Path information in the Drs Object

[nmalfroy]

Hi all, Not sure if I'm missing something so feel free to correct me.

It seems that the spec allows for, essentially, a directory structure that lets you nest objects. My question is: given this, it seems that we should provide: a) a path and / or b) a parent id This means that given a drs id, you can't get a sense where in the dataset hierarchy a given object lives.

Is that something that's planned in the spec at some point or am I thinking about this wrong?

The usecase is that a user may get a DRS ID for a file that has a dependency on another file in the same "directory" and with the current spec, I don't really know how to model that without the starting file knowing what it's "directory" even is.

[nmalfroy]

In the mean time, we're using the aliases field to store this path information but that feels wrong. Is there an accepted pattern where we could a an extra namespaced field (e.g. x-object-path) to the DrsObject? Are we allowed to add fields to objects? I couldn't find anything in the spec saying one way or another

[kv076]

@briandoconnor - Any take on this issue? I think this is a suggestion for this in DRS 1.3?