Function: Difference between revisions

From Bohemia Interactive Community
Jump to navigation Jump to search
(→‎Introduction: Updated link)
m (Text replacement - "[[Image:" to "[[File:")
 
(105 intermediate revisions by 18 users not shown)
Line 1: Line 1:
==Introduction==
{{TOC|side}}


Functions were first introduced into an OFP: Resistance patch.
{{Feature|important|For a proper writing tutorial, see [[Arma 3: Writing a Function|Writing a Function]].<br>
See also:
* [[Arma 2: Functions Library]]
* [[Take On Helicopters: Functions Library]]
* [[Arma 3: Functions Library]]
}}


Function files use the [[sqf syntax|SQF syntax]]


==Usage==
A '''function''' contains [[Code|code]] which usually consists of an ''input'', ''processing'' and ''output'' part. Functions were first introduced in the [[Operation Flashpoint: Resistance Version History|{{ofpr}}]] patch.
The main advantages of functions are:


A function is much like a regular scripting command except that you can use functions to create something like a custom command. A '''function''' is simply a chunk of code that does something, the function can then return a value to the point which 'called' that function or it can simply return [[Nothing]].
; Improved legibility
: By writing the code once and re-using it by calling the function


Functions differ from scripts in that they halt all other game engine processes until the function has completed its instructions.  This means functions run faster than scripts, and the result of functions is immediate and unambiguous. It can also mean that if a function takes too long to run it will have an adverse effect on game play - large functions or CPU intensive functions can cause the game to seize up until it completes.  When creating a functions you want the function to be short and sweet to achieve the best results.
; Easier debugging
: If your function contains an error, you only have to fix it in one place


Functions have a limitation of 10,000 loops before they are forced to exit by the game engine in order to maintain some level of stability.


===Operation Flashpoint===
== Types of function ==


Functions are first compiled via [[preprocessfile]] command or loaded dynamically via the [[loadfile]] command. They are then executed via the [[call]] command. Functions in OFP cannot be paused and must finish execution before the game will continue.
=== Functions-as-files ===


===Armed Assault===
:''See [[Function#Call|calling functions]] for how a function-as-file is called.
Functions as files are functions stored within a file. These are usually used for larger and more complex functions. The code is evaluated in the same way, however, there are additional commands which must include the file before the function itself can be called.


Starting with Armed Assault, functions can wait suspended (like SQS or SQF-style scripts), using [[sleep]] or [[waitUntil]].  
<sqf>
// Code inside fn_showHint.sqf
hint "Function was executed!"; // Function will show a hint when executed
</sqf>


