Merge branch 'json-view/08-view-builder' into json-view/11-view-access

Signed-off-by: Niels Lohmann <mail@nlohmann.me>

README.md

@@ -1402,7 +1402,7 @@ THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR I
 
 - The class contains the UTF-8 Decoder from Bjoern Hoehrmann which is licensed under the [MIT License](https://opensource.org/licenses/MIT) (see above). Copyright &copy; 2008-2009 [Björn Hoehrmann](https://bjoern.hoehrmann.de/) <bjoern@hoehrmann.de>
 - The class contains a slightly modified version of the Grisu2 algorithm from Florian Loitsch which is licensed under the [MIT License](https://opensource.org/licenses/MIT) (see above). Copyright &copy; 2009 [Florian Loitsch](https://florian.loitsch.com/)
-- The class contains a port of the shortest double-to-decimal conversion of [Żmij](https://github.com/vitaut/zmij) by Victor Zverovich, which is licensed under the [MIT License](https://opensource.org/licenses/MIT) (see above). Copyright &copy; 2025 [Victor Zverovich](https://github.com/vitaut)
+- The class contains a port of the shortest double-to-decimal conversion of [Żmij](https://github.com/vitaut/zmij) by Victor Zverovich, including the conversion of the digits to text by Xiang JunBo and the SIMD instruction sequence of Dougall Johnson, which is licensed under the [MIT License](https://opensource.org/licenses/MIT) (see above). Copyright &copy; 2025 [Victor Zverovich](https://github.com/vitaut)
 - The class contains a copy of [Hedley](https://nemequ.github.io/hedley/) from Evan Nemerson which is licensed as [CC0-1.0](https://creativecommons.org/publicdomain/zero/1.0/).
 - The class contains parts of [Google Abseil](https://github.com/abseil/abseil-cpp) which is licensed under the [Apache 2.0 License](https://opensource.org/licenses/Apache-2.0).
 - The class contains an adapted version of the Eisel-Lemire algorithm, its table of powers of five, and its digit comparison for long numbers from [fast_float](https://github.com/fastfloat/fast_float) by Daniel Lemire and contributors, which is available under the [MIT License](https://opensource.org/licenses/MIT) (used here), the Apache 2.0 License, and the Boost Software License. Copyright &copy; 2021 The fast_float authors

docs/mkdocs/docs/api/basic_json/number_float_t.md

@@ -23,10 +23,10 @@ type to use.
 ## Template parameters
 
 `NumberFloatType`
-:   the type to store floating-point numbers. The parser converts `#!cpp float`, `#!cpp double`, and a
-    `#!cpp long double` that is IEEE 754 binary64 itself and other `#!cpp long double` formats with
-    `#!cpp std::from_chars` or `#!cpp std::strtold`, and serialization falls back to `#!cpp std::snprintf`, so the
-    type must be `#!cpp float`, `#!cpp double`, or `#!cpp long double`. The
+:   the type to store floating-point numbers. The type must be `#!cpp float`, `#!cpp double`, or
+    `#!cpp long double`. The parser converts `#!cpp float`, `#!cpp double`, and a `#!cpp long double` that is IEEE 754
+    binary64 itself. It converts other `#!cpp long double` formats with `#!cpp std::from_chars` where available, or
+    with `#!cpp std::strtold` otherwise. Serialization falls back to `#!cpp std::snprintf`. The
     [binary formats](../../features/binary_formats/index.md) additionally require `#!cpp float` or `#!cpp double`,
     because they have no encoding for `#!cpp long double`. See
     [Template Parameter Requirements](../../features/types/template_parameters.md#numberfloattype).

docs/mkdocs/docs/api/basic_json_document/accept.md

@@ -43,6 +43,11 @@ input's own copy (for inputs that are always read into a buffer) throws.
 
 Linear in the length of the input.
 
+## Notes
+
+An integer argument that is not a `#!cpp bool` where the flags are expected, such as `#!cpp accept(ptr, len)`, does not
+compile; see [`parse`](parse.md#notes).
+
 ## Examples
 
 ??? example

docs/mkdocs/docs/api/basic_json_document/parse.md

@@ -19,24 +19,33 @@ static basic_json_document parse(IteratorType first, IteratorType last,
 1. Deserialize from a compatible input, borrowing or owning it depending on its value category and type (see Notes).
 2. Deserialize from a pair of input iterators.
 
-Both overloads accept exactly what [`BasicJsonType::parse()`](../basic_json/parse.md) accepts, with the same
+Both overloads accept the same JSON text as [`BasicJsonType::parse()`](../basic_json/parse.md), with the same
 `ignore_comments`/`ignore_trailing_commas` options, but build a [`basic_json_document`](index.md) (a flat index into
-the input) instead of a tree of `BasicJsonType` values.
+the input) instead of a tree of `BasicJsonType` values. The input must be byte-oriented (see the template parameters
+below): not every input type of `BasicJsonType::parse()` is supported.
 
 ## Template parameters
 
 `InputType`
-:   A compatible input, for instance:
+:   A byte-oriented input, one of:
 
-    - a `#!cpp std::string`, `#!cpp std::string_view`, or a C-style array of characters
-    - a pointer to a null-terminated string of single byte characters
+    - a `#!cpp std::string`, `#!cpp std::string_view`, or a C-style array of single-byte characters
+    - a pointer to a null-terminated string of single-byte characters (`#!cpp char`, `#!cpp signed char`,
+      `#!cpp unsigned char`, `#!cpp std::uint8_t`)
     - a container for which `#!cpp obj.data()` and `#!cpp obj.size()` give contiguous single-byte access, e.g.
       `#!cpp std::vector<char>` or `#!cpp std::vector<std::uint8_t>`
-    - an `#!cpp std::istream` object, or anything else [`BasicJsonType::parse()`](../basic_json/parse.md) accepts
+    - an `#!cpp std::istream` object
+    - a wide string object (`#!cpp std::wstring`, `#!cpp std::u16string`, `#!cpp std::u32string`), which is converted
+      to UTF-8
+
+    Other inputs are not supported: a `#!cpp FILE*`, and pointers to or arrays of wide characters (`#!cpp wchar_t`,
+    `#!cpp char16_t`, `#!cpp char32_t`) are rejected at compile time by a `#!cpp static_assert`. (Use
+    [`BasicJsonType::parse()`](../basic_json/parse.md) for these.)
 
 `IteratorType`
-:   a compatible iterator type, for instance a pair of pointers such as `ptr` and `ptr + len`, or a pair of
-    `#!cpp std::string::iterator`
+:   an input iterator type, for instance a pair of pointers such as `ptr` and `ptr + len`, or a pair of
+    `#!cpp std::string::iterator`; the iterators of single-byte characters are borrowed or read like the byte inputs
+    above, those of wide characters are converted to UTF-8
 
 ## Parameters
 
@@ -70,8 +79,8 @@ discarded; see [`is_discarded`](is_discarded.md).
 Throws the same exception [`BasicJsonType::parse()`](../basic_json/parse.md) throws for the same input and options --
 the same exception id, message, and position -- because on a failing input the library's own parser is run on the
 same bytes to produce the diagnostic. Additionally throws
-[`out_of_range.416`](../../home/exceptions.md#jsonexceptionout_of_range416) if the input is 4 GiB or larger, a size
-[`BasicJsonType::parse()`](../basic_json/parse.md) does not reject.
+[`out_of_range.416`](../../home/exceptions.md#jsonexceptionout_of_range416) if the input is 4294967280 bytes (4 GiB
+minus 16 bytes) or larger, a size [`BasicJsonType::parse()`](../basic_json/parse.md) does not reject.
 
 ## Complexity
 
@@ -84,8 +93,8 @@ Linear in the length of the input.
 | `input`                                                                             | ownership                                                    |
 |--------------------------------------------------------------------------------------|--------------------------------------------------------------|
 | lvalue byte container (`std::string`, `std::vector<char>`, ...), `std::string_view`, C string, character array | **borrowed** -- `input` must outlive the document |
-| rvalue `#!cpp std::string`                                                            | **owned**, moved in without a copy                            |
-| rvalue byte container other than `#!cpp std::string`                                  | **owned**, copied                                             |
+| non-const rvalue `#!cpp std::string`                                                  | **owned**, moved in without a copy                            |
+| other rvalue byte container (including a `#!cpp const` rvalue `#!cpp std::string`)  | **owned**, copied                                             |
 | stream, wide string, or anything else read through the general input adapter          | **owned**, read into a buffer (a stream is read to its end)   |
 
 For overload (2), a pair of pointers to single-byte integers (e.g. `#!cpp const char*`, `#!cpp std::uint8_t*`) is
@@ -100,6 +109,12 @@ See [`owns_source`](owns_source.md) to check which happened after a call, and th
 **Numbers.** As for [`BasicJsonType::parse()`](../basic_json/parse.md), an integer literal too large for the 64-bit
 integer type becomes a floating-point value.
 
+**No lengths.** An integer argument that is not a `#!cpp bool` where the flags are expected -- for example
+`#!cpp parse(ptr, len)` -- does not compile (the overload is deleted). Such a call would convert `len` to
+`allow_exceptions` and read `ptr` as a null-terminated string, past the end of a buffer that has none. To parse a
+buffer of a given length, pass a pair of pointers: `#!cpp parse(ptr, ptr + len)`. The same holds for
+[`parse_copy`](parse_copy.md), [`accept`](accept.md), and [`read`](read.md).
+
 ## Examples
 
 ??? example "Example: (1) borrowed vs. owned input, and errors identical to `BasicJsonType::parse()`"

docs/mkdocs/docs/api/basic_json_document/parse_copy.md

@@ -51,6 +51,9 @@ Linear in the length of the input.
 only differs in that the input is always copied rather than sometimes borrowed. Prefer [`parse()`](parse.md) when the
 input's lifetime already covers the document's, since it avoids the copy for borrowed inputs.
 
+An integer argument that is not a `#!cpp bool` where the flags are expected, such as `#!cpp parse_copy(ptr, len)`, does not
+compile; see [`parse`](parse.md#notes).
+
 ## Examples
 
 ??? example

docs/mkdocs/docs/api/basic_json_document/read.md

@@ -49,6 +49,9 @@ whether or not the new parse succeeds; take fresh views from [`root()`](root.md)
 `input` is borrowed or owned by the same rules as [`parse()`](parse.md#notes); a document can borrow on one call and
 own on the next, since ownership is decided freshly each time.
 
+An integer argument that is not a `#!cpp bool` where the flags are expected, such as `#!cpp read(ptr, len)`, does not
+compile; see [`parse`](parse.md#notes).
+
 ## Examples
 
 ??? example

docs/mkdocs/docs/api/basic_json_document/root.md

@@ -1,10 +1,15 @@
 # <small>nlohmann::basic_json_document::</small>root
 
 ```cpp
-view_type root() const noexcept;
+// (1)
+view_type root() const& noexcept;
+
+// (2)
+view_type root() const&& = delete;
 ```
 
-Returns a view of the root value of the document.
+1. Returns a view of the root value of the document.
+2. Deleted: the view of a temporary document would dangle.
 
 ## Return value
 
@@ -21,6 +26,16 @@ Constant.
 
 ## Notes
 
+**Lifetime.** A view refers into the document, so the document must outlive it. `root()` can therefore only be called
+on a document that has a name (an lvalue); calling it on a temporary does not compile:
+
+```cpp
+auto v = json_document::parse(text).root();  // error: the document is destroyed at the end of the statement
+
+auto doc = json_document::parse(text);       // OK: keep the document alive
+auto v = doc.root();
+```
+
 `root()` is a cheap handle into the document's index, not a copy of anything; call it as often as needed. The
 returned view is valid under the same conditions as any other view of the document -- see
 [Object inspection](../basic_json_view/index.md) -- in particular, it is invalidated by the next

