The Wiert Corner – irregular stream of stuff

Jeroen W. Pluimers on .NET, C#, Delphi, databases, and personal interests

  • My badges

  • Twitter Updates

  • My Flickr Stream

  • Pages

  • All categories

  • Enter your email address to subscribe to this blog and receive notifications of new posts by email.

    Join 1,860 other subscribers

Archive for the ‘C# 1.0’ Category

Yes Dorothy, the .NET System.Array Class can throw you a NotSupportedException

Posted by jpluimers on 2016/09/21

It’s been in the System.Array class forever, but remarkably few people do know that it can throw you a NotSupportedException (for instance when calling Add, Insert, Remove, etc).

It does because it implements IList, but not all methods implemented from IList are valid.

And it also indicates that, as the IList Properties allows for IsFixedSize to return false.

A similar case is there for IsReadOnly: then you cannot even modify the values.

Ever since I started teaching .NET/C# classes almost 15 years ago, I warned:

beware when you use IList as not everybody implements all methods.

–jeroen

via:

Posted in .NET, .NET 1.x, .NET 2.0, .NET 3.0, .NET 3.5, .NET 4.0, .NET 4.5, C#, C# 1.0, C# 2.0, C# 3.0, C# 4.0, C# 5.0, C# 6 (Roslyn), Development, Software Development | Leave a Comment »

.NET/C#: two ways of creating an empty array

Posted by jpluimers on 2016/08/18

Empty arrays are not used often as arrays usually are about the presence data, not about the absence.

Here are two ways based on the int data type in C# (the original [WayBackexamples [WayBack] are using string, but since string itself is also a kind of array…):

Specify a size of zero:

int[] a = new int[0];

Specify an empty initialisation:

int[] a = new int[] { };

Though many people think arrays are a thing of the past, I think it is one of the first generic types and have their place. For one, enumerating over arrays using foreach is a lot faster in many environments than enumerating over other data types. Another thing is that the fixed nature of arrays can be very beneficial in setting constraints.

That’s why I like the balanced view from Eric Lippert [WayBack] in Arrays considered somewhat harmful – Fabulous Adventures In Coding – Site Home – MSDN Blogs [WayBack]

–jeroen

via:

Posted in .NET, .NET 1.x, .NET 2.0, .NET 3.0, .NET 3.5, .NET 4.0, .NET 4.5, C#, C# 1.0, C# 2.0, C# 3.0, C# 4.0, C# 5.0, C# 6 (Roslyn), Development, Software Development | Leave a Comment »

c# – How can I create a temp file with a specific extension with .net? – Stack Overflow

Posted by jpluimers on 2016/01/27

You’d think this is a simple question.

Be amazed by the many ways leading into to creating a temporary file with a specific extension.

This list doesn’t even cover all of them:

  • Create a file name based on GetTempPath, a Guid and an extension.
  • Use the TempFileCollection from the Compiler in System.CodeDom.
  • Get a random file name, then change the extension. Loop until it is unique.
  • Use a timestamp to generate unique file names.

All via: c# – How can I create a temp file with a specific extension with .net? – Stack Overflow.

Which one would you choose?

–jeroen

Posted in .NET, .NET 1.x, .NET 2.0, .NET 3.0, .NET 3.5, .NET 4.0, .NET 4.5, C#, C# 1.0, C# 2.0, C# 3.0, C# 4.0, C# 5.0, C# 6 (Roslyn), Development, RemObjects C#, Software Development | Leave a Comment »

Ensure you know the state of an instance.

Posted by jpluimers on 2015/11/24

A while ago, I came across a class having (among other members) two methods named like this:

  • Start
  • Stop

Within one of the other members of the class, I had to (temporarily) Stop processing, then Start it again.

But I couldn’t, as neither Start, nor Stop would make a record of the state it left the instance in.

Always ensure you know the state of an instance.

So I added the state, and tests to ensure a Stop/Start change was indeed not breaking things.

