|
I have a C program that has some static arrays of Number/String pairs (error number to descriptive text mapping). I thought that this would be great to have in a C++ collection class (std::map, or CMap in windows). Now apart from the weird template syntax, this would work nicely.
However, to initialise the collection class, I have to run an initialiser function that loads in the collection entries, either as hard-coded values in the source code, or by reading a file.
In the case of hard-coded values in the source code, the application ends up with two copies of the data in memory. Not as efficient as old C static arrays.
In the case of file reading, there is the overhead of the load procedure.
IS there any way, or proposed C++ standard, to initialise a collection/classes at compile-time - so you could pre-fill the data elements as easily as you can currently define array elements?
|
|
|
|
|
One solution is a DLL. Initialize the containers in the class contructor.
Kuphryn
|
|
|
|
|
I've been handed some code in Matlab to
convert to c++. Although I am familiar with Matlab,
I have come across:
while 1 % forever do...
% carry out a calculation
end % end forever do
with no obvious termination condition. My question
is, how will this loop ever terminate as I need
the c++ function version of it to eventually return
a value? I suspect I should be looking for a
return or break statement
but can't seem to locate one inside the while loop.
I am thinking this loop is not coded correctly, but
have been assured that it works... 
I'm not exactly a Matlab Guru, so maybe I am missing
something?
John Theal
Physicist/Mathematical Programmer
Digital Immersion Software Corporation
Got CAD?
http://www.presenter3d.com[^]
http://www.merlin3d.com[^]
|
|
|
|
|
I don't know Metlab at all but just looking at your post, could it be possible that it will break out of the loop if the condition "l" becomes false?
Rob
Whoever said nothing's impossible never tried slamming a revolving door!
|
|
|
|
|
|
look for a break or a return within the loop
J.
----------------------------
|
|
|
|
|
Welcome!!! I looking for description of UPX compressed exe files. How can I - programmatically - read PE header of oryginal file - before compression. In most cases that is not protected, so in theory it's possible - but how ?
Sorry 4 my polish 
Mila
|
|
|
|
|
Hi
I have a program which runs in command line version, I need to collect the
output which commandline version is displaying with "cout<<" and have to store it in a file, I could do it with the ios::sync_with_stdio()
If I need to dispay the same output on some control say Edit control in runtime, is there any way??
Thanks in advance
Warm Regards,
R.Krishnan
|
|
|
|
|
Hello, after finally getting the Winsock Layered Service Provider sample from MSDN to work i'm now facing another problem. I modified the WSPRecvFrom function to be able to intercept UDP packets that are being read by an application. I then write the data to a logfile to see what is being read. But when i do this i end up with files that have vast planes of zero buffers in it, and inbetween this large pieces of zeros theres the data (so yea it works). My question is though, how can i get rid of these zeros, they are only in the way, is there a way to filter them out or am i missing something? Also, i thought when u intercept the data like this u only get like the exact data that is being read, and this is weird because i did the same for the WSPRecv function before and then opened my browser and i got like some binary data first, hows that possible? since the HTTP protocol is text based... I really don't get it. Here's how i modified the function so far:
inline INT
DPROVIDER::WSPRecvFrom(
IN SOCKET s,
IN LPWSABUF lpBuffers,
IN DWORD dwBufferCount,
IN LPDWORD lpNumberOfBytesRecvd,
IN OUT LPDWORD lpFlags,
OUT struct sockaddr FAR * lpFrom,
IN LPINT lpFromlen,
IN LPWSAOVERLAPPED lpOverlapped,
IN LPWSAOVERLAPPED_COMPLETION_ROUTINE lpCompletionRoutine,
IN LPWSATHREADID lpThreadId,
OUT INT FAR *lpErrno
)
{
INT ReturnValue;
if (PREAPINOTIFY(( DTCODE_WSPRecvFrom,
&ReturnValue,
