Object Browser: The World's Worst-Named Tool

So, you know the class you need but you don't know what class library it's in. How do you add the right reference to your project? Object Browser will let you do it in two steps.

You can do that because "Object Browser" is patently misnamed -- to begin with, it displays classes, not objects. Just as obviously, it isn't just limited to classes (objects) but also displays namespaces, enums, structs, interfaces, and class members (e.g. events, properties, etc.).

If you know what class (or interface or enum or, even, member) you want, you can search for it in Object Browser using the search box at the top of Object Browser's window. Once you find what you're looking for, just click on the Add to References icon at the top of Object Browser to add a reference to the relevant library to whatever project you have selected in Solution Explorer.

So, it isn't just a browser, either.

Worst. Name. Ever.

Posted by Peter Vogel on 04/05/2018 at 5:18 AM0 comments


Inheriting from Generics: Set the Datatype ... Or Not

When you're creating a derived class and your base type is a generic class, you have two choices in implementing your derived class: You can set the type of your derived class or you can make your derived class another generic class.

For example, imagine that you have a class called ReadRepository that accepts a variety of types:

Public Class ReadRepository(Of T)

End Class

If you create a CustomerRepository that inherits from ReadRepository, you might choose to set the type of your base class:

Public Class CustomerRepository 
    Inherits ReadRepository(Of Customer)
End Class

That's the strategy to follow when your derived class adds functionality specific to a datatype (in this case, I'm adding functionality specific to the Customer class).

On the other hand, if you wanted to create an UpdateAndReadRepository, you might choose to have your new class also be a generic class. In that case, your derived class also accepts a type placeholder and passes that placeholder to the base class:

Public Class UpdateAndReadRepository(Of T) 
    Inherits ReadRepository(Of T)
End Class

This is the strategy to follow if you're extending the base class with functionality that can be used with a variety of classes. In this case, adding Update capabilities for any class.

Posted by Peter Vogel on 04/02/2018 at 5:44 AM0 comments


Collecting Collections: The Lookup Collection

Every once in a while, I end up with a bunch of collections in memory and need the ability to pick the collection I want by name. For example, I might need a bunch of Customer collections, one for each city where I have a customer.

Furthermore, I'd like to have those collections organized into a larger collection called CityCusts so I can pull out all of the Customers for any specific city.

The code I want to use to retrieve the Customers collection for a city would look like this:

Dim lstCusts As List(Of Customer)
lstCusts = CityCusts("Regina")

I could build that CityCusts collection myself by defining a Dictionary collection that holds values that are, themselves, a collection of Customers:

Private CityCusts As Dictionary(Of String, List(Of Customer))

Initializing the individual collections for each Dictionary entry and adding the right customer to the right City collection would be a pain, though. Even using LINQ's Group By syntax is awkward, in both languages.

Fortunately, the Lookup collection makes building that collection a snap. To declare a Lookup class you just specify the type of the Dictionary's key (pretty much always a string) and the type held in the collection. Here's the declaration to create a Lookup collection that holds a collection of Customers associated with a city:

Private CityCusts As Lookup(Of String, Customer)

To load the class, I just start with a collection of Customer objects and call the collection's ToLookup method. I must pass the method a lambda expression that specifies which property to use to organize the Customers collection.

This example creates collections of Customer objects that share the same value in their City property and then stores those collections in CityCusts, under the name of the shared city:

CityCusts = Custs.ToLookup(Function(c) c.City)

In C#, the two lines of code to declare and load the collection would look like this:

private Lookup<string, Customer> CityCusts;
CityCusts = Custs.ToLookup(c => c.City);

Bada bing (as they say), bada boom.

Posted by Peter Vogel on 03/19/2018 at 5:51 AM0 comments


Finding What's Changed in Your Code

You (or someone else) rolled out a new version of an application and it's behaving…oddly. You ask the people involved (or yourself) "What changed?" and the answer you get is "Nothing." This is, of course, not true. Users do this to us all the time though the question we ask is slightly different ("What did you do differently?", "Nothing").

