Github user ahgittin commented on the issue:
https://github.com/apache/brooklyn-server/pull/338
Very cool. Haven't reviewed in depth but looks on point at a high level.
A few specific comments:
* this will conflict with
https://github.com/apache/brooklyn-server/pull/573 -- should be easy to
resolve, and this is a better solution to the underlying problem
* I think we want _both_ `catalogItemId` with the semantics of #573 and the
`stack` introduced here for searching, on the spec (that's what we have on
entity)
* there may be persistence errors migrating from older version if we
persist specs and remove the `catalogItemId` here (?); the previous suggestion
to keep both should resolve this (perhaps with an additional check when
initializing an entity that if `catalogItemId` is set then `stack.get(0)`
equals that id
And one general comment -- there is an issue if e.g. `my-cluster` and
`my-node` are set in the same BOM, both at version `1` with the former
referring to the latter omitting the version. If we then install a version `2`
of that bom, then `my-cluster:1` creates a cluster of `my-node:2` which is
surprising. It would be nice to solve that (and also be able to retrieve the
`bom` file and bundle list where an entity is defined, for subsequent
editting). It's a separate problem to this, but the solutions may be
interrelated. Very happy to have this merged (once someone else does a
detailed review and we resolve #573 conflicts) without solving the version and
catalog-group / bom questions, expecting a bit more change afterwards, but
wanted to flag that. cc @geomacy @neykov @aledsage
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