m_lib_name,
&s,
&lpBuffers,
&dwBufferCount,
&lpNumberOfBytesRecvd,
&lpFlags,
&lpFrom,
&lpFromlen,
&lpOverlapped,
&lpCompletionRoutine,
&lpThreadId,
&lpErrno)) ) {
return(ReturnValue);
}
FILE* logf;
logf = fopen("D:\\lsplog.txt","a");
for(int i=0;i<(int)dwBufferCount;i++)
{
if(lpBuffers[i].len > 0)
{
fwrite("\r\n",strlen("\r\n"),1,logf);
fwrite(lpBuffers[i].buf,lpBuffers[i].len,1,logf);
}
}
fclose(logf);
ReturnValue = m_proctable->lpWSPRecvFrom(
s,
lpBuffers,
dwBufferCount,
lpNumberOfBytesRecvd,
lpFlags,
lpFrom,
lpFromlen,
lpOverlapped,
lpCompletionRoutine,
lpThreadId,
lpErrno);
POSTAPINOTIFY(( DTCODE_WSPRecvFrom,
&ReturnValue,
m_lib_name,
&s,
&lpBuffers,
&dwBufferCount,
&lpNumberOfBytesRecvd,
&lpFlags,
&lpFrom,
&lpFromlen,
&lpOverlapped,
&lpCompletionRoutine,
&lpThreadId,
&lpErrno));
return(ReturnValue);
}
Kuniva
--------------------------------------------
|
|
|
|
|
Does anyone know of a C/C++ utility/library which can convert a ddif document to a tif?
|
|
|
|
|
Hi all,
We're trying to make a different language version of our product. We've implemented dynamic .DLL resource loading.
The problem is editing the resource file, which has to be done by people outside the company (since none of us speak Japanese). VS .NET 2003 (and below) don't seem to allow us to enter Katakana characters, instead, it replaces them with ????. Even on Japanese systems. Even running Japanese Visual Studio.
We tried converting the resource file to UNICODE, and then we can have whatever characters we want in the file, but the UNICODE file can't be edited in Visual Studio - which won't do for translation purposes.
Does anyone have any experience with this sort of thing? Any third-party resource editor reccommendataions? Any experience trying to edit a resource file in a different locale?
Thanks,
Eco
|
|
|
|
|
... we needed the Japanese version of .NET to enter Japanese multibyte resources. Thanks, Microsoft.
Eco
|
|
|
|
|
Hi ,
im Trying to Make an Right-to-Left MFC Application , But I got a probelm which it cant be viewd correctly
I Tried the Right-to-Left Reading Order option under the extend Styles , but it failed and I always got it view like this ---> E?IE?
Any idea What Can I do to make it View correct ?
Thanks in advanced
|
|
|
|
|
E?IE? makes no sense. What are you trying to display?
The WS_EX_RTLREADING style is only supported for shell languages in Hebrew, Arabic, or another languages that support reading-order alignment.
Five birds are sitting on a fence.
Three of them decide to fly off.
How many are left?
|
|
|
|
|
yeah , But How can I manage to make my application write and read such letters ?
|
|
|
|
|
I'd imagine you'd have to install a version of Windows that supported Hebrew, Arabic, or another languages that support reading-order alignment.
http://support.microsoft.com/default.aspx?scid=kb;en-us;311019
Five birds are sitting on a fence.
Three of them decide to fly off.
How many are left?
|
|
|
|
|
i am writing a very simple program which allocates some memory for an array...
it is running under vc++ 6.0 or 7.1, on a athlon xp , 512 mb of ram, win xp pro. the big problem is that when i try to allocate something like 200 mb of ram memory (out of ~380 mb available), the program takes forever just to allocate the memory, and after that the computer begins to run extremely slow, even after i deallocate the memory. and if i run the program again and try to allocate memory again, windows will only allow me to allocate virtual memory, althought there's enough ram memory free that could be allocated... that's how i have written the program:
...
int main()
{
long size=25000;
long *array;
cout << "Will now allocate memory for the array..." << endl;
array=new long [size*sizeof(long)];
cout << "Memory allocated sucessfully..." << endl;
cout << "Will now deallocate the memory..." << endl;
delete[] array;
system("pause");
return 0;
}
i used a program (memtest) to test if this could be caused by defective ram memory, but i dont think thats the cause of that...
thanx
|
|
|
|
|
Anonymous wrote:
the program takes forever just to allocate the memory
What did you expect?
