rm-old-base: remove ifdefs for pre-4.13 bases

[NadiaYvette]

Template B: This PR does not modify behaviour or interface

4.13 and older have fallen out of the support window. Hence this commit removes code only conditionally included for base 4.13 and older. Some occasional transitive removals were implied and done in this same commit.

• Patches conform to the coding conventions.
• Is this a PR that fixes CI? If so, it will need to be backported to older cabal release branches (ask maintainers for directions).

[geekosaur]

You're supposed to pick a template based on whether your PR changes the user-visible or programming API. This one should not, so you would use template B.

[andreabedini]

LGTM. Thank you so much! <3

[andreabedini]

@fgaz Can you have another look when you have time?

[andreabedini]

@fgaz @ulysses4ever @Mikolaj @Kleidukos could you give a review?

[NadiaYvette]

@fgaz @ulysses4ever @Mikolaj @Kleidukos could you give a review?

Thank you so much Andrea! That's a lot of LOC you put into the fixups!

[geekosaur]

It might be worth looking for more instances of FOURMOLU_DISABLE and see if CPP has been removed; in general, CPP causes fourmolu to choke.

[Kleidukos]

Yes! Less CPP makes happier tooling

[Mikolaj]

I see "base >= 4.11" in Cabal/Cabal.cabal. Can we update it, too? Probably in a separate PR, because it would modify the interface. It would also need a little changelog file stating what it does.

[andreabedini]

Maybe something went wrong with the rebase I had changed the bounds the cabal files in my commit acffcd4.

Edit: I pushed my last commit (9ce4213) to https://github.com/andreabedini/cabal/tree/rm-old-base-001

[fgaz]

Some of my previous comments are still unresolved, I marked them as such.

[Mikolaj]

Maybe something went wrong with the rebase I had changed the bounds the cabal files in my commit acffcd4.

Edit: I pushed my last commit (9ce4213) to https://github.com/andreabedini/cabal/tree/rm-old-base-001

Right here it's as it should be. That's rather disturbing. Was this mergify's rebase or your manual one? Can you recover from this corruption?

[andreabedini]

@fgaz @Mikolaj I restored the lower bounds on base.

[ulysses4ever]

@fgaz you requested changes in this PR, and the issue, they claim, was fixed (I didn't check). Could you see if you could reevaluate it and maybe approve?

[geekosaur]

FWIW everything except the import is marked Outdated, suggesting they've already been fixed. I'm validating locally with the import removed. (ETA: build fails without the import.)

[geekosaur]

I stand corrected; I misunderstood what was being requested, and removed the wrong thing. Testing again now. (ETA: builds with 9.6.6; trying with 8.8.4 locally now.)

[fgaz]

@fgaz you requested changes in this PR, and the issue, they claim, was fixed (I didn't check). Could you see if you could reevaluate it and maybe approve?

All opened conversations (3 at this time) are still unresolved

[geekosaur]

[27 of 51] Compiling Distribution.Solver.Types.ConstraintSource ( src/Distribution/Solver/Types/ConstraintSource.hs, /home/allbery/Sources/cabal/dist-newstyle/build/x86_64-linux/ghc-8.8.4/cabal-install-solver-3.13.0.0/build/Distribution/Solver/Types/ConstraintSource.o )

src/Distribution/Solver/Types/ConstraintSource.hs:48:3: error:
    parse error on input ‘-- | The source of the constraint is not specified.’
   |
48 |   -- | The source of the constraint is not specified.
   |   ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

[geekosaur]

@fgaz, I think we've dealt with the remaining issues, could you confirm?

[andreabedini]

@geekosaur @fgaz @ulysses4ever Thank you for the help. I am afraid some changes kept getting lost in the rebases :-/

[fgaz]

Looks good!