C++におけるstringのsplit

c++にはstring内にsplitメソッドが定義されていない。そのため自分で実装しなければならない。

実装

stringstreamを用いる

split.cpp

#include <vector>
#include <string>
#include <sstream>

using namespace std;

vector<string> split(const string &s, char delim) {
    vector<string> elems;
    stringstream ss(s);
    string item;
    while (getline(ss, item, delim)) {
    if (!item.empty()) {
            elems.push_back(item);
        }
    }
    return elems;
}

愚直に実装

split_naive.cpp

#include <vector>
#include <string>

using namespace std;

vector<string> split_naive(const string &s, char delim) {
    vector<string> elems;
    string item;
    for (char ch: s) {
        if (ch == delim) {
            if (!item.empty())
                elems.push_back(item);
            item.clear();
        }
        else {
            item += ch;
        }
    }
    if (!item.empty())
        elems.push_back(item);
    return elems;
}

ベンチマーク

bench1: 入力文字は短く、関数の実行回数が多い

bench1.cpp

void bench1() {
    string input = "This is  the part when I say I don't want ya";
    char delim = ' ';

    cout << "split: " << measeure(split, input, delim, 100000) / (CLOCKS_PER_SEC / 1000) << "ms" << endl;
    cout << "naive: " << measeure(split_naive, input, delim, 100000) / (CLOCKS_PER_SEC / 1000) << "ms" << endl;
}

bench2: 入力文字は長く、関数の実行回数は少ない。

bench2.cpp

void bench2() {
    string random_input = "";
    int len = 1000000;
    mt19937 engine;
    uniform_int_distribution<int> distribution(0, 25);
    char delim = distribution(engine) + 'a';
    for (int p=0; p<len; ++p) {
        int ch = distribution(engine);
    random_input += ch + 'a';
    }

    cout << "split: " << measeure(split, random_input, delim, 10) / (CLOCKS_PER_SEC / 1000) << "ms" << endl;
    cout << "naive: " << measeure(split_naive, random_input, delim, 10) / (CLOCKS_PER_SEC / 1000) << "ms" << endl;
}

結果:

	bench1[ms]	bench2[ms]
split	604	477
naive	384	188

Reference

[1]http://stackoverflow.com/questions/236129/how-to-split-a-string-in-c