–jeroen

Posted in .NET, .NET 1.x, .NET 2.0, .NET 3.0, .NET 3.5, .NET 4.0, .NET 4.5, C#, C# 1.0, C# 2.0, C# 3.0, C# 4.0, C# 5.0, C# 6 (Roslyn), Delphi, Delphi 10 Seattle, Delphi 2007, Delphi 2009, Delphi 2010, Delphi XE, Delphi XE2, Delphi XE3, Delphi XE4, Delphi XE5, Delphi XE6, Delphi XE7, Delphi XE8, Development, Software Development | 1 Comment »

Do not expose properties with writeable fields without a chance to react on the writes.

Posted by jpluimers on 2015/11/04

From a chat with a co-worker a while ago:

I’m not against properties. Just something against properties properties that are objects with writeable fields.

So even if you expose such a property as read-only, it can still get its writeable fields overwritten.

That is a pain when those are part of the state of the underlying object.

In other words: encapsulate your state changes.

Here we solved it by making

  • the type of the property immutable
  • the property writeable
  • react on state changes during the write

Proper encapsulation.

In this case it was a project mixing C# and Delphi, but you can easily apply the above to any language featuring classes and properties.

Another solution would have been to extend the type of the property so it can expose an event that fires during change. Much more convoluted.

–jeroen

Posted in .NET, C#, C# 1.0, C# 2.0, C# 3.0, C# 4.0, C# 5.0, C# 6 (Roslyn), Delphi, Delphi 10 Seattle, Delphi 2007, Delphi 2009, Delphi 2010, Delphi XE, Delphi XE2, Delphi XE3, Delphi XE4, Delphi XE5, Delphi XE6, Delphi XE7, Delphi XE8, Development, Software Development | 2 Comments »

c#: Lazy is not constrained to static contexts, instance field initialisers are – via: Stack Overflow

Posted by jpluimers on 2015/11/03

Lazy<T> is not constrained to static contexts.

Instance field initialisers cannot use instance references (but can use static references) as they run outside of the constructor.Though there are arguments for instance field initialisers too., I think this is a good reason to initialise fields inside the constructor: there you do have access to instance references (but should not call virtual instance methods or properties) which leads to another reason: consistency as field initialisers run in the opposite hierarchy order as constructors (incidentally causing this virtual method restriction).

Boy, that was a long sentence (:

–jeroen

via:

Posted in .NET, .NET 1.x, .NET 2.0, .NET 3.0, .NET 3.5, .NET 4.0, .NET 4.5, C#, C# 1.0, C# 2.0, C# 3.0, C# 4.0, C# 5.0, C# 6 (Roslyn), Development, Software Development | Leave a Comment »

Unless you write an installer with the right manifest, don’t include Installer, Update, Upgrade, Setup, … in your EXE name

Posted by jpluimers on 2015/09/28

I’ve seen this question coming up a few times, and bumped into this at a client recently: the UAC dialog coming up when debugging a 32-bit executable.

This is caused (more details below) by Installer Detection Technology introduced in Windows Vista (with UAC) and tightened in more modern Windows versions.

The solution is to either:

  • not include Installer, Patch, Update, Upgrade, Setup, … in your EXE name
  • provide a correct manifest to your EXE (getting this right can be hard)
  • don’t use x86 as platform target

For software you don’t have source code for, you can alter the manifest with a requestedExecutionLevel elementFixing the way Vista Auto-detects Installers – Ben’s Writing.

A few links on Installer Detection Technology in Windows:

Read the rest of this entry »

Posted in .NET, .NET 1.x, .NET 2.0, .NET 3.0, .NET 3.5, .NET 4.0, .NET 4.5, .NET CF, C#, C# 1.0, C# 2.0, C# 3.0, C# 4.0, C# 5.0, C# 6 (Roslyn), Delphi, Delphi 10 Seattle, Delphi 2, Delphi 2005, Delphi 2006, Delphi 2007, Delphi 2009, Delphi 2010, Delphi 3, Delphi 4, Delphi 5, Delphi 6, Delphi 7, Delphi XE, Delphi XE2, Delphi XE3, Delphi XE4, Delphi XE5, Delphi XE6, Delphi XE7, Delphi XE8, Development, RemObjects C#, Software Development | 1 Comment »

.NET enable and disable Fusion log to investigate assembly loading issues

Posted by jpluimers on 2015/08/18

Had to investigate some Assembly Loading issues, so I wrote two batch files to enable and disable the .NET Fusion Log:


reg add "HKLM\Software\Microsoft\Fusion" /v EnableLog /t REG_DWORD /d 0 /f


reg add "HKLM\Software\Microsoft\Fusion" /v EnableLog /t REG_DWORD /d 1 /f

They modify the HKLM\Software\Microsoft\Fusion key REG_DWORD value EnableLog.

