Firstly, I've very little experience with MFC and realize my answer may not be helpful.
However, when I fire up a plain ol' win32 program and click on the header of a list-view, there are indeed two WM_NOTIFY in a row in the messages list.
Looking at the messages in Spy++, I get:
<00824> 01BC0A0C S WM_NOTIFY idCtrl: 0 pnmh:0028DD08
<00825> 01BC0A0C R WM_NOTIFY
Checking the help for spy++, gives this explanation of the 'S' and the 'R' Code Meaning P The message was posted to the queue with the PostMessage function. No information is available concerning the ultimate disposition of the message. S The message was sent with the SendMessage function. This means that the sender doesn’t regain control until the receiver processes and returns the message. The receiver can, therefore, pass a return value back to the sender. s The message was sent, but security prevents access to the return value. R Each ‘S’ line has a corresponding ‘R’ (return) line that lists the message return value. Sometimes message calls are nested, which means that one message handler sends another message.
Reading the code meanings seems to suggest that the second instance, the 'R' one is the return value of a call to SendMessage (when the child notified the parent)
Now I really am intrigued...
Are you able to report on what spy++ tells you about the messages being passed in your program?
Which perhaps unsurprisingly points to the below code:
ON_NOTIFY_REFLECT_EX(LVN_COLUMNCLICK, OnColumnclick) /* This is the offending line *///}}AFX_MSG_MAP
Tried it with VS2010.
Upon digging deeper, I came across this page[^] at msdn, that seems to indicate that your function OnColumnClick should have a return-type of BOOL
Hmmmmmm - just tried editing the definition of the OnColumnClick function so that it has a return type of BOOL, instead of LRESULT - that worked!
Now, finally I can see where you're coming from. Rather than the pair of messages I was seeing in my app, I can see that your app is generating 2 pairs of messages. It seems that the messages are being sent from two different controls
Each click on a column header is giving me (amongst other messages) this:
<00323> 015A080C S WM_NOTIFY idCtrl:59680 pnmh:0040E220
<00324> 015A080C R WM_NOTIFY
<00325> 015A080C S WM_NOTIFY idCtrl: 0 pnmh:0040F96C
<00326> 015A080C R WM_NOTIFY
Perhaps an investigation into what these two controls are is in order..
Perhaps one of these may help? (1)[^] (2)[^] (In german, use Google translate - for some reason had trouble pasting the url of the translated page)
It seems that the messages are being sent from two different controls
Each click on a column header is giving me (amongst other messages) this: < 00323> 015A080C S WM_NOTIFY idCtrl:59680 pnmh:0040E220
< 00324> 015A080C R WM_NOTIFY
< 00325> 015A080C S WM_NOTIFY idCtrl: 0 pnmh:0040F96C
< 00326> 015A080C R WM_NOTIFY
Perhaps an investigation into what these two controls are is in order..
The ID of the header control within a listview is always 0.
"One man's wage rise is another man's price increase." - Harold Wilson
"Fireproof doesn't mean the fire will never come. It means when the fire comes that you will be able to withstand it." - Michael Simmons
"Some people are making such thorough preparation for rainy days that they aren't enjoying today's sunshine." - William Feather
I need to be able to stop COM port from sending the Morse code using space character from keyboard.
I use a button change event to execute the code sending function, however, I am unable to detect the keyboard since the sending function is still executing.
I tried to “insert” PostMessage custom WM_USER message but Windows still waits until the sending function is finished before recognizing the keyboard.
I am using PreTranslateMessage in MFC view.
Would adding another thread work?
I do not have much experience in multithreading, but willing to do it if it solves the problem.
Any other constructive suggestions will be appreciated.
That may work... what you have going on is a classic case of a C/C++ blocking call keeping you from doing what you need... in which case the threading should help. I would however, move the COM interactions to their own thread and stop when needed instead of leaving the blocking call in the main thread.
OK,I'll will do multithreading. May as well learn it now and I will need to stop other types of transmission in future anyway.
Just a thought - would it be possible to "chop" the Morse code into event function for each element instead of sending the whole string of Morse characters in one event function? Probably too messy and it would still not be stoppable in the middle of dash character anyway.
I got the new thread running with one problem.
Just for test I tried to intercept the keyboard via PreTranslateMessage ( in the new thread) and it did not work.
As suggested , eventually the COM stop will go into another thread, hopefully I can figure out how to communicate between threads.
For now I need to “redirect” the keyboard messages into the new thread.
They are still going ( being intercepted ) to the main thread.
Because I still cannot get the main thread to intercept the keyboard while the sending function is executing even in the new thread.
I must be doing it wrong - here is the code snippet
// implementing multiuthreading
TRACE("\n Start the C_MorseCode_Thread thread");
m_C_MorseCode_Thread = (C_MorseCode_Thread*) AfxBeginThread(RUNTIME_CLASS(C_MorseCode_Thread));
m_C_MorseCode_Thread->m_C_Morse_Class->C_ConvertText("Test Test Test Test ",m_CFD2008Doc);
Do I need to start the execution in the new thread itself?
I am making progress, however, have some minor questions.
1.How does the parameter gets into worker thread control function ?
2.Why does this function need to be defined as static?
3.Per article I am using , the worker thread can be “easily stopped”, however, since the thread sets the COM port (DTR) how do I know it will disable the DTR not just quit the thread? ( I am guessing I need to make sure it will and stopping the thread is no guarantee.)
4.If I use user-interface thread I need to create Cwnd ( Morse code class) It fails but “getLastError” reports “function completed successfully”.
I am using MFC document / view to set / control COM port (DTR) ( outputting Morse code string) and have to be able to stop it ( reset DTR via keyboard ) from the main thread.
I think I will need to use user-interface thread to get better control.
One more – if I want the user interface thread to intercept keyboard messages I am planning to intercept them in the main thread (Cview) via PreTranslateMessage and than PostThreadMessage to thread of interest. Is that the only way to accomplish this?
Thanks for reading,
Any constructive help will be as always appreciated.
In my aplication menu had a ribbon edit with spin button, I'm using this:
But when I increment with up arrow over 999 the value is changed to 1,000
I used GetEditText to change this value but what I have is 1
I can't change increment or ignore ',' or remove ',' from value, that is a bug with ribbon.
In old MFC application I have Spinbuttonctrl in create method I have used UDS_NOTHOUSANDS but in ribbon I don't have this option.
Seems the comma in 1,000 is his problem (ie GetEditText reads "1,000" then parseInt() or whatever stops at the comma). If there's no suitable style/attribute on the control he's using, then I guess it requires a bit of string-munching.
The problem is comma, when I use GetEditText() returns '1' and not '1,000', I want 1000 because I'm editing a object and when reach 1000 returns to value 1, that's wrong. I don't have parseInt() method I'm using mfc c++.
This seems like such an obvious bug... why would GetEditText() parse out the text (CString never ends on a ',' so I don't see how this bug could occur)? where are you checking your return? Maybe post your actual code and point out where you see the 1.
Last Visit: 31-Dec-99 18:00 Last Update: 2-Oct-23 16:22