Plan for a "less ambitious" version of D8 Libraries API [#3039570]

Over in #2939919: Tag an 8.x-3.0 release, the following conversation took place re: the future of the D8 Libraries API module:

@tstoeckler

I am not really happy with the current state of the module, so I am actually not comfortable with creating a stable release. We did have some quite ambitious plans for this module at some point, unfortunately we never achieved them. But even if we decrease the scope of the module I'm not happy creating a stable release with all the stuff that is already in there that is broken. So I am going to mark this issue postponed for now. If any of the other maintainers feel differently I certainly don't want to hold anyone up.

@sun

If possible in some way, we should release a tagged release of 8.x-2.x in some stable-ish form to allow the whole (Drupal) project and community to make progress on many other fronts. The comments on this issue sound like that release could even be a 8.x-2.0.0 after quickly cleaning up some experiments? All code targeting a major rewrite / refactoring should be postponed to a new major 3.x at this point in time of the Drupal Core release cycle.

@tstoeckler

So the problem is that the asset system got drastically revamped between D7 and D8 so a straight port is simply not possible. That's not just in terms of removed/revamped APIs, but conceptually Libraries API cannot work as it did in D7, as it relied on the runtime adding of assets (drupal_add_js/css()) which got completely removed (i.e. not just replaced with an OO-version). If someone has concrete plans/ideas or even code for a less ambitious version of a D8 Libraries API than what is currently in 8.x-3.x I'm fine with going for a 8.x-2.x, but so far that simply doesn't exist.

The goal of this issue is simply to discuss and establish "concrete plans/ideas for a less ambitious version of a D8 Libraries API.

Comments

Comment #1

12 March 2019 at 19:50

Chris Matthews created an issue. See original summary.

Comment #2

Chris Matthews CreditAttribution: Chris Matthews as a volunteer and at City of Oaks Design commented 11 April 2019 at 04:04

Would it be possible to simply take the current 8.x-3.x-dev, plus the pending commits from the current 8.x-3.x issues that have RTBC status, and create a full 8.x-2.0 recommended release that can be more easily developed and maintained by the community?

Comment #3

rooby CreditAttribution: rooby commented 11 April 2019 at 22:59

Not that it's the main concern, but it would probably be better to go with 4.x instead of 2.x I think, regardless of how similar it is to the D7 2.x branch or whether it's simpler than 3.x.

More and more people are starting to use the current version though so it would definitely be good to either make a proper release or tell people not to use it at all, if that's the recommendation.