docs/mkdocs/docs/api/basic_json_view/basic_json_view.md

@@ -4,8 +4,8 @@
 basic_json_view() noexcept = default;
 ```
 
-Creates an invalid (discarded) view: [`type()`](type.md) is `#!cpp value_t::discarded`,
-[`is_discarded()`](is_discarded.md) is `#!cpp true`, and `#!cpp explicit operator bool()` is `#!cpp false`.
+Creates an invalid (discarded) view: [`type()`](type.md) is `#!cpp value_t::discarded` and
+[`is_discarded()`](is_discarded.md) is `#!cpp true`.
 
 This is the only constructor a caller can use directly. Every other view is obtained from a
 [`basic_json_document`](../basic_json_document/index.md), via [`root()`](../basic_json_document/root.md) or by
@@ -44,7 +44,6 @@ placeholder for "no value yet" and later be assigned a real view.
 ## See also
 
 - [is_discarded](is_discarded.md) - return whether the view is invalid
-- [operator bool](operator_bool.md) - return whether the view refers to a value
 - [root](../basic_json_document/root.md) - the view of a document's root value
 
 ## Version history

docs/mkdocs/docs/api/basic_json_view/index.md

@@ -69,7 +69,6 @@ comparison.
 - [**is_primitive**](is_primitive.md) - return whether the type is primitive
 - [**is_structured**](is_structured.md) - return whether the type is structured
 - [**is_discarded**](is_discarded.md) - return whether the view is invalid
-- [**operator bool**](operator_bool.md) - return whether the view refers to a value
 
 ### Element access
 

docs/mkdocs/docs/api/basic_json_view/is_discarded.md

@@ -27,8 +27,9 @@ Constant.
 
 ## Notes
 
-`#!cpp v.is_discarded()` and `#!cpp !static_cast<bool>(v)` are equivalent; use whichever reads better at the call
-site.
+A `basic_json_view` is not convertible to `#!cpp bool`: such a conversion would mean "refers to a value", whereas
+`basic_json` converts to the `#!cpp bool` it holds, so the same code would silently behave differently. Test
+`#!cpp !v.is_discarded()` explicitly.
 
 ## Examples
 
@@ -50,7 +51,6 @@ site.
 ## See also
 
 - [operator[]](operator[].md) - access specified element; yields a discarded view where an element is missing
-- [operator bool](operator_bool.md) - return whether the view refers to a value
 - [(constructor)](basic_json_view.md) - the default constructor creates a discarded view
 - [is_discarded (basic_json_document)](../basic_json_document/is_discarded.md) - return whether the last parse failed
 - [`BasicJsonType::is_discarded`](../basic_json/is_discarded.md) - the corresponding function of `basic_json`

docs/mkdocs/docs/api/basic_json_view/operator_bool.md

@@ -1,46 +0,0 @@
-# <small>nlohmann::basic_json_view::</small>operator bool
-
-```cpp
-explicit operator bool() const noexcept;
-```
-
-Returns whether this view refers to a value, i.e. the negation of [`is_discarded()`](is_discarded.md). Being
-`#!cpp explicit`, this conversion is only considered in a boolean context (`#!cpp if (v)`, `#!cpp !v`, `#!cpp v &&
-...`), not for implicit conversions to other types.
-
-## Return value
-
-`#!cpp true` if the view refers to a value, `#!cpp false` if it is [discarded](is_discarded.md).
-
-## Exception safety
-
-No-throw guarantee: this function never throws exceptions.
-
-## Complexity
-
-Constant.
-
-## Examples
-
-??? example
-
-    The example below classifies several parsed documents by the type of their root value, without materializing any
-    of them into a `BasicJsonType` value.
-
-    ```cpp
-    --8<-- "examples/basic_json_view__type_predicates.cpp"
-    ```
-
-    Output:
-
-    ```json
-    --8<-- "examples/basic_json_view__type_predicates.output"
-    ```
-
-## See also
-
-- [is_discarded](is_discarded.md) - return whether the view is invalid
-
-## Version history
-
-- Added in version 3.13.0.

docs/mkdocs/docs/examples/basic_json_view__basic_json_view.cpp

@@ -8,10 +8,10 @@ int main()
     // the default constructor is the only public one: it creates an invalid
     // (discarded) view, useful as a "no value yet" placeholder
     nlohmann::json_view v;
-    std::cout << static_cast<bool>(v) << ' ' << v.is_discarded() << '\n';
+    std::cout << v.is_discarded() << '\n';
 
     // views are trivially copyable handles (two pointers); the document owns
     // the actual data
     nlohmann::json_view copy = v;
-    std::cout << static_cast<bool>(copy) << '\n';
+    std::cout << copy.is_discarded() << '\n';
 }

docs/mkdocs/docs/examples/basic_json_view__basic_json_view.output

@@ -1,2 +1,2 @@
-false true
-false
+true
+true

docs/mkdocs/docs/examples/basic_json_view__type_predicates.cpp

@@ -35,8 +35,8 @@ int main()
     // parse without exceptions, are both discarded
     nlohmann::json_view invalid;
     json_document failed = json_document::parse("not json", /* allow_exceptions */ false);
-    std::cout << static_cast<bool>(invalid) << ' ' << invalid.is_discarded() << '\n';
-    std::cout << static_cast<bool>(failed.root()) << ' ' << failed.root().is_discarded() << '\n';
+    std::cout << invalid.is_discarded() << '\n';
+    std::cout << failed.root().is_discarded() << '\n';
 
     // type() returns the same value_t enumeration as basic_json::type()
     std::cout << (d_object.root().type() == nlohmann::json::value_t::object) << '\n';

docs/mkdocs/docs/examples/basic_json_view__type_predicates.output

@@ -7,6 +7,6 @@ true
 true true
 true false
 false
-false true
-false true
+true
+true
 true

docs/mkdocs/docs/features/json_view.md

@@ -81,6 +81,9 @@ Moving the document itself is fine and does **not** invalidate its views: the in
 that keeps its address across the move. Take a fresh view from [`root()`](../api/basic_json_document/root.md)
 whenever any of the other conditions above was not met.
 
+Because a view dies with its document, [`root()`](../api/basic_json_document/root.md) is not callable on a temporary
+document: `#!cpp auto v = json_document::parse(text).root();` does not compile. Give the document a name first.
+
 ??? example "Example: borrowed and owned documents, and when views become invalid"
 
     ```cpp
@@ -113,7 +116,7 @@ whenever any of the other conditions above was not met.
 
 - **Only 64-bit integers.** `basic_json_document<BasicJsonType>` requires `BasicJsonType::number_integer_t` and
   `number_unsigned_t` to both be 64 bits wide; this is a compile-time `#!cpp static_assert`.
-- **A 4 GiB input limit.** An input of 4 GiB or more throws
+- **A 4 GiB input limit.** An input of 4294967280 bytes (4 GiB minus 16 bytes) or more throws
   [`out_of_range.416`](../home/exceptions.md#jsonexceptionout_of_range416), a limit
   `#!cpp basic_json::parse()` does not have.
 - **A stream is always read to its end.** There is no partial/streaming read of an `#!cpp std::istream`.

docs/mkdocs/docs/features/types/template_parameters.md

@@ -353,16 +353,21 @@ using array_t = ArrayType<basic_json, AllocatorType<basic_json>>;
 ### Always required
 
 - A member type `value_type` that is one byte wide and `char`-compatible. The library stores and processes UTF-8
-  encoded `char` data and passes `data()` to functions that take a `#!cpp const char*`, such as `#!cpp std::strtod`.
+  encoded `char` data and passes `data()` to functions that take a `#!cpp const char*`, such as `#!cpp std::strtold`
+  (only used to parse a `#!cpp long double` that is not IEEE 754 binary64, see
+  [`NumberFloatType`](#numberfloattype)).
   `#!cpp std::wstring`, `#!cpp std::u16string`, and `#!cpp std::u32string` are **not** valid choices; see the FAQ on
   [wide string handling](../../home/faq.md#wide-string-handling).
 - Constructors: default, copy, move, from `#!cpp const char*` (which must not be `#!cpp explicit`), from
   `#!cpp (const char*, size_type)`, and from `#!cpp (size_type, char)`; and copy or move assignment.
 - Member functions `size()`, `clear()`, `resize(n, c)`, `data()`, `push_back(char)`, and `operator[]`
   (const and non-const, returning references). `c_str()` and `back()` are **not** required.
-- `data()` must return a pointer to a contiguous, **null-terminated** buffer -- the parser may hand it to
-  `#!cpp std::strtod`, which reads up to the null character. A type whose `data()` is not null-terminated does not
-  fail to compile; it can silently misparse floating-point numbers.
+- `data()` must return a pointer to a contiguous, **null-terminated** buffer. `#!cpp float`, `#!cpp double`, and a
+  `#!cpp long double` that is IEEE 754 binary64 are converted by the library itself and do not depend on this. For any
+  other `NumberFloatType` (a `#!cpp long double` of another format), the parser falls back to `#!cpp std::strtold` when
+  `#!cpp std::from_chars` is not available or declines the token, and `std::strtold` reads up to the null character. A type whose `data()`
+  is not null-terminated does not fail to compile; with such a `NumberFloatType` it can silently misparse
+  floating-point numbers.
 - `append(const char*, size_type)`, used by [`dump`](../../api/basic_json/dump.md), and `append(const StringType&)`,
   used by the CBOR reader for indefinite-length strings. The library's internal string concatenation additionally has
   to append a `#!cpp char` and a `#!cpp const char*`; for each it selects between `append(arg)`, `#!cpp operator+=`,

docs/mkdocs/docs/home/architecture.md

@@ -210,7 +210,8 @@ packet-beta
 - **Navigation** needs no pointers: the elements of an array or object follow its node, and the node after a value's
   subtree is `next` nodes further for an array or object, and the next node otherwise (`document_data::after`). Views
   step from element to element this way and skip whole subtrees in constant time.
-- **Offsets** are 32 bits wide, so a document is limited to 4 GiB (`out_of_range.416`).
+- **Offsets** are 32 bits wide, so a document is limited to 4294967279 bytes, 4 GiB minus 16 bytes (a margin below
+  2^32 for positions one scanner step past the end of the text; `out_of_range.416`).
 
 For example, `#!json {"a": [1, 2.5]}` becomes five nodes. Each node's elements follow it, and `next` leads from an
 array or object past its subtree:

docs/mkdocs/docs/home/exceptions.md

@@ -1048,12 +1048,12 @@ MessagePack's ext type and BSON's binary subtype are each stored in a single byt
 
 [`basic_json_document::parse()`](../api/basic_json_document/parse.md) and the other parsing functions of
 [`basic_json_document`](../api/basic_json_document/index.md) index a value's position in the source text in 32 bits,
-so they do not support an input of 4 GiB or more.
+so they do not support an input of 4294967280 bytes (4 GiB minus 16 bytes) or more.
 
 !!! failure "Example message"
 
     ```
-    [json.exception.out_of_range.416] input of 4 GiB or more is not supported by json_document
+    [json.exception.out_of_range.416] input of 4294967280 bytes or more is not supported by json_document
     ```
 
 !!! note

docs/mkdocs/docs/home/license.md

@@ -18,7 +18,7 @@ The class contains the UTF-8 Decoder from Bjoern Hoehrmann which is licensed und
 
 The class contains a slightly modified version of the Grisu2 algorithm from Florian Loitsch which is licensed under the [MIT License](https://opensource.org/licenses/MIT) (see above). Copyright © 2009 [Florian Loitsch](https://florian.loitsch.com/)
 
-The class contains a port of the shortest double-to-decimal conversion of [Żmij](https://github.com/vitaut/zmij) by Victor Zverovich, which is licensed under the [MIT License](https://opensource.org/licenses/MIT) (see above). Copyright © 2025 [Victor Zverovich](https://github.com/vitaut)
+The class contains a port of the shortest double-to-decimal conversion of [Żmij](https://github.com/vitaut/zmij) by Victor Zverovich, including the conversion of the digits to text by Xiang JunBo and the SIMD instruction sequence of Dougall Johnson, which is licensed under the [MIT License](https://opensource.org/licenses/MIT) (see above). Copyright © 2025 [Victor Zverovich](https://github.com/vitaut)
 
 The class contains a copy of [Hedley](https://nemequ.github.io/hedley/) from Evan Nemerson which is licensed as [CC0-1.0](https://creativecommons.org/publicdomain/zero/1.0/).
 

docs/mkdocs/mkdocs.yml

@@ -281,7 +281,6 @@ nav:
           - 'items': api/basic_json_view/items.md
           - 'materialize': api/basic_json_view/materialize.md
           - 'number_token': api/basic_json_view/number_token.md
-          - 'operator bool': api/basic_json_view/operator_bool.md
           - 'operator[]': api/basic_json_view/operator[].md
           - 'size': api/basic_json_view/size.md
           - 'source_offset': api/basic_json_view/source_offset.md

include/nlohmann/detail/abi_config.hpp

@@ -15,10 +15,11 @@ namespace detail
 {
 
 /*!
-@brief the configuration macros that change the library's behavior
+@brief the configuration macros that json_view.hpp reads
 
 json.hpp undefines these macros at its end (see macro_unscope.hpp), so code
-that builds on the library after it (json_view.hpp) reads them here. Like the
+that builds on the library after it (json_view.hpp) reads them here. A macro
+is added when the view starts to depend on it. Like the
 macros, they are part of the ABI namespace, so they always match the
 basic_json they are used with.
 */
@@ -26,8 +27,6 @@ struct abi_config
 {
     /// JSON_STRICT_NUL_HANDLING: a null byte is an error, not the end of input
     static constexpr bool strict_nul_handling = JSON_STRICT_NUL_HANDLING != 0;
-    /// JSON_USE_LEGACY_DISCARDED_VALUE_COMPARISON
-    static constexpr bool legacy_discarded_value_comparison = JSON_USE_LEGACY_DISCARDED_VALUE_COMPARISON != 0;
 };
 
 }  // namespace detail

include/nlohmann/detail/bit_ops.hpp

@@ -9,8 +9,9 @@
 #pragma once
 
 #include <cstdint> // uint64_t
-#if !defined(__SIZEOF_INT128__) && defined(_MSC_VER) && (defined(_M_X64) || defined(_M_ARM64))
-    #include <intrin0.h> // __umulh, _umul128
+#include <cstring> // memcpy
+#if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_ARM64)) && (!defined(__SIZEOF_INT128__) || (!defined(__GNUC__) && !defined(__clang__)))
+    #include <intrin0.h> // __umulh, _umul128, _BitScanForward64, _BitScanReverse64
 #endif
 
 #include <nlohmann/detail/macro_scope.hpp> // JSON_HEDLEY_ALWAYS_INLINE, NLOHMANN_JSON_NAMESPACE_BEGIN
@@ -28,6 +29,10 @@ inline int count_leading_zeros(std::uint64_t x) noexcept
 {
