Arsenal - Reuse current inventory containers when importing invalid loadouts

[mrschick]

When merged this pull request will:

• When importing a stored loadout which has a uniform/vest/backpack that is missing or from an unloaded mod, retain the currently equipped container and fill it with available items from the stored loadout.
• Fix Arsenal Import button ignores limited arsenal #10673, also attempt to recover missing containers in the imported inventory.

The use case for this change is missions in which the arsenal is restricted to clothing from a specific faction, but still contains all the usual medical and other specialized equipment for a given role.
Previously, even if one had saved a very similar loadout, but with a different camo, it was impossible to load it without losing all clothing items. Forcing one to redo the entire loadout from scratch.
With this change, it will be possible to import such a loadout, so that the numerous equipment items will be placed in already worn clothing of the specific faction, saving time.

If "" is given as a container, an error is thrown, therefore, filter out all unavailable/null containers

I realize that this was part of optimization work in #9316. Since it hinders this implementation, I removed that check. I have not found it causing errors when importing loadouts.

[LinkIsGrim]

I realize that this was part of optimization work in #9316.

I think loadout validation is spread out over multiple frames now so it doesn't matter as much

Just for confirmation's sake: does this only apply if the class doesn't exist, or also for items that aren't available in the arsenal as well?

Not the problem if the latter, it opens up an interesting possibility (layered loadouts)

[mrschick]

does this only apply if the class doesn't exist, or also for items that aren't available in the arsenal as well?

It applies whenever fnc_verifyLoadout.sqf returns an inventory container's config name as "", which happens in both cases.

[mrschick]

Could this make 3.19?

[LinkIsGrim]

I ended up refactoring verifyLoadout again while reviewing this.

I have a minor nitpick with preventOverfilling (could be less code) and I'm not a fan of recoverInvalidContainers atm, but I don't think there's any working around the latter.

I'll leave my drastic changes to another PR, LGTM.

[mrschick]

(could be less code)

Please do tell ✍🏻

[LinkIsGrim]

See below:

I'd have pushed to branch but no write perms

[mrschick]

Sadly still an issue with organization forks.
Now please don't make me write it off that picture 😁

[PabstMirror]

looks ok to me