MethodBuilder.GetILGenerator Método
Importante
Algumas informações se referem a produtos de pré-lançamento que podem ser substancialmente modificados antes do lançamento. A Microsoft não oferece garantias, expressas ou implícitas, das informações aqui fornecidas.
Retorna um ILGenerator
para este método.
GetILGenerator() |
Retorna um |
GetILGenerator(Int32) |
Retorna um |
- Origem:
- MethodBuilder.cs
- Origem:
- MethodBuilder.cs
- Origem:
- MethodBuilder.cs
Retorna um ILGenerator
para esse método com um tamanho de fluxo MSIL (linguagem intermediária da Microsoft) padrão de 64 bytes.
public:
System::Reflection::Emit::ILGenerator ^ GetILGenerator();
public System.Reflection.Emit.ILGenerator GetILGenerator ();
member this.GetILGenerator : unit -> System.Reflection.Emit.ILGenerator
Public Function GetILGenerator () As ILGenerator
Retornos
Retorna um objeto ILGenerator
para este método.
Exceções
O método não deve ter um corpo devido aos seus sinalizadores MethodAttributes ou MethodImplAttributes, por exemplo, porque ele tem o sinalizador PinvokeImpl.
- ou -
O método é genérico, mas não é uma definição de método genérico. Ou seja, a propriedade IsGenericMethod é true
, mas a propriedade IsGenericMethodDefinition é false
.
Exemplos
O exemplo de código abaixo demonstra o uso contextual do GetILGenerator
método, criando e emitindo um assembly dinâmico que calculará o produto de ponto de dois pontos no espaço 3D.
using namespace System;
using namespace System::Threading;
using namespace System::Reflection;
using namespace System::Reflection::Emit;
Type^ DynamicDotProductGen()
{
Type^ ivType = nullptr;
array<Type^>^temp0 = {int::typeid,int::typeid,int::typeid};
array<Type^>^ctorParams = temp0;
AppDomain^ myDomain = Thread::GetDomain();
AssemblyName^ myAsmName = gcnew AssemblyName;
myAsmName->Name = "IntVectorAsm";
AssemblyBuilder^ myAsmBuilder = myDomain->DefineDynamicAssembly( myAsmName, AssemblyBuilderAccess::RunAndSave );
ModuleBuilder^ IntVectorModule = myAsmBuilder->DefineDynamicModule( "IntVectorModule", "Vector.dll" );
TypeBuilder^ ivTypeBld = IntVectorModule->DefineType( "IntVector", TypeAttributes::Public );
FieldBuilder^ xField = ivTypeBld->DefineField( "x", int::typeid, FieldAttributes::Private );
FieldBuilder^ yField = ivTypeBld->DefineField( "y", int::typeid, FieldAttributes::Private );
FieldBuilder^ zField = ivTypeBld->DefineField( "z", int::typeid, FieldAttributes::Private );
Type^ objType = Type::GetType( "System.Object" );
ConstructorInfo^ objCtor = objType->GetConstructor( gcnew array<Type^>(0) );
ConstructorBuilder^ ivCtor = ivTypeBld->DefineConstructor( MethodAttributes::Public, CallingConventions::Standard, ctorParams );
ILGenerator^ ctorIL = ivCtor->GetILGenerator();
ctorIL->Emit( OpCodes::Ldarg_0 );
ctorIL->Emit( OpCodes::Call, objCtor );
ctorIL->Emit( OpCodes::Ldarg_0 );
ctorIL->Emit( OpCodes::Ldarg_1 );
ctorIL->Emit( OpCodes::Stfld, xField );
ctorIL->Emit( OpCodes::Ldarg_0 );
ctorIL->Emit( OpCodes::Ldarg_2 );
ctorIL->Emit( OpCodes::Stfld, yField );
ctorIL->Emit( OpCodes::Ldarg_0 );
ctorIL->Emit( OpCodes::Ldarg_3 );
ctorIL->Emit( OpCodes::Stfld, zField );
ctorIL->Emit( OpCodes::Ret );
// This method will find the dot product of the stored vector
// with another.
array<Type^>^temp1 = {ivTypeBld};
array<Type^>^dpParams = temp1;
// Here, you create a MethodBuilder containing the
// name, the attributes (public, static, private, and so on),
// the return type (int, in this case), and a array of Type
// indicating the type of each parameter. Since the sole parameter
// is a IntVector, the very class you're creating, you will
// pass in the TypeBuilder (which is derived from Type) instead of
// a Type object for IntVector, avoiding an exception.
// -- This method would be declared in C# as:
// public int DotProduct(IntVector aVector)
MethodBuilder^ dotProductMthd = ivTypeBld->DefineMethod( "DotProduct", MethodAttributes::Public, int::typeid, dpParams );
// A ILGenerator can now be spawned, attached to the MethodBuilder.
ILGenerator^ mthdIL = dotProductMthd->GetILGenerator();
// Here's the body of our function, in MSIL form. We're going to find the
// "dot product" of the current vector instance with the passed vector
// instance. For reference purposes, the equation is:
// (x1 * x2) + (y1 * y2) + (z1 * z2) = the dot product
// First, you'll load the reference to the current instance "this"
// stored in argument 0 (ldarg.0) onto the stack. Ldfld, the subsequent
// instruction, will pop the reference off the stack and look up the
// field "x", specified by the FieldInfo token "xField".
mthdIL->Emit( OpCodes::Ldarg_0 );
mthdIL->Emit( OpCodes::Ldfld, xField );
// That completed, the value stored at field "x" is now atop the stack.