A few notes:

  • It is very important to turn of the Fusion log settings as soon as you are finished investigating. Fusion logs potentially take a lot of resources.
  • When you have a .NET host like ISS, you have to restart that host (for instance by running iisreset)
  • There is also Fuslogvw.exe Assembly Binding Log Viewer, but I like logging over viewing as logs are persistent.
  • There are more values under HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Fusion you can configure; see the answer by Gary Kindel on StackOverflow:
    • DWORD ForceLog set value to 1
    • DWORD LogFailures set value to 1
    • DWORD LogResourceBinds set value to 1
    • String LogPath set value to folder for logs e.g. C:\FusionLog\ (ensure the final backslash is there and the folder exists).

–jeroen

via:

Posted in .NET, .NET 1.x, .NET 2.0, .NET 3.0, .NET 3.5, .NET 4.0, .NET 4.5, C#, C# 1.0, C# 2.0, C# 3.0, C# 4.0, C# 5.0, C# 6 (Roslyn), Development, Software Development | Leave a Comment »

Difference Between Int32.Parse, Convert.ToInt32, and Int32.TryParse – CodeProject

Posted by jpluimers on 2015/08/05

Every C#/VB/.NET developer should read Difference Between Int32.Parse, Convert.ToInt32, and Int32.TryParse – CodeProject.

Then also read TryParse with default values.

It is all about handling values that are not Integers, Overflow values and Nulls. There are subtle differences, in the handling of the methods, and the exceptions they could throw: ArgumentNullException, FormatException and OverflowException.

Finally read all about the NumberStyles enumeration, IFormatProvider interface and CultureInfo (especially the difference between InvariantCulture, CurrentCulture, CurrentUICulture and InstalledUICulture).

Because getting your conversions right matters.

–jeroen

Posted in .NET, .NET 1.x, .NET 2.0, .NET 3.0, .NET 3.5, .NET 4.0, .NET 4.5, C#, C# 1.0, C# 2.0, C# 3.0, C# 4.0, C# 5.0, C# 6 (Roslyn), Development, Mono, Mono for Android, Prism, RemObjects C#, Software Development, VB.NET, VB.NET 10.0, VB.NET 11.0, VB.NET 7.0, VB.NET 7.1, VB.NET 8.0, VB.NET 9.0 | Leave a Comment »

Boolean parameters usualy are a sign of weakness (via: Avoiding Booleans)

Posted by jpluimers on 2015/07/16

During a recent code review, I bumped into a couple of C# constructors having boolean parameters, leading to the dreaded magic booleans code smell.

This reminded me of the infamous Avoiding Booleans post on Coding Horror, which now is almost 10 years old.

To celebrate, I will coin the Dutch phrase when marking these in a review:

Boolean parameters en literals zijn vrijwel altijd een zwaktebod. Een teken dat beter nagedacht moet worden over het doel van de code.

The Dutch word zwaktebod is used when bidding Bridge using the Acol system. It is the equivalent of a “weak takeout” response to a bid of 1 NT (notrump or no trump, in other languages sometimes sans atout).

The English translation is just about this:

Boolean parameters and literals virtually always are a sign of weakness. It indicates you need to give more thought to the goal of the code.

–jeroen

Posted in .NET, .NET 1.x, .NET 2.0, .NET 3.0, .NET 3.5, .NET 4.0, .NET 4.5, C#, C# 1.0, C# 2.0, C# 3.0, C# 4.0, C# 5.0, C# 6 (Roslyn), Development, Software Development | Leave a Comment »