Background: earlier today I posted an open-source app, written in C# .NET with WinForms, on github. The first couple of people who downloaded it found that it crashed immediately on startup. Assuming I'd screwed up the distribution somehow, I posted a question here. Some helpful people downloaded the project and quickly determined that the problem was caused by a stack overflow.
So that's resolved, but the question remains: why doesn't this code blow up when I run it?
The specific problem was caused by code that attempted to resize the rightmost column of a details-mode ListView whenever the size of the window or of another column changed. A ColumnWidthChanged event handler was setting the width of a column, resulting in infinite recursion.
On my system, it works just fine (here's video). One possible explanation would be that I have a special version of WinForms (or the underlying control) that doesn't post an event change notification if the column width hasn't changed.
What I really want is to have a high degree of confidence that code that works on my system will work elsewhere. The people who answered my question seemed to have no difficulty reproducing the crash on their development systems. Did I manage to install some sort of "bulletproof edition"?
The app and library projects target either .NET Framework 4.6.1 or .NET Standard 2.0. Changing the app to target 4.7.1 had no effect. I'm using Visual Studio Community 2017, updated through 15.8.3, on a Windows 10 Pro system. This is my first attempt at a desktop C# .NET app, so I may be doing something silly.
Update: inspired by a comment from #jwdonahue I conducted an experiment: in the event handler, grab a StackTrace and check the FrameCount, recording the largest value seen. With one of the column updates enabled, I consistently see a max frame count of 414, with multiple instances of the ColumnWidthChanged event handler present. If I comment out the column update, the max count drops to 174, and I only see one instance of the event.
The stack trace from a failed run forwarded by #LexLi has 5087 entries, so it's not a simple matter of my stack limit being higher.
It might be that on your system it will eventually set the column width to the existing value. If the new value is the same as the current value then it might not trigger the changed event.
Possible reasons are:
The startup window size might be different because of different DPI.
The data might be different causing different widths on different systems.
On the failing systems the width might not converge but alternate between two values.
Theoretically it could depend on the version of the .net framework or OS.
Either way, from a practical standpoint it is best to design your software such that there are no circular calculations.
Related
I am designing a Windows Form app. I have an MDIParent form that loads in a maximized state, and loads its child forms in a maximized state as well. However, when I open an OpenFileDialog, or any datareader object, the MDIParent shrinks to a smaller size with all its forms and controls.
This solution Opening child form is causing mdiform to change size and shrink does not apply/work in my situation.
Also this solution https://support.microsoft.com/en-nz/help/967173/restoring-a-maximized-or-minimized-mdi-parent-form-causes-its-height-t did not work for me.
Some background: I have seen this behaviour in almost all my WinForm applications but I have never been keen to sort it out. I was able to narrow down to the causes as highlighted above when I started investigate it. Some posts are describing it as a windows bug, but it has existed for as long as the screen resolutions started going above 1024 (VS 2010) for my case. I hoped it is not just a windows bug...
I hoped it is not just a windows bug...
Feature, not a bug, but it is not one that Winforms programmers like very much. Notable is that there have been several questions about mystifying window shrinkage in the past few months. I think it is associated with the release of Win10 Fall Creators edition. It has deep changes to the legacy Win32 api layer and they've caused plenty of upheaval.
In your specific case, the "feature" is enabled by a shell extension. They get injected into your process when you use OpenFileDialog. The one that does this is very, very evil and does something that a shell extension must absolutely never do. It calls SetProcessDPIAware(). Notable is that it might have been written in WPF, it has a very sneaky backdoor to declare itself dpiAware. Just loading the PresentationCore assembly is enough. But not otherwise limited to WPF code, any code can do this and that might have been undetected for a long time.
One way to chase down this evil extension is by using SysInternals' AutoRuns utility. It lets you selectively disable extensions. But there is also a programmer's way, you can debug this in VS.
Use Project > Properties > Debug tab > tick the "Enable native code debugging" checkbox. Named slightly different in old VS versions btw. Then Debug > New Breakpoint > Function Breakpoint. Function name = user32!SetProcessDPIAware, Language = C. You can exercise this in a do-nothing WPF app to ensure that everything is set correctly. For completeness you can also add the breakpoint for SetProcessDPIAwareness, the new flavor.
Press F5 to start debugging and trigger the OpenFileDialog.ShowDialog() call. The breakpoint should now hit, use Debug > Windows > Call Stack to look at the stack trace. You typically will not see anything very recognizable in your case since the evil code lives in a DLL that you don't have a PDB for. But the DLL name and location (visible in Debug > Windows > Modules) ought to be helpful to identify the person you need to file a bug with. Uninstall it if you can live without it.
