PEGTL VS C++ Format

Very cool, and I like the name!

I'd be interested in reading about how Matcheroni compares with PEGTL and Lexy.

https://github.com/taocpp/PEGTL

I found a library I wanted to test: Pegtl

I like PEGTL

Sure - libraries that are expected to be entirely self-contained. The one that comes to mind is PEGTL, a parser combinator library that is intended to be embedded inside a larger program. Making it import more dependencies would break this philosophy. Similarly, in the Rust world, there are a variety of "no-std" crates that should be able to be imported even if the standard library is not available on the target platform.

The program in the post was just an example meant to illustrate the problem. Originally, this (new) behavior of MSVC broke my code in the PEGTL, see [this commit](https://github.com/taocpp/PEGTL/commit/e3c8cb499dc3d1d76d23f2d5d79469dcb15550c5) that I needed to apply to fix it.

As a professional C++ programmer I feel a lot of the reasons C++ gets this response is because it's simply not "batteries included" like Go or Rust.

C++ is a very powerful, unopinionated language, that gives you a lot of freedom to attack your problem domain the way you best see fit.

If you're writing a networked application, don't use POSIX sockets, go and find a higher level library. If you're parsing complex text formats, don't iterate over buffers with char*'s, go pick up PEGTL[0]. If you're working on graphs, or need to properly index in-memory data, go pick up Boost[1][2]. If you need a GUI, go pick up Qt.

It's extremely common in C++, due to the lack of a universal package management solution, for people to try and "muddle through" and do shit themselves when it's far outside their core competency.

At one of my last employers, the core product was parsing JSON with std::regex, simply because they couldn't be bothered to integrate a JSON library.

[0] https://github.com/taocpp/PEGTL

[1] https://www.boost.org/doc/libs/1_76_0/libs/graph/

[2] https://www.boost.org/doc/libs/1_76_0/libs/multi_index/doc/i...

You are looking for Boost.Spirit (https://www.boost.org/doc/libs/1_76_0/libs/spirit/doc/x3/html/index.html) or PEGTL (https://github.com/taocpp/PEGTL)

I personally prefer to use parsing combinator libraries in C++, where the "grammar" is just part of normal C++ and directly integrate. Examples are Boost.Spirit, pegtl, or (my own) lexy.

While I'm not quite sure how this might transfer to your approach, with your Haskell-inspired style being quite different from our C++ templates, in the PEGTL our equivalent to your Char, which is called one, is variadic (true to the T in PEGTL a variadic template) and takes a list of possible matches.

Continuation passing monads form the basis of a perfectly valid and usable software architecture and programming pattern.

In the case of ostream and operator<<, this pattern reduces the number of intermediate objects that would otherwise be constructed.

If you object to iostream on religious or stylistic grounds, there's always fmt which is more like Go or Python string interpolation.[0]

0. https://fmt.dev

GitHub repo: fmtlib/fmt

cmake_minimum_required(VERSION 3.5) project(knn_cpp CXX) # Set up C++ version and properties include(CheckIncludeFileCXX) check_include_file_cxx(any HAS_ANY) check_include_file_cxx(string_view HAS_STRING_VIEW) check_include_file_cxx(coroutine HAS_COROUTINE) set(CMAKE_CXX_STANDARD 20) set(CMAKE_BUILD_TYPE Debug) set(CMAKE_CXX_STANDARD_REQUIRED ON) set(CMAKE_CXX_EXTENSIONS OFF) # Copy data file to build directory file(COPY ${CMAKE_CURRENT_SOURCE_DIR}/iris.data DESTINATION ${CMAKE_CURRENT_BINARY_DIR}) # Download library usinng FetchContent include(FetchContent) FetchContent_Declare(matplotplusplus GIT_REPOSITORY https://github.com/alandefreitas/matplotplusplus GIT_TAG origin/master) FetchContent_GetProperties(matplotplusplus) if(NOT matplotplusplus_POPULATED) FetchContent_Populate(matplotplusplus) add_subdirectory(${matplotplusplus_SOURCE_DIR} ${matplotplusplus_BINARY_DIR} EXCLUDE_FROM_ALL) endif() FetchContent_Declare( fmt GIT_REPOSITORY https://github.com/fmtlib/fmt.git GIT_TAG 7.1.3 # Adjust the version as needed ) FetchContent_MakeAvailable(fmt) # Add executable and link project libraries and folders add_executable(${PROJECT_NAME} main.cc) target_link_libraries(${PROJECT_NAME} PUBLIC matplot fmt::fmt) aux_source_directory(lib LIB_SRC) target_include_directories(${PROJECT_NAME} PRIVATE ${CMAKE_CURRENT_SOURCE_DIR}) target_sources(${PROJECT_NAME} PRIVATE ${LIB_SRC}) add_subdirectory(tests)

Good catch, thanks! Fixed now. This explains why the difference was kinda low compared to another benchmark: https://github.com/fmtlib/fmt?tab=readme-ov-file#compile-tim....

> This is from C++23, right?

std::println is, yes.

> I wonder how available this is within compilers

https://en.cppreference.com/w/cpp/compiler_support says clang, gcc, and msvc all support it, though I don't know how recent those versions are off the top of my head.

In my understanding, with this specific feature, if you want a polyfill for older compilers, or to use some more cutting-edge features that haven't been standardized yet, https://github.com/fmtlib/fmt is available to you.

{fmt} has internal buffering but it's not yet exposed to users. There is a feature request for it: https://github.com/fmtlib/fmt/issues/2354. FILE buffering is not too bad but it can be easily optimized: https://www.zverovich.net/2020/08/04/optimal-file-buffer-siz....

Automatic use of operator<< when that exists was present in fmt until version 9.0.0. In version 9 you could use FMT_DEPRECATED_OSTREAM to opt in the old behaviour, but this too was removed in version 10.0.0. Now there is no way to automatically use operator<<.

You can install it https://github.com/fmtlib/fmt

Additionally, if you like low level stuff, check out libfmt (https://github.com/fmtlib/fmt) - not a big project, not difficult to understand. Or something like simdjson (https://github.com/simdjson/simdjson).