estree-util-visit

What is estree-util-visit?

The estree-util-visit package is a utility for visiting nodes in an ESTree-compliant AST (Abstract Syntax Tree). It provides a simple way to traverse and manipulate JavaScript code structures programmatically, which is useful in tasks such as code analysis, transformation, and optimization.

What are estree-util-visit's main functionalities?

Node Traversal

This feature allows you to traverse the AST nodes. The 'enter' method is called for each node in the tree, where you can perform actions like logging node types or modifying nodes.

import {visit} from 'estree-util-visit';

visit(tree, {
  enter(node) {
    console.log(node.type);
  }
});

Selective Traversal

This feature enables traversal of specific types of nodes. In this example, it logs the names of all function declarations, allowing targeted analysis or transformations.

import {visit} from 'estree-util-visit';

visit(tree, {
  enter(node) {
    if (node.type === 'FunctionDeclaration') {
      console.log('Function name:', node.id.name);
    }
  }
});

Skip Subtree

This feature allows you to skip the traversal of a subtree. If a function declaration is encountered, its body is skipped, optimizing traversal by avoiding unnecessary nodes.

import {visit} from 'estree-util-visit';

visit(tree, {
  enter(node, ancestors) {
    if (node.type === 'FunctionDeclaration') {
      return visit.SKIP;
    }
  }
});

Other packages similar to estree-util-visit

estraverse

Estraverse is another popular package for traversing and manipulating JavaScript ASTs. It offers similar functionalities but differs in API design and extensibility. Estraverse provides more granular control over traversal with separate enter and leave functions.

acorn-walk

Acorn-walk, derived from the Acorn JS parser, provides utilities for walking ASTs generated by Acorn. It is tightly coupled with Acorn's AST structure, whereas estree-util-visit is more generic and works with any ESTree-compliant AST.

README

estree-util-visit

esast (and estree) utility to visit nodes.

Similar to unist-util-visit. Basically the tiniest you can go (while being usable). Also has nice stack traces if something crashes.

Install

This package is ESM only: Node 12+ is needed to use it and it must be imported instead of required.

npm:

npm install estree-util-visit

Use

import {parse} from 'acorn'
import {visit} from 'estree-util-visit'

var tree = parse(
  'export function x() { console.log(1 + "2"); process.exit(3) }',
  {sourceType: 'module'}
)

visit(tree, function (node) {
  if (node.type === 'Literal') console.log(node.value)
})

// Both enter and leave:
// walk(tree, {
//   enter(node, field, index, parents) {},
//   leave(node, field, index, parents) {}
// })

Yields:

1
"2"
3

API

This package exports the following identifiers: visit, EXIT, CONTINUE, and SKIP. There is no default export.

visit(tree, visitor|visitors)

Visit nodes (inclusive descendants of tree), with ancestral information.

This algorithm performs depth-first tree traversal in preorder (NLR) and/or postorder (LRN).

Compared to other estree walkers, this does not need a dictionary of which fields are nodes, because it ducktypes instead.

Walking the tree is an intensive task. Make use of the return values of the visitor(s) when possible. Instead of walking a tree multiple times, walk it once, use unist-util-is to check if a node matches, and then perform different operations.

Parameters

• tree (Node) — Tree to traverse
• visitor (Function) — Same as passing {enter: visitor}
• visitors ({enter: visitor, exit: visitor}) — Two functions, respectively called when entering a node (preorder) or before leaving a node (postorder)

next? = visitor(node, key, index, ancestors)

Called when a node is found.

Visitors are free to transform node. They can also transform the parent of node (the last of ancestors). Replacing node itself, if SKIP is not returned, still causes its descendants to be walked. If adding or removing previous siblings of node, visitor should return a new index (number) to specify the sibling to traverse after node is traversed. Adding or removing next siblings of node is handled as expected without needing to return a new index.

Parameters

• node (Node) — Found node
• key (string?) — Field at which node lives in its parent
• index (number?) — Index at which node lives if parent[key] is an array
• ancestors (Array.<Node>) — Ancestors of node

Returns

The return value can have the following forms:

• index (number) — Treated as a tuple of [CONTINUE, index]
• action (symbol) — Treated as a tuple of [action]
• tuple (Array.<symbol|number>) — List with one or two values, the first an action, the second and index. Note that passing a tuple only makes sense if the action is SKIP. If the action is EXIT, that action can be returned. If the action is CONTINUE, index can be returned.

action

An action can have the following values:

• EXIT (symbol) — Stop traversing immediately
• CONTINUE (symbol) — Continue traversing as normal (same behaviour as not returning an action)
• SKIP (symbol) — Do not traverse this node’s children. Has no effect in leave

Related

Contribute

See contributing.md in syntax-tree/.github for ways to get started. See support.md for ways to get help.

This project has a code of conduct. By interacting with this repository, organization, or community you agree to abide by its terms.

License

MIT © Titus Wormer