Of course, the answer we're really giving is "Nothing relevant," which isn't quite the same thing as "Nothing." In both cases, the person asking the question has unconsciously edited out of recollection anything that doesn't seem relevant to the problem. The only solution is to force ourselves to look at everything that changed, rather than just "what's relevant" (it's like looking for something you've lost: Once you've looked in all the places where it should be then you have to look at all the places where it shouldn't be).

Your source control system should be able to tell you all the lines of code that have changed between one version of your code and another. If you're not using a source control system, you can use Visual Studio's Compare Files option, though it's hard to get to in more recent versions of Visual Studio. To compare two files without source control, first click in the Command window (if the window isn't open, use View | Other Windows | Command Window to open it). In the Command window, type this:

Tools.DiffFiles <filename1> <fileName2>

That command isn't as bad as it looks: As soon as you start typing a file name, you'll get IntelliSense support to finish off the file's name. After you hit the Enter key, the command will open a tab with two panes showing both files with your changed/added/deleted lines highlighted.

Of course, even when you find the changed lines you have to stop thinking "Oh, well that can't be the problem." Yes, it can. In fact, it probably is.

Posted by Peter Vogel on 03/15/2018 at 2:16 PM0 comments


Converting Your ASP.NET Web Forms Application to ASP.NET MVC

If you want to build a Web application quickly, do it with ASP.NET Web Forms. However, you have to be willing to give up a lot: client-side coding and Ajax is more awkward in Web Forms than MVC, you won't have as complete control over your HTML/CSS as in MVC, and you'll have to be careful about what code goes into your code-behind file if you want to do automated testing.

Because those are areas that people do care about, the future of Web development in .NET belongs to ASP.NET MVC (it's telling that there is no "MVC" in .NET Core -- the Web application framework is just called "ASP.NET Core").

I made a good living out of Web Forms, but, except for the occasional fix to legacy applications, my clients all want me to work in ASP.NET MVC now. That being the case, I occasionally get asked "How do I migrate from Web Forms to MVC?" Here's what I tell my clients (and I won't charge you a thing):

  1. Don't. There is no migration path and your Web Forms application has that whole "working" feature that users like.
  2. If it's just a need to add client-side support or to look up-to-date, consider replacing the default Microsoft Web Form controls with controls from third-party providers. A lot of those third-party controls offer features that the default Microsoft Web Form controls don't (rich client-side models, for example). They are also often "plug-compatible" with your current controls so you can just replace your existing Microsoft control with a third-party control and start doing cool stuff.
  3. If you must convert (because you don't want to support two toolsets or there's something you want your application to do that Web Forms doesn't support or your commercial application looks old fashioned or management has mandated it or ...), remember that, since Visual Studio 2012, you can combine MVC and Web Forms in the same project. Do the conversion form by form rather than creating a whole new site. There will be some forms (those forms maintaining your lookup tables, for example) that you may never bother converting.
  4. Where you need to rewrite an existing Web Form to ASP.NET MVC, remember that you wrote that form the way you did because you wrote it in Web Forms. If you'd written it in MVC, you'd have done it differently. Do it that "differently way" now.
  5. Do create all your new pages in ASP.NET MVC even in existing Web Form projects (no sense in making the problem larger).

Posted by Peter Vogel on 03/12/2018 at 8:45 AM0 comments


Brush Up Your SQL

Thanks to LINQ and Entity Framework, I don't write a lot of online, transactional SQL any more (I like to think that, thanks to Entity Framework, I have Microsoft's ADO.NET team doing that for me). But there are features in SQL that just aren't available in the current version of LINQ (see my columns on SQL Server 2016's support for JSON and temporal tables for examples).

But it isn't just new features that may drive you to issuing raw SQL through Entity Framework, though.

For example, I had someone attending one of the SQL Server querying classes I teach with a problem that LINQ couldn't solve efficiently (I also teach the 2014 course, though it doesn't really matter because they're exactly the same course ... though, as I remember, the 2016 certification exam was easier).

His problem was that he was receiving a constant stream of reports from various vehicles and needed, from that received data to determine when the vehicle had stopped to refuel. The only way to answer this question is to compare the amount of gas in the tank at (time A) with the amount in the tank at (time A – 5 minutes).

