html-bust

html-bust

Cache-busts URLs in HTML files by appending a query string with a hash of the referenced resources. Similar to grunt-cache-bust, but runs on its own.

Because parsing and re-marshaling HTML documents is brittle, URLs are replaced through regular expressions. As a way to mitigate against unwanted substitutions, URLs that are to be busted can be marked with a special suffix.

Usage

var bust = require('html-bust');

bust(inPath, outPath, [options], [done])

Rewrites the HTML file inPath to outPath, cache-busting URLs by appending a query string with a hash of the referenced resource. Relative URLs are resolved against the location of the HTML file; absolute URLs are ignored.

Available options:

• tagTypes (default: [ 'img', 'script', 'link' ])

  An array of HTML tag types to cache-bust. Currently, img, script and link are supported. Tags of other types are ignored.

• urlHint (default: '?bust')

  If set to a string, only URLs ending with that string are busted. The string is removed from the processed HTML file. If null, URLs are busted inconditionally.

• mode (default: hash)

  One of hash or string. In hash mode, references are busted with a hash of the respective file contents, according to the hashAlgorithm and hashString options. In string mode, references are busted with the fixed string given by the fixedString option.

• hashAlgorithm (default: 'sha1')

  The hash algorithm to use in hash mode. Check the documentation for crypto.createHash.

• hashLength (default: 8)

  The number of hash characters to use in hash mode as the cache-busting string.

• fixedString (default: '')

  The fixed string to use in string mode.