|
dcof wrote: Can you tell me what would be an 'odd' setup in my environment?
Someone messed up the virtual memory setup (they would need to do this explicitly) or it is being run in a situation, such as a hosted environment, that explicitly limits the memory.
dcof wrote: If app1 is using alot of memory before it makes the call, can you tell me how to check on the memory that app1 is using?
Pretty sure there is a C# api that tells you that. Google for it.
|
|
|
|
|
I'm trying to use the child class method instead of the parent class for some child class cases. For some reason, I get to the parent class method, but it never gets to the override in the child. Any ideas why? I've been referring to http://stackoverflow.com/questions/9067885/calling-child-class-method-from-parent[^]
but it's not getting to the child's method.
In my PC.cs of the child class (Ko):
public partial class PC : GenericPC
{
public override void ValidateDynData(TextBox[] tb_dynData_Array, ref int result)
{
Console.WriteLine("Ko PC::ValidateDynData");
}
}
In my GenericPC.cs of the parent class:
private int processDynData()
{
ValidateDynData(itsEnteredDynData, ref returnCode);
}
public virtual void ValidateDynData(TextBox[] tb_dynData_Array, ref int result)
{
Console.WriteLine("Generic::ValidateDynData passing off to child to validate special data");
}
Any ideas why it's not going to the child class implementation of ValidateDynData when I call ValidateDynData in my parent class? This is the only area in my code where I am trying to have a child class override a parent implementation, so I'm not sure if I'm doing something wrong?
|
|
|
|
|
I tried it myself, by pasting your code into an app, and adding a few bits to prevent compilation errors:
private void button2_Click(object sender, EventArgs e)
{
PC pc = new PC();
int i = 4;
pc.ValidateDynData(null, ref i);
}
public partial class PC : GenericPC
{
public override void ValidateDynData(TextBox[] tb_dynData_Array, ref int result)
{
Console.WriteLine("Ko PC::ValidateDynData");
}
}
public class GenericPC
{
TextBox[] itsEnteredDynData;
int returnCode;
private int processDynData()
{
ValidateDynData(itsEnteredDynData, ref returnCode);
return returnCode;
}
public virtual void ValidateDynData(TextBox[] tb_dynData_Array, ref int result)
{
Console.WriteLine("Generic::ValidateDynData passing off to child to validate special data");
}
}Ignoring the "not assigned" warning message, when I press the button I get:
Ko PC::ValidateDynData Which is what I would expect - the overloaded version.
Could it be that you are using an instance declared as the base class, rather than the derived?
GenericPC gpc = new GenericPC();
gpc.ValidateDynData(null, ref i);
would give you
Generic::ValidateDynData passing off to child to validate special data
If you get an email telling you that you can catch Swine Flu from tinned pork then just delete it. It's Spam.
|
|
|
|
|
It's possible that's the problem. My object instance is created with Reflection:
_assembly = Assembly.LoadFrom(programDll);
_type = _assembly.GetType("CrWriter.PC");
_objectInstance = Activator.CreateInstance(_type);
_parameters = new Object[] { cb1, programDll, templateArr, itsDll, cert, i, cb1.cbInfos[i].boxID };
controls[i] = new Control();
controls[i] = (Control)_type.InvokeMember("GetPC", BindingFlags.Default | BindingFlags.InvokeMethod, null, _objectInstance, _parameters);
controls[i].Dock = DockStyle.None;
this.Controls.Add(controls[i]);
Since my Invocation of GetPC lands in the child class and not the parent, it must be created correctly. So I'm not sure why it's not landing in the correct ValidateDynData method. Maybe I need to cast my object to the programDll somehow. When I run the program and inspect the _objectInstance it could be a problem:
variable.............................................................value
base: {GenericCrWriter.GenericPC} ......CrWriter.PC
baseInst: .........................................................GenericCrWriter.GenericPC
But then, the _assembly is referring to Ko/PC and not Generic/GenericPC.
Also, my _assembly.GetType looks good. My Generic/parent doesn't have anything named CrWriter.PC
I wonder if there is a way to cast an object obtained through reflection?
modified 1-Mar-13 15:56pm.
|
|
|
|
|
MichCl wrote: Since my Invocation of GetPC lands in the child class and not the parent, it must be created correctly. So I'm not sure why it's not landing in the correct ValidateDynData method.
Presuming the former is really true then the possibilities are.
- The child signature of ValidateDynData is wrong
- You are not running the code that you think you are (binary does not match source.)
|
|
|
|
|
Do you see any problems with my method signatures?
This is the parent, GenericPC.cs:
public virtual void ValidateDynData(TextBox[] tb_dynData_Array, ref int result)
{
Console.WriteLine("Generic::ValidateDynData passing off to child to validate special dynamic data");
}
This is the child, PC.cs:
public override void ValidateDynData(TextBox[] tb_dynData_Array, ref int result)
{
Console.WriteLine("Ko PC::ValidateDynData");
}
The parent, GenericPC.cs calls the method this way:
ValidateDynData(itsEnteredDynamicData, ref returnCode);
I checked the correct version of Generic.dll is referenced in the child's project/class. I did a clean build in the child class. Any other binaries that should be checked?
|
|
|
|
|
I've been looking into this some more, and I think the reason flow in the parent is not seeing the overridden ValidateDynData in the child is because of the way I get flow into the parent from the child class:
Is there another way to get to the parent's method without creating an instance of the parent?
