1.1 What is .NET?
.NET is a general-purpose software development platform, similar to Java. At its core is a virtual machine that turns intermediate language (IL) into machine code. High-level language compilers for C#, VB.NET and C++ are provided to turn source code into IL. C# is a new programming language, very similar to Java. An extensive class library is included, featuring all the functionality one might expect from a contempory development platform - windows GUI development (Windows Form s), database access (ADO.NET), web development (ASP.NET), web services, XML etc.
1.2 When was .NET announced?
Bill Gates delivered a keynote at Forum 2000, held June 22, 2000, outlining the .NET 'vision'. The July 2000 PDC had a number of sessions on .NET technology, and delegates were given CDs containing a pre-release version of the .NET framework/SDK and Visual Studio.NET.
1.3 What versions of .NET are there?
The final version of the 1.0 SDK and runtime was made publicly available around 6pm PST on 15-Jan-2002. At the same time, the final version of Visual Studio.NET was made available to MSDN subscribers.
.NET 1.1 was released in April 2003 - it's mostly bug fixes for 1.0.
.NET 2.0 is expected in 2005.
1.4 What operating systems does the .NET Framework run on?
The runtime supports Windows Server 2003, Windows XP, Windows 2000, NT4 SP6a and Windows ME/98. Windows 95 is not supported. Some parts of the framework do not work on all platforms - for example, ASP.NET is only supported on XP and Windows 2000/2003. Windows 98/ME cannot be used for development. IIS is not supported on Windows XP Home Edition, and so cannot be used to host ASP.NET. However, the ASP.NET Web Matrix web server does run on XP Home.
The .NET Compact Framework is a version of the .NET Framework for mobile devices, running Windows CE or Windows Mobile.
The Mono project has a version of the .NET Framework that runs on
Linux.
1.5 What tools can I use to develop .NET applications?
There are a number of tools, described here in ascending order of cost:
The .NET Framework SDK is free and includes command-line compilers for C++, C#, and VB.NET and various other utilities to aid development.
ASP.NET Web Matrix is a free ASP.NET development environment from
Microsoft. As well as a GUI development environment, the download includes a simple web server that can be used instead of IIS to host ASP.NET apps. This opens up ASP.NET development to users of Windows XP Home Edition, which cannot run IIS.
Microsoft Visual C# .NET Standard 2003 is a cheap (around $100) version of Visual Studio limited to one language and also with limited wizard support. For example, there's no wizard support for class libraries or custom UI controls. Useful for beginners to learn with, or for savvy developers who can work around the deficiencies in the supplied wizards. As well as C#, there are VB.NET and C++ versions. Microsoft Visual Studio.NET Professional 2003. If you have a license for Visual Studio 6.0, you can get the upgrade. You can also upgrade from VS.NET 2002 for a token $30. Visual Studio.NET includes support for all the MS languages (C#, C++, VB.NET) and has extensive wizard support.
At the top end of the price spectrum are the Visual Studio.NET 2003
Enterprise and Enterprise Architect editions. These offer extra features such as Visual Sourcesafe (version control), and performance and analysis tools. Check out the Visual Studio.NET Feature Comparison at http://msdn.microsoft.com/vstudio/howtobuy/choosing.asp
Terminology
2.1 What is the CLI? Is it the same as the CLR?
The CLI (Common Language Infrastructure) is the definition of the fundamentals of the .NET framework - the Common Type System (CTS), metadata, the Virtual Execution Environment (VES) and its use of intermediate language (IL), and the support of multiple programming languages via the Common Language Specification (CLS). The CLI is documented through ECMA - see http://msdn.microsoft.com/net/ecma/ for more details.
The CLR (Common Language Runtime) is Microsoft's primary implementation of the CLI. Microsoft also have a shared source implementation known as ROTOR, for educational purposes, as well as the .NET Compact Framework for mobile devices. Non-Microsoft CLI implementations include Mono and DotGNU Portable. NET.
