From: Stuart Henderson Subject: Re: lang/zig moved to Codeberg To: Thomas Frohwein Cc: Volker Schlecht , ports@openbsd.org, semarie@openbsd.org, bmercer@eutonian.com Date: Mon, 30 Mar 2026 10:00:54 +0100 On 2026/03/29 22:53, Thomas Frohwein wrote: > On Sat, 28 Mar 2026 18:04:02 +0000 > Stuart Henderson wrote: > > > > +WRKDIST = ${WRKDIR}/zig > > > > this one is so annoying, it seems like it should be simple to fix > > so that dist-tuple.port.mk sets it correctly, but nothing I try > > works... (all the ports using codeberg DIST_TUPLE explicitly set > > WRKDIST to what you get from ${WRKDIR}/${_project}) > > I have a draft for this since the codeberg post-extract fix when I > noticed that annoying bit, too. codeberg is an outlier here. For now I > suggest the diff below. It seems like it should be that simple, doesn't it. That was one of the "nothing I try works". > The codeberg ports' dist-tuple variables change along these lines, as > shown by running the change through my > https://github.com/rfht/dist-tuple.test: > > -_DT_WRKDIST = **WRKDIR**/gumbo-parser-0.13.2 > +_DT_WRKDIST = **WRKDIR**/gumbo-parser > > -_DT_WRKDIST = **WRKDIR**/leiningen-2.12.0 > +_DT_WRKDIST = **WRKDIR**/leiningen > > etc... make show=WRKDIST tells a different story. > Probably a project for after ports are past release lock/unlock, but > would be good to give it a try. > > Index: dist-tuple.port.mk > =================================================================== > RCS file: /cvs/ports/infrastructure/mk/dist-tuple.port.mk,v > diff -u -p -r1.20 dist-tuple.port.mk > --- dist-tuple.port.mk 7 Jan 2026 16:49:11 -0000 1.20 > +++ dist-tuple.port.mk 30 Mar 2026 05:52:56 -0000 > @@ -30,7 +30,11 @@ _subdir = > _distname ?= ${_project}-${_id:C/^(v|V|ver|[Rr]el|[Rr]elease)[-._]?([0-9])/\2/:S,/,-,g} > DISTNAME ?= ${_distname} > _subdir = refs/tags/ > +. if "${_template}" == "codeberg" > +_DT_WRKDIST ?= ${WRKDIR}/${_project} > +. else > _DT_WRKDIST ?= ${WRKDIR}/${_distname} > +. endif > . else > _DT_WRKDIST ?= ${WRKDIR}/${_project:C,^.*/,,}-${_id} > . endif