Kelly Kinkade
Kelly Kinkade
I would use the "pragma" approach to disable the warnings instead of using an attribute, since (I sincerely hope) that no compiler author is doofus enough to issue even so...
i don't think that would be a bad approach i dunno if there are material changes between what we're using and what is in tronkko's repo; i haven't looked that...
yeah i was holding off on the administrivia until i get some intel that this actually works
> assuming it works, code looks ok i got assurances on discord from a couple of testers that it worked, which is good enough for me given the nature of...
i honestly don't know. the accuracy display is in the viewscreen code, not the caravan handling code, and i don't remember if we even looked into that. we definitely figured...
this is caused by stonesense's use of a CoreSuspender in its fetch and draw loops, which results in a deadlock. the specific scenario is that some other thread (typically either...
> we can't forward declare enums, so we'll have to include some SDL headers in DFSDL.h?, if want this. :/ this isn't entirely true, you can declare an "opaque" enum...
> @lethosor @ab9rf now that DFHack no longer shadows SDL entrypoints, is DFSDL still necessary? What exactly are we gaining by avoiding the link-time dependency? You're right, we probably don't...