|
I tried:
call "C:\Program Files\Microsoft Visual Studio .NET 2003\Common7\IDE\devenv.exe" "CMM Tool Kit 7.10.sln" /rebuild release /D "PCM_MULTILINEAR" /out test.log
But I got the following error message:
Invalid Command Line. Unknown Switch : D.
As far as : "Alternatively, you would set up proper project-configs for that, and call devenv with it:"
Is project-configs simply a text file or a file with a specific extension? Could you please give an example of how the content of project-configs would look like.
Thanks.
|
|
|
|
|
Mohamed Douffir wrote: Is project-configs simply a text file or a file with a specific extension?
Look at one of the projects set up with the VC++-GUI and open the *.vcproj-file.
Alternatively, there is nmake, the Microsoft make-implementation.
Look for help on it in MSDN.
Though I speak with the tongues of men and of angels, and have not money, I am become as a sounding brass, or a tinkling cymbal.
George Orwell, "Keep the Aspidistra Flying", Opening words
|
|
|
|
|
Try here.
"A good athlete is the result of a good and worthy opponent." - David Crow
"To have a respect for ourselves guides our morals; to have deference for others governs our manners." - Laurence Sterne
|
|
|
|
|
Hi guys,
I'm currently figuring how to use of genetic algorithm to produce a image recognition application for my project.
Simply to say, is something like handwriting recognition or neural netwrk.
But to all my researches and understanding, GA are only applicable to path finding alghorithm and optimization. GA works well with binary. My intention was to make use of pixel count by "1s" for blank pixel and "0s" for black pixels.
Can anyone give me guidances or any tips/ related projects as references. Thanks
Ray Luee
|
|
|
|
|
A genetic algorithm does, loosley speaking, try to find a global maximum by looking at a number of random spots in the "solution space".
These spots are called individuals, and how "good" they are is determined.
In the next step ("generation"), it exchanges the worst individuals by some new individuals near the position of the best individuals known so far.
You have no guarantee whatsoever that you will find the global maximum, but you will find one or more local maxima.
Also, you will need to determine what is a "good" and what is "bad" fast and reliably - its called the "fitness function", and it is evaluated for each individual in each generation.
Can you give a suitable fitness function? Then a GA might work very well for you.
Though I speak with the tongues of men and of angels, and have not money, I am become as a sounding brass, or a tinkling cymbal.
George Orwell, "Keep the Aspidistra Flying", Opening words
|
|
|
|
|
Yes, as expected from you, I was told to compared random sample of points on the two images to find a set of relationship that will transform one images to fit onto another using GA.
My understanding of GA was to
Based on fitness of the guesses from past guesses.
Encode using binary strings.
By swapping and recombining.
How will you approach to this situation to get the suitable fitness function?
Rayluee
|
|
|
|
|
rayluee wrote: How will you approach to this situation to get the suitable fitness function?
Actually, when I was teached GA, I was told to find out a fitness-function beforehand.
When a fitness function would "spring to mind", would be fast and "naturally" to find, the problem would be well suited to a GA.
If not, one might force it, but probably the GA would ever be suboptimal.
For your problem of transforming pictures, do you have a "solution space" with many dimensions?
Probably: each pixel is a dimension.
Your fitness function would involve computing some sort of a difference of the pictures, where a minimal difference would be best.
So each individual would consist of a set of pixel manipulations done on pic1, and you would then calculate the difference between the manipulated picture and pic2.
But whether that can be done fast?!?
Also, I have no idea if "crossover" would be sensible in this context.
</Brainstorming>
I hope to have given you something to start thinking.
Though I speak with the tongues of men and of angels, and have not money, I am become as a sounding brass, or a tinkling cymbal.
George Orwell, "Keep the Aspidistra Flying", Opening words
|
|
|
|
|
How do you traverse, or loop through the controls on the dialog?
