Structure of an assembly

The structure of an assembly

Assemblies contain code that is executed by the Common Language Runtime. The great thing about assemblies is that they are self-describing. All the details about the assembly are stored within the assembly itself. In the Windows DNA world, COM stored all its self-describing
data within the server’s registry, and so installing (as well as uninstalling) COM components meant shutting down IIS. Because .a NET assembly stores this information within itself, it makes XCOPY functionality possible. Installing an assembly is as simple as copying it, and there is no need to stop or start IIS while this is going on.

Assemblies are made up of the following parts:

• The assembly manifest
• Type metadata
• Microsoft Intermediate Language (MSIL) code

The assembly manifest is where the details of the assembly are stored. The assembly is stored within the DLL or EXE itself. Assemblies can either be single or multifile assemblies and, therefore, assembly manifests can either be stored in the assembly or as a separate file. The assembly manifest also stores the version number of the assembly to ensure that the application always uses the correct version. When you are going to have multiple versions of an assembly on the same machine, it is important to label them carefully so that the CLR Common Language Runtime knows which one to use. Version numbers in assemblies are constructed in the following manner:

...

Type metadata was explained earlier in this chapter as data about data. This metadata contains
information on the types that are exposed by the assembly such as security permission information, class and interface information, and other assembly information.