json_view: no per-object allocations in ordered_json comparisons

For an object type that keeps the order of its members (ordered_json), resolve_indices() allocated a vector per compared object, and resolve_members() two. They now use one working vector owned by equal(), whose capacity is reused, and resolve_indices() needs none when the keys are unique. resolve_members() sorts small objects by insertion, as the other paths do. Instructions per comparison (callgrind, GCC 13 and clang 18, x86-64): twitter, ordered_json view == ordered_json: 0.939 / 0.941; 20000 small objects with a duplicate key: 0.741-0.768; all other rows (json, citm_catalog, canada, view == view) within 0.988-1.003. Signed-off-by: Niels Lohmann <mail@nlohmann.me>
Merge develop into perf/view-compare
2026-10-26 12:14:46 +00:00 · 2026-10-25 21:56:14 +01:00 · 2026-10-25 12:13:26 +01:00 · 2026-10-25 12:04:43 +01:00 · 2026-10-25 12:00:01 +01:00 · 2026-10-24 20:03:58 +02:00
21 changed files with 2911 additions and 1036 deletions
@@ -90,9 +90,8 @@ type `#!cpp binary_t*` must be dereferenced.

 - MessagePack
    - If a subtype is given and the binary array contains exactly 1, 2, 4, 8, or 16 elements, the fixext family (fixext1,
-      fixext2, fixext4, fixext8, fixext16) is used. For other sizes, the ext family (ext8, ext16, ext32) is used. The
-      subtype is then written as the signed 8-bit ext type: subtypes 128 to 255 map to the negative (reserved) ext
-      types, and on reading the ext type is exposed as an unsigned subtype from 0 to 255.
+      fixext2, fixext4, fixext8) is used. For other sizes, the ext family (ext8, ext16, ext32) is used. The subtype is
+      then added as a signed 8-bit integer.
    - If no subtype is given, the bin family (bin8, bin16, bin32) is used.

 - BSON
@@ -15,13 +15,11 @@ error
 :   report a parse error in case of a tag (the `from_cbor` overloads throw a `parse_error` exception by default)

 ignore
-:   ignore tags (also in front of a text string map key)
+:   ignore tags

 store
 :   store tagged byte strings (for bytes 0xd8..0xdb) as binary values with the tag as subtype; other tagged values are
-    read as if the tag were ignored. If several tags precede a byte string, only the tag directly in front of it is
-    stored, and only when it has the form 0xd8..0xdb (a tag 0..23 written as 0xc0..0xd7 is skipped). Tags in front of a
-    text string map key are skipped.
+    read as if the tag were ignored. If several tags precede a byte string, only the innermost one is stored.

 ## Examples

@@ -32,10 +32,8 @@ The exact mapping and its limitations are described on a [dedicated page](../../
    
    - an `std::istream` object
    - a `FILE` pointer
-    - a C-style array of characters (its terminating NUL is read as trailing data: `parse_error.110` unless `strict` is
-      `false`)
-    - a pointer to a null-terminated string of single byte characters (read up to the first 0x00 byte, which is a valid
-      CBOR byte)
+    - a C-style array of characters
+    - a pointer to a null-terminated string of single byte characters
    - a container `obj` for which `begin(obj)` and `end(obj)` produce a valid pair of iterators
      (as found via ADL or member functions, with semantics compatible to `std::begin` and `std::end`)

@@ -91,14 +89,8 @@ Strong guarantee: if an exception is thrown, there are no changes in the JSON va
 - Throws [parse_error.112](../../home/exceptions.md#jsonexceptionparse_error112) if unsupported features from CBOR were
  used in the given input or if the input is not valid CBOR
 - Throws [parse_error.113](../../home/exceptions.md#jsonexceptionparse_error113) if a map key is not a string (keys of
-  other types are not supported, as JSON object keys are always strings), if the head of a string or byte string is
-  malformed (including a chunk of the wrong type or an indefinite-length string nested inside an indefinite-length
-  string), or if a string value or object key (or one chunk of an indefinite-length string) is not valid UTF-8 and
-  `error_handler` is `strict`
- Throws [out_of_range.408](../../home/exceptions.md#jsonexceptionout_of_range408) if an array or map announces
-  a length that `std::size_t` cannot hold, or that equals `std::size_t(-1)` (for example `9B FF FF FF FF FF FF FF FF`)
- Throws [out_of_range.406](../../home/exceptions.md#jsonexceptionout_of_range406) if a floating-point number does not
-  fit into `number_float_t`
+  other types are not supported, as JSON object keys are always strings), or if a string value or object key is not
+  valid UTF-8 and `error_handler` is `strict`

