•
•
•
•
What is DaniWeb IT Discussion Community?
You're currently browsing the C++ section within the Software Development category of DaniWeb, a massive community of 391,596 software developers, web developers, Internet marketers, and tech gurus who are all enthusiastic about making contacts, networking, and learning from each other. In fact, there are 2,706 IT professionals currently interacting right now! Registration is free, only takes a minute and lets you enjoy all of the interactive features of the site.
Please support our C++ advertiser:
Views: 4852 | Replies: 7
 |
•
•
Join Date: Aug 2005
Posts: 10
Reputation: 
Rep Power: 4
Solved Threads: 0
Newbie Poster
Dear all:
I've created TCPsocket in linux using C++. It receives an array of unsigned characters. I need the corresponding hex values in character format.
i.e. "ABC" in hex it is "414243". Now I want to treat each of "414243" as character.
So I have done the coding as follow ( taking into cosideration that the TCP socket is established and the data is received in an unsigned character pointer m_cpSockData sized same to Buffer Size of the socket. And the number of byte read is m_iReadLen).
<< moderator edit: added [code][/code] tags >>
I also have similar thing while writing back to the socket.
But this reduces the efficiency of the code..
Can any body help to provide a mechanism where efficiency can be increased.
I've created TCPsocket in linux using C++. It receives an array of unsigned characters. I need the corresponding hex values in character format.
i.e. "ABC" in hex it is "414243". Now I want to treat each of "414243" as character.
So I have done the coding as follow ( taking into cosideration that the TCP socket is established and the data is received in an unsigned character pointer m_cpSockData sized same to Buffer Size of the socket. And the number of byte read is m_iReadLen).
char* m_cHexvalue = new char[2*BufSize +1];
char *m_cCurrCharVal=new char[3];
*(m_cCurrCharVal+2)=0X00;
memset(m_cHexvalue ,0,2);
for(int count=0;count<BufSize ;count++)
{
memset(m_cCurrCharVal,0,2);
sprintf(m_cCurrCharVal,"%x",*(m_cpSockData+count));
strcat(m_cHexvalue ,m_cCurrCharVal);
}I also have similar thing while writing back to the socket.
But this reduces the efficiency of the code..
Can any body help to provide a mechanism where efficiency can be increased.
•
•
Join Date: Apr 2004
Posts: 3,449
Reputation: 
Rep Power: 16
Solved Threads: 138
This is pretty much the same, but it's all I've got at the moment.
#include <stdio.h>
char *foo(char *dst, const char *src)
{
char *start = dst;
while ( *src )
{
dst += sprintf(dst, "%x", (unsigned)*src++);
}
return start;
}
char *bar(char *dst, const char *src)
{
char *start = dst;
while ( *src )
{
unsigned value;
if ( sscanf(src, "%2x", &value) == 1 )
{
src += 2;
*dst++ = value;
}
}
*dst = '\0';
return start;
}
int main(void)
{
const char abc[] = "ABC";
char over [ sizeof abc * 2 - 1 ], back [ sizeof abc ];
printf("\"%s\" -> \"%s\"\n", abc, foo(over, abc));
printf("\"%s\" -> \"%s\"\n", over, bar(back, over));
return 0;
}
/* my output
"ABC" -> "414243"
"414243" -> "ABC"
*/ Last edited by Dave Sinkula : Aug 19th, 2005 at 6:27 pm. Reason: Added 'bar'.
•
•
Join Date: Aug 2005
Posts: 10
Reputation:
Rep Power: 4
Solved Threads: 0
Newbie Poster
Thanks for the reply.
I understand you have removed some operations. That reduces quite a big time.
I think this can further be minimized. Do you have any idea of doing this with the help of asm codes in C.
I've not used so far but I think that may help the roundtrip time.
Thanks again.
Sutanu
I understand you have removed some operations. That reduces quite a big time.
I think this can further be minimized. Do you have any idea of doing this with the help of asm codes in C.
