|
What is looks like your missing is context.
To call a member function(method) it must be called in association with an object of the class it belongs to.
You declare, say
class A {
..
public:
void member_function(void);
...
};
you have to call it like this:
{
A anInstanceOfa( ... ); // create an instance of A
...
anInstanceOfa.member_function();
...
}
or maybe: A *ptrA = new A(.. )
ptrA->member_function();
the member_function is only meaningfull in the context of a specific object of
type A.
|
|
|
|
|
hia,
how do we reset a file pointer to the beginning of the stream?
if we have an input file stream fin, then the following code complains about ios_base. what do i do???
fin.seekg(0, ios_base::beq);
please help. thanx.
|
|
|
|
|
Hi !
Try ios_base::beg instead of ios_base::beq ... or it just was a misprint ?
Best regards,
-----------
Igor Soukhov (Brainbench/Tekmetrics ID:50759)
igor_soukhov@yahoo.com | ICQ:57404554 | http://siv.da.ru
|
|
|
|
|
hi,
you're right, i was misreading - and hence, mistyping - the beg thing. but it doesn't work that way either... the error message is
error C2653: 'ios_base' : is not a class or namespace name
i'm sure you know what it means. i guess i have to include something or the other but am not sure what. please oblige.
imran.
|
|
|
|
|
Hi again !
What headers files you've included into you source code where you're using ? (e.g. #include <fstream>).
Take at this - it was compiled OK by VC++6
#include <fstream>
using namespace std;
fstream file;
file.seekg(0, ios_base::beg);
Best regards,
-----------
Igor Soukhov (Brainbench/Tekmetrics ID:50759)
igor_soukhov@yahoo.com | ICQ:57404554 | http://siv.da.ru
|
|
|
|
|
Check "Differences in iostream Implementation" in the Visual C++ Programmer's Guide
If you are using the old iostream library ( <fstream.h> ) you have to use ios::beg instead.
If you are using the standard library ( <fstream> ) you can use either.
|
|
|
|
|
i have a problem maybe you will be able to solve it, i have a pointer to the internet explorer i want to copycat the ctrl+n key and make another window, but hidden ..
that's means to take the first document(+cookie and other things) and pass it to a hidden internet explorer that still i will have a pointer to him.
do you have any idea?
thanks,
|
|
|
|
|
Hi All,
Again problem with Threads.In an application I am creating three threads each thread trying to call a interface function(This function tries to add records to a Database) of a Component.CCI is succeeding ,I am getting Interface pointer.Since each Thread tries to add the Record there should be 3 Records in Database.But, there are only 2 records.I am using CreateThread(...) for creating the Threads.In the Debug mode there is no such problem.There are 3 Records added to the Database.Also, if in this scenario I give Sleep(0) then no problem .
Where am I going wrong ?
Thanx in advance for the help.
Regards,
yamini
|
|
|
|
|
What kind of database is it, if it's sql server, maybe try to put a store procedure and then the interface should call it.
|
|
|
|
|
Hi,
I am using MS Access Database(no stored queries).I am sure the problem is not with Interface pointer as while Debugging the program works perfectly.So,pls tell me if there is problem with some other part
if the application I have mentioned .
|
|
|
|
|
Need to know a bit more. Is the component shared among threads? What threading model is being
used? Is a shared DB connection being used? One thing that springs to mind about debug ~vs~ release is the speed. As release is faster, it may be causing a problem. How does Access do
its locking, is it per row, table or block? if per table or block you may need
to synchronise access to the DB to prevent collisions.
HTH.
|
|
|
|
|
Without knowing all of the facts here is my guess:
This sounds like a classic "race" condition. The database record lock and your writes in the threads collide with each other. To prove if this is the case try adding three different, long delays to the threads (e.g. Sleep( 100 ), Sleep( 200 ), Sleep( 300 ). If this works, your database and interface is OK and you have a timing problem. You must synchronise the writes or retry on error.
Have Fun!!!
|
|
|
|
|
Hello every body,
My questio is: how can I insert in a View in a Dialog Window?
Thank you for your help
Giancarlo 
Giancarlo Amati --> amati@iei.pi.cnr.it
|
|
|
|
|
What kind of View? A view as in Doc/View?
Michael 
|
|
|
|
|
Yes,
I need to build a Dialog in which I would build a view to draw or write something. For example a preview o something similar.
Thank
Giancarlo
|
|
|
|
|
Have a look at http://www.codeproject.com/miscctrl/customcontrol.asp. It should give you a good starting point.
