10.0.0

• Replaced Grisu with a new floating-point formatting algorithm for given precision (#3262, #2750, #3269, #3276). The new algorithm is based on Dragonbox already used for the shortest representation and gives substantial performance improvement:

  • Red: new algorithm

  • Green: new algorithm with FMT_USE_FULL_CACHE_DRAGONBOX defined to 1

  • Blue: old algorithm

  Thanks @jk-jeon (Junekey Jeon).

• Replaced snprintf-based hex float formatter with an internal implementation (#3179, #3203). This removes the last usage of s(n)printf in {fmt}. Thanks @phprus (Vladislav Shchapov).

• Fixed alignment of floating-point numbers with localization (#3263, #3272). Thanks @ShawnZhong (Shawn Zhong).

• Made handling of # consistent with std::format.

• Improved C++20 module support (#3134, #3254, #3386, #3387, #3388, #3392, #3397, #3399, #3400). Thanks @laitingsheng (Tinson Lai), @Orvid (Orvid King), @DanielaE (Daniela Engert). Switched to the modules CMake library which allows building {fmt} as a C++20 module with clang:

  CXX=clang++ cmake -DFMT_MODULE=ON .
  make
  

• Made format_as work with any user-defined type and not just enums. For example (godbolt):

  #include <fmt/format.h>
  
  struct floaty_mc_floatface {
    double value;
  };
  
  auto format_as(floaty_mc_floatface f) { return f.value; }
  
  int main() {
    fmt::print("{:8}\n", floaty_mc_floatface{0.42}); // prints "    0.42"
  }

• Removed deprecated implicit conversions for enums and conversions to primitive types for compatibility with std::format and to prevent potential ODR violations. Use format_as instead.

• Added support for fill, align and width to the time point formatter (#3237, #3260, #3275). For example (godbolt):

  #include <fmt/chrono.h>
  
  int main() {
    // prints "    2023"
    fmt::print("{:>8%Y}\n", std::chrono::system_clock::now());
  }

  Thanks @ShawnZhong (Shawn Zhong).

• Implemented formatting of subseconds (#2207, #3117, #3115, #3143, #3144, #3349). For example (godbolt):

  #include <fmt/chrono.h>
  
  int main() {
    // prints 01.234567
    fmt::print("{:%S}\n", std::chrono::microseconds(1234567));
  }

  Thanks @patrickroocks (Patrick Roocks) @phprus (Vladislav Shchapov), @BRevzin (Barry Revzin).

• Added precision support to %S (#3148). Thanks @SappyJoy (Stepan Ponomaryov)

• Added support for std::utc_time (#3098, #3110). Thanks @patrickroocks (Patrick Roocks).

• Switched formatting of std::chrono::system_clock from local time to UTC for compatibility with the standard (#3199, #3230). Thanks @ned14 (Niall Douglas).

• Added support for %Ez and %Oz to chrono formatters. (#3220, #3222). Thanks @phprus (Vladislav Shchapov).

• Improved validation of format specifiers for std::chrono::duration (#3219, #3232). Thanks @ShawnZhong (Shawn Zhong).

• Fixed formatting of time points before the epoch (#3117, #3261). For example (godbolt):

  #include <fmt/chrono.h>
  
  int main() {
    auto t = std::chrono::system_clock::from_time_t(0) -
             std::chrono::milliseconds(250);
    fmt::print("{:%S}\n", t); // prints 59.750000000
  }

  Thanks @ShawnZhong (Shawn Zhong).

• Experimental: implemented glibc extension for padding seconds, minutes and hours (#2959, #3271). Thanks @ShawnZhong (Shawn Zhong).

• Added a formatter for std::exception (#2977, #3012, #3062, #3076, #3119). For example (godbolt):

  #include <fmt/std.h>
  #include <vector>
  
  int main() {
    try {
      std::vector<bool>().at(0);
    } catch(const std::exception& e) {
      fmt::print("{}", e);
    }
  }

  prints:

  vector<bool>::_M_range_check: __n (which is 0) >= this->size() (which is 0)
  

  on libstdc++. Thanks @zach2good (Zach Toogood) and @phprus (Vladislav Shchapov).

• Moved std::error_code formatter from fmt/os.h to fmt/std.h. (#3125). Thanks @phprus (Vladislav Shchapov).

• Added formatters for standard container adapters: std::priority_queue, std::queue and std::stack (#3215, #3279). For example (godbolt):

  #include <fmt/ranges.h>
  #include <stack>
  #include <vector>
  
  int main() {
    auto s = std::stack<bool, std::vector<bool>>();
    for (auto b: {true, false, true}) s.push(b);
    fmt::print("{}\n", s); // prints [true, false, true]
  }

  Thanks @ShawnZhong (Shawn Zhong).

• Added a formatter for std::optional to fmt/std.h. Thanks @tom-huntington.

• Fixed formatting of valueless by exception variants (#3347). Thanks @TheOmegaCarrot.

