Last active
January 16, 2024 15:29
-
-
Save ribomation/e24c62899cd3120a890a9133be54161c to your computer and use it in GitHub Desktop.
Simple implementation of 1BRC in Moden C++
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#include <iostream> | |
#include <fstream> | |
#include <string> | |
#include <unordered_map> | |
#include <map> | |
#include <algorithm> | |
#include <format> | |
#include <limits> | |
#include "util.hxx" | |
using std::cout; | |
using std::string; | |
using namespace std::string_literals; | |
namespace rm = ribomation::util; | |
struct Aggregation { | |
unsigned count = 0; | |
double sum = 0; | |
double min = std::numeric_limits<double>::max(); | |
double max = std::numeric_limits<double>::min(); | |
void operator+=(double t) { | |
++count; | |
sum += t; | |
min = std::min(min, t); | |
max = std::max(max, t); | |
} | |
friend auto operator<<(std::ostream& os, Aggregation const& a) -> std::ostream& { | |
return os << std::format("{:+.2f}C, {:+.1f}/{:+.1f} ({:Ld})", | |
(a.sum / a.count), a.min, a.max, a.count); | |
} | |
}; | |
int main(int argc, char** argv) { | |
auto filename = rm::getFilename(argc, argv); | |
cout << "filename: " << filename << "\n----\n"; | |
rm::elapsed([filename]() { | |
auto f = std::ifstream{filename}; | |
if (!f) throw std::invalid_argument{"cannot open "s + filename}; | |
auto data = std::unordered_map<string, Aggregation>{}; | |
data.reserve(500); | |
auto count = 0U; | |
for (string line; std::getline(f, line);) { | |
++count; | |
auto semi = line.find(';'); | |
auto name = line.substr(0, semi); | |
auto temp = std::stod(line.substr(semi + 1)); | |
data[name] += temp; | |
} | |
// auto sorted = std::map<string, Aggregation>{data.begin(), data.end()}; | |
auto sorted = std::vector<std::pair<string, Aggregation>>{data.begin(), data.end()}; | |
std::sort(sorted.begin(), sorted.end(), [](auto&& lhs, auto&& rhs){ | |
return lhs.first < rhs.first; | |
}); | |
for (auto&& [name, aggr]: sorted) { | |
cout << name << ": " << aggr << "\n"; | |
} | |
cout << "------\n# records: " << count << "\n"; | |
}); | |
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#include <iostream> | |
#include <string> | |
#include <chrono> | |
#include "util.hxx" | |
namespace ribomation::util { | |
using namespace std::string_literals; | |
namespace cr = std::chrono; | |
using std::cout; | |
using std::string; | |
void elapsed(std::function<void()> const& stmts) { | |
auto startTime = cr::high_resolution_clock::now(); | |
stmts(); | |
auto endTime = cr::high_resolution_clock::now(); | |
auto elapsed = cr::duration<double, std::ratio<1, 1>>{endTime - startTime}; | |
cout << std::format("Elapsed time: {:.3f} seconds\n", elapsed.count()); | |
} | |
auto getFilename(int argc, char** argv) -> string { | |
auto filename = "data/weather-data-100K.csv"s; | |
for (auto k = 1; k < argc; ++k) { | |
auto arg = string{argv[k]}; | |
if (arg == "-f"s) { | |
filename = argv[++k]; | |
} else { | |
std::cerr << "usage: " << argv[0] << " [-f <str>]\n"; | |
throw std::invalid_argument{"usage"}; | |
} | |
} | |
return filename; | |
} | |
} |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#pragma once | |
#include <functional> | |
#include <string> | |
namespace ribomation::util { | |
void elapsed(const std::function<void()>& stmts); | |
auto getFilename(int argc, char** argv) -> std::string; | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment