Am 10.04.25 um 10:29 schrieb Fiona Ebner: > Agreed and I get your point. It just felt more natural to go with the > object approach since all call sites have that very object and would > need to do the very same deconstruction if going with function(volume, > format, subtype). Independent of the chosen approach, keeping it > consistent with the volume_is_lxc_backup() helper is also an advantage.
Hmm, that one could be reworked too to a simpler/clearer signature. Such methods in Utils are far from ideal in the first place, having just an opaque signature does not really help here. But yeah, does not really matter, one can hold it rather easily wrong either way. _______________________________________________ pve-devel mailing list pve-devel@lists.proxmox.com https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel
