recrawl

recrawl v1.1.1

Collect the descendants of a directory.

import { recrawl } from 'recrawl'

// Create a crawl function.
// These are the default options.
const crawl = recrawl({
  only: [],
  skip: [],
  deep: true,
  follow: false,
})

// The result is an array when `follow` is false, else an object.
const files = crawl(root)

// Provide your own array/object.
crawl(root, files)

// Provide an iterator.
crawl(root, (file, link) => {
  // The `file` argument is relative to the root.
  // The `link` argument is null for non-symlinks. It will be absolute if the target is outside the root.
})

You can use the crawl() export if you don't want to reuse the configured crawler.

import { crawl } from 'recrawl'

crawl(root, {
  only: [],
  skip: [],
  deep: true,
  follow: false,
})

Options

• only?: string|string[]
• skip?: string|string[]
• deep?: boolean
• depth?: number
• enter?: function
• filter?: function
• follow?: boolean|number|function
• adapter?: FileAdapter

The only and skip options should be self-explanatory. Paths matching any of the only patterns are good. When only is an empty array, all paths are good. Paths matching any of the skip patterns are bad. When skip is an empty array, no paths are bad. The skip patterns override the only patterns.

To avoid crawling sub-directories, set deep to false or depth to 0. You should never define both deep and depth, because the depth option implies deep when it's greater than zero. If neither deep nor depth are defined, the default depth is infinite.

The enter option is called whenever a directory is encountered. It's passed the directory path and the current depth. You may return a falsy value to avoid crawling a directory.

The filter option is called whenever a filename is encountered. It's passed the filename and its basename. You may return a falsy value to skip a filename. The only and skip options are applied before this option is called.

To follow all symlinks, set follow to true. For greater control, use a function. It's called whenever a symlink is encountered. You may return a falsy value to avoid following a symlink. It's passed the symlink path and the current link depth. If you only need to limit the link depth, you can set follow to a number, where zero is equivalent to false.

The adapter option lets you provide your own filesystem.

Gotchas

• Directory symlinks are treated the same as real directories
• Directories are not affected by the only option

Pattern syntax

Recrawl has its own take on globbing.

1. When a path has no separators (/), only the basename is matched.

'*.js' // matches 'a.js' and 'a/b.js'

1. Recursivity is implicit.

'a/b' // identical to '**/a/b'

1. Use a leading separator to match against the root.

'/*.js' // matches 'a.js' not 'a/b.js'

1. Use a trailing separator to match all descendants.

'foo/' // matches 'foo/bar' and 'foo/bar/baz' etc

1. Regular expression syntax is supported. (except dot-all)

'*.jsx?' // matches 'a.js' and 'b.jsx'
'*.(js|ts)' // matches 'a.js' and 'b.ts'

1. Recursive globbing is supported.

'foo/**/bar' // matches 'foo/bar' and 'foo/a/b/c/bar' etc