Closed ocharles closed 2 years ago
I think your best option might be to depend explicitly on aeson, and then you can put a suitable bound on (ideally updating code to support aeson-2). If you agree with this, I'd be happy to do the work and send a PR.
Thanks, that sounds good.
This is a bit of a weird one! Because
fakedata
depends onyaml
, it indirectly depends onaeson
. However,yaml
just re-exports combinators like.:
fromaeson
. Therefore, depending on the solvers choice ofaeson
,fakedata
might or might not build. Ifaeson-2
is chosen, thenfakedata
will fail to build with:(This is just an example, many many modules fail with similar errors).
I think your best option might be to depend explicitly on
aeson
, and then you can put a suitable bound on (ideally updating code to supportaeson-2
). If you agree with this, I'd be happy to do the work and send a PR.