fast_float

What's Changed

• Trimming out one eight-digit optimization. by @lemire in https://github.com/fastfloat/fast_float/pull/152

Full Changelog: https://github.com/fastfloat/fast_float/compare/v3.6.0...v3.7.0

What's Changed

• Simplified Clinger's fast path for increased portability in https://github.com/fastfloat/fast_float/pull/150 credit to @jakubjelinek
• Fix some typos by @striezel in https://github.com/fastfloat/fast_float/pull/140
• 🔨 add library ALIAS FastFloat::fast_float by @OlivierLDff in https://github.com/fastfloat/fast_float/pull/145
• update actions/checkout in GitHub Actions to v3 by @striezel in https://github.com/fastfloat/fast_float/pull/141
• Don't use __umulh() with MinGW on ARM64 by @kou in https://github.com/fastfloat/fast_float/pull/147
• 🔨 Don't write to global CMAKE_CXX_STANDARD by @OlivierLDff in https://github.com/fastfloat/fast_float/pull/144
• 🔨 Introduce FASTFLOAT_INSTALL to make install optional by @OlivierLDff in https://github.com/fastfloat/fast_float/pull/143

New Contributors

• @striezel made their first contribution in https://github.com/fastfloat/fast_float/pull/140
• @OlivierLDff made their first contribution in https://github.com/fastfloat/fast_float/pull/145
• @kou made their first contribution in https://github.com/fastfloat/fast_float/pull/147

Full Changelog: https://github.com/fastfloat/fast_float/compare/v3.5.1...v3.6.0

Fixing typo with respect to emscripten.

Full Changelog: https://github.com/fastfloat/fast_float/compare/v3.5.0...v3.5.1

What's Changed

• Better support for emscripten by @lemire
• Make endianness detection more portable by @jwakely in https://github.com/fastfloat/fast_float/pull/123
• Fix deduction failure for std::min call by @jwakely in https://github.com/fastfloat/fast_float/pull/122
• Fixed installation on other than Ubuntu GNU/Linux distributions by @xvitaly in https://github.com/fastfloat/fast_float/pull/127
• Export CMake targets as architecture independent by @xvitaly in https://github.com/fastfloat/fast_float/pull/129
• python: set file encoding by @fargies in https://github.com/fastfloat/fast_float/pull/126

New Contributors

• @jwakely made their first contribution in https://github.com/fastfloat/fast_float/pull/123
• @xvitaly made their first contribution in https://github.com/fastfloat/fast_float/pull/127
• @fargies made their first contribution in https://github.com/fastfloat/fast_float/pull/126

Full Changelog: https://github.com/fastfloat/fast_float/compare/v3.4.0...v3.4.1

Patch release.

Fix https://github.com/fastfloat/fast_float/issues/117

Removing broken and unsupported C++20 code.

Patch release for version 3.0.

Fixes https://github.com/fastfloat/fast_float/issues/106

Fixes https://github.com/fastfloat/fast_float/issues/105

Special thanks to @mumbleskates for reporting these issues !

This adds a much faster parser for long input strings (e.g., 132.312312321321321321321312321321e131) credit: @Alexhuszagh

A new function was added which allows you to specify that the decimal separator is the comma. The from_chars function is unchanged.

credit @pitrou

Fixed build issue when compiling for 32-bit ARM platforms under Visual Studio.

Fixed "permissive" issue under Visual Studio whereas some users would get link errors with some flags due to static constexpr not being considered inline by default under Visual Studio.

In this version, as in the C++17 standard, we forbid leading spaces as well as the leading '+' character.

NaN/infinity parsing has been revised. In particular, parse "nan(n-char-seq-opt)" as required by C++17 20.19.3.7, credit @eugenegff.

Update to CMake to allow package installation @georg-emg

Fixes a potential issue raised by Ivan Smirnov and improved minutely code clarity.

Minor fix in the fall back code (issue 47).

Better support for clang under msys2.

• Better support for 32-bit systems (@biojppm)
• Removed iostream dependency (@biojppm)
• The algorithmic fast paths are more common
• Adopted slightly faster SWAR code (@aqrit)