 #if defined(__GNUC__) || defined(__clang__)
     return __builtin_clzll(x);
+#elif defined(_MSC_VER) && (defined(_M_X64) || defined(_M_ARM64))
+    unsigned long index = 0;
+    _BitScanReverse64(&index, x);
+    return 63 - static_cast<int>(index);
 #else
     int n = 0;
     for (int shift = 32; shift != 0; shift >>= 1)
@@ -47,6 +52,10 @@ inline int count_trailing_zeros(std::uint64_t x) noexcept
 {
 #if defined(__GNUC__) || defined(__clang__)
     return __builtin_ctzll(x);
+#elif defined(_MSC_VER) && (defined(_M_X64) || defined(_M_ARM64))
+    unsigned long index = 0;
+    _BitScanForward64(&index, x);
+    return static_cast<int>(index);
 #else
     int n = 0;
     for (int shift = 32; shift != 0; shift >>= 1)
@@ -94,15 +103,21 @@ inline uint128_parts full_multiplication(std::uint64_t a, std::uint64_t b) noexc
 #endif
 }
 
-/// eight bytes as a little-endian word (compilers fold this into one load on
-/// little-endian targets; always inlined, as GCC otherwise calls it in the
-/// number loops)
+/// eight bytes as a little-endian word (a single load on little-endian
+/// targets; always inlined, as GCC otherwise calls it in the number loops)
 JSON_HEDLEY_ALWAYS_INLINE std::uint64_t read_eight_bytes(const unsigned char* b) noexcept
 {
+#if defined(_MSC_VER) || defined(__x86_64__) || defined(__i386__) || (defined(__BYTE_ORDER__) && defined(__ORDER_LITTLE_ENDIAN__) && __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
+    // the byte order already matches (all MSVC targets are little-endian)
+    std::uint64_t result = 0;
+    std::memcpy(&result, b, sizeof(result));
+    return result;
+#else
     return static_cast<std::uint64_t>(b[0]) | (static_cast<std::uint64_t>(b[1]) << 8u)
            | (static_cast<std::uint64_t>(b[2]) << 16u) | (static_cast<std::uint64_t>(b[3]) << 24u)
            | (static_cast<std::uint64_t>(b[4]) << 32u) | (static_cast<std::uint64_t>(b[5]) << 40u)
            | (static_cast<std::uint64_t>(b[6]) << 48u) | (static_cast<std::uint64_t>(b[7]) << 56u);
+#endif
 }
 
 /// eight bytes as a little-endian word

include/nlohmann/detail/conversions/to_chars.hpp

@@ -4,6 +4,7 @@
 // |_____|_____|_____|_|___|  https://github.com/nlohmann/json
 //
 // SPDX-FileCopyrightText: 2009 Florian Loitsch <https://florian.loitsch.com/>
+// SPDX-FileCopyrightText: 2025 Victor Zverovich <https://github.com/vitaut/zmij>
 // SPDX-FileCopyrightText: 2013-2026 Niels Lohmann <https://nlohmann.me>
 // SPDX-License-Identifier: MIT
 
@@ -939,88 +940,6 @@ void grisu2(char* buf, int& len, int& decimal_exponent, FloatType value)
     grisu2(buf, len, decimal_exponent, w.minus, w.w, w.plus);
 }
 
-/*!
-@brief the shortest digits of a positive finite float (other than double): Grisu2
-*/
-template<typename FloatType>
-JSON_HEDLEY_NON_NULL(1)
-void shortest_digits(char* buf, int& len, int& decimal_exponent, FloatType value)
-{
-    grisu2(buf, len, decimal_exponent, value);
-}
-
-/*!
-@brief the shortest digits of a positive finite double: the conversion of
-Zmij (see zmij.hpp), which always finds the shortest digits that read back as
-the same value (Grisu2 does not for about one double in a thousand), and the
-closest of them if there are several
-
-v = buf * 10^decimal_exponent, as for grisu2()
-*/
-JSON_HEDLEY_NON_NULL(1)
-inline void shortest_digits(char* buf, int& len, int& decimal_exponent, double value)
-{
-    static_assert(std::numeric_limits<double>::is_iec559 && std::numeric_limits<double>::digits == 53,
-                  "internal error: the conversion of Zmij needs IEEE 754 binary64 doubles");
-    JSON_ASSERT(std::isfinite(value));
-    JSON_ASSERT(value > 0);
-
-    std::uint64_t bits = 0;
-    std::memcpy(&bits, &value, sizeof(bits));
-    zmij::decimal d = zmij::to_decimal(bits);
-    // without trailing zeros (up to 16): 8, 4, 2, 1 at a time
-    while (d.significand % 100000000 == 0)
-    {
-        d.significand /= 100000000;
-        d.exponent += 8;
-    }
-    if (d.significand % 10000 == 0)
-    {
-        d.significand /= 10000;
-        d.exponent += 4;
-    }
-    if (d.significand % 100 == 0)
-    {
-        d.significand /= 100;
-        d.exponent += 2;
-    }
-    if (d.significand % 10 == 0)
-    {
-        d.significand /= 10;
-        d.exponent += 1;
-    }
-    // at most 17 digits, written from the back two at a time
-    static constexpr const char* pairs =
-        "00010203040506070809101112131415161718192021222324252627282930313233343536373839"
-        "40414243444546474849505152535455565758596061626364656667686970717273747576777879"
-        "8081828384858687888990919293949596979899";
-    std::array<char, 20> digits{};
-    std::size_t n = digits.size();
-    while (d.significand >= 100)
-    {
-        const std::uint64_t two_digits = d.significand % 100; // a variable: GCC calls a cast of the remainder useless where std::uint64_t is std::size_t
-        const auto i = static_cast<std::size_t>(two_digits) * 2;
-        d.significand /= 100;
-        n -= 2;
-        digits[n] = pairs[i];
-        digits[n + 1] = pairs[i + 1];
-    }
-    if (d.significand >= 10)
-    {
-        const auto i = static_cast<std::size_t>(d.significand) * 2;
-        n -= 2;
-        digits[n] = pairs[i];
-        digits[n + 1] = pairs[i + 1];
-    }
-    else
-    {
-        digits[--n] = static_cast<char>('0' + d.significand);
-    }
-    len = static_cast<int>(digits.size() - n);
-    std::memcpy(buf, digits.data() + n, static_cast<std::size_t>(len));
-    decimal_exponent = d.exponent;
-}
-
 /*!
 @brief appends a decimal representation of e to buf
 @return a pointer to the element following the exponent.
@@ -1423,53 +1342,30 @@ inline char* write_shortest(char* first, const zmij::shortest_decimal d) noexcep
     return end + (three ? 5 : 4);
 }
 
-/// the powers of ten up to 10^16
-inline const std::array<std::uint64_t, 17>& powers_of_ten_16() noexcept
-{
-    static const std::array<std::uint64_t, 17> powers =
-    {
-        {
-            1u, 10u, 100u, 1000u, 10000u, 100000u, 1000000u, 10000000u, 100000000u, 1000000000u, 10000000000u,
-            100000000000u, 1000000000000u, 10000000000000u, 100000000000000u, 1000000000000000u, 10000000000000000u
-        }
-    };
-    return powers;
-}
-
 /*!
-@brief digits * 10^exp, as write_decimal() writes it, for the digits of a
-double that need no conversion (count digits, at most 15, the first not 0;
-trailing zeros allowed): extended to 16 digits and written by write_shortest()
+@brief whether FloatType is an IEEE 754 binary64 type (a double, or a long double
+that has the same format, as with MSVC and on Apple's Arm CPUs)
 
-@return a pointer past the text; up to 41 bytes at @a first are written
-        (some beyond the returned end)
+These are the types the conversion of Zmij (see zmij.hpp) is used for; all
+others (binary32, or a format the library does not know) use Grisu2.
 */
-JSON_HEDLEY_NON_NULL(1)
-JSON_HEDLEY_RETURNS_NON_NULL
-inline char* write_short_decimal(char* first, std::uint64_t digits, int count, int exp) noexcept
+template<typename FloatType>
+constexpr bool has_binary64_format() noexcept
 {
-    JSON_ASSERT(digits >= powers_of_ten_16()[static_cast<std::size_t>(count - 1)] && count <= 15);
-    const int scale = 16 - count;
-    return write_shortest(first, zmij::shortest_decimal{digits * powers_of_ten_16()[static_cast<std::size_t>(scale)], exp - scale - 1, 0, false});
+    return std::numeric_limits<FloatType>::is_iec559
+           && std::numeric_limits<FloatType>::digits == 53
+           && std::numeric_limits<FloatType>::max_exponent == 1024
+           && sizeof(FloatType) == sizeof(std::uint64_t);
 }
 
-/// as write_short_decimal(), counting the digits (not 0, less than 10^15)
-JSON_HEDLEY_NON_NULL(1)
-JSON_HEDLEY_RETURNS_NON_NULL
-inline char* write_short_decimal(char* first, std::uint64_t digits, int exp) noexcept
-{
-    JSON_ASSERT(digits != 0 && digits < 1000000000000000u);
-    // floor(log10(2^bits)) + 1 digits, or one less
-    const int log2_bound = ((64 - count_leading_zeros(digits)) * 1233) >> 12;
-    const int count = log2_bound + (digits >= powers_of_ten_16()[static_cast<std::size_t>(log2_bound)] ? 1 : 0);
-    return write_short_decimal(first, digits, count, exp);
-}
+template<typename FloatType>
+struct is_binary64 : std::integral_constant<bool, has_binary64_format<FloatType>()> {};
 
-/// a positive finite float (other than double): Grisu2 and format_buffer()
+/// a positive finite float (other than binary64): Grisu2 and format_buffer()
 template<typename FloatType>
 JSON_HEDLEY_NON_NULL(1, 2)
 JSON_HEDLEY_RETURNS_NON_NULL
-char* write_positive(char* first, const char* last, FloatType value)
+char* write_positive_grisu2(char* first, const char* last, FloatType value)
 {
     JSON_ASSERT(last - first >= std::numeric_limits<FloatType>::max_digits10);
     static_cast<void>(last); // (only used in the assertion)
@@ -1480,7 +1376,7 @@ char* write_positive(char* first, const char* last, FloatType value)
     // len is the length of the buffer, i.e., the number of decimal digits.
     int len = 0;
     int decimal_exponent = 0;
-    shortest_digits(first, len, decimal_exponent, value);
+    grisu2(first, len, decimal_exponent, value);
 
     JSON_ASSERT(len <= std::numeric_limits<FloatType>::max_digits10);
 
@@ -1496,15 +1392,16 @@ char* write_positive(char* first, const char* last, FloatType value)
     return format_buffer(first, len, decimal_exponent, kMinExp, kMaxExp);
 }
 
-/// a positive finite double: the shortest digits (Zmij), laid out by
+/// a positive finite binary64 number: the shortest digits (Zmij), laid out by
 /// write_shortest() (through a local buffer if [first, last) is shorter than
 /// the 41 bytes it may write)
+template<typename FloatType>
 JSON_HEDLEY_NON_NULL(1, 2)
 JSON_HEDLEY_RETURNS_NON_NULL
-inline char* write_positive(char* first, const char* last, double value)
+char* write_positive_zmij(char* first, const char* last, FloatType value)
 {
-    static_assert(std::numeric_limits<double>::is_iec559 && std::numeric_limits<double>::digits == 53,
-                  "internal error: the conversion of Zmij needs IEEE 754 binary64 doubles");
+    static_assert(is_binary64<FloatType>::value,
+                  "internal error: the conversion of Zmij needs IEEE 754 binary64 numbers");
     std::uint64_t bits = 0;
     std::memcpy(&bits, &value, sizeof(bits));
     const zmij::shortest_decimal d = zmij::to_shortest(bits);
@@ -1519,6 +1416,34 @@ inline char* write_positive(char* first, const char* last, double value)
     return first + len;
 }
 
+/// a positive finite binary64 number: Zmij (as a long double has the format of
+/// a double here, its bits are those of the double of the same value)
+template<typename FloatType>
+JSON_HEDLEY_NON_NULL(1, 2)
+JSON_HEDLEY_RETURNS_NON_NULL
+char* write_positive(char* first, const char* last, FloatType value, std::true_type /*is_binary64*/)
+{
+    return write_positive_zmij(first, last, value);
+}
+
+/// a positive finite float of any other format: Grisu2
+template<typename FloatType>
+JSON_HEDLEY_NON_NULL(1, 2)
+JSON_HEDLEY_RETURNS_NON_NULL
+char* write_positive(char* first, const char* last, FloatType value, std::false_type /*is_binary64*/)
+{
+    return write_positive_grisu2(first, last, value);
+}
+
+/// a positive finite float: Zmij for binary64 numbers, Grisu2 otherwise
+template<typename FloatType>
+JSON_HEDLEY_NON_NULL(1, 2)
+JSON_HEDLEY_RETURNS_NON_NULL
+char* write_positive(char* first, const char* last, FloatType value)
+{
+    return write_positive(first, last, value, is_binary64<FloatType> {});
+}
+
 }  // namespace dtoa_impl
 
 /*!

include/nlohmann/detail/conversions/zmij.hpp

@@ -36,13 +36,6 @@ computed from the compressed tables of Zmij beyond it.
 namespace zmij
 {
 
-/// significand * 10^exponent
-struct decimal
-{
-    std::uint64_t significand;
-    int exponent;
-};
-
 /// the compressed powers of ten of Zmij
 inline const std::array<std::uint64_t, 28>& pow10_minor() noexcept
 {
@@ -221,18 +214,6 @@ JSON_HEDLEY_ALWAYS_INLINE shortest_decimal to_shortest(std::uint64_t bits) noexc
     return shortest_decimal{integral, dec_exp, static_cast<unsigned char>(digit), !round_up && !round_down};
 }
 
-/// The shortest decimal in the rounding interval of a positive finite double
-/// given by its bits, as one number. The significand can end in zeros.
-inline decimal to_decimal(std::uint64_t bits) noexcept
-{
-    const shortest_decimal d = to_shortest(bits);
-    if (d.has_digit)
-    {
-        return decimal{(d.integral * 10) + d.digit, d.exponent};
-    }
-    return decimal{d.integral, d.exponent + 1};
-}
-
 }  // namespace zmij
 }  // namespace detail
 NLOHMANN_JSON_NAMESPACE_END

include/nlohmann/detail/view/builder.hpp

@@ -9,7 +9,7 @@
 
 #pragma once
 
-#include <algorithm> // find, find_if, max
+#include <algorithm> // find, find_if, max, min
 #include <array> // array
 #include <cstddef> // size_t, ptrdiff_t
 #include <cstdint> // int64_t, uint8_t, uint16_t, uint32_t, uint64_t
@@ -195,8 +195,18 @@ class builder
         const std::uint64_t done = static_cast<std::uint64_t>(at - b) + 1;
         const std::uint64_t guess = static_cast<std::uint64_t>(n) * static_cast<std::uint64_t>(e - b + 1) / done;
         const std::uint64_t grown = guess + (guess / 4) + 64; // a variable: GCC calls a cast of the sum useless where std::uint64_t is std::size_t
+        // (n is below 2^32: the input is smaller than 4 GiB; the sum cannot wrap)
+        const std::uint64_t wanted = (std::max)(grown, static_cast<std::uint64_t>(n) + (n / 2) + 64);
+        const std::uint64_t limit = document_data::max_nodes();
         doc.tape_size = n;
-        doc.reserve((std::max)(static_cast<std::size_t>(grown), n + (n / 2) + 64));
+        // LCOV_EXCL_START (a node array that fills the address space)
+        if (NLOHMANN_VIEW_UNLIKELY(n >= limit))
+        {
+            document_data::throw_bad_alloc(); // no room for another node
+        }
+        // LCOV_EXCL_STOP
+        // (a count beyond the limit is cut: the index does not grow beyond what can be addressed)
+        doc.reserve(static_cast<std::size_t>((std::min)(wanted, limit)));
         return doc.tape;
     }
 
