[v4] Fingerprinting the output html file name with a hash determined by the html contents conflicts with webpack's `contenthash` variable

[pegiadise]

Expected behaviour

It should fingerprint the output html file name with a hash determined by the html contents after the scripts have been injected

Current behaviour

Webpack throws error:

UnhandledPromiseRejectionWarning: Error: Path variable [contenthash] not implemented in this context: html/app.[contenthash].html

Environment

node: v8.11.2
npm: v5.6
yarn: v1.9.2
webpack@4.17.1
html-webpack-plugin@4.0.0-alpha

Config

module.exports = {
  ...,
  plugins: [
    new HtmlWebpackPlugin(
      template: './app/assets/html/app.html',
      chunks: ['app-bundle', 'vendor~app-bundle'],
      filename: IS_DEV ? 'index.html' : 'html/app.[contenthash].html'
    ...
  ]
}

Similar issues

There was a similar behaviour from extract-text-webpack-plugin in the past when webpack-4 introduced this internally in version 4.3.0:

webpack-contrib/extract-text-webpack-plugin#763

[jantimon]

Do you have any idea how we might fix that?

[pegiadise]

The simplest approach would be my PR which does nothing else but renaming the key there. However, it is up to you for the long term solution. Thanks for the awesome plugin! 🎉