And wow the book is third on this weeks bestselling early access titles. Juhu!

But there is no Windows Resource Kit for 2008. You might use the sc.exe and get the old srvany.exe (with compatibility issues according to Microsoft itself). It would work but why bother when there is a Non-Sucking Service Manager! All you have to do to install a service with this tool is to download it, issue a

nssm.exe install SeleniumRC

and edit this dialog box:

Click Install service and you are done. Selenium RC Server is installed. All you have to do is to start it. Voila!

C:\Program Files\svn\bin>svn log https://your_server/svn/your_repository/trunk –username username –password password
Error validating server certificate for ‘https://your_server:443′:
- The certificate is not issued by a trusted authority. Use the
fingerprint to validate the certificate manually!
- The certificate hostname does not match.
Certificate information:
- Hostname: your_server
- Valid: from Sat, 26 Sep 2009 17:24:27 GMT until Tue, 24 Sep 2019 17:24:27 GMT

- Issuer: your_server
- Fingerprint: 24:8e:f6:ba:c7:a6:3f:69:32:c0:21:92:64:44:62:fe:2c:bb:b4:69
(R)eject, accept (t)emporarily or accept (p)ermanently?

If you accept you will not be bothered again. But CCNet works as a Windows Service. There is no one to make the decision. How to deal with this issue. Well earlier it was easy enough. You had to use one of the security holes and start cmd.exe in interactive mode wit at command (look here for more details). But with Windows Server 2008 it is not possible you will simply get this:

C:\Users\Administrator>time
The current time is: 23:31:11.59
Enter the new time:

C:\Users\Administrator>at 22:32 /interactive cmd.exe
Warning: Due to security enhancements, this task will run at the time
expected but not interactively.
Use schtasks.exe utility if interactive task is required (’schtasks /?’
for details).
Added a new job with job ID = 1

How to deal with this. There is very easy solution. Set the CruiseContril.NET service “Allow to interact with desktop” flag (Start –> Control Panel –> Administrative Tools –> Services –CruiseControl.NET) like this

Restart the service and wait a while for this windows to appear:

Select show me the message.

Voila! You have command line as Local System user available. You can now issue the

C:\Program Files\svn\bin>svn log https://your_server/svn/your_repository/trunk –username username –password password
command and accept the SSL certificate permanently.

From this time on you CCNet server will not have any problems with accessing your secured repository.

The Continuous Integration in .NET book is available as an early access edition from MEAP. For those of you that are not familiar with MEAP it stands for Manning Early Access Program and it is the place where you can download a free chapter and buy the whole book printed or as an ebook before it gets published. You will get continuously chapter after chapter successively as they will be written.

We are working with Oracle Database and .NET project using NHibernate as a object-relation mapper. We have a function in Oracle that returns an Oracle Object. Objects in Oracle are user defined types. They are defined like this:

CREATE OR REPLACE TYPE Foo as object(
Value1 number,
Value2 date,
Value3 varchar(10)
);

And now Oracle Function with signature like this:

CREATE OR REPLACE FUNCTION Bar (id int)
return Foo pipelined

This function returns a Foo object and gets the id key from a table VT (some table). The 3 values from Foo object should extend an NHibernate VT object. So I want to have 3 new properties in the NHibernate entity.

Sound easy? Well it wasn’t for me. At first how to get the values? You can get an instance of IDBConnection down from NHibernate context and fight your way with ordinary ADO.NET (which is not so easy for functions with user defined return values – at least I don’t know how to do this). But you can use a TABLE() function of Oracle. You can issue this SQL select statement:

select * from Table(Bar(1))

It will return a result set as a table you can select from. Cool stuff! But to use it you have to know it exists ;)

Ok, now the easy part. I though. I define a property in NHibernate with a formula attribute and I will map this on a property in my class. Something like this:

<property
  name=“Value1”
  type=“Int64”
  formula=“(select Value1 from Table(Bar(id)))“>
</property>

Not so easy! If you do this you will get the Oracle exception. NHibernate thinks that Value1 from your formula is a part of NHibernate object and gives it a suffix. It is transformed to something like that:

select _vt0.Value1 from Table(Bar(4711))

_vt0 being the object alias. It is obviously not the case. There is no Value1 field physically in the database. There is a custom made property with SQL formula in NHibernate. As you can se the id was translated to actual key in the table.

The general rule for the formula attribute is everything tat NHibernate does not understand is send direct to database as is. The trick was to make NHibernate think it does not understands the statement completely.

