Let transforms consider mixin types #214
Conversation
ghost
assigned 
| final NodeType primaryNodeType = node.getPrimaryNodeType(); | ||
|
|
||
| final NodeType[] supertypes = primaryNodeType.getSupertypes(); | ||
|
|
||
| final Iterable<NodeType> nodeTypes = | ||
| concat(of(primaryNodeType), copyOf(supertypes)); | ||
| final Set<NodeType> nodeTypes = |
There was a problem hiding this comment.
What sort of ordering guarantee do we get here? At the very least, are the results deterministic? I'm not sure how we ought to balance the priorities of primary vs mixin types, but we should at least have consistent behavior.
There was a problem hiding this comment.
We're not guaranteeing right now, and that was okay for my simple tests. I'll change to a List to guarantee mixins first, then primary type, then primary's supertypes. That would be best for the indexing workflow, which is the chief use case for fcr:transform right now.
Do you think I should order mixins by name and supertypes by name, just to get more predictability? Do you think we should include the supertypes of mixin's (which are formally types of the node as well, just as the supertypes of the primary are)? I think the answer to both questions is yes, but I'll defer to you.
There was a problem hiding this comment.
Thanks. That seems good enough for now, I think. There's plenty of time to be clever later.
Let transforms consider mixin types
https://www.pivotaltracker.com/story/show/62512344