|
Using Events is also another option for notifying another window.
|
|
|
|
|
Hello, everyone. I am designing an application that is intended for use with a touch-screen monitor, so all of the GUI elements need to be big, friendly, and easy to poke.
I would like to provide context-sensitive help for various controls on the main form, and thought that I might just watch for the WM_HELP message that is sent whenever a control is clicked on when in "What's This?" mode, entered by clicking the "?" help button on the title bar. That bit works fine, but I am running into an unexpected problem.
Since the "?" button is much too small for my touch-screen requirements, I thought that I would include a larger help button elsewhere on the form that would accomplish the same thing. I have been totally unable to find any way to programmatically trigger "What's This?" mode, though. The "?" button doesn't even seem to send a message to WndProc() when it is clicked. Does anyone know what occurs when this button is clicked, and how to accomplish the same thing programmatically?
If not, does anyone have any suggestions for providing context-sensitive help that doesn't require something silly, like putting a check in the form of if(helpMode) { ShowHelpMessage(this); return; } at the start of the click event for every single control?
Thanks very much!
|
|
|
|
|
Oddball wrote: and thought that I might just watch for the WM_HELP message
I don't know but I would not think that at all. If I needed to know I would start here[^].
led mike
|
|
|
|
|
Mike,
Thank you for the suggestion, but you really did not address my question. As I stated, I am recognizing when a user clicks on a control when in "What's This?" mode with no problems. Everything works great if the user can click the help ("?") button in the title bar and then on another control. Since my requirements are that a touchscreen be used for the interface, however, I need to create a larger button that operates like the titlebar help button.
I am unable to enter "What's This?" mode programmatically. That is what I was asking about. I've searched all over, including in reference books, with no results, and thought that I'd seek expert experience as a last resort. Respectfully, directing me to Google with a search phrase that general is insulting.
|
|
|
|
|
Oddball wrote: directing me to Google with a search phrase that general is insulting.
Yes it is, to me since I obviously did not read your post thoroughly. I apologize for that but I was not trying to insult you.
led mike
|
|
|
|
|
Ok, I read your OP again and this several times. Since you are on a touch screen you wan the user to click once on your button and then a second time on the control to kick off Context Sensitive Help yes? If that's correct then I don't know if this will help you but the '?' button sends a WM_SYSCOMMAND message.[^]
led mike
|
|
|
|
|
Mike,
I apologize for not responding sooner, but I was a victim of whatever happened to the boards yesterday, and couldn't log in. Sorry for getting snippy back there, too. I should know better than to check up on message boards when my day isn't going well to begin with. I suppose I could have made my OP a little more clear, too.
In any event, the WM_SYSCOMMAND message sounds like exactly what I need. I came to the conclusion before that the "?" button was not sending a message by trying to watch the behavior of the WndProc method when it was pressed. With how many messages go through that puppy, I missed it.
I've been playing around with Win32 messaging since yesterday, which is new to me, and I can't quite seem to get this rolling. Perhaps you could shed some more light on this for me? I believe that this is the last part of the solution to my problem. I've cobbled together the following lines in a button click handler to attempt to activate "What's This?" mode:
int WM_SYSCOMMAND = 0x112;
IntPtr SC_CONTEXTHELP = IntPtr(0xF180);
Windows::Forms::Message msg = Windows::Forms::Message::Create(this->Handle, WM_SYSCOMMAND, SC_CONTEXTHELP, IntPtr(0));
this->WndProc(msg);
It compiles without complaint, but nothing happens (that I can detect, anyway) when the button is clicked. Can you see what I am screwing up or missing?
Thanks for your patience, and I appreciate the help.
|
|
|
|
|
Oddball wrote: I've been playing around with Win32 messaging since yesterday
Well I have no idea if this is going to work since I've never tried it either but:
Oddball wrote: this->WndProc(msg);
You can't just send the message to your windows procedure. I would imagine if it's going to work at all you have to post or send the message.
PostMessage[^]
That said you may not be able to leverage the system in that way if they never intended for it to be used that way. You might have to track the state yourself and then invoke the help after they click on the control they want the help on.
Does that make sense?
led mike
|
|
|
|
|
Mike,
led mike wrote: You might have to track the state yourself and then invoke the help after they click on the control they want the help on.
