FS#8897 - Some asymmetries of the current conflict resolving

Attached to Project: Pacman
Opened by Nagy Gabor (combo) - Tuesday, 11 December 2007, 15:47 GMT
Last edited by Dan McGee (toofishes) - Sunday, 27 January 2008, 18:16 GMT
Task Type Bug Report
Category General
Status Closed
Assigned To Xavier (shining)
Architecture All
Severity Medium
Priority Normal
Reported Version 3.0.6
Due in Version 3.2.0
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

Well, I don't like provide == depend case at all, but I set rules to show your expected result.
Pactests attached.
   sync1007.py (0.3 KiB)
   sync1008.py (0.3 KiB)
   sync1009.py (0.3 KiB)
This task depends upon

Closed by  Dan McGee (toofishes)
Sunday, 27 January 2008, 18:16 GMT
Reason for closing:  Fixed
Additional comments about closing:  Fixed by commit 88cbee3c24768470cc0d4abe050e98b989807a67
Comment by Nagy Gabor (combo) - Friday, 21 December 2007, 14:56 GMT
Well, I replace sync1009.py with the following:
------sync1009.py------
self.description = "Provide == conflict (3)"

sp1 = pmpkg("pkg1")
sp1.conflicts = [ "pkg2" ]
sp1.provides = [ "pkg2" ]
self.addpkg2db("sync", sp1)

sp2 = pmpkg("pkg2")
sp2.conflicts = [ "pkg1" ]
self.addpkg2db("sync", sp2)

self.args = "-S pkg2 pkg1"

self.addrule("PKG_EXIST=pkg1")
self.addrule("!PKG_EXIST=pkg2")
------------------------

I hope that this indicates better then the current conflict resolving _depends on_ the order of packages in the target list (see sync1007.py vs. sync1009.py)
Comment by Xavier (shining) - Monday, 31 December 2007, 17:37 GMT
I don't get this. It doesn't depend on the order, it just installs the package that provides the other.
However, I don't understand what is the reasoning behind this behavior, as I previously said there:
http://www.archlinux.org/pipermail/pacman-dev/2007-December/010393.html
Comment by Nagy Gabor (combo) - Wednesday, 02 January 2008, 14:20 GMT
The only difference between sync1007.py and (new) sync1009.py is self.args.
And sync1007 passes, sync1009 fails; this is really ugly.
Comment by Xavier (shining) - Wednesday, 02 January 2008, 17:12 GMT
Oops, you're right, maybe I was looking at the old sync1009 instead of the new one.
This is very odd indeed..

Loading...