|
It's not a memory leak it's just that the GC hasn't reclaimed the memory the form that gone out of scope.
Here's a test you can do... Instantiate and destroy the offending form about 10x's, then call GC.Collect() and see what happens.
|
|
|
|
|
It's recommended that you don't call GC.Collect except in dire circumstances. For any class that implements IDisposable - like many of the Windows Forms classes including derivatives of Control - you should call Dispose when done with the class instance. The using block statement in C# is helpful in this way.
This will release the native resources immediately instead of waiting of the object to be finalized which happens when the application is idle or memory is near exhaustion.
This posting is provided "AS IS" with no warranties, and confers no rights.
Software Design Engineer
Developer Division Sustained Engineering
Microsoft
[My Articles] [My Blog]
|
|
|
|
|
You may also want to refer to this CP article[^] for a brief discussion of how and when to implement your own Dispose method - particularly with respect to suppressing finalization when Dispose is used.
The most exciting phrase to hear in science, the one that heralds the most discoveries, is not 'Eureka!' ('I found it!') but 'That's funny...’
|
|
|
|
|
Don't tell me - tell the other poster. I'm well aware of the disposable pattern[^] as Microsoft defines and implements it.
This posting is provided "AS IS" with no warranties, and confers no rights.
Software Design Engineer
Developer Division Sustained Engineering
Microsoft
[My Articles] [My Blog]
|
|
|
|
|
My comment was associated with the context of yours, so the reply was attached under yours. I see no reason to create a peer comment, then reference your comment to achieve the same thing.
And yes, I'm positive you know all about implementing Dispose.
The most exciting phrase to hear in science, the one that heralds the most discoveries, is not 'Eureka!' ('I found it!') but 'That's funny...’
|
|
|
|
|
Thanks. I'll be trying it out tonite.
"Being smart is no good until you find out who else is smarter?"
|
|
|
|
|
Hi
I wrote a C# application that export grid data to an excel workbook. The application works fine when I run it on my local machine and fail with "System.Security.SecurityException" error if I try to lunch it from a network drive folder.
How do I solve the problem? Is there easy to follow reading material about this on the web? most of my search are not getting me anywhere.
Thanks
|
|
|
|
|
The error you're encountering is most likely arising because you don't have launch permission for Excel on the remote machine.
Since you're executing your app from a remote share, when you attempt to launch Excel (unless you've done something to avoid it) you're launching it on the remote machine. If that's really what you are aiming for, you should be able to set launch permissions for Microsoft Excel by running dcomcnfg.exe.
Best of luck.
The most exciting phrase to hear in science, the one that heralds the most discoveries, is not 'Eureka!' ('I found it!') but 'That's funny...’
|
|
|
|
|
None of that is true. First of all, the SecurityException is thrown because the permissions granted to managed code for a network address are few.
Second, when you launch an application from a remote machine the application files are loaded into memory on your local machine unless you're using Terminal Services (or something similar, like VNC).
This posting is provided "AS IS" with no warranties, and confers no rights.
Software Design Engineer
Developer Division Sustained Engineering
Microsoft
[My Articles] [My Blog]
|
|
|
|
|
The previous post is incorrect. The problem you're having has to do with permissions granted to your managed code. Code Access Security (CAS) is a mechanism that grants permissions to managed code based on evidence. For managed code running over the LAN (Intranet), the default permissions are few. For code run over the Internet, the permissions are even tighter (in .NET 1.0 such code gets no permissions to run).
You need to grant your code based on evidence - like a UrlMembershipCondition using "file://SERVER/SHARE" - permissions to run.
Read Understanding .NET Code Access Security[^] here on Code Project for more information.
This posting is provided "AS IS" with no warranties, and confers no rights.
Software Design Engineer
Developer Division Sustained Engineering
Microsoft
[My Articles] [My Blog]
|
|
|
|
|
I have an integer, say 0x819A, which represents the Japanese shift-jis coded star character. How do I convert this to the System.Char? (that is the unicode encoding)
Best regards,
Paul.
