If a variable is enclosed in percent signs within an expression (e.g. %Var%), whatever that variable contains is assumed to be the name or partial name of another variable (if there is no such variable, %Var% resolves to a blank string). This is most commonly used to reference pseudo-array elements such as the following example:

Var := MyArray%A_Index% + 100

For backward compatibility, command parameters that are documented as "can be an expression" treat an isolated name in percent signs (e.g. %Var%, but not Array%i%) as though the percent signs are absent. This can be avoided by enclosing the reference in parentheses; e.g. Sleep (%Var%).

[AHK_L 52+]: In addition to normal variables, %Var% may resolve to an environment variable, the clipboard, or any reserved/read-only variable. Prior to revision 52, %Var% yielded an empty string in these cases.

Power. Both the base and the exponent may contain a decimal point. If the exponent is negative, the result will be formatted as a floating point number even if the base and exponent are both integers. Since ** is of higher precedence than unary minus, -2**2 is evaluated like -(2**2) and so yields -4. Thus, to raise a literal negative number to a power, enclose it in parentheses such as (-2)**2.

Note: A negative base combined with a fractional exponent such as (-2)**0.5 is not supported; it will yield an empty string. But both (-2)**2 and (-2)**2.0 are supported.

Unary minus (-): Although it uses the same symbol as the subtract operator, unary minus applies to only a single item or sub-expression as shown twice in this example: -(3 / -x). On a related note, any unary plus signs (+) within an expression are ignored.

Logical-not (!): If the operand is blank or 0, the result of applying logical-not is 1, which means "true". Otherwise, the result is 0 (false). For example: !x or !(y and z). Note: The word NOT is synonymous with ! except that ! has a higher precedence. [v1.0.46+]: Consecutive unary operators such as !!Var are allowed because they are evaluated in right-to-left order.

Bitwise-not (~): This inverts each bit of its operand. If the operand is a floating point value, it is truncated to an integer prior to the calculation. If the operand is between 0 and 4294967295 (0xffffffff), it will be treated as an unsigned 32-bit value. Otherwise, it is treated as a signed 64-bit value. For example, ~0xf0f evaluates to 0xfffff0f0 (4294963440).

Address (&): &MyVar retrieves the address of MyVar's contents in memory, which is typically used with DllCall structures. &MyVar also disables the caching of binary numbers in that variable, which can slow down its performance if it is ever used for math or numeric comparisons. Caching is re-enabled for a variable whenever its address changes (e.g. via VarSetCapacity()).

Dereference (*): *Expression assumes that Expression resolves to a numeric memory address; it retrieves the byte at that address as a number between 0 and 255 (0 is always retrieved if the address is 0; but any other invalid address must be avoided because it might crash the script). However, NumGet() generally performs much better when retrieving binary numbers.

Multiply (*): The result is an integer if both inputs are integers; otherwise, it is a floating point number.

True divide (/): Unlike EnvDiv, true division yields a floating point result even when both inputs are integers. For example, 3/2 yields 1.5 rather than 1, and 4/2 yields 2.0 rather than 2.

Floor divide (//): The double-slash operator uses high-performance integer division if the two inputs are integers. For example, 5//3 is 1 and 5//-3 is -1. If either of the inputs is in floating point format, floating point division is performed and the result is truncated to the nearest integer to the left. For example, 5//3.0 is 1.0 and 5.0//-3 is -2.0. Although the result of this floating point division is an integer, it is stored in floating point format so that anything else that uses it will see it as such. For modulo, see mod().

The *= and /= operators are a shorthand way to multiply or divide the value in a variable by another value. For example, Var*=2 produces the same result as Var:=Var*2 (though the former performs better).

Division by zero yields a blank result (empty string).

Add (+) and subtract (-). On a related note, the += and -= operators are a shorthand way to increment or decrement a variable. For example, Var+=2 produces the same result as Var:=Var+2 (though the former performs better). Similarly, a variable can be increased or decreased by 1 by using Var++, Var--, ++Var, or --Var.

