Scripting Discussion (dev branch)

[fn_Quiksilver 1636]

8 hours ago, f2k sel said:

According  to Hint F1 editor  the following command should return an array of known targets  within a specified distance

 

player targets [[EAST, GUER], 100]

However when run it gives:-  error type array expected bool

 

full description

Unit targets [[side..],distance,age,centre]

 

I suspect this command hasn't been implemented, any chance it could be finished or fixed.     

 

the command is fine but the syntax documentation is wrong

 

proper use looks more like 

<unit> targets [true,300];

 

[f2k sel 164]

15 hours ago, fn_Quiksilver said:

 

the command is fine but the syntax documentation is wrong

 

proper use looks more like 

<unit> targets [true,300];

 

Thanks I got close myself I did try using true but  placed it before the Filter array which gives a different error.

 

It should have worked though according to the WIKI

_targets = _unit targets [true, [], 300]; //enemy targets in 300m

 

I did look for the  WIKI entry for the targets description but I must have looked in A2/OA section previously.

 

 

[R3vo 2654]

Can anyone explain to me how BIS_fnc_saveGame is supposed to return a boolean?

 

test = [player] call BIS_fnc_saveGame; //at the time it's executed, return value is BIS_fnc_kbisSpeaking, after saving happened, test becomes <NULL-script>

 

It would be nice if this function would return true if saving was successful and false if it wasn't (saving disabled)

[killzone_kid 1330]

8 hours ago, R3vo said:

Can anyone explain to me how BIS_fnc_saveGame is supposed to return a boolean?

If you think this is important enough to be looked at, consider making a ticket on FT

[Sniperwolf572 758]

Can we please get some docs on the new force and torque functions that were recently added?

Some yuuuuuge numbers are needed to get any kind of effect from them, I'm concerned I'm using them wrong. 

[SilentSpike 84]

36 minutes ago, Sniperwolf572 said:

Can we please get some docs on the new force and torque functions that were recently added?

Some yuuuuuge numbers are needed to get any kind of effect from them, I'm concerned I'm using them wrong. 

 

Have you tried putting them into the debug console and hitting F1 with the caret on them?

[x3kj 1247]

On 30.5.2017 at 9:28 AM, silentspike said:

 

Have you tried putting them into the debug console and hitting F1 with the caret on them?

this doesnt help much, because it only tells you what input it expects, but not what the input represents (for the addForce).

 

On 30.5.2017 at 8:51 AM, Sniperwolf572 said:

some docs on the new force and torque functions

 reyhard notified me that this has been added today:

https://community.bistudio.com/wiki/addForce

https://community.bistudio.com/wiki/addTorque

 

 

[Grumpy Old Man 3545]

11 minutes ago, x3kj said:

https://community.bistudio.com/wiki/addForce

https://community.bistudio.com/wiki/addTorque

Played around with those for a bit.

I guess torque makes an object rotate, factoring in its mass,

no idea about addForce, that one just made the vehicle turn towards the position.

At least that's what I think it does.

Having those wiki entries is great but they don't tell you what these commands actually do.

 

Cheers

[x3kj 1247]

50 minutes ago, Grumpy Old Man said:

no idea about addForce, that one just made the vehicle turn towards the position.

At least that's what I think it does.

Having those wiki entries is great but they don't tell you what these commands actually do.

 

addForce is a bit quirky, as it uses the Worldspace for the direction of force, but a modelspace position on the vehicle for force application.

 

A force that is not applied at the objects center of mass will result in a torque as a byproduct.  So depening on the models orientation, it will lead to different results for the same addForce parameters.  Since center of mass does not coincide with center of the models origin for every model, it is very tricky to just apply a force without creating a "by-product" torque.

 

Also, addforce and torque is quirky when applied to vehicles with "active" simulation like helicopters that are quirky in their own way. These simulation may have artificial limiters and so on to deal with simulation instabilities, and their simulation is not necessarily physically correct in certain aspects. So adding certain forces/torques may not have a reallistic effect.

 

Edit: I actually dont know if the force is applied to center of origin of model, or to center of mass. I assume however that its center of origin

[Sniperwolf572 758]

2 hours ago, x3kj said:

this doesnt help much, because it only tells you what input it expects, but not what the input represents (for the addForce).

 

 reyhard notified me that this has been added today:

https://community.bistudio.com/wiki/addForce

https://community.bistudio.com/wiki/addTorque

 

 

Haven't had much time to respond to that, but that was exactly the problem! Thanks for this info that this was documented.