The answer is to enclose Value1 with quotation marks. To achieve this you will have to use the XML escape marks. Just like this:

    <property
      name=“Value1”
      type=“Int64”
      formula=“(select “Value1” from Table(Bar(id)))“>

A lot of tiny puzzles to put together. But it works!

I’ve once written an article for Software Developers Journal about Continuous Integration in .NET. It was very well received. I always thought I have a lot more to say in this topic. The idea to write a book came while speaking with my good friend hsd (well I have more and more to thank you for :). I’ve tried to propose the book to a polish publisher Helion. Very fortunately for me they said there is not enough place on the polish marketplace to get enough sell (I’ve blogged about this in polish). I presume they know better. I’ve already buried the idea about the book when I saw the ASP.NET MVC 1.0 Quickly book from Packt Publication. It was written by a guy from Belgium (Maarten Balliauw). I’ve read his blog and I realized has not native English speaker. “If he can, why not to try myself?” I thought and written a short proposal for Packt. They were interested! I’ve worked on the proposal with the Acquisition Editor and I’ve shortly after that I’ve got the information that the book was “committed”. I was very, very happy until something went wrong at Packt (the went scared by a download stats of one of the tools I wanted to write about). The book was “uncommitted” and we didn’t sign the agreement. I don’t throw my hands up! I thought if they are interested probably someone else will be too. I’ve send view more proposals to other publishers and the Manning appeared. In the mean time Packt turned to be after all interested in me writing the book. The Acquisition Editor at Packt Publishing was a great guy. I’m sure it would be great to write a book for them but it was to late. Manning found me a co-writer Craig Berntson. Someone that speaks English a lot better than I and has a know how in continuous integration. The way the Manning deals with his authors seems very provisional and I’ve very glad to get this opportunity.

So be aware. The best Continuous Integration in .NET book comes sooner than you expected!

public class TestModel
{
    public double NumberField
    {
        get;
        set;
    }

    public double CurrencyField
    {
        get;
        set;
    }
}

The MVC view is strongly typed with this TestModel. And the view model value is formatted like this:

<%=Html.TextBox(“NumberField”, Model.NumberField)%>

<%=Html.TextBox(“CurrencyField”, Model.CurrencyField.ToString(“c”))%>

If you set the current culture to German-Swiss, for example in (base)controller like that:

protected override void Initialize(System.Web.Routing.RequestContext requestContext)
{
    base.Initialize(requestContext);

    string culture = “de-CH”;

    CultureInfo ci = CultureInfo.GetCultureInfo(culture);

    Thread.CurrentThread.CurrentCulture = ci;
    Thread.CurrentThread.CurrentUICulture = ci;
}

Done so we will get a text boxes looking like this:

What will happen if you use a default binder to get this value? The default binder will try to parse the string as a double value and get an ModelState error. The string does not represent a double value.

To deal with this issue we have to write our own model binder. This can be done by implementing the IModelBinder. But I don’t want to reimplement this how DefaultModelBinder reads other values than my currency doubles. So I though I will set a special attribute to the “special” properties. Something like this:

[AttributeUsage(AttributeTargets.Property)]
public class CurrencyAttribute : Attribute
{
}

And my object:

public class TestModel
{
    public double NumberField
    {
        get;
        set;
    }

    [Currency]
    public double CurrencyField
    {
        get;
        set;
    }
}

And now we can overload the DefaultModelBinder and implement our TestModelBinder like this:

public class TestModelBinder : DefaultModelBinder
    {
        public override object BindModel(ControllerContext controllerContext,
            ModelBindingContext bindingContext)
        {
            object bindingObject = base.BindModel(controllerContext,
                bindingContext);

            foreach (System.Reflection.PropertyInfo propInfo
                in bindingObject.GetType().GetProperties())
            {
                object[] attributes =
                    propInfo.GetCustomAttributes(typeof(CurrencyAttribute), false);

                foreach (object attribute in attributes)
                {
                    CurrencyAttribute currAtt = attribute as CurrencyAttribute;

                    if (currAtt != null)
                    {
                        bindingContext.ModelState[propInfo.Name].Errors.Clear();

                        string attempted =
                            bindingContext.ValueProvider[propInfo.Name].AttemptedValue;
                        CultureInfo ci =
                            bindingContext.ValueProvider[propInfo.Name].Culture;

                        propInfo.SetValue(
                            bindingObject,
                            double.Parse(attempted, NumberStyles.Currency, ci),
                            null);

                    }
                }
            }

            return bindingObject;
        }
    }

And now we have to set the attribute to our object to tell MVC that it has to use your binder.

