[Rpm-ecosystem] hawkey query performance

Michael Schroeder mls at suse.de
Tue Sep 15 11:42:22 UTC 2015


On Tue, Sep 15, 2015 at 06:02:08AM -0400, Honza ??ilhan wrote:
> Thanks for taking care. I am not even sure if someone is / will be using this
> tool in Fedora. If you still plan to add this support it would be great to
> incorporate it into depsolver (if possible) so the file conflicts errors would
> be reported during / after resolution not by rpm after all packages are already
> downloaded [1]. Or is there any function callable from C code on some package
> set so we can run it after depsolving?
>
> [1] https://bugzilla.redhat.com/show_bug.cgi?id=1219911#c0

The filelist.xml data is needed so that file requires can be solved.
It's impossible to do file conflict detection with just the repository
metadata, as it lacks the file checksum information.

Regarding the bug report, it's probably just that dnf's do_transaction
code (in base.py) uses the last problem when throwing the exception
instead of concatenating the problems.

Cheers,
  Michael.

-- 
Michael Schroeder                                   mls at suse.de
SUSE LINUX GmbH,           GF Jeff Hawn, HRB 16746 AG Nuernberg
main(_){while(_=~getchar())putchar(~_-1/(~(_|32)/13*2-11)*13);}


More information about the Rpm-ecosystem mailing list