C++でsplit (3) – utf16版

char16_t* next_c_u16(const char16_t* c) {
  if (IS_SURROGATE_PAIR(c[0], c[1]) == true)
    return (char16_t*)c + 2;
  return (char16_t*)c + 1;
}

//utf16文字を一文字取り出す
void ext_c_u16(char16_t* const dst, const char16_t* src,int* size) {
  int sz;
  if (IS_SURROGATE_PAIR(src[0], src[1]) == true)
    sz = 4;
  else
    sz = 2;

  memcpy(dst, src, sz);
  dst[sz/2] = char16_t(0);

  if(size)
    *size = sz;
}

//utf16文字を比較する
bool cmp_c_u16(const char16_t* c1, const char16_t* c2) {
  int K = bytes_c_u16(c1);
  int L = bytes_c_u16(c2);

  if (K != L)
    return false;

  bool issame = (memcmp(c1, c2, K) == 0);
  return issame;
}

//文字のバイト数を取得
int bytes_c_u16(const char16_t* c) {
  if (IS_SURROGATE_PAIR(c[0], c[1]) == true)
    return 4;
  return 2;
}

std::vector<std::u16string> split_u16(const char16_t* src, const char16_t* del) {

  char16_t tmp[10];

  std::vector<std::u16string> result;

  std::u16string tmps;
  while (*src) {

    //デリミタを飛ばす
    const char16_t* p = src;
    while (cmp_c_u16(src, del) == true && *src != '\0')
      src = next_c_u16(src);

    //デリミタに遭遇するまで文字を追加し続ける
    while (cmp_c_u16(src, del) != true && *src != '\0') {
      ext_c_u16(tmp, src, nullptr);//一文字取り出す
      tmps += tmp;
      src = next_c_u16(src);
    }
    if (tmps.size()) {
      result.push_back(tmps);
    }
    tmps.clear();
  }

  return result;
}

int main()
{

  setlocale(LC_ALL, "");

  char16_t* u16 = (char16_t*)L"C:\\いろは\\山田\\abc\\";

  std::ofstream f(L"c:\\test\\test.txt", std::ios::out | std::ios::binary);

  std::vector<std::u16string> ret = split_u16(u16, (char16_t*)L"\\");

  for (size_t i = 0; i < ret.size(); i++) {
    wprintf(L"[%s]", (wchar_t*)ret[i].c_str());
  }

  f.close();

  return 0;
}