Update MapLibre GL JS to eeefc070d7333b21b205be3882df9b2e3cd444d1

[github-actions[bot]]

[wipfli]

The only render test that fails here is render-tests/icon-text-fit/enlargen-both, which already fails on main. So far so good...

[ntadej]

Can we get the tests in the CI first :)

[wipfli]

#357 aims to do that, yes

[ntadej]

Also let's maybe tag a release before that.

[wipfli]

We should be able to merge this once we have render tests in CI #357

[wipfli]

Converting to draft until then...

[birkskyum]

@wipfli The render tests are merged, so can this be converted back from draft now?

[wipfli]

I think the render tests should fail on this one so maybe it is better to keep it as a draft. But lets see...

[birkskyum]

What workflow does the render test run in?

[wipfli]

I think it is node

[wipfli]

https://github.com/maplibre/maplibre-gl-native/blob/ad91ff235d2a25e86ed3e94b2ce7593fdc10ed8b/.github/workflows/node-ci.yml#L166-L175

[birkskyum]

Great, so that looks like a success so far

[ntadej]

I think we should still tag core before we start updating JS.

[birkskyum]

@ntadej , I'm all for it if you think it will improve maintainability, but what does it look like exactly? It was a bit tedious to update the maplibre-native-base upstream of the core, and with this, would we then have a 3-repo long chain if it becomes it's own submodule sitting between the base and the sdk's? Could the core maybe merge with the base in this refactoring? (is there a discussion thread about this anywhere?)

[ntadej]

The discussion is kind of here: https://github.com/maplibre/maplibre-gl-native/discussions/371

The issue I see is that we are nearing a relatively stable native "core". It may be good to make a snapshot so people can stay with this version for a while. Maybe it would even be a motivation to release all SDKs so tagging also @roblabs for his opinion.

[ntadej]

I guess this PR can be closed and branch deleted?

[birkskyum]

yes, without maplibre-gl-js in Native, it makes little sense to keep.