GenericPC baseInst = new GenericPC();
return baseInst.GetPC(cbInst, dllSel, templ, dll, cert0, slaveIndex, BoxID);
|
|
|
|
|
Is it possible to pass parameters from unmanaged code to managed code by reference?
I am programming agents for the Metta Trader 4 (MT4) platform using MetaQuotes Language 4 (MQL4); a C like language.
Normally one would use C++ to generate DLLs but I would like to use C# if I could because when you release something in C++ it is frozen in time as opposed to when you release something in C#. With C# if problems are detected in the underlieing classes they are corrected by normal updates of the .NET platform. This is an advantage I would like to take advantage of to simplify maintenance of products I hope to sell.
I will need to pass large arrays of market data to these DLLs for analysis and would like to do this by reference.
Here is an example of how to pass an array by value:
[DllExport(EntryPoint = "SumIntegerArray")]
public static int Sum([MarshalAs(UnmanagedType.LPArray, SizeParamIndex = 1)]int[] values, int count)
{
int result = 0;
for (int i = 0; i < count; i++)
result += values[i];
return result;
}
How would I pass the array by reference to the DLL?
|
|
|
|
|
You could pass a pointer and use unsafe on the C# end instead of arrays. AFAIK you can't (easily) pretend that a native array is a managed array.
|
|
|
|
|
Thanks for the information, I will try it. 
I am just unsure how to accomplish it in C#. I didn't know you could use a pointer in managed code.
Off to the articles to see if I can find out how.
|
|
|
|
|
It mostly works as you'd expect, same syntax and all, but you can't take the address of managed objects and you need "unsafe" keywords (and enable unsafe code in the build options).
Also, "int *x, y" makes both of them pointers, not just x.
|
|
|
|
|
Thanks for the information.
|
|
|
|
|
As an alternative, you could put it in a memory mapped file[^]
Bastard Programmer from Hell 
If you can't read my code, try converting it here[^]
|
|
|
|
|
I don't think it would be applicable in my case but I will look into them.
Thanks for the information.
|
|
|
|
|
I want to get the cursor coordinates when the mouse is clicked in inkcanvas.here i used mouse event handler and i am trying to display cordinates in a textbox but somehow its not working.i have used the following code snippet.can someone please help
private void inkCanvas1_MouseLeftButtonDown(object sender, MouseButtonEventArgs e)
{
System.Windows.Point position = e.GetPosition(this);
this.textBox1.Text = position.X.ToString()+","+position.Y.ToString();
}
|
|
|
|
|
I am assuming its winforms
but have a look a
Cursor.Position.X
Further reading System.Windows.Forms.Cursort.Position[^]
Lobster Thermidor aux crevettes with a Mornay sauce, served in a Provençale manner with shallots and aubergines, garnished with truffle pate, brandy and a fried egg on top and Spam - Monty Python Spam Sketch
|
|
|
|
|
|
|
I've inherited a wrapper for a gps interpretter. The guy who wrote it is dispose happy, I'm guess from the old days with .net. I once had to go back and change all the close calls in managed database connections to .dispose(). This was code we were maintaining, but not code that was touched in my update. So I had to go through the whole project and change them to prevent memory leaks that wasn't there?
Anyway, I found the msdn page, after, that said that wasn't necessary. That still wasn't proof enough?
Ok, so this wrapper is a static class. It has a member NmeaInterpreter, and a method dispose. Dispose says :
if( GpsInterpreter.Device != null)
{
GpsInterpreter.Device.Dispose();
}
GpsInterpreter.Dispose();
I don't even think that is necessary. Is it?
If it moves, compile it
|
|
|
|
|
loctrice wrote: I don't even think that is necessary. Is it?
Depends. Not disposing means that it'll be disposed of when the GC runs, which usually happens when the system is low on memory. That might also cause other apps to run into problems when trying to allocate memory.
There's another problem with not disposing objects, if they reference unmanaged resources; a stream or a SqlConnection needs be disposed of. Given a file-stream, the file can be "locked" until the reader is disposed of.
The advantage of disposing aggressively is performance. Still, I don't implement dispose unless I have to.
Bastard Programmer from Hell 
If you can't read my code, try converting it here[^]
|
|
|
|
|
I will leave the dispose in there, just to be on the safe side. It's a mobile device that we are using, so resources are resources.
I always thought, and programmed, that an Sql connection and/or file stream can be .close() 'ed without issue. They don't need both close and dispose called, it's one or the other and close is "good enough"
If it moves, compile it
|
|
|
|
|
loctrice wrote: They don't need both close and dispose called, it's one or the other and close is "good enough"
Dispose would also close it. Closing it will not dispose it, but will return the connection to the pool. This will not work for all classes in the same way though.
Bastard Programmer from Hell
If you can't read my code, try converting it here[^]
|
|
|
|
|
am created sdk for rfid card reader it reads the rfid value and display the output in that sdk screen i want to use the rfid values in my form to use the rfid value
|
|
|
|
|
And? What's your question?
|
|
|
|
|
You don't really know what an SDK is, do you?
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
