Date: Fri, 22 Aug 1997 14:43:34 -0400 (EDT) From: Tim Vanderhoek <hoek@hwcn.org> To: Warner Losh <imp@rover.village.org> Cc: Satoshi Asami <asami@cs.berkeley.edu>, fenner@parc.xerox.com, ports@FreeBSD.ORG Subject: Re: Versioning bsd.port.mk Message-ID: <Pine.GSO.3.96.970822142902.7088A-100000@james.freenet.hamilton.on.ca> In-Reply-To: <E0x1wpq-00052M-00@rover.village.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 22 Aug 1997, Warner Losh wrote: > Hmmm. I'm not sure that that is even foolproof. If you are making > changes to the bsd.port.mk file while I'm making changes to the > gerbils port. My changes get in first, and then yours. My changes > depend on version n-1 of the bsd.port.mk file, but version n breaks I think it's necessary to assume that version n+1 of bsd.port.mk will work with a Makefile designed for version n. If you use strict version n for version n matching it'll become difficult to keep everything sync'd (for every revision of bsd.port.mk, each Makefile (or files/VER) would need updating, plus expected user frustration at having to upgrade their ports subsystem considerably more often than necessary). I think if you want to catch cases where a Makefile depending on bsd.port.mk version n-1 doesn't work with bsd.port.mk version n you have to use either a feature-for-feature compatibility check (yuck) or what Bill suggested -- just print out a warning "try upgrading bsd.port.mk??" on errors. The one problem there is with matching RCS strings, which I mentioned before, is catching the case where a port is updated but bsd.port.mk isn't touched. Like I said, then, if you're willing to let that case slide.... [although I still prefer Itojun's last patch to all other solutions :-] -- Outnumbered? Maybe. Outspoken? Never! tIM...HOEk
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.GSO.3.96.970822142902.7088A-100000>