Attributes in .Net

Attributes are a mechanism for adding metadata, such as compiler instructions and other data about your data, methods, and classes, to the program itself. Attributes are inserted into the metadata and are visible through ILDasm and other metadata-reading tools.

Reflection is the process by which a program can read its own metadata. A program is said to reflect on itself, extracting metadata from its assembly and using that metadata either to inform the user or to modify its own behavior.

An attribute is an object that represents data you want to associate with an element in your program. The element to which you attach an attribute is referred to as the target of that attribute.

Using Attributes
Attributes can be placed on most any declaration (though a specific attribute might restrict the types of declarations on which it is valid). Syntactically, an attribute is specified by placing the name of the attribute, enclosed in square brackets, in front of the declaration of the entity to which it applies. For example, a class with the attribute DllImport is declared like this:
[DllImport] public class MyDllimportClass { ... }

Many attributes have parameters, which can be either positional (unnamed) or named.
Any positional parameters must be specified in a certain order and cannot be omitted; named parameters are optional and can be specified in any order. Positional parameters are specified first. For example, these three attributes are equivalent:
[DllImport("user32.dll", SetLastError=false, ExactSpelling=false)]
[DllImport("user32.dll", ExactSpelling=false, SetLastError=false)]
[DllImport("user32.dll")]

The first parameter, the DLL name, is positional and always comes first; the others are named. In this case, both named parameters default to false, so they can be omitted (refer to the individual attribute's documentation for information on default parameter values).

More than one attribute can be placed on a declaration, either separately or within the same set of brackets:

bool AMethod([In][Out]ref double x);
bool AMethod([Out][In]ref double x);
bool AMethod([In,Out]ref double x);

Creating Custom Attributes

You can create your own custom attributes by defining an attribute class, a class that derives directly or indirectly from System.Attribute (which makes identifying attribute definitions in metadata fast and easy). Suppose you want to tag classes and structs with the name of the programmer who wrote the class or struct. You might define a custom Author attribute class:

using System;
[AttributeUsage(AttributeTargets.Class|AttributeTargets.Struct)]
public class Author : Attribute
{
public Author(string name) { this.name = name; version = 1.0; }
public double version;
string name;
}


The class name is the attribute's name, Author. It is derived from System.Attribute, so it is a custom attribute class. The constructor's parameters are the custom attribute's positional parameters (in this case, name), and any public read-write fields or properties are named parameters (in this case, version is the only named parameter). Note the use of the AttributeUsage attribute to make the Author attribute valid only on class and struct declarations.