I was afraid of that. I thought there might be a more elegant solution with hijacking the "?" button functionality, but I am getting the impression that I am now just over-complicating the issue.
I'll give it a shot, but one last question. We're rapidly leaving my realm of C++/CLI experience, (which isn't that broad anyway, since I am much more familiar with C#) so I apologize if this next one is borne of ignorance. PostMessage does not appear to be a .Net method, and MSDN indicates that it lives in user32.dll. How can I use it? Is this where C++/CLI's ability to use unmanaged code comes in? If a complete answer to this would be too lengthy, I'll be satisfied with a "you need to go learn about X some more" response this time. 
Thanks again.
|
|
|
|
|
Oddball wrote: Is this where C++/CLI's ability to use unmanaged code comes in?
Yes but I think the idea of tracking the state yourself and then using Luc's provided API is far simpler and should satisfy your requirements.
led mike
|
|
|
|
|
Mike,
It appears that the system was just not designed to do what I am after. I will have to come up with a different way to provide the functionality I need.
Thank you for your help.
|
|
|
|
|
Hi,
Google told me there is a CodeProject article[^] that should interest you.
Luc Pattyn [Forum Guidelines] [My Articles]
This month's tips:
- before you ask a question here, search CodeProject, then Google;
- the quality and detail of your question reflects on the effectiveness of the help you are likely to get;
- use PRE tags to preserve formatting when showing multi-line code snippets.
|
|
|
|
|
Luc,
So much of that article was just so close to what I needed that it was a frustrating read! It had a great deal of information on how to provide help once context-sensitive help mode is engaged by the user, but not how to engage it programmatically.
I appreciate all the knowledge I can get, though, and I picked up a few ideas from there for other projects. Thanks very much!
|
|
|
|
|
You're welcome.
I don't quite understand how you fail to engage the help mode; AFAIK the article provides
context sensitive help triggered by the F1 key, and it implements a ShowContextHelp() method
to do the work. You should be able to do similar things triggered by another key,
mouse gesture, whatever you choose.
Luc Pattyn [Forum Guidelines] [My Articles]
This month's tips:
- before you ask a question here, search CodeProject, then Google;
- the quality and detail of your question reflects on the effectiveness of the help you are likely to get;
- use PRE tags to preserve formatting when showing multi-line code snippets.
|
|
|
|
|
Luc,
The project that the article is discussing is much more complicated than what I was looking for. If I understand it correctly, it is providing home-brewed context-sensitive help functionality. I was trying to use the built-in "What's This?" mode, but with my own button to start it, rather than the '?' button in the title bar.
Between you and Mike, it is clear to me that the system was not written with this goal in mind, and there is no easy way to accomplish what I am after. I will have to find a different way to provide the functionality I am looking for.
Thank you both for your help.
|
|
|
|
|
You're welcome.
If the article's approach is more complex than you anticipated, there is probably good
cause for it. Why not adopt their methods? I trust it works...
Luc Pattyn [Forum Guidelines] [My Articles]
This month's tips:
- before you ask a question here, search CodeProject, then Google;
- the quality and detail of your question reflects on the effectiveness of the help you are likely to get;
- use PRE tags to preserve formatting when showing multi-line code snippets.
|
|
|
|
|
Hi,
I'm using Windows Forms C++.
I'm getting the user to add commands to a listbox then they click on a button and this runs the commands in the listbox.
The problem I have is, I don't know how to run these commands from an index number because they are currently being run in the order shown below and not in the order of where they are in the listbox.
Does anyone have any ideas?
do<br />
{<br />
if(lstCommands->Items->Count > 0)<br />
{<br />
<br />
if(lstCommands->Items->Contains("Move forward")) <br />
{<br />
protocol_test(2, distance, 0);<br />
<br />
}<br />
else if(lstCommands->Items->Contains("Move Back"))<br />
{<br />
<br />
protocol_test(3, distance, 0);<br />
<br />
}<br />
else if(lstCommands->Items->Contains("Turn Right"))<br />
{<br />
protocol_test(4, distance, degrees);<br />
<br />
lstCommands->Items->Remove("Turn Right");<br />
}<br />
else if(lstCommands->Items->Contains("Turn Left"))<br />
{<br />
<br />
protocol_test(5, distance, degrees);<br />
}<br />
<br />
}<br />
else if(lstCommands->Text->Empty)<br />
{<br />
MessageBox::Show("No Commands selected","No Commands",MessageBoxButtons::OKCancel,MessageBoxIcon::Exclamation);<br />
} <br />
} while(!lstCommands->Items->Count == 0);<br />
|
|
|
|
|
newkid wrote: Does anyone have any ideas?