 

As for the center of mass, it can be retrieved and set if necessary.

[x3kj 1247]

oh ok, didnt have those commands on the radar.  This makes it easier. You basically have to create your new coordinate reference to the center of mass all the time then when applying force.

Smaller amounts of torque can still result, because of numerical inaccuracy. Depending on the objects moments of inertia this effect may be smaller or greater.

[dscha 147]

When your FPS is > 60, problems occur with addForce (not reacting/no Force added).

I was in VR, ~200-250FPS -> maybe 1 of 10 addForce commands had an effect on the model.

While with VSync-Enabled (max 60FPS) -> 10/10 had an effect.

 

 

 

EDIT:
also: Since the MTW Positions can change in a model, while it is moving, it can have some "unwanted Results".

 

Draw3D Visualization with HEMTT Mover while some Tests:
https://i.gyazo.com/b1ba068b9bca3aafcff47b679d6d050e.mp4
 

Old BugReport from ~ 3 Years ago

 

[fn_Quiksilver 1636]

really impressed with the improvements to SQF and new scripting commands added in the past 12 months, great work

[Grumpy Old Man 3545]

On 7.6.2017 at 1:43 PM, fn_Quiksilver said:

really impressed with the improvements to SQF and new scripting commands added in the past 12 months, great work

params, select, apply...

The amount of convenience these 3 commands alone added to the scripting engine is incredible!

Then improvement on BIS functions by glorious KK on top of that!

Can't praise good work enough, so thanks anyone involved in this!

 

On a sidenote, would it be possible to enhance getCompatiblePylonMagazines to accept classname as alternative syntax?

 

Cheers

[R3vo 2654]

Can anyone tell me if BIS_fnc_cutDecimals is working correctly?

 

154 / 60 = 2.56667;

 

[154 / 60,0] call BIS_fnc_cutDecimals;// 3

 

Shouldn't it return 2 ? Otherwise I could simply use round.

 

toFixed seems to do the same.

 

 

Edit: Nevermind. Read the wiki description

Quote

Function that rounds specified number to specified amount of decimals.

 

[pierremgi 4857]

1 hour ago, R3vo said:

Can anyone tell me if BIS_fnc_cutDecimals is working correctly?

 

154 / 60 = 2.56667;

 

[154 / 60,0] call BIS_fnc_cutDecimals;// 3

 

Shouldn't it return 2 ? Otherwise I could simply use round.

 

toFixed seems to do the same.

 

 

Edit: Nevermind. Read the wiki description

 

Not a bug but you're right, cutting should not round anything. Round and toFixed already do that. So this function could be different.

[killzone_kid 1330]

On 6/12/2017 at 7:15 AM, Grumpy Old Man said:

On a sidenote, would it be possible to enhance getCompatiblePylonMagazines to accept classname as alternative syntax?

Apparently this was possible from the start. Updated biki https://community.bistudio.com/wiki/getCompatiblePylonMagazines

[froggyluv 2136]

 

 Did I go full ReeRee or is getUnitTrait "camouflageCoef " returning Bool rather than numbers?

 

_ct = cursortarget getunittrait "camouflageCoef "; hint str _ct

 

 Returns: False as before I used to get a value??

[bad benson 1733]

got a question about BIS_fnc_getTurrets.

 

when i make it output config paths it seems to give me the path to the vehicle itself, if it has no turrets. is that intended or am i doing something wrong?

 

it's kind of weird because when i make it output the turret path array it outputs an empty array for those vehicles without turrets. seems kind of counter intuitive. but maybe i did something wrong. was late yesterday and i can't test again right now because it's loading the update right now.

[killzone_kid 1330]

2 hours ago, bad benson said:

is that intended

This is how it has always been and most likely intended, though I don't know exact reason myself. What were you going to use this function for, if not secret?

[bad benson 1733]

Just now, killzone_kid said:

This is how it has always been and most likely intended, though I don't know exact reason myself. What were you intended to use this function for, if not secret?

 

using it as part of my function to get all vehicle seats, their turret paths (if turret) and the name of the seat in the scroll menu. i do that to populate my own seat UI. i was just wondering why. maybe because of the way it is used in another BIS_fnc or something.

 

thx for the answer though.

[killzone_kid 1330]

1 minute ago, bad benson said:

 

using it as part of my function to get all vehicle seats, their turret paths (if turret) and the name of the seat in the scroll menu. i do that to populate my own seat UI. i was just wondering why. maybe because of the way it is used in another BIS_fnc or something.

