1 Introduction

With the adoption of [P2741R3], static_assert can take a std::string. And the standard library has a very convenient way of producing a std::string by way of std::format. Except that it’s not constexpr, so it’s not suitable for that purpose. Let’s change that.

std::format is specified to use type-erasure, by way of a handle type as specified in 22.14.8.1 [format.arg]/10:

namespace std {
  template<class Context>
  class basic_format_arg<Context>::handle {
    const void* ptr_;                                           // exposition only
    void (*format_)(basic_format_parse_context<char_type>&,
                    Context&, const void*);                     // exposition only

    template<class T> explicit handle(T& val) noexcept;         // exposition only

    friend class basic_format_arg<Context>;                     // exposition only

  public:
    void format(basic_format_parse_context<char_type>&, Context& ctx) const;
  };
}

Such a type was unusable during constant evaluate due to the to cast ptr_ from void const* to some T const* in the format_ function. But with the adoption of [P2738R1], that cast is now allowed. And, if such handles were constructed in place, such construction is now allowed too with the adoption of [P2747R2].

There’s really nothing that stands in the way of making std::format fully constexpr.

Except… there are two categories of types that we cannot immediately support without further work.

1.1 Floating Point

Integral and floating point types are specified to use std::to_chars. However, only one overload of to_chars is currently declared constexpr in 22.13.1 [charconv.syn]:

constexpr to_chars_result to_chars(char* first, char* last,           // freestanding
                                   integer-type value, int base = 10);
to_chars_result to_chars(char* first, char* last,                     // freestanding
                         bool value, int base = 10) = delete;
to_chars_result to_chars(char* first, char* last,                     // freestanding-deleted
                         floating-point-type value);
to_chars_result to_chars(char* first, char* last,                     // freestanding-deleted
                         floating-point-type value, chars_format fmt);
to_chars_result to_chars(char* first, char* last,                     // freestanding-deleted
                         floating-point-type value, chars_format fmt, int precision);

That was added by [P2291R3] which explicitly noted the difficulty of implementing floating point support. That paper was adopted in 2021, and a lot has chanced since then (including multiple floating point formatting algorithms). So perhaps this decision should be revisited at some point.

1.2 Chrono Types

The chrono types (all of them) are specified to be formatted as if by streaming through basic_ostringstream<char> (see 29.12 [time.format]) and absolutely nothing in that type is constexpr. That won’t work without further changes — most likely by changing how chrono type formatting works rather than by changing basic_ostringstream.

1.3 Other Types

There are a few other types in the standard library that have formatters that rely on functionality that is not currently constexpr: stacktrace_entry, filesystem::path, and thread::id. Those will remain non-constexpr formattable. Additionally, void const* cannot be formatted at compile time because it cannot be converted to an address.

1.4 Implementation Experience

The {fmt} library has supported compile-time formatting for a while, just through a different API with the format string annotated: as format_to(out, FMT_COMPILE("x={}"), x). That implementation even supports floating point types at compile time (example), but not the chrono types.

Implementing this in libstdc++ was mostly a matter of marking a lot of functions constexpr (which is easy thanks to -fimplicit-constexpr). There were only two specific changes I had to make:

When assigning into the union, the current implementation does this:
```
template<typename _Tp>
 [[__gnu__::__always_inline__]]
 void
 _M_set(_Tp __v) noexcept
 {
   if constexpr (derived_from<_Tp, _HandleBase>)
     std::construct_at(&_M_handle, __v);
   else
     _S_get<_Tp>(*this) = __v;
 }
```
In that last assignment, _S_get<_Tp>(*this) returns the appropriate member of the union for the type _Tp. That doesn’t work during constant evaluation (although it can probably be made to) since there’s no active member yet. So I had to change that to be a new function invoked like _S_set(*this, __v).
Like {fmt}, libstdc++ has a derived scanner type that is only constructed during constant evaluation time and is only used for parsing. This gets confusing when we also do formatting at compile time, because we don’t have that compile-time scanner type, so casting down to it will fail. In libstdc++, this is easy to fix though since there is a member pointer to array of types that is just nullptr if that scanner isn’t used — so we can check that before downcasting.