The problem I have is that I have a spreadsheet control on a dialog, and when the dialog opens, the spreadsheet control is loaded before the header information is displayed. I thought I could grab the indexes for the controls located in the header and loop thru them to implicitly show them before loading the sheet.
for(int lcv = IDC_VENDID; lcv <= IDC_BTNHEADDOWN; lcv++)<br />
{<br />
GetDlgItem(lcv)->ShowWindow(SW_SHOW);<br />
}
The problem with this approach is that the control indexes aren't sequential, and an exception is thrown when lcv hits a value that references nothing (its not really nothing its a control id for another dialog).
|
|
|
|
|
littleGreenDude wrote: for(int lcv = IDC_VENDID; lcv <= IDC_BTNHEADDOWN; lcv++)
since you have the ID's why don't you just use the one for the window you are after and get it?
|
|
|
|
|
 Dude, I'm a programmer, and programmers are fundamentally lazy in nature (so as not to offend anybody in our overly PC society, allow me to rephrase and state that all the good ones I know display this characteristic). And it serves them well to be that way. Think about it, there could be 300 controls in the header of the screen. Would you want to type 300 hundred Show Window statements. I think not. Also, it would be kind of ugly looking code.
Okay, so I'm exaggerating with the 300 number, but even 3 is one too many. I'd match rather have the ability to step from the low control to the high control and jump over the invalid references along the way.
|
|
|
|
|
littleGreenDude wrote: the controls located in the header and loop thru them
If they are real child windows you only need the header/parent then there are API's for like FindFirstChild/FindNext etc.
|
|
|
|
|
I dont understand your problem fully. Would something like this help you?
for(int lcv = IDC_VENDID; lcv <= IDC_BTNHEADDOWN; lcv++)
{
CWnd* pCtrl = GetDlgItem(lcv);
if( pCtrl->IsKindOf(RUNTIME_CLASS(CHeaderCtrl))) {
pCtrl->ShowWindow(SW_SHOW);
}
}Or do you simply need to change the Tab-Order of your controls?
Though I speak with the tongues of men and of angels, and have not money, I am become as a sounding brass, or a tinkling cymbal.
George Orwell, "Keep the Aspidistra Flying", Opening words
|
|
|
|
|
Thank you, that's close to what I was looking for...
The following calls showwindow for all the valid controls
for(int lcv = IDC_VENDID; lcv <= IDC_BTNHEADDOWN; lcv++)<br />
{<br />
CWnd* pCtrl = GetDlgItem(lcv);<br />
if( pCtrl )<br />
pCtrl->ShowWindow(SW_SHOW); <br />
}<br />
However, for some reason it is still not displaying the header information prior to loading the spreadsheet control. I guess I'll have to post a message to spreadsheet vendor's forum to see what the deal is.
Thanks for the help.
|
|
|
|
|
littleGreenDude wrote: ...header information...
"A good athlete is the result of a good and worthy opponent." - David Crow
"To have a respect for ourselves guides our morals; to have deference for others governs our manners." - Laurence Sterne
|
|
|
|
|
Sorry, header information - all the labels, edit fields, combos, etc. located above the spreadsheet control on the dialog. Specifically in this case it is a Purchase Order dialog so the spreadsheet consists of inventory items to order and the header is the typical PO header stuff (vendor, PO#, order date, delivery date, terms, shipping, etc.).
|
|
|
|
|
Do those controls happen to be contained within a group control? Or, can you change the controls such that their IDs are sequential? Or, do they all have the same parent?
"A good athlete is the result of a good and worthy opponent." - David Crow
"To have a respect for ourselves guides our morals; to have deference for others governs our manners." - Laurence Sterne
|
|
|
|
|
Not within a group. I could changed the IDs to be sequential, but I was looking for another solution.
The response provided above, creating a pointer, and if the pointer exists perform a the ShowWindow is the answer I was looking for.