Concatenate. The period (dot) operator is used to combine two items into a single string (there must be at least one space on each side of the period). You may also omit the period to achieve the same result (except where ambiguous such as x -y, or when the item on the right side has a leading ++ or --). When the dot is omitted, there should be at least one space between the items to be merged.
Example (expression method): Var := "The color is " . FoundColor
Example (traditional method): Var = The color is %FoundColor%

Sub-expressions can also be concatenated. For example: Var := "The net price is " . Price * (1 - Discount/100).

A line that begins with a period (or any other operator) is automatically appended to the line above it.

Equal (=), case-sensitive-equal (==), and not-equal (<> or !=). The operators != and <> are identical in function. The == operator behaves identically to = except when either of the inputs is not a number, in which case == is always case sensitive and = is always case insensitive (the method of insensitivity depends on StringCaseSense). By contrast, <> and != obey StringCaseSense.

Note: A quoted literal string such as "55" is always considered non-numeric in this context.

Assign. Performs an operation on the contents of a variable and stores the result back in the same variable (but in versions prior to 1.0.46, these could only be used as the leftmost operator on a line, and only the first five operators were supported). The simplest assignment operator is colon-equals (:=), which stores the result of an expression in a variable. For a description of what the other operators do, see their related entries in this table. For example, Var //= 2 performs floor division to divide Var by 2, then stores the result back in Var. Similarly, Var .= "abc" is a shorthand way of writing Var := Var . "abc".

Unlike most other operators, assignments are evaluated from right to left. Consequently, a line such as Var1 := Var2 := 0 first assigns 0 to Var2 then assigns Var2 to Var1.

If an assignment is used as the input for some other operator, its value is the variable itself. For example, the expression (Var+=2) > 50 is true if the newly-increased value in Var is greater than 50. This also allows an assignment to be passed ByRef, or its address taken; for example: &(x:="abc").

The precedence of the assignment operators is automatically raised when it would avoid a syntax error or provide more intuitive behavior. For example: not x:=y is evaluated as not (x:=y). Similarly, ++Var := X is evaluated as ++(Var := X); and Z>0 ? X:=2 : Y:=2 is evaluated as Z>0 ? (X:=2) : (Y:=2).

Known limitations caused by backward compatibility (these may be resolved in a future release): 1) When /= is the leftmost operator in an expression and it is not part of a multi-statement expression, it performs floor division unless one of the inputs is floating point (in all other cases, /= performs true division); 2) Date/time math is supported by += and -= only when that operator is the leftmost one on a line; 3) The operators +=, -=, and *= treat blank variables as zero, but only when they are alone on a line; for example, y:=1, x+=1 and MsgBox % x-=3 both produce a blank result when x is blank.

Comma (multi-statement) [v1.0.46+]. Commas may be used to write multiple sub-expressions on a single line. This is most commonly used to group together multiple assignments or function calls. For example: x:=1, y+=2, ++index, MyFunc(). Such statements are executed in order from left to right.

Note: A line that begins with a comma (or any other operator) is automatically appended to the line above it. See also: comma performance.

[v1.0.46.01+]: When a comma is followed immediately by a variable and an equal sign, that equal sign is automatically treated as an assignment (:=). For example, all of the following are assignments: x:=1, y=2, a=b=c. New scripts should not rely on this behavior as it may change. The rule applies only to plain variables and not double-derefs, so the following contains only one assignment: x:=1, %y%=2

Any sub-expression enclosed in parentheses. For example, (3 + 2) * 2 forces 3 + 2 to be evaluated first.

mod()
round()
abs()

%func%()

Deprecated: This syntax is not recommended for use. Use %func%() (for function names and objects) or func.Call() (for function objects) instead.

[AHK_L 48+]: Attempts to call an empty-named method of the object func. By convention, this is the object's "default" method. If func does not contain an object, the default base object is invoked instead.

[v1.0.95+]: If func contains a function name, the named function is called.

[AHK_L 60+]: Variadic function call. Params is an array (object) containing parameter values.

