Don't add supervisor more than once. #9962
Conversation
| final def ++[B](that: Supervisor[B]): Supervisor[(A, B)] = | ||
| Supervisor.Zip(self, that) |
There was a problem hiding this comment.
I don't think you need patchAdd - we should just change the behavior of ++.
There was a problem hiding this comment.
Yep, we can, but it that requires making the return type Supervisor[Any]. While nothing in ZIO seems to actually use the returned tuple, it's a public API so we can't change it now.
There was a problem hiding this comment.
@quelgar this method is still being used in 2 places:
zio/core/shared/src/main/scala/zio/ZIO.scala
Line 2040 in c6f342e
The second one I'm not sure we need to change it but the first one definitely.
There was a problem hiding this comment.
Yep changed the first one, thanks.
For the second, yeah it's not actually adding anything, it's just rebuilding based on the existing graph, so no need to change it 👍 .
kyri-petrou
left a comment
There was a problem hiding this comment.
LGTM, just a minor comment
| final def ++[B](that: Supervisor[B]): Supervisor[(A, B)] = | ||
| Supervisor.Zip(self, that) |
There was a problem hiding this comment.
@quelgar this method is still being used in 2 places:
zio/core/shared/src/main/scala/zio/ZIO.scala
Line 2040 in c6f342e
The second one I'm not sure we need to change it but the first one definitely.
b4b6937 to
0678ec6
Compare
See #9668 for rationale.
This adds
Supervisor#addPatchthat works the same as++except: if the supervisor to be added is already present, the current supervisor is returned unchanged. This is used to implementAddSupervisorpatching, so that complex layer graphs never result in a supervisor being added more than once.@kyri-petrou please take a look at your convenience.