@@ -801,7 +811,10 @@ indent_done:
             n->flags = flags;
             n->extra = extra;
             n->off = static_cast<std::uint32_t>(off);
-            set_integer_bits(*n, second);
+            // len is the low half of the second word, next the high half
+            // (not a native word over both, which swaps them on big-endian)
+            n->len = static_cast<std::uint32_t>(second);
+            n->next = static_cast<std::uint32_t>(second >> 32);
 #endif
             return n;
         }

include/nlohmann/detail/view/document_data.hpp

@@ -11,7 +11,8 @@
 #include <array> // array
 #include <cstddef> // size_t
 #include <cstring> // memcpy
-#include <new> // operator new, placement new
+#include <limits> // numeric_limits
+#include <new> // bad_alloc, operator new, placement new
 #include <string> // string
 
 #include <nlohmann/json.hpp>
@@ -84,13 +85,30 @@ struct document_data
         tape_cap = inline_cap;
     }
 
-    /// make room for n nodes; keeps the first tape_size nodes
+    /// the largest node count whose size in bytes fits a std::size_t
+    static constexpr std::size_t max_nodes() noexcept
+    {
+        return (std::numeric_limits<std::size_t>::max)() / sizeof(node);
+    }
+
+    [[noreturn]] NLOHMANN_VIEW_NOINLINE static void throw_bad_alloc()
+    {
+        NLOHMANN_VIEW_THROW(std::bad_alloc());
+    }
+
+    /// make room for n nodes; keeps the first tape_size nodes (throws
+    /// std::bad_alloc for a count that does not fit the address space,
+    /// instead of wrapping around in n * sizeof(node))
     void reserve(std::size_t n)
     {
         if (n <= tape_cap)
         {
             return;
         }
+        if (NLOHMANN_VIEW_UNLIKELY(n > max_nodes()))
+        {
+            throw_bad_alloc();
+        }
         node* fresh = static_cast<node*>(::operator new (n * sizeof(node)));
         if (tape_size != 0)
         {

include/nlohmann/detail/view/errors.hpp

@@ -61,9 +61,8 @@ template<typename BasicJsonType>
 {
     if (f.code == error_code::input_too_large)
     {
-        // LCOV_EXCL_START (4 GiB)
-        NLOHMANN_VIEW_THROW(out_of_range::create(416, "input of 4 GiB or more is not supported by json_document", nullptr));
-        // LCOV_EXCL_STOP
+        // (the limit is detail::view::max_input_size: 4 GiB minus 16 bytes)
+        NLOHMANN_VIEW_THROW(out_of_range::create(416, "input of 4294967280 bytes or more is not supported by json_document", nullptr));
     }
     const BasicJsonType accepted = BasicJsonType::parse(src, src + size, nullptr, true, ignore_comments, ignore_trailing_commas);
     // LCOV_EXCL_START (only if parse() accepts what the view rejects: a bug)

include/nlohmann/detail/view/input.hpp

@@ -9,7 +9,7 @@
 #pragma once
 
 #include <string> // basic_string, char_traits, string
-#include <type_traits> // decay, integral_constant, is_array, is_lvalue_reference, is_pointer, is_same, remove_reference
+#include <type_traits> // decay, integral_constant, is_array, is_const, is_integral, is_lvalue_reference, is_pointer, is_same, remove_reference
 #include <utility> // forward
 
 #include <nlohmann/json.hpp>
@@ -28,12 +28,12 @@ namespace view
 /// how a document takes its input
 enum class input_kind
 {
-    move_string,  ///< rvalue std::string: owned without a copy
+    move_string,  ///< non-const rvalue std::string: owned without a copy
     c_string,     ///< const char* (NUL-terminated): borrowed
     char_array,   ///< char array (e.g. a string literal): borrowed
     borrow_range, ///< lvalue contiguous byte container, or std::string_view: borrowed
-    copy_range,   ///< rvalue contiguous byte container: copied
-    adapter,      ///< anything else parse() accepts (streams, wide strings, ...): read into a buffer
+    copy_range,   ///< rvalue contiguous byte container (a const rvalue std::string too): copied
+    adapter,      ///< streams, wide strings, and the rest of what the library's input adapter reads: read into a buffer
 };
 
 template<typename InputType>
@@ -52,13 +52,20 @@ struct classify_input
     static constexpr input_kind value =
         std::is_array<R>::value ? input_kind::char_array
         : std::is_pointer<D>::value ? input_kind::c_string
-        : (is_rvalue && std::is_same<D, std::string>::value) ? input_kind::move_string
+        : (is_rvalue && !std::is_const<R>::value && std::is_same<D, std::string>::value) ? input_kind::move_string
         : (is_bytes && (!is_rvalue || is_string_view)) ? input_kind::borrow_range
         : is_bytes ? input_kind::copy_range
         : input_kind::adapter;
     // NOLINTEND(readability-avoid-nested-conditional-operator)
 };
 
+/// an integer type other than bool: a length passed where a flag is expected
+template<typename T>
+struct is_integer_not_bool : std::is_integral<T> {};
+
+template<>
+struct is_integer_not_bool<bool> : std::false_type {};
+
 /// std::basic_string guarantees a NUL at data()[size()] (the parser's sentinel)
 template<typename T>
 struct is_std_string : std::false_type {};

include/nlohmann/detail/view/node.hpp

@@ -29,6 +29,11 @@ static_assert(static_cast<std::uint8_t>(value_t::null) == 0 && static_cast<std::
               && static_cast<std::uint8_t>(value_t::number_unsigned) == 6 && static_cast<std::uint8_t>(value_t::number_float) == 7,
               "the node format depends on the numbering of value_t");
 
+/// The largest input a document accepts, in bytes. Offsets and node counts are
+/// 32 bits wide; the limit keeps 16 bytes (the width of the scanner's steps)
+/// below 2^32, so that a position one step past the end of the text fits.
+static constexpr std::size_t max_input_size = 0xFFFFFFEFu;
+
 /// node flags
 struct node_flags
 {
@@ -57,17 +62,27 @@ NLOHMANN_VIEW_ALWAYS_INLINE bool is_container(const node& n) noexcept
     return static_cast<unsigned>(n.kind) - 1u <= 1u;
 }
 
-/// the converted value of an integer node (stored in len/next)
+/// the converted value of an integer node: len is its low half, next its high
+/// half (on little-endian targets the two words are the value in memory)
 NLOHMANN_VIEW_ALWAYS_INLINE std::uint64_t integer_bits(const node& n) noexcept
 {
+#if NLOHMANN_VIEW_LITTLE_ENDIAN
     std::uint64_t v = 0;
     std::memcpy(&v, reinterpret_cast<const unsigned char*>(&n) + 8, 8); // NOLINT(cppcoreguidelines-pro-type-reinterpret-cast)
     return v;
+#else
+    return static_cast<std::uint64_t>(n.len) | (static_cast<std::uint64_t>(n.next) << 32);
+#endif
 }
 
 NLOHMANN_VIEW_ALWAYS_INLINE void set_integer_bits(node& n, std::uint64_t v) noexcept
 {
+#if NLOHMANN_VIEW_LITTLE_ENDIAN
     std::memcpy(reinterpret_cast<unsigned char*>(&n) + 8, &v, 8); // NOLINT(cppcoreguidelines-pro-type-reinterpret-cast)
+#else
+    n.len = static_cast<std::uint32_t>(v);
+    n.next = static_cast<std::uint32_t>(v >> 32);
+#endif
 }
 
 /// token length of a number node

include/nlohmann/json_view.hpp

@@ -16,7 +16,7 @@
  * the read-only part of the basic_json interface; materialize() turns a     *
  * subtree into the basic_json value that parse() would produce.             *
  *                                                                           *
- * The source text must outlive a document that borrows it (lvalue byte     *
+ * The source text must outlive a document that borrows it (lvalue byte      *
  * containers, C strings); rvalue strings, streams, and other inputs are     *
  * owned by the document.                                                    *
 \****************************************************************************/
@@ -169,12 +169,6 @@ class basic_json_view
         return type() == value_t::discarded;
     }
 
-    /// false for discarded views
-    explicit operator bool() const noexcept
-    {
-        return m_node != nullptr;
-    }
-
     /// the name of the type, as basic_json::type_name()
     const char* type_name() const noexcept
     {
@@ -733,6 +727,13 @@ class basic_json_document
         return d;
     }
 
+    /// parse(ptr, len) does not compile: len would convert to allow_exceptions
+    /// and ptr be read as a C string (as for the overloads of parse_copy,
+    /// accept, and read below)
+    template<typename InputType, typename IntegerType, typename... Flags>
+    static typename std::enable_if<detail::view::is_integer_not_bool<IntegerType>::value, basic_json_document>::type
+    parse(InputType&& input, IntegerType value, Flags&&... flags) = delete;
+
     /// parse [first, last)
     template<typename IteratorType, typename std::enable_if<
                  std::is_base_of<std::input_iterator_tag, typename std::iterator_traits<IteratorType>::iterator_category>::value, int>::type = 0>
@@ -760,6 +761,10 @@ class basic_json_document
         return d;
     }
 
+    template<typename InputType, typename IntegerType, typename... Flags>
+    static typename std::enable_if<detail::view::is_integer_not_bool<IntegerType>::value, basic_json_document>::type
+    parse_copy(InputType&& input, IntegerType value, Flags&&... flags) = delete;
+
     /// check whether the input is valid JSON (the result of basic_json::accept)
     template<typename InputType>
     static bool accept(InputType&& input, const bool ignore_comments = false, const bool ignore_trailing_commas = false)
@@ -769,6 +774,10 @@ class basic_json_document
         return !d.is_discarded();
     }
 
+    template<typename InputType, typename IntegerType, typename... Flags>
+    static typename std::enable_if<detail::view::is_integer_not_bool<IntegerType>::value, bool>::type
+    accept(InputType&& input, IntegerType value, Flags&&... flags) = delete;
+
     /// parse into this document, reusing its memory
     template<typename InputType>
     // flawfinder: ignore (a member function, not POSIX read())
@@ -781,12 +790,17 @@ class basic_json_document
                   std::integral_constant<detail::view::input_kind, detail::view::classify_input<InputType>::value> {});
     }
 
+    template<typename InputType, typename IntegerType, typename... Flags>
+    // flawfinder: ignore (a member function, not POSIX read())
+    typename std::enable_if<detail::view::is_integer_not_bool<IntegerType>::value, void>::type
+    read(InputType&& input, IntegerType value, Flags&&... flags) = delete;
+
     ////////////
     // access //
     ////////////
 
     /// the root value (discarded if parsing failed without exceptions)
-    view_type root() const noexcept
+    view_type root() const& noexcept
     {
         if (!m_data || m_data->discarded)
         {
@@ -795,6 +809,9 @@ class basic_json_document
         return view_type(m_data.get(), m_data->tape);
     }
 
+    /// deleted: the view of a temporary document would dangle
+    view_type root() const&& = delete;
+
     bool is_discarded() const noexcept
     {
         return !m_data || m_data->discarded;
@@ -900,9 +917,9 @@ class basic_json_document
         d.discarded = true;
         detail::view::parse_failure failure;
         bool ok = false;
-        if (NLOHMANN_VIEW_UNLIKELY(size >= 0xFFFFFFF0u))
+        if (NLOHMANN_VIEW_UNLIKELY(size > detail::view::max_input_size))
         {
-            failure.code = detail::view::error_code::input_too_large; // LCOV_EXCL_LINE (4 GiB)
+            failure.code = detail::view::error_code::input_too_large;
         }
         else
         {

single_include/nlohmann/json.hpp

@@ -7544,10 +7544,11 @@ namespace detail
 {
 
 /*!
-@brief the configuration macros that change the library's behavior
+@brief the configuration macros that json_view.hpp reads
 
 json.hpp undefines these macros at its end (see macro_unscope.hpp), so code
-that builds on the library after it (json_view.hpp) reads them here. Like the
+that builds on the library after it (json_view.hpp) reads them here. A macro
+is added when the view starts to depend on it. Like the
 macros, they are part of the ABI namespace, so they always match the
 basic_json they are used with.
 */
@@ -7555,8 +7556,6 @@ struct abi_config
 {
     /// JSON_STRICT_NUL_HANDLING: a null byte is an error, not the end of input
     static constexpr bool strict_nul_handling = JSON_STRICT_NUL_HANDLING != 0;
-    /// JSON_USE_LEGACY_DISCARDED_VALUE_COMPARISON
-    static constexpr bool legacy_discarded_value_comparison = JSON_USE_LEGACY_DISCARDED_VALUE_COMPARISON != 0;
 };
 
 }  // namespace detail
@@ -8863,8 +8862,9 @@ NLOHMANN_JSON_NAMESPACE_END
 
 
 #include <cstdint> // uint64_t
-#if !defined(__SIZEOF_INT128__) && defined(_MSC_VER) && (defined(_M_X64) || defined(_M_ARM64))
-    #include <intrin0.h> // __umulh, _umul128
+#include <cstring> // memcpy
+#if defined(_MSC_VER) && (defined(_M_X64) || defined(_M_ARM64)) && (!defined(__SIZEOF_INT128__) || (!defined(__GNUC__) && !defined(__clang__)))
+    #include <intrin0.h> // __umulh, _umul128, _BitScanForward64, _BitScanReverse64
 #endif
 
 // #include <nlohmann/detail/macro_scope.hpp>
@@ -8883,6 +8883,10 @@ inline int count_leading_zeros(std::uint64_t x) noexcept
 {
 #if defined(__GNUC__) || defined(__clang__)
     return __builtin_clzll(x);
+#elif defined(_MSC_VER) && (defined(_M_X64) || defined(_M_ARM64))
+    unsigned long index = 0;
+    _BitScanReverse64(&index, x);
+    return 63 - static_cast<int>(index);
 #else
     int n = 0;
     for (int shift = 32; shift != 0; shift >>= 1)
@@ -8902,6 +8906,10 @@ inline int count_trailing_zeros(std::uint64_t x) noexcept
 {
 #if defined(__GNUC__) || defined(__clang__)
     return __builtin_ctzll(x);
+#elif defined(_MSC_VER) && (defined(_M_X64) || defined(_M_ARM64))
+    unsigned long index = 0;
+    _BitScanForward64(&index, x);
+    return static_cast<int>(index);
 #else
     int n = 0;
     for (int shift = 32; shift != 0; shift >>= 1)
@@ -8949,15 +8957,21 @@ inline uint128_parts full_multiplication(std::uint64_t a, std::uint64_t b) noexc
 #endif
 }
 
-/// eight bytes as a little-endian word (compilers fold this into one load on
-/// little-endian targets; always inlined, as GCC otherwise calls it in the
-/// number loops)
+/// eight bytes as a little-endian word (a single load on little-endian
+/// targets; always inlined, as GCC otherwise calls it in the number loops)
 JSON_HEDLEY_ALWAYS_INLINE std::uint64_t read_eight_bytes(const unsigned char* b) noexcept
 {
