Re: git: 42a2e855eff0 - main - lang/purescript: fix build on powerpc64le

On 26-06-24 11:40:31, Gleb Popov wrote:
> On Wed, Jun 24, 2026 at 11:22 AM Piotr Kubaj <pkubaj@freebsd.org> wrote:
> >
> > The branch main has been updated by pkubaj:
> >
> > URL: https://cgit.FreeBSD.org/ports/commit/?id=42a2e855eff014b2603fcf8318e898ddb8cfb6c1
> >
> > commit 42a2e855eff014b2603fcf8318e898ddb8cfb6c1
> > Author:     Piotr Kubaj <pkubaj@FreeBSD.org>
> > AuthorDate: 2026-06-24 08:20:56 +0000
> > Commit:     Piotr Kubaj <pkubaj@FreeBSD.org>
> > CommitDate: 2026-06-24 08:22:35 +0000
> >
> >     lang/purescript: fix build on powerpc64le
> >
> >     powerpc64le is a 64-bit architecture as well.
> > ---
> >  .../files/patch-__cabal__deps_tasty-1.5.3_tasty.cabal         | 11 +++++++++++
> >  1 file changed, 11 insertions(+)
> >
> > diff --git a/lang/purescript/files/patch-__cabal__deps_tasty-1.5.3_tasty.cabal b/lang/purescript/files/patch-__cabal__deps_tasty-1.5.3_tasty.cabal
> > new file mode 100644
> > index 000000000000..67950ca8d8e2
> > --- /dev/null
> > +++ b/lang/purescript/files/patch-__cabal__deps_tasty-1.5.3_tasty.cabal
> > @@ -0,0 +1,11 @@
> > +--- _cabal_deps/tasty-1.5.3/tasty.cabal.orig   2026-05-29 10:05:13 UTC
> > ++++ _cabal_deps/tasty-1.5.3/tasty.cabal
> > +@@ -69,7 +69,7 @@ library
> > +     ansi-terminal        >= 0.9  && < 1.2
> > +
> > +   -- No reason to depend on unbounded-delays on 64-bit architecture
> > +-  if(!arch(x86_64) && !arch(aarch64) && !arch(ppc64) && !arch(s390x) && !arch(riscv64))
> > ++  if(!arch(x86_64) && !arch(aarch64) && !arch(ppc64le) && !arch(ppc64) && !arch(s390x) && !arch(riscv64))
> > +     build-depends:
> > +       unbounded-delays >= 0.1 && < 0.2
> > +
> >
> 
> The 1.5.4 release is out and it seems to contain a fix for that:
> https://github.com/UnkindPartition/tasty/commit/88f435b6374f62e9c87919fec9c9106ba26f80ac
> Does simply switching to tasy 1.5.4 work?
> 

Yes, it does. But since tasty 1.5.4 probably also brings other fixes, it
would change the package and require bumping PORTREVISION, so I wanted
to avoid doing that. The fix I committed is noop for other
architectures.