Andreas Abel

Even though this issue is older as #10379, I am marking this one as duplicate since #10379 describes the problem (custom setup) already in the OP.

Me too, trying @phadej's example at: - https://github.com/haskell-CI/haskell-ci/pull/582#issuecomment-1113991712 ``` cabal v1-haddock --all ``` leaves the HTML docs in `dist/...`. ``` cabal v2-haddock --haddock-all ``` succeeds but does not produce any...

`new-` prefixes, yes, certainly should go, because what was new in the past isn't new today. However, I see no advantage in removing the `v2-` prefixes (as alternative to no...

Well this issue is about deprecation in 3.12 and removal in 4.0. I am not opposed to deprecation warnings, but what purpose do they serve if there is no removal...

@gbaz wrote: > the original plan was _never_ to remove the `v2-` prefix. Good plan! One can think about the prefixes in the following way: With `v1-` and `v2-` (and...

Pondering whether to implement `--set-rev=N`, I got doubts that it is worthwhile: - It only sensibly applies to pushing a single cabal file (but maybe exactly this was the intended...

Thanks for the contribution! So when I do `cabal run all -- get-bounds hackage-cli.cabal` it prints nothing. Am I doing something wrong or have a misunderstood the description of the...

E.g. `cabal run all -- add-bound aeson '>=2.0' hackage-cli.cabal` just reports an error: ``` Cannot find library section in hackage-cli.cabal ``` Likely it does not work with named (internal) libraries....

I added a commit that removes some code duplication you introduced in the previous commits. It wasn't hard to do the necessary factoring for this. Maybe next time you can...

`cabal check` says: > Common stanza imports should be at the top of the enclosing section I'd say this is also an upstream problem; - https://github.com/haskell/cabal/issues/5564 it is questionable for...