In LINQ, this could be solved by joining every report row with all the report row from the same vehicle more than 5 minutes ago ... but that query had a response time that could be measured with a calendar, not with a stop watch. However, a raw SQL query that used the Preceding keyword (not available in LINQ), gave a very snappy response.

I love LINQ. I love Entity Framework. You're still going to need SQL.

Posted by Peter Vogel on 03/05/2018 at 7:20 AM0 comments


Avoiding Entity Framework Slowdown

Generally speaking, I don't worry much about tweaking my LINQ queries when working with Entity Framework (this is also true when I'm working with SQL directly, by the way). I'm always telling my clients that if they want to speed up their data access they should look at their database design and, especially, how they're using indexes.

There is one exception to that rule, though: If you've got a query doing a comparison to a char or varchar column, then you may be unnecessarily slowing down your LINQ queries. The problem is that Entity Framework assumes that the database equivalent to your string property is a Unicode (nchar or nvarchar) column. If that's not the case -- if your columns are char or varchar -- then you're incurring some data conversion overhead when you use those properties in a Where clause like this:

Dim res = From cust In db.Customers
 	   Where cust.Name = "Vogel"
	   Select cust

That Where clause is going to get converted into a SQL statement that looks like this:

Select *
From Customers
Where Name = N'Vogel'

Because Entity Framework assumes my column is a Unicode column, it slaps the N in front of my string constant ('Vogel') to turn it into a Unicode literal ... and then tries to compare that Unicode literal to my non-Unicode (char or varchar) Name column. SQL Server then has to do the work to convert 'Vogel' back out of Unicode (or it may be converting my Name column to Unicode, either of which is bad).

The solution is to tell Entity Framework that your column is char or varchar by decorating the property with the Column attribute and setting the attribute's TypeName property. This example tells Entity Framework that my Name column is varchar:

<Column(TypeName="varchar")>
Public Property Name As String

For a longer discussion, see Brian Sullivan's blog post on how he got burned. And stop using char and varchar!

Posted by Peter Vogel on 02/23/2018 at 8:59 AM0 comments


The C# Scopes for Privileged Inheritance

You can't really combine the various scopes that you can apply to a C# member because, I suspect, they wouldn't make sense (what's the scope of a public private method, for example?). However, you can combine internal with private and protected like this:

public class CustomerBase
{
  protected internal void DeleteCustomer() {

or

public class CustomerBase
{
  private protected void DeleteCustomer() {

In my first example, because of its internal scope, DeleteCustomer can be used by code that instantiates CustomerBase but only if that code is in the same project as CustomerBase. Because of the protected declaration, however, code in other projects can access DeleteCustomer through a class that inherits from CustomerBase.

Code in other projects that instantiate CustomerBase directly won't see DeleteCustomer. Essentially, code in the same project as CustomerBase has a privilege denied to code in other projects (the ability to delete customers, in this case).

The scope private protected goes even further and is probably more useful. This scope prevents DeleteCustomer from being accessed except through classes that derive from CustomerBase and only if those derived classes are in the same project as CustomerBase.

Code in some other project in a class that inherits from CustomerBase won't be able to access DeleteCustomer. This allows you to create derived classes in the same project as CustomerBase that have privileges denied to classes in other projects that inherit from CustomerBase.

Posted by Peter Vogel on 02/21/2018 at 6:47 AM0 comments


Getting Data from the Request Object in ASP.NET MVC

Most of the time in ASP.NET MVC I can count on model binding to fill in the values for the parameters to my Action methods. Every once in a while, though, model binding doesn't do what I expect. You can create your own custom modelbinder to solve this problem (I've even discussed how to do that for the ASP.NET Web API). However, that may be overkill.

Often you can solve your problem by accessing one (or all) of the collections in ASP.NET's Request object (available in your controller through its Request property). There are a number of collections you can use to access data sent from the browser and their names do a pretty good job of describing what data they hold:

  • Cookies
  • Files
  • Form
  • Headers
  • QueryString
  • ServerVariables

You can also use the Params collection (which combines the values from Cookies, Form, QueryString, and ServerVariables) but you'd be safer with one of the more specific collections. The problem with the Params collection is that you can have the same data stored under the same name in multiple collections (Cookies and QueryString, for example). If you use Params you can't be sure which of those collections your data will come from.

