#include <scn/scn.h>
#include <cstdio>

int main() {
    int i;
    // Read an integer from stdin
    // with an accompanying message
    scn::prompt("What's your favorite number? ", "{}", i);
    printf("Oh, cool, %d!", i);
}

// Example result:
// What's your favorite number? 42
// Oh, cool, 42!

scnlib is a modern C++ library for replacing scanf and std::istream. This library attempts to move us ever so closer to replacing iostreams and C stdio altogether. It's faster than iostream (see Benchmarks) and type-safe, unlike scanf. Think {fmt} but in the other direction.

This library is the reference implementation of the ISO C++ standards proposal P1729 "Text Parsing".

This library is currently of pre-release quality (version 0.3). It is usable in its present state, but do not expect a bug-free experience. Additionally, I reserve the right to change the interfaces in incompatible ways in future minor version releases (e.g. 0.4).

The documentation can be found online, from https://scnlib.dev. The documentation includes a tutorial.

To build the docs yourself, build the doc target generated by CMake. Building the documentation requires Doxygen.

Rather lacking examples can be found from the examples/ folder.

Should you get stuck somewhere, you can find me on the CppLang Slack, or by DMing me on Twitter.

#include <scn/scn.h>
#include <iostream>
#include <string_view>

using namespace std::string_view_literals; // For sv

int main() {
    auto str = "Hello world!"sv;

    std::string word;
    scn::scan(str, "{}", word);

    std::cout << word << '\n'; // Will output "Hello"
    std::cout << str << '\n';  // Will output " world!"
}

#include <scn/scn.h>

int main() {
    int i, j;
    auto ret = scn::scan("123 456 foo", "{} {}", i, j);
    // ret == true
    // i == 123
    // j == 456

    std::string str;
    ret = scn::scan(ret.range(), "{}", str);
    // ret == true
    // str == "foo"
}

#include <scn/scn.h>
#include <scn/tuple_return.h>

int main() {
    auto [r, i] = scn::scan_tuple<int>("42", "{}");
    // r is a result object, contextually convertible to `bool`
    // i == 42
}

#include <scn/scn.h>
#include <string_view>
#include <iostream>

using namespace std::literals;

int main() {
    auto str = "foo"sv;
    int i;
    // "foo" is not a valid integer
    auto result = scn::scan(str, "{}", i);
    if (!result) {
        // i is not touched (still unconstructed)
        // str == "foo" (range not advanced)
        std::cout << "Integer parsing failed with message: " << result.error().msg() << '\n';
    }
}

• Blazing-fast parsing of values (see benchmarks)
• Modern C++ interface, featuring type safety (variadic templates), convenience (ranges) and customizability
  • No << chevron >> hell
  • Requires C++11 or newer
• "{python}"-like (recommended) or "%scanf"-like format string syntax
• Optionally header only
• Minimal code size increase
• No exceptions (supports building with -fno-exceptions -fno-rtti with minimal loss of functionality)
  • Localization requires exceptions, because of the way std::locale is

The easiest way of integrating scnlib to your project would likely be to use a git submodule.

# Replace `external/scnlib` with the directory where you'd like the submodule be located
$ git submodule add https://github.com/eliaskosunen/scnlib external/scnlib

Then just do the following in your CMakeLists.txt

# Include the submodule directory
add_subdirectory(external/scnlib EXCLUDE_FROM_ALL)

# Target which you'd like to use scnlib
add_executable(my_program ...)
# Use scn::scn-header-only if you'd prefer
target_link_libraries(my_program scn::scn)

You can also build scnlib separately.

# Whereever you cloned scnlib to
$ mkdir build
$ cd build
$ cmake ..
$ make -j
$ make install

Then, in your project:

# Find scnlib package
find_package(scn CONFIG REQUIRED)

# Target which you'd like to use scnlib
add_executable(my_program ...)
target_link_libraries(my_program scn::scn)

Every commit is tested with

• gcc 5.5 and newer
• clang 3.7 and newer
• Visual Studio 2017 and 2019

with very extreme warning flags (see cmake/flags.cmake) and with multiple build configurations for each compiler.

Older compilers may work, but it is not guaranteed. GCC 4.x support will not be provided, as its C++11 support is too buggy. Same thing applies for VS 2015, due to its inability to handle templates.

These benchmarks were run on a Ubuntu 19.04 machine running kernel version 5.0.0-32, with an Intel Core i5-6600K processor, and compiled with gcc version 8.3.0, with -O3 -march=native. The source code for the benchmarks can be seen in the benchmark directory.

Times are in nanoseconds of CPU time. Lower is better.

scn::scan is calling the scn::scan(..., scn::default_tag, ...) overload.

You can run the benchmarks yourself by enabling SCN_BUILD_BENCHMARKS and building the target bench. SCN_BUILD_BENCHMARKS is enabled by default if scn is the root CMake project, and disabled otherwise.

Performance comparison benchmarks with Boost.Spirit.x3 can be found here

Code size benchmarks test code bloat for nontrivial projects. It generates 25 translation units and reads values from stdin* five times to simulate a medium sized project. The resulting executable size is shown in the following tables.

The code was compiled on Ubuntu 19.04 with g++ 8.3.0. scnlib is linked dynamically to level out the playing field compared to already dynamically linked libc and libstdc++. See the directory benchmark/bloat for more information, e.g. templates for each TU.

To run these tests yourself:

$ cd build
$ cmake -DCMAKE_BUILD_TYPE=Debug -DBUILD_SHARED_LIBS=ON -DSCN_INSTALL=OFF -DSCN_BLOAT=ON ..
$ make -j
$ ./benchmark/bloat/run-bloat-tests.py ./benchmark/bloat

The contents of this library are heavily influenced by fmtlib and its derivative works.

https://github.com/fmtlib/fmt
https://fmt.dev

fmtlib is licensed under the BSD 2-clause license.
Copyright (c) 2012-2019 Victor Zverovich

The ranges implementation found from this library is based on NanoRange:

https://github.com/tcbrindle/NanoRange

NanoRange is licensed under the Boost Software License, version 1.0.
Copyright (c) 2018 Tristan Brindle

scnlib is licensed under the Apache License, version 2.0.
Copyright (c) Elias Kosunen 2017-2019
See LICENSE for further details