• Made fmt::ptr accept unique_ptr with a custom deleter (#3177). Thanks @hmbj (Hans-Martin B. Jensen).

• Fixed formatting of noncopyable ranges and nested ranges of chars (#3158 #3286, #3290). Thanks @BRevzin (Barry Revzin).

• Fixed issues with formatting of paths and ranges of paths (#3319, #3321 #3322). Thanks @phprus (Vladislav Shchapov).

• Improved handling of invalid Unicode in paths.

• Enabled compile-time checks on Apple clang 14 and later (#3331). Thanks @cloyce (Cloyce D. Spradling).

• Improved compile-time checks of named arguments (#3105, #3214). Thanks @rbrich (Radek Brich).

• Fixed formatting when both alignment and 0 are given (#3236, #3248). Thanks @ShawnZhong (Shawn Zhong).

• Improved Unicode support in the experimental file API on Windows (#3234, #3293). Thanks @Fros1er (Froster).

• Unified UTF transcoding (#3416). Thanks @phprus (Vladislav Shchapov).

• Added support for UTF-8 digit separators via an experimental locale facet (#1861). For example (godbolt):

  auto loc = std::locale(
    std::locale(), new fmt::format_facet<std::locale>("’"));
  auto s = fmt::format(loc, "{:L}", 1000);

  where ’ is U+2019 used as a digit separator in the de_CH locale.

• Added an overload of formatted_size that takes a locale (#3084, #3087). Thanks @gerboengels.

• Removed the deprecated FMT_DEPRECATED_OSTREAM.

• Fixed a UB when using a null std::string_view with fmt::to_string or format string compilation (#3241, #3244). Thanks @phprus (Vladislav Shchapov).

• Added starts_with to the fallback string_view implementation (#3080). Thanks @phprus (Vladislav Shchapov).

• Added fmt::basic_format_string::get() for compatibility with basic_format_string (#3111). Thanks @huangqinjin.

• Added println for compatibility with C++23 (#3267). Thanks @ShawnZhong (Shawn Zhong).

• Renamed the FMT_EXPORT macro for shared library usage to FMT_LIB_EXPORT.

• Improved documentation (#3108, #3169, #3243). #3404). Thanks @Cleroth and @Vertexwahn.

• Improved build configuration and tests (#3118, #3120, #3188, #3189, #3198, #3205, #3207, #3210, #3240, #3256, #3264, #3299, #3302, #3312, #3317, #3328, #3333, #3369, #3373, #3395, #3406, #3411). Thanks @dimztimz (Dimitrij Mijoski), @phprus (Vladislav Shchapov), @DavidKorczynski, @ChrisThrasher (Chris Thrasher), @FrancoisCarouge (François Carouge), @kennyweiss (Kenny Weiss), @luzpaz, @codeinred (Alecto Irene Perez), @Mixaill (Mikhail Paulyshka), @joycebrum (Joyce), @kevinhwang (Kevin Hwang), @Vertexwahn.

• Fixed a regression in handling empty format specifiers after a colon ({:}) (#3086). Thanks @oxidase (Michael Krasnyk).

• Worked around a broken implementation of std::is_constant_evaluated in some versions of libstdc++ on clang (#3247, #3281). Thanks [@phprus (Vladislav Shc...

9.1.0

• fmt::formatted_size now works at compile time (#3026). For example (godbolt):

  #include <fmt/compile.h>
  
  int main() {
    using namespace fmt::literals;
    constexpr size_t n = fmt::formatted_size("{}"_cf, 42);
    fmt::print("{}\n", n); // prints 2
  }

  Thanks @marksantaniello (Mark Santaniello).

• Fixed handling of invalid UTF-8 (#3038, #3044, #3056). Thanks @phprus (Vladislav Shchapov) and @skeeto (Christopher Wellons).

• Improved Unicode support in ostream overloads of print (#2994, #3001, #3025). Thanks @dimztimz (Dimitrij Mijoski).

• Fixed handling of the sign specifier in localized formatting on systems with 32-bit wchar_t (#3041).

• Added support for wide streams to fmt::streamed (#2994). Thanks @phprus (Vladislav Shchapov).

• Added the n specifier that disables the output of delimiters when formatting ranges (#2981, #2983). For example (godbolt):

  #include <fmt/ranges.h>
  #include <vector>
  
  int main() {
    auto v = std::vector{1, 2, 3};
    fmt::print("{:n}\n", v); // prints 1, 2, 3
  }

  Thanks @BRevzin (Barry Revzin).

• Worked around problematic std::string_view constructors introduced in C++23 (#3030, #3050). Thanks @strega-nil-ms (nicole mazzuca).

• Improve handling (exclusion) of recursive ranges (#2968, #2974). Thanks @Dani-Hub (Daniel Krügler).

• Improved error reporting in format string compilation (#3055).

• Improved the implementation of Dragonbox, the algorithm used for the default floating-point formatting (#2984). Thanks @jk-jeon (Junekey Jeon).

