2 Commits
Author SHA1 Message Date
san7890andGitHub 0cd21f110c fixes silly mistake in can_be_held element (#96960)
## About The Pull Request

Error from #96917 (cb535cdfa6)

We shouldn't be calling `mob_try_pickup` on the other pathways, that's
the responsibility of the mousedrop proc. The other two signal handling
procs on this element exist only to block storage dumping/stripping when
the mousedrop proc is working. I think I didn't catch this in testing
because it's an invisible bug that gets obfuscated by the `do_after()`
but I pretty obviously fell asleep at the wheel here and added this
extra code when it's definitely not meant to work that way.

I'm gonna do a CL because I forgot to mention something in the original
refactor's changelog so people might be more aware of the new mechanic
that I forgot to mention (implemented to better reliably parse intent
and be less ambiguous)

🆑
refactor: Picking up your pets (or any holdable mob) requires an
aggressive grab now.
/🆑
2026-07-15 00:05:03 +01:00
san7890andGitHub cb535cdfa6 Refactor var/can_be_held to an element (#96917)
## About The Pull Request

We were doing a lot of mental gymnastics in a bunch of other places, so
let's change this wonky var to a streamlined element that will rely on
the same signals that a lot of stuff was already using/accounting for in
its own signal handling pathways, instead of being a weird coverage gap.
This patch should also make the whole "checking if someone is attempting
to pick a mob up" thing make a lot more sense and use a unified proc
instead of spot-checking whatever random things it wants to spot-check.
## Why It's Good For The Game

I didn't know this was a thing until I looked at #96873 and it made me
sad because literally everything else involving mob drag-and-drop is
already signal-based except this weird stinker that relied on proc
overrides. Never mind that now, let's use nice traits to avoid
typecasting and elements to avoid duplicating code. It should also be
much cleaner to add holdability to a mob isntead of having to do
`can_be_held = FALSE` as a weird behavior (at least one instance had
this non-necessarily). All this really is is just middleware on the
extant /mob/living code but still making it in proper lockstep with the
other signalling procs.

I did port over raptor code faithfully but I'm not 100% sure if it was
meant to be like this? Regardless, that's how it is.
## Changelog
🆑
refactor: Picking up mobs has been altered a bit, please report any bugs
or glitches.
/🆑
2026-07-13 16:57:47 +02:00