Yes. An Object Oriented Design would be far easier to understand. In this regard finding errors would be relatively simple.
If/then and switch statements are the simplest form of instance –specific behavior….
The more paths through a program the less likely the program is to be correct….
The proliferation of conditionals reduces reliability….
This problem is compounded when conditionals are duplicated.
These problems can all be eliminated by converting the conditional
logic to messages, either with subclasses or delegation. Kent Beck - Implmentation Patterns
led mike
|
|
|
|
|
hi
I am working in complete 'C' environment on UNIX. I have to convert signed char and signed char* to unsigned char and unsigned char* respectively.
I have read that using (unsigned char) is not very safe and can cause.
Please suggest me any way I can do this conversion. Is there some function I have to write?
Please help
think because thats what matters
|
|
|
|
|
Welcome to CodeProject ![Beer | [beer]](https://codeproject.global.ssl.fastly.net/script/Forums/Images/beer.gif)
The Windows platform now includes the .NET platform and this forum is specifically for using Microsoft Visual C++ versions with language extensions for the .NET platform. That is why the forum name says "managed C++" in reference to the managed memory (garbage collected) enforcement of the platform. For C or standard C++ questions you want to use the Visual C++ / MFC forum.
vikramaditya234 wrote: I have to convert signed char and signed char* to unsigned char and unsigned char* respectively.
There are two things in play when casting from signed to unsigned. First the sizeof() of the source and target types. If the source is larger than the target then you can lose data since you have less bits to hold the data. The second is the use of the sign bit called Two's Complement[^].
led mike
|
|
|
|
|
I am using Managed C++ under Microsoft Visual 2005 on a Windows forms application.
I have a BufferedGraphics object where I paint the information (in a double buffer approach), while in OnPaint function I am rendering it to the display.
I would like to use more graphics objects to be rendered in an “Overwrite” way to the display. This is similar to different layers of image to be painted together.
In Win32 I was able to do this using BitBlt, specifying SRCCOPY as raster operation.
How can I do it using BufferedGraphics ?
Thank you.
|
|
|
|
|
The equivalent to BitBlt() is Graphics::DrawImage(). You should be able to use that to draw
multiple bitmaps to your BufferedGraphics Graphics object.
Mark
Mark Salsbery
Microsoft MVP - Visual C++
![Java | [Coffee]](https://codeproject.global.ssl.fastly.net/script/Forums/Images/coffee.gif)
|
|
|
|
|
Thank you, Mark.
Indeed, DrawImage can be used for this but I am not happy about how it works for me: after clearing the bitmaps using FillRectangle, the bitmaps are not drawn one "on top" of the other; instead I only get the last one.
Regards,
Cristian
|
|
|
|
|
cristian.fatu wrote: the bitmaps are not drawn one "on top" of the other; instead I only get the last one.
Where do the rest of them go then? If the last one is larger than the previous ones, then of course you'll only see the last one.
That's the same behavior as BitBlt() (which you can use as well - just get a DC from the Graphics object).
Otherwise something is wrong with the code.
If you want them to blend, then you'll need to use 32bpp ARGB bitmaps.
Mark
Mark Salsbery
Microsoft MVP - Visual C++
![Java | [Coffee]](https://www.codeproject.com/script/Forums/Images/coffee.gif)
|
|
|
|
|
All the bitmaps are equals in size, they are created by code.
Finnaly I implemented my functionality by avoiding this "overwritten" approach. I am using an "empty" screen BufferedGraphics object that I am rendering to the current BufferedGraphics object every time I clear the screen and then I repaint all the layers to the current BufferedGraphics object (being double buffered I can afford it).
Thank you for the involvment.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