 

thx for the answer though.

Maybe so that one can query driver turret somehow? I don't know. What about allTurrets or even fullCrew? Those are faster and probably better than any config searching

[bad benson 1733]

15 minutes ago, killzone_kid said:

Maybe so that one can query driver turret somehow? I don't know. What about allTurrets or even fullCrew? Those are faster and probably better than any config searching

 

 

i just started testing methods and stuff.

 

so this is what i did for now.

 

fnc_getTurretDta =
{
	params ["_veh"];
	
	_paths = [_veh, []] call BIS_fnc_getTurrets;
	_cfgs = [_veh, config] call BIS_fnc_getTurrets;
	_turrets = _cfgs apply 
	{
		[
			getText (configfile >> "CfgVehicles" >> typeOf _veh >> "Turrets" >> configname _x >> "gunnername"),
			_paths select (_cfgs find _x)
		]
	};

	_turrets
};

hint str (cursorObject call fnc_getTurretDta);

 

i will take a look at allTurrets  but as you can see above i need some more data.

 

you can see the first entry there being weird. and i also just noticed the last one returns null for the path. not sure why. i might've fucked it up along the way.

 

EDIT:

 

ok so just adding

 

_cfgs deleteAt 0;

 

right after _cfgs is declared pretty much solved it. i guess i was looking for meaning too much. will test and see, if it causes any anomalies this way but so far it makes sense.

[f2k sel 164]

Why don't the AI turn around when they hear friendly gun fire it looks silly to have one unit possibly engaging an enemy  and the other units a few feet away not even looking at what he's shooting at.  Only if the AI return fire do they respond, I just don't find it realistic.

 

I know I could use the fired EVH and then run a script to alert others in the area but wouldn't that be a resource eater all those units running scripts on every shot.

Maybe an EVH that would run once when a unit spots and engages an enemy unit and resets when area is clear.

[das attorney 858]

Hi,

 

I wasn't sure where to put this question - if it's in the wrong place, then mods please move it.

 

I'm using remoteExecCall to delete groups (given that in the mission I'm working on, the groups could be created on server,hc, or client).

 

I keep seeing these messages in the report file (note that the timestamp is identical):

 

 0:10:54 A nil group passed as a target to RemoteExec(Call) 'deletegroup'
 0:10:54 A nil group passed as a target to RemoteExec(Call) 'deletegroup'
 0:10:54 A nil group passed as a target to RemoteExec(Call) 'deletegroup'
 0:10:54 A nil group passed as a target to RemoteExec(Call) 'deletegroup'
 0:10:54 A nil group passed as a target to RemoteExec(Call) 'deletegroup'
 0:10:54 A nil group passed as a target to RemoteExec(Call) 'deletegroup'
 0:10:54 A nil group passed as a target to RemoteExec(Call) 'deletegroup'
 0:10:54 A nil group passed as a target to RemoteExec(Call) 'deletegroup'
 0:10:54 A nil group passed as a target to RemoteExec(Call) 'deletegroup'

 

This is strange to me as the groups passed to it have definitely been previously created, so I would expect them to be grpNull, and not nil.

 

Plus the amount of messages I get in the report file are larger than the amount of groups that are scheduled to be deleted.  I have checked to make sure that the code is not executed multiple times via debug messages so everything seems ok there.

 

Here is my code (pretty standard I think).

 

    if ([] isEqualTo units _grp) then {
        _grp remoteExecCall [
            "deleteGroup",
            _grp
        ];
    };

 

I have a debug hint onscreen to keep track of how many empty groups there are, and it always says: [] so it looks like remoteExec is doing it's job, but then why the nil group... log messages?

 

I recently changed over to using remoteExec from an older function that worked OK.  It checks if the group is null and if not, then sends it to group owner to delete.  I would have thought that if any group was nil, there would be an error in the log because there is no check for nil variable passed to it.

 

Here is the older (but working) function:

// _this = group;   local can be used for groups (since Arma 3 v1.31.127204)
if not (isNull _this) then {
    if (local _this) then {
        deleteGroup _this
    } else {
        if (isServer) then {
            _this remoteExecCall [
                "horde_fnc_deleteGroupGlobal",
                groupOwner _this
            ];
        } else {
            _this remoteExecCall [
                "horde_fnc_deleteGroupGlobal",
                2
            ];
        };
    };
};
true

  Am I doing something wrong, or is there some sort of engine quirk with remoteExec??? 

 

Many thanks.