I have a sack with packages of all possible arches. I always run queries this way: q.filter(arch=[<supported arches>]).filter(...) Running multiple queries is *very* slow due to the arch filtering. I'd like to populate per-arch sacks. In general, it would be nice to create a sack with user-defined content. s = hawkey.Sack() # populate with individual packages s.add_packages([pkg, pkg, ...]) # populate with query results s.add_packages(q.filter(arch=[<supported arches>])
We rather speed up performance of Query. Adding `.save()` method that would compute and hold the result of the Query for later use. That would be easier to implement and more versatile for other use cases.
Do you have any ETA for this change? Hawkey performance issues are blocking Pungi yum->dnf transition for more than a year. In my use cases is hawkey roughly 5x slower than yum.
no more than another year, I hope. It will be on my TODO list.
This bug appears to have been reported against 'rawhide' during the Fedora 22 development cycle. Changing version to '22'. More information and reason for this action is here: https://fedoraproject.org/wiki/Fedora_Program_Management/HouseKeeping/Fedora22