|
Dear everyone,
I have to receive a very large data from my sensor via Serial Port. Here below is the structure of my data :
10 bytes HEADER + ( 4 bytes BEGIN + 240bytes DATA )*280 + 4 byes FOOTER
=> The total byte is : 10 + ( 4 + 240 )*280 + 4 = 68334 byes !
And, with such a large data, I don't know how to receive it. My buffer always overflow !!! So, now, I want to create a independent thread for reading data out of buffer, saving to other place and free the buffer sothat it can receive other data from my sensor without overflowing. I tried search "Serial Port" on this website, and I find the article about my problem. That article is "Serial Communication in Windows", posted by Ashish Dhar. I tried to use it, but I can't succeed !
So, please kindly show me how to trap create an independent thread to achieve my purpose.
Please don't ignore my question because this is my final project for graduating from my university ...
Please help me !!! My project's deadline is coming .... I really need that code ...
Thank you very much !!!
Best regards,
Quang Dien
Email to me at : quangdien12jan@yahoo.com
|
|
|
|
|
You can use AfxBeginThread() or CreateThread to invoke a new thread.
In my suggestion Create threads for reading from COM port, Parsing (if any)the received data and processing the data. So u can start 3 threads ( must be ssynchronized). The read thread reads bytes of data only and keep in a byte pointer. The parsing thread read from the byte pointer and remove any header/paddings etc & convert to pure data u need and keep. The processing thread must read from it or assign to proper data structure and do it work..
Anilkumar
|
|
|
|
|
I have a finished implementation for activation that relies on the manufacturer assigned non-changable serial number of a hard drive. During testing I discovered that my method of discovering this number does not work on all versions of Windows that are going to be supported. The WMI claass Win32_PhysicalMedia does not exist in Win2k. The workaround for this is also very painful, requiring DeviceIoControl and probably administrative priveleges at runtime. I noticed that there is Wmi32_BaseBoard class that provides a serial number. However, not all motherboard provide this information (driver code never sets the number).
What is a good, simply way to identify a machine that works consistently on Windows 2000 and newer?
Alex Korchemniy
|
|
|
|
|
|
All the network cards I own (3COM PCI, REALTEK PCIE, 3COM PCMCIA models) allow me to set the MAC-Address on my own. Yes, they DO originally have a "unique"-MAC, but its changeable.
Cheers,
Sebastian
--
Contra vim mortem non est medicamen in hortem.
|
|
|
|
|
look into using following Win32 API GetVolumneInformation[^]
Note that the forth parameter gives you the volume serial number that the operating system assigns when a hard disk is formatted. To programmatically obtain the hard disk's serial number that the manufacturer assigns, use the Windows Management Instrumentation (WMI) Win32_PhysicalMedia property SerialNumber. But obviously you already know that Win32_PhysicalMedia doesn't work in less than WinXP
|
|
|
|
|
GetVolumeInformation can be easily changed and does get changed after formating. I got a good suggested at ms newsgroups... using Win32_ComputerSystemProduct. Its a class that is designed to uniquely identify a computer. So far so good.
Alex Korchemniy
|
|
|
|
|
Hello!
I am searching for a way to rearrange crossed-bounds text so that they don't appear on eachother. Your informations would be very helpful. 
|
|
|
|
|
Sorry if this is the incorrect forum - there are several that are labelled not MFC, but non that are specifically MFC or ActiveX.
The company where I have just started working would like to enable a part of their program to be usable over the web.
As a proof of concept I took the MFC client/server demo by Hans Dietrich and attempted to port the client part of it an ActiveX control.
To aid me I have found Steve Zimmerman's port of the Scribble tutorial from MInD (0497).
My knowledge of MFC and ActiveX is very minimal at best. I have been trying my best but getting no further to getting a working demo running, despite the apparent simplicity of it.
The problem seems to be that the client part of Han's code is the fact that it is done as a CDialog app while the Scribble demo is done as a CFrame. Also the ActiveX initialisation is hidden away inside InitInstance.
