R
Ron Natalie
You can write one in a few lines. strtok bites because it maintains a
hidden internal state and that it modifies it's input.
hidden internal state and that it modifies it's input.
R
red floyd
Ron said:
IANALG (I am not a language guru), but I'd bet that your best bet is to write a
"Tokenizer" class. The advantage of that over "strtok" would be that (if done right)
each Tokenizer would have its own state, so you could have several different "parsings"
going on simultaneously, plus you wouldn't corrupt your original string.
M
Mike Wahler
I know this now, which is why I said my original plan was flawed (because I
*did* want to shorten the string...)
#include <iostream>
#include <string>
int main()
{
std::string s("blah");
std::string::size_type pos(s.find('a'));
if(pos != std::string::npos)
s.erase(pos);
std::cout << s << '\n'; /* prints "bl" */
return 0;
}
Recommendation:
http://www.josuttis.com/libbook/index.html
-Mike
A
Agent Mulder
AM> You produced some very ugly code, Mike ;-)7
MW> Why do you find it 'ugly'? What would you do instead
MW> that you'd consider less 'ugly'?
At least you took the effort to type code. That's good.
#include <iostream>
#include <algorithm>
#include <string>
int main(int argc,char**argv)
{
std::string s("Bach");
std::cout<<"\nBefore: "<<s.c_str();
std::replace(s.begin(),s.end(),'B','J');
std::replace(s.begin(),s.end(),'c','h');
std::replace(s.begin(),s.end(),'h','z');
std::cout<<"\nAfter: "<<s.c_str();
}
______
output
Before: Bach
After: Jazz
MW> Why do you find it 'ugly'? What would you do instead
MW> that you'd consider less 'ugly'?
At least you took the effort to type code. That's good.
#include <iostream>
#include <algorithm>
#include <string>
int main(int argc,char**argv)
{
std::string s("Bach");
std::cout<<"\nBefore: "<<s.c_str();
std::replace(s.begin(),s.end(),'B','J');
std::replace(s.begin(),s.end(),'c','h');
std::replace(s.begin(),s.end(),'h','z');
std::cout<<"\nAfter: "<<s.c_str();
}
______
output
Before: Bach
After: Jazz
D
Default User
I know this now, which is why I said my original plan was flawed (because I
*did* want to shorten the string...)
If you want to use std::string, you need to sit and study them
thoroughly. Stop worrying about C-style strings.
Brian Rodenborn
R
Ron Natalie
red floyd said:
Yep, and I believe if you google back a bit in this group for "stringtok" you'll find one has
been posted.
M
Mike Wahler
Agent Mulder said:
Why is my code 'uglier' than yours?
I find yours 'uglier' if for no other reason than
that you failed to indent. You also declared parameters
to main() which you never use. Yuck!
Why do you make the unnecessary call to 'c_str()'
for output?
(My computer can beat up your computer!
)-Mike
D
Default User
Nothing built-in. Here's a utility I wrote, it's not equivilent to
strtok() but rather ones like PHP explode():
#include <vector>
#include <string>
// breaks apart a string into substrings separated by a character string
// does not use a strtok() style list of separator characters
// returns a vector of std::strings
std::vector<std::string> Explode (const std::string &inString,
const std::string &separator)
{
std::vector<std::string> returnVector;
std::string::size_type start = 0;
std::string::size_type end = 0;
while ((end=inString.find (separator, start)) != std::string::npos)
{
returnVector.push_back (inString.substr (start, end-start));
start = end+separator.size();
}
returnVector.push_back (inString.substr (start));
return returnVector;
}
Brian Rodenborn
M
Mike Wahler
Default User said:
I guess I'll throw mine away now.
I started tohack one out, which was beginning to look very much
like this.
Thanks.
-Mike
A
Agent Mulder
Duh, I forgot the return statement:
#include <string>
#include <iostream>
#include <algorithm>
static std::string s="Bach";
int main(int argc,char**argv)
{
std::cout<<"\nBefore: "<<s.c_str();
std::replace(s.begin(),s.end(),'B','J');
std::replace(s.begin(),s.end(),'c','h');
std::replace(s.begin(),s.end(),'h','z');
std::cout<<"\nAfter: "<<s.c_str();
return 0;
}
______
output
Before: Bach
After: Jazz
#include <string>
#include <iostream>
#include <algorithm>
static std::string s="Bach";
int main(int argc,char**argv)
{
std::cout<<"\nBefore: "<<s.c_str();
std::replace(s.begin(),s.end(),'B','J');
std::replace(s.begin(),s.end(),'c','h');
std::replace(s.begin(),s.end(),'h','z');
std::cout<<"\nAfter: "<<s.c_str();
return 0;
}
______
output
Before: Bach
After: Jazz
D
Default User
Mike said:I guess I'll throw mine away now.
hack one out, which was beginning to look very much
like this.
I had a project last year that required parsing long strings (flight
plans) with nested delimiters, first "\r\n", then "..", then ":",
finally ",".
Brian Rodenborn
D
Default User
Explode? Oh, you mean like "get bigger" and not "make pretty exceptions all
over the place"
Like bust it into pieces.
Brian Rodenborn
D
Default User
Explode? Oh, you mean like "get bigger" and not "make pretty exceptions all
over the place"
Cool, thank you...
One major difference between that code and strtok() (besides string
separators vs. character delimiters) is the behavior with adjacent
delimiters.
Example, the string "one,two,,three" gives the tokens:
one
two
three
When using strtok().
And:
one
two
three
With the Explode() routine. That's by design and desire, so that "empty"
fields are represented, not just ignored.
Brian Rodenborn