// Now, you'll do the same for the Object reference we passed as a
// parameter, stored in argument 1 (ldarg.1). After Ldfld executed,
// you'll have the value stored in field "x" for the passed instance
// atop the stack.
mthdIL->Emit( OpCodes::Ldarg_1 );
mthdIL->Emit( OpCodes::Ldfld, xField );
// There will now be two values atop the stack - the "x" value for the
// current vector instance, and the "x" value for the passed instance.
// You'll now multiply them, and push the result onto the evaluation stack.
mthdIL->Emit( OpCodes::Mul_Ovf_Un );
// Now, repeat this for the "y" fields of both vectors.
mthdIL->Emit( OpCodes::Ldarg_0 );
mthdIL->Emit( OpCodes::Ldfld, yField );
mthdIL->Emit( OpCodes::Ldarg_1 );
mthdIL->Emit( OpCodes::Ldfld, yField );
mthdIL->Emit( OpCodes::Mul_Ovf_Un );
// At this time, the results of both multiplications should be atop
// the stack. You'll now add them and push the result onto the stack.
mthdIL->Emit( OpCodes::Add_Ovf_Un );
// Multiply both "z" field and push the result onto the stack.
mthdIL->Emit( OpCodes::Ldarg_0 );
mthdIL->Emit( OpCodes::Ldfld, zField );
mthdIL->Emit( OpCodes::Ldarg_1 );
mthdIL->Emit( OpCodes::Ldfld, zField );
mthdIL->Emit( OpCodes::Mul_Ovf_Un );
// Finally, add the result of multiplying the "z" fields with the
// result of the earlier addition, and push the result - the dot product -
// onto the stack.
mthdIL->Emit( OpCodes::Add_Ovf_Un );
// The "ret" opcode will pop the last value from the stack and return it
// to the calling method. You're all done!
mthdIL->Emit( OpCodes::Ret );
ivType = ivTypeBld->CreateType();
return ivType;
}
int main()
{
Type^ IVType = nullptr;
Object^ aVector1 = nullptr;
Object^ aVector2 = nullptr;
array<Type^>^temp2 = {int::typeid,int::typeid,int::typeid};
array<Type^>^aVtypes = temp2;
array<Object^>^temp3 = {10,10,10};
array<Object^>^aVargs1 = temp3;
array<Object^>^temp4 = {20,20,20};
array<Object^>^aVargs2 = temp4;
// Call the method to build our dynamic class.
IVType = DynamicDotProductGen();
Console::WriteLine( "---" );
ConstructorInfo^ myDTctor = IVType->GetConstructor( aVtypes );
aVector1 = myDTctor->Invoke( aVargs1 );
aVector2 = myDTctor->Invoke( aVargs2 );
array<Object^>^passMe = gcnew array<Object^>(1);
passMe[ 0 ] = dynamic_cast<Object^>(aVector2);
Console::WriteLine( "(10, 10, 10) . (20, 20, 20) = {0}", IVType->InvokeMember( "DotProduct", BindingFlags::InvokeMethod, nullptr, aVector1, passMe ) );
}
// +++ OUTPUT +++
// ---
// (10, 10, 10) . (20, 20, 20) = 600
using System;
using System.Threading;
using System.Reflection;
using System.Reflection.Emit;
class TestILGenerator
{
public static Type DynamicDotProductGen()
{
Type ivType = null;
Type[] ctorParams = new Type[] { typeof(int),
typeof(int),
typeof(int)};
AppDomain myDomain = Thread.GetDomain();
AssemblyName myAsmName = new AssemblyName();
myAsmName.Name = "IntVectorAsm";
AssemblyBuilder myAsmBuilder = myDomain.DefineDynamicAssembly(
myAsmName,
AssemblyBuilderAccess.RunAndSave);
ModuleBuilder IntVectorModule = myAsmBuilder.DefineDynamicModule("IntVectorModule",
"Vector.dll");
TypeBuilder ivTypeBld = IntVectorModule.DefineType("IntVector",
TypeAttributes.Public);
FieldBuilder xField = ivTypeBld.DefineField("x", typeof(int),
FieldAttributes.Private);
FieldBuilder yField = ivTypeBld.DefineField("y", typeof(int),
FieldAttributes.Private);
FieldBuilder zField = ivTypeBld.DefineField("z", typeof(int),
FieldAttributes.Private);
Type objType = Type.GetType("System.Object");
ConstructorInfo objCtor = objType.GetConstructor(new Type[0]);
ConstructorBuilder ivCtor = ivTypeBld.DefineConstructor(
MethodAttributes.Public,
CallingConventions.Standard,
ctorParams);
ILGenerator ctorIL = ivCtor.GetILGenerator();
ctorIL.Emit(OpCodes.Ldarg_0);
ctorIL.Emit(OpCodes.Call, objCtor);
ctorIL.Emit(OpCodes.Ldarg_0);
ctorIL.Emit(OpCodes.Ldarg_1);
ctorIL.Emit(OpCodes.Stfld, xField);
ctorIL.Emit(OpCodes.Ldarg_0);
ctorIL.Emit(OpCodes.Ldarg_2);
ctorIL.Emit(OpCodes.Stfld, yField);
ctorIL.Emit(OpCodes.Ldarg_0);
ctorIL.Emit(OpCodes.Ldarg_3);
ctorIL.Emit(OpCodes.Stfld, zField);
ctorIL.Emit(OpCodes.Ret);
// This method will find the dot product of the stored vector
// with another.
Type[] dpParams = new Type[] { ivTypeBld };
// Here, you create a MethodBuilder containing the
// name, the attributes (public, static, private, and so on),
// the return type (int, in this case), and a array of Type
// indicating the type of each parameter. Since the sole parameter
// is a IntVector, the very class you're creating, you will
// pass in the TypeBuilder (which is derived from Type) instead of
// a Type object for IntVector, avoiding an exception.
