Daniel Ahlberg posted <[EMAIL PROTECTED]>, excerpted below, on Mon, 17 Jan 2005 04:23:14 +0100:
> TOP TEN: > ======== > > games-mud/gMOO-0.4.8-r1 - 1081 days (~amd64) > net-misc/smssend-3.2 - 1081 days (~sparc) > x11-libs/buffy-0.2-r1 - 1081 days (~ppc) > sci-chemistry/babel-1.6 - 1081 days (~amd64) > app-office/gnofin-0.8.4 - 1081 days (~sparc) > media-sound/id3-0.12-r1 - 1081 days (~ppc64) > app-office/sc-7.12 - 1081 days (~amd64) > dev-util/mergetrees-0.9.3 - 1081 days (~mips ~ppc) > app-text/duconv-1.1 - 1081 days (~mips) > sci-chemistry/moldy-2.16e - 1081 days (~ppcmacos) > > Questions and comments may be directed to [EMAIL PROTECTED] [posted to list and to aliz directly, as this is a reply to an automated posting] OK, from previous discussion, we know that the script, due to portage, can't tell how long an ebuild has actually been keyworded ~arch, if the keyword was just added to the package (or something like that.. I may be misunderstanding the details). We all know most or all of these haven't /actually/ been unstable arch keyworded that long on those archs, with at least Gentoo amd64 not having even existed that long, AFAIK. Thus, the mail is of little use as is, and because of this bug with the way portage works, and the limit to ten, it isn't going to be of much use any time soon. Here's a suggestion. In your script, do a grep --invert-match (or use a similar sed command) on "1081 days", which appears to be the default max time period. That will kill all the default matches due to that portage bug, which should result in a top-ten that's actually meaningful (well, unless there's another such issue). -- Duncan - List replies preferred. No HTML msgs. "They that can give up essential liberty to obtain a little temporary safety, deserve neither liberty nor safety." -- Benjamin Franklin -- [email protected] mailing list
