Xia Li-yao

Now that I think about it, you may want manage the installation of Why3 and Coq manually with a global opam switch instead, because it will make it less of...

- Recent features (`#[logic(inline)]`) make `compute_specified` much more effective to simplify the goal. - We don't have plans to explicitly support working with Rocq, but in principle it's just a...

It may be related to the fact that `PutObject` is a record type but gets rendered as a non-record type.

What's the issue with windows builds? I don't recall one in recent memory.

Thank you so much!

The failing test `t_toUpper_char` is guarded by a [`MIN_VERSION_base(...)` condition](https://github.com/haskell/text/blob/3cd9ed3c1d16ba18986bccbafbb34a4c0b89c564/tests/Tests/Properties/Text.hs#L143) which goes out of date almost every time we upgrade the Unicode version for the case mappings in text. -...

I rebased it. This looks pretty good to me so I propose to merge this soon as long as the tests pass.

Thanks a lot @lweqx !

The details are not settled, but right now that applies up to the next `QCPreprocess` command, so you can undo it.

Indeed, the instances for `(:*:)` (the `GHC.Generics`-specific variant of `Product`) are also too strict: https://hackage.haskell.org/package/ghc-internal-9.1001.0/docs/src//GHC.Internal.Generics.html#local-6989586621679638273 This can only be fixed in base.