I was told on another forum that the dialog controls actually paint in reverse order of the tab order. Is that true?
|
|
|
|
|
littleGreenDude wrote: The response provided above, creating a pointer, and if the pointer exists perform a the ShowWindow is the answer I was looking for.
Hence my question about them all having the same parent. However, you indicated here that it did not work.
littleGreenDude wrote: I was told on another forum that the dialog controls actually paint in reverse order of the tab order. Is that true?
Yes, with certain versions of VS.
"A good athlete is the result of a good and worthy opponent." - David Crow
"To have a respect for ourselves guides our morals; to have deference for others governs our manners." - Laurence Sterne
|
|
|
|
|
Try code like this:
HWND hwndDialog = GetSafeHwnd();
for (HWND hwndChild=::GetWindow(hwndDialog, GW_CHILD); hwndChild!=NULL; hwndChild=::GetWindow(hwndChild, GW_HWNDNEXT))
{
}
Steve
|
|
|
|
|
Does anybody know of any good articles about communication with a PDA attached to the PC?
I have an EASTCOM ES2100 ( Chinese brand ) PDA with a Hopen 2.0 OS. Now the guys at Hopen have created an SDK for this OS but I have no way of installing my software on the PDA. I'm going to have to create a windows driver to connect to open a port to the device, but after that I have no idea as to what to do.
I do know that the OS is based on Linux is thats of any use :/
|
|
|
|
|
I'm no Linux expert but if Hopen is Linux based I would expect it to be running server type services of some sort telnet, ssh or even vnc. I'd start from there, consider what type the physical ports are at either end and see what 3rd party tools like USB serial drivers or PUTTY you could use.
Nothing is exactly what it seems but everything with seems can be unpicked.
|
|
|
|
|
Hi everybody,
i made a buffer-CEdit where i rely a char* with a CEdit.
how can i store the char* from the View into the CEdit?
i create the View-Variables like this:
char *var = (char*)malloc(sizeof(char)*50);
My own CEdit derived texbox:
MyCEdit tbNumber (Creation via DDX)
then i "link" the string with the Textbox
tbNumber.Link(&var);
where MyCEdit has a
char** InternBuffer;
and the "linkage" is made this way:
void MyCEdit::Link(char** var_t)
{
InternBuffer = var_t;
}
But it doesn't work, i tried some other possibilities, but i always get another pointer value
There is seriously a simple * mismatch
Thanks for help
|
|
|
|
|
Why are you doing this? One of the major selling points of using MFC is that you don't need to mess around with char-type variables, especially where the memory manager is concerned.
if you need the txt within an edit control, simply use:
CString str;
tbNumber.GetWindowText(str);
"A good athlete is the result of a good and worthy opponent." - David Crow
"To have a respect for ourselves guides our morals; to have deference for others governs our manners." - Laurence Sterne
|
|
|
|
|
Thanks for your answer,
i like to develop this kind of trading to minimize the lines of code.
at the window i have class-member strings which are always up-to-date with the
content of the textbox.
Because if i have 10 textboxes on a view and they'll filled with data from out a database
then the user can modify the content and at the destruction of the view all the contents of the
textboxes are re-stored into the database.
In this case i have to GetWindowText 10times and SetWindowText 10times.
If the view performs other special functionalities i Get/Set more than 2-3times.
Since now i always do it in your way, but i try to keep the DOS-Programming way, for that
my work-partners, which works only in DOS programming could understand more faster my application.
Because i copy the existing DOS-Applic 1:1 into MFC
And i don't have the time to change all the logic from DOS-to-Windows Programmatic-System
I know that all this is strange, but it's my work 
|
|
|
|
|
I fail to see what any of this has to do with using char* rather than CString variables. If your coworkers are unable to work with CString, then they are also unable to work with CEdit.
"A good athlete is the result of a good and worthy opponent." - David Crow
"To have a respect for ourselves guides our morals; to have deference for others governs our manners." - Laurence Sterne
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