// -- This method would be declared in C# as:
// public int DotProduct(IntVector aVector)
MethodBuilder dotProductMthd = ivTypeBld.DefineMethod(
"DotProduct",
MethodAttributes.Public,
typeof(int),
dpParams);
// A ILGenerator can now be spawned, attached to the MethodBuilder.
ILGenerator mthdIL = dotProductMthd.GetILGenerator();
// Here's the body of our function, in MSIL form. We're going to find the
// "dot product" of the current vector instance with the passed vector
// instance. For reference purposes, the equation is:
// (x1 * x2) + (y1 * y2) + (z1 * z2) = the dot product
// First, you'll load the reference to the current instance "this"
// stored in argument 0 (ldarg.0) onto the stack. Ldfld, the subsequent
// instruction, will pop the reference off the stack and look up the
// field "x", specified by the FieldInfo token "xField".
mthdIL.Emit(OpCodes.Ldarg_0);
mthdIL.Emit(OpCodes.Ldfld, xField);
// That completed, the value stored at field "x" is now atop the stack.
// Now, you'll do the same for the object reference we passed as a
// parameter, stored in argument 1 (ldarg.1). After Ldfld executed,
// you'll have the value stored in field "x" for the passed instance
// atop the stack.
mthdIL.Emit(OpCodes.Ldarg_1);
mthdIL.Emit(OpCodes.Ldfld, xField);
// There will now be two values atop the stack - the "x" value for the
// current vector instance, and the "x" value for the passed instance.
// You'll now multiply them, and push the result onto the evaluation stack.
mthdIL.Emit(OpCodes.Mul_Ovf_Un);
// Now, repeat this for the "y" fields of both vectors.
mthdIL.Emit(OpCodes.Ldarg_0);
mthdIL.Emit(OpCodes.Ldfld, yField);
mthdIL.Emit(OpCodes.Ldarg_1);
mthdIL.Emit(OpCodes.Ldfld, yField);
mthdIL.Emit(OpCodes.Mul_Ovf_Un);
// At this time, the results of both multiplications should be atop
// the stack. You'll now add them and push the result onto the stack.
mthdIL.Emit(OpCodes.Add_Ovf_Un);
// Multiply both "z" field and push the result onto the stack.
mthdIL.Emit(OpCodes.Ldarg_0);
mthdIL.Emit(OpCodes.Ldfld, zField);
mthdIL.Emit(OpCodes.Ldarg_1);
mthdIL.Emit(OpCodes.Ldfld, zField);
mthdIL.Emit(OpCodes.Mul_Ovf_Un);
// Finally, add the result of multiplying the "z" fields with the
// result of the earlier addition, and push the result - the dot product -
// onto the stack.
mthdIL.Emit(OpCodes.Add_Ovf_Un);
// The "ret" opcode will pop the last value from the stack and return it
// to the calling method. You're all done!
mthdIL.Emit(OpCodes.Ret);
ivType = ivTypeBld.CreateType();
return ivType;
}
public static void Main() {
Type IVType = null;
object aVector1 = null;
object aVector2 = null;
Type[] aVtypes = new Type[] {typeof(int), typeof(int), typeof(int)};
object[] aVargs1 = new object[] {10, 10, 10};
object[] aVargs2 = new object[] {20, 20, 20};
// Call the method to build our dynamic class.
IVType = DynamicDotProductGen();
Console.WriteLine("---");
ConstructorInfo myDTctor = IVType.GetConstructor(aVtypes);
aVector1 = myDTctor.Invoke(aVargs1);
aVector2 = myDTctor.Invoke(aVargs2);
object[] passMe = new object[1];
passMe[0] = (object)aVector2;
Console.WriteLine("(10, 10, 10) . (20, 20, 20) = {0}",
IVType.InvokeMember("DotProduct",
BindingFlags.InvokeMethod,
null,
aVector1,
passMe));
// +++ OUTPUT +++
// ---
// (10, 10, 10) . (20, 20, 20) = 600
}
}
Imports System.Threading
Imports System.Reflection
Imports System.Reflection.Emit
_
Class TestILGenerator
Public Shared Function DynamicDotProductGen() As Type
Dim ivType As Type = Nothing
Dim ctorParams() As Type = {GetType(Integer), GetType(Integer), GetType(Integer)}
Dim myDomain As AppDomain = Thread.GetDomain()
Dim myAsmName As New AssemblyName()
myAsmName.Name = "IntVectorAsm"
Dim myAsmBuilder As AssemblyBuilder = myDomain.DefineDynamicAssembly( _
myAsmName, _
AssemblyBuilderAccess.RunAndSave)
Dim IntVectorModule As ModuleBuilder = myAsmBuilder.DefineDynamicModule( _
"IntVectorModule", _
"Vector.dll")
Dim ivTypeBld As TypeBuilder = IntVectorModule.DefineType("IntVector", TypeAttributes.Public)
Dim xField As FieldBuilder = ivTypeBld.DefineField("x", _
GetType(Integer), _
FieldAttributes.Private)
Dim yField As FieldBuilder = ivTypeBld.DefineField("y", _
GetType(Integer), _
FieldAttributes.Private)
Dim zField As FieldBuilder = ivTypeBld.DefineField("z", _
GetType(Integer), _
FieldAttributes.Private)
Dim objType As Type = Type.GetType("System.Object")
Dim objCtor As ConstructorInfo = objType.GetConstructor(New Type() {})
Dim ivCtor As ConstructorBuilder = ivTypeBld.DefineConstructor( _
MethodAttributes.Public, _
CallingConventions.Standard, _
ctorParams)
Dim ctorIL As ILGenerator = ivCtor.GetILGenerator()
ctorIL.Emit(OpCodes.Ldarg_0)
ctorIL.Emit(OpCodes.Call, objCtor)
ctorIL.Emit(OpCodes.Ldarg_0)
ctorIL.Emit(OpCodes.Ldarg_1)
ctorIL.Emit(OpCodes.Stfld, xField)
ctorIL.Emit(OpCodes.Ldarg_0)
ctorIL.Emit(OpCodes.Ldarg_2)
ctorIL.Emit(OpCodes.Stfld, yField)
ctorIL.Emit(OpCodes.Ldarg_0)
ctorIL.Emit(OpCodes.Ldarg_3)
ctorIL.Emit(OpCodes.Stfld, zField)
ctorIL.Emit(OpCodes.Ret)
' Now, you'll construct the method find the dot product of two vectors. First,
' let's define the parameters that will be accepted by the method. In this case,
' it's an IntVector itself!