2.2 What is the CTS, and how does it relate to the CLS?
CTS = Common Type System. This is the full range of types that the .NET runtime understands. Not all .NET languages support all the types in the CTS.
CLS = Common Language Specification. This is a subset of the CTS which all
.NET languages are expected to support. The idea is that any program which uses CLS-compliant types can interoperate with any .NET program written in any language. This interop is very fine-grained - for example a VB.NET class can inherit from a C# class.
2.3 What is IL?
IL = Intermediate Language. Also known as MSIL (Microsoft Intermediate Language) or CIL (Common Intermediate Language). All .NET source code (of any language) is compiled to IL during development. The IL is then converted to machine code at the point where the software is installed, or (more commonly) at run-time by a Just-In-Time (JIT) compiler.
2.4 What is C#?
C# is a new language designed by Microsoft to work with the .NET framework. In their "Introduction to C#" whitepaper, Microsoft describe C# as follows:
"C# is a simple, modern, object oriented, and type-safe programming language derived from C and C++. C# (pronounced “C sharp”) is firmly planted in the C and C++ family tree of languages, and will immediately be familiar to C and C++ programmers. C# aims to combine the high productivity of Visual Basic and the raw power of C++."
Substitute 'Java' for 'C#' in the quote above, and you'll see that the statement still works pretty well :-).
2.5 What does 'managed' mean in the .NET context?
The term 'managed' is the cause of much confusion. It is used in various places within .NET, meaning slightly different things.
Managed code: The .NET framework provides several core run-time services to the programs that run within it - for example exception handling and security. For these services to work, the code must provide a minimum level of information to the runtime. Such code is called managed code.
Managed data: This is data that is allocated and freed by the .NET runtime's garbage collector.
Managed classes: This is usually referred to in the context of Managed Extensions (ME) for C++. When using ME C++, a class can be marked with the gc keyword. As the name suggests, this means that the memory for instances of the class is managed by the garbage collector, but it also means more than that. The class becomes a fully paid-up member of the .NET community with the benefits and restrictions that brings. An example of a benefit is proper interop with classes written in other languages - for example, a managed C++ class can inherit from a VB class. An example of a restriction is that a managed class can only inherit from one base class.
2.6 What is reflection?
All .NET compilers produce metadata about the types defined in the modules they produce. This metadata is packaged along with the module (modules in turn are packaged together in assemblies), and can be accessed by a mechanism called reflection. The System.Reflection namespace contains classes that can be used to interrogate the types for a module/assembly.
Using reflection to access .NET metadata is very similar to using ITypeLib/ITypeInfo to access type library data in COM, and it is used for similar purposes - e.g. determining data type sizes for marshaling data across context/process/machine boundaries.
Reflection can also be used to dynamically invoke methods (see System.Type.InvokeMember), or even create types dynamically at run-time (see System.Reflection.Emit.TypeBuilder).
3. Assemblies
3.1 What is an assembly?
An assembly is sometimes described as a logical .EXE or .DLL, and can be an application (with a main entry point) or a library. An assembly consists of one or more files (dlls, exes, html files etc), and represents a group of resources, type definitions, and implementations of those types. An assembly may also contain references to other assemblies. These resources, types and references are described in a block of data called a manifest. The manifest is part of the assembly, thus making the assembly self-describing.
An important aspect of assemblies is that they are part of the identity of a type. The identity of a type is the assembly that houses it combined with the type name. This means, for example, that if assembly A exports a type called T, and assembly B exports a type called T, the .NET runtime sees these as two completely different types. Furthermore, don't get confused between assemblies and namespaces - namespaces are merely a hierarchical way of organising type names. To the runtime, type names are type names, regardless of whether namespaces are used to organise the names. It's the
assembly plus the typename (regardless of whether the type name belongs to a namespace) that uniquely indentifies a type to the runtime.
Assemblies are also important in .NET with respect to security - many of the security restrictions are enforced at the assembly boundary. Finally, assemblies are the unit of versioning in .NET - more on this below.