Last but not least, it is getting pretty important to start creating Winforms apps that are dpiAware so such a bug can never byte. You kick this off by declaring your app to be dpiAware so DPI virtualization is disabled. Plus whatever you need to do in your code to ensure the UI design scales properly.
I have a winform within a remote desktop application that is built dynamically with a container panel for a toolbar with large (regular) buttons. The application is writen in C# and .NET 4.0. The application is viewed as it should only if I do not maximize the window (meaning restored view is fine).
Now, after RDP login - within most client computers - the application looks fine (even when we maximize and restoring). Within low resolution client computers (800x600), the buttons on the toolbar are not displayed in the right position (such as in the 1024x768 clients).
I used double buffering and it didn't help... the problem may or may not be depended on the screen resolution (it can be memory issue or something). It is not a problem of localization layout.
The following image shows the Fine and Bad states: http://imgur.com/gO4UJ,v7O9K
What can I do to resolve this issue?
OK, so in case anybody gets to this kind of case, this is the thing that made the issue:
In the form, there was an empty repaint override. The programmer that created the form was overriding the repaint with an empty method (don't know why, I guess I'll ask my team for the business logic).
How did I reproduce it in the development environment? Following Albin Sunnanbo's note, I ran the RDP application in minimized version during initialization, and I noticed that it occurs with large resolution clients also in this use case... So, I know repaint occurs while performing a window resize - and this made me suspicious about the method...
==> On a side note, I guess this is why inheritance is so bad, and going on to WPF is better. It took me 3 days to understand why this issue occurred.
The main form of my application constantly turns white in the designer when I perform the following steps:
1) Open form in design mode (All controls are visible at this point)
2) View source code for form
3) Switch back to design view
After performing the steps above, there is nothing visible in design mode for the form (not even the form). I only have this issue with one form in my application and, unfortunately it is my main form where there is a lot of logic. The workaround is to always close the form and reopen it in design mode show that all the components are visible again. I have experienced erratic errors with the designer when I am low on available memory but, this form always produces this symptom and is the only one. Any ideas as to what is causing this? I guess I could always create a new form and try moving all the controls and logic over but, I'd prefer to avoid the work if there is a simpler option.
The diagnostic is that the Paint event or OnPaint method of a control is misbehaving. These methods run at design time so you'll get an accurate visual representation of the control, the way for example that you can see the Image property of a PictureBox at design time. When such a paint event gets stuck in a loop then the entire form stops rendering properly. Beyond a simple bug, the typical reason is that the code is getting confuzzled by the non-standard runtime environment in design mode. You use the DesignMode property to ensure that such code won't cause trouble and is disabled in design mode.
Finding the misbehaving code is the challenge, especially when these are not controls you wrote yourself. Short from removing controls one by one to find the troublemaker, you can use the debugger by starting another instance of Visual Studio and use Tools + Attach to Process to attach to the first one.
From personal experience, I can confirm that this is an occasional issue in both Visual Studio 2003 & 2005 whether VB or C# is used. We patched both versions to the latest service pack and even got hotfixes directly from Microsoft, neither of which resolved the issue.
In the case of VB in Visual Studio 2003, the disappearance of the controls also removed the underlying designer code so we kept having to restore the deleted code from our version control system. Very annoying - as we'd often lose code changes and have to start over.
You defiantly should check the next link:
https://weblog.west-wind.com/posts/2019/Feb/14/WPF-Hanging-in-Infinite-Rendering-Loop?fbclid=IwAR23ZnUrz7buVpFLXOX2qQin1WcifQ6h280EO25URO74NTGDkTedx1TDRb4
here's a quote from the page:
Using the StarDefinitionsCanExceedAvailableSpace Override This setting
overrides the new GridRendering behavior and basically lets you run
with a .NET 4.7.x target in your project, but keeps the old behavior
that was used in previous versions.
There is a configuration setting that can be set in app.config for
your application:
<configuration>
<runtime>
<AppContextSwitchOverrides value="Switch.System.Windows.Controls.Grid.StarDefinitionsCanExceedAvailableSpace=true"
/>
</runtime>
</configuration>
I can verify that using that switch lets me run 4.7.1 and not see the lock up in any scaling mode. After I
had my running version in 4.6.2 back, I once again moved up to 4.7.1
in a separate branch to try this out and sure enough the switch made
the application run targeted with 4.7.1. So there's a hacky
workaround.
It's a workaround though. This 'fix' according to Microsoft improves
grid rendering allocations, providing more precise sizing and also
improves performance and reduces memory usage. On paper this is a
great improvement, but... well, side effects 😃
I suspect this issue is not wildly common as there was not very much
info to be found about it. I think Markdown Monster makes this issue
come up because the startup sequence has a lot of window manipulation.