• Fixed issues with floating-point formatting on exotic platforms.

• Improved the implementation of chrono formatting (#3010). Thanks @phprus (Vladislav Shchapov).

• Improved documentation (#2966, #3009, #3020, #3037). Thanks @mwinterb, @jcelerier (Jean-Michaël Celerier) and @remiburtin (Rémi Burtin).

• Improved build configuration (#2991, #2995, #3004, #3007, #3040). Thanks @dimztimz (Dimitrij Mijoski) and @hwhsu1231 (Haowei Hsu).

• Fixed various warnings and compilation issues (#2969, #2971, #2975, #2982, #2985, #2988, #3000, #3006, #3014, #3015, #3021, #3023, #3024, #3029, #3043, #3052, #3053, #3054). Thanks @h-friederich (Hannes Friederich), @dimztimz (Dimitrij Mijoski), @olupton (Olli Lupton), @bernhardmgruber (Bernhard Manfred Gruber), @phprus (Vladislav Shchapov).

Pull Requests

• Add xchar support for fmt::streamed(). by @phprus in #2961
• Fixed typo in changelog example by @mwinterb in #2966
• Exclude recursive ranges from the formatter specialization for ranges by @Dani-Hub in #2974
• Pointless comparison warnings by @federico-busato in #2971
• The n specifier for ranges. by @brevzin in #2981
• Fix for EDG frontend (Intel, NVHPC compilers) by @phprus in #2982
• Yet another simplification of Dragonbox by @jk-jeon in #2984
• Add additional search paths for doxygen on Windows by @hwhsu1231 in #2991
• Fixes IBM XLC behavior with uint128 fallback by @federico-busato in #2985
• Reduce filesize of the tests on MinGW by @dimztimz in #2995
• Improve Unicode handling when writing to an ostream on Windows by @dimztimz in #2994
• Add fmt:: namespace to doc by @jcelerier in #3009
• Range formatter by @brevzin in #2983
• Fix Unicode handling for ostream under Windows with libc++. by @dimztimz in #3001
• Improve CI on Windows, deprecate AppVeyor by @dimztimz in #3007
• Simplify and improve chrono formatting by @phprus in #3010
• Suppress unused typedef warning by @h-friederich in #3021
• Remove -Wl,--as-needed linker option by @phprus in #3024
• Use is_utf8() in print(std::ostream&, ...) by @dimztimz in #3025
• Constexpr formatted_size by @marksantaniello in #3026
• Fix testsuite on MinGW + MSVCRT by @dimztimz in #3029
• Fix docs by @remiburtin in #3037
• Prepare for deprecating FindPythonInterp module. by @hwhsu1231 in #3040
• nvhpc/22.3: workaround for c++17 mode by @olupton in #3043
• Fix decoder on broken utf8 sequences. by @phprus in #3044
• Disable non-type template args for nvhpc by @bernhardmgruber in #3053
• Disable bogus -Wstringop-overflow on GCC 11 by @phprus in #3054
• Fix bugs in utf8 decoder by @phprus in #3056

New Contributors

• @hwhsu1231 made their first contribution in #2991
• @dimztimz made their first contribution in #2995
• @h-friederich made their first contribution in #3021
• @marksantaniello made their first contribution in #3026
• @remiburtin made their first contribution in #3037
• @bernhardmgruber made their first contribution in #3053

Full Changelog: 9.0.0...9.1.0

9.0.0

• Switched to the internal floating point formatter for all decimal presentation formats. In particular this results in consistent rounding on all platforms and removing the s[n]printf fallback for decimal FP formatting.

• Compile-time floating point formatting no longer requires the header-only mode. For example (godbolt):

  #include <array>
  #include <fmt/compile.h>
  
  consteval auto compile_time_dtoa(double value) -> std::array<char, 10> {
    auto result = std::array<char, 10>();
    fmt::format_to(result.data(), FMT_COMPILE("{}"), value);
    return result;
  }
  
  constexpr auto answer = compile_time_dtoa(0.42);

  works with the default settings.

• Improved the implementation of Dragonbox, the algorithm used for the default floating-point formatting (#2713, #2750). Thanks @jk-jeon (Junekey Jeon).

• Made fmt::to_string work with __float128. This uses the internal FP formatter and works even on system without __float128 support in [s]printf.

• Disabled automatic std::ostream insertion operator (operator<<) discovery when fmt/ostream.h is included to prevent ODR violations. You can get the old behavior by defining FMT_DEPRECATED_OSTREAM but this will be removed in the next major release. Use fmt::streamed or fmt::ostream_formatter to enable formatting via std::ostream instead.