3.2 How can I produce an assembly?
The simplest way to produce an assembly is directly from a .NET compiler. For example, the following C# program:
public class CTest
{
public CTest() { System.Console.WriteLine( "Hello from CTest" ); }
}
can be compiled into a library assembly (dll) like this:
csc /t:library ctest.cs
You can then view the contents of the assembly by running the "IL Disassembler" tool that comes with the .NET SDK.
Alternatively you can compile your source into modules, and then combine the modules into an assembly using the assembly linker (al.exe). For the C# compiler, the /target:module switch is used to generate a module instead of an assembly.
3.3 What is the difference between a private assembly and a shared assembly?
Location and visibility: A private assembly is normally used by a single application, and is stored in the application's directory, or a sub- directory beneath. A shared assembly is normally stored in the global assembly cache, which is a repository of assemblies maintained by the
.NET runtime. Shared assemblies are usually libraries of code which many applications will find useful, e.g. the .NET framework classes.
Versioning: The runtime enforces versioning constraints only on
shared assemblies, not on private assemblies.
3.4 How do assemblies find each other?
By searching directory paths. There are several factors which can affect the path (such as the AppDomain host, and application configuration files), but for private assemblies the search path is normally the application's directory and its sub-directories. For shared assemblies, the search path is normally same as the private assembly path plus the shared assembly cache.
3.5 How does assembly versioning work?
Each assembly has a version number called the compatibility version. Also each reference to an assembly (from another assembly) includes both the name and version of the referenced assembly.
The version number has four numeric parts (e.g. 5.5.2.33). Assemblies with either of the first two parts different are normally viewed as incompatible. If the first two parts are the same, but the third is different, the assemblies are deemed as 'maybe compatible'. If only the fourth part is different, the assemblies are deemed compatible. However, this is just the default guideline - it is the version policy that decides to what extent these rules are enforced. The version policy can be specified via the application configuration file.
Remember: versioning is only applied to shared assemblies, not private assemblies.
3.6 How can I develop an application that automatically updates itself from the web?
4. Application Domains
4.1 What is an application domain?
An AppDomain can be thought of as a lightweight process. Multiple AppDomains can exist inside a Win32 process. The primary purpose of the AppDomain is to isolate applications from each other, and so it is particularly useful in hosting scenarios such as ASP.NET. An AppDomain can be destroyed by the host without affecting other AppDomains in the process.
Win32 processes provide isolation by having distinct memory address spaces. This is effective, but expensive. The .NET runtime enforces AppDomain isolation by keeping control over the use of memory - all memory in the AppDomain is managed by the .NET runtime, so the runtime can ensure that AppDomains do not access each other's memory.
One non-obvious use of AppDomains is for unloading types. Currently the only way to unload a .NET type is to destroy the AppDomain it is loaded into. This is particularly useful if you create and destroy types on-the-fly via reflection.
4.2 How does an AppDomain get created?
AppDomains are usually created by hosts. Examples of hosts are the Windows Shell, ASP.NET and IE. When you run a .NET application from the command-line, the host is the Shell. The Shell creates a new AppDomain for every application.
AppDomains can also be explicitly created by .NET applications. Here is a C# sample which creates an AppDomain, creates an instance of an object inside it, and then executes one of the object's methods:
using System;
using System.Runtime.Remoting;
using System.Reflection;
public class CAppDomainInfo : MarshalByRefObject
{
public string GetName() { return AppDomain.CurrentDomain.FriendlyName; }
}
public class App
{
public static int Main()
{
AppDomain ad = AppDomain.CreateDomain( "Andy's new domain" ); CAppDomainInfo adInfo = (CAppDomainInfo)ad.CreateInstanceAndUnwrap(
Assembly.GetCallingAssembly().GetName().Name, "CAppDomainInfo" ); Console.WriteLine( "Created AppDomain name = " + adInfo.GetName() ); return 0;
}
}
4.3 Can I write my own .NET host?
Yes. For an example of how to do this, take a look at the source for the dm.net moniker developed by Jason Whittington and Don Box. There is also a code sample in the .NET SDK called CorHost.
