实现字符串任意分割,如下程序,并有使用举例。
#include <vector>
#include
<string>
#include <iostream>
#include
<algorithm>
using namespace std;
typedef basic_string<char>::size_type S_T;
static const
S_T npos = -1;
////trim指示是否保留空串,默认为保留。tok可以为任意多个字符
vector<string> tokenize(const
string& src, string
tok,
bool trim=false, string
null_subst="")
{
if(
src.empty() || tok.empty() )
throw "tokenize: empty
string/0";
vector<string> v;
S_T pre_index = 0,
index = 0, len = 0;
while( (index =
src.find_first_of(tok, pre_index)) !=npos )
{
if( (len =
index-pre_index)!=0
)
v.push_back(src.substr(pre_index,
len));
else
if(trim==false)
v.push_back(null_subst);
pre_index = index+1;
}
string endstr =
src.substr(pre_index);
if( trim==false )
v.push_back( endstr.empty()?null_subst:endstr
);
else if( !endstr.empty()
)
v.push_back(endstr);
return
v;
}
//delimit为一个字符,严格分割
vector<string>
split(const string& src, string delimit, string
null_subst="")
{
if(
src.empty() || delimit.empty()
)
throw "split:empty
string/0";
vector<string>
v;
S_T deli_len =
delimit.size();
long index = npos,
last_search_position = 0;
while(
(index=src.find(delimit,
last_search_position))!=npos )
{
if(index==last_search_position)
v.push_back(null_subst);
else
v.push_back( src.substr(last_search_position,
index-
last_search_position)
);
last_search_position = index + deli_len;
}
string last_one =
src.substr(last_search_position);
v.push_back(
last_one.empty()? null_subst:last_one );
return v;
}
测试如下:
int main(int argc, char* argv[])
{
string
src = ",ab,cde;,,fg,," ;
string tok = ",;"
;
vector<string> v1 =
tokenize(src, tok ,true);
vector<string>
v2 = tokenize(src, tok
,false,
"<null>");
cout<<"-------------v1:"<<endl;
for(int i=0; i<v1.size();i++)
{
cout<<v1[i].c_str()<<endl;
}
cout<<"-------------v2:"<<endl;
for(int j=0; j<v2.size();j++)
{
cout<<v2[j].c_str()<<endl;
}
try{
string s = "1;2;3;4";
string del =
";";//"###";
vector<string> v3 = split(s, del,
"<null>");
cout<<"-------------v3:"<<endl;
for(int k=0;
k<v3.size();k++)
{
cout<<v3[k].c_str()<<endl;
}
}
catch
(char *s) {
cout<<s<<endl;
}
return 0;
}