And with that, this works (you can see the changes in question on line 3322 for the changed call to _S_set and 4392 for the check, at compile-time only, of _M_types):

template <auto F>
constexpr auto formatted = []{
    static constexpr auto array = []{
        std::array<char, 100> a = {};
        F(a.data());
        return a;
    }();

    return std::string_view(array.data());
}();

static_assert(formatted<[](char* p){std::format_to(p, "x={}", 42);}> == "x=42");
static_assert(formatted<[](char* p){std::format_to(p, "x={:{}}", 42, 5);}> == "x=   42");

2 Proposal

Make std::format constexpr, with the understanding that we will not (yet) be able to format floating point and chrono types during constant evaluation time, nor the locale-aware overloads. The facility is still plenty useful even if we can’t format everything quite yet!

2.1 Wording

[ Drafting note: I’m introducing the term “constexpr-enabled” to describe the standard library formatters that have a constexpr format function. This will include most types, but not the floating-point or chrono types mentioned earlier. As such, there are no changes to 29.12 [time.format] here. Also, most of the wording is simply adding constexpr to a lot of places — only the synopses are shown in the diff below. ].

Add constexpr to a lot of places in 22.14.1 [format.syn]:

  namespace std {
    // [format.context], class template basic_format_context
    template<class Out, class charT> class basic_format_context;
    using format_context = basic_format_context<unspecified, char>;
    using wformat_context = basic_format_context<unspecified, wchar_t>;

    // [format.args], class template basic_format_args
    template<class Context> class basic_format_args;
    using format_args = basic_format_args<format_context>;
    using wformat_args = basic_format_args<wformat_context>;

    // [format.fmt.string], class template basic_format_string
    template<class charT, class... Args>
      struct basic_format_string;

    template<class charT> struct runtime-format-string {                  // exposition only
    private:
      basic_string_view<charT> str;                                       // exposition only
    public:
-     runtime-format-string(basic_string_view<charT> s) noexcept : str(s) {}
+     constexpr runtime-format-string(basic_string_view<charT> s) noexcept : str(s) {}
      runtime-format-string(const runtime-format-string&) = delete;
      runtime-format-string& operator=(const runtime-format-string&) = delete;
    };
-   runtime-format-string<char> runtime_format(string_view fmt) noexcept { return fmt; }
-   runtime-format-string<wchar_t> runtime_format(wstring_view fmt) noexcept { return fmt; }
+   constexpr runtime-format-string<char> runtime_format(string_view fmt) noexcept { return fmt; }
+   constexpr runtime-format-string<wchar_t> runtime_format(wstring_view fmt) noexcept { return fmt; }

    template<class... Args>
      using format_string = basic_format_string<char, type_identity_t<Args>...>;
    template<class... Args>
      using wformat_string = basic_format_string<wchar_t, type_identity_t<Args>...>;

    // [format.functions], formatting functions
    template<class... Args>
-     string format(format_string<Args...> fmt, Args&&... args);
+     constexpr string format(format_string<Args...> fmt, Args&&... args);
    template<class... Args>
-     wstring format(wformat_string<Args...> fmt, Args&&... args);
+     constexpr wstring format(wformat_string<Args...> fmt, Args&&... args);
    template<class... Args>
      string format(const locale& loc, format_string<Args...> fmt, Args&&... args);
    template<class... Args>
      wstring format(const locale& loc, wformat_string<Args...> fmt, Args&&... args);