[AHK_L 31+]: Member access. Get or set a value or call a method of object x, where y is a parameter list (typically an array index or key) or an expression which returns a method name.

[v1.0.97+]: Array literal. If the open-bracket is not preceded by a value (or a sub-expression which yields a value), it is interpreted as the beginning of an array literal. For example, [a, b, c] is equivalent to Array(a, b, c) (a, b and c are variables).

See array syntax and object syntax for more details.

[v1.0.97+]: Object literal. Create an object or associative array. For example, x := {a: b} is equivalent to x := Object("a", b) or x := Object(), x.a := b. See Associative Arrays for details.

The number of the currently executing line within the script (or one of its #Include files). This line number will match the one shown by ListLines; it can be useful for error reporting such as this example: MsgBox Could not write to log file (line number %A_LineNumber%).

Since a compiled script has merged all its #Include files into one big script, its line numbering may be different than when it is run in non-compiled mode.

For non-compiled scripts: The full path and name of the EXE file that is actually running the current script. For example: C:\Program Files\AutoHotkey\AutoHotkey.exe

For compiled scripts: The same as the above except the AutoHotkey directory is discovered via the registry entry HKLM\SOFTWARE\AutoHotkey\InstallDir. If there is no such entry, A_AhkPath is blank.

Contains 1 if strings are Unicode (16-bit) and an empty string (which is considered false) if strings are ANSI (8-bit). The format of strings depends on the version of AutoHotkey.exe which is used to run the script, or if it is compiled, which bin file was used to compile it.

For ANSI executables prior to [v1.1.06], A_IsUnicode was left undefined; that is, the script could assign to it, and attempting to read it could trigger a UseUnsetGlobal warning. In later versions it is always defined and is read-only.

Contains 1 if the script is running as a compiled EXE and an empty string (which is considered false) if it is not.

For non-compiled scripts prior to [v1.1.06], A_IsCompiled was left undefined; that is, the script could assign to it, and attempting to read it could trigger a UseUnsetGlobal warning. In later versions it is always defined and is read-only.

Current 4-digit year (e.g. 2004). Synonymous with A_Year.

Note: To retrieve a formatted time or date appropriate for your locale and language, use FormatTime, OutputVar (time and long date) or FormatTime, OutputVar,, LongDate (retrieves long-format date).

Current 2-digit minute (00-59).

The current local time in YYYYMMDDHH24MISS format.

Note: Date and time math can be performed with EnvAdd and EnvSub. Also, FormatTime can format the date and/or time according to your locale or preferences.

The number of milliseconds since the computer was rebooted. By storing A_TickCount in a variable, elapsed time can later be measured by subtracting that variable from the latest A_TickCount value. For example:

StartTime := A_TickCount
Sleep, 1000
ElapsedTime := A_TickCount - StartTime
MsgBox,  %ElapsedTime% milliseconds have elapsed.

If you need more precision than A_TickCount's 10ms, use QueryPerformanceCounter().

The type of event that launched the current thread. If the thread was not launched via GUI action, this variable is blank. Otherwise, it contains one of the following strings:

Normal: The event was triggered by a single left-click or via keystrokes (↑, →, ↓, ←, Tab, Space, underlined shortcut key, etc.). This value is also used for menu bar items and the special events such as GuiClose and GuiEscape.

DoubleClick: The event was triggered by a double-click. Note: The first click of the click-pair will still cause a Normal event to be received first. In other words, the subroutine will be launched twice: once for the first click and again for the second.

RightClick: Occurs only for GuiContextMenu, ListViews, and TreeViews.

Context-sensitive values: For details see GuiContextMenu, GuiDropFiles, Slider, MonthCal, ListView, and TreeView.

Contains additional information about the following events:

• The OnClipboardChange label
• Mouse wheel hotkeys (WheelDown/Up/Left/Right)
• OnMessage()
• RegisterCallback()
• Regular Expression Callouts
• GUI events, namely GuiContextMenu, GuiDropFiles, ListBox, ListView, TreeView, and StatusBar. If there is no additional information for an event, A_EventInfo contains 0.

