epic/OpenRA
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Actors will lose targets and AI won't follow indefinitely

Browse Source
This commit is contained in:
pevers
2015-04-09 16:00:47 +02:00
committed by Oliver Brakmann
parent 1cbde65a08
commit 6ed0273656
8 changed files with 93 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

23
OpenRA.Mods.Common/AI/States/ProtectionStates.cs
View File

@@ -19,6 +19,9 @@ namespace OpenRA.Mods.Common.AI
class UnitsForProtectionAttackState : GroundStateBase, IState
{
public const int BackoffTicks = 4;
internal int Backoff = BackoffTicks;
public void Activate(Squad owner) { }
public void Tick(Squad owner)
@@ -26,7 +29,7 @@ namespace OpenRA.Mods.Common.AI
if (!owner.IsValid)
return;
if (!owner.TargetIsValid)
if (!owner.IsTargetValid)
{
owner.TargetActor = owner.Bot.FindClosestEnemy(owner.CenterPosition, WRange.FromCells(8));
@@ -37,8 +40,22 @@ namespace OpenRA.Mods.Common.AI
}
}
foreach (var a in owner.Units)
owner.World.IssueOrder(new Order("AttackMove", a, false) { TargetLocation = owner.TargetActor.Location });
if (!owner.IsTargetVisible)
{
if (Backoff < 0)
{
owner.FuzzyStateMachine.ChangeState(owner, new UnitsForProtectionFleeState(), true);
Backoff = BackoffTicks;
return;
}
Backoff--;
}
else
{
foreach (var a in owner.Units)
owner.World.IssueOrder(new Order("AttackMove", a, false) { TargetLocation = owner.TargetActor.Location });
}
}
public void Deactivate(Squad owner) { }