-   string vformat(string_view fmt, format_args args);
-   wstring vformat(wstring_view fmt, wformat_args args);
+   constexpr string vformat(string_view fmt, format_args args);
+   constexpr wstring vformat(wstring_view fmt, wformat_args args);
    string vformat(const locale& loc, string_view fmt, format_args args);
    wstring vformat(const locale& loc, wstring_view fmt, wformat_args args);

    template<class Out, class... Args>
-     Out format_to(Out out, format_string<Args...> fmt, Args&&... args);
+     constexpr Out format_to(Out out, format_string<Args...> fmt, Args&&... args);
    template<class Out, class... Args>
-     Out format_to(Out out, wformat_string<Args...> fmt, Args&&... args);
+     constexpr Out format_to(Out out, wformat_string<Args...> fmt, Args&&... args);
    template<class Out, class... Args>
      Out format_to(Out out, const locale& loc, format_string<Args...> fmt, Args&&... args);
    template<class Out, class... Args>
      Out format_to(Out out, const locale& loc, wformat_string<Args...> fmt, Args&&... args);

    template<class Out>
-     Out vformat_to(Out out, string_view fmt, format_args args);
+     constexpr Out vformat_to(Out out, string_view fmt, format_args args);
    template<class Out>
-     Out vformat_to(Out out, wstring_view fmt, wformat_args args);
+     constexpr Out vformat_to(Out out, wstring_view fmt, wformat_args args);
    template<class Out>
      Out vformat_to(Out out, const locale& loc, string_view fmt, format_args args);
    template<class Out>
      Out vformat_to(Out out, const locale& loc, wstring_view fmt, wformat_args args);

    template<class Out> struct format_to_n_result {
      Out out;
      iter_difference_t<Out> size;
    };
    template<class Out, class... Args>
-     format_to_n_result<Out> format_to_n(Out out, iter_difference_t<Out> n,
+     constexpr format_to_n_result<Out> format_to_n(Out out, iter_difference_t<Out> n,
                                          format_string<Args...> fmt, Args&&... args);
    template<class Out, class... Args>
-     format_to_n_result<Out> format_to_n(Out out, iter_difference_t<Out> n,
+     constexpr format_to_n_result<Out> format_to_n(Out out, iter_difference_t<Out> n,
                                          wformat_string<Args...> fmt, Args&&... args);
    template<class Out, class... Args>
      format_to_n_result<Out> format_to_n(Out out, iter_difference_t<Out> n,
                                          const locale& loc, format_string<Args...> fmt,
                                          Args&&... args);
    template<class Out, class... Args>
      format_to_n_result<Out> format_to_n(Out out, iter_difference_t<Out> n,
                                          const locale& loc, wformat_string<Args...> fmt,
                                          Args&&... args);

    template<class... Args>
-     size_t formatted_size(format_string<Args...> fmt, Args&&... args);
+     constexpr size_t formatted_size(format_string<Args...> fmt, Args&&... args);
    template<class... Args>
-     size_t formatted_size(wformat_string<Args...> fmt, Args&&... args);
+     constexpr size_t formatted_size(wformat_string<Args...> fmt, Args&&... args);
    template<class... Args>
      size_t formatted_size(const locale& loc, format_string<Args...> fmt, Args&&... args);
    template<class... Args>
      size_t formatted_size(const locale& loc, wformat_string<Args...> fmt, Args&&... args);

    // [format.formatter], formatter
    template<class T, class charT = char> struct formatter;

    // [format.formatter.locking], formatter locking
    template<class T>
      constexpr bool enable_nonlocking_formatter_optimization = false;

    // [format.formattable], concept formattable
    template<class T, class charT>
      concept formattable = see below;

    template<class R, class charT>
      concept const-formattable-range =                                   // exposition only
        ranges::input_range<const R> &&
        formattable<ranges::range_reference_t<const R>, charT>;

    template<class R, class charT>
      using fmt-maybe-const =                                             // exposition only
        conditional_t<const-formattable-range<R, charT>, const R, R>;

    // [format.parse.ctx], class template basic_format_parse_context
    template<class charT> class basic_format_parse_context;
    using format_parse_context = basic_format_parse_context<char>;
    using wformat_parse_context = basic_format_parse_context<wchar_t>;

