OPERATOR
Declares or defines an overloaded operator.
Syntax:
{ Type | Class | Union } typename
{ Type | Class | Union } typename
declare Operator unary_op ( [ byref | byval ] rhs as datatype ) as datatype
declare Operator binary_op ( [ byref | byval ] lhs as datatype, [ byref | byval ] rhs as datatype ) as datatype
Operator typename.cast () [ byref ] as datatype [ Export ]
Operator typename.@ () [ byref ] as datatype ptr [ Export ]
Operator typename.assignment_op ( [ byref | byval ] rhs as datatype ) [ Export ]
Operator typename.[] ( index as datatype ) [ byref ] as datatype [ Export ]
Operator unary_op ( [ byref | byval ] rhs as datatype ) as datatype [ Export ]
Operator binary_op ( [ byref | byval ] lhs as datatype, [ byref | byval ] rhs as datatype ) as datatype [ Export ]
Operator typename.new ( size as uinteger ) as any ptr [ Export ]
Operator typename.new[] ( size as uinteger ) as any ptr [ Export ]
Operator typename.delete ( buf as any ptr ) [ Export ]
Operator typename.delete[] ( buf as any ptr ) [ Export ]
declare Operator cast () [ byref ] as datatype
declare Operator @ () [ byref ] as datatype ptr
declare Operator assignment_op ( [ byref | byval ] rhs as datatype )
declare Operator [] ( index as datatype ) [ byref ] as datatype
declare Operator new ( size as uinteger ) as any ptr
declare Operator new[] ( size as uinteger ) as any ptr
declare Operator delete ( buf as any ptr )
declare Operator delete[] ( buf as any ptr )
End { Type | Class | Union }declare Operator @ () [ byref ] as datatype ptr
declare Operator assignment_op ( [ byref | byval ] rhs as datatype )
declare Operator [] ( index as datatype ) [ byref ] as datatype
declare Operator new ( size as uinteger ) as any ptr
declare Operator new[] ( size as uinteger ) as any ptr
declare Operator delete ( buf as any ptr )
declare Operator delete[] ( buf as any ptr )
{ Type | Class | Union } typename
declare Operator For ()
declare Operator For ( [ byref | byval ] stp as typename )
declare Operator Step ()
declare Operator Step ( [ byref | byval ] stp as typename )
declare Operator Next ( [ byref | byval ] cond as typename ) as Integer
declare Operator Next ( [ byref | byval ] cond as typename, [ byref | byval ] stp as typename ) as Integer
End { Type | Class | Union }declare Operator For ( [ byref | byval ] stp as typename )
declare Operator Step ()
declare Operator Step ( [ byref | byval ] stp as typename )
declare Operator Next ( [ byref | byval ] cond as typename ) as Integer
declare Operator Next ( [ byref | byval ] cond as typename, [ byref | byval ] stp as typename ) as Integer
declare Operator unary_op ( [ byref | byval ] rhs as datatype ) as datatype
declare Operator binary_op ( [ byref | byval ] lhs as datatype, [ byref | byval ] rhs as datatype ) as datatype
Operator typename.cast () [ byref ] as datatype [ Export ]
Operator typename.@ () [ byref ] as datatype ptr [ Export ]
Operator typename.assignment_op ( [ byref | byval ] rhs as datatype ) [ Export ]
Operator typename.[] ( index as datatype ) [ byref ] as datatype [ Export ]
Operator unary_op ( [ byref | byval ] rhs as datatype ) as datatype [ Export ]
Operator binary_op ( [ byref | byval ] lhs as datatype, [ byref | byval ] rhs as datatype ) as datatype [ Export ]
Operator typename.new ( size as uinteger ) as any ptr [ Export ]
Operator typename.new[] ( size as uinteger ) as any ptr [ Export ]
Operator typename.delete ( buf as any ptr ) [ Export ]
Operator typename.delete[] ( buf as any ptr ) [ Export ]
Parameters:
typename
assignment_op
assignment_op
let += -= *= &= /= \= mod= shl= shr= and= or= xor= imp= eqv= ^=
unary_op
- not * -> abs sgn fix frac int exp log sin asin cos acos tan atn len sqr
binary_op
+ - * & / \ mod shl shr and or xor imp eqv ^ = <> < > <= >=
Description:
The built in operators like =, +, and cast have predefined behaviors when used in expressions. These operators can be overloaded to do something other than predefined operations when at least one of the arguments to the operator is a Type, Class, Enum, or Union data type.
Operators are just functions. The operator '+' has functionality like Function Plus( A as DataType, B as DataType ) as DataType. See Operator Overloading for more information. Operators can be overloaded to accept different data types as parameters. The Cast Operator is the only operator (or function) that can be declared multiple times when only the return type differs, but not the same as the Type, Class, or Union it is declared in (for not explicit usage, the compiler may decide which cast overload to call based on how the object is used).
