Pacman

Historical bug tracker for the Pacman package manager.

The pacman bug tracker has moved to gitlab:
https://gitlab.archlinux.org/pacman/pacman/-/issues

This tracker remains open for interaction with historical bugs during the transition period. Any new bugs reports will be closed without further action.
Tasklist

FS#14887 - Using timestamps to help pacman offer a solution to packages that have weird upstream versioning

Attached to Project: Pacman
Opened by jasin (rooloo) - Saturday, 30 May 2009, 23:06 GMT
Last edited by Dan McGee (toofishes) - Wednesday, 12 January 2011, 06:24 GMT
Task Type Feature Request
Category Backend/Core
Status Closed
Assigned To No-one
Architecture All
Severity Low
Priority Normal
Reported Version 3.2.2
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

Description: Feature request for pacman.

warning: oss: local (4.1_1052-1) is newer than community (4.1_1052b-1)

After some discussion on irc it sort of hit me that maybe these sort of warnings can be squished/eliminated or at the very least have pacman offer a solution as to which is the newer package by using timestamps on when packages where built. I don't really see an issue with this unless the packagers system clock is off. But even this can be screened for, one could check not so much the time but maybe just the date of packages being uploaded to the official repositories. Just a thought


Additional info:
Latest pacman package.
This task depends upon

Closed by  Dan McGee (toofishes)
Wednesday, 12 January 2011, 06:24 GMT
Reason for closing:  Won't implement
Additional comments about closing:  Will be fixed by epoch in 3.5.0
Comment by Gerardo Exequiel Pozzi (djgera) - Saturday, 30 May 2009, 23:31 GMT
Maybe this is that this package needs...

options=('force')

from man:PKGBUILD

options (array)
<snip>
force
Force the package to be upgraded by a pacman system upgrade operation, even if the version number would normally not trigger such an
upgrade. This is useful when the version numbering scheme of a package changes (or is alphanumeric). See linkman:pacman[8] for more
infomation on version comparisons.
Comment by jasin (rooloo) - Sunday, 31 May 2009, 02:27 GMT
I don't really have a problem with that 'force' option, But it has to be implimented by the packager. A timestamp check would eliminate the possiblity of human error, as in this paticular case. Is it really that much trouble to automate this process a little more and remove the human error factor?

In any reguard, this presents now as a bug report for the above aformentioned reason.
Comment by Nagy Gabor (combo) - Sunday, 31 May 2009, 22:39 GMT
Well, timestamps don't help here. A very old package can be built today etc.

RPM uses EPOCH to workaround this problem: http://sial.org/howto/rpm/epoch/

I don't like the frequent use of force, since the dependency check ignores force field. And two foo packages with force cannot be compared.
Comment by jasin (rooloo) - Monday, 01 June 2009, 17:44 GMT
A very old package can be built today etc.

Well I didn't think anyone would actually upload a old package to the server.

I still think this needs more thought on what can be done. The end user really shouldn't have to worry about running pacman -S oss after running -Syu when oss is installed. Whether the fix be on the packagers side or the pacman side. Arch is supposed to adhere to the KISS method. There is nothing intuitive or simple about installing something that is already installed just to see it being updated.

I guess the complexity of fixing the issue is greater then the effect to the end user, thus it should remain as is?
Comment by Dan McGee (toofishes) - Saturday, 06 June 2009, 16:12 GMT
It would be epoch or nothing (sticking with the current 'force' option) if we wish to fix this. Timestamps just don't make sense in all cases.

Loading...