    // [format.range], formatting of ranges
    // [format.range.fmtkind], variable template format_kind
    enum class range_format {
      disabled,
      map,
      set,
      sequence,
      string,
      debug_string
    };

    template<class R>
      constexpr unspecified format_kind = unspecified;

    template<ranges::input_range R>
        requires same_as<R, remove_cvref_t<R>>
      constexpr range_format format_kind<R> = see below;

    // [format.range.formatter], class template range_formatter
    template<class T, class charT = char>
      requires same_as<remove_cvref_t<T>, T> && formattable<T, charT>
    class range_formatter;

    // [format.range.fmtdef], class template range-default-formatter
    template<range_format K, ranges::input_range R, class charT>
      struct range-default-formatter;                                     // exposition only

    // [format.range.fmtmap], [format.range.fmtset], [format.range.fmtstr], specializations for maps, sets, and strings
    template<ranges::input_range R, class charT>
      requires (format_kind<R> != range_format::disabled) &&
               formattable<ranges::range_reference_t<R>, charT>
    struct formatter<R, charT> : range-default-formatter<format_kind<R>, R, charT> { };

    template<ranges::input_range R>
        requires (format_kind<R> != range_format::disabled)
      inline constexpr bool enable_nonlocking_formatter_optimization<R> = false;

    // [format.arguments], arguments
    // [format.arg], class template basic_format_arg
    template<class Context> class basic_format_arg;

    // [format.arg.store], class template format-arg-store
    template<class Context, class... Args> class format-arg-store;        // exposition only

    template<class Context = format_context, class... Args>
-     format-arg-store<Context, Args...>
+     constexpr format-arg-store<Context, Args...>
        make_format_args(Args&... fmt_args);
    template<class... Args>
-     format-arg-store<wformat_context, Args...>
+     constexpr format-arg-store<wformat_context, Args...>
        make_wformat_args(Args&... args);

    // [format.error], class format_error
    class format_error;
  }

Apply the same changes where these functions are referenced.

Add to 22.14.6.4 [format.formatter.spec]:

2 Let charT be either char or wchar_t. Each specialization of formatter is either enabled or disabled, as described below. A debug-enabled specialization of formatter additionally provides a public, constexpr, non-static member function set_debug_format() which modifies the state of the formatter to be as if the type of the std-format-spec parsed by the last call to parse were ?. A constexpr-enabled specialization of formatter has its format member function declared constexpr. Each header that declares the template formatter provides the following enabled specializations:
(2.1) The debug-enabled and constexpr-enabled specializations
template<> struct formatter<char, char>;
template<> struct formatter<char, wchar_t>;
template<> struct formatter<wchar_t, wchar_t>;
(2.2) For each charT, the debug-enabled and constexpr-enabled string type specializations
template<> struct formatter<charT*, charT>;
template<> struct formatter<const charT*, charT>;
template<size_t N> struct formatter<charT[N], charT>;
template<class traits, class Allocator>
  struct formatter<basic_string<charT, traits, Allocator>, charT>;
template<class traits>
  struct formatter<basic_string_view<charT, traits>, charT>;
(2.3) For each charT, for each cv-unqualified arithmetic type ArithmeticT other than char, wchar_t, char8_t, char16_t, or char32_t, a specialization that is constexpr-enabled unless ArithmeticT is a floating-point type
template<> struct formatter<ArithmeticT, charT>;
(2.4) For each charT, the constexpr-enabled pointer type specializations
template<> struct formatter<nullptr_t, charT>;
(2.5) For each charT, the pointer type specializations
template<> struct formatter<void*, charT>;
template<> struct formatter<const void*, charT>;
The parse member functions of these formatters interpret the format specification as a std-format-spec as described in [format.string.std].

