Date: Fri, 30 Aug 1996 10:15:18 -0500 (EST) From: John Fieber <jfieber@indiana.edu> To: Chuck Robey <chuckr@glue.umd.edu> Cc: Satoshi Asami <asami@FreeBSD.ORG>, jkh@time.cdrom.com, freebsd-ports@FreeBSD.ORG Subject: Re: Should this port go in ? Message-ID: <Pine.BSI.3.95.960830095804.5578T-100000@fallout.campusview.indiana.edu> In-Reply-To: <Pine.OSF.3.95.960830105001.19949A-100000@baud.eng.umd.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 30 Aug 1996, Chuck Robey wrote: > I see one problem with this. If you make it too automated, then stuff the > erroneously gets installed to dumb places like /usr/lib or /etc will not > be noticed by porters, who are forced to notice it now. The problem I was Have bsd.ports.mk review the installation and sound and sound an alarm if anything lands outside of /usr/local? The manual system may help avoid problems like that but its a PITA and very error prone, particularly when ports are upgraded and new files appear but the PLIST isn't updated. I think we can get rid of the PITA factor and errors while making sure things get installed in reasonable places. > addressing above was not the convenience of making PLIST (which I admit > I'd like to have) but the ability for a package to be more agile in > installing options, as a port can be. The obvious example is whether or Mm... I guess I wasn't thinking of ports more than packages and I'm not sure if or how a modified install would help in this respect. Another thing the scheme doesn't catch is symlinks that are, out of necessity, made in the target directory after the real files have been installed. -john == jfieber@indiana.edu =========================================== == http://fallout.campusview.indiana.edu/~jfieber ================
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSI.3.95.960830095804.5578T-100000>