• Added fmt::ostream_formatter that can be used to write formatter specializations that perform formatting via std::ostream. For example (godbolt):

  #include <fmt/ostream.h>
  
  struct date {
    int year, month, day;
  
    friend std::ostream& operator<<(std::ostream& os, const date& d) {
      return os << d.year << '-' << d.month << '-' << d.day;
    }
  };
  
  template <> struct fmt::formatter<date> : ostream_formatter {};
  
  std::string s = fmt::format("The date is {}", date{2012, 12, 9});
  // s == "The date is 2012-12-9"

• Added the fmt::streamed function that takes an object and formats it via std::ostream. For example (godbolt):

  #include <thread>
  #include <fmt/ostream.h>
  
  int main() {
    fmt::print("Current thread id: {}\n",
               fmt::streamed(std::this_thread::get_id()));
  }

  Note that fmt/std.h provides a formatter specialization for std::thread::id so you don't need to format it via std::ostream.

• Deprecated implicit conversions of unscoped enums to integers for consistency with scoped enums.

• Added an argument-dependent lookup based format_as extension API to simplify formatting of enums.

• Added experimental std::variant formatting support (#2941). For example (godbolt):

  #include <variant>
  #include <fmt/std.h>
  
  int main() {
    auto v = std::variant<int, std::string>(42);
    fmt::print("{}\n", v);
  }

  prints:

  variant(42)
  

  Thanks @jehelset.

• Added experimental std::filesystem::path formatting support (#2865, #2902, #2917, #2918). For example (godbolt):

  #include <filesystem>
  #include <fmt/std.h>
  
  int main() {
    fmt::print("There is no place like {}.", std::filesystem::path("/home"));
  }

  prints:

  There is no place like "/home".
  

  Thanks @phprus (Vladislav Shchapov).

• Added a std::thread::id formatter to fmt/std.h. For example (godbolt):

  #include <thread>
  #include <fmt/std.h>
  
  int main() {
    fmt::print("Current thread id: {}\n", std::this_thread::get_id());
  }

• Added fmt::styled that applies a text style to an individual argument (#2793). For example (godbolt):

  #include <fmt/chrono.h>
  #include <fmt/color.h>
  
  int main() {
    auto now = std::chrono::system_clock::now();
    fmt::print(
      "[{}] {}: {}\n",
      fmt::styled(now, fmt::emphasis::bold),
      fmt::styled("error", fg(fmt::color::red)),
      "something went wrong");
  }

  prints

  Thanks @rbrugo (Riccardo Brugo).

• Made fmt::print overload for text styles correctly handle UTF-8 (#2681, #2701). Thanks @AlexGuteniev (Alex Guteniev).

• Fixed Unicode handling when writing to an ostream.

• Added support for nested specifiers to range formatting (#2673). For example (godbolt):

  #include <vector>
  #include <fmt/ranges.h>
  
  int main() {
    fmt::print("{::#x}\n", std::vector{10, 20, 30});
  }

  prints [0xa, 0x14, 0x1e].

  Thanks @BRevzin (Barry Revzin).

• Implemented escaping of wide strings in ranges (#2904). Thanks @phprus (Vladislav Shchapov).

• Added support for ranges with begin / end found via the argument-dependent lookup (#2807). Thanks @rbrugo (Riccardo Brugo).

• Fixed formatting of certain kinds of ranges of ranges (#2787). Thanks @BRevzin (Barry Revzin).

• Fixed handling of maps with element types other than std::pair (#2944). Thanks @BrukerJWD (Jonathan W).

• Made tuple formatter enabled only if elements are formattable (#2939, #2940). Thanks @jehelset.

• Made fmt::join compatible with format string compilation (#2719, #2720). Thanks @phprus (Vladislav Shchapov).

• Made compile-time checks work with named arguments of custom types and std::ostream print overloads (#2816, #2817, #2819). Thanks @timsong-cpp.

• Removed make_args_checked because it is no longer needed for compile-time checks (#2760). Thanks @phprus (Vladislav Shchapov).

• Removed the following deprecated APIs: _format, arg_join, the format_to overload that takes a memory buffer, [v]fprintf that takes an ostream.

• Removed the deprecated implicit conversion of [const] signed char* and [const] unsigned char* to C strings.

• Removed the deprecated fmt/locale.h.

• Replaced the deprecated fileno() with descriptor() in buffered_file.

• Moved to_string_view to the detail namespace since it's an implementation detail.

• Made access mode of a created file consistent with fopen by setting S_IWGRP and S_IWOTH (#2733). Thanks @arogge (Andreas Rogge).

• Removed a redundant buffer resize when formatting to std::ostream (#2842, #2843). Thanks @jcelerier (Jean-Michaël Celerier).

• Made precision computation for strings consistent with width (#2888).

• Fixed handling of locale separators in floating point formatting (#2830).

• Made sign specifiers work with __int128_t (#2773).

• Improved support for systems such as CHERI with extra data stored in pointers (#2932). Thanks @davidchisnall (David Chisnall).

• Improved documentation (#2706, #2712, #2789, #2803, #2805, #2815, #2924). Thanks @BRevzin (Barry Revzin), @Pokechu22, @setoye (Alta), @rtobar, @rbrugo (Riccardo Brugo), @anoonD (cre), @leha-bot (Alex).