Mark basic_format_context as mostly constexpr in 22.14.6.7 [format.context] (everything but locale(), and repeated in the specification of these functions):

  namespace std {
    template<class Out, class charT>
    class basic_format_context {
      basic_format_args<basic_format_context> args_;      // exposition only
      Out out_;                                           // exposition only

      basic_format_context(const basic_format_context&) = delete;
      basic_format_context& operator=(const basic_format_context&) = delete;

    public:
      using iterator = Out;
      using char_type = charT;
      template<class T> using formatter_type = formatter<T, charT>;

-     basic_format_arg<basic_format_context> arg(size_t id) const noexcept;
+     constexpr basic_format_arg<basic_format_context> arg(size_t id) const noexcept;
      std::locale locale();

-     iterator out();
-     void advance_to(iterator it);
+     constexpr iterator out();
+     constexpr void advance_to(iterator it);
    };
  }

Mark range_formatter::format as constexpr in 22.14.7.2 [format.range.formatter] and repeated in its specification:

  namespace std {
    template<class T, class charT = char>
      requires same_as<remove_cvref_t<T>, T> && formattable<T, charT>
    class range_formatter {
      formatter<T, charT> underlying_;                                          // exposition only
      basic_string_view<charT> separator_ = STATICALLY-WIDEN<charT>(", ");      // exposition only
      basic_string_view<charT> opening-bracket_ = STATICALLY-WIDEN<charT>("["); // exposition only
      basic_string_view<charT> closing-bracket_ = STATICALLY-WIDEN<charT>("]"); // exposition only

    public:
      constexpr void set_separator(basic_string_view<charT> sep) noexcept;
      constexpr void set_brackets(basic_string_view<charT> opening,
                                  basic_string_view<charT> closing) noexcept;
      constexpr formatter<T, charT>& underlying() noexcept { return underlying_; }
      constexpr const formatter<T, charT>& underlying() const noexcept { return underlying_; }

      template<class ParseContext>
        constexpr typename ParseContext::iterator
          parse(ParseContext& ctx);

      template<ranges::input_range R, class FormatContext>
          requires formattable<ranges::range_reference_t<R>, charT> &&
                   same_as<remove_cvref_t<ranges::range_reference_t<R>>, T>
-       typename FormatContext::iterator
+       constexpr typename FormatContext::iterator
          format(R&& r, FormatContext& ctx) const;
    };
  }

And likewise with range-default-formatter in 22.14.7.3 [format.range.fmtdef]:

  namespace std {
    template<ranges::input_range R, class charT>
    struct range-default-formatter<range_format::sequence, R, charT> {    // exposition only
    private:
      using maybe-const-r = fmt-maybe-const<R, charT>;                    // exposition only
      range_formatter<remove_cvref_t<ranges::range_reference_t<maybe-const-r>>,
                      charT> underlying_;                                 // exposition only

    public:
      constexpr void set_separator(basic_string_view<charT> sep) noexcept;
      constexpr void set_brackets(basic_string_view<charT> opening,
                                  basic_string_view<charT> closing) noexcept;

      template<class ParseContext>
        constexpr typename ParseContext::iterator
          parse(ParseContext& ctx);

      template<class FormatContext>
-       typename FormatContext::iterator
+       constexpr typename FormatContext::iterator
          format(maybe-const-r& elems, FormatContext& ctx) const;
    };
  }

And the range-default-formatter for maps in 22.14.7.4 [format.range.fmtmap]:

  namespace std {
    template<ranges::input_range R, class charT>
    struct range-default-formatter<range_format::map, R, charT> {
    private:
      using maybe-const-map = fmt-maybe-const<R, charT>;                  // exposition only
      using element-type =                                                // exposition only
        remove_cvref_t<ranges::range_reference_t<maybe-const-map>>;
      range_formatter<element-type, charT> underlying_;                   // exposition only

    public:
      constexpr range-default-formatter();

      template<class ParseContext>
        constexpr typename ParseContext::iterator
          parse(ParseContext& ctx);

      template<class FormatContext>
-       typename FormatContext::iterator
+       constexpr typename FormatContext::iterator
          format(maybe-const-map& r, FormatContext& ctx) const;
    };
  }

