Luc Pattyn wrote: So what is the problem ?? Error in my post: It should be: "If the program is run in release mode, it prints out "NextDouble: Not infinity"!!". I modified this. Luc Pattyn wrote: And did you ever care to print out range (that is, if it ever goes wrong) ? Yes I did. The last line in NextDouble prints out range (if uncommented of course). It always gives that range is positive infinity and as I said in this case the program works as expected. "Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning." - Rick Cook www.troschuetz.de

What version of the framework is installed on your computer? Any language packs? "Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning." - Rick Cook www.troschuetz.de

Stefan Troschütz wrote: tried the first option before posting here, but it doesn't work (get a compiler error that the tag prefix asp is unknown). Don't worry about this error, it still works just fine, make sure that you don't have any other error when the build is OK when you compile your web site.

You're right. Thanks for the advice :) www.troschuetz.de

You need to include a copy of the licence when distributing the source code. (ianal) And we are happy if you add a licence notice to your article - It's a pain to find n article claiming to be "free", and ffinding out it's (L)GPL'd after digging deeper into it. Waiting for your article :) Pandoras Gift #44: Hope. The one that keeps you on suffering. boost your code || Fold With Us! || sighist | doxygen

He has all rights on this directory. In fact if he hasn't enough rights the whole thing wouldn't work when normally closing the application. I think the whole mess has something to do with the uncorrect initialization of the "csc.exe". But I don't know why this happens and how to prevent it? www.troschuetz.de

The possibility with the command line parameter came to my mind too shortly after posting. Guess I will use it, cause the TickCount may be a bit imprecise :). THX www.troschuetz.de

To learn what the FCL has - browse through the class library documentation. You don't have to memorize everything, but at least understand what's there and were. As you develop .NET applications and libraries, you should also see a pattern that Microsoft uses to try to maintain consistency. Like for provider classes: you'll typically see a static Create or CreateInstance method. If you want to serialize your class in such a way, then you'll have to do it yourself. But take a lesson from both XML and runtime serialization. A good solution would be to create a new attribute - say XmlCommentAttribute - that stores the comment for a field: [AttributeUsage(AttributeTargets.Property | AttributeTargets.Field)] public class XmlCommentAttribute : System.Attribute { private string comment; public XmlCommentAttribute(string comment) { this.comment = comment; } public string Comment { get { return this.comment; } } } This would allow you to declaritivly define the comments for your fields: [XmlRoot("myClass", Namespace="urn:myClass"] public class MyClass { [XmlComment("This is parsed as a string.")]private string field1; [XmlElement("myField2"), XmlComment("This is an int.")] private int field2; // ... } Noticed how I used the XML serialization attributes in places? This allows you to easily rename members or declare how things are stored, whether they should be elements or attributes, etc. If they aren't, you can have your serialization routine default to one or the other. Then just do something like this: FieldInfo[] fields = myClass1.GetType().GetFields( BindingFlags.Instance | BindingFlags.NonPublic | BindingFlags.Public); foreach (FieldInfo field in fields) { if (!field.IsNotSerialized) // At least honor the NotSerializedAttribute { string name = field.Name; bool useElement = true; string comment = null;   // Get optional element name from custom attributes. XmlElementAttribute[] elemAttrs = (XmlElementAttribute[]) field.GetCustomAttributes(typeof(XmlElementAttribute)) if (elemAttrs != null && elemAttrs.Length > 0) name = elemAttrs[0].ElementName; else { // Get optional attribute name from custom attributes. XmlAttributeAttribute[] attrAttrs = (XmlAttributeAttribute[]) field.GetCustomAttributes(typeof(XmlAttributeAttribute)); if (attrAttrs != null && attrAttrs.Length > 0) { useElement = false

I believe it was VS.NET 2003 that finally fixed a lot of the bugs in Microsofts C++ compiler. If you're passing a reference to a reference, just change the code to not do that. :-) Christian I have drunk the cool-aid and found it wan and bitter. - Chris Maunder

There are several ways. One is through markup like with HTML that uses the <sup> and <sub> tags. I doubt that's appropriate here, however. Another is to get a font that is full of these. It's not uncommon to have a font represent symbols. Microsoft uses several fonts for glyphs in Outlook and other applications. Since .NET supports Unicode, there is the superscript and subscript diacritics in the U+2070 through U+209F range. The problem is that you need a font that supports them. Tahoma (default font) and Arial Unicode MS currently do not, nor do any of the other common ones. If you had a font that supported this Unicode range and displayed the text in a TextBox (for example), then you could use: "n\u2081 + n\u2082 =" Finally, about your only other option is to owner-draw the text yourself in parts using Graphics.DrawString and either off-setting the coordinates of the upper-left corner or using a StringFormat with a LineAlignment set accordingly. Microsoft MVP, Visual C# My Articles

If you can't dictate how the algorithm is developed, don't fret. You have several options: create a MC++ assembly that uses the C/C++ APIs even in a __gc class (managed class declaration). Compile it as an assembly and you'll have no problems. If the algorithm is comprised of only a few exported functions (hopefully C-style, but C++ name mangling is not really a problem either), then consider writing your wrapper component in C# and just P/Invoke the functions you need, correctly defining any structs and consts you might need to marshal. Microsoft MVP, Visual C# My Articles

There is a sample code in C++ in MSDN,youu can use thos API,search this topic in MSDN: HOWTO: Ejecting Removable Media in Windows NT/Windows 2000 Mazy "A bank is a place that will lend you money if you can prove that you don't need it." - Bob Hope

Thanks for info! I'm now using this code in my server application and it functions very well. this.remoteObject = new RemoteObject(); System.Runtime.Remoting.RemotingServices.Marshal(this.remoteObject, "Server"); System.Runtime.Remoting.RemotingServices.Disconnect(this.remoteObject);

To expand on what Niels said, you'll want to P/Invoke that API: [DllImport("user32.dll")] private static extern IntPtr SendMessage( IntPtr hWnd, uint msg, IntPtr wParam, IntPtr lParam); There are some shortcuts, too. If a param takes an address of a struct, rather than make several unsafe calls from the System.Runtime.InteropServices namespace, you can instead overload SendMessage. For instance, the TVM_SETITEM message for a TreeView Common Control (which System.Windows.Forms.TreeView encapsulates) requires the address of an `LPTVITEM` for the `lParam`. You could overload `SendMessage` like so: [DllImport("user32.dll")] private static extern IntPtr SendMessage) IntPtr hWnd, uint msg, IntPtr wParam, ref LVITEM lvItem); You'll still have to define the `LVITEM` struct, but this is pretty easy to do. See more information about that in the `StructLayoutAttribute` documentation. -----BEGIN GEEK CODE BLOCK----- [Version: 3.21](http://www.geekcode.com/geek.html) GCS/G/MU d- s: a- C++++ UL@ P++(+++) L+(--) E--- W+++ N++ o+ K? w++++ O- M(+) V? PS-- PE Y++ PGP++ t++@ 5 X+++ R+@ tv+ b(-)>b++ DI++++ D+ G e++>+++ h---* r+++ y+++ -----END GEEK CODE BLOCK-----