Jesus Christ is LOVE! Please tell somebody.
|
|
|
|
|
If you want to specify the character in code, use the \uXXXX notation in C# like string star = "\u819a";. As you should notice, the character format uses a hexadecimal representation.
If you want to convert any bytes to a particular encoding, you use the System.Text.Encoding class that's appropriate, like the UnicodeEncoding (which you can get from Encoding.Unicode as well).
Since a char in managed code is actually a 2-byte character (wide character), you can simply cast it: char c = (char)0x819a;.
If you want to convert arbitrary characters you can also convert the 2-byte integer (short, or Int16) to a byte[] array and use UnicodeEncoding.GetString to decode the character:
byte[] buf = BitConverter.GetBytes(0x819a);
string s = Encoding.Unicode.GetString(buf); But there's an easier way than that, too, using the BitConverter class:
char c = BitConverter.ToChar(0x819a); The important thing is to understand just what a char is in native or manage code and about encodings.
This posting is provided "AS IS" with no warranties, and confers no rights.
Software Design Engineer
Developer Division Sustained Engineering
Microsoft
[My Articles] [My Blog]
|
|
|
|
|
Hello Heath,
Thanks for the response. It is a good attempt, but I have played with all these already 
The main point in my post is that the number is the shift-jis coded.
What works is Chr in the Microsoft.VisualBasic.dll. You just wanna be a VB.NET programmer for after wasting half a day on C#.
Well, with Reflector, I was able to get the C# version of Chr to continue my work 
Best regards,
Paul.
Jesus Christ is LOVE! Please tell somebody.
|
|
|
|
|
If it uses a different encoding then create that instance of the Encoding you need. Taking a dependency on another assembly just to use Chr is absurd. Are you now going to bloat your software distribution and force that assembly on your clients?
It all comes down to encodings. If the character is encoded differently than UTF8 or Unicode, then use the right encoding:
Encoding enc = Encoding.GetEncoding(932);
string s = enc.GetString(BitConverter.GetBytes(0x819a));
This posting is provided "AS IS" with no warranties, and confers no rights.
Software Design Engineer
Developer Division Sustained Engineering
Microsoft
[My Articles] [My Blog]
|
|
|
|
|
Hello Heath,
Thanks for the support.
Heath Stewart wrote:
Taking a dependency on another assembly just to use Chr is absurd. Are you now going to bloat your software distribution and force that assembly on your clients?
You really believe I will be doing that? Chr is a static and almost independent.
Heath Stewart wrote:
It all comes down to encodings. If the character is encoded differently than UTF8 or Unicode, then use the right encoding:
If you are using an international version of XP, just set the font of the Form to "MS Mincho" or similar with the character set to Japanese and view the output of your code snip in a paint handler as:
e.Graphics.DrawString(s, this.Font, Brushes.Red, 20, 20);
Best regards,
Paul.
Jesus Christ is LOVE! Please tell somebody.
|
|
|
|
|
Paul Selormey wrote:
Chr is a static and almost independent.
It doesn't matter if it's static. There is no "static linking" like with native code. Or are you saying that you'll just duplicate what it does. I've done this for an old project for a GetDeviceCaps wrapper.
The character encoding is a different issue than the font issue, which I didn't realize you were having problems. Yes, a font file must have defined a glyph for that font, but how the character is encoded is a different matter. When you decode it you have a char. The cast I showed you initially would've worked had the character been ASCII (7-bit) or Unicode (native to .NET) encoded. Characters in .NET are all stored as Unicode, so you just need to make sure you decode it from the proper encoding. Whatever font you use should be based on what characters you expect.
In Windows XP several of them have a wide range of Unicode characters. Font files, BTW, are typically encoded for a particular encoding. For the Unicode font files, they are coded for Unicode encoded characters, of course.
This posting is provided "AS IS" with no warranties, and confers no rights.
Software Design Engineer
Developer Division Sustained Engineering
Microsoft
[My Articles] [My Blog]
|
|
|
|
|
Heath Stewart wrote:
Or are you saying that you'll just duplicate what it does. I've done this for an old project for a GetDeviceCaps wrapper.
