C Board

Resize console window

4c0 — Wed, 16 May 2012 22:07:31 GMT

How to change console resolution to 640x480 when application start in ubuntu using C?Is that possible with ncurses library?

Advice

TwoBabyJedis — Wed, 16 May 2012 20:06:13 GMT

I've been studying C++ for a few months, but still am in the beginning stages. I've decided to pursue my bachelors. There is a nagging question on my mind I must ask. Anyway, the degree is not a BSCS, but a general purpose CIS/business degree. My question... armed with a shiny bachelors degree and some in depth self study of C++ over the next 3 years (Im anticipating on atleast reaching a very solid intermediate level based on my learning curve) am I marketable for a decent paying programming job? Whats are some things I can do to enhance my resume? I have zero work experience in the field and cant leave my current job for helpdesk and such. I got a Network+ certification for the heck of it, but its only worth making into a paper airplane. Thanks for your help. If my post is in the wrong area or annoying, please delete it.

Which Gui toolkit should I use.

freiza — Wed, 16 May 2012 19:29:53 GMT

What gui tool kit should I use?

For now I my project is windows based. And I will stick to windows for atleast 1 year. I know only C++ and Win32 api. My project is not a simple database application. It is heavily relying on windows memory management , process management and diagnostic function. So I have to use win32.

So kindly suggest what Gui tool kit should I use with my project so that I have least gui overhead.

Qt or WxWidgets or MFC or WTL ?

8086 emulator CPU command verification files (testsuite)?

superfury — Wed, 16 May 2012 19:13:18 GMT

Are there any data files (load at either 0x0000:0x7C00 or 0x0000:0x0000) for testing a 8086 CPU emulator for errors?
I've found one testsuite for the 80186 here.
Is there any for the 8086 too?

Somehow when I try to load DOS 3.1 on my emulator (boot sector loaded at 0x0000:0x7C00), it loops with a labelx: PUSH CX, LOOPNZ labelx. Since CX>(0x1000/2), this would result in a stack overflow (with the stack overwriting the opcodes), since CS=DS=ES=SS.

Anybody knows how to find the bugs in the 8086 emulated CPU?
Or does anyone has a testsuite for 8086 emulator opcodes?

keypress simulation needed to complete process automation

Annonymous — Wed, 16 May 2012 19:06:12 GMT

Hey y'all. I wrote a daemon script to launch a second script that is responsible for monitoring/controlling a UI program. The problem is that if i run the daemon script on the command line like so: sh /path/to/daemon_script.sh, the monitor script causes the UI to not work correctly. Now if I run the contents of the daemon script manually on the command line like so: screen watch -n 60 /path/to/monitor_script.sh it works properly and runs every 60 seconds like it is supposed to. So, I am going to make a c program to see if it will work that way. I plan on using execl calls or something similar. Now here is where the problem lies; In order for the process to become daemonized, I need to manually hit control + A and control + D. That is what the screen command is for. Is there a way to simulate these key presses with C? I'll even settle for python, but would prefer C since that is what my entire program is based off of.

Even a better way to do this would be cool. Suggestions, ideas?

Thanks!

Simple Routing Table Performance

SMurf — Wed, 16 May 2012 18:45:19 GMT

Hello,

I'm looking at creating a routing system for use with an inter-thread communication (asynchronous method calling) framework that I am working on.

I have a function that is analogous to Windows's PostMessage. It dispatches method calls to the incoming queue of the appropriate instance. That instance may be a single thread doing the work of a specific instance, or it may be a thread that handles multiple instances. In the latter case, I need a way to quickly find which thread is responsible for a given instance.

The best thing that I can come up with at the moment is a per-thread structure that stores the thread ID for a given instance ID. But what is the fastest way to perform a search (this function would be called several times on average per method call, per thread)?

A binary tree initially came to mind, but I'm not sure. This is a fairly simple framework (number of instances will never exceed 255, instance IDs are generated sequentially until they wrap around), so the work required to maintain a balanced tree for each thread may not be worth it. Would it be better to simply have an array of 255 entries and index directly into it (would cost a couple KB in memory per thread)?

I don't think actual network routers use trees, do they?

What about some sort of structure where each successful search for an instance ID promotes that ID, making it faster to find it the next time?

Sorry about the questions, but this is probably the most performance critical aspect of what I'm doing. :o

Basic pointers to structures question...

superbbrr — Wed, 16 May 2012 17:31:00 GMT

header:

Code:

struct termlist

{

      uint32_t num_terms; 

      char **term;           

};



int getMetaData( uint32_t specific_tag, termlist **terms);

main:

Code:

int main(int argc, char **argv)

{



uint32_t specificTag;

specificTag = atoi(argv[1]);



termlist *sample1;



getMetaData( specificTag, sample1 );



...



}

Hi, I'm really confused about pointers to structures. I'm trying to understand this snippet of code above.

The function prototype requires a double pointer to the structure, termlist **terms. But in main when calling the function only a pointer to the structure is passed, sample1.

