如何有效地识别字符串命令? [英] How to efficiently identify string commands?

问题描述

给出一系列必须针对每个命令运行的命令和非常独特的代码:

Given a series of commands and very unique code that must be run for each:

if(cmd == "cmd.setBoosterRocket")
    ...
else if(cmd == "cmd.windSales")
    ...
else if(cmd == "cmd.selfDustruct")
    ...
else if(cmd == "cmd.unleashHounds")
    ...

如何优化?放在switch语句中，是吗?

How might this be optimized? Be put into a switch statement, that is?

我考虑过制作哈希向量:

I considered making a vector of hashes:

std::hash<std::string> hasher;
for(std::string command : m_commandList)
    m_mashes.push_back(hasher(command)

但是向量不能作为switch case语句的一部分进行访问，因为它不是constexpr.字符串命令列表在编译时就已知道，我可能会对哈希值进行硬编码...但这似乎不是一个好主意.

But a vector cannot be accessed as part of a switch case statement as it is not a constexpr. The list of string commands is known at compile time, and I could potentially hardcode the hash values...but that does not seem like a great idea.

推荐答案

一种可能的方法是标记化:创建 enum 类型和字典.这样一来，您可以利用切换功能(与硬编码的散列相比，对程序员和编译器更友好)，并且具有对数复杂性.

One possible approach is tokenization: make an enum type and a dictionary. This way you take advantage of the switch (in a more programmer and compiler friendly way than hard-coded hashes) and have just logarithmic complexity.

enum Command {SET_BOOSTER_ROCKET, WINDSALES, ETC};
const std::map<std::string, Command> commands = {
  {"cmd.setBoosterRocket", SET_BOOSTER_ROCKET},
  {"cmd.windsales", WINDSALES},
  {"othercommands", ETC},
};

然后

auto cmd_it = commands.find(cmd);
if(cmd_it == commands.end()) { // ERROR }
switch(cmd_it->second){
  case SET_BOOSTER_ROCKET:
    // your code
  break;
  case WINDSALES:
    // your code
  break;
  // etc
}

像这样的标记化如果您需要大量的操作，您的命令可能会有些乏味，但是在可伸缩性和可读性之间要有一个很好的平衡.

Tokenizing like this your commands might be a little tedious if you have a lot to start, but then it has a good balance between scalability and readability.

这篇关于如何有效地识别字符串命令?的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！