"It is better to remain silent and be thought a fool than to open one's mouth and remove all doubt." - Abraham Lincoln
|
|
|
|
|
Good God! Why do you need 25000 * 4 = 100,000 places for data?!?!?
If there is no other option for you here, why don't you break
it up into manageable chunks, say, I don't know, 100 blocks
of 1000 elements or something, then process each one on it's
own...
The slowness you are experiencing is due to the sheer size of the array
you are trying to create. You are trying to fill 200 MB of RAM,
so the operating system will attempt to swap the contents of your memory disk to make room for it all. Since you have the 200 MB of the array and the OS occupying the bulk of the rest, you are probably running short on RAM...
BTW: There's nothing wrong with your RAM here, just the pure size of the array you are trying to allocate.
John Theal
Physicist/Mathematical Programmer
Digital Immersion Software Corporation
Got CAD?
http://www.presenter3d.com[^]
http://www.merlin3d.com[^]
|
|
|
|
|
What are you trying to achieve here? Why would you need 25000 x 4 = 100000 longs? IMO it is unusual to see sizeof() within the brackets of a new [] statement.
When creating arrays, the number in the [] brackets, is the number of elements in the array. If you need 25000 longs then the line should read
long *array = new long [25000];
If you were using malloc, then the same allocation would be
long *array = malloc( 25000 * sizeof(long) );
Hope this might help.
|
|
|
|
|
i am writing a very simple program which allocates some memory for an array...
this is running on a athlon xp , 512 mb of ram, win xp pro. the big problem is that when i try to allocate something like 200 mb of ram memory (out of ~380 mb available), the program takes forever just to allocate the memory, and after that the computer begins to run extremely slow, even after i deallocate the memory. and if i run the program again and try to allocate memory again, windows will only allow me to allocate virtual memory, althought there's enough ram memory free that could be allocated... that's how i have written the program:
...
int main()
{
long size=25000;
long *array;
cout << "Will now allocate memory for the array..." << endl;
array=new long [size*sizeof(long)];
cout << "Memory allocated sucessfully..." << endl;
cout << "Will now deallocate the memory..." << endl;
delete[] array;
system("pause");
return 0;
}
i used a program (memtest) to test if this could be caused by defective ram memory, but i dont think thats the cause of that...
thanx
|
|
|
|
|
So what's your question? Allocating 200MB out of 380MB is a task that the memory manager is going to have to work at, especially if other applications are running. The more applications that are running, the more stuff will need to be swapped to disk before 200MB is freed up.
While the delete operator does free up the allocated memory (i.e., your application no longer owns it), the memory manager does not guarantee that it is immediately available again.
BTW, 25000 * 4 * 4 is only four hundred thousand, a trivial amount of RAM to obtain from the heap.
Five birds are sitting on a fence.
Three of them decide to fly off.
How many are left?
|
|
|
|
|
Hi
I have this dialog where the size can be adjusted.
So i've written some event handlers to handle the messages (vscroll, size & hscroll).
This works very good.
But what if the dialog is already too big to fit into the screen!?!!. Then there have to be scrollbars already.If i resize the dialog, then there are some scrollbars but at default not.
How can i work around this? or has anyone written a class derived from CDialog which implements automatic scrollbar options.
Greetings
|
|
|
|
|
Hi! Anyone knows how to programmaticly read one message from DBX file (Outlook Express) in my application. In OE all messages (in folder) are stored in one file. How can I take out one message ???
Sorry for my polish
Mila
|
|
|
|
|
If I use the NM_CLICK event to do a chkBox.GeCheck() (for a listCtrl)soon as I check the box, it claims that check is FALSE. When I check the second item, the first checkbox is stated to have been checked by the debugger. SO basically the checkmark seems to apppear AFTER the NM_CLICK is triggered. So what message do I want to handle to detect instantly that the checkbox got checked in the listCtrl?
Thanks,
ns
|
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