I'm quite confused on how the function prototype requires a double pointer, but in main only a single pointer is passed.

Thanks!

redefinition of 'main' error

Gil Carvalho — Wed, 16 May 2012 16:49:08 GMT

Hi

I have tow files with each one int main() function, the second file has an header file, when i compile i have this error

redefinition of 'main' error

I can't use main in both source files??

Regards

Float that initializes to NaN

then — Wed, 16 May 2012 15:22:29 GMT

Hi! I want to make a class that works exactly like a float, only that it is initialized to NaN when the default constructor is called and if DEBUG is defined so that I can easily detect when I have an uninitialized float. Is there some simple way to do this? Thanks in advance!

Sending Data to Excel with C++

chopshardiman — Wed, 16 May 2012 14:38:38 GMT

Hi folks,

I'm working on a program that reads a text file which has columns of dates, times and associated values, then sends the columns to an excel sheet. The entire file follows the same pattern.

Text file sample:

Code:

2012-05-14 14:08:12.498 8 InfoMSG1   N  2.719  7.117 11.335 12.795 16.228 16.619 18.117 19.608 20.899   6883      0   2994   2153   2089      0 {InfoMSG2}

2012-08-14 14:08:38.644 8 InfoMSG3 N  2.974  6.732  9.577 10.596 13.355 13.737 14.871 16.327 17.641   7142      0   1399   1709   1637      0 {InfoMSG4}

At the moment, I'm using a fairly inelegant solution of breaking each sentence into tokens delimited by whitespace and storing each column in a separate array as follows:

Code:

    string            line;

    string            *col1, *col2, *col4, *col5, *col21;        





        col1 = new string[fileLength];

    col2 = new string[fileLength];

    col3 = new LONG[fileLength];

    col4 = new string[fileLength];

    col5 = new string[fileLength];

    col6 = new FLOAT[fileLength];

    col7 = new FLOAT[fileLength];

    col8 = new FLOAT[fileLength];

    col9 = new FLOAT[fileLength];

    col10 = new FLOAT[fileLength];

    col11 = new FLOAT[fileLength];

    col12 = new FLOAT[fileLength];

    col13 = new FLOAT[fileLength];

    col14 = new FLOAT[fileLength];

    col15 = new FLOAT[fileLength];

    col16 = new FLOAT[fileLength];

    col17 = new FLOAT[fileLength];

    col18 = new FLOAT[fileLength];

    col19 = new FLOAT[fileLength];

    col20 = new FLOAT[fileLength];

    col21 = new string[fileLength];





while(getline(file, line))

    {

        std::stringstream    linestream(line);

                

        //Reading space separated columns into string arrays

        linestream >> col1[i] >> col2[i] >> col3[i] >> col4[i] >> col5[i] >> col6[i] >> col7[i] >> col8[i] >> col9[i] >> col10[i] >> col11[i];

        linestream >> col11[i] >> col13[i] >> col14[i] >> col15[i] >> col16[i] >> col17[i] >> col18[i] >> col19[i] >> col20[i] >> col21[i];





        i++;

    }

In my research of the COM interface process from C++ to Excel I haven't been able to find a method of passing single byte strings to excel. The only method I've been able to have any success with is detailed in this MSDN article - How to automate Excel from C++ without using MFC or #import

It uses a VARIANT safearray to pass data to Excel, and the only way I can see to use my arrays as above is to convert the std::strings in each element first to wstrings, then to BSTR.

I would think there's probably a better way to do this conversion but haven't come up with one.

Anyway, my conversion form single byte to wide string is achieved using the following function:

Code:

wstring stws(const string &src_string)

{

    size_t src_len = src_string.length();





    if(0 == src_len)

        return L"";





    wchar_t *buf = new(std::nothrow) wchar_t[src_len + 1];





    if(0 == buf)

        return L"";





    mbstowcs(buf, src_string.c_str(), src_len);

    buf[src_len] = L'\0';





    wstring final_string = buf;





    if(0 != buf)

        delete [] buf;





    return final_string;

}

The further conversion from wide string to BSTR is carried out like so:

Code:

    

    wcol1 = new wstring[fileLength];

    wcol2 = new wstring[fileLength];

    wcol4 = new wstring[fileLength];

    wcol5 = new wstring[fileLength];

    wcol21 = new wstring[fileLength];





    bcol1 = new BSTR[fileLength];

    bcol2 = new BSTR[fileLength];

    bcol4 = new BSTR[fileLength];

    bcol5 = new BSTR[fileLength];

    bcol21 = new BSTR[fileLength];





for(int i=0; i<10; i++)

    {

        wcol1[i] = stws(col1[i]);

        wcol2[i] = stws(col2[i]);

        wcol4[i] = stws(col4[i]);

        wcol5[i] = stws(col5[i]);

        wcol21[i] = stws(col21[i]);





        bcol1[i] = SysAllocString(wcol1[i].c_str());

        bcol2[i] = SysAllocString(wcol1[i].c_str());

        bcol4[i] = SysAllocString(wcol1[i].c_str());

        bcol5[i] = SysAllocString(wcol1[i].c_str());

        bcol21[i] = SysAllocString(wcol1[i].c_str());

    }

I've printed the BSTR values to the console using wcout to confirm that they are correct.

