asset-cache-bust/README.md

Bust your asset cache using querystring hashes based on actual asset content

# Asset-cache-bust
Bust the browser cache of your assets using query strings based on actual asset content.

## What it does
This module parses HTML and finds `<link>` and `<script>` tags that contain a `href` or `src` 
and `data-finger-print` attribute. It then extracts the URL from those attributes  and 
tries to read those files from disk. If a file was successfully read, it uses its content 
to create an MD5 hash. This hash is then used to append to the original URL as a query 
parameter in the form of: `v=md5stringGoesHere`. When all matched URLs have been replaced in 
the original HTML, it will be passed to the success callback.


## How to use it
This module can used as a node module or through its CLI.

> The given HTML should contain `<link>` or `<script>` tags in the following format for this package to
> replace their URLs: `<link href="asset.css" data-finger-print />` or `<script src="some.js" data-finger-print></script>`.

To use it as a node module, install it using: `npm install asset-cache-bust`

See following example:
```
const assetCacheBust = require('asset-cache-bust');
assetCacheBust(validHtml)
    .fork(
        error => console.log(error), // Do something usefull with the error
        cacheBustedHtml => console.log(cacheBustedHtml) // Do something with the cache busted HTML
    );
```

To use the CLI you have 2 options:

* Install this package globally:
    1. Install this module using `npm install -g asset-cache-bust`
    2. Cache bust html files using `asset-cache-bust *.html` for example
    > type `asset-cache-bust --help` for help and options
* Install this package locally:
    1. Install this module using `npm install asset-cache-bust`
    2. Create an npm script in your `package.json` file with a fitting name and corresponding value.
    e.g. `"bust-cache": "asset-cache-bust *.html"`
    3. Run your npm script using `npm run <script name from above goes here>`
    
> The default behaviour of the CLI is to replace the given HTML files with the cache
> busted version. See the CLI help for more information and options.

> If no matching tags were found and thus no replacements can be made, no files will be
> changed.

## Contributing
I will create a seperate markdown file with documentation for contribution if
I get requests from people who would like to contribute.

## Licence
MIT (See LICENCE file)