We have run into a problem using <string> in place of <string.h>
Hence we need to be careful replacing that.

As a correct fix, we should be replacinf <string.h> with  <cstring>
and not <string>

Thanks,
Samisa...

On 7/22/05, Henrik Nordberg <[EMAIL PROTECTED]> wrote:
>  
> The two libraries that are incompatible are the old and the new io stream
> libraries. 
> <string.h> and <string> can be used together because they don't do the same
> thing. <string.h> contains the old C string functions (strcpy(), et al.) and
> <string> contains std::string (C++), and again, you can mix these freely
> (though we are now supposed to include <cstring> instead of <string.h>). 
>   
> So, where you really have to watch out is if you use the old stream headers:
> <iostream.h> should be replaced by <iostream> etc. But I did a quick search
> for iostream.h and it doesn't seem to be used in axis. 
>   
>  - Henrik
>  
>  
> ----- Original Message ----- 
> From: Fred Preston 
> To: [email protected] 
> Sent: Friday, July 22, 2005 2:00 AM 
> Subject: Using consistent libraries... changing from <???.h> to <???> 
> 
> 
> Hi All, 
>         We are using a combination of <string.h> and <string> in our code
> and this is now leading to problems in testing as the two libraries (at
> least the Microsoft C++ run-time libraries) are incompatible (check out
> http://msdn.microsoft.com/library/default.asp?url=/library/en-us/vccore98/html/_crt_c_run.2d.time_libraries.asp).
> 
> To avoid future problems, I vote that we move to the newer incantation of
> the libraries i.e. we move from <string.h> to <string>.  If anyone thinks
> that this may cause problems please let me know, but from the web page, it
> seems that we have little choice... 
> 
> P.S. 
>         I guess this also means changing <algorithm.h> to <algorithm>,
> <bitset.h> to <bitset>, <cassert.h> to <cassert>, <cctype.h> to <cctype>,
> <cerrno.h> to <cerrno>, <cfloat.h> to <cfloat>, <ciso646.h> to <ciso646>,
> <climits.h> to <climits>, <clocale.h> to <clocale>, <cmath.h> to <cmath>,
> <complex.h> to <complex>, <csetjmp.h> to <csetjmp>, <csignal.h> to
> <csignal>, <cstdarg.h> to <cstdarg>, <cstddef.h> to <cstddef>, <cstdio.h> to
> <cstdio>, <cstdlib.h> to <cstdlib>, <cstring.h> to <cstring>, <ctime.h> to
> <ctime>, <cwchar.h> to <cwchar>, <cwctype.h> to <cwctype>, <deque.h> to
> <deque>, <exception.h> to <exception>, <fstream.h> to <fstream>,
> <functional.h> to <functional>, <iomanip.h> to <iomanip>, <ios.h> to <ios>,
> <iosfwd.h> to <iosfwd>, <iostream.h> to <iostream>, <istream.h> to
> <istream>, <iterator.h> to <iterator>, <limits.h> to <limits>, <list.h> to
> <list>, <locale.h> to <locale>, <map.h> to <map>, <memory.h> to <memory>,
> <numeric.h> to <numeric>, <ostream.h> to <ostream>, <queue.h> to <queue>,
> <set.h> to <set>, <sstream.h> to <sstream>, <stack.h> to <stack>,
> <stdexcept.h> to <stdexcept>, <streambuf.h> to <streambuf>, <string.h> to
> <string>, <strstream.h> to <strstream>, <utility.h> to <utility>,
> <valarray.h> to <valarray> and <vector.h> to <vector>. 
> 
> Regards, 
> 
> Fred Preston.
>

Reply via email to