• Improved build configuration (#2766, #2772, #2836, #2852, #2907, #2913, #2914). Thanks @kambala-decapitator (Andrey Filipenkov), @mattiasljungstrom (Mattias Ljungström), @kieselnb (Nick Kiesel), @nathannaveen, @Vertexwahn.

• Fixed various warnings and compilation issues (#2408, #2507, #2697, #2715, #2717, #2722, #2724, #2725, #2726, #2728, #2732, #2738, #2742, #2744, #2745, #2746, #2754, #2755, #2757, #2758, #2761, #2762, #2763, #2765, #2769, #2770, #2771, #2777, #2779, #2782, #2783, #2794, #2796, #2797, #2801, #2802, #2808, #2818, #2819, #2829, #2835, #2848, #2860, #2861, #2882, #2886, #2891, #2892, #2895, #2896, #2903, #2906, #2908, #2909, #2920, #2922, #2927, #2929, #2936, #2937, #2938, #2951, #2954, #2957, #2958, #2960). Thanks @matrackif @Tobi823 (Tobias Hellmann), @ivan-volnov (Ivan Volnov), @VasiliPupkin256, @federico-busato (Federico), @barcharcraz (Charlie Barto), @jk-jeon (Junekey Jeon), @HazardyKnusperkeks (Björn Schäpers), @dalboris (Boris Dalstein), @seanm (Sean McBride), @gsjaardema (Greg Sjaardema), @timsong-cpp, [@seanm (Sean Mc...

8.1.1

• Restored ABI compatibility with version 8.0.x (#2695, #2696). Thanks @saraedum (Julian Rüth).

• Fixed chrono formatting on big endian systems (#2698, #2699). Thanks @phprus (Vladislav Shchapov) and @xvitaly (Vitaly Zaitsev).

• Fixed a linkage error with mingw (#2691, #2692). Thanks @rbberger (Richard Berger).

8.1.0

• Optimized chrono formatting (#2500, #2537, #2541, #2544, #2550, #2551, #2576, #2577, #2586, #2591, #2594, #2602, #2617, #2628, #2633, #2670, #2671).

  Processing of some specifiers such as %z and %Y is now up to 10-20 times faster, for example on GCC 11 with libstdc++:

  ----------------------------------------------------------------------------
  Benchmark                                  Before             After
  ----------------------------------------------------------------------------
  FMTFormatter_z                             261 ns             26.3 ns
  FMTFormatterCompile_z                      246 ns             11.6 ns
  FMTFormatter_Y                             263 ns             26.1 ns
  FMTFormatterCompile_Y                      244 ns             10.5 ns
  ----------------------------------------------------------------------------
  

  Thanks @phprus (Vladislav Shchapov) and @toughengineer (Pavel Novikov).

• Implemented subsecond formatting for chrono durations (#2623). For example (godbolt):

  #include <fmt/chrono.h>
  
  int main() {
    fmt::print("{:%S}", std::chrono::milliseconds(1234));
  }

  prints "01.234".

  Thanks @matrackif.

• Fixed handling of precision 0 when formatting chrono durations (#2587, #2588). Thanks @lukester1975.

• Fixed an overflow on invalid inputs in the tm formatter (#2564). Thanks @phprus (Vladislav Shchapov).

• Added fmt::group_digits that formats integers with a non-localized digit separator (comma) for groups of three digits. For example (godbolt):

  #include <fmt/format.h>
  
  int main() {
    fmt::print("{} dollars", fmt::group_digits(1000000));
  }

  prints "1,000,000 dollars".

• Added support for faint, conceal, reverse and blink text styles (#2394):

  blink.mp4

  Thanks @benit8 (Benoît Lormeau) and @data-man (Dmitry Atamanov).

• Added experimental support for compile-time floating point formatting (#2426, #2470). It is currently limited to the header-only mode. Thanks @alexezeder (Alexey Ochapov).

• Added UDL-based named argument support to compile-time format string checks (#2640, #2649). For example (godbolt):

  #include <fmt/format.h>
  
  int main() {
    using namespace fmt::literals;
    fmt::print("{answer:s}", "answer"_a=42);
  }

  gives a compile-time error on compilers with C++20 consteval and non-type template parameter support (gcc 10+) because s is not a valid format specifier for an integer.

  Thanks @alexezeder (Alexey Ochapov).

• Implemented escaping of string range elements. For example (godbolt):

  #include <fmt/ranges.h>
  #include <vector>
  
  int main() {
    fmt::print("{}", std::vector<std::string>{"\naan"});
  }

  is now printed as:

  ["\naan"]
  

  instead of:

  ["
  aan"]
  

• Switched to JSON-like representation of maps and sets for consistency with Python's str.format. For example (godbolt):

  #include <fmt/ranges.h>
  #include <map>
  
  int main() {
    fmt::print("{}", std::map<std::string, int>{{"answer", 42}});
  }

  is now printed as:

  {"answer": 42}
  

• Extended fmt::join to support C++20-only ranges (#2549). Thanks @BRevzin (Barry Revzin).