+#if defined(_MSC_VER) || defined(__x86_64__) || defined(__i386__) || (defined(__BYTE_ORDER__) && defined(__ORDER_LITTLE_ENDIAN__) && __BYTE_ORDER__ == __ORDER_LITTLE_ENDIAN__)
+    // the byte order already matches (all MSVC targets are little-endian)
+    std::uint64_t result = 0;
+    std::memcpy(&result, b, sizeof(result));
+    return result;
+#else
     return static_cast<std::uint64_t>(b[0]) | (static_cast<std::uint64_t>(b[1]) << 8u)
            | (static_cast<std::uint64_t>(b[2]) << 16u) | (static_cast<std::uint64_t>(b[3]) << 24u)
            | (static_cast<std::uint64_t>(b[4]) << 32u) | (static_cast<std::uint64_t>(b[5]) << 40u)
            | (static_cast<std::uint64_t>(b[6]) << 48u) | (static_cast<std::uint64_t>(b[7]) << 56u);
+#endif
 }
 
 /// eight bytes as a little-endian word
@@ -25172,6 +25186,7 @@ NLOHMANN_JSON_NAMESPACE_END
 // |_____|_____|_____|_|___|  https://github.com/nlohmann/json
 //
 // SPDX-FileCopyrightText: 2009 Florian Loitsch <https://florian.loitsch.com/>
+// SPDX-FileCopyrightText: 2025 Victor Zverovich <https://github.com/vitaut/zmij>
 // SPDX-FileCopyrightText: 2013-2026 Niels Lohmann <https://nlohmann.me>
 // SPDX-License-Identifier: MIT
 
@@ -25247,13 +25262,6 @@ computed from the compressed tables of Zmij beyond it.
 namespace zmij
 {
 
-/// significand * 10^exponent
-struct decimal
-{
-    std::uint64_t significand;
-    int exponent;
-};
-
 /// the compressed powers of ten of Zmij
 inline const std::array<std::uint64_t, 28>& pow10_minor() noexcept
 {
@@ -25432,18 +25440,6 @@ JSON_HEDLEY_ALWAYS_INLINE shortest_decimal to_shortest(std::uint64_t bits) noexc
     return shortest_decimal{integral, dec_exp, static_cast<unsigned char>(digit), !round_up && !round_down};
 }
 
-/// The shortest decimal in the rounding interval of a positive finite double
-/// given by its bits, as one number. The significand can end in zeros.
-inline decimal to_decimal(std::uint64_t bits) noexcept
-{
-    const shortest_decimal d = to_shortest(bits);
-    if (d.has_digit)
-    {
-        return decimal{(d.integral * 10) + d.digit, d.exponent};
-    }
-    return decimal{d.integral, d.exponent + 1};
-}
-
 }  // namespace zmij
 }  // namespace detail
 NLOHMANN_JSON_NAMESPACE_END
@@ -26351,88 +26347,6 @@ void grisu2(char* buf, int& len, int& decimal_exponent, FloatType value)
     grisu2(buf, len, decimal_exponent, w.minus, w.w, w.plus);
 }
 
-/*!
-@brief the shortest digits of a positive finite float (other than double): Grisu2
-*/
-template<typename FloatType>
-JSON_HEDLEY_NON_NULL(1)
-void shortest_digits(char* buf, int& len, int& decimal_exponent, FloatType value)
-{
-    grisu2(buf, len, decimal_exponent, value);
-}
-
-/*!
-@brief the shortest digits of a positive finite double: the conversion of
-Zmij (see zmij.hpp), which always finds the shortest digits that read back as
-the same value (Grisu2 does not for about one double in a thousand), and the
-closest of them if there are several
-
-v = buf * 10^decimal_exponent, as for grisu2()
-*/
-JSON_HEDLEY_NON_NULL(1)
-inline void shortest_digits(char* buf, int& len, int& decimal_exponent, double value)
-{
-    static_assert(std::numeric_limits<double>::is_iec559 && std::numeric_limits<double>::digits == 53,
-                  "internal error: the conversion of Zmij needs IEEE 754 binary64 doubles");
-    JSON_ASSERT(std::isfinite(value));
-    JSON_ASSERT(value > 0);
-
-    std::uint64_t bits = 0;
-    std::memcpy(&bits, &value, sizeof(bits));
-    zmij::decimal d = zmij::to_decimal(bits);
-    // without trailing zeros (up to 16): 8, 4, 2, 1 at a time
-    while (d.significand % 100000000 == 0)
-    {
-        d.significand /= 100000000;
-        d.exponent += 8;
-    }
-    if (d.significand % 10000 == 0)
-    {
-        d.significand /= 10000;
-        d.exponent += 4;
-    }
-    if (d.significand % 100 == 0)
-    {
-        d.significand /= 100;
-        d.exponent += 2;
-    }
-    if (d.significand % 10 == 0)
-    {
-        d.significand /= 10;
-        d.exponent += 1;
-    }
-    // at most 17 digits, written from the back two at a time
-    static constexpr const char* pairs =
-        "00010203040506070809101112131415161718192021222324252627282930313233343536373839"
-        "40414243444546474849505152535455565758596061626364656667686970717273747576777879"
-        "8081828384858687888990919293949596979899";
-    std::array<char, 20> digits{};
-    std::size_t n = digits.size();
-    while (d.significand >= 100)
-    {
-        const std::uint64_t two_digits = d.significand % 100; // a variable: GCC calls a cast of the remainder useless where std::uint64_t is std::size_t
-        const auto i = static_cast<std::size_t>(two_digits) * 2;
-        d.significand /= 100;
-        n -= 2;
-        digits[n] = pairs[i];
-        digits[n + 1] = pairs[i + 1];
-    }
-    if (d.significand >= 10)
-    {
-        const auto i = static_cast<std::size_t>(d.significand) * 2;
-        n -= 2;
-        digits[n] = pairs[i];
-        digits[n + 1] = pairs[i + 1];
-    }
-    else
-    {
-        digits[--n] = static_cast<char>('0' + d.significand);
-    }
-    len = static_cast<int>(digits.size() - n);
-    std::memcpy(buf, digits.data() + n, static_cast<std::size_t>(len));
-    decimal_exponent = d.exponent;
-}
-
 /*!
 @brief appends a decimal representation of e to buf
 @return a pointer to the element following the exponent.
@@ -26835,53 +26749,30 @@ inline char* write_shortest(char* first, const zmij::shortest_decimal d) noexcep
     return end + (three ? 5 : 4);
 }
 
-/// the powers of ten up to 10^16
-inline const std::array<std::uint64_t, 17>& powers_of_ten_16() noexcept
-{
-    static const std::array<std::uint64_t, 17> powers =
-    {
-        {
-            1u, 10u, 100u, 1000u, 10000u, 100000u, 1000000u, 10000000u, 100000000u, 1000000000u, 10000000000u,
-            100000000000u, 1000000000000u, 10000000000000u, 100000000000000u, 1000000000000000u, 10000000000000000u
-        }
-    };
-    return powers;
-}
-
 /*!
-@brief digits * 10^exp, as write_decimal() writes it, for the digits of a
-double that need no conversion (count digits, at most 15, the first not 0;
-trailing zeros allowed): extended to 16 digits and written by write_shortest()
+@brief whether FloatType is an IEEE 754 binary64 type (a double, or a long double
+that has the same format, as with MSVC and on Apple's Arm CPUs)
 
-@return a pointer past the text; up to 41 bytes at @a first are written
-        (some beyond the returned end)
+These are the types the conversion of Zmij (see zmij.hpp) is used for; all
+others (binary32, or a format the library does not know) use Grisu2.
 */
-JSON_HEDLEY_NON_NULL(1)
-JSON_HEDLEY_RETURNS_NON_NULL
-inline char* write_short_decimal(char* first, std::uint64_t digits, int count, int exp) noexcept
+template<typename FloatType>
+constexpr bool has_binary64_format() noexcept
 {
-    JSON_ASSERT(digits >= powers_of_ten_16()[static_cast<std::size_t>(count - 1)] && count <= 15);
-    const int scale = 16 - count;
-    return write_shortest(first, zmij::shortest_decimal{digits * powers_of_ten_16()[static_cast<std::size_t>(scale)], exp - scale - 1, 0, false});
+    return std::numeric_limits<FloatType>::is_iec559
+           && std::numeric_limits<FloatType>::digits == 53
+           && std::numeric_limits<FloatType>::max_exponent == 1024
+           && sizeof(FloatType) == sizeof(std::uint64_t);
 }
 
-/// as write_short_decimal(), counting the digits (not 0, less than 10^15)
-JSON_HEDLEY_NON_NULL(1)
-JSON_HEDLEY_RETURNS_NON_NULL
-inline char* write_short_decimal(char* first, std::uint64_t digits, int exp) noexcept
-{
-    JSON_ASSERT(digits != 0 && digits < 1000000000000000u);
-    // floor(log10(2^bits)) + 1 digits, or one less
-    const int log2_bound = ((64 - count_leading_zeros(digits)) * 1233) >> 12;
-    const int count = log2_bound + (digits >= powers_of_ten_16()[static_cast<std::size_t>(log2_bound)] ? 1 : 0);
-    return write_short_decimal(first, digits, count, exp);
-}
+template<typename FloatType>
+struct is_binary64 : std::integral_constant<bool, has_binary64_format<FloatType>()> {};
 
-/// a positive finite float (other than double): Grisu2 and format_buffer()
+/// a positive finite float (other than binary64): Grisu2 and format_buffer()
 template<typename FloatType>
 JSON_HEDLEY_NON_NULL(1, 2)
 JSON_HEDLEY_RETURNS_NON_NULL
-char* write_positive(char* first, const char* last, FloatType value)
+char* write_positive_grisu2(char* first, const char* last, FloatType value)
 {
     JSON_ASSERT(last - first >= std::numeric_limits<FloatType>::max_digits10);
     static_cast<void>(last); // (only used in the assertion)
@@ -26892,7 +26783,7 @@ char* write_positive(char* first, const char* last, FloatType value)
     // len is the length of the buffer, i.e., the number of decimal digits.
     int len = 0;
     int decimal_exponent = 0;
-    shortest_digits(first, len, decimal_exponent, value);
+    grisu2(first, len, decimal_exponent, value);
 
     JSON_ASSERT(len <= std::numeric_limits<FloatType>::max_digits10);
 
@@ -26908,15 +26799,16 @@ char* write_positive(char* first, const char* last, FloatType value)
     return format_buffer(first, len, decimal_exponent, kMinExp, kMaxExp);
 }
 
-/// a positive finite double: the shortest digits (Zmij), laid out by
+/// a positive finite binary64 number: the shortest digits (Zmij), laid out by
 /// write_shortest() (through a local buffer if [first, last) is shorter than
 /// the 41 bytes it may write)
+template<typename FloatType>
 JSON_HEDLEY_NON_NULL(1, 2)
 JSON_HEDLEY_RETURNS_NON_NULL
-inline char* write_positive(char* first, const char* last, double value)
+char* write_positive_zmij(char* first, const char* last, FloatType value)
 {
-    static_assert(std::numeric_limits<double>::is_iec559 && std::numeric_limits<double>::digits == 53,
-                  "internal error: the conversion of Zmij needs IEEE 754 binary64 doubles");
+    static_assert(is_binary64<FloatType>::value,
+                  "internal error: the conversion of Zmij needs IEEE 754 binary64 numbers");
     std::uint64_t bits = 0;
     std::memcpy(&bits, &value, sizeof(bits));
     const zmij::shortest_decimal d = zmij::to_shortest(bits);
@@ -26931,6 +26823,34 @@ inline char* write_positive(char* first, const char* last, double value)
     return first + len;
 }
 
+/// a positive finite binary64 number: Zmij (as a long double has the format of
+/// a double here, its bits are those of the double of the same value)
+template<typename FloatType>
+JSON_HEDLEY_NON_NULL(1, 2)
+JSON_HEDLEY_RETURNS_NON_NULL
+char* write_positive(char* first, const char* last, FloatType value, std::true_type /*is_binary64*/)
+{
+    return write_positive_zmij(first, last, value);
+}
+
+/// a positive finite float of any other format: Grisu2
+template<typename FloatType>
+JSON_HEDLEY_NON_NULL(1, 2)
+JSON_HEDLEY_RETURNS_NON_NULL
+char* write_positive(char* first, const char* last, FloatType value, std::false_type /*is_binary64*/)
+{
+    return write_positive_grisu2(first, last, value);
+}
+
+/// a positive finite float: Zmij for binary64 numbers, Grisu2 otherwise
+template<typename FloatType>
+JSON_HEDLEY_NON_NULL(1, 2)
+JSON_HEDLEY_RETURNS_NON_NULL
+char* write_positive(char* first, const char* last, FloatType value)
+{
+    return write_positive(first, last, value, is_binary64<FloatType> {});
+}
+
 }  // namespace dtoa_impl
 
 /*!

single_include/nlohmann/json_view.hpp

@@ -16,7 +16,7 @@
  * the read-only part of the basic_json interface; materialize() turns a     *
  * subtree into the basic_json value that parse() would produce.             *
  *                                                                           *
- * The source text must outlive a document that borrows it (lvalue byte     *
+ * The source text must outlive a document that borrows it (lvalue byte      *
  * containers, C strings); rvalue strings, streams, and other inputs are     *
  * owned by the document.                                                    *
 \****************************************************************************/
@@ -55,7 +55,7 @@
 
 
 
-#include <algorithm> // find, find_if, max
+#include <algorithm> // find, find_if, max, min
 #include <array> // array
 #include <cstddef> // size_t, ptrdiff_t
 #include <cstdint> // int64_t, uint8_t, uint16_t, uint32_t, uint64_t
@@ -79,7 +79,8 @@
 #include <array> // array
 #include <cstddef> // size_t
 #include <cstring> // memcpy
-#include <new> // operator new, placement new
+#include <limits> // numeric_limits
+#include <new> // bad_alloc, operator new, placement new
 #include <string> // string
 
 // #include <nlohmann/json.hpp>
@@ -189,6 +190,11 @@ static_assert(static_cast<std::uint8_t>(value_t::null) == 0 && static_cast<std::
               && static_cast<std::uint8_t>(value_t::number_unsigned) == 6 && static_cast<std::uint8_t>(value_t::number_float) == 7,
               "the node format depends on the numbering of value_t");
 
+/// The largest input a document accepts, in bytes. Offsets and node counts are
+/// 32 bits wide; the limit keeps 16 bytes (the width of the scanner's steps)
+/// below 2^32, so that a position one step past the end of the text fits.
+static constexpr std::size_t max_input_size = 0xFFFFFFEFu;
+
 /// node flags
 struct node_flags
 {
@@ -217,17 +223,27 @@ NLOHMANN_VIEW_ALWAYS_INLINE bool is_container(const node& n) noexcept
     return static_cast<unsigned>(n.kind) - 1u <= 1u;
 }
 
-/// the converted value of an integer node (stored in len/next)
+/// the converted value of an integer node: len is its low half, next its high
+/// half (on little-endian targets the two words are the value in memory)
 NLOHMANN_VIEW_ALWAYS_INLINE std::uint64_t integer_bits(const node& n) noexcept
 {
+#if NLOHMANN_VIEW_LITTLE_ENDIAN
     std::uint64_t v = 0;
     std::memcpy(&v, reinterpret_cast<const unsigned char*>(&n) + 8, 8); // NOLINT(cppcoreguidelines-pro-type-reinterpret-cast)
     return v;
+#else
+    return static_cast<std::uint64_t>(n.len) | (static_cast<std::uint64_t>(n.next) << 32);
+#endif
 }
 
 NLOHMANN_VIEW_ALWAYS_INLINE void set_integer_bits(node& n, std::uint64_t v) noexcept
 {
+#if NLOHMANN_VIEW_LITTLE_ENDIAN