[ModelBinder(typeof(TestModelBinder))]
public class TestModel
{
    public double NumberField
    {
        get;
        set;
    }

    [Currency]
    public double CurrencyField
    {
        get;
        set;
    }
}

Done! Although there are a view issues with this solution (if you are using Entity Framework you cannot set custom attributes, and in the binder above you will have to implement checking if the field is present on the view), but you get the general idea!

    <h2><%= Html.Encode(ViewData[“Message”]) %></h2>

    <script language=”javascript” type=”text/javascript”>
        function ChangeFormAction(sender, url) {
            sender.form.action = url;
        }
    </script>

    <form method=”post”>
        <input id=”text” name=”text” type=”text” value=”Hello from Action” />
        <br />
        <input type=”submit” value=”Go to action 1″
            onclick=”ChangeFormAction(this, ‘/Home/Action1′)” />
        <input type=”submit” value=”Go to action 2″
            onclick=”ChangeFormAction(this, ‘/Home/Action2′)” />
    </form>

And the actions that responds to this form are here:

public class HomeController : Controller
{
    public ActionResult Action1(string text)
    {
        ViewData[“Message”] = text + “1″;

        return View(“Index”);
    }

    public ActionResult Action2(string text)
    {
        ViewData[“Message”] = text + “2″;

        return View(“Index”);
    }
}

Nice!

But be aware! In such scenario (and, well… in any other scenario as well) you have to keep your lib up to date. Take a look at this nasty bug. I was using the NUnit version 2.4.6.0

Take a look at this test method:

[NUnit.Framework.Test]
public void Test()
{
    System.Text.ASCIIEncoding ASCIIEncoding = new System.Text.ASCIIEncoding();

    string str1 = “abc”;

    byte[] array = ASCIIEncoding.GetBytes(str1.ToCharArray());

    System.Array.Resize<byte>(ref array, 10);

    string str2 = ASCIIEncoding.GetString(array);

    NUnit.Framework.Assert.AreEqual(str1, str2);
}

What would you expect?

Fail! Isn’t it?

No! 1 passed, 0 failed, 0 skipped, took 0,54 seconds.

Even NUnit hast bugs to! I’ve updated to the newest version and it works as a charm! So keep you lib folder up to date!

Red (cicstarter) – Schell command name

Blue (startf) – Action name

Brown (cicvlm.exe f=\”%1\”) – Application path (parameter inside f=)

Green (.cic) – File extension

You can manipulate the colored parts to feet your needs. Here is the script. Good luck!

If you are using SQL Server its easy. You start the SQL Profiler and you are done. If you are using Oracle the task is not so easy. You can:

1. start the trace on the Oracle server and get the huge barely readable text file physically on the server and use TKPROF

2. look up the lat SQL query from session using the Enterprise Manager-Konsole

3. buy a 3rd party tool like FlexTracer to trace the OCI communication.

Well yeah! Not fun.

But there is another option inside Entity Framework itself! It is not quite obvious, so it took me a little to figure it out:

((System.Data.Objects.ObjectQuery)Query).ToTraceString()

where Query is something like System.Linq.IQueryable<T>.

public class MyClass
{
    public static void Main()
    {
        System.Console.WriteLine(“Foo”);
    }
}

instead of

using System;
public class MyClass
{
    public static void Main()
    {
        Console.WriteLine(“Foo”);
    }
}

Unfortunately over time it proofed no value to me. I was not happy with this rule, but I thought I’ll consult the brother audience for opinion before I start a holly war against something that I might not fully understand.

I decided to give stackoverflow.com a try. It is quite new Q&A Service dedicated specially for software developers. I write my question I was given a flood of answers. Some better, some worse. Some hitting the bottom line, some pouring “divine knowledge” of individuals that think they know things better.

Nevertheless the discussion on stackoverflow.com helped my back my opinion. Give it a try!