5. Garbage Collection
5.1 What is garbage collection?
Garbage collection is a heap-management strategy where a run-time component takes responsibility for managing the lifetime of the memory used by objects. This concept is not new to .NET - Java and many other languages/runtimes have used garbage collection for some time.
5.2 Is it true that objects don't always get destroyed immediately when the last reference goes away?
Yes. The garbage collector offers no guarantees about the time when an object will be destroyed and its memory reclaimed.
There was an interesting thread on the DOTNET list, started by Chris Sells, about the implications of non-deterministic destruction of objects in C#. In October 2000, Microsoft's Brian Harry posted a lengthy analysis of the problem. Chris Sells' response to Brian's posting is here.
5.3 Why doesn't the .NET runtime offer deterministic destruction?
Because of the garbage collection algorithm. The .NET garbage collector works by periodically running through a list of all the objects that are currently being referenced by an application. All the objects that it doesn't find during this search are ready to be destroyed and the memory reclaimed. The implication of this algorithm is that the runtime doesn't get notified immediately when the final reference on an object goes away - it only finds out during the next 'sweep' of the heap.
Futhermore, this type of algorithm works best by performing the garbage collection sweep as rarely as possible. Normally heap exhaustion is the trigger for a collection sweep.
5.4 Is the lack of deterministic destruction in .NET a problem?
It's certainly an issue that affects component design. If you have objects that maintain expensive or scarce resources (e.g. database locks), you need to provide some way to tell the object to release the resource when it is done. Microsoft recommend that you provide a method called Dispose() for this purpose. However, this causes problems for distributed objects - in a distributed system who calls the Dispose() method? Some form of reference- counting or ownership-management mechanism is needed to handle distributed objects - unfortunately the runtime offers no help with this.
5.5 Should I implement Finalize on my class? Should I
implement IDisposable?
This issue is a little more complex than it first appears. There are really two categories of class that require deterministic destruction - the first category manipulate unmanaged types directly, whereas the second category manipulate managed types that require deterministic destruction. An example of the first category is a class with an IntPtr member representing an OS file handle. An example of the second category is a class with a System.IO.FileStream member.
For the first category, it makes sense to implement IDisposable and override Finalize. This allows the object user to 'do the right thing' by calling Dispose, but also provides a fallback of freeing the unmanaged resource in the Finalizer, should the calling code fail in its duty. However this logic does not apply to the second category of class, with only managed resources. In this case implementing Finalize is pointless, as managed member objects cannot be accessed in the Finalizer. This is because there is no guarantee about the ordering of Finalizer execution. So only the Dispose method should be implemented. (If you think about it, it doesn't really make sense to call Dispose on member objects from a Finalizer anyway, as the member object's Finalizer will do the required cleanup.) For classes that need to implement IDisposable and override Finalize, see Microsoft's documented pattern.
Note that some developers argue that implementing a Finalizer is always a bad idea, as it hides a bug in your code (i.e. the lack of a Dispose call). A less radical approach is to implement Finalize but include a Debug.Assert at the start, thus signalling the problem in developer builds but allowing the cleanup to occur in release builds.
5.6 Do I have any control over the garbage collection algorithm?
A little. For example the System.GC class exposes a Collect method, which forces the garbage collector to collect all unreferenced objects immediately.
Also there is a gcConcurrent setting that can be specified via the application configuration file. This specifies whether or not the garbage collector performs some of its collection activities on a separate thread. The setting only applies on multi-processor machines, and defaults to true.
5.7 How can I find out what the garbage collector is doing?
Lots of interesting statistics are exported from the .NET runtime via the '.NET CLR xxx' performance counters. Use Performance Monitor to view them.