     std::memcpy(reinterpret_cast<unsigned char*>(&n) + 8, &v, 8); // NOLINT(cppcoreguidelines-pro-type-reinterpret-cast)
+#else
+    n.len = static_cast<std::uint32_t>(v);
+    n.next = static_cast<std::uint32_t>(v >> 32);
+#endif
 }
 
 /// token length of a number node
@@ -323,13 +339,30 @@ struct document_data
         tape_cap = inline_cap;
     }
 
-    /// make room for n nodes; keeps the first tape_size nodes
+    /// the largest node count whose size in bytes fits a std::size_t
+    static constexpr std::size_t max_nodes() noexcept
+    {
+        return (std::numeric_limits<std::size_t>::max)() / sizeof(node);
+    }
+
+    [[noreturn]] NLOHMANN_VIEW_NOINLINE static void throw_bad_alloc()
+    {
+        NLOHMANN_VIEW_THROW(std::bad_alloc());
+    }
+
+    /// make room for n nodes; keeps the first tape_size nodes (throws
+    /// std::bad_alloc for a count that does not fit the address space,
+    /// instead of wrapping around in n * sizeof(node))
     void reserve(std::size_t n)
     {
         if (n <= tape_cap)
         {
             return;
         }
+        if (NLOHMANN_VIEW_UNLIKELY(n > max_nodes()))
+        {
+            throw_bad_alloc();
+        }
         node* fresh = static_cast<node*>(::operator new (n * sizeof(node)));
         if (tape_size != 0)
         {
@@ -736,8 +769,18 @@ class builder
         const std::uint64_t done = static_cast<std::uint64_t>(at - b) + 1;
         const std::uint64_t guess = static_cast<std::uint64_t>(n) * static_cast<std::uint64_t>(e - b + 1) / done;
         const std::uint64_t grown = guess + (guess / 4) + 64; // a variable: GCC calls a cast of the sum useless where std::uint64_t is std::size_t
+        // (n is below 2^32: the input is smaller than 4 GiB; the sum cannot wrap)
+        const std::uint64_t wanted = (std::max)(grown, static_cast<std::uint64_t>(n) + (n / 2) + 64);
+        const std::uint64_t limit = document_data::max_nodes();
         doc.tape_size = n;
-        doc.reserve((std::max)(static_cast<std::size_t>(grown), n + (n / 2) + 64));
+        // LCOV_EXCL_START (a node array that fills the address space)
+        if (NLOHMANN_VIEW_UNLIKELY(n >= limit))
+        {
+            document_data::throw_bad_alloc(); // no room for another node
+        }
+        // LCOV_EXCL_STOP
+        // (a count beyond the limit is cut: the index does not grow beyond what can be addressed)
+        doc.reserve(static_cast<std::size_t>((std::min)(wanted, limit)));
         return doc.tape;
     }
 
@@ -1342,7 +1385,10 @@ indent_done:
             n->flags = flags;
             n->extra = extra;
             n->off = static_cast<std::uint32_t>(off);
-            set_integer_bits(*n, second);
+            // len is the low half of the second word, next the high half
+            // (not a native word over both, which swaps them on big-endian)
+            n->len = static_cast<std::uint32_t>(second);
+            n->next = static_cast<std::uint32_t>(second >> 32);
 #endif
             return n;
         }
@@ -1643,9 +1689,8 @@ template<typename BasicJsonType>
 {
     if (f.code == error_code::input_too_large)
     {
-        // LCOV_EXCL_START (4 GiB)
-        NLOHMANN_VIEW_THROW(out_of_range::create(416, "input of 4 GiB or more is not supported by json_document", nullptr));
-        // LCOV_EXCL_STOP
+        // (the limit is detail::view::max_input_size: 4 GiB minus 16 bytes)
+        NLOHMANN_VIEW_THROW(out_of_range::create(416, "input of 4294967280 bytes or more is not supported by json_document", nullptr));
     }
     const BasicJsonType accepted = BasicJsonType::parse(src, src + size, nullptr, true, ignore_comments, ignore_trailing_commas);
     // LCOV_EXCL_START (only if parse() accepts what the view rejects: a bug)
@@ -1684,7 +1729,7 @@ NLOHMANN_JSON_NAMESPACE_END
 
 
 #include <string> // basic_string, char_traits, string
-#include <type_traits> // decay, integral_constant, is_array, is_lvalue_reference, is_pointer, is_same, remove_reference
+#include <type_traits> // decay, integral_constant, is_array, is_const, is_integral, is_lvalue_reference, is_pointer, is_same, remove_reference
 #include <utility> // forward
 
 // #include <nlohmann/json.hpp>
@@ -1704,12 +1749,12 @@ namespace view
 /// how a document takes its input
 enum class input_kind
 {
-    move_string,  ///< rvalue std::string: owned without a copy
+    move_string,  ///< non-const rvalue std::string: owned without a copy
     c_string,     ///< const char* (NUL-terminated): borrowed
     char_array,   ///< char array (e.g. a string literal): borrowed
     borrow_range, ///< lvalue contiguous byte container, or std::string_view: borrowed
-    copy_range,   ///< rvalue contiguous byte container: copied
-    adapter,      ///< anything else parse() accepts (streams, wide strings, ...): read into a buffer
+    copy_range,   ///< rvalue contiguous byte container (a const rvalue std::string too): copied
+    adapter,      ///< streams, wide strings, and the rest of what the library's input adapter reads: read into a buffer
 };
 
 template<typename InputType>
@@ -1728,13 +1773,20 @@ struct classify_input
     static constexpr input_kind value =
         std::is_array<R>::value ? input_kind::char_array
         : std::is_pointer<D>::value ? input_kind::c_string
-        : (is_rvalue && std::is_same<D, std::string>::value) ? input_kind::move_string
+        : (is_rvalue && !std::is_const<R>::value && std::is_same<D, std::string>::value) ? input_kind::move_string
         : (is_bytes && (!is_rvalue || is_string_view)) ? input_kind::borrow_range
         : is_bytes ? input_kind::copy_range
         : input_kind::adapter;
     // NOLINTEND(readability-avoid-nested-conditional-operator)
 };
 
+/// an integer type other than bool: a length passed where a flag is expected
+template<typename T>
+struct is_integer_not_bool : std::is_integral<T> {};
+
+template<>
+struct is_integer_not_bool<bool> : std::false_type {};
+
 /// std::basic_string guarantees a NUL at data()[size()] (the parser's sentinel)
 template<typename T>
 struct is_std_string : std::false_type {};
@@ -2928,12 +2980,6 @@ class basic_json_view
         return type() == value_t::discarded;
     }
 
-    /// false for discarded views
-    explicit operator bool() const noexcept
-    {
-        return m_node != nullptr;
-    }
-
     /// the name of the type, as basic_json::type_name()
     const char* type_name() const noexcept
     {
@@ -3492,6 +3538,13 @@ class basic_json_document
         return d;
     }
 
+    /// parse(ptr, len) does not compile: len would convert to allow_exceptions
+    /// and ptr be read as a C string (as for the overloads of parse_copy,
+    /// accept, and read below)
+    template<typename InputType, typename IntegerType, typename... Flags>
+    static typename std::enable_if<detail::view::is_integer_not_bool<IntegerType>::value, basic_json_document>::type
+    parse(InputType&& input, IntegerType value, Flags&&... flags) = delete;
+
     /// parse [first, last)
     template<typename IteratorType, typename std::enable_if<
                  std::is_base_of<std::input_iterator_tag, typename std::iterator_traits<IteratorType>::iterator_category>::value, int>::type = 0>
@@ -3519,6 +3572,10 @@ class basic_json_document
         return d;
     }
 
+    template<typename InputType, typename IntegerType, typename... Flags>
+    static typename std::enable_if<detail::view::is_integer_not_bool<IntegerType>::value, basic_json_document>::type
+    parse_copy(InputType&& input, IntegerType value, Flags&&... flags) = delete;
+
     /// check whether the input is valid JSON (the result of basic_json::accept)
     template<typename InputType>
     static bool accept(InputType&& input, const bool ignore_comments = false, const bool ignore_trailing_commas = false)
@@ -3528,6 +3585,10 @@ class basic_json_document
         return !d.is_discarded();
     }
 
+    template<typename InputType, typename IntegerType, typename... Flags>
+    static typename std::enable_if<detail::view::is_integer_not_bool<IntegerType>::value, bool>::type
+    accept(InputType&& input, IntegerType value, Flags&&... flags) = delete;
+
     /// parse into this document, reusing its memory
     template<typename InputType>
     // flawfinder: ignore (a member function, not POSIX read())
@@ -3540,12 +3601,17 @@ class basic_json_document
                   std::integral_constant<detail::view::input_kind, detail::view::classify_input<InputType>::value> {});
     }
 
+    template<typename InputType, typename IntegerType, typename... Flags>
+    // flawfinder: ignore (a member function, not POSIX read())
+    typename std::enable_if<detail::view::is_integer_not_bool<IntegerType>::value, void>::type
+    read(InputType&& input, IntegerType value, Flags&&... flags) = delete;
+
     ////////////
     // access //
     ////////////
 
     /// the root value (discarded if parsing failed without exceptions)
-    view_type root() const noexcept
+    view_type root() const& noexcept
     {
         if (!m_data || m_data->discarded)
         {
@@ -3554,6 +3620,9 @@ class basic_json_document
         return view_type(m_data.get(), m_data->tape);
     }
 
+    /// deleted: the view of a temporary document would dangle
+    view_type root() const&& = delete;
+
     bool is_discarded() const noexcept
     {
         return !m_data || m_data->discarded;
@@ -3659,9 +3728,9 @@ class basic_json_document
         d.discarded = true;
         detail::view::parse_failure failure;
         bool ok = false;
-        if (NLOHMANN_VIEW_UNLIKELY(size >= 0xFFFFFFF0u))
+        if (NLOHMANN_VIEW_UNLIKELY(size > detail::view::max_input_size))
         {
-            failure.code = detail::view::error_code::input_too_large; // LCOV_EXCL_LINE (4 GiB)
+            failure.code = detail::view::error_code::input_too_large;
         }
         else
         {

tests/fuzzing.md

@@ -6,7 +6,10 @@ Each parser of the library (JSON, BJData, BON8, BSON, CBOR, MessagePack, and UBJ
 Additionally, `parse_json_view_fuzzer` (`tests/src/fuzzer-parse_json_view.cpp`) cross-checks `json_document`/`json_view`
 (the zero-copy, read-only view declared in `json_view.hpp`) against `basic_json` on the same JSON text: it asserts that
 `json_document::accept` agrees with `json::accept`, that an accepted input materializes to the same value `json::parse`
-produces, and that a rejected input makes both parsers throw with an identical `what()`. It takes plain JSON text, so it
+produces, and that a rejected input makes both parsers throw with an identical `what()`. It checks this for a
+`std::string` input (borrowed, with a NUL after the last byte) and for an exact-size `std::vector<std::uint8_t>`
+(borrowed, with nothing after the last byte), and for the `ignore_comments` and `ignore_trailing_commas` options, which
+are taken from the low bits of the first input byte (the byte stays part of the text). It takes plain JSON text, so it
 reuses the `corpus_json` corpus rather than a format of its own.
 
 ## What the fuzzers check

tests/src/fuzzer-parse_json_view.cpp

@@ -9,7 +9,9 @@
 /*
 This file implements a parser test suitable for fuzz testing. It checks that
 json_document (the zero-copy, read-only view of a parsed JSON text declared in
-json_view.hpp) agrees with basic_json on every input:
+json_view.hpp) agrees with basic_json on every input, for the parse options
+selected by the low bits of the first input byte (bit 0: ignore_comments, bit 1:
+ignore_trailing_commas; the byte stays part of the text):
 
 - json_document::accept(data) must equal json::accept(data)
 - if the input is accepted, json_document::parse(data).root().materialize()
@@ -18,12 +20,20 @@ json_view.hpp) agrees with basic_json on every input:
   enabled) must throw a json::parse_error or json::out_of_range whose what()
   is identical to the one json::parse(data) throws
 
+This is checked for two kinds of input: a std::string, which the document
+borrows and which ends in the NUL the parser uses as sentinel, and an
+exact-size byte vector, which has no NUL after its last byte and takes the
+parser's bounds-checked path (AddressSanitizer reports any read past the end).
+
 The provided function `LLVMFuzzerTestOneInput` can be used in different fuzzer
 drivers.
 */
 
 #include <cassert>
+#include <cstddef>
+#include <cstdint>
 #include <string>
+#include <vector>
 #include <nlohmann/json.hpp>
 #include <nlohmann/json_view.hpp>
 
@@ -35,65 +45,110 @@ drivers.
 using json = nlohmann::json;
 using json_document = nlohmann::json_document;
 
-// see http://llvm.org/docs/LibFuzzer.html
-extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size)
+namespace
 {
-    // json_document::accept only has a single-argument overload; wrap the raw
-    // bytes in a (borrowed) std::string so the same bytes can be handed to it
-    const std::string input(reinterpret_cast<const char*>(data), size); // NOLINT(cppcoreguidelines-pro-type-reinterpret-cast)
+// what json::parse does with a text: the value, or the message of the exception
+struct reference_result
+{
+    bool accepted = false;
+    json value{};
+    std::string what{}; // NOLINT(readability-redundant-member-init)
+};
 
-    const bool accepted_by_json = json::accept(data, data + size);
-    const bool accepted_by_view = json_document::accept(input);
+reference_result parse_reference(const std::uint8_t* data, std::size_t size, bool comments, bool trailing_commas)
+{
+    reference_result r;
+    r.accepted = json::accept(data, data + size, comments, trailing_commas);
+    bool json_threw = false;
+    try
+    {
+        r.value = json::parse(data, data + size, nullptr, true, comments, trailing_commas);
+    }
+    catch (const json::parse_error& e)
+    {
+        r.what = e.what();
+        json_threw = true;
+    }
+    catch (const json::out_of_range& e)
+    {
+        r.what = e.what();
+        json_threw = true;
+    }
+    // json::accept and json::parse must agree
+    assert(json_threw == !r.accepted);
+    static_cast<void>(json_threw);
+    return r;
+}
 
+// json_document must agree with the reference for this input (a container
+// that json_document::parse borrows)
+template<typename Input>
+void check_input(const Input& input, const reference_result& expected, bool comments, bool trailing_commas)