Non-static operator members are declared inside the Type, Class, or Union. Global operators are declared outside. All operator definitions (procedure bodies) must appear outside.
Let, Cast, and other assignment operators must be declared inside the Type, Class, or Union. As all non-static member procedures, they have passed a hidden this parameter.
Unary operators must be declared outside the Type, Class, or Union and have a return data type explicitly declared. Unary operators can be overloaded to return any valid data type, except for Operator -> (Pointer to member access) which must return a Type, Class, or Union data type.
Binary operators must be declared outside the Type, Class, or Union and have a return data type explicitly declared. Binary operators can be overloaded with valid data types, including for relational operators, which can also return any valid data type.
Let refers to the assignment operator, as in LET a=b. The Let keyword is omitted in common practice, and is not allowed in the -lang fb dialect. However, Let() can be used to assign the fields of a UDT to multiple variables.
See For, Step, and Next for more information on overloading the For..Next statement for use with user defined types.
Although declared inside the Type, Class, or Union, the member operators New, New[], Delete, and Delete[] are always static, even if not explicitly declared (static keyword is unnecessary but allowed).
Operators are just functions. The operator '+' has functionality like Function Plus( A as DataType, B as DataType ) as DataType. See Operator Overloading for more information. Operators can be overloaded to accept different data types as parameters. The Cast Operator is the only operator (or function) that can be declared multiple times when only the return type differs, but not the same as the Type, Class, or Union it is declared in (for not explicit usage, the compiler may decide which cast overload to call based on how the object is used).
Non-static operator members are declared inside the Type, Class, or Union. Global operators are declared outside. All operator definitions (procedure bodies) must appear outside.
Let, Cast, and other assignment operators must be declared inside the Type, Class, or Union. As all non-static member procedures, they have passed a hidden this parameter.
Unary operators must be declared outside the Type, Class, or Union and have a return data type explicitly declared. Unary operators can be overloaded to return any valid data type, except for Operator -> (Pointer to member access) which must return a Type, Class, or Union data type.
Binary operators must be declared outside the Type, Class, or Union and have a return data type explicitly declared. Binary operators can be overloaded with valid data types, including for relational operators, which can also return any valid data type.
Let refers to the assignment operator, as in LET a=b. The Let keyword is omitted in common practice, and is not allowed in the -lang fb dialect. However, Let() can be used to assign the fields of a UDT to multiple variables.
See For, Step, and Next for more information on overloading the For..Next statement for use with user defined types.
Although declared inside the Type, Class, or Union, the member operators New, New[], Delete, and Delete[] are always static, even if not explicitly declared (static keyword is unnecessary but allowed).
Examples:
'' operator1.bas
Type Vector2D
As Single x, y
'' Return a string containing the vector data.
Declare Operator Cast() As String
'' Multiply the vector by a scalar.
Declare Operator *= ( ByVal rhs As Single )
End Type
'' Allow two vectors to be able to be added together.
Declare Operator + ( ByRef lhs As Vector2D, ByRef rhs As Vector2D ) As Vector2D
'' Return the modulus (single) of the vector using the overloaded operator abs().
Declare Operator Abs ( ByRef rhs As Vector2D ) As Single
Operator Vector2D.Cast () As String
Return "(" + Str(x) + ", " + Str(y) + ")"
End Operator
Operator Vector2D.*= ( ByVal rhs As Single )
This.x *= rhs
This.y *= rhs
End Operator
Operator + ( ByRef lhs As Vector2D, ByRef rhs As Vector2D ) As Vector2D
Return type<Vector2D>( lhs.x + rhs.x, lhs.y + rhs.y )
End Operator
Operator Abs ( ByRef rhs As Vector2D ) As Single
Return Sqr( rhs.x * rhs.x + rhs.y * rhs.y )
End Operator
Dim a As Vector2D = type<Vector2D>( 1.2, 3.4 )
Dim b As Vector2D = type<Vector2D>( 8.9, 6.7 )
Dim c As Vector2D = type<Vector2D>( 4.3, 5.6 )
Print "a = "; a, "abs(a) ="; Abs( a )
Print "b = "; b, "abs(b) ="; Abs( b )
Print "a + b = "; a + b, "abs(a+b) ="; Abs( a + b )
Print "c = "; c, "abs(c) ="; Abs( c )
Print "'c *= 3'"
c *= 3
Print "c = "; c, "abs(c) ="; Abs( c )
Type Vector2D
As Single x, y
'' Return a string containing the vector data.
Declare Operator Cast() As String
'' Multiply the vector by a scalar.
Declare Operator *= ( ByVal rhs As Single )
End Type
'' Allow two vectors to be able to be added together.
Declare Operator + ( ByRef lhs As Vector2D, ByRef rhs As Vector2D ) As Vector2D
'' Return the modulus (single) of the vector using the overloaded operator abs().
Declare Operator Abs ( ByRef rhs As Vector2D ) As Single
Operator Vector2D.Cast () As String
Return "(" + Str(x) + ", " + Str(y) + ")"
End Operator
Operator Vector2D.*= ( ByVal rhs As Single )
This.x *= rhs
This.y *= rhs
End Operator
Operator + ( ByRef lhs As Vector2D, ByRef rhs As Vector2D ) As Vector2D
Return type<Vector2D>( lhs.x + rhs.x, lhs.y + rhs.y )
End Operator
Operator Abs ( ByRef rhs As Vector2D ) As Single
Return Sqr( rhs.x * rhs.x + rhs.y * rhs.y )
End Operator
Dim a As Vector2D = type<Vector2D>( 1.2, 3.4 )
Dim b As Vector2D = type<Vector2D>( 8.9, 6.7 )
Dim c As Vector2D = type<Vector2D>( 4.3, 5.6 )
Print "a = "; a, "abs(a) ="; Abs( a )
Print "b = "; b, "abs(b) ="; Abs( b )
Print "a + b = "; a + b, "abs(a+b) ="; Abs( a + b )
Print "c = "; c, "abs(c) ="; Abs( c )
Print "'c *= 3'"
c *= 3
Print "c = "; c, "abs(c) ="; Abs( c )
Small use case of the operator "[]": simplest smart pointers for byte buffers.
'' operator3.bas
'' A smart pointer is an object which behaves like a pointer but does more than a pointer:
'' - This object is flexible as a pointer and has the advantage of being an object,
'' like constructor and destructor called automatically.
'' - Therefore, the destructor of the smart pointer will be automatically called
'' when this object goes out of scope, and it will delete the user pointer.
'' Example of simplest smart pointers for byte buffers:
'' - Constructor and destructor allow to allocate, deallocate, and resize the byte buffer.
'' - Pointer index operator allows to access buffer elements.
'' - Copy-constructor and let-operator are just declared in private section,
'' in order to disallow copy construction and any assignment.
Type smartByteBuffer
Public:
Declare Constructor (ByVal size As UInteger = 0)
Declare Operator [] (ByVal index As UInteger) ByRef As Byte
Declare Destructor ()
Private:
Declare Constructor (ByRef rhs As smartByteBuffer)
Declare Operator Let (ByRef rhs As smartByteBuffer)
Dim As Byte Ptr psbb
End Type
Constructor smartByteBuffer (ByVal size As UInteger = 0)
This.Destructor()
If size > 0 Then
This.psbb = New Byte[size]
Print "Byte buffer allocated"
End If
End Constructor
Operator smartByteBuffer.[] (ByVal index As UInteger) ByRef As Byte
Return This.psbb[index]
End Operator
Destructor smartByteBuffer ()
If This.psbb > 0 Then
Delete[] This.psbb
This.psbb = 0
Print "Byte buffer deallocated"
End If
End Destructor
Scope
Dim As smartByteBuffer sbb = smartByteBuffer(256)
For I As Integer = 0 To 255
sbb[I] = I - 128
Next I
Print
For I As Integer = 0 To 255
Print Using "#####"; sbb[I];
Next I
Print
End Scope
'' A smart pointer is an object which behaves like a pointer but does more than a pointer:
'' - This object is flexible as a pointer and has the advantage of being an object,
'' like constructor and destructor called automatically.
'' - Therefore, the destructor of the smart pointer will be automatically called
'' when this object goes out of scope, and it will delete the user pointer.
'' Example of simplest smart pointers for byte buffers:
'' - Constructor and destructor allow to allocate, deallocate, and resize the byte buffer.
'' - Pointer index operator allows to access buffer elements.
'' - Copy-constructor and let-operator are just declared in private section,
'' in order to disallow copy construction and any assignment.
Type smartByteBuffer
Public:
Declare Constructor (ByVal size As UInteger = 0)
Declare Operator [] (ByVal index As UInteger) ByRef As Byte
Declare Destructor ()
Private:
Declare Constructor (ByRef rhs As smartByteBuffer)
Declare Operator Let (ByRef rhs As smartByteBuffer)
Dim As Byte Ptr psbb
End Type
Constructor smartByteBuffer (ByVal size As UInteger = 0)
This.Destructor()
If size > 0 Then
This.psbb = New Byte[size]
Print "Byte buffer allocated"
End If
End Constructor
Operator smartByteBuffer.[] (ByVal index As UInteger) ByRef As Byte
Return This.psbb[index]
End Operator
Destructor smartByteBuffer ()
If This.psbb > 0 Then
Delete[] This.psbb
This.psbb = 0
Print "Byte buffer deallocated"
End If
End Destructor
Scope
Dim As smartByteBuffer sbb = smartByteBuffer(256)
For I As Integer = 0 To 255
sbb[I] = I - 128
Next I
For I As Integer = 0 To 255
Print Using "#####"; sbb[I];
Next I
End Scope
Dialect Differences:
- Only available in the -lang fb dialect.
See also:
Back to User Defined Types
