Date: Wed, 19 Jun 1996 12:23:56 -0700 From: "Jordan K. Hubbard" <jkh@time.cdrom.com> To: Nate Williams <nate@sri.MT.net> Cc: Mark Murray <mark@grondar.za>, current@FreeBSD.org Subject: Re: tcl -- what's going on here. Message-ID: <25390.835212236@time.cdrom.com> In-Reply-To: Your message of "Wed, 19 Jun 1996 12:43:27 MDT." <199606191843.MAA06814@rocky.sri.MT.net>
next in thread | previous in thread | raw e-mail | index | archive | help
> Because all of the 'Berkeley' targets don't work. > > 'make depend all install clean cleandir obj' etc.. are *useful* (nay > critical) for some installations. Another strawman argument, I'm afraid. Through encapsulation it would easily be possible to make ALL of these work, and with minimal perturberation to the package. You want NOPROFILE to be passed through when that's applicable? Fine. You want obj, clean and cleandir to work? Fine. I already did that for bsd.port.mk, in fact. You want your debugging flags passed in? No problem. I looked at providing *all* of these for the TCL port and, while I only implemented the ones I thought were critical (obj and cleandir), the rest were pretty trivial. For most, if not all, of the GNU ports I've looked at analogs exist for pretty much all of our standard bmake operations. As I said in my last mail, the obj link wart is our biggest hurdle and I'd just as soon see it die anyway (and that is NOT a new position on my part - I've been complaining about the stupid things for 2 years now). I'm also all for backing this out and getting on with our lives, but I just thought I'd defend what I thought was a little unfair criticism of the proposed mechanism's shortcomings. Whatever else its shortcomings, you can do a LOT of "homogenization" through the ports-style Makefile wrapper. Jordan
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?25390.835212236>