5.8 What is the lapsed listener problem?
The lapsed listener problem is one of the primary causes of leaks in .NET applications. It occurs when a subscriber (or 'listener') signs up for a publisher's event, but fails to unsubscribe. The failure to unsubscribe means that the publisher maintains a reference to the subscriber as long as the publisher is alive. For some publishers, this may be the duration of the application.
This situation causes two problems. The obvious problem is the leakage of the subscriber object. The other problem is the performance degredation due to the publisher sending redundant notifications to 'zombie' subscribers.
There are at least a couple of solutions to the problem. The simplest is to make sure the subscriber is unsubscribed from the publisher, typically by adding an Unsubscribe() method to the subscriber. Another solution, documented here by Shawn Van Ness, is to change the publisher to use weak references in its subscriber list.
5.9 When do I need to use GC.KeepAlive?
It's very unintuitive, but the runtime can decide that an object is garbage much sooner than you expect. More specifically, an object can become garbage while a method is executing on the object, which is contrary to most developers' expectations. Chris Brumme explains the issue on his blog. I've taken Chris's code and expanded it into a full app that you can play with if you want to prove to yourself that this is a real problem:
using System;
using System.Runtime.InteropServices;
class Win32
{
[DllImport("kernel32.dll")]
public static extern IntPtr CreateEvent( IntPtr lpEventAttributes, bool bManualReset,bool bInitialState, string lpName);
[DllImport("kernel32.dll", SetLastError=true)]
public static extern bool CloseHandle(IntPtr hObject);
[DllImport("kernel32.dll")]
public static extern bool SetEvent(IntPtr hEvent);
}
class EventUser
{
public EventUser()
{
hEvent = Win32.CreateEvent( IntPtr.Zero, false, false, null );
}
~EventUser()
{
Win32.CloseHandle( hEvent ); Console.WriteLine("EventUser finalized");
}
public void UseEvent()
{
UseEventInStatic( this.hEvent );
}
static void UseEventInStatic( IntPtr hEvent )
{
//GC.Collect();
bool bSuccess = Win32.SetEvent( hEvent );
Console.WriteLine( "SetEvent " + (bSuccess ? "succeeded" : "FAILED!") );
}
IntPtr hEvent;
}
class App
{
static void Main(string[] args)
{
EventUser eventUser = new EventUser();
eventUser.UseEvent();
}
}
If you run this code, it'll probably work fine, and you'll get the following output:
SetEvent succeeded
EventDemo finalized
However, if you uncomment the GC.Collect() call in the UseEventInStatic()
method, you'll get this output:
EventDemo finalized
SetEvent FAILED!
(Note that you need to use a release build to reproduce this problem.)
So what's happening here? Well, at the point where UseEvent() calls UseEventInStatic(), a copy is taken of the hEvent field, and there are no further references to the EventUser object anywhere in the code. So as far as the runtime is concerned, the EventUser object is garbage and can be collected. Normally of course the collection won't happen immediately, so you'll get away with it, but sooner or later a collection will occur at the wrong time, and your app will fail.
A solution to this problem is to add a call to GC.KeepAlive(this) to the end of the UseEvent method, as Chris explains.
6. Serialization
6.1 What is serialization?
Serialization is the process of converting an object into a stream of bytes. Deserialization is the opposite process, i.e. creating an object from a stream of bytes. Serialization/Deserialization is mostly used to transport objects (e.g. during remoting), or to persist objects (e.g. to a file or database).
6.2 Does the .NET Framework have in-built support for serialization?
There are two separate mechanisms provided by the .NET class library - XmlSerializer and SoapFormatter/BinaryFormatter. Microsoft uses XmlSerializer for Web Services, and SoapFormatter/BinaryFormatter for remoting. Both are available for use in your own code.
6.3 I want to serialize instances of my class. Should I use
XmlSerializer, SoapFormatter or BinaryFormatter?