And the range-default-formatter for sets in 22.14.7.5 [format.range.fmtset]:

  namespace std {
    template<ranges::input_range R, class charT>
    struct range-default-formatter<range_format::set, R, charT> {
    private:
      using maybe-const-set = fmt-maybe-const<R, charT>;                  // exposition only
      range_formatter<remove_cvref_t<ranges::range_reference_t<maybe-const-set>>,
                      charT> underlying_;                                 // exposition only

    public:
      constexpr range-default-formatter();

      template<class ParseContext>
        constexpr typename ParseContext::iterator
          parse(ParseContext& ctx);

      template<class FormatContext>
-       typename FormatContext::iterator
+       constexpr typename FormatContext::iterator
          format(maybe-const-set& r, FormatContext& ctx) const;
    };
  }

And the range-default-formatter for strings in 22.14.7.6 [format.range.fmtstr]:

  namespace std {
    template<range_format K, ranges::input_range R, class charT>
      requires (K == range_format::string || K == range_format::debug_string)
    struct range-default-formatter<K, R, charT> {
    private:
      formatter<basic_string<charT>, charT> underlying_;                  // exposition only

    public:
      template<class ParseContext>
        constexpr typename ParseContext::iterator
          parse(ParseContext& ctx);

      template<class FormatContext>
-       typename FormatContext::iterator
+       constexpr typename FormatContext::iterator
          format(see below& str, FormatContext& ctx) const;
    };
  }

The basic_format_arg class template can be made entirely constexpr too, in 22.14.8.1 [format.arg]:

  namespace std {
    template<class Context>
    class basic_format_arg {
    public:
      class handle;

    private:
      using char_type = typename Context::char_type;                              // exposition only

      variant<monostate, bool, char_type,
              int, unsigned int, long long int, unsigned long long int,
              float, double, long double,
              const char_type*, basic_string_view<char_type>,
              const void*, handle> value;                                         // exposition only

-     template<class T> explicit basic_format_arg(T& v) noexcept;                 // exposition only
+     template<class T> constexpr explicit basic_format_arg(T& v) noexcept;       // exposition only

    public:
-     basic_format_arg() noexcept;
+     constexpr basic_format_arg() noexcept;

-     explicit operator bool() const noexcept;
+     constexpr explicit operator bool() const noexcept;

      template<class Visitor>
-       decltype(auto) visit(this basic_format_arg arg, Visitor&& vis);
+       constexpr decltype(auto) visit(this basic_format_arg arg, Visitor&& vis);
      template<class R, class Visitor>
-       R visit(this basic_format_arg arg, Visitor&& vis);
+       constexpr R visit(this basic_format_arg arg, Visitor&& vis);
    };
  }

And the handle type introduced in 22.14.8.1 [format.arg]/10:

10 The class handle allows formatting an object of a user-defined type.

  namespace std {
    template<class Context>
    class basic_format_arg<Context>::handle {
      const void* ptr_;                                           // exposition only
      void (*format_)(basic_format_parse_context<char_type>&,
                      Context&, const void*);                     // exposition only
-     template<class T> explicit handle(T& val) noexcept;         // exposition only
+     template<class T> constexpr explicit handle(T& val) noexcept; // exposition only
      friend class basic_format_arg<Context>;                     // exposition only
    public:
-     void format(basic_format_parse_context<char_type>&, Context& ctx) const;
+     constexpr void format(basic_format_parse_context<char_type>&, Context& ctx) const;
    };
  }

And basic_format_args in 22.14.8.3 [format.args]:

  namespace std {
    template<class Context>
    class basic_format_args {
      size_t size_;                               // exposition only
      const basic_format_arg<Context>* data_;     // exposition only

    public:
      template<class... Args>
-       basic_format_args(const format-arg-store<Context, Args...>& store) noexcept;
+       constexpr basic_format_args(const format-arg-store<Context, Args...>& store) noexcept;

-     basic_format_arg<Context> get(size_t i) const noexcept;
+     constexpr basic_format_arg<Context> get(size_t i) const noexcept;
    };

    template<class Context, class... Args>
      basic_format_args(format-arg-store<Context, Args...>) -> basic_format_args<Context>;
  }