It's all very frustrating. What I would like is either a link to a demo/tutorial where someone converts a simple MFC app to ActiveX - Steve Zimmerman's article isn't very in depth and adds to much clutter that confuses the issue. Or I would like a very clear tutorial. A lot of the tutorials I have read are very terse or don't touch any of the concepts that I would like to know about.
While I don't want to be spoon-fed I would like a few pointers to where I can find the information or specific terms I should be searching for.
Thanks, Graham Reeds.
|
|
|
|
|
This is as good as any place to ask, ActiveX is realy a ATL/COM thing thou.
I realy could not explane it with out actualy doing it, and then I would probably need a good book on ATL/COM handy. In VC6, I would probably start by creating an ActiveX control shell, using the "MFC ActiveX ControlWizard", and then porting the code into it piece mill (testing as I went), if possible.
Sorry, I do not know any totorials on porting like this. Most assume that you are building an ActiveX control from the start.
Good luck!
INTP
Every thing is relative...
|
|
|
|
|
John R. Shaw wrote: In VC6, I would probably start by creating an ActiveX control shell, using the "MFC ActiveX ControlWizard", and then porting the code into it piece mill (testing as I went), if possible.
That's the problem - ones done in a Dialog and it seems you can only use dialogs as a dialog in ActiveX.
C# is a lot nicer as it hides all the cruft away and you can get on with the coding. However MFC generates zillion comments and macros which I have no idea what they all do. Some articles delete them while others say to leave them alone.
Now Borland Builder (5 or 6) doesn't seem so bad...
|
|
|
|
|
Generalizing/Seperating is a problem, in MFC you can create an CRichEditView. But the CRichEditView is the top of the food chain, at the bottom is the actual RichEdit control. You can use the control seperate from the the view (like in a dialog), but you would might have to provide your own programmers interface.
CRichEditView provides a programmers interface that provides wrapping for the CRichEditCtrl programmers interface, which provides wrapping for the actual RichEdit control interface (IRichEditOle). The RichEdit control is not dependent on MFC (as far as I know).
The general idea is to seperate the code that actualy does the work from the type of window where the results are displayed, if possible.
Yes, I know it sounds very complicate (and it is), but that is what it amounts to.
Personaly I hate dealing with COM at the lowest level, that's why wizards are so important. Imagine trying to write all that code youself.
If you can and are allowed to convert the code into C#, provided it will make life easier, then go ahead. From what I have seen, sooner or later, some one will be asking if there is a programmer that can convert the C# code to C++.
I know none of this realy helps you, but I thought I would give you my general veiws on the subject.
INTP
Every thing is relative...
|
|
|
|
|
Hi
i have a little problem using the list<>::iterator.
Assume you have a class CARS{ virtual void info();...}
and some public derived classes as GM{void info();...},BMW{void info();...}
<br />
...<br />
list<CARS> carpark;<br />
GM car1();<br />
GM car2();<br />
BMW car3();<br />
BMW car4();<br />
<br />
carpark..push_back(car1);<br />
carpark..push_back(car2);<br />
carpark..push_back(car3);<br />
carpark..push_back(car4);<br />
<br />
list<CARS>::iterator it;<br />
for(it=carpark.begin();it != carpark.end(); it++)<br />
{<br />
cout << it->info() << endl;<br />
}<br />
...<br />
so far so fine.
Now how to call the info() from GM or BMW using the iterator ??
(not the one from CARS)
THX
Tim
-- modified at 17:50 Thursday 5th January, 2006
|
|
|
|
|
It should just be called for you automatically. That's the point of an overrided method.
However, if all cars have a type, then it should be stored in a variable that exists in the base class, and the GetType method should exist only in the base class. The derived classes should just set that variable.
Christian Graus - Microsoft MVP - C++
-- modified at 17:58 Thursday 5th January, 2006
|
|
|
|
|
Hallo
THX so far, but how do i do this with the GetType Methode and how can this help me calling the method of the derived class ??