MM uses the MahApps UI framework which uses Window Animation and extra
rendering frames for the main window, and MM itself moves the window
offscreen for initial render and adjusts window sizing based on screen
sizes and DPI settings if the window doesn't fit on the screen or
would otherwise be offscreen. IOW, there's a bit of gyration to get
the initial window onto the screen that is more likely to hit this bug
than a simple WPF form.
So I doubt that every application needs to worry about this, but if
you have a 4.7.x WPF app it might be a good idea to try it out at
various resolutions and scale levels just to see how it fares.
I have what seems to be a common problem. I am running Windows 7 Home Premium on one of the most awesomest computers (when it was bought last year) and certain visual effects just automatically turn themselves off.
My average user experience rating is high, so it doesn't explain why this happens. The only feature that ever gets turned off is the 'Show window contents while dragging' option. And it really annoys me.
There are currently no working solutions to this problem online. Other than to "there must be a conflict with another app installed on your machine."
And yes, I do know what app is causing this conflict. It's my bloody Internet Provider's software - you know... that app that you absolutely MUST have open at all times when you're connected to the net.
So, I had a thought. What if I could subscribe to an event so that my app that runs in the background will detect when this 'show window contents while dragging' option is turned off - and then my app will simply turn it back on again.
When I do this manually, it seems to stay in effect for about an hour or two, then it gets switched off again.
Is it possible to handle these types of events, and re-start certain visual effect features? If so, are there any resources on this?
I have not been able to find anything on this sibject yet.
Yes the WM_SETTINGSCHANGE message is sent to all windows when a system setting is changed. Then you can call SystemParametersInfo with SPI_GETDRAGFULLWINDOWS to determine if the "Show window contents while dragging" is disabled and use SPI_SETDRAGFULLWINDOWS to enable it.
So all that you will need to do is create an application with a form (that can even stay hidden) and override the forms WndProc and handle the WM_SETTINGSCHANGE message and call SystemParametersInfo using p/Invoke. The p/Invoke definition for SystemParamtersInfo is available at pinvoke.net
Altough what may be easier is change security on the HKCU\Control Panel\Desktop\DragFullWindows registry value so that it can't be changed.
I have a piece of software that has worked fine on many machines, althoughon one machine there is a problem that appears to occur occasionaly, the MenuStrip and the ToolStrip both appear as a blank white background with a red cross over it, as a custom control would if you created a null object. This doesn't happen whilst I am debugging and I don't know how to make the problem re-occur, but it does happen. I was wondering if anyone knew what could be the problem?
Would it be the version of the .NET framework?
Thanks
This is a common occurrence when there's a GDI+ problem ("The Red X of Death"). Are you doing any custom drawing in OnPaint? Or perhaps there's a graphic resource or a glyph which is corrupt or being improperly disposed of.
Edit: I've re-read your question. You seem to have this only on one of the machines. I've googled around a bit for this, and I stumbled upon this old thread. The post at the bottom suggests that there might be an issue with Virtual Memory turned off:
We did manage to solve this - we were
seeing the problem on a device running
XP embedded. The XPe image developer
had turned off Virtual Memory and as
soon as we turned it on the problem
went away. I believe it is just a
symptom of the system running out of
memory to display the graphics (maybe
particularly if you use a lot of
double buffering)
Hope that helps.
Sounds like a symptom of an Out Of Memory Exception to me.
Edit:
Which can sometimes lead onto a System.InvalidOperationException: BufferedGraphicsContext
Are you trying to update the GUI controls from a thread other than the GUI thread? Combine the cross thread operation with an exception handler that swallowed everything and we had the behavior you describe (on a grid control, not a menustrip bar) on an app I was maintaining.
Definitively sounds like a cross-thread problem.
Make sure you use Invoke when accessing controls and/or firing events from a thread that is not the main UI thread.
Seeing this happen on just one computer of more than a 1000 that have our prouducts.
On that one computer I am seeing a .NET 3.5 program occassionally show the red X on its datagrid. And another far simpler .NET 2.0 program got the red X on its menuStrip. I only have source code for the simpler program but I can say that there isn't any user code at all which affects that component. No cross-thread stuff because nothing updates it. It's contents are set at development time with one item added to it at program load. The Red X failure was well after program load.
I was very surprised to see the problem across two different frameworks and on one program that has no data bindings. I am very wishfully hoping that the computer has its virtual memory turned off.
If it isn't that then any guidance on system parts that are shared across .NET 2.0 and .NET 3.5 would be appreciated.
Update: The user with the problem retired the computer and replaced it (which solved the problem)