Adminlte: Contributing to Bootstrap 4

[JayteeNZ]

I wanted to start contributing to the Bootstrap 4 branch. I have a couple of questions:

1. What about moving away from bower_components to node_modules ?
2. What components are required to stay
3. What components are required to be removed
4. What components are required to be added / modified
5. If v3 going to take the same shape as v2? E.G: as in the same styles, or perhaps suggest a new theme completely or modify just a little bit
6. Icons: This is my personal preference, but I now often just pull in icons as i go, either from Zondicons.com, or the new Heroicons UI set. Is font-awesome still going to be a standard approach?

[schmunk42]

Additional question: Will a v4 version of AdminLTE use https://getbootstrap.com/docs/4.0/examples/dashboard/# as a base implementation?

[EliuTimana]

we should start using webpack too 😃

[JayteeNZ]

Here's a roadmap that i'm working on for AdminLTE v3. Some of these might not be wanted/required in the V3 version, but i'll do a PR when i'm ready and a decision can be made:

• use webpack
• use official bootstrap 4 stable
• change project structure to be more like bootstrap 4 (e.g: scss as a parent directory instead of within the build directory) Then for people using scss in their custom projects, they can just do something like: @import "~AdminLTE/scss/AdminLTE". The same goes for JavaScript, just a require('adminlte') to load all js
• rename skins to themes (it's more evident to some people of what it is)
• add a card variant for boxes

• remove the use of bower_components. use node_modules and an optional plugins directory for those downloading the repository

• change documentation to something like Jekyll with AdminLTE theme??

• various other changes

@almasaeed2010 What do you think?

[schmunk42]

rename skins to themes (it's more evident to some people of what it is)

I'd keep skins, since AdminLTE is a theme, but its different colors are skins.

Just my 2 cents ;)

[JayteeNZ]

@schmunk42 Yep okay,

What about everything else i've listed? Cool with that? Also, anything else to add in mind?

I think one of the bigger things with bootstrap 4 is the use of utility classes on components (e.g: AdminLTE has a mixin for progress bar for the different colours, but now BS4 makes use of .bg-{colour} instead).

I figure, keeping everything to the same sort of standard BS4 is would be best, and a bonus of it is it removes a lot of no-longer-needed styles.

Oh and I was gonna pull in Laravel Mix instead of straight webpack, I think for a lot of developers it would be useful since it offers a lot of different compiling (scss, less, stylus, vue, react etc). Problem is, some might think of it as unnecessary even though it's actually fairly straight-forward and great. So stick with webpack, or laravel mix?

[schmunk42]

Sorry, haven't reviewed all the other points in detail.

So stick with webpack, or laravel mix?

If changes to the current build process are required then webpack.

[almasaeed2010]

Hello @JThompsonNZ

It's great to see enthusiasm about AdminLTE 3 :D

You bring up some good points here. I am still in the process of deciding how to structure v3 and what tools to use. I plan on opening a discussion "issue" soon to address the compiling process and other things like the upgrade process and what it entails.

As to contributing to building v3, I think it's best to do that when we have an agreed upon base to build from. Once that's in place, I'd appreciate all the help I can get to upgrade pages to use Bootstrap 4 and AdminLTE 3 elements.

Here is what I think regarding Laravel-mix vs WebPack. I love Laravel Mix and really wish I could use it here but as you mentioned it's somewhat of an overkill and not necessary for many projects. I started implementing a WebPack solution for v3 already that I will push to v3-dev branch soon.

Thanks! Let's continue this discussion once we have a good base for v3.

[pioz]

Any expected date for Admin LTE with bootstrap 4?