Hello,

I've been searching for some way to get all possible substrings out of
an input string.  I've found information and scripts on suffix trees,
but the examples I found only return an array of suffixes (hence the
name I guess)...that is, substrings from progressively splitting off
an initial letter from the string.  Has anyone run across a C++
implementation of perl's String::Splitter function?  That's the kind
of functionality that I'm looking for.

Thanks,

Robert

On 26 =D1=8F=D0=BD=D0=B2, 03:14, Robert Reno <nebnetjeru...@gmail.com> wrot=
e:
> Hello,
>
> I've been searching for some way to get all possible substrings out of
> an input string.

Something like this:

$fullLength=3Dstrlen($string);
for($begin=3D0;$begin<$fullLength;++$begin)
 for($l=3D$fullLength-$begin;$l>0;--$l)
  echo substr($string,$begin,$l)."\n";

On 26 =D1=8F=D0=BD=D0=B2, 11:04, "criman...@googlemail.com"
<criman...@googlemail.com> wrote:
> Something like this:
>
> $fullLength=3Dstrlen($string);
> for($begin=3D0;$begin<$fullLength;++$begin)
> =C2=A0for($l=3D$fullLength-$begin;$l>0;--$l)
> =C2=A0 echo substr($string,$begin,$l)."\n";

Oh, no! I was in PHP conference now and this is result. %-)

On Jan 26, 9:14=A0am, Robert Reno <nebnetjeru...@gmail.com> wrote:
> Hello,
>
> I've been searching for some way to get all possible substrings out of
> an input string. =A0I've found information and scripts on suffix trees,
> but the examples I found only return an array of suffixes (hence the
> name I guess)...that is, substrings from progressively splitting off
> an initial letter from the string. =A0Has anyone run across a C++
> implementation of perl's String::Splitter function? =A0That's the kind
> of functionality that I'm looking for.

Have a look at boost's tokenize library.

Cheers,
Tony

Robert Reno wrote:
> I've been searching for some way to get all possible substrings out of
> an input string.  I've found information and scripts on suffix trees,
> but the examples I found only return an array of suffixes (hence the
> name I guess)...that is, substrings from progressively splitting off
> an initial letter from the string.  Has anyone run across a C++
> implementation of perl's String::Splitter function?  That's the kind
> of functionality that I'm looking for.

I'm not sure I understood you correctly -- is this what you need?

#include <cassert>
#include <algorithm>
#include <vector>
#include <string>

typedef std::vector<std::string> substrVector;
typedef std::pair<substrVector, substrVector> substrHelperType;

class Prepend : public std::unary_function<std::string, std::string> {
    private: std::string prefix_;
    public:
    Prepend(const std::string &prefix): prefix_(prefix) {}
    std::string operator() (const std::string &str) {
        return prefix_ + str;
    }
};

substrHelperType substringsHelper(const std::string &str) {
    size_t len = str.length();
    assert(len > 0);
    std::string x(str.begin(), str.begin() + 1);
    if (len > 1) {
        substrHelperType rest =
            substringsHelper(std::string(str.begin() + 1, str.end()));
        substrVector p(1, x), np(rest.first);
        p.resize(rest.first.size() + 1);
        std::transform(rest.first.begin(), rest.first.end(),
            p.begin() + 1, Prepend(x));
        np.resize(np.size() + rest.second.size());
        std::copy(rest.second.begin(), rest.second.end(),
            np.end() - rest.second.size());
        return substrHelperType(substrVector(p), substrVector(np));
    } else {
        return substrHelperType(substrVector(1, x), substrVector());
    }
}

substrVector substrings(const std::string &str) {
    substrHelperType res = substringsHelper(str);
    res.first.resize(res.first.size() + res.second.size());
    std::copy(res.second.begin(), res.second.end(),
        res.first.end() - res.second.size());
    return substrVector(res.first);
}

-- 
P. Lepin