[[Image:Function_Execution_Diagram.jpg]]
=== Inline functions ===


Inline functions are functions are technically [[Code|code]] which is often stored within a variable or declared as a function parameter. Inline functions operate the same way as functions-as-files as both are evaluated in the same way, but the difference is that inline functions are stored within parentheses {{hl|{}}}, whereas functions-as-files do not require these.
<sqf>
// Inline function as variable
TAG_fnc_showHint =
{
hint "Function was executed!"; // Function will show a hint when executed
};
call TAG_fnc_showHint;
</sqf>
== Anatomy of a function ==
When scripting, there are two types of functions: functions-as-files and inline functions.
Functions-as-files are instances where the a whole file itself is used to house a function, whereas inline functions are either contained within a variable or as a parameter of a function.
Some built-in functions require functions-as-files, whereas most will support both.
=== Parameters (input) ===
Parameters for functions are available to the function via the [[Magic Variables|magic variable]] [[Magic Variables#this|_this]].
Declaration of parameters can be done using the [[params]] command. Common practice for defining parameters is done via the use of [[private]] variables and defined variables.
<sqf>
// Code inside fn_showHint.sqf
params ["_text"];
hint _text;
</sqf>
<sqf>
// Inline function as variable
TAG_fnc_showHint =
{
params ["_text"];
hint _text;
};
</sqf>
<sqf>"Function was executed!" call TAG_fnc_showHint;</sqf>
Parameters passed to functions are passed '''before''' the function, rather than after (such as the mathematical or c-like syntax ''f(x)'').
<sqf>
// Array variable as parameter
_myTempParams = [_parameterOne, _parameterTwo];
_myTempVariableTwo = _myTempParams call myInlineFunction;
</sqf>
<sqf>
// Array as parameter
_myTempVariable = [_parameterOne,_parameterTwo] call myInlineFunction;
</sqf>
=== Return Values (output) ===
The value of the last executed statement in a function is returned to the calling instance.
<sqf>
my_fnc = {
if (_this > 0) exitWith {
_this + 1
};
_this - 1
};
hint str ( 5 call my_fnc); //  6
hint str (-5 call my_fnc); // -6
</sqf>
In the first example "_this + 1" is the last executed statement in my_fnc, in the second example it is "_this - 1".
{{Feature|informative|It was believed at one point that the returning statement should not have a semicolon {{hl|;}} after it; this is false and having a semicolon does not change anything and is recommended.}}
<sqf>
my_fnc = {
a = 1;
b = 2;
c = a + b;
c // ← fine
};
</sqf>
<sqf>
my_fnc = {
a = 1;
b = 2;
c = a + b;
c; // ← better
};
</sqf>
More examples:
<sqf>
// myCode.sqf
private _myName = _this select 0;
private _returnMe = "FAIL";
if (_myName == "Test") then
{
_returnMe = "PASS";
};
_returnMe;
</sqf>
<sqf>
// myCodeInline
myCodeReturnValue = {
private _myName = _this select 0;
private _returnMe = "FAIL";
if (_myName == "Kaboom") then
{
_returnMe = "PASS";
};
_returnMe
};
_myCalledVariable = ["Kaboom"] call myCodeReturnValue; // "PASS"
_myCalledVariableFail = ["Blah"] call myCodeReturnValue; // "FAIL"
</sqf>
<sqf>
// return.sqf
STATEMENT 1;
STATEMENT 2;
RETURN_VALUE
</sqf>
<sqf>
// test.sqf
value = call compile preprocessFile "return.sqf";
// value is now RETURN_VALUE
call compile preprocessFile "return.sqf";
// valid, but RETURN_VALUE is not saved anywhere
</sqf>
=== Execution ===
[[File: Function_Execution.png|frame|right||Function Execution Diagram in [[Scheduler#Scheduled_Environment|scheduled environment]]<br><br>'''Executing Instance :''' [[Script File|script]], function or game engine]]
Functions can be executed from several points in the game:
* Other [[Script File|scripts]]
* Other functions
* Scripting lines in the [[ArmA:_Mission_Editor|Mission Editor]]
* [[:Category:Event Handlers|Event Handlers]] in addon config files
Functions are first loaded as [[String]] from a file via [[preprocessFile]] or [[loadFile]]. They are then executed via the [[call]] or [[spawn]] command. Since Armed Assault the loaded [[String]] needs to be [[compile|compiled]] in order to convert it to [[Code]], which is required for [[call]] or [[spawn]].
==== Call ====
Example (Operation Flashpoint):
<sqf>
myFunction1 = loadFile "myFunction1.sqf";
myFunction2 = preprocessFile "myFunction2.sqf";
call myFunction1;
[1, 2] call myFunction2;
</sqf>
Example (Armed Assault):
<sqf>
myFunction1 = compile loadFile "myFunction1.sqf";
myFunction2 = compile preprocessFile "myFunction2.sqf";
_result1 = call myFunction1;
_result2 = [1, 2] call myFunction2;
</sqf>
Functions executed using [[call]] are run ''within'' the executing instance, which waits for the result of the function.
Unlike scripts, functions halt all other game engine processes until the function has completed its instructions.
This means functions run faster than scripts, and the result of functions is immediate and unambiguous.
It can also mean that if a function takes too long to run it will have an adverse effect on game play - large functions or CPU intensive functions can cause the game to seize up until it completes.
When creating a functions you want the function to be short and sweet to achieve the best results.
{{Feature|informative|You can still use the special variables and commands of [[Script File|scripts]] in functions ({{arma1}} only)!}}
==== Spawn ====
Functions may also be executed using [[spawn]], but then the function result is not accessible, making it behave more like a procedure.
Spawned functions will run asynchronously or ''alongside'' the executing instance. This helps prevent large CPU intensive functions from seizing up the game.
Example (Armed Assault):
<sqf>
myFunction1 = compile loadFile "myFunction1.sqf";
myFunction2 = compile preprocessFile "myFunction2.sqf";
_param spawn myFunction1;
[1, 2] spawn myFunction2;
</sqf>
== Examples ==
=== Example 1: max.sqf ===


