pkgsrc-Bugs archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: pkg/44698: Alternative PKGFILE [patch]



The following reply was made to PR pkg/44698; it has been noted by GNATS.

From: David Holland <dholland-pbugs%netbsd.org@localhost>
To: gnats-bugs%NetBSD.org@localhost
Cc: 
Subject: Re: pkg/44698: Alternative PKGFILE [patch]
Date: Wed, 9 Mar 2011 10:27:29 +0000

 On Tue, Mar 08, 2011 at 12:10:01PM +0000, cheusov%tut.by@localhost wrote:
  > @@ -11,7 +11,8 @@
  >  .  endif
  >  .endif
  >  PKG_SUFX?=         .tgz
  > -PKGFILE?=          ${PKGREPOSITORY}/${PKGNAME}${PKG_SUFX}
  > +FILEBASE?=         ${PKGBASE}
  > +PKGFILE?=          ${PKGREPOSITORY}/${FILEBASE}-${PKGVERSION}${PKG_SUFX}
  >  PKGREPOSITORY?=            ${PACKAGES}/${PKGREPOSITORYSUBDIR}
  >  PKGREPOSITORYSUBDIR?=      All
 
 So, much as it's useful to be able to build such packages
 automatically, I don't think this is the right way to handle them.
 
 The name of the resulting package should be derived deterministically
 from the options (and also, any other build switches resulting in a
 different package, like X11_TYPE) so that binary package tools have
 some hope of being able to process the results.
 
 That is, if you build a bunch of variant packages with options and
 upload them to a package repository it should be possible for people
 using pkg_install to download and install them without specifically
 having to know the special names; in fact, ideally, pkg_install would
 respond to non-default PKG_OPTIONS by looking for the corresponding
 variant binary package.
 
 This all hinges on getting the naming right so I think we need to
 first figure out what the naming scheme should be, which isn't
 entirely trivial to sort out, then arrange to have the package
 filename be set automatically rather than by user configuration.
 
 -- 
 David A. Holland
 dholland%netbsd.org@localhost
 


Home | Main Index | Thread Index | Old Index