Open PatrykWalach opened 12 hours ago
should we name the field link
? I think __link
looks ugly... :-D
And do we fail if there's a link already, or do we ignore? Ignoring will panic when adding server pointer, but we can warn in that case.
type Link is not imported right know, but it's not an issue since client pointers are not implemented yet
And this is an issue, because __link
can be selected manually so I guess I should fix this too
I think we should panic with a helpful message. Eventually we can allow the user to provide a directive in a schema extension to rename the link
field if they have a clash
And maybe once we have namespaces, we can have an "internal" namespace, and the user can selected internal.link
and internal.asUser
etc or something
No more conflicts! Idk
This is a bit awkward at places because
Loadability
, should we do that?__link
\@loadably, we should validate it as part of this PRtype Link
is not imported right know, but it's not an issue since client pointers are not implemented yetother then that It seems to work