Julian Rohrhuber
Julian Rohrhuber
To me it seems that this is a good occasion to remove some boundary cases that are confusing both in use and implementation. Unfortunately, this means a bit of breakage....
It is good to enforec a bit of semantic consistency here.
> Additionally — in another PR — we could actually test that **all** classes that have a 'subclass responsibility' method in their parent (and no child classes) actually implement it....
> One thing to note immediately is that isAssociationArray isn't define on many classes — this is probably a mistake because its a boolean, but it doesn't have any documentation....
I haven't requested another review. There ad been just one thing pending which was resolved, and I approved. So far, the idea was that the person who approves does not...
> > If you approve a PR but do not merge it please leave a comment what are the missing steps towards a merge (e.g. "looks fine to me, but...
> This could be a little surprising in some circumstances, but I don't _think_ it will cause any confusion? So this means there are cases in which you call *newCopyArgs...
Yes, that is unchanged – thanks fo rthe clear example, I think this should go somewhere in the help files. What I meant was the `Could not find instance variable`...
Sorry! Yes, we want this warning. If this warning is always there when we want it to be there, all is well.
yes, ready to merge, maybe someone else can take a brief look and merge.