That is the simply story.
public static char Chr(int CharCode)
{
if ((CharCode < -32768) || (CharCode > 0xffff))
{
throw new ArgumentException();
}
if ((CharCode >= 0) && (CharCode <= 0x7f))
{
return Convert.ToChar(CharCode);
}
int nTemp;
Encoding enc = Encoding.GetEncoding(932);
if ((enc.GetMaxByteCount(1) == 1) && ((CharCode < 0) || (CharCode > 0xff)))
{
throw new InvalidOperationException();
}
char[] chTemp = new char[2];
byte[] buffer = new byte[2];
Decoder decoder = enc.GetDecoder();
if ((CharCode >= 0) && (CharCode <= 0xff))
{
buffer[0] = (byte) (CharCode & 0xff);
nTemp = decoder.GetChars(buffer, 0, 1, chTemp, 0);
}
else
{
buffer[0] = (byte) ((CharCode & 65280) / 0x100);
buffer[1] = (byte) (CharCode & 0xff);
nTemp = decoder.GetChars(buffer, 0, 2, chTemp, 0);
}
return chTemp[0];
}
Best regards,
Paul.
Jesus Christ is LOVE! Please tell somebody.
|
|
|
|
|
Hello,
I would like to do the following thing: Suppose I have a .Net program running. From another program I want to be able to get access to the first program's instantiated objects. My utlimate goal is to have something like a real time debugger where I can see the actual object's properties and fields values, without having to pause the program and use the VS debugger. Right now I have done something that can do exactly what I describe here, however from the program I want to debug I need to give reference to the objects I want to analyse to my real time debugger.
I realy want to achieve the same thing with my real time debugger but without having to have a link between the debugge and and the debugger.
I have investigated the following:
- Use Process.GetProcesses() then get a module from the process with: ProcessModuleCollection modules = process.Modules.
From there I don't know where to go.... 
Any Ideas ?
- Dany McCarthy -
|
|
|
|
|
A process is not the same as an AppDomain, in which the managed objects are actually hosted. For this you need to implement .NET Remoting, which you can read about at http://msdn.microsoft.com/library/en-us/cpguide/html/cpconaccessingobjectsinotherapplicationdomainsusingnetremoting.asp[^]. Just having a Process instance (in the native world, this is a process handle, or HANDLE) won't help you acheive that end.
If it's a debugger or debugger-like application your after, then write one. It's actually not too difficult if all you want to do is examine objects. View the search results for managed debugger[^] for a large list of examples about how to implement a debugger, as well as some blog entries from developers on the CLR team.
This will require native code and can tell a lot more about an object than what you can get through managed code, even with .NET Remoting - which would require that you establish a remoting infrastructure for both applications.
You can use other IPC (inter-process communications) mechanisms to communicate between applications, but .NET Remoting is the recommend way, and is the best way to marshal managed objects across application boundaries, rather than marshaling memory through global locks and other ways that you'd have to do in managed code. Keep in mind that what managed applications see is much different than what's actually on the heap or in the stack for managed objects. That managed code is JIT'd to native code, which works much differently under the covers.
This posting is provided "AS IS" with no warranties, and confers no rights.
Software Design Engineer
Developer Division Sustained Engineering
Microsoft
[My Articles] [My Blog]
|
|
|
|
|
Thanks Heath, for sure I do not want to go through the unmanage solution. I think I will keep going with what I've done so far. I already planned to implement remoting to do real time remote debugging. I will send an article about my results if they are concluant.
Best regards,
Dany McCarthy
|
|
|
|
|
I have to do a stock market demstration for class, and I am fairly new to C#(have made one little guessing game) and i need some help. I would like for the user to start with some money, purchase stocks of a few possible companies, be able to sell them, and then give a total score at the end of 30 days in the game. It would be random, so you never know what could happen. Anyway to do this simply?
Doesn't have to be pretty.
(Im only 13 so i dont know that much)
|
|
|
|
|
Well, the easiest way is to take the problem and break it into smaller problems, then work on the smaller problems. If you still can't work it out then perhaps the problem is still too big and needs to be broken down further.