Dim dpParams() As Type = {ivTypeBld}
' Here, you create a MethodBuilder containing the
' name, the attributes (public, static, private, and so on),
' the return type (int, in this case), and a array of Type
' indicating the type of each parameter. Since the sole parameter
' is a IntVector, the very class you're creating, you will
' pass in the TypeBuilder (which is derived from Type) instead of
' a Type object for IntVector, avoiding an exception.
' -- This method would be declared in VB.NET as:
' Public Function DotProduct(IntVector aVector) As Integer
Dim dotProductMthd As MethodBuilder = ivTypeBld.DefineMethod("DotProduct", _
MethodAttributes.Public, GetType(Integer), _
dpParams)
' A ILGenerator can now be spawned, attached to the MethodBuilder.
Dim mthdIL As ILGenerator = dotProductMthd.GetILGenerator()
' Here's the body of our function, in MSIL form. We're going to find the
' "dot product" of the current vector instance with the passed vector
' instance. For reference purposes, the equation is:
' (x1 * x2) + (y1 * y2) + (z1 * z2) = the dot product
' First, you'll load the reference to the current instance "this"
' stored in argument 0 (ldarg.0) onto the stack. Ldfld, the subsequent
' instruction, will pop the reference off the stack and look up the
' field "x", specified by the FieldInfo token "xField".
mthdIL.Emit(OpCodes.Ldarg_0)
mthdIL.Emit(OpCodes.Ldfld, xField)
' That completed, the value stored at field "x" is now atop the stack.
' Now, you'll do the same for the object reference we passed as a
' parameter, stored in argument 1 (ldarg.1). After Ldfld executed,
' you'll have the value stored in field "x" for the passed instance
' atop the stack.
mthdIL.Emit(OpCodes.Ldarg_1)
mthdIL.Emit(OpCodes.Ldfld, xField)
' There will now be two values atop the stack - the "x" value for the
' current vector instance, and the "x" value for the passed instance.
' You'll now multiply them, and push the result onto the evaluation stack.
mthdIL.Emit(OpCodes.Mul_Ovf_Un)
' Now, repeat this for the "y" fields of both vectors.
mthdIL.Emit(OpCodes.Ldarg_0)
mthdIL.Emit(OpCodes.Ldfld, yField)
mthdIL.Emit(OpCodes.Ldarg_1)
mthdIL.Emit(OpCodes.Ldfld, yField)
mthdIL.Emit(OpCodes.Mul_Ovf_Un)
' At this time, the results of both multiplications should be atop
' the stack. You'll now add them and push the result onto the stack.
mthdIL.Emit(OpCodes.Add_Ovf_Un)
' Multiply both "z" field and push the result onto the stack.
mthdIL.Emit(OpCodes.Ldarg_0)
mthdIL.Emit(OpCodes.Ldfld, zField)
mthdIL.Emit(OpCodes.Ldarg_1)
mthdIL.Emit(OpCodes.Ldfld, zField)
mthdIL.Emit(OpCodes.Mul_Ovf_Un)
' Finally, add the result of multiplying the "z" fields with the
' result of the earlier addition, and push the result - the dot product -
' onto the stack.
mthdIL.Emit(OpCodes.Add_Ovf_Un)
' The "ret" opcode will pop the last value from the stack and return it
' to the calling method. You're all done!
mthdIL.Emit(OpCodes.Ret)
ivType = ivTypeBld.CreateType()
Return ivType
End Function 'DynamicDotProductGen
Public Shared Sub Main()
Dim IVType As Type = Nothing
Dim aVector1 As Object = Nothing
Dim aVector2 As Object = Nothing
Dim aVtypes() As Type = {GetType(Integer), GetType(Integer), GetType(Integer)}
Dim aVargs1() As Object = {10, 10, 10}
Dim aVargs2() As Object = {20, 20, 20}
' Call the method to build our dynamic class.
IVType = DynamicDotProductGen()
Dim myDTctor As ConstructorInfo = IVType.GetConstructor(aVtypes)
aVector1 = myDTctor.Invoke(aVargs1)
aVector2 = myDTctor.Invoke(aVargs2)
Console.WriteLine("---")
Dim passMe(0) As Object
passMe(0) = CType(aVector2, Object)
Console.WriteLine("(10, 10, 10) . (20, 20, 20) = {0}", _
IVType.InvokeMember("DotProduct", BindingFlags.InvokeMethod, _
Nothing, aVector1, passMe))
End Sub
End Class
' +++ OUTPUT +++
' ---
' (10, 10, 10) . (20, 20, 20) = 600
Aplica-se a
.NET 9 e outras versões
Produto | Versões |
---|---|
.NET | Core 1.0, Core 1.1, Core 2.0, Core 2.1, Core 2.2, Core 3.0, Core 3.1, 5, 6, 7, 8, 9 |
.NET Framework | 1.1, 2.0, 3.0, 3.5, 4.0, 4.5, 4.5.1, 4.5.2, 4.6, 4.6.1, 4.6.2, 4.7, 4.7.1, 4.7.2, 4.8, 4.8.1 |
.NET Standard | 2.0, 2.1 |
- Origem:
- MethodBuilder.cs
- Origem:
- MethodBuilder.cs
- Origem:
- MethodBuilder.cs
Retorna um ILGenerator
pra esse método com o tamanho de fluxo MSIL (linguagem intermediária da Microsoft) especificado.
public:
System::Reflection::Emit::ILGenerator ^ GetILGenerator(int size);
public System.Reflection.Emit.ILGenerator GetILGenerator (int size);
member this.GetILGenerator : int -> System.Reflection.Emit.ILGenerator
Public Function GetILGenerator (size As Integer) As ILGenerator
Parâmetros
- size
- Int32
O tamanho do fluxo MSIL, em bytes.