Michael
|
|
|
|
|
Chaps and fellow geeks,
I have a pointer to an array of 10 chars (note that this is different from an array of ten chars, which allocates 10 bytes, the former allocates 4 bytes). This means, whenever I increment the pointer, I can skip by 10 bytes. I then proceed to allocate 100 bytes of storage (as I want to treat it as a 10 by 10 array). Trouble is (well, not really trouble), when I use new to allocate 100 bytes for the pointer I need to typecast the pointer returned by the new operator to the type of "pointer to an array". Towards this end, I have typedefed CHARTENARRAY as a pointer to an array of 10 bytes. I then typecast the return value of new like so: pArray = (CHARTENARRAY) new char[100];
I would like to know if there is a way of doing this *without* the typedef. How would I typecast it? In general, what would be the cast for a pointer to an array of x bytes that is returned, by the new operator when it is used to allocate x bytes?
To make things clear, here is the source (with lots of comments), you can compile it and run it if you'd like to experiment:
typedef char (*CHARTENARRAY)[10]; // typedefines CHARTENARRAY to be a pointer to an array of 10 b
CHARTENARRAY pArray = NULL;
pArray = (CHARTENARRAY) new char[100]; // need to do this WITHOUT the (CHARTENARRAY) typedef.
for(int i = 0; i < 10; ++i)
{
strcpy(*pArray, "HELLO!"); // Fill in 10 "HELLO!" strings in 10 slots
++pArray; // increment by 10 bytes (scalar)
}
pArray -= 10; // Go to array start (will decrement by 100 bytes, since scalar is 10)
for(i = 0; i < 10; ++i) // Display routine
{
puts(*pArray);
++pArray; // Go to next string
}
pArray -= 10; // Go to array start
delete [] pArray; // Free
|
|
|
|
|
Don't try to do it without the typedef. Using the typedef makes the code easier to understand.
Well, ok, to answer, I think the cast would be (char (*)[10]). But honestly, only Bjarne Stroustroup would know what that means.  Stick with the typedef.
--Mike--
http://home.inreach.com/mdunn/
#include "buffy_sig"
|
|
|
|
|
Won't forget that in a hurry-it works, thanx Mike. I did try (char*)[10] - this didn't work, though I thought that was intuitive Looks like the right to left rule of reading declarations might come in handy even in areas such as this!
Cheers,
Shanker.
|
|
|
|
|
#include <iostream.h>
#include <string.h>
void main()
{
char *pstr="test" ;
cout<<strrev(pstr)<<endl;
}
This program isn't executed in VC++6.0, but in C++ Builder5.0, Why?
|
|
|
|
|
#include <iostream.h>
#include <string.h>
void main()
{
char *pstr="test" ;
cout<<strrev(pstr)<<endl;
}
OK, then Borland is not comforming to C. You wrote code that is attempting to modify a constant string. _strrev actually modifies the character array/pointer, so an error occurs here because you are modifying a constant. The correct way of doing this problem would be:
#include <iostream.h>
#include <string.h>
void main()
{
char str[] = "test";
cout << _strrev(str) << endl;
}
Frank
|
|
|
|
|
Thanks, Frank.
There is another question .
The program taken from an article "A Sample Generic-Text Program" from MSDN isn't executed in VC++ 6.0 , Whether it modifies a constant string either?
#include <stdio.h >
#include <stdlib.h >
#include <string.h >
#include <direct.h >
#include <errno.h >
#include <tchar.h >
int __cdecl _tmain(int argc, _TCHAR **argv, _TCHAR **envp)
{
_TCHAR buff[_MAX_PATH];
_TCHAR *str = _T("Astring");
char *amsg = "Reversed";
wchar_t *wmsg = L"Is";
#ifdef _UNICODE
printf("Unicode version\n");
#else /* _UNICODE */
#ifdef _MBCS
printf("MBCS version\n");
#else
printf("SBCS version\n");
#endif
#endif /* _UNICODE */
if (_tgetcwd(buff, _MAX_PATH) == NULL)
printf("Can't Get Current Directory - errno=%d\n", errno);
else
_tprintf(_T("Current Directory is '%s'\n"), buff);
_tprintf(_T("'%s' %hs %ls:\n"), str, amsg, wmsg);
_tprintf(_T("'%s'\n"), _tcsrev(str));
return 0;
}
|
|
|
|
|
Because when you declare:
TCHAR* str = "blah blah";
the literal string "blah blah" is stored in the executable. When loaded into memory, that string may, at the whim of the OS, be stored in a read-only section of memory. Thus your program fails when it tries to modify the string.
--Mike--
http://home.inreach.com/mdunn/
#include "buffy_sig"
|
|
|
|
|
Same Problem.
_TCHAR *str = _T("Astring");
is a pointer to a string constant. What you really want is
_TCHAR str[] = _T("Astring");
This will store the string into an array. Therefore, it allows _tcsrev (or now _wcsrev) to modify the array. Of course, I can't say that it's "ANSI C" because this is a MS-specific type.
Heh, it seems that the sample program is trying to get away with this business. In practice, you should never modify a pointer to a constant string.
Frank
|
|
|
|
|
#include <iostream.h>
#include <string.h>
void main()
{
char *pstr = new char[5];
strcpy(pstr,"test");
cout<< strrev(pstr) <
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