Microsoft Technology Summit 2008 (8th and 9th November), the biggest  Microsoft technology event in Poland went quite well. It was my first MTS and I was trying not to expect anything special. And indeed I haven’t experienced anything  earth shaking but altogether it was quite interesting and a good investment of time. I surely learned a few new things and meet interesting people there. I was astonished by the sheer number of attendees that gathered in there. I presume the old communist “castle” in the center of Warsaw the “Palace of Culture and Science” was the only place to accommodate this whole crowd, but it is not a cushy place. It’s big and cold and far from perfect to host a technology event like that. And well… despite the overwhelming size of this place, you almost have to eat you lunch on the floor, because there was no enough tables ;-)

But I passed the 70-543 Exam there and gained the Microsoft Certified Professional Developer in ASP.NET title. So altogether it was a full success!

int? i = null;
int? j = 0;

Is null smaller than 0

Console.WriteLine(i < j);

False – no it is not.

So probably null is greater than 0

Console.WriteLine(i > j);

False – no it is not greater as well.

All right! So null is equal 0. It has to be, JIT has no other choice, right?

Console.WriteLine(i == j);

Well False too! This two little fellows are not equal too. What? Is it raining frogs and we about to experience Armageddon?

No! We have to use Nullable.Compare() and we will by back in normal world:

switch (Nullable.Compare(i, j))
        {
               case -1:
                  Console.WriteLine(”i < j“);
              break;
               case 1:
                  Console.WriteLine(”i > j“);
              break;
               case 0:
                  Console.WriteLine(”i == j“);
              break;
        }

0 is a little more than null. null is less than 0. null equals null and 0 equals 0.

Uff!

using System;
using System.Collections.Generic;

public class MyClass
{
    public enum FooBar
    {
        Foo,
        Bar,
        FooBar
    }

    public static void Main()
    {
        WL(FooBarFactory(”Foo“));
        RL();
    }

    public static FooBar FooBarFactory(string init)
    {
        try
        {
            return (FooBar)System.Enum.Parse(typeof(FooBar), init);
        }
        catch
        {
            // Do something
            throw;
        }
    }

    #region Helper methods

    private static void WL(object text, params object[] args)
    {
        Console.WriteLine(text.ToString(), args);
    }

    private static void RL()
    {
        Console.ReadLine();
    }

    private static void Break()
    {
        System.Diagnostics.Debugger.Break();
    }

    #endregion
}

string s = “abcd“;

Its something like this:

If we try to interpret this as a uint variable:

uint ui = (uint)((s[0] << 24) | (s[1] << 16) | (s[2] << 8) | s[3]);

we will get something like this number 1633837924.

Lets look at this from the other side. Our string goes direct into memory:

Using Win32 compiler we want to read it to a ulong variable x from a px address. In Clarion the code to do this will look like this:

x &ulong

long px

x &= (px)

What do we get in x?

1684234849

Hmm, variable ui in C# and x in Wind32 are not the same. Why? Because Intel is using a small endian way to store the numbers. So it means that the highest byte is saved at the right side and the lowest at the left. Like this (small endian):

and not (big endian)

* In Jonathan Swift’s Gulliver’s Travels the Little Endians broke their eggs at the small end, where the Big Endians broke theirs at the large end. And the ones were not very fond of the others ;-)

We work with Visual Studio as our main development environment and Subversion as our source control system. What we needed was a systematic approach where we share our framework across all of our projects. We are actively developing our framework so simple dll reference is not an option. Our rules according to framework development make it fairly save to share the same code across multiple projects and to develop at the same time.

We decided that the framework should by placed in separate SVN repository and a separate Visual Studio project. But we wanted the framework and the project being pullable from source control as a whole. Firstly because we are (I’m!) lazy and I don’t want to remember that I have to check out the framework and than the project separately. Secondly we have a living Continuous Integration system running on CruiseControl.NET and it should by able to detect the changes and build to project in one shot.

Our way is to add the framework as an existing project from separate solutions…

and to connect the main SVN project repository with framework repository as a svn:external.

Step by step solution:

1. Create “Framework” Project in Visual Studio

2. Create a SVN repository for this project (check everything you need in)

3. Create new solution in Visual Studio that will hold your “Project”

4. Create a SVN repository for this project

5. Create new folder on hard drive inside the “Project” folder

6. Check out the framework from repository into newly created folder

7. In Visual Studio add the framework to you solution as a existing project.

8. Link the Framework repository with the Project repository (using svn propset or with TortoiseSVN like this)

You are done!

Now if you check out the “Project” you will get the “Framework” automatically. Unfortunately if you will make changes into “Framework” you will not by able to checkc everything in. You will have to separately check the “Framework” changes and “Project” changes. But if you are using Tortoise SVN you will by warned.

Is is cool or not!?