Retornos
Retorna um objeto ILGenerator
para este método.
Exceções
O método não deve ter um corpo devido aos seus sinalizadores MethodAttributes ou MethodImplAttributes, por exemplo, porque ele tem o sinalizador PinvokeImpl.
- ou -
O método é genérico, mas não é uma definição de método genérico. Ou seja, a propriedade IsGenericMethod é true
, mas a propriedade IsGenericMethodDefinition é false
.
Exemplos
O exemplo de código abaixo demonstra o uso contextual do GetILGenerator
método, criando e emitindo um assembly dinâmico que calculará o produto de ponto de dois pontos no espaço 3D.
using namespace System;
using namespace System::Threading;
using namespace System::Reflection;
using namespace System::Reflection::Emit;
Type^ DynamicDotProductGen()
{
Type^ ivType = nullptr;
array<Type^>^temp0 = {int::typeid,int::typeid,int::typeid};
array<Type^>^ctorParams = temp0;
AppDomain^ myDomain = Thread::GetDomain();
AssemblyName^ myAsmName = gcnew AssemblyName;
myAsmName->Name = "IntVectorAsm";
AssemblyBuilder^ myAsmBuilder = myDomain->DefineDynamicAssembly( myAsmName, AssemblyBuilderAccess::RunAndSave );
ModuleBuilder^ IntVectorModule = myAsmBuilder->DefineDynamicModule( "IntVectorModule", "Vector.dll" );
TypeBuilder^ ivTypeBld = IntVectorModule->DefineType( "IntVector", TypeAttributes::Public );
FieldBuilder^ xField = ivTypeBld->DefineField( "x", int::typeid, FieldAttributes::Private );
FieldBuilder^ yField = ivTypeBld->DefineField( "y", int::typeid, FieldAttributes::Private );
FieldBuilder^ zField = ivTypeBld->DefineField( "z", int::typeid, FieldAttributes::Private );
Type^ objType = Type::GetType( "System.Object" );
ConstructorInfo^ objCtor = objType->GetConstructor( gcnew array<Type^>(0) );
ConstructorBuilder^ ivCtor = ivTypeBld->DefineConstructor( MethodAttributes::Public, CallingConventions::Standard, ctorParams );
ILGenerator^ ctorIL = ivCtor->GetILGenerator();
ctorIL->Emit( OpCodes::Ldarg_0 );
ctorIL->Emit( OpCodes::Call, objCtor );
ctorIL->Emit( OpCodes::Ldarg_0 );
ctorIL->Emit( OpCodes::Ldarg_1 );
ctorIL->Emit( OpCodes::Stfld, xField );
ctorIL->Emit( OpCodes::Ldarg_0 );
ctorIL->Emit( OpCodes::Ldarg_2 );
ctorIL->Emit( OpCodes::Stfld, yField );
ctorIL->Emit( OpCodes::Ldarg_0 );
ctorIL->Emit( OpCodes::Ldarg_3 );
ctorIL->Emit( OpCodes::Stfld, zField );
ctorIL->Emit( OpCodes::Ret );
// This method will find the dot product of the stored vector
// with another.
array<Type^>^temp1 = {ivTypeBld};
array<Type^>^dpParams = temp1;
// Here, you create a MethodBuilder containing the
// name, the attributes (public, static, private, and so on),
// the return type (int, in this case), and a array of Type
// indicating the type of each parameter. Since the sole parameter
// is a IntVector, the very class you're creating, you will
// pass in the TypeBuilder (which is derived from Type) instead of
// a Type object for IntVector, avoiding an exception.
// -- This method would be declared in C# as:
// public int DotProduct(IntVector aVector)
MethodBuilder^ dotProductMthd = ivTypeBld->DefineMethod( "DotProduct", MethodAttributes::Public, int::typeid, dpParams );
// A ILGenerator can now be spawned, attached to the MethodBuilder.
ILGenerator^ mthdIL = dotProductMthd->GetILGenerator();
// Here's the body of our function, in MSIL form. We're going to find the
// "dot product" of the current vector instance with the passed vector
// instance. For reference purposes, the equation is:
// (x1 * x2) + (y1 * y2) + (z1 * z2) = the dot product
// First, you'll load the reference to the current instance "this"
// stored in argument 0 (ldarg.0) onto the stack. Ldfld, the subsequent
// instruction, will pop the reference off the stack and look up the
// field "x", specified by the FieldInfo token "xField".
mthdIL->Emit( OpCodes::Ldarg_0 );
mthdIL->Emit( OpCodes::Ldfld, xField );
// That completed, the value stored at field "x" is now atop the stack.
// Now, you'll do the same for the Object reference we passed as a
// parameter, stored in argument 1 (ldarg.1). After Ldfld executed,
// you'll have the value stored in field "x" for the passed instance
// atop the stack.
mthdIL->Emit( OpCodes::Ldarg_1 );
mthdIL->Emit( OpCodes::Ldfld, xField );
// There will now be two values atop the stack - the "x" value for the
// current vector instance, and the "x" value for the passed instance.
