Date: Sat, 08 Jan 2000 15:46:34 -0500 (EST) From: Will Andrews <andrews@TECHNOLOGIST.COM> To: "Matthew D. Fuller" <fullermd@futuresouth.com> Cc: ports@FreeBSD.ORG, Satoshi - Ports Wraith - Asami <asami@FreeBSD.ORG> Subject: Re: multi-level categories Message-ID: <XFMail.000108154634.andrews@TECHNOLOGIST.COM> In-Reply-To: <20000108131719.A22210@futuresouth.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 08-Jan-00 Matthew D. Fuller wrote: > In at least most cases, I support this anyway :) > I remember taking 20 minutes to track down Mosaic, simply because it was > capitalized for no real reason I could discern.. *nod* I think all port dirs should be lowercased. PKGNAME, however.. can be capitalized in any form. > ???? > The CVS/ dir is created by CVS when you checkout the tree. They have > nothing to do with the port itself. In fact, most of us *USE* those CVS/ > dirs to create patches for send-pr (read: 'cvs diff'). I don't get them (CVS/) from cvsup.. and most patches I've seen (for -ports anyway) don't use cvs diff. Further, porting.html doesn't mention that you can use cvs diff instead of creating a whole new copy.. like I've always done. On a side note, do most people do their cvs diffs between the repo and the "locally modified" copies, rather than commit their changes on a certain vendor branch and do a cvs diff between these? > AFA filesystem performance it would make a BIG difference. All ports > have files/, and a large number (most?) have patches/. I'd say that > sticking everything in files/ would cut the number of dirs practically > in half. Yay, a /usr/ports that uses a less-than-obscene number of > inodes! Would this combination of directories reduce fs delays? I am no filesystem expert. But.. if it would reduce the amount of time it takes to traverse the ports tree, it would kick ass. Of course, we'd need to do some testing, and bsd.port.mk would need some (serious?) modifications, and (all?) 2900+ ports would have to be updated... > That doesn't work when we have multiple files for a port. It also makes > it easy to transition to another hash scheme for checksums in the future > if necessary; just use a different first token. I don't think forking > off a few extra processes during the checksum check is any huge performance > sink. Even musca (my 2.1.5 386/16 with 4 megs of RAM) can handle that > in a fairly trivial amount of time. Ah shit, I forgot about that. -- but, we can still remove "MD5 " and " =". ;) -- Will Andrews <andrews@technologist.com> GCS/E/S @d- s+:+>+:- a--->+++ C++ UB++++ P+ L- E--- W+++ !N !o ?K w--- ?O M+ V-- PS+ PE++ Y+ PGP+>+++ t++ 5 X++ R+ tv+ b++>++++ DI+++ D+ G++>+++ e->++++ h! r-->+++ y? To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?XFMail.000108154634.andrews>