Now, to pass this data into an excel sheet. I'm using the method outlined in the MSDN article linked above, creating a safe array with my list of values from the BSTR arrays:

Code:

    

// Create a safearray of variants...

    VARIANT arr;

    arr.vt = VT_ARRAY | VT_VARIANT;

    {

        SAFEARRAYBOUND sab[2];

        sab[0].lLbound = 1; sab[0].cElements = 10;

        sab[1].lLbound = 1; sab[1].cElements = 1;

        arr.parray = SafeArrayCreate(VT_VARIANT, 2, sab);

    }





// Fill safearray with some values...

    for(int i=0; i<10; i++) {

        for(int j=0; j<2; j++) {

            // Create entry value for (i,j)

            VARIANT tmp;

            tmp.vt = VT_BSTR;

            tmp.bstrVal = bcol2[i];

            // Add to safearray...

            long indices[] = {i,j};

            SafeArrayPutElement(arr.parray, indices, (void *)&tmp);

        }

    }

Then I do the usual COM tasks, creating a new Workbook, acquiring the Active Worksheet, getting a range object, before setting the values in the range as follows:

Code:

    

// Set range with our safearray...

    AutoWrap(DISPATCH_PROPERTYPUT, NULL, pXlRange, L"Value", 1, arr);

For the scenario described above, where I only copy a single column to excel at a time, what I have does the trick.

My major problem at this stage is finding a way to loop in the two dimensions of the safearray to copy the rest of the columns at the same time. I know I can do it by creating a safearray for each column following a similarly inelegant approach to how I dealt with splitting the data into columns in the first place, but I'm sure there's a better way to do it than that.

Any help would be greatly appreciated, ty.

Debugging help, pointer forwarding

Hoang Bao Ngo — Wed, 16 May 2012 14:31:55 GMT

There's the code under, but I'm getting an invalid read size of 4, and it seems because I forwarded the pointer a little far ahead... I think.
It's just a little part of my network code, but it's pretty easy to understand what I'm doing under right?

LSP_buffer is a char *, that's why I have to cast it.

allocated memory for a struct LSP and how many counts edges there is around this router(nodes).

Not so sure if I thought it out right, by adding sizeof(struct LSP) to pointer it would mean forwarding the pointer by so many bytes as the size of struct right?, why then would I get "Invalid write of size 4" ?

thanks for replying.

Code:



   LSP_buffer = malloc((sizeof(struct LSP) + (sizeof(struct EDGE) * count))); 

    lsp = (struct LSP*) LSP_buffer;                                        /*cast it to struct LSP and write in the necessary info about this LSP*/

    lsp->source = htonl(own);

    lsp->seq = htonl(LSP_generation++);                                    /*set which generation this LSP is, and increment it*/

    lsp->nodes = htonl(count);

    node = (struct NODE *) (LSP_buffer + sizeof(struct LSP));                /*cast LSP_buffer to struct EDGE and move the pointer forward so we don't overwrite any data*/

   

    fprintf(stderr, "Own address: %d LSP generation is %d\n", own , LSP_generation);

    



    /*store info about the edges around this router*/

    for(i = 0; i < MAX_ADDRESSES; i++){     

      if(distance[own][i].cost > 0){

    node->address = htonl(i);   <---------------------------    

    node->cost = htonl(distance[own][i].cost);

    node = node + sizeof(struct NODE);                                /*forward pointer by sizeof struct NODE so we don't overwrite old NODE data*/

      }

    }

how to create a 3 x 3 matrix

kavitabix — Wed, 16 May 2012 12:10:03 GMT

how do i create a 3 x 3 matrix and find the sum of elements in all rows?

how to count number of strings being used in an array of strings

kiwisfan — Wed, 16 May 2012 11:44:56 GMT

hi everyone, my title seems confusing but i didnt know how to word it any other way so ill explain it a bit better here.

i have declared an array of strings to be,dir[30][13]. Now my program is supposed load file names into those strings and i want to know how i can count the number of strings being used in that array of strings.

ive tried doing while(dir[i]!=NULL) and various other combinations of functions but i cant think of how to solve this problem.

In my program if i want to add a new file to the dir with user input and the dir is full (all string arrays are being used) then an error message of "directory is full" will show up.

your help is appreciated

Hi!

Nogmor — Wed, 16 May 2012 10:46:17 GMT

Whats up? Frist time learning C should be interesting. starting a years course in development, should include alot of C, C++, java and suchlikes.

Anyways, this is what I came up with on the first day of C. :D

cya around

Attached Images

Books on Data Structures

smoker91 — Wed, 16 May 2012 10:16:36 GMT

hello guys,
Can anyone please recommend me some good books on Data Structures Using C Language.
Thank you.
Please help me ASAP. :)