Search found 2440 matches

by Celtic_Minstrel
May 11th, 2026, 2:44 pm
Forum: WML Workshop
Topic: Checking the feasibility of a complex Custom Project
Replies: 7
Views: 99

Re: Checking the feasibility of a complex Custom Project

Nearly all of what you want is probably possible without modifying the engine, except for points 1-3… which, I presume, are the most important points. Though as Ravana says you miiight be able to manage that with a bot, that would basically mean anyone can pop in and submit a command – though you ...
by Celtic_Minstrel
May 3rd, 2026, 11:03 pm
Forum: WML Workshop
Topic: AI in add-ons
Replies: 6
Views: 95

Re: AI in add-ons

I'm not 100% sure if there's an official policy on this, so take my words with a grain of salt.

I think the policy is roughly "no output from an AI may be uploaded". This implies to me that (for example) asking the AI what's wrong with your code may be acceptable, as long as you wrote the actual ...
by Celtic_Minstrel
April 17th, 2026, 1:40 am
Forum: WML Workshop
Topic: Prevent Map is revealed on Player defeat
Replies: 2
Views: 53

Re: Prevent Map is revealed on Player defeat

I think you can only disable revealing the map with an explicit defeat, not with default defeat conditions.

In short: assuming you want the player to be defeated when any leader dies, adding the following to your scenario should work:

[side]
side=1
defeat_condition=never # disable default ...
by Celtic_Minstrel
April 15th, 2026, 10:06 pm
Forum: WML Workshop
Topic: Quick syntax question(s).
Replies: 3
Views: 98

Re: Quick syntax question(s).

I'm not sure what you mean by "override/supersede" or what "[avoid] [ai]" is supposed to be (perhaps you meant "[ai] [avoid]"), but… the simple attack AI does not take [avoid] into account at all. It ignores most other aspects too, like aggression.
by Celtic_Minstrel
April 12th, 2026, 3:11 pm
Forum: WML Workshop
Topic: Quick syntax question(s).
Replies: 3
Views: 98

Re: Quick syntax question(s).

Looks correct to me, as long as corner_skeleton_quota has a value.
by Celtic_Minstrel
April 12th, 2026, 3:09 pm
Forum: WML Workshop
Topic: Changing recruitment with [modify_ai].
Replies: 14
Views: 313

Re: Changing recruitment with [modify_ai].

That's strange; it feels like it could actually be a bug.

Just checking, but… in that example, you've opened inspect immediately after starting the scenario? I just want to make sure there's no other code that might be intervening to add blocker=yes.
by Celtic_Minstrel
March 22nd, 2026, 2:49 pm
Forum: WML Workshop
Topic: Getting Lich recruitment animation to work.
Replies: 4
Views: 99

Re: Getting Lich recruitment animation to work.

Seems like the key issue is that the recruiting animation requires a second unit, but the [animate_unit] tag offers no way to provide one that's not on the map? The Lua API underlying [animate_unit] could probably work by passing in a temporary unit, but the WML tag designates the unit by a filter ...
by Celtic_Minstrel
February 16th, 2026, 9:44 pm
Forum: WML Workshop
Topic: effect → apply_to=type
Replies: 6
Views: 139

Re: effect → apply_to=type

Although advances_to can indeed be placed inside [unit], it's one of those things that could be randomly overwritten when the unit is "rebuilt". I'd recommend instead using [effect]apply_to=remove_advancement and/or [effect]apply_to=new_advancement.
by Celtic_Minstrel
February 15th, 2026, 3:19 pm
Forum: WML Workshop
Topic: effect → apply_to=type
Replies: 6
Views: 139

Re: effect → apply_to=type

Why?

Can't you get the same effect just by adding that unit to advances_to? Why use an effect?
by Celtic_Minstrel
February 11th, 2026, 2:21 pm
Forum: WML Workshop
Topic: Removing attacks from unit variation.
Replies: 26
Views: 756

Re: Removing attacks from unit variation.

Yes, every unit has a standing animation, because one is synthesized if you don't explicitly define one. That's not the case for an idling animation.
by Celtic_Minstrel
February 11th, 2026, 5:10 am
Forum: WML Workshop
Topic: Removing attacks from unit variation.
Replies: 26
Views: 756

Re: Removing attacks from unit variation.

Actually, I doubt it works for the idling animation either (unless the unit didn't have an idling animation to begin with).

As I mentioned previously, that only adds a new animation. It doesn't remove the old animation. And the way the animation system works is that it collects all the possible ...
by Celtic_Minstrel
February 7th, 2026, 8:18 pm
Forum: WML Workshop
Topic: Removing attacks from unit variation.
Replies: 26
Views: 756

Re: Removing attacks from unit variation.

I think the only animations that go behind units are terrain animations; and dynamically altering those is difficult (and only possible if the hex has no overlay terrain).
by Celtic_Minstrel
February 5th, 2026, 11:19 pm
Forum: WML Workshop
Topic: Removing attacks from unit variation.
Replies: 26
Views: 756

Re: Removing attacks from unit variation.


Is there a way to make it use the new standing animation instead of the old one?

Well, if I'm not mistaken, what should actually happen when you add a new standing animation with an effect is that the unit uses the old one 50% of the time and the new one 50% of the time. Unfortunately, there is ...
by Celtic_Minstrel
February 1st, 2026, 8:36 pm
Forum: WML Workshop
Topic: Changing recruitment with [modify_ai].
Replies: 14
Views: 313

Re: Changing recruitment with [modify_ai].

Hmm. Thinking about it, the problem is probably that your recruitment instructions aspect is a composite aspect , so a path without a facet won't match. Try something like this:

The definition (the only change is adding an ID to the facet):

[aspect]
id=recruitment_instructions
[facet]
id=my ...
by Celtic_Minstrel
February 1st, 2026, 7:08 am
Forum: WML Workshop
Topic: Changing recruitment with [modify_ai].
Replies: 14
Views: 313

Re: Changing recruitment with [modify_ai].


Hello, I am trying to follow the wiki https://wiki.wesnoth.org/Modifying_AI_Components#The_.5Bmodify_ai.5D_Tag, however there is something I am doing wrong. May I ask for help with the following blocks? I would be very grateful.

I can see at least two issues in the code in your opening post ...