Date: Wed, 24 Jan 2001 00:15:27 -0800 From: Jordan Hubbard <jkh@winston.osd.bsdi.com> To: Jeremy Lea <reg@FreeBSD.org>, Maxim Sobolev <sobomax@FreeBSD.org>, cvs-committers@FreeBSD.org, cvs-all@FreeBSD.org Subject: Re: cvs commit: src/usr.sbin/pkg_install/add extract.c main.c src/usr.sbin/pkg_install/create create.h main.c perform.c pkg_create.1 src/usr.sbin/pkg_install/lib file.c plist.c Message-ID: <37689.980324127@winston.osd.bsdi.com> In-Reply-To: Message from Jeremy Lea <reg@FreeBSD.org> of "Wed, 24 Jan 2001 00:06:17 PST." <20010124000617.C63549@shale.csir.co.za>
next in thread | previous in thread | raw e-mail | index | archive | help
Well, I do review patches occasionally. :) But thanks for bringing this up. I looked at this set, in fact, and sort of went "Ew. Is that really necessary?" since it would also entail that anyone you *give* a bzip port to also know how to unbzip and re-gzip it manually or have the most up-to-date pkg_add tools as well. The nice thing about handing people a foo.tgz file is that packages have existed in that format since the dawn of time and you always know that 100% of your target audience (those with enough disk space and the inclination to do so) can install it. With bzip'd packages, I wouldn't even feel comfortable using the format in any kind of official media for several years at least, making it less useful than many other things I could think of. But all that said, I also don't care enough about the pkg_install tools to defend them all that aggressively anymore. It's just crocks layered on top of crocks in there and I'm focusing what extremely small bits of time I have for the subject on what comes next. :) - Jordan > Hi, > > On Mon, Jan 22, 2001 at 04:01:55AM -0800, Maxim Sobolev wrote: > > Modified files: > > usr.sbin/pkg_install/add extract.c main.c > > usr.sbin/pkg_install/create create.h main.c perform.c > > pkg_create.1 > > usr.sbin/pkg_install/lib file.c plist.c > > Log: > > - Add ability to handle bzip2-compressed packages; > > The code in create/main.c never turns UseBzip2 = TRUE. Which implies > it was never tested, which also implies that the command line switch is > redundant. We can guess bzip2 compression from the suffix. In fact we > have to have a bzip2 suffix, else we can't guess how to extract it, so > having a command line switch can only cause bad things to happen. > > > No response from: -ports > > This should have read > > Approved by: jkh > > He is the maintainer of this code, not -ports. (Even if he never > changes it or reviews patches :-( ) > > Regards, > -Jeremy > > -- > FreeBSD - Because the best things in life are free... > http://www.freebsd.org/ > To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?37689.980324127>