Open bbarker opened 3 years ago
If you use unexceptionalid
functions to build a stack based on UIO
I'm curious how you end up with a SomeException
at all? Should always end up with at most a SomeNonPseudoException
because PseudoException
should not be caught in UIO
context.
Perhaps I went wrong somewhere, but here's what I did:
xsd <- liftEither $ parseText def $ TL.fromStrict xsdTxt
Due to parseText
(being a function from a 3rd party library) possibly returning Left SomeException
as part of it's signature, I then needed a function like mapZErrorOrExit :: ZIO r SX.SomeException a -> ZIO r SomeNonPseudoException a
to absolve the error as in:
xsd <- mapZErrorOrExit $ liftEither $ parseText def $ TL.fromStrict xsdTxt
I can imagine other alternatives, though I haven't really imagined any obviously better alternatives so far.
Oh interesting. That's a case I guess I hadn't considered: when a library already catches/returns exceptions but doesn't use anything more specific than SomeException
.
Here's an option:
xsd <- UIO.fromIO $ UIO.runEitherIO $ return $ parseText def $ TL.fromStrict xsdTxt
I recently had (possibly good) cause to have a function like this:
The context is that I can use seOrExit with a
mapError
that works on my monad transformer stack that is built on top of UIO. If you're chugging along, and get a psuedo exception, I'm not sure what else there is to be done about it anyway, other than performing an alternative form of logging.Anyway, wanted to get your thoughts on this use case. Maybe this is terrible.