• Optimized handling of non-const-iterable ranges and implemented initial support for non-const-formattable types.

• Disabled implicit conversions of scoped enums to integers that was accidentally introduced in earlier versions (#1841).

• Deprecated implicit conversion of [const] signed char* and [const] unsigned char* to C strings.

• Deprecated _format, a legacy UDL-based format API (#2646). Thanks @alexezeder (Alexey Ochapov).

• Marked format, formatted_size and to_string as [[nodiscard]] (#2612). @0x8000-0000 (Florin Iucha).

• Added missing diagnostic when trying to format function and member pointers as well as objects convertible to pointers which is explicitly disallowed (#2598, #2609, #2610). Thanks @AlexGuteniev (Alex Guteniev).

• Optimized writing to a contiguous buffer with format_to_n (#2489). Thanks @Roman-Koshelev.

• Optimized writing to non-char buffers (#2477). Thanks @Roman-Koshelev.

• Decimal point is now localized when using the L specifier.

• Improved floating point formatter implementation (#2498, #2499). Thanks @Roman-Koshelev.

• Fixed handling of very large precision in fixed format (#2616).

• Made a table of cached powers used in FP formatting static (#2509). Thanks @jk-jeon (Junekey Jeon).

• Resolved a lookup ambiguity with C++20 format-related functions due to ADL (#2639, #2641). Thanks @mkurdej (Marek Kurdej).

• Removed unnecessary inline namespace qualification (#2642, #2643). Thanks @mkurdej (Marek Kurdej).

• Implemented argument forwarding in format_to_n (#2462, #2463). Thanks @owent (WenTao Ou).

• Fixed handling of implicit conversions in fmt::to_string and format string compilation (#2565).

• Changed the default access mode of files created by fmt::output_file to -rw-r--r-- for consistency with fopen (#2530).

• Make fmt::ostream::flush public (#2435).

• Improved C++14/17 attribute detection (#2615). Thanks @AlexGuteniev (Alex Guteniev).

• Improved consteval detection for MSVC (#2559). Thanks @DanielaE (Daniela Engert).

• Improved documentation (#2406, #2446, #2493, #2513, #2515, #2522, #2562, #2575, #2606, #2620, #2676). Thanks @sobolevn (Nikita Sobolev), @UnePierre (Max FERGER), @zhsj, @phprus (Vladislav Shchapov), @ericcurtin (Eric Curtin), @Lounarok.

• Improved fuzzers and added a fuzzer for chrono timepoint formatting (#2461, #2469). @pauldreik (Paul Dreik),

• Added the FMT_SYSTEM_HEADERS CMake option setting which marks {fmt}'s headers as system. It can be used to suppress warnings (#2644, #2651). Thanks @alexezeder (Alexey Ochapov).

• Added the Bazel build system support (#2505, #2516). Thanks @Vertexwahn.

• Improved build configuration and tests (#2437, #2558, #2648, #2650, #2663, #2677). Thanks @DanielaE (Daniela Engert), @alexezeder (Alexey Ochapov), @phprus (Vladislav Shchapov).

• Fixed various warnings and compilation issues (#2353, #2356, #2399, #2408, #2414, #2427, #2432, #2442, #2434, #2439, #2447, #2450, #2455, #2465, #2472, #2474, #2476, #2478, #2479, #2481, #2482, #2483, #2490, #2491, #2510, #2518, #2528, #2529, #2539, #2540, #2545, #2555, #2557, #2570, #2573, #2582, #2605, #2611, #2647, #2627, #2630, #2635, #2638, #2653, #2654, #2661, #2664, #2684). Thanks @DanielaE (Daniela Engert), @mwinterb, @cdacamar (Cameron DaCamara), @TrebledJ (Johnathan), @bodomartin (brm), @cquammen (Cory Quammen), @white238 (Chris White), @mmarkeloff (Max), @palacaze (Pierre-Antoine Lacaze), @jcelerier (Jean-Michaël Celerier), @mborn-adi (Mathias Born), @BrukerJWD (Jonathan W), @spyridon97 (Spiros Tsalikis), @phprus (Vladislav Shchapov), @oliverlee (Oliver Lee), @joshessman-llnl (Josh Essman), @akohlmey (Axel Kohlmeyer), @timkalu, @olupton (Olli Lupton), @Acretock, @alexezeder (Alexey Ochapov), @andrewcorrigan (Andrew Corrigan), @lucpelletier, @HazardyKnusperkeks (Björn Schäpers).

8.0.1

• Fixed the version number in the inline namespace (#2374).

• Added a missing presentation type check for std::string (#2402).

• Fixed a linkage error when mixing code built with clang and gcc (#2377).

• Fixed documentation issues (#2396, #2403, #2406). Thanks @mkurdej (Marek Kurdej).

• Removed dead code in FP formatter ( #2398). Thanks @javierhonduco (Javier Honduvilla Coto).

• Fixed various warnings and compilation issues (#2351, #2359, #2365, #2368, #2370, #2376, #2381, #2382, #2386, #2389, #2395, #2397, #2400 #2401, #2407). Thanks @zx2c4 (Jason A. Donenfeld), @AidanSun05 (Aidan Sun), @mattiasljungstrom (Mattias Ljungström), @joemmett (Jonathan Emmett), @erengy (Eren Okka), @patlkli (Patrick Geltinger), @gsjaardema (Greg Sjaardema), @phprus (Vladislav Shchapov).

8.0.0

• Enabled compile-time format string check by default. For example (godbolt):

  #include <fmt/core.h>
  
  int main() {
    fmt::print("{:d}", "I am not a number");
  }

  gives a compile-time error on compilers with C++20 consteval support (gcc 10+, clang 11+) because d is not a valid format specifier for a string.

  To pass a runtime string wrap it in fmt::runtime:

  fmt::print(fmt::runtime("{:d}"), "I am not a number");

• Added compile-time formatting (#2019, #2044, #2056, #2072, #2075, #2078, #2129, #2326). For example (godbolt):

  #include <fmt/compile.h>
  
  consteval auto compile_time_itoa(int value) -> std::array<char, 10> {
    auto result = std::array<char, 10>();
    fmt::format_to(result.data(), FMT_COMPILE("{}"), value);
    return result;
  }
  
  constexpr auto answer = compile_time_itoa(42);

  Most of the formatting functionality is available at compile time with a notable exception of floating-point numbers and pointers. Thanks @alexezeder (Alexey Ochapov).

• Optimized handling of format specifiers during format string compilation. For example, hexadecimal formatting ("{:x}") is now 3-7x faster than before when using format_to with format string compilation and a stack-allocated buffer (#1944).

  Before (7.1.3):

  ----------------------------------------------------------------------------
  Benchmark                                  Time             CPU   Iterations
  ----------------------------------------------------------------------------
  FMTCompileOld/0                         15.5 ns         15.5 ns     43302898
  FMTCompileOld/42                        16.6 ns         16.6 ns     43278267
  FMTCompileOld/273123                    18.7 ns         18.6 ns     37035861
  FMTCompileOld/9223372036854775807       19.4 ns         19.4 ns     35243000
  ----------------------------------------------------------------------------
  

  After (8.x):

  ----------------------------------------------------------------------------
  Benchmark                                  Time             CPU   Iterations
  ----------------------------------------------------------------------------
  FMTCompileNew/0                         1.99 ns         1.99 ns    360523686
  FMTCompileNew/42                        2.33 ns         2.33 ns    279865664
  FMTCompileNew/273123                    3.72 ns         3.71 ns    190230315
  FMTCompileNew/9223372036854775807       5.28 ns         5.26 ns    130711631
  ----------------------------------------------------------------------------
  

  It is even faster than std::to_chars from libc++ compiled with clang on macOS:

  ----------------------------------------------------------------------------
  Benchmark                                  Time             CPU   Iterations
  ----------------------------------------------------------------------------
  ToChars/0                               4.42 ns         4.41 ns    160196630
  ToChars/42                              5.00 ns         4.98 ns    140735201
  ToChars/273123                          7.26 ns         7.24 ns     95784130
  ToChars/9223372036854775807             8.77 ns         8.75 ns     75872534
  ----------------------------------------------------------------------------
  

  In other cases, especially involving std::string construction, the speed up is usually lower because handling format specifiers takes a smaller fraction of the total time.

• Added the _cf user-defined literal to represent a compiled format string. It can be used instead of the FMT_COMPILE macro (#2043, #2242):

  #include <fmt/compile.h>
  
  using namespace fmt::literals;
  auto s = fmt::format(FMT_COMPILE("{}"), 42); // 🙁 not modern
  auto s = fmt::format("{}"_cf, 42);           // 🙂 modern as hell

  It requires compiler support for class types in non-type template parameters (a C++20 feature) which is available in GCC 9.3+. Thanks @alexezeder (Alexey Ochapov).

• Format string compilation now requires format functions of formatter specializations for user-defined types to be const:

  template <> struct fmt::formatter<my_type>: formatter<string_view> {
    template <typename FormatContext>
    auto format(my_type obj, FormatContext& ctx) const {  // Note const here.
      // ...
    }
  };

• Added UDL-based named argument support to format string compilation (#2243, #2281). For example:

  #include <fmt/compile.h>
  
  using namespace fmt::literals;
  auto s = fmt::format(FMT_COMPILE("{answer}"), "answer"_a = 42);

  Here the argument named "answer" is resolved at compile time with no runtime overhead. Thanks @alexezeder (Alexey Ochapov).

• Added format string compilation support to fmt::print (#2280, #2304). Thanks @alexezeder (Alexey Ochapov).

