如何在文档中查找某些关键字并将其读入程序（C++）

Question

I've been given a document, from which I have to read certain Information into my C++ program that are preceded by appropriate Keywords looking like this:

...
*useless information*
...
KEYWORD_A : VALUE_A
...
*useless information*
...
KEYWORD_B : VALUE_B
...

What would be the most efficient/correct way for filtering these Keywords, that do not have to appear in a certain order? I currently use following scheme to do so

    std::string content_of_line = "";
    while(content_of_line.find("KEYWORD_A") == std::string::npos){
        std::getline(file,content_of_line);
    }
    std::stringstream stream(content_of_line);
    stream >> variable_a;

    std::getline(file,content_of_line);
    while(content_of_line.find("KEYWORD_B") == std::string::npos){
        std::getline(file,content_of_line);
    }
    stream.clear();
    stream << content_of_line;
    stream >> variable_b;

// and so on....

This, however, produces a lot of redundant code and also relies on a specific order of the keywords (which doesnt necessarily have to exist)... I hope you can show me a much nicer way of solving my problem. Thank you very much in advance for trying to help me on my problem!

Answer 1

Ok, I'll take a shot

#include <fstream>
#include <iostream>
#include <string>
#include <unordered_map>

int main()
{
    std::ifstream inp("inp.txt");
    std::string word;
    using valueType = std::string;
    //Assuming keywords only appear once in the file
    std::unordered_map<std::string, valueType> keywords {{"KEYWORD_A", ""},{"KEYWORD_B", ""},};
    while(inp>>std::ws>>word) {
        if(keywords.find(word) != keywords.cend()) {
            //Assuming schema of lines with keywords is KEY : VALUE
            inp.ignore(std::numeric_limits<std::streamsize>::max(), ':');
            inp>>std::ws>>keywords[word];
        }
    }
    for(const auto& pair: keywords) {
        std::cout<<pair.first<<' '<<pair.second<<'\n';
    }
    return 0;
}