It depends. XmlSerializer has severe limitations such as the requirement that the target class has a parameterless constructor, and only public read/write properties and fields can be serialized. However, on the plus side, XmlSerializer has good support for customising the XML document that is produced or consumed. XmlSerializer's features mean that it is most suitable for cross-platform work, or for constructing objects from existing XML documents.
SoapFormatter and BinaryFormatter have fewer limitations than XmlSerializer. They can serialize private fields, for example. However they both require that the target class be marked with the [Serializable] attribute, so like XmlSerializer the class needs to be written with serialization in mind. Also there are some quirks to watch out for - for example on deserialization the constructor of the new object is not invoked.
The choice between SoapFormatter and BinaryFormatter depends on the application. BinaryFormatter makes sense where both serialization and deserialization will be performed on the .NET platform and where performance is important. SoapFormatter generally makes more sense in all other cases, for ease of debugging if nothing else.
6.4 Can I customise the serialization process?
Yes. XmlSerializer supports a range of attributes that can be used to configure serialization for a particular class. For example, a field or property can be marked with the [XmlIgnore] attribute to exclude it from serialization. Another example is the [XmlElement] attribute, which can be used to specify the XML element name to be used for a particular property or field.
Serialization via SoapFormatter/BinaryFormatter can also be controlled to some extent by attributes. For example, the [NonSerialized] attribute is the equivalent of XmlSerializer's [XmlIgnore] attribute. Ultimate control of the serialization process can be acheived by implementing the the ISerializable interface on the class whose instances are to be serialized.
6.5 Why is XmlSerializer so slow?
There is a once-per-process-per-type overhead with XmlSerializer. So the first time you serialize or deserialize an object of a given type in an application, there is a significant delay. This normally doesn't matter, but it may mean, for example, that XmlSerializer is a poor choice for loading configuration settings during startup of a GUI application.
6.6 Why do I get errors when I try to serialize a Hashtable?
XmlSerializer will refuse to serialize instances of any class that implements IDictionary, e.g. Hashtable. SoapFormatter and BinaryFormatter do not have this restriction.
6.7 XmlSerializer is throwing a generic "There was an error reflecting MyClass" error. How do I find out what the problem is?
Look at the InnerException property of the exception that is thrown to get a more specific error message.
6.8 Why am I getting an InvalidOperationException when I
serialize an ArrayList?
XmlSerializer needs to know in advance what type of objects it will find in an
ArrayList. To specify the type, use the XmlArrayItem attibute like this:
public class Person{ public string Name; public int Age;
}
public class Population{
[XmlArrayItem(typeof(Person))] public ArrayList People;
}
7. Attributes
7.1 What are attributes?
There are at least two types of .NET attribute. The first type I will refer to as a metadata attribute - it allows some data to be attached to a class or method. This data becomes part of the metadata for the class, and (like other class metadata) can be accessed via reflection. An example of a metadata attribute is [serializable], which can be attached to a class and means that instances of the class can be serialized.
[serializable] public class CTest {}
The other type of attribute is a context attribute. Context attributes use a similar syntax to metadata attributes but they are fundamentally different. Context attributes provide an interception mechanism whereby instance activation and method calls can be pre- and/or post-processed. If you have encountered Keith Brown's universal delegator you'll be familiar with this idea.
7.2 Can I create my own metadata attributes?
Yes. Simply derive a class from System.Attribute and mark it with the
AttributeUsage attribute. For example:
[AttributeUsage(AttributeTargets.Class)]
public class InspiredByAttribute : System.Attribute {
public string InspiredBy;
public InspiredByAttribute( string inspiredBy ){ InspiredBy = inspiredBy;
}
}
[InspiredBy("Andy Mc's brilliant .NET FAQ")]
class CTest{
}
class CApp{
public static void Main(){
object[] atts = typeof(CTest).GetCustomAttributes(true);
foreach( object att in atts )
if( att is InspiredByAttribute )
Console.WriteLine( "Class CTest was inspired by {0}", ((InspiredByAttribute)att).InspiredBy );
}
}
No comments:
Post a Comment