Tim
|
|
|
|
|
#include <string>
class CAR
{
private:
std::string type;
public:
std::string GetType();
}
class BMW
{
public:
BMW() { type = "BMW"; }
}
and so on. you don't need to use std::string, obviously.
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
HI
OK have somthing like this but how to call the individual info() method ??
Tim
|
|
|
|
|
Well, my way, there is no individual info method. However, in either case, you should find you're calling it in your original code. If not, perhaps you need to store a pointer instead of a class instance, perhaps it's being downcast by std::list at the moment ?
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
OK
THX alot
Will write my own list and iterator
That might be no problem then i just liked to know how to use list and iterator
CU
|
|
|
|
|
tbrake wrote: Will write my own list and iterator
No, please don't do that. What I was suggesting was that you should perhaps try storing a CAR * instead of a CAR in std::list.
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
OK, I've done some testing:
this works:
#include <string>
#include <iostream>
#include <list>
class CAR
{
public:
virtual char * GetInfo() { return "BASE"; };
};
class BMW: public CAR
{
char * GetInfo() { return "BMW"; }
};
class Gemini :public CAR
{
char * GetInfo() { return "Gemini"; }
};
class Benz :public CAR
{
char * GetInfo() { return "Benz"; }
};
int _tmain(int argc, _TCHAR* argv[])
{
std::list<CAR*> cars;
cars.push_back(new BMW());
cars.push_back(new BMW());
cars.push_back(new Gemini());
cars.push_back(new Benz());
std::list<CAR*>::iterator it = cars.begin();
while(it != cars.end())
{
std::cout << (*it)->GetInfo() << std::endl;
++it;
}
int i;
std::cin >> i;
return 0;
}
but if you don't store pointers in the list, then it does indeed downcast to the base class.
Christian Graus - Microsoft MVP - C++
|
|
|
|
|
Hello all,
(My first post)
I am developing an application for WinCE 3.0 for a Pocket PC 2000 device in C++ using MFC / WinAPI.
In my application I use a Bluetooth CF card to communicate with a wireless printer. The drivers installed assign a COM port to the printer so I may
read / write to it using WinAPI calls for Files such as:
CreateFile( ... );<br />
ReadFile( ... );<br />
WriteFle( ... );
The wireless printer conserves battery by automatically turning off after a period of inactivity, so at times I must check to see if it is there before the user tries to print something.
The problem is that if I want to know anything about the file I must first get a handle with the CreateFile( ... ) function, which lags about 10-15 seconds before it times out if the printer is off, so usually the user is already trying to print something by the time I get back from that function call. Or if I make the user wait until I get back, they will get frustrated after so much time.
I tried using
CFile::GetStatus( ... )<br />
but it doesn't seem to work for COM ports, eg. I can call:
HANDLE hPrinter = CreateFile( _T("COM4:"),...);
and I get the handle but if I do this:
BOOL bIsPortAvailable = CFile::GetStatus( _T("COM4:"), ...);
I get FALSE back;
Is there a way to specify the timeout for the CreateFile( ... ) function or another way to check if that COM port is available?
Thanks Much,
T.J. Barbour
|
|
|
|
|
You could try polling the printer to see when it's off. That way when a user tried to print, the status would already be off unless it just turned off a few seconds before the user tried to print.
I think you might get more responses if this post were moved to the WinCE forum? I've noticed lots and lots of function calls give me major lag on WinCE when they don't on WinXP, so you're more likely to get someone who knows how to solve the problem over there.
Kelly Ryan
|
|
|
|
|
I'll look into that.
Polling is an option that I may persue, but we all know the disadvantages of polling : )
I appreciate the reply!
Thanks Much,
T.J. Barbour
|
|
|
|
|
Maybe you can get the printer driver manufacturer to support WM_DEVICECHANGE then you can actually be notified that the power has come or gone...
Marriage slows down your coding, a baby slows it down even more!
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
