@jonschlinkert I'm not sure of the exact meaning of the question but will interpret just as how you could have a local development environment where Assemble builds templates/watches & reloads templates on changes, and where Webpack builds JS/CSS and does some livereload/hot reload. Then in prod you might hash your asset paths for your JS/CSS and then build HTML that puts these asset paths into <link> and <script> tags.

The tasks in this repo show a bit of an overly complicated example https://github.com/build-boiler/making-gulp-suck-less/tree/master/build/tasks but essentially if one uses a tool like webpack-isomorphic-tools or some other webpack stats generation plugin to generate stats on their webpack compiled assets then they can consume these stats via their templating language either by loading them into the assemble context or just reading them inside of some sort of helper/template tag. In this case I made a nunjucks tag that gets the stats off of the template context https://github.com/build-boiler/making-gulp-suck-less/blob/master/build/tasks/assemble/tags/get-assets.js so in the HTML you could do something like

<head>
  {% get_asset type="css" %}
</head>
<body>
  {% get_asset type="js" %}
</body>

Everything else is just typical config for assemble and webpack, depending upon how you want to setup your local environment.

I hope that helps. The repo example would most likely be difficult to get up and running as it is a mono repo using lerna but I think it would work on first clone of the repo via

npm i -g gulp-cli # uses gulp 4
npm i
npm run lerna:bootstrap
gulp watch

The process could be stepped through if someone installs iron-node:

npm i -g iron-node@latest
# place a debugger somewhere in the webpack or assemble task
iron-node `which gulp` <task_name>

Uh oh!

Usage Exemple for webpack #16

Description

Metadata

Metadata

Assignees

Labels

Type

Projects

Milestone

Relationships

Development

Issue actions