• Added initial support for compiling {fmt} as a C++20 module (#2235, #2240, #2260, #2282, #2283, #2288, #2298, #2306, #2307, #2309, #2318, #2324, #2332, #2340). Thanks @DanielaE (Daniela Engert).

• Made symbols private by default reducing shared library size (#2301). For example there was a ~15% reported reduction on one platform. Thanks @sergiud (Sergiu Deitsch).

• Optimized includes making the result of preprocessing fmt/format.h ~20% smaller with libstdc++/C++20 and slightly improving build times (#1998).

• Added support of ranges with non-const begin / end (#1953). Thanks @kitegi (sarah).

• Added support of std::byte and other formattable types to fmt::join (#1981, #2040, #2050, #2262). For example:

  #include <fmt/format.h>
  #include <cstddef>
  #include <vector>
  
  int main() {
    auto bytes = std::vector{std::byte(4), std::byte(2)};
    fmt::print("{}", fmt::join(bytes, ""));
  }

  prints "42".

  Thanks @kamibo (Camille Bordignon).

• Implemented the default format for std::chrono::system_clock (#2319, #2345). For example:

  #include <fmt/chrono.h>
  
  int main() {
    fmt::print("{}", std::chrono::system_clock::now());
  }

  prints "2021-06-18 15:22:00" (the output depends on the current date and time). Thanks @sunmy2019.

• Made more chrono specifiers locale independent by default. Use the 'L' specifier to get localized formatting. For example:

  #include <fmt/chrono.h>
  
  int main() {
    std::locale::global(std::locale("ru_RU.UTF-8"));
    auto monday = std::chrono::weekday(1);
    fmt::print("{}\n", monday);   // prints "Mon"
    fmt::print("{:L}\n", monday); // prints "пн"
  }

• Improved locale handling in chrono formatting (#2337, #2349, #2350). Thanks @phprus (Vladislav Shchapov).

• Deprecated fmt/locale.h moving the formatting functions that take a locale to fmt/format.h (char) and fmt/xchar (other overloads). This doesn't introduce a dependency on <locale> so there is virtually no compile time effect.

• Made parameter order in vformat_to consistent with format_to (#2327).

• Added support for time points with arbitrary durations (#2208). For example:

  #include <fmt/chrono.h>
  
  int main() {
    using tp = std::chrono::time_point<
      std::chrono::system_clock, std::chrono::seconds>;
    fmt::print("{:%S}", tp(std::chrono::seconds(42)));
  }

  prints "42".

• Formatting floating-point numbers no longer produces trailing zeros by default for consistency with std::format. For example:

  #include <fmt/core.h>
  
  int main() {
    fmt::print("{0:.3}", 1.1);
  }

  prints "1.1". Use the '#' specifier to keep trailing zeros.

• Dropped a limit on the number of elements in a range and replaced {} with [] as range delimiters for consistency with Python's str.format.

• The 'L' specifier for locale-specific numeric formatting can now be combined with presentation specifiers as in std::format. For example:

  #include <fmt/core.h>
  #include <locale>
  
  int main() {
    std::locale::global(std::locale("fr_FR.UTF-8"));
    fmt::print("{0:.2Lf}", 0.42);
  }

  prints "0,42". The deprecated 'n' specifier has been removed.

• Made the 0 specifier ignored for infinity and NaN (#2305, #2310). Thanks @Liedtke (Matthias Liedtke).

• Made the hexfloat formatting use the right alignment by default (#2308, #2317). Thanks @Liedtke (Matthias Liedtke).

• Removed the deprecated numeric alignment ('='). Use the '0' specifier instead.

• Removed the deprecated fmt/posix.h header that has been replaced with fmt/os.h.

• Removed the deprecated format_to_n_context, format_to_n_args and make_format_to_n_args. They have been replaced with format_context, format_args and make_format_args respectively.

• Moved wchar_t-specific functions and types to fmt/xchar.h. You can define FMT_DEPRECATED_INCLUDE_XCHAR to automatically include fmt/xchar.h from fmt/format.h but this will be disabled in the next major release.

• Fixed handling of the '+' spec...

7.1.3

• Fixed handling of buffer boundaries in format_to_n (#1996, #2029).

• Fixed linkage errors when linking with a shared library (#2011).

• Reintroduced ostream support to range formatters (#2014).

• Worked around an issue with mixing std versions in gcc (#2017).

7.1.2

• Fixed floating point formatting with large precision (#1976).

7.1.1

• Fixed ABI compatibility with 7.0.x (#1961).

• Added the FMT_ARM_ABI_COMPATIBILITY macro to work around ABI incompatibility between GCC and Clang on ARM (#1919).

• Worked around a SFINAE bug in GCC 8 (#1957).

• Fixed linkage errors when building with GCC's LTO (#1955).

• Fixed a compilation error when building without __builtin_clz or equivalent (#1968). Thanks @tohammer (Tobias Hammer).

• Fixed a sign conversion warning (#1964). Thanks @OptoCloud.