+{
     // json_document::accept must agree with json::accept on every input
-    assert(accepted_by_json == accepted_by_view);
+    const bool accepted_by_view = json_document::accept(input, comments, trailing_commas);
+    assert(expected.accepted == accepted_by_view);
+    static_cast<void>(accepted_by_view);
 
-    if (accepted_by_json)
+    if (expected.accepted)
     {
         // both parsers must agree on the resulting value
-        json const j1 = json::parse(data, data + size);
-        json_document const doc = json_document::parse(input);
+        json_document const doc = json_document::parse(input, true, comments, trailing_commas);
+        assert(!doc.is_discarded());
         json const j2 = doc.root().materialize();
-        assert(j1 == j2);
+        assert(expected.value == j2);
+        static_cast<void>(j2);
+
+        // (without exceptions, the same document)
+        json_document const quiet = json_document::parse(input, false, comments, trailing_commas);
+        assert(!quiet.is_discarded());
+        assert(quiet.node_count() == doc.node_count());
     }
     else
     {
         // both parsers must reject the input the same way when exceptions are used
-        std::string expected_what;
-        bool json_threw = false;
-        try
-        {
-            static_cast<void>(json::parse(data, data + size));
-        }
-        catch (const json::parse_error& e)
-        {
-            expected_what = e.what();
-            json_threw = true;
-        }
-        catch (const json::out_of_range& e)
-        {
-            expected_what = e.what();
-            json_threw = true;
-        }
-        assert(json_threw);
-
         bool view_threw = false;
         try
         {
-            static_cast<void>(json_document::parse(input));
+            static_cast<void>(json_document::parse(input, true, comments, trailing_commas));
         }
         catch (const json::parse_error& e)
         {
-            assert(e.what() == expected_what);
+            assert(e.what() == expected.what);
             view_threw = true;
         }
         catch (const json::out_of_range& e)
         {
-            assert(e.what() == expected_what);
+            assert(e.what() == expected.what);
             view_threw = true;
         }
         assert(view_threw);
+        static_cast<void>(view_threw);
+
+        // and without exceptions, the document is discarded
+        json_document const quiet = json_document::parse(input, false, comments, trailing_commas);
+        assert(quiet.is_discarded());
+        static_cast<void>(quiet);
     }
+}
+} // namespace
+
+// see http://llvm.org/docs/LibFuzzer.html
+extern "C" int LLVMFuzzerTestOneInput(const uint8_t* data, size_t size)
+{
+    // the parse options are taken from the low bits of the first byte
+    const bool comments = size > 0 && (data[0] & 1U) != 0;
+    const bool trailing_commas = size > 0 && (data[0] & 2U) != 0;
+
+    const reference_result expected = parse_reference(data, size, comments, trailing_commas);
+
+    // a std::string: borrowed, with the NUL of std::string as sentinel
+    const std::string input(reinterpret_cast<const char*>(data), size); // NOLINT(cppcoreguidelines-pro-type-reinterpret-cast)
+    check_input(input, expected, comments, trailing_commas);
+
+    // an exact-size byte vector: borrowed, with nothing after its last byte
+    const std::vector<std::uint8_t> exact(data, data + size);
+    check_input(exact, expected, comments, trailing_commas);
 
     // return 0 - non-zero return values are reserved for future use
     return 0;

tests/src/unit-class_lexer.cpp

@@ -1792,4 +1792,11 @@ TEST_CASE("string scanning kernels")
         CHECK(nlohmann::detail::count_trailing_zeros(bit) == k);
         CHECK(nlohmann::detail::count_trailing_zeros(bit | (bit << 1u) | 0x8000000000000000u) == k);
     }
+
+    // eight bytes as a little-endian word, at any alignment
+    const unsigned char bytes[16] = {0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, 0xFF};
+    CHECK(nlohmann::detail::read_eight_bytes(bytes) == 0x0807060504030201u);
+    CHECK(nlohmann::detail::read_eight_bytes(bytes + 1) == 0x0908070605040302u);
+    CHECK(nlohmann::detail::read_eight_bytes(bytes + 8) == 0xFF0F0E0D0C0B0A09u);
+    CHECK(nlohmann::detail::read_eight_bytes(reinterpret_cast<const char*>(bytes) + 3) == 0x0B0A090807060504u);
 }

tests/src/unit-json_view.cpp

@@ -19,6 +19,7 @@ using nlohmann::ordered_json_view;
 #include <algorithm>
 #include <array>
 #include <cmath>
+#include <cstddef>
 #include <cstdint>
 #include <cstdio>
 #include <cstring>
@@ -29,6 +30,7 @@ using nlohmann::ordered_json_view;
 #include <random>
 #include <sstream>
 #include <string>
+#include <type_traits>
 #include <unordered_map>
 #include <utility>
 #include <vector>
@@ -39,6 +41,55 @@ using nlohmann::ordered_json_view;
 
 namespace
 {
+// the value of a text, through a named document: the views of a temporary
+// document would dangle (root() of an rvalue document does not compile)
+template<typename Document, typename... Args>
+auto materialized(Args&& ... args) -> decltype(std::declval<typename Document::view_type>().materialize())
+{
+    const Document d = Document::parse(std::forward<Args>(args)...);
+    return d.root().materialize();
+}
+
+template<typename Document, typename Input>
+auto materialized_copy(Input&& input) -> decltype(std::declval<typename Document::view_type>().materialize())
+{
+    const Document d = Document::parse_copy(std::forward<Input>(input));
+    return d.root().materialize();
+}
+
+// a "byte container" that claims to hold `size` bytes, to reach the limit on
+// the size of the input without allocating gigabytes; nothing past the first
+// bytes is ever read, because the size is checked before the parse starts
+struct oversized_input
+{
+    using value_type = char;
+    std::size_t claimed;
+
+    const char* data() const
+    {
+        return "[1]";
+    }
+
+    std::size_t size() const
+    {
+        return claimed;
+    }
+};
+
+// detection of calls that must not compile
+template<typename... Args>
+using parse_call_t = decltype(json_document::parse(std::declval<Args>()...));
+template<typename... Args>
+using parse_copy_call_t = decltype(json_document::parse_copy(std::declval<Args>()...));
+template<typename... Args>
+using accept_call_t = decltype(json_document::accept(std::declval<Args>()...));
+template<typename... Args>
+using read_call_t = decltype(std::declval<json_document&>().read(std::declval<Args>()...));
+template<typename Document>
+using root_call_t = decltype(std::declval<Document>().root());
+template<typename View>
+using bool_conversion_t = decltype(static_cast<bool>(std::declval<View>()));
+
 #if !defined(JSON_NOEXCEPTION)
 // the exception parse() throws for a text, or "" if it accepts it
 std::string parse_exception(const std::string& text, bool comments = false, bool trailing_commas = false)
@@ -152,7 +203,6 @@ TEST_CASE("json_view")
             CHECK(v.is_primitive() == j.is_primitive());
             CHECK(v.is_structured() == j.is_structured());
             CHECK(!v.is_discarded());
-            CHECK(static_cast<bool>(v));
             CHECK(v.size() == j.size());
             CHECK(v.empty() == j.empty());
             CHECK(v.materialize() == j);
@@ -160,7 +210,6 @@ TEST_CASE("json_view")
 
         const json_view invalid{};
         CHECK(invalid.is_discarded());
-        CHECK(!static_cast<bool>(invalid));
         CHECK(invalid.type() == json::value_t::discarded);
         CHECK(invalid.size() == 0);
         CHECK(invalid.empty());
@@ -176,19 +225,19 @@ TEST_CASE("json_view")
             std::string text;
             g.value(text, 0);
             CAPTURE(text)
-            CHECK(json_document::parse(text).root().materialize() == json::parse(text));
+            CHECK(materialized<json_document>(text) == json::parse(text));
             // member order as ordered_json::parse keeps it
-            CHECK(ordered_json_document::parse(text).root().materialize().dump() == ordered_json::parse(text).dump());
+            CHECK(materialized<ordered_json_document>(text).dump() == ordered_json::parse(text).dump());
         }
         // duplicate keys: the last value, at the position of the first key
-        CHECK(json_document::parse(R"({"a":1,"b":2,"a":3})").root().materialize() == json::parse(R"({"a":1,"b":2,"a":3})"));
-        CHECK(ordered_json_document::parse(R"({"a":1,"b":2,"a":3})").root().materialize().dump() == R"({"a":3,"b":2})");
+        CHECK(materialized<json_document>(R"({"a":1,"b":2,"a":3})") == json::parse(R"({"a":1,"b":2,"a":3})"));
+        CHECK(materialized<ordered_json_document>(R"({"a":1,"b":2,"a":3})").dump() == R"({"a":3,"b":2})");
         // very deep nesting (iterative, as parse())
         const std::string deep = std::string(100000, '[') + std::string(100000, ']');
-        CHECK(json_document::parse(deep).root().materialize() == json::parse(deep));
+        CHECK(materialized<json_document>(deep) == json::parse(deep));
 #if JSON_DIAGNOSTICS
         // the parents are set, so errors name the path
-        const json m = json_document::parse(R"({"a":{"b":[1]}})").root().materialize();
+        const json m = materialized<json_document>(R"({"a":{"b":[1]}})");
         CHECK_THROWS_WITH_AS(m.at("a").at("b").at(0).at("x"), "[json.exception.type_error.304] (/a/b/0) cannot use at() with number", json::type_error&);
 #endif
     }
@@ -261,7 +310,7 @@ TEST_CASE("json_view")
             CHECK(json_document::accept(text));
             if (accepted)
             {
-                CHECK(float_document::parse(text).root().materialize() == json_float::parse(text));
+                CHECK(materialized<float_document>(text) == json_float::parse(text));
             }
         }
         float_document f;
@@ -275,7 +324,7 @@ TEST_CASE("json_view")
         CHECK(json_document::accept(with_nul) == json::accept(with_nul));
         const std::string nul_in_comment("[1, // c\0\n2]", 12);
         CHECK(json_document::accept(nul_in_comment, true) == json::accept(nul_in_comment, true));
-        CHECK(json_document::parse("\xEF\xBB\xBF[1]").root().materialize() == json::parse("\xEF\xBB\xBF[1]"));
+        CHECK(materialized<json_document>("\xEF\xBB\xBF[1]") == json::parse("\xEF\xBB\xBF[1]"));
 #if !defined(JSON_NOEXCEPTION)
         CHECK(view_exception("\xEF\xBB") == parse_exception("\xEF\xBB"));
 #endif
@@ -291,18 +340,18 @@ TEST_CASE("json_view")
         CHECK(!borrowed.owns_source());
         CHECK(borrowed.source().data() == text.data());
         CHECK(borrowed.root().materialize() == expected);
-        CHECK(json_document::parse(text.c_str()).root().materialize() == expected);
-        CHECK(json_document::parse(R"([1, "two", {"three": 3.5}])").root().materialize() == expected);
-        CHECK(json_document::parse(text.data(), text.data() + text.size()).root().materialize() == expected);
+        CHECK(materialized<json_document>(text.c_str()) == expected);
+        CHECK(materialized<json_document>(R"([1, "two", {"three": 3.5}])") == expected);
+        CHECK(materialized<json_document>(text.data(), text.data() + text.size()) == expected);
         const std::vector<char> chars(text.begin(), text.end());
         CHECK(!json_document::parse(chars).owns_source());
-        CHECK(json_document::parse(chars).root().materialize() == expected);
+        CHECK(materialized<json_document>(chars) == expected);
         const std::vector<std::uint8_t> bytes(text.begin(), text.end());
-        CHECK(json_document::parse(bytes).root().materialize() == expected);
+        CHECK(materialized<json_document>(bytes) == expected);
 #ifdef JSON_HAS_CPP_17
         const std::string_view sv = text;
         CHECK(!json_document::parse(sv).owns_source());
-        CHECK(json_document::parse(sv).root().materialize() == expected);
+        CHECK(materialized<json_document>(sv) == expected);
 #endif
 
         // owned
@@ -311,15 +360,28 @@ TEST_CASE("json_view")
         CHECK(from_rvalue.owns_source());
         CHECK(from_rvalue.root().materialize() == expected);
         CHECK(json_document::parse(std::vector<char>(text.begin(), text.end())).owns_source());
+        // a const rvalue cannot be moved from, and is not borrowed (it may be a
+        // temporary): it is copied, as is a const rvalue of any container
+        const std::string const_text = text;
+        const json_document from_const_rvalue = json_document::parse(std::move(const_text)); // NOLINT(performance-move-const-arg,hicpp-move-const-arg)
+        CHECK(from_const_rvalue.owns_source());
+        CHECK(from_const_rvalue.source().data() != const_text.data());
+        CHECK(from_const_rvalue.root().materialize() == expected);
+        json_document read_const_rvalue;
+        read_const_rvalue.read(std::move(const_text)); // NOLINT(performance-move-const-arg,hicpp-move-const-arg)
+        CHECK(read_const_rvalue.owns_source());
+        CHECK(read_const_rvalue.root().materialize() == expected);
+        const std::vector<char> const_chars(text.begin(), text.end());
+        CHECK(json_document::parse(std::move(const_chars)).owns_source()); // NOLINT(performance-move-const-arg,hicpp-move-const-arg)
         CHECK(json_document::parse_copy(text).owns_source());
-        CHECK(json_document::parse_copy(text).root().materialize() == expected);
+        CHECK(materialized_copy<json_document>(text) == expected);
         std::istringstream stream(text);
         const json_document from_stream = json_document::parse(stream);
         CHECK(from_stream.owns_source());
         CHECK(from_stream.root().materialize() == expected);
         const std::list<char> list(text.begin(), text.end());
         CHECK(json_document::parse(list.begin(), list.end()).owns_source());
-        CHECK(json_document::parse(list.begin(), list.end()).root().materialize() == expected);
+        CHECK(materialized<json_document>(list.begin(), list.end()) == expected);
 
         // iterator pairs: pointers are borrowed, and so are contiguous library
         // iterators where the input adapter detects them (C++20)
@@ -330,14 +392,85 @@ TEST_CASE("json_view")
