freqs.cpp

#include <string>
#include <iostream>
#include <fstream>
#include <map>
#include <locale>
#include <vector>
#include <algorithm>

void countWords(std::ifstream &in, std::map<std::string, int> &result) {
	std::string word;
	std::locale loc("C");
	while (!in.eof()) {
		auto c = std::tolower(in.get(), loc);
		if (std::isalpha(c, loc)) {
			word.append(1, c);
		}
		else if (word.length() > 0) {
			result[word]++;
			word = "";
		}
	}
}

bool comparePairs(std::pair<std::string, int> &a, std::pair<std::string, int> &b) {
	return a.second == b.second ? a.first < b.first : a.second > b.second;
}

void sortWords(std::map<std::string, int> &wordMap, std::vector<std::pair<std::string, int>> &pairList) {
	for (auto pair = wordMap.begin(); pair != wordMap.end(); ++pair) {
		pairList.push_back(*pair);
	}
	std::sort(pairList.begin(), pairList.end(), comparePairs);
}

int main(int argc, char* argv[]) {
	if (argc != 3) {
		std::cout << "usage: freqs [input] [output]";
	}
	else {
		auto inFilename = argv[1];
		auto outFilename = argv[2];

		std::ifstream in;
		in.open(inFilename);
		if (!in.is_open()) {
			std::cout << "cannot open file '" << inFilename << "'";
		}
		else {
			std::map<std::string, int> wordMap;
			countWords(in, wordMap);
			std::vector<std::pair<std::string, int>> pairList;
			sortWords(wordMap, pairList);

			std::ofstream out;
			out.open(outFilename);
			if (!out.is_open()) {
				std::cout << "cannot open file '" << outFilename << "'";
			}
			else {
				for (auto pair = pairList.begin(); pair != pairList.end(); ++pair) {
					out << pair->second << ' ' << pair->first << std::endl;
				}
			}
		}
	}

    return 0;
}