// You'll now multiply them, and push the result onto the evaluation stack.
mthdIL->Emit( OpCodes::Mul_Ovf_Un );
// Now, repeat this for the "y" fields of both vectors.
mthdIL->Emit( OpCodes::Ldarg_0 );
mthdIL->Emit( OpCodes::Ldfld, yField );
mthdIL->Emit( OpCodes::Ldarg_1 );
mthdIL->Emit( OpCodes::Ldfld, yField );
mthdIL->Emit( OpCodes::Mul_Ovf_Un );
// At this time, the results of both multiplications should be atop
// the stack. You'll now add them and push the result onto the stack.
mthdIL->Emit( OpCodes::Add_Ovf_Un );
// Multiply both "z" field and push the result onto the stack.
mthdIL->Emit( OpCodes::Ldarg_0 );
mthdIL->Emit( OpCodes::Ldfld, zField );
mthdIL->Emit( OpCodes::Ldarg_1 );
mthdIL->Emit( OpCodes::Ldfld, zField );
mthdIL->Emit( OpCodes::Mul_Ovf_Un );
// Finally, add the result of multiplying the "z" fields with the
// result of the earlier addition, and push the result - the dot product -
// onto the stack.
mthdIL->Emit( OpCodes::Add_Ovf_Un );
// The "ret" opcode will pop the last value from the stack and return it
// to the calling method. You're all done!
mthdIL->Emit( OpCodes::Ret );
ivType = ivTypeBld->CreateType();
return ivType;
}
int main()
{
Type^ IVType = nullptr;
Object^ aVector1 = nullptr;
Object^ aVector2 = nullptr;
array<Type^>^temp2 = {int::typeid,int::typeid,int::typeid};
array<Type^>^aVtypes = temp2;
array<Object^>^temp3 = {10,10,10};
array<Object^>^aVargs1 = temp3;
array<Object^>^temp4 = {20,20,20};
array<Object^>^aVargs2 = temp4;
// Call the method to build our dynamic class.
IVType = DynamicDotProductGen();
Console::WriteLine( "---" );
ConstructorInfo^ myDTctor = IVType->GetConstructor( aVtypes );
aVector1 = myDTctor->Invoke( aVargs1 );
aVector2 = myDTctor->Invoke( aVargs2 );
array<Object^>^passMe = gcnew array<Object^>(1);
passMe[ 0 ] = dynamic_cast<Object^>(aVector2);
Console::WriteLine( "(10, 10, 10) . (20, 20, 20) = {0}", IVType->InvokeMember( "DotProduct", BindingFlags::InvokeMethod, nullptr, aVector1, passMe ) );
}
// +++ OUTPUT +++
// ---
// (10, 10, 10) . (20, 20, 20) = 600
using System;
using System.Threading;
using System.Reflection;
using System.Reflection.Emit;
class TestILGenerator
{
public static Type DynamicDotProductGen()
{
Type ivType = null;
Type[] ctorParams = new Type[] { typeof(int),
typeof(int),
typeof(int)};
AppDomain myDomain = Thread.GetDomain();
AssemblyName myAsmName = new AssemblyName();
myAsmName.Name = "IntVectorAsm";
AssemblyBuilder myAsmBuilder = myDomain.DefineDynamicAssembly(
myAsmName,
AssemblyBuilderAccess.RunAndSave);
ModuleBuilder IntVectorModule = myAsmBuilder.DefineDynamicModule("IntVectorModule",
"Vector.dll");
TypeBuilder ivTypeBld = IntVectorModule.DefineType("IntVector",
TypeAttributes.Public);
FieldBuilder xField = ivTypeBld.DefineField("x", typeof(int),
FieldAttributes.Private);
FieldBuilder yField = ivTypeBld.DefineField("y", typeof(int),
FieldAttributes.Private);
FieldBuilder zField = ivTypeBld.DefineField("z", typeof(int),
FieldAttributes.Private);
Type objType = Type.GetType("System.Object");
ConstructorInfo objCtor = objType.GetConstructor(new Type[0]);
ConstructorBuilder ivCtor = ivTypeBld.DefineConstructor(
MethodAttributes.Public,
CallingConventions.Standard,
ctorParams);
ILGenerator ctorIL = ivCtor.GetILGenerator();
ctorIL.Emit(OpCodes.Ldarg_0);
ctorIL.Emit(OpCodes.Call, objCtor);
ctorIL.Emit(OpCodes.Ldarg_0);
ctorIL.Emit(OpCodes.Ldarg_1);
ctorIL.Emit(OpCodes.Stfld, xField);
ctorIL.Emit(OpCodes.Ldarg_0);
ctorIL.Emit(OpCodes.Ldarg_2);
ctorIL.Emit(OpCodes.Stfld, yField);
ctorIL.Emit(OpCodes.Ldarg_0);
ctorIL.Emit(OpCodes.Ldarg_3);
ctorIL.Emit(OpCodes.Stfld, zField);
ctorIL.Emit(OpCodes.Ret);
// This method will find the dot product of the stored vector
// with another.
Type[] dpParams = new Type[] { ivTypeBld };
// Here, you create a MethodBuilder containing the
// name, the attributes (public, static, private, and so on),
// the return type (int, in this case), and a array of Type
// indicating the type of each parameter. Since the sole parameter
// is a IntVector, the very class you're creating, you will
// pass in the TypeBuilder (which is derived from Type) instead of
// a Type object for IntVector, avoiding an exception.
// -- This method would be declared in C# as:
// public int DotProduct(IntVector aVector)
MethodBuilder dotProductMthd = ivTypeBld.DefineMethod(
"DotProduct",
MethodAttributes.Public,
typeof(int),
dpParams);
// A ILGenerator can now be spawned, attached to the MethodBuilder.
