haplo-org / haplo-plugins Goto Github PK

Above service allows other plugins to add roles as a last resort/if it can't fit the model - however any plugin adding roles this way doesn't have a way to flush the cache, so as far as I can see you may make changes to data that causes a cached role (added by add_roles_to_user service) to be removed, but the permissions/role will persist until something else triggers a cache flush?

Think implementing an invalidation service available for other plugins to call solves this.

Not sure if this is a known thing or not, but proves to be a little bit of an issue when setting up new selfLabelled child objects.

I've provided some test schema and code for reproduction, but worth noting I had the same issue/outcome using Research Institutes in haplo_research_manager (ie, adding a department to a faculty)

type test:type:test-one as TestOne
    title: Test one
    search-name: test one
    attribute dc:attribute:title [sort=1000]
    attribute std:attribute:parent [sort=1100]
    attribute std:attribute:notes [sort=1600]
    render-type book
    render-icon: E210,1,f
    render-category 0
    create-position common

P.implementService("haplo:descriptive_object_labelling:setup", function(type) {
    type(T.TestOne, {
        selfLabelling: true
    });
});

Repro:

1. Create object of type
2. Create object of same type, use object created above as A.Parent
3. New object will have labels: T.TestOne and itself but not the Parent object created in 1)
4. Edit object, Save without changes, parent will now be included