces, but ${...} expression content is preserved

Description

• Newlines are preserved to maintain line number alignment for error location
• Escape sequences in strings (e.g., \n, \t) are treated as 2-character length to keep length consistent
• ${expression} content in template literals is preserved as-is because it may contain business logic
• Suitable for preprocessing before static code analysis to avoid comments and strings interfering with identifier extraction

Examples

const code = `
  // single-line comment
  const name = "hello"
  /* multi-line
     comment */
  const template = \`prefix \${value} suffix\`
`

const stripped = stripCommentsAndStrings(code)
// Result:
//   (spaces)
//   const name =           (string content removed)
//   (spaces, newline preserved)
//   (spaces)
//   const template = `prefix ${value} suffix`  (template string content removed, but ${value} preserved)