==Examples==
<b>Example 1</b> (max.sqf)
In this example the function returns maximum of first and second argument.
In this example the function returns maximum of first and second argument.


comment "Return maximum of first and second argument"
max.sqf
[[private variableNameList|private]] {"_a","_b"};
<sqf>
_a = _this [[select]] 0;
//"Return maximum of first and second argument";
_b = _this [[select]] 1;
params ["_a", "_b"];
[[if]] (_a>_b) [[then]] {_a} [[else]] {_b}
[_b, _a] select (_a > _b);
</sqf>
 
alternative max.sqf (big boys code :))
<sqf>(_this select 0) max (_this select 1);</sqf>
 
executing script:
<sqf>
fMax = compile preprocessFile "max.sqf";
maxValue = [3, 5] call fMax;
 
// maxValue is now 5
</sqf>


=== Example 2: infantrySafe.sqf ===


<b>Example 2</b> (infantrySafe.sqf)
In this example the function returns no value and switches all units to safe mode.
In this example the function returns no value and switches all units to safe mode.
<sqf>
// Switch all infantry units to safe mode
{
if (vehicle _x == _x) then
{
_x setBehaviour "safe";
}
} forEach _this;
</sqf>


comment "Switch all infantry units to safe mode";
=== Example 3: Inline Function ===
{
  [[if]] ([[vehicle]] _x == _x) [[then]]
  {
  _x [[setBehaviour]] "safe"
  }
} [[forEach]] _this
 


<b>Example 3</b> (inline function)
An inline-function can be created in any script:
An inline-function can be created in any script:
<sqf>FNC_sayhello = {hint format ["hello %1", _this]};</sqf>


FNC_sayhello = {hint format["hello %1",_this];}
This function can then be called (in other scripts, functions, unit's init lines, trigger activation fields, etc.) via:
<sqf>name player call FNC_sayhello</sqf>


This function can then be called (in other scripts, functions, unit's init lines, trigger activation fields, etc.) via:
In case the function doesn't require any arguments you can just call the function.
<sqf>call FNC_helloall</sqf>


name player call FNC_sayhello


Notice that there are '''no''' brackets around the functions arguments which precede the call command.<br>In case the function doesn't require any arguments you can use empty brackets instead (<code>[] call FNC_helloall</code>).
== See also ==


* [[Script File|Scripts]]
* [[SQF Syntax]]
* [[call]]


==See also==
[[call]]


[[Category: Scripting_Topics]]
[[Category: Scripting Topics]]

Latest revision as of 23:11, 20 November 2023


A function contains code which usually consists of an input, processing and output part. Functions were first introduced in the Operation Flashpoint: Resistance patch. The main advantages of functions are:

Improved legibility
By writing the code once and re-using it by calling the function
Easier debugging
If your function contains an error, you only have to fix it in one place


Types of function

Functions-as-files

See calling functions for how a function-as-file is called.

Functions as files are functions stored within a file. These are usually used for larger and more complex functions. The code is evaluated in the same way, however, there are additional commands which must include the file before the function itself can be called.

// Code inside fn_showHint.sqf hint "Function was executed!"; // Function will show a hint when executed

Inline functions

Inline functions are functions are technically code which is often stored within a variable or declared as a function parameter. Inline functions operate the same way as functions-as-files as both are evaluated in the same way, but the difference is that inline functions are stored within parentheses {}, whereas functions-as-files do not require these.

// Inline function as variable TAG_fnc_showHint = { hint "Function was executed!"; // Function will show a hint when executed }; call TAG_fnc_showHint;


Anatomy of a function

When scripting, there are two types of functions: functions-as-files and inline functions. Functions-as-files are instances where the a whole file itself is used to house a function, whereas inline functions are either contained within a variable or as a parameter of a function. Some built-in functions require functions-as-files, whereas most will support both.

Parameters (input)

Parameters for functions are available to the function via the magic variable _this. Declaration of parameters can be done using the params command. Common practice for defining parameters is done via the use of private variables and defined variables.

// Code inside fn_showHint.sqf params ["_text"]; hint _text;

// Inline function as variable TAG_fnc_showHint = { params ["_text"]; hint _text; };

"Function was executed!" call TAG_fnc_showHint;

Parameters passed to functions are passed before the function, rather than after (such as the mathematical or c-like syntax f(x)).

