Call for feedback on a Ports-collection change

[Alexander Leidinger]

On Fri, 09 Jan 2004 06:54:42 +0300
Sergey Matveychuk <sem at ciam.ru> wrote:

> Even separated pkg-install and pkg-deinstall has a logic: pkg-install 
> will run only when you install a package not a port. pkg-deistall will 
> run always you deinstall the port.

---snip---
/usr/ports/chinese/pine4/Makefile:      ${SH} pkg-install
/usr/ports/databases/postgresql-devel/Makefile: @${SH} ${PKGDIR}/pkg-install ${PORTNAME} BACKUPWARNING
/usr/ports/databases/postgresql-devel/Makefile:         ${SH} ${PKGDIR}/pkg-install ${PORTNAME} PRE-INSTALL
/usr/ports/databases/postgresql72/Makefile:             ${SH} ${PKGDIR}/pkg-install ${PORTNAME} PRE-INSTALL
/usr/ports/devel/perforce/Makefile:             ${SH} ${PKGDIR}/pkg-install ${PORTNAME} PRE-INSTALL
/usr/ports/games/lbreakout2/Makefile:               ${SH} pkg-install ${PKGNAME} POST-INSTALL
[...]
---snip---

So I think we can collapse both into one, as both can get called
regardless of port of package.

But I don't see much more potential in the actual ports infrastructure
to safe inodes while maintaining the actual ease of use for a port
developer. For simple ports which just need a barebones Makefile, a
plist, a distinfo an a descr, a collapsed format may be easy to handle,
but we have a lot of ports which are not that simple. Creating lang/icc
in such a scheme would have been much more time consuming for me.

Bye,
Alexander.

-- 
           I will be available to get hired in April 2004.

http://www.Leidinger.net                       Alexander @ Leidinger.net
  GPG fingerprint = C518 BC70 E67F 143F BE91  3365 79E2 9C60 B006 3FE7