And the tuple formatter in 22.14.9 [format.tuple]:

  namespace std {
    template<class charT, formattable<charT>... Ts>
    struct formatter<pair-or-tuple<Ts...>, charT> {
    private:
      tuple<formatter<remove_cvref_t<Ts>, charT>...> underlying_;               // exposition only
      basic_string_view<charT> separator_ = STATICALLY-WIDEN<charT>(", ");      // exposition only
      basic_string_view<charT> opening-bracket_ = STATICALLY-WIDEN<charT>("("); // exposition only
      basic_string_view<charT> closing-bracket_ = STATICALLY-WIDEN<charT>(")"); // exposition only

    public:
      constexpr void set_separator(basic_string_view<charT> sep) noexcept;
      constexpr void set_brackets(basic_string_view<charT> opening,
                                  basic_string_view<charT> closing) noexcept;

      template<class ParseContext>
        constexpr typename ParseContext::iterator
          parse(ParseContext& ctx);

      template<class FormatContext>
-       typename FormatContext::iterator
+       constexpr typename FormatContext::iterator
          format(see below& elems, FormatContext& ctx) const;
    };

    template<class... Ts>
      inline constexpr bool enable_nonlocking_formatter_optimization<pair-or-tuple<Ts...>> =
        (enable_nonlocking_formatter_optimization<Ts> && ...);
  }

Lastly, format_error for now will remain untouched — pending a combination of [P3068R1] and [P3295R0], since runtime_error will also have to be marked.

Mark the vector<bool>::reference formatter constexpr in 24.3.13.2 [vector.bool.fmt]:

  namespace std {
    template<class T, class charT>
      requires is-vector-bool-reference<T>
    struct formatter<T, charT> {
    private:
      formatter<bool, charT> underlying_;       // exposition only

    public:
      template<class ParseContext>
        constexpr typename ParseContext::iterator
          parse(ParseContext& ctx);

      template<class FormatContext>
-       typename FormatContext::iterator
+       constexpr typename FormatContext::iterator
          format(const T& ref, FormatContext& ctx) const;
    };
  }

And the container adaptors in 24.6.13 [container.adaptors.format]:

1 For each of queue, priority_queue, and stack, the library provides the following constexpr-enabled formatter specialization where adaptor-type is the name of the template:

  namespace std {
    template<class charT, class T, formattable<charT> Container, class... U>
    struct formatter<adaptor-type<T, Container, U...>, charT> {
    private:
      using maybe-const-container =                                             // exposition only
        fmt-maybe-const<Container, charT>;
      using maybe-const-adaptor =                                               // exposition only
        maybe-const<is_const_v<maybe-const-container>,                          // see [ranges.syn]
                    adaptor-type<T, Container, U...>>;
      formatter<ranges::ref_view<maybe-const-container>, charT> underlying_;    // exposition only

    public:
      template<class ParseContext>
        constexpr typename ParseContext::iterator
          parse(ParseContext& ctx);

      template<class FormatContext>
-       typename FormatContext::iterator
+       constexpr typename FormatContext::iterator
          format(maybe-const-adaptor& r, FormatContext& ctx) const;
    };
  }

3 References

[P2291R3] Daniil Goncharov, Karaev Alexander. 2021-09-23. Add Constexpr Modifiers to Functions to_chars and from_chars for Integral Types in Header.

https://wg21.link/p2291r3

[P2738R1] Corentin Jabot, David Ledger. 2023-02-13. constexpr cast from void*: towards constexpr type-erasure.

https://wg21.link/p2738r1

[P2741R3] Corentin Jabot. 2023-06-16. user-generated static_assert messages.

https://wg21.link/p2741r3

[P2747R2] Barry Revzin. 2024-03-19. constexpr placement new.

https://wg21.link/p2747r2

[P3068R1] Hana Dusíková. 2024-03-30. Allowing exception throwing in constant-evaluation.

https://wg21.link/p3068r1

[P3295R0] Ben Craig. 2024-05-21. Freestanding constexpr containers and constexpr exception types.

https://wg21.link/p3295r0

`constexpr std::format`

Contents