// Array variable as parameter _myTempParams = [_parameterOne, _parameterTwo]; _myTempVariableTwo = _myTempParams call myInlineFunction;

// Array as parameter _myTempVariable = [_parameterOne,_parameterTwo] call myInlineFunction;

Return Values (output)

The value of the last executed statement in a function is returned to the calling instance.

my_fnc = { if (_this > 0) exitWith { _this + 1 }; _this - 1 }; hint str ( 5 call my_fnc); // 6 hint str (-5 call my_fnc); // -6

In the first example "_this + 1" is the last executed statement in my_fnc, in the second example it is "_this - 1".

It was believed at one point that the returning statement should not have a semicolon
after it; this is false and having a semicolon does not change anything and is recommended.

my_fnc = { a = 1; b = 2; c = a + b; c // ← fine };
my_fnc = { a = 1; b = 2; c = a + b; c; // ← better };

More examples:

// myCode.sqf private _myName = _this select 0; private _returnMe = "FAIL"; if (_myName == "Test") then { _returnMe = "PASS"; }; _returnMe;

// myCodeInline myCodeReturnValue = { private _myName = _this select 0; private _returnMe = "FAIL"; if (_myName == "Kaboom") then { _returnMe = "PASS"; }; _returnMe }; _myCalledVariable = ["Kaboom"] call myCodeReturnValue; // "PASS" _myCalledVariableFail = ["Blah"] call myCodeReturnValue; // "FAIL"

// return.sqf STATEMENT 1; STATEMENT 2; RETURN_VALUE

// test.sqf value = call compile preprocessFile "return.sqf"; // value is now RETURN_VALUE call compile preprocessFile "return.sqf"; // valid, but RETURN_VALUE is not saved anywhere

Execution

Function Execution Diagram in scheduled environment

Executing Instance : script, function or game engine

Functions can be executed from several points in the game:

Functions are first loaded as String from a file via preprocessFile or loadFile. They are then executed via the call or spawn command. Since Armed Assault the loaded String needs to be compiled in order to convert it to Code, which is required for call or spawn.

Call

Example (Operation Flashpoint):

myFunction1 = loadFile "myFunction1.sqf"; myFunction2 = preprocessFile "myFunction2.sqf"; call myFunction1; [1, 2] call myFunction2;

Example (Armed Assault):

myFunction1 = compile loadFile "myFunction1.sqf"; myFunction2 = compile preprocessFile "myFunction2.sqf"; _result1 = call myFunction1; _result2 = [1, 2] call myFunction2;

Functions executed using call are run within the executing instance, which waits for the result of the function. Unlike scripts, functions halt all other game engine processes until the function has completed its instructions. This means functions run faster than scripts, and the result of functions is immediate and unambiguous. It can also mean that if a function takes too long to run it will have an adverse effect on game play - large functions or CPU intensive functions can cause the game to seize up until it completes. When creating a functions you want the function to be short and sweet to achieve the best results.

You can still use the special variables and commands of scripts in functions (Armed Assault only)!

Spawn

Functions may also be executed using spawn, but then the function result is not accessible, making it behave more like a procedure. Spawned functions will run asynchronously or alongside the executing instance. This helps prevent large CPU intensive functions from seizing up the game.

Example (Armed Assault):

myFunction1 = compile loadFile "myFunction1.sqf"; myFunction2 = compile preprocessFile "myFunction2.sqf"; _param spawn myFunction1; [1, 2] spawn myFunction2;


Examples

Example 1: max.sqf

In this example the function returns maximum of first and second argument.

max.sqf

//"Return maximum of first and second argument"; params ["_a", "_b"]; [_b, _a] select (_a > _b);

alternative max.sqf (big boys code :))

executing script:

fMax = compile preprocessFile "max.sqf"; maxValue = [3, 5] call fMax; // maxValue is now 5

Example 2: infantrySafe.sqf

In this example the function returns no value and switches all units to safe mode.

// Switch all infantry units to safe mode { if (vehicle _x == _x) then { _x setBehaviour "safe"; } } forEach _this;

Example 3: Inline Function

An inline-function can be created in any script:

FNC_sayhello = {hint format ["hello %1", _this]};

This function can then be called (in other scripts, functions, unit's init lines, trigger activation fields, etc.) via:

name player call FNC_sayhello

In case the function doesn't require any arguments you can just call the function.

call FNC_helloall


See also