I've not used so far but I think that may help the roundtrip time.
Thanks again.
Sutanu
•
•
Join Date: Apr 2004
Posts: 3,449
Reputation:
Rep Power: 16
Solved Threads: 138
Perhaps discard the standard routines and roll your own. This is for ASCII, and I think it is right. But I was wrong with it a minute ago.
#include <stdio.h>
#include <ctype.h>
unsigned char *foo(unsigned char *dst, const unsigned char *src)
{
static const unsigned char nibble[] =
{
'0','1','2','3','4','5','6','7','8','9','A','B','C','D','E','F'
};
unsigned char *start = dst;
while ( *src )
{
*dst++ = nibble [ *src >> 4 ];
*dst++ = nibble [ *src++ & 0xF ];
}
*dst = '\0';
return start;
}
unsigned char *bar(unsigned char *dst, const unsigned char *src)
{
unsigned char *start = dst;
while ( *src )
{
if ( isdigit(*src) )
{
*dst = *src++ - '0';
}
else
{
*dst = *src++ - 'A' + 10;
}
*dst <<= 4;
if ( isdigit(*src) )
{
*dst++ += *src++ - '0';
}
else
{
*dst++ += *src++ - 'A' + 10;
}
}
*dst = '\0';
return start;
}
int main(void)
{
const unsigned char abc[] = "ABCMNO";
unsigned char over [ sizeof abc * 2 - 1 ], back [ sizeof abc ];
printf("\"%s\" -> \"%s\"\n", abc, foo(over, abc));
printf("\"%s\" -> \"%s\"\n", over, bar(back, over));
return 0;
}
/* my output
"ABCMNO" -> "4142434D4E4F"
"4142434D4E4F" -> "ABCMNO"
*/ Last edited by Dave Sinkula : Aug 19th, 2005 at 11:23 pm. Reason: Got rid of temporary variables.
•
•
Join Date: Aug 2005
Posts: 10
Reputation:
Rep Power: 4
Solved Threads: 0
Newbie Poster
Thanks again.
I performed a dry run. I think it would be better to use
if (*src <= '9')
rather than
if (isdigit(*src))
What do you think...
Sutanu
I performed a dry run. I think it would be better to use
if (*src <= '9')
rather than
if (isdigit(*src))
What do you think...
Sutanu
•
•
Join Date: Apr 2004
Posts: 3,449
Reputation:
Rep Power: 16
Solved Threads: 138
•
•
•
•
Originally Posted by Sutanu
I performed a dry run. I think it would be better to use
if (*src <= '9')
rather than
if (isdigit(*src))
What do you think...
Sutanu
•
•
Join Date: Aug 2005
Location: near St Louis, Missouri, USA
Posts: 10,546
Reputation:
Rep Power: 36
Solved Threads: 860
if( *src <= '9' )
what if *src < '0' ? isdigit() will catch that, but your if statement won't.
•
•
Join Date: Aug 2005
Posts: 10
Reputation:
Rep Power: 4
Solved Threads: 0
Newbie Poster
Thanks for the concern.
The scenario I'm talking about doesn't permit you to have any values other than 0-9 and A-F.
In other scenarios or to have a generic one I think using a isdigit() is a better option. But then also you can see what happens if any character as * ( ASCII value is 42) comes . Then you probably will have a negative value (42 - 64 +10 = -12).
The scenario I'm talking about doesn't permit you to have any values other than 0-9 and A-F.
In other scenarios or to have a generic one I think using a isdigit() is a better option. But then also you can see what happens if any character as * ( ASCII value is 42) comes . Then you probably will have a negative value (42 - 64 +10 = -12).
|
•
•
•
•
Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
•
•
•
•
•
•
•
•
DaniWeb C++ Marketplace
Linear Mode
Similar Threads
Other Threads in the C++ Forum
- Previous Thread: what is classes?
- Next Thread: Urgent Pls Help: My codes are in errors!