So your user has:
money
choice of stocks
ownership of stock
The program runs for 30 cycles (days). In each cycle the user has the option to buy or sell stock. (I suggest allowing the user to sell first, before buying). At the end of each cycle the each stock price changes by a random amount (to simulate market movements)
At the start of your program you need to set up the variables (money, stocks, and owenership).
You may wish to look at NameObjectCollection[^] so you can create a look up based on the stock name. So you have a lookup of stocks (the key is the stock name) with the object being the current stock price. The same goes for ownership (the key is the stock name, with the object being the current quantity of stock owned) [Is this making any sense? If I'm talking at the wrong level let me know]
Each cycle you get the user input for the stocks they wish to sell. You adjust the ownership list for these stocks by the amount of stock they sell and credit the money they have with the sale price.
You then allow the user to buy stock. Add the stock to ownership list, and deduct the amout of money it cost from their account.
At the end of the program you can list how much money the user has, and how much of each stock (and the value of each, if you want).
I hope this has helped. It may help if you write out some pseudo code (a kind of English-like code that reads like bullet points on paper) that can help to organise thoughts better. Like this:
--Initial setup
* Set money to $1000
* Set stock_price item "MSFT" to $25
* Set stock_price item "RYA" to $36
* Set stock_price item "HBOS" to $12
* Set ownership item "MSFT" to $0
* Set ownership item "RYA" to $0
* Set ownership item "HBOS" to $0
-- Main loop
* Loop 30 times
-- Sell Stock
** Ask for stock to sell
** Get stock_name as input
** Get quantity as input
** Subtract quantity from ownership item [stock_name]
** Add quantity * stock_price [stock_name] to money
-- Buy Stock
** Ask for stock to buy
** Get stock_name as input
** Get quantity as input
** Add quantity from ownership item [stock_name]
** Subtract quantity * stock_price [stock_name] from money
-- Update stock prices
** Change stock_price "MSFT" by random amount
** Change stock_price "RYA" by random amount
** Change stock_price "HBOS" by random amount
* End loop
* Display results
You can use what ever notation works for you. What ever way helps you organise your thoughts...
Does this help?
Do you want to know more?
WDevs.com - Open Source Code Hosting, Blogs, FTP, Mail and Forums
Upcoming talk:
SELECT UserName, Password FROM Users
-- Getting unauthorised access to a SQL Server, and how to prevent it.
|
|
|
|
|
I kind of get it, but how would i stop the loop to allow selling/buying?
I am a little confused.
|
|
|
|
|
Code Project lost my original post so I'll try again.
To loop a set number of times use a for loop like this:
for(int i=0; i<30; i++)
{
}I'll break each section down so you can see what it does.
int i=0; This sets up a variable i which we use to count the iterations (i.e. the number of times the loop will repeat itself) and we set that variable to zero to start with (you can set it to any number you like depending on your circumstances, but because our problem is simple zero will do just fine.
i<30; This is the condition that is used to determine whether the loop is to keep repeating. This says that while i is less than 30 then repeat.
i++ This is what happens to i at the end of each iteration. In this case we add one (++ is the increment operator). You can specify anything you like depending on your situation, however since this is a simple repeat 30 times we'll just add one.
Does this help?
Do you want to know more?
WDevs.com - Open Source Code Hosting, Blogs, FTP, Mail and Forums
Upcoming talk:
SELECT UserName, Password FROM Users
-- Getting unauthorised access to a SQL Server, and how to prevent it.
|
|
|
|
|
Hi folks:
I have simple question for you: how can I find the position to a newly created record in a DataSet in the corresponding DataGrid?
My DataGrid is binded to a SQL DataSource. Another application creates a new record into the database. I need to refresh the DataSet and, after this, I need to obtain the position of the new record in the DataGrid in order to show to the user the new record.
I assumed it was as easy as to search the new record in the dataset, but the position of the record in the dataset doesn't correspond (necessarily) to the position in the DataGrid because it may have another sort order.
Hope somebody can help with this simple question,
Regards,
Enrique  .
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