         CHECK((from_iterators.source().data() == chars.data()) == contiguous);
         CHECK(from_iterators.root().materialize() == expected);
         const std::string padded = "x" + text + "x";
-        CHECK(json_document::parse(padded.begin() + 1, padded.end() - 1).root().materialize() == expected);
+        CHECK(materialized<json_document>(padded.begin() + 1, padded.end() - 1) == expected);
         CHECK(json_document::parse(chars.cbegin(), chars.cbegin(), false).is_discarded());
         const std::wstring wide = L"[\"\u00e4\u20ac\", 1]";
-        CHECK(json_document::parse(wide).root().materialize() == json::parse(wide));
+        CHECK(materialized<json_document>(wide) == json::parse(wide));
         CHECK(json_document::parse(static_cast<const char*>(nullptr), false).is_discarded());
         CHECK(json_document::parse("", false).is_discarded());
     }
 
+    SECTION("integer arguments do not compile")
+    {
+        using nlohmann::detail::is_detected;
+
+        // a length is not a flag: parse(ptr, len) would convert len to
+        // allow_exceptions and read ptr as a C string, which need not end
+        static_assert(is_detected<parse_call_t, const char*, bool>::value, "parse(ptr, bool) is valid");
+        static_assert(is_detected<parse_call_t, const char*, bool, bool, bool>::value, "parse(ptr, bool, bool, bool) is valid");
+        static_assert(is_detected<parse_call_t, const char*, const char*>::value, "parse(first, last) is valid");
+        static_assert(is_detected<parse_call_t, const char*, const char*, bool>::value, "parse(first, last, bool) is valid");
+        static_assert(!is_detected<parse_call_t, const char*, std::size_t>::value, "parse(ptr, len) must not compile");
+        static_assert(!is_detected<parse_call_t, const char*, int>::value, "parse(ptr, int) must not compile");
+        static_assert(!is_detected<parse_call_t, const char*, char>::value, "parse(ptr, char) must not compile");
+        static_assert(!is_detected<parse_call_t, const char*, std::size_t, bool>::value, "parse(ptr, len, bool) must not compile");
+        static_assert(!is_detected<parse_call_t, const std::string&, std::size_t>::value, "parse(string, len) must not compile");
+        static_assert(!is_detected<parse_call_t, const std::vector<char>&, std::size_t>::value, "parse(vector, len) must not compile");
+
+        static_assert(is_detected<parse_copy_call_t, const char*, bool>::value, "parse_copy(ptr, bool) is valid");
+        static_assert(!is_detected<parse_copy_call_t, const char*, std::size_t>::value, "parse_copy(ptr, len) must not compile");
+
+        static_assert(is_detected<accept_call_t, const char*, bool>::value, "accept(ptr, bool) is valid");
+        static_assert(!is_detected<accept_call_t, const char*, std::size_t>::value, "accept(ptr, len) must not compile");
+
+        static_assert(is_detected<read_call_t, const char*, bool>::value, "read(ptr, bool) is valid");
+        static_assert(!is_detected<read_call_t, const char*, std::size_t>::value, "read(ptr, len) must not compile");
+
+        // json_view has no conversion to bool: unlike basic_json's, it would
+        // mean "exists", not "is not null"; use is_discarded()
+        static_assert(!is_detected<bool_conversion_t, json_view>::value, "json_view must not convert to bool");
+
+        // the valid calls still work
+        const char* const text = "[1]";
+        CHECK(materialized<json_document>(text, true) == json::parse(text));
+        CHECK(json_document::accept(text, true, true));
+    }
+
+    SECTION("root of a temporary document does not compile")
+    {
+        using nlohmann::detail::is_detected;
+
+        // the view would dangle: auto v = json_document::parse(text).root();
+        static_assert(is_detected<root_call_t, json_document&>::value, "root() of an lvalue is valid");
+        static_assert(is_detected<root_call_t, const json_document&>::value, "root() of a const lvalue is valid");
+        static_assert(!is_detected<root_call_t, json_document>::value, "root() of an rvalue must not compile");
+        static_assert(!is_detected < root_call_t, json_document && >::value, "root() of an rvalue must not compile");
+        static_assert(!is_detected < root_call_t, const json_document && >::value, "root() of a const rvalue must not compile");
+        static_assert(!is_detected<root_call_t, ordered_json_document>::value, "root() of an rvalue must not compile");
+
+        // a named document is fine, also after a move
+        json_document d = json_document::parse("[1]");
+        CHECK(d.root().size() == 1);
+        const json_document moved = std::move(d);
+        CHECK(moved.root().size() == 1);
+    }
+
+    SECTION("input size limit")
+    {
+        // 32-bit offsets: the limit is 4 GiB minus 16 bytes (a margin below 2^32),
+        // which is what the exception message and the documentation say
+        const std::size_t limit = nlohmann::detail::view::max_input_size;
+        CHECK(limit == std::size_t{4294967279u});
+
+        const oversized_input input{limit + 1};
+        CHECK(!json_document::accept(input));
+        CHECK(json_document::parse(input, false).is_discarded());
+#if !defined(JSON_NOEXCEPTION)
+        json_document d;
+        CHECK_THROWS_WITH_AS(d = json_document::parse(input), "[json.exception.out_of_range.416] input of 4294967280 bytes or more is not supported by json_document", json::out_of_range&);
+#endif
+    }
+
     SECTION("document lifetime and reuse")
     {
         json_document d;

tests/src/unit-json_view_builder.cpp

@@ -19,8 +19,10 @@ using nlohmann::json;
 
 #include <cstdint>
 #include <fstream>
+#include <limits>
 #include <map>
 #include <memory>
+#include <new>
 #include <random>
 #include <sstream>
 #include <string>
@@ -401,3 +403,89 @@ TEST_CASE("json_view builder")
         }
     }
 }
+
+TEST_CASE("json_view node integer bits")
+{
+    using nlohmann::detail::view::integer_bits;
+    using nlohmann::detail::view::set_integer_bits;
+
+    // an integer lives in len (low half) and next (high half), on any byte
+    // order; a big-endian target must not store the native word over both
+    SECTION("set_integer_bits and integer_bits")
+    {
+        node n = {};
+        for (const std::uint64_t v :
+                {
+                    std::uint64_t{0}, std::uint64_t{1}, std::uint64_t{0xFFFFFFFFu}, std::uint64_t{0x100000000u},
+                    std::uint64_t{0x0000000200000003u}, std::uint64_t{0x0123456789ABCDEFu}, std::uint64_t{0xFFFFFFFFFFFFFFFEu}
+                })
+        {
+            CAPTURE(v)
+            n.kind = 0x5A;
+            n.flags = 0xA5;
+            n.extra = 0x1234;
+            n.off = 0x89ABCDEFu;
+            set_integer_bits(n, v);
+            CHECK(integer_bits(n) == v);
+            CHECK(n.len == static_cast<std::uint32_t>(v));
+            CHECK(n.next == static_cast<std::uint32_t>(v >> 32));
+            // the other fields are untouched
+            CHECK(n.kind == 0x5A);
+            CHECK(n.flags == 0xA5);
+            CHECK(n.extra == 0x1234);
+            CHECK(n.off == 0x89ABCDEFu);
+        }
+    }
+
+    SECTION("parsed integers")
+    {
+        struct integer_case
+        {
+            const char* text;
+            std::uint64_t bits;
+        };
+        for (const integer_case c :
+                {
+                    integer_case{"[8589934595]", 0x0000000200000003u}, integer_case{"[4294967296]", 0x100000000u}, integer_case{"[4294967295]", 0xFFFFFFFFu},
+                    integer_case{"[-2]", 0xFFFFFFFFFFFFFFFEu}, integer_case{"[-4294967297]", 0xFFFFFFFEFFFFFFFFu}, integer_case{"[18446744073709551615]", 0xFFFFFFFFFFFFFFFFu},
+                    integer_case{"[7]", 7u}
+                })
+        {
+            CAPTURE(c.text)
+            const built b = build(c.text, false, false, true);
+            REQUIRE(b.ok);
+            const node& n = b.data->tape[1];
+            CHECK(integer_bits(n) == c.bits);
+            CHECK(n.len == static_cast<std::uint32_t>(c.bits));
+            CHECK(n.next == static_cast<std::uint32_t>(c.bits >> 32));
+        }
+    }
+}
+
+TEST_CASE("json_view node array size limit")
+{
+    // a node array larger than the address space is refused, not wrapped to a
+    // small allocation (the size computation overflows on 32-bit targets, and
+    // for absurd counts everywhere)
+    std::unique_ptr<document_data, document_data::deleter> d(document_data::create(0));
+    d->reserve(8);
+    REQUIRE(d->tape_cap >= 8);
+    d->tape_size = 2;
+    const std::size_t cap = d->tape_cap;
+    node* const tape = d->tape;
+
+#if !defined(JSON_NOEXCEPTION)
+    const std::size_t too_many = document_data::max_nodes() + 1;
+    CHECK_THROWS_AS(d->reserve(too_many), std::bad_alloc&);
+    CHECK_THROWS_AS(d->reserve((std::numeric_limits<std::size_t>::max)()), std::bad_alloc&);
+    // the array is unchanged
+    CHECK(d->tape == tape);
+    CHECK(d->tape_cap == cap);
+    CHECK(d->tape_size == 2);
+#endif
+
+    // the largest count that fits is not refused by the check (nothing is
+    // allocated for a count that is already there)
+    d->reserve(cap);
+    CHECK(d->tape == tape);
+}

tests/src/unit-to_chars.cpp

@@ -15,6 +15,7 @@
 #include <nlohmann/json.hpp>
 using nlohmann::detail::dtoa_impl::reinterpret_bits;
 
+#include <algorithm>
 #include <array>
 #include <cmath>
 #include <cstdint>
@@ -666,13 +667,24 @@ void check_shortest(double v)
     const std::string text(buf.data(), end);
     CAPTURE(text)
     CHECK(parse_double(text) == v);
-    // the layout is that of format_buffer() for the same digits
+    // the layout is that of format_buffer() for the digits of Zmij
+    const auto sd = nlohmann::detail::zmij::to_shortest(reinterpret_bits<std::uint64_t>(v));
+    const std::uint64_t significand = sd.has_digit ? (sd.integral * 10) + sd.digit : sd.integral;
+    int exponent = sd.has_digit ? sd.exponent : sd.exponent + 1;
+    std::string significand_digits = std::to_string(significand);
+    while (significand_digits.size() > 1 && significand_digits.back() == '0')
+    {
+        significand_digits.pop_back();
+        ++exponent;
+    }
     std::array<char, 64> reference{};
-    int len = 0;
-    int exponent = 0;
-    nlohmann::detail::dtoa_impl::shortest_digits(reference.data(), len, exponent, v);
-    const char* const reference_end = nlohmann::detail::dtoa_impl::format_buffer(reference.data(), len, exponent, -4, 15);
+    std::copy(significand_digits.begin(), significand_digits.end(), reference.begin());
+    const char* const reference_end = nlohmann::detail::dtoa_impl::format_buffer(reference.data(), static_cast<int>(significand_digits.size()), exponent, -4, 15);
     CHECK(text == std::string(reference.data(), static_cast<std::size_t>(reference_end - reference.data())));
+    // and write_positive() is what to_chars() calls
+    std::array<char, 64> positive{};
+    const char* const positive_end = nlohmann::detail::dtoa_impl::write_positive(positive.data(), positive.data() + positive.size(), v);
+    CHECK(text == std::string(positive.data(), static_cast<std::size_t>(positive_end - positive.data())));
     const auto de = digits_and_exponent(text);
     const std::string& digits = de.first;
     if (digits.size() > 1)
@@ -785,3 +797,58 @@ TEST_CASE("shortest digits of doubles")
         }
     }
 }
+
+TEST_CASE("choice of the conversion")
+{
+    using nlohmann::detail::dtoa_impl::is_binary64;
+
+    SECTION("by the format of the type")
+    {
+        // Zmij needs binary64 numbers; everything else uses Grisu2
+        static_assert(!is_binary64<float>::value, "float is not binary64");
+        static_assert(is_binary64<double>::value == (std::numeric_limits<double>::is_iec559 && std::numeric_limits<double>::digits == 53),
+                      "double is binary64 where it is IEEE 754 with 53 digits");
+        static_assert(!is_binary64<int>::value, "integers are not binary64");
+        static_assert(is_binary64<long double>::value == (std::numeric_limits<long double>::is_iec559 && std::numeric_limits<long double>::digits == 53 && sizeof(long double) == 8),
+                      "long double is binary64 where it has the format of a double");
+        CHECK(!is_binary64<float>::value);
+        CHECK(is_binary64<double>::value);
+    }
+
+    SECTION("float: Grisu2, double: Zmij")
+    {
+        // 5.3165205877497296e+16 is one of the doubles for which Grisu2 does not find the shortest digits
+        constexpr double value = 5.3165205877497296e+16;
+        std::array<char, 64> buf{};
+        const char* const last = buf.data() + buf.size();
+
+        char* end = nlohmann::detail::dtoa_impl::write_positive(buf.data(), last, value);
+        CHECK(std::string(buf.data(), end) == "5.31652058774973e+16");
+        end = nlohmann::detail::dtoa_impl::write_positive_grisu2(buf.data(), last, value);
+        CHECK(std::string(buf.data(), end) == "5.3165205877497296e+16");
+
+        constexpr float f = 1.1754944e-38f;
+        end = nlohmann::detail::dtoa_impl::write_positive(buf.data(), last, f);
+        const std::string dispatched(buf.data(), end);
+        end = nlohmann::detail::dtoa_impl::write_positive_grisu2(buf.data(), last, f);
+        CHECK(dispatched == std::string(buf.data(), end));
+    }
+
+    SECTION("long double with the format of a double: Zmij")
+    {
+        // (on platforms where long double is wider, Grisu2 does not apply either: the snprintf fallback does)
+        if (std::numeric_limits<long double>::digits == 53 && std::numeric_limits<long double>::is_iec559)
+        {
+            using long_double_json = nlohmann::json::with_float_t<long double>;
+            for (const double d :
+                    {
+                        5.3165205877497296e+16, 1.0, 0.1, 123456.789, 2.2250738585072014e-308, 1.7976931348623157e+308, -5.3165205877497296e+16
+                    })
+            {
+                CAPTURE(d)
+                CHECK(long_double_json(static_cast<long double>(d)).dump() == nlohmann::json(d).dump());
+            }
+            CHECK(long_double_json(5.3165205877497296e+16L).dump() == "5.31652058774973e+16");
+        }
+    }
+}