Posted by Peter Vogel on 02/20/2018 at 5:52 AM0 comments


Paste JSON and XML as a Class

I don't use Visual Studio's Edit menu enough, obviously: For some time there's been a Paste Special option on that menu that I didn't know about. It has up to two choices depending on what kind of file you're editing: Paste XML as Classes and Paste JSON as Classes. Using the Paste JSON as Classes option, I can copy some JSON in my JavaScript code and create a corresponding C# or Visual Basic class in my server-side code just by picking a menu the appropriate menu choice.

However, I'd rather be able to use this functionality without having to drag my mouse to the Edit menu. To support that, I added a keystroke combination for the underlying commands related to these menu choices.

If you want to do the same, first go to Tools | Options | Environment | Keyboard. To grab the commands, in the "Show commands containing" textbox, type in AsClasses and select one. Then, from the "Use new shortcut in" dropdown list, pick your language's editor (for example, CSharp Editor or Microsoft Visual Basic Editor). Now you can click in the "Press shortcut keys" textbox to select your shortcut key(s): Just press the keys you want to use. Finally, click the Assign button. Lots of people forget that step and just click the OK button at the bottom of the dialog (by "lots of people" I mean: me).

You probably don't want to override any existing shortcut keys so I'd recommend picking a "chord": two or more keys pressed sequentially while holding down one of the control keys (the dialog box will tell you if you pick a shortcut that's already in use). I picked Ctrl+K, J for pasting JSON and Ctrl+K, M for pasting XML (Ctrl+K, X was already taken for inserting snippets and, who knows, I might start using that some day).

My life would be complete if there was a Paste Class as JSON when I was in a TypeScript or JavaScript file. In the meantime, I'm settling for Edit | Insert File as Text and doing some editing or using Nha Bui Doc's TypeScriptSyntaxPaste.

Posted by Peter Vogel on 02/16/2018 at 6:16 AM0 comments


Manipulating URLs on the Server

There's nothing stopping you from working with URLs in code using standard string functions. If you want to create a URL, you should use the URL object. On the other hand, if you want to analyze a URL -- or, more properly, a URI -- you should use the Uri class.

To use the Uri class, you just instantiate it, passing a URI. Once you've done that, you can check whether the URI's scheme name (ftp, http and so on) is valid, that the host name is a valid DNS name, or even dismantle the URI to get its individual components (to get, for example, the port or host name).

You can even use the Uri object to compare two URLs to see if they are significantly different (the Uri object is smart enough to recognize that two URLs are identical if one URL omits a port number while the other one specifies the default).

Posted by Peter Vogel on 02/13/2018 at 10:27 AM0 comments


Best Practice: Declare Variables as Constants

In the bad old days, I only used constants as a way to assign names to values. Things like this, for example:

Private Const DaysInWeek As Integer = 7

However, these days most of my variables are holding objects; Some of my other variables are holding things like collections or delegates. That changes what we mean by "constant."

When a variable is declared as a constant, it doesn't mean that the object it's holding is immutable. You can, for example, still change values in the properties of an object held in a constant or add items to a collection held in a constant. It just means that you can't replace the object (or collection) that was assigned when the variable was initialized.

Especially in TypeScript (where many of my variables are holding functions), I'm switching to declaring my variables as constants. A subset of these variables are fields (class-level variables) that are set precisely once, in an object's constructor, and I'm taking the time to mark as those read only also.

If I want an actual immutable object I'll declare the object's properties as read only or use one of .NET's read only collections.

In part, I'm doing this because I'm thinking more in terms of value objects (an important concept in Domain Driven Design). However, there are other benefits. I'm told, for example, it makes life easier for the compiler. But it's also because this practice sends a clear message to the next developer (either using my code or modifying it) about how I think that the object should be used.

Posted by Peter Vogel on 02/12/2018 at 7:20 AM0 comments


Most   Popular
Upcoming Events

.NET Insight

Sign up for our newsletter.

Terms and Privacy Policy consent

I agree to this site's Privacy Policy.