🇩🇪Germany @Harlor

harlor → created an issue.

harlor → created an issue.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

harlor → created an issue.

harlor → created an issue.

harlor → created an issue.

harlor → made their first commit to this issue’s fork.

Sorry accidentally changed status...

Great! - @graber are you planning to release a 4.3.2 for this?

harlor → created an issue.

harlor → made their first commit to this issue’s fork.

harlor → created an issue.

harlor → made their first commit to this issue’s fork.

Code looks good.

I think we should add a custom error message though - if one uses a invalid slot-id currently the message reads:

The machine-readable name must contain only lowercase letters, numbers, and underscores.

Which is not correct because we allow "-" too

harlor → created an issue.

harlor → created an issue.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

Yes - I think I wanted to avoid merge conflicts :D

harlor → created an issue.

harlor → created an issue.

harlor → created an issue.

harlor → created an issue.

This makes sense to me! I made a MR for the suggested changes.

harlor → made their first commit to this issue’s fork.

harlor → created an issue.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

harlor → made their first commit to this issue’s fork.

Remove the implicit dependency on devel and avoid PHP warnings.

harlor → created an issue.

harlor → made their first commit to this issue’s fork.

harlor → created an issue.

@fmb, I tried to write a draft for the suggestions in #9

Is this going into the right direction?

fmb → credited harlor → .

I discussed this with @marcus_johansson - the main points were:

- An agent for enabling/disabling modules makes sense
- Enabling modules is easy to achieve
- When disabling modules one must make sure that the user is informed about consequences (Content loss)

In order to get things done I'd reduce the scope to enabling modules which seems to work:

harlor → created an issue.

I successfully enabled a module with the prototype already :D

But it needs cleanup and the disable action is missing too.

harlor → created an issue.

harlor → changed the visibility of the branch 3473821-missing-cache-context to hidden.

Since the token data can vary per URL (entity) it makes sense to me that the cache should respect that.

Harlor → created an issue.

Harlor → created an issue.

Not sure how the test results have to be interpreted but here is the MR.

Harlor → created an issue.