The most recently executed hotkey or non-auto-replace hotstring (blank if none), e.g. #z. This value will change if the current thread is interrupted by another hotkey, so be sure to copy it into another variable immediately if you need the original value for later use in a subroutine.

When a hotkey is first created -- either by the Hotkey command or a double-colon label in the script -- its key name and the ordering of its modifier symbols becomes the permanent name of that hotkey, shared by all variants of the hotkey.

See also: A_ThisLabel

Contains the same string as the environment's ComSpec variable (e.g. C:\Windows\system32\cmd.exe). Often used with Run/RunWait.

One of the following strings, if appropriate: WIN_7 [requires AHK_L 42+], WIN_8 [requires v1.1.08+], WIN_8.1 [requires v1.1.15+], WIN_VISTA, WIN_2003, WIN_XP, WIN_2000.

Applying compatibility settings in the AutoHotkey executable or compiled script's properties causes the OS to report a different version number, which is reflected by A_OSVersion.

[v1.1.20+]: If the OS version is not recognized as one of those listed above, a string in the format "major.minor.build" is returned. For example, 10.0.14393 is Windows 10 build 14393, also known as version 1607.

; This example is obsolete as these operating systems are no longer supported.
if A_OSVersion in WIN_NT4,WIN_95,WIN_98,WIN_ME  ; Note: No spaces around commas.
{
    MsgBox This script requires Windows 2000/XP or later.
    ExitApp
}

The Program Files directory (e.g. C:\Program Files or C:\Program Files (x86)). This is usually the same as the ProgramFiles environment variable.

On 64-bit systems (and not 32-bit systems), the following applies:

• If the executable (EXE) that is running the script is 32-bit, A_ProgramFiles returns the path of the "Program Files (x86)" directory.
• For 32-bit processes, the ProgramW6432 environment variable contains the path of the 64-bit Program Files directory. On Windows 7 and later, it is also set for 64-bit processes.
• The ProgramFiles(x86) environment variable contains the path of the 32-bit Program Files directory.

[v1.0.43.08+]: The A_ prefix may be omitted, which helps ease the transition to #NoEnv.

If the current user has admin rights, this variable contains 1. Otherwise, it contains 0.

To have the script restart itself as admin (or show a prompt to the user requesting admin), use Run *RunAs. However, note that running the script as admin causes all programs launched by the script to also run as admin. For a possible alternative, see the FAQ.

A_ScreenWidth
A_ScreenHeight

The width and height of the primary monitor, in pixels (e.g. 1024 and 768).

To discover the dimensions of other monitors in a multi-monitor system, use SysGet.

To instead discover the width and height of the entire desktop (even if it spans multiple monitors), use the following example:

SysGet, VirtualWidth, 78
SysGet, VirtualHeight, 79

In addition, use SysGet to discover the work area of a monitor, which can be smaller than the monitor's total area because the taskbar and other registered desktop toolbars are excluded.

The type of mouse cursor currently being displayed. It will be one of the following words: AppStarting, Arrow, Cross, Help, IBeam, Icon, No, Size, SizeAll, SizeNESW, SizeNS, SizeNWSE, SizeWE, UpArrow, Wait, Unknown. The acronyms used with the size-type cursors are compass directions, e.g. NESW = NorthEast+SouthWest. The hand-shaped cursors (pointing and grabbing) are classified as Unknown.

The current X and Y coordinates of the caret (text insertion point). The coordinates are relative to the active window unless CoordMode is used to make them relative to the entire screen. If there is no active window or the caret position cannot be determined, these variables are blank.

The following script allows you to move the caret around to see its current position displayed in an auto-update tooltip. Note that some windows (e.g. certain versions of MS Word) report the same caret position regardless of its actual position.

#Persistent
SetTimer, WatchCaret, 100
return
WatchCaret:
    ToolTip, X%A_CaretX% Y%A_CaretY%, A_CaretX, A_CaretY - 20
return

Contain 1 and 0. They can be used to make a script more readable. See Boolean Values for details.