ILGenerator mthdIL = dotProductMthd.GetILGenerator();
// Here's the body of our function, in MSIL form. We're going to find the
// "dot product" of the current vector instance with the passed vector
// instance. For reference purposes, the equation is:
// (x1 * x2) + (y1 * y2) + (z1 * z2) = the dot product
// First, you'll load the reference to the current instance "this"
// stored in argument 0 (ldarg.0) onto the stack. Ldfld, the subsequent
// instruction, will pop the reference off the stack and look up the
// field "x", specified by the FieldInfo token "xField".
mthdIL.Emit(OpCodes.Ldarg_0);
mthdIL.Emit(OpCodes.Ldfld, xField);
// That completed, the value stored at field "x" is now atop the stack.
// Now, you'll do the same for the object reference we passed as a
// parameter, stored in argument 1 (ldarg.1). After Ldfld executed,
// you'll have the value stored in field "x" for the passed instance
// atop the stack.
mthdIL.Emit(OpCodes.Ldarg_1);
mthdIL.Emit(OpCodes.Ldfld, xField);
// There will now be two values atop the stack - the "x" value for the
// current vector instance, and the "x" value for the passed instance.
// You'll now multiply them, and push the result onto the evaluation stack.
mthdIL.Emit(OpCodes.Mul_Ovf_Un);
// Now, repeat this for the "y" fields of both vectors.
mthdIL.Emit(OpCodes.Ldarg_0);
mthdIL.Emit(OpCodes.Ldfld, yField);
mthdIL.Emit(OpCodes.Ldarg_1);
mthdIL.Emit(OpCodes.Ldfld, yField);
mthdIL.Emit(OpCodes.Mul_Ovf_Un);
// At this time, the results of both multiplications should be atop
// the stack. You'll now add them and push the result onto the stack.
mthdIL.Emit(OpCodes.Add_Ovf_Un);
// Multiply both "z" field and push the result onto the stack.
mthdIL.Emit(OpCodes.Ldarg_0);
mthdIL.Emit(OpCodes.Ldfld, zField);
mthdIL.Emit(OpCodes.Ldarg_1);
mthdIL.Emit(OpCodes.Ldfld, zField);
mthdIL.Emit(OpCodes.Mul_Ovf_Un);
// Finally, add the result of multiplying the "z" fields with the
// result of the earlier addition, and push the result - the dot product -
// onto the stack.
mthdIL.Emit(OpCodes.Add_Ovf_Un);
// The "ret" opcode will pop the last value from the stack and return it
// to the calling method. You're all done!
mthdIL.Emit(OpCodes.Ret);
ivType = ivTypeBld.CreateType();
return ivType;
}
public static void Main() {
Type IVType = null;
object aVector1 = null;
object aVector2 = null;
Type[] aVtypes = new Type[] {typeof(int), typeof(int), typeof(int)};
object[] aVargs1 = new object[] {10, 10, 10};
object[] aVargs2 = new object[] {20, 20, 20};
// Call the method to build our dynamic class.
IVType = DynamicDotProductGen();
Console.WriteLine("---");
ConstructorInfo myDTctor = IVType.GetConstructor(aVtypes);
aVector1 = myDTctor.Invoke(aVargs1);
aVector2 = myDTctor.Invoke(aVargs2);
object[] passMe = new object[1];
passMe[0] = (object)aVector2;
Console.WriteLine("(10, 10, 10) . (20, 20, 20) = {0}",
IVType.InvokeMember("DotProduct",
BindingFlags.InvokeMethod,
null,
aVector1,
passMe));
// +++ OUTPUT +++
// ---
// (10, 10, 10) . (20, 20, 20) = 600
}
}
Imports System.Threading
Imports System.Reflection
Imports System.Reflection.Emit
_
Class TestILGenerator
Public Shared Function DynamicDotProductGen() As Type
Dim ivType As Type = Nothing
Dim ctorParams() As Type = {GetType(Integer), GetType(Integer), GetType(Integer)}
Dim myDomain As AppDomain = Thread.GetDomain()
Dim myAsmName As New AssemblyName()
myAsmName.Name = "IntVectorAsm"
Dim myAsmBuilder As AssemblyBuilder = myDomain.DefineDynamicAssembly( _
myAsmName, _
AssemblyBuilderAccess.RunAndSave)
Dim IntVectorModule As ModuleBuilder = myAsmBuilder.DefineDynamicModule( _
"IntVectorModule", _
"Vector.dll")
Dim ivTypeBld As TypeBuilder = IntVectorModule.DefineType("IntVector", TypeAttributes.Public)
Dim xField As FieldBuilder = ivTypeBld.DefineField("x", _
GetType(Integer), _
FieldAttributes.Private)
Dim yField As FieldBuilder = ivTypeBld.DefineField("y", _
GetType(Integer), _
FieldAttributes.Private)
Dim zField As FieldBuilder = ivTypeBld.DefineField("z", _
GetType(Integer), _
FieldAttributes.Private)
Dim objType As Type = Type.GetType("System.Object")
Dim objCtor As ConstructorInfo = objType.GetConstructor(New Type() {})
Dim ivCtor As ConstructorBuilder = ivTypeBld.DefineConstructor( _
MethodAttributes.Public, _
CallingConventions.Standard, _
ctorParams)
Dim ctorIL As ILGenerator = ivCtor.GetILGenerator()
ctorIL.Emit(OpCodes.Ldarg_0)
ctorIL.Emit(OpCodes.Call, objCtor)
ctorIL.Emit(OpCodes.Ldarg_0)
ctorIL.Emit(OpCodes.Ldarg_1)
ctorIL.Emit(OpCodes.Stfld, xField)
ctorIL.Emit(OpCodes.Ldarg_0)
ctorIL.Emit(OpCodes.Ldarg_2)
ctorIL.Emit(OpCodes.Stfld, yField)
ctorIL.Emit(OpCodes.Ldarg_0)
ctorIL.Emit(OpCodes.Ldarg_3)
ctorIL.Emit(OpCodes.Stfld, zField)
ctorIL.Emit(OpCodes.Ret)
' Now, you'll construct the method find the dot product of two vectors. First,
' let's define the parameters that will be accepted by the method. In this case,
' it's an IntVector itself!
