[Kata 8 C++]Correct the mistakes of the character recognition software

问题描述

Character recognition software is widely used to digitise printed texts. Thus the texts can be edited, searched and stored on a computer.
When documents (especially pretty old ones written with a typewriter), are digitised character recognition softwares often make mistakes.
Your task is correct the errors in the digitised text. You only have to handle the following mistakes:

• S is misinterpreted as 5
• O is misinterpreted as 0
• I is misinterpreted as 1
  The test cases contain numbers only by mistake.

测试用例

#include <random>
#include <iostream>

auto randint(int min, int max) {
  static std::random_device rd;
  static std::mt19937 rng(rd());
  std::uniform_int_distribution<int> uni(min, max);
  return uni(rng);
}

std::string sol(std::string s){
  std::transform(s.begin(), s.end(), s.begin(), [](char c){return c == '5' ? 'S' : c == '0' ? 'O' : c == '1' ? 'I' : c;});
  return s;
}

Describe(Correct){
  It(BasicTests){
    Assert::That(correct("1F-RUDYARD K1PL1NG"), Equals("IF-RUDYARD KIPLING"));
    Assert::That(correct("R0BERT MERLE - THE DAY 0F THE D0LPH1N"), Equals("ROBERT MERLE - THE DAY OF THE DOLPHIN"));
    Assert::That(correct("R1CHARD P. FEYNMAN - THE FEYNMAN LECTURE5 0N PHY51C5"), Equals("RICHARD P. FEYNMAN - THE FEYNMAN LECTURES ON PHYSICS"));
    Assert::That(correct("R1CHARD P. FEYNMAN - 5TAT15T1CAL MECHAN1C5"), Equals("RICHARD P. FEYNMAN - STATISTICAL MECHANICS"));
    Assert::That(correct("5TEPHEN HAWK1NG - A BR1EF H15T0RY 0F T1ME"), Equals("STEPHEN HAWKING - A BRIEF HISTORY OF TIME"));
    Assert::That(correct("5TEPHEN HAWK1NG - THE UN1VER5E 1N A NUT5HELL"), Equals("STEPHEN HAWKING - THE UNIVERSE IN A NUTSHELL"));
    Assert::That(correct("ERNE5T HEM1NGWAY - A FARWELL T0 ARM5"), Equals("ERNEST HEMINGWAY - A FARWELL TO ARMS"));
    Assert::That(correct("ERNE5T HEM1NGWAY - F0R WH0M THE BELL T0LL5"), Equals("ERNEST HEMINGWAY - FOR WHOM THE BELL TOLLS"));
    Assert::That(correct("ERNE5T HEM1NGWAY - THE 0LD MAN AND THE 5EA"), Equals("ERNEST HEMINGWAY - THE OLD MAN AND THE SEA"));
    Assert::That(correct("J. R. R. T0LK1EN - THE L0RD 0F THE R1NG5"), Equals("J. R. R. TOLKIEN - THE LORD OF THE RINGS"));
  }
  It(RandomTests){
    std::string base="abcdefghijklmnopqrstuvwxyz ABCDEFGHIJKLMNOPQRSTUVWXYZ - ";
    for (int i = 0; i < 40; i++) {
      std::string s;
      int sLen = randint(1, 35);
      for (int j = 0; j < sLen; j++) s += randint(0, 9) == 0 ? "501"[randint(0, 2)] : base[randint(0, base.size() - 1)];
      std::cout << "Testing for '" << s << "'\n";
      Assert::That(correct(s), Equals(sol(s)));
    }
  }
};

解答

高赞replace函数

#include <string>

std::string correct(std::string str){
  replace(str.begin(), str.end(), '5', 'S');
  replace(str.begin(), str.end(), '0', 'O');
  replace(str.begin(), str.end(), '1', 'I');
  return str;
}

replace函数包含于头文件#include中。泛型算法replace把队列中与给定值相等的所有值替换为另一个值，整个队列都被扫描，即此算法的各个版本都在线性时间内执行———其复杂度为O(n)。
但是上面的replace 用法是在 algorithm中

template <class _FwdIt, class _Ty>
_CONSTEXPR20 void replace(const _FwdIt _First, const _FwdIt _Last, const _Ty& _Oldval, const _Ty& _Newval) {
    // replace each matching _Oldval with _Newval
    _Adl_verify_range(_First, _Last);
    auto _UFirst      = _Get_unwrapped(_First);
    const auto _ULast = _Get_unwrapped(_Last);
    for (; _UFirst != _ULast; ++_UFirst) {
        if (*_UFirst == _Oldval) {
            *_UFirst = _Newval;
        }
    }
}

这个方案比较易读，但有3次迭代，会损失一些效率。

次高赞

#include <string>
#include <algorithm>

std::string correct(std::string str){
for(size_t i=0; i<str.size(); i++)
{
  switch (str[i])
  {
    case '1': str[i]='I'; break; 
    case '0': str[i]='O'; break; 
    case '5': str[i]='S'; break; 
  }
}
return str;
}

简单的替换，使用switch 替换了if else，提高了易读性。

map

#include <string>

std::string correct(std::string str)
{
  std::map<char, char> corrections{ {'5', 'S'}, {'0', 'O'}, {'1', 'I'} };
  for(auto& c : str)
  {
    for(auto const cor : corrections)
      if(c == cor.first)
      {
        c = cor.second;
        break;
      }
  }
  return str;
}

使用了map 来存放条件， auto& c 来遍历

transform

#include <string>
#include <algorithm>

std::string correct(std::string str){
  std::transform(str.begin(), str.end(), str.begin(), [](auto c){ 
    if(c == '5') return 'S';
    if(c == '0') return 'O';
    if(c == '1') return 'I';
    return c;
    });
  return str;
}

transform函数的作用是：将某操作应用于指定范围的每个元素。transform函数有两个重载版本：
transform(first,last,result,op);//first是容器的首迭代器，last为容器的末迭代器，result为存放结果的容器，op为要进行操作的一元函数对象或sturct、class。

transform(first1,last1,first2,result,binary_op);//first1是第一个容器的首迭代 器，last1为第一个容器的末迭代器，first2为第二个容器的首迭代器，result为存放结果的容器，binary_op为要进行操作的二元函数 对象或sturct、class。

注意：第二个重载版本必须要保证两个容器的元素个数相等才行，否则会抛出异常。

for_each版本

#include <string>
#include <algorithm>

std::string correct(std::string str){
  std::for_each( str.begin(), str.end(), [](char& c){
    switch( c ) 
    {
      case '5': c = 'S'; break;
      case '0': c = 'O'; break;
      case '1': c = 'I'; break;
      default: break;
    }
  });
  
  return str;
}

lambda表达式中传入了引用，原地修改字符串。

其它

std::string correct(const std::string& str){
    std::string r { str };
    for (int i { 0 }; i < r.size(); i++){
        if (r[i] == '5') { r[i] = 'S'; }
        else if ( r[i] == '0') { r[i] = 'O'; }
        else if ( r[i] == '1') { r[i] = 'I'; }
    }
    return r;
}

这里有一个特殊的写法：int i { 0 }; 貌似是C++ 11 的初始化方法 如 vector<int> p={1,2,3};