On Fri, 2019-08-23 at 10:22 -0400, Pedro Vanzella wrote: > I actually resubmitted by Filipe's request, since the patches weren't > applying cleanly anymore. The idea was to apply these patches and in the > future refactor the code to use the feature discovery routines. Yes, I want to refactor everything so I though there was no point in us changing the patch set again. I did not review the first revision of the patch set so if that works for you (Benjamin) we can just merge that. > So we (you, me and Filipe) should probably come up with an action plan > here. The way I see it there are two issues here: one is adding this > feature, and the other is refactoring to use feature discovery for all > features. There are advantages and disadvantages to doing one or another > first and we might want to discuss that. > > By merging this first (probably after I resubmit it as a single squashed > patch) we get to test it a bit better and have a usable feature sooner. > Plenty of people have been requesting this and there is plenty of stuff > that can be built on top of it, but only once this is actually merged I > think. > > On the other hand, by first refactoring the rest of the code to use > 0x0001 we avoid some rework on this patch. It should be minor, as most > functions here do all the heavy lifting after the initial feature > discovery, and are thus mostly independent from how that is done. > > I'm happy either way, so just let me know what you guys decide. I am also fine either way so I think we should just re-send the first revision of your patch set as Benjamin requested. Thank you, Filipe Laíns
Attachment:
signature.asc
Description: This is a digitally signed message part