Dim dpParams() As Type = {ivTypeBld}
' Here, you create a MethodBuilder containing the
' name, the attributes (public, static, private, and so on),
' the return type (int, in this case), and a array of Type
' indicating the type of each parameter. Since the sole parameter
' is a IntVector, the very class you're creating, you will
' pass in the TypeBuilder (which is derived from Type) instead of
' a Type object for IntVector, avoiding an exception.
' -- This method would be declared in VB.NET as:
' Public Function DotProduct(IntVector aVector) As Integer
Dim dotProductMthd As MethodBuilder = ivTypeBld.DefineMethod("DotProduct", _
MethodAttributes.Public, GetType(Integer), _
dpParams)
' A ILGenerator can now be spawned, attached to the MethodBuilder.
Dim mthdIL As ILGenerator = dotProductMthd.GetILGenerator()
' Here's the body of our function, in MSIL form. We're going to find the
' "dot product" of the current vector instance with the passed vector
' instance. For reference purposes, the equation is:
' (x1 * x2) + (y1 * y2) + (z1 * z2) = the dot product
' First, you'll load the reference to the current instance "this"
' stored in argument 0 (ldarg.0) onto the stack. Ldfld, the subsequent
' instruction, will pop the reference off the stack and look up the
' field "x", specified by the FieldInfo token "xField".
mthdIL.Emit(OpCodes.Ldarg_0)
mthdIL.Emit(OpCodes.Ldfld, xField)
' That completed, the value stored at field "x" is now atop the stack.
' Now, you'll do the same for the object reference we passed as a
' parameter, stored in argument 1 (ldarg.1). After Ldfld executed,
' you'll have the value stored in field "x" for the passed instance
' atop the stack.
mthdIL.Emit(OpCodes.Ldarg_1)
mthdIL.Emit(OpCodes.Ldfld, xField)
' There will now be two values atop the stack - the "x" value for the
' current vector instance, and the "x" value for the passed instance.
' You'll now multiply them, and push the result onto the evaluation stack.
mthdIL.Emit(OpCodes.Mul_Ovf_Un)
' Now, repeat this for the "y" fields of both vectors.
mthdIL.Emit(OpCodes.Ldarg_0)
mthdIL.Emit(OpCodes.Ldfld, yField)
mthdIL.Emit(OpCodes.Ldarg_1)
mthdIL.Emit(OpCodes.Ldfld, yField)
mthdIL.Emit(OpCodes.Mul_Ovf_Un)
' At this time, the results of both multiplications should be atop
' the stack. You'll now add them and push the result onto the stack.
mthdIL.Emit(OpCodes.Add_Ovf_Un)
' Multiply both "z" field and push the result onto the stack.
mthdIL.Emit(OpCodes.Ldarg_0)
mthdIL.Emit(OpCodes.Ldfld, zField)
mthdIL.Emit(OpCodes.Ldarg_1)
mthdIL.Emit(OpCodes.Ldfld, zField)
mthdIL.Emit(OpCodes.Mul_Ovf_Un)
' Finally, add the result of multiplying the "z" fields with the
' result of the earlier addition, and push the result - the dot product -
' onto the stack.
mthdIL.Emit(OpCodes.Add_Ovf_Un)
' The "ret" opcode will pop the last value from the stack and return it
' to the calling method. You're all done!
mthdIL.Emit(OpCodes.Ret)
ivType = ivTypeBld.CreateType()
Return ivType
End Function 'DynamicDotProductGen
Public Shared Sub Main()
Dim IVType As Type = Nothing
Dim aVector1 As Object = Nothing
Dim aVector2 As Object = Nothing
Dim aVtypes() As Type = {GetType(Integer), GetType(Integer), GetType(Integer)}
Dim aVargs1() As Object = {10, 10, 10}
Dim aVargs2() As Object = {20, 20, 20}
' Call the method to build our dynamic class.
IVType = DynamicDotProductGen()
Dim myDTctor As ConstructorInfo = IVType.GetConstructor(aVtypes)
aVector1 = myDTctor.Invoke(aVargs1)
aVector2 = myDTctor.Invoke(aVargs2)
Console.WriteLine("---")
Dim passMe(0) As Object
passMe(0) = CType(aVector2, Object)
Console.WriteLine("(10, 10, 10) . (20, 20, 20) = {0}", _
IVType.InvokeMember("DotProduct", BindingFlags.InvokeMethod, _
Nothing, aVector1, passMe))
End Sub
End Class
' +++ OUTPUT +++
' ---
' (10, 10, 10) . (20, 20, 20) = 600
Aplica-se a
.NET 9 e outras versões
Produto | Versões |
---|---|
.NET | Core 1.0, Core 1.1, Core 2.0, Core 2.1, Core 2.2, Core 3.0, Core 3.1, 5, 6, 7, 8, 9 |
.NET Framework | 1.1, 2.0, 3.0, 3.5, 4.0, 4.5, 4.5.1, 4.5.2, 4.6, 4.6.1, 4.6.2, 4.7, 4.7.1, 4.7.2, 4.8, 4.8.1 |
.NET Standard | 2.0, 2.1 |
Comentários do .NET
O .NET é um projeto código aberto. Selecione um link para fornecer comentários: