Development

Public notes on the past, present and future development of the PEGTL.

C++ Standard

Version 0.x of the PEGTL requires at least C++11.

Version 1.x of the PEGTL requires at least C++11.

Version 2.x of the PEGTL requires at least C++11.

Version 3.x of the PEGTL requires at least C++17.

Version 4.x of the PEGTL requires at least C++17.

Version 5.x of the PEGTL will jump to C++20 or C++23.

C++ Features

• Add support for C++20 char8_t where appropriate.
• Use C++20 std::span in inputs and everywhere else it makes sense.
• Investigate whether there is anything useful we can do with Ranges in the PEGTL.
• Use C++20 Concepts instead of all the SFINAE and meta-programming where possible.
• Give examples for C++20 "lambdas in unevaluated contexts" in conjunction with tao::pegtl::function.
• Keep an open eye for opportunities to use C++20 spaceship operator. Spaceship!
• Keep an open eye for opportunities to use C++20 defaulted comparison operators.
• Keep an open eye for opportunities to use C++20 [[likely]] and [[unlikely]].
• Keep an open eye for opportunities to use C++20 constinit and consteval, and
• keep an open eye for opportunities to use the extended constexpr facilities.
• For example make the bulk() and size() functions in the peek classes consteval!
• Keep an open eye for opportunities to use the extended CTAD facilities from C++20.
• Keep an open eye for opportunities to use class types as non-type template parameters.
• Replace the hand-crafted endian facilities with C++20 std::endian and C++23 std::byteswap.
• Investigate how C++20 and C++23 compile-time facilities can help with compile-time strings.
• Investigate whether we can use C++20 std::bit_cast to improve some of the low-level code.
• Use C++23 "deducing this" feature to let base class make_rewind_guard() return a rewind guard for a derived class.
• When can we assume the C++17 charconv facilities are universally available?
• Auto-detect internal rules insead of internal::enable_control?

Other Things

• Investigate whether we are crazy enough to implement parsing non-contiguous memory like linked lists?
• Investigate whether we are crazy enough to implement parsing non-linear data structures like trees?

Stream Parsing

A couple of things that could be done in the area of stream parsing.

• Use the double-mapped ring-buffer to prevent discard() having to move data.
• Debug input and related facilities that detect when data in the input buffer is accessed after being discarded or moved (by a discard).
• Investigate the use of ("stackful") coroutines for parsing from a network socket, and
• investigate whether this can also be used for stream parsing that keeps everything.

Development Tools

We use the code coverage facilities of the GCC and Clang compilers to meet our goal of 100% code coverage with unit tests. Unfortunately the coverage reports are not 100% reliable.

We also try to use clang-format to ensure consistent formatting of our source code. Unfortunately it, too, can get confused and result in incorrectly formatted code.

For example under some circumstances clang-format does not understand whether a && is the logical operator or an r-value reference declarator, or whether a < is a comparison operator or introduces a template argument list.

---

This page is part of the PEGTL and its documentation.

Copyright (c) 2017-2026 Dr. Colin Hirsch and Daniel Frey
Distributed under the Boost Software License, Version 1.0
See accompanying file LICENSE_1_0.txt or copy at https://www.boost.org/LICENSE_1_0.txt