 ## Complexity

@@ -42,8 +42,7 @@ The exact mapping and its limitations are described on a [dedicated page](../../

 ## Exception safety

-Strong guarantee: if an exception is thrown, there are no changes in the JSON value. Bytes that were already written to
-the output adapter are not taken back.
+Strong guarantee: if an exception is thrown, there are no changes in the JSON value.

 ## Exceptions

@@ -53,7 +53,7 @@ Strong guarantee: if an exception is thrown, there are no changes in the JSON va
  value, array, or object exceeds 4294967295, the maximum MessagePack can store; example:
  `"MessagePack length 4294967296 exceeds maximum of 4294967295"`
 - Throws [`out_of_range.415`](../../home/exceptions.md#jsonexceptionout_of_range415) if the subtype of a binary value
-  exceeds 255, the largest value the library can store in the one-byte MessagePack ext type; example:
+  exceeds 255, the maximum of the MessagePack ext type; example:
  `"subtype 70000 is too large for the MessagePack ext type (max 255)"`
 - Throws [type_error.316](../../home/exceptions.md#jsonexceptiontype_error316) if a string or object key in `j` is
  not valid UTF-8 and `error_handler` is `strict`
@@ -11,7 +11,7 @@ extremely small code sizes, fairly small message size, and extensibility without

 ## Byte layout

-CBOR encodes every data item as a *head* followed by optional content. The first byte of the head holds the *major type* in its upper three bits (unsigned integer, negative integer, byte string, text string, array, map, tag, or simple/float) and an *additional information* value in its lower five bits. Additional information below 24 is the argument itself (for example, a small integer or a short length); the values 24 to 27 announce that the argument follows in 1, 2, 4, or 8 big-endian bytes. Strings, arrays, and maps are prefixed with their length (or count of elements/pairs), and content follows directly. A reader of definite-length items knows how many items to expect without a terminator; indefinite-length items (additional information 31) are instead closed by the "break" byte 0xFF, which `from_cbor` reads but `to_cbor` never writes.
+CBOR encodes every data item as a *head* followed by optional content. The first byte of the head holds the *major type* in its upper three bits (unsigned integer, negative integer, byte string, text string, array, map, tag, or simple/float) and an *additional information* value in its lower five bits. Additional information below 24 is the argument itself (for example, a small integer or a short length); the values 24 to 27 announce that the argument follows in 1, 2, 4, or 8 big-endian bytes. Strings, arrays, and maps are prefixed with their length (or count of elements/pairs), and content follows directly, so a reader always knows how many items to expect without a terminator.

 Consider the following JSON value (shown formatted for readability):

@@ -52,16 +52,16 @@ The library uses the following mapping from JSON values types to CBOR types acco
 | null            | `null`                                     | Null                              | 0xF6       |
 | boolean         | `true`                                     | True                              | 0xF5       |
 | boolean         | `false`                                    | False                             | 0xF4       |
-| number_integer  | -9223372036854775808..-4294967297          | Negative integer (8 bytes follow) | 0x3B       |
-| number_integer  | -4294967296..-65537                        | Negative integer (4 bytes follow) | 0x3A       |
-| number_integer  | -65536..-257                               | Negative integer (2 bytes follow) | 0x39       |
-| number_integer  | -256..-25                                  | Negative integer (1 byte follow)  | 0x38       |
+| number_integer  | -9223372036854775808..-2147483649          | Negative integer (8 bytes follow) | 0x3B       |
+| number_integer  | -2147483648..-32769                        | Negative integer (4 bytes follow) | 0x3A       |
+| number_integer  | -32768..-129                               | Negative integer (2 bytes follow) | 0x39       |
+| number_integer  | -128..-25                                  | Negative integer (1 byte follow)  | 0x38       |
 | number_integer  | -24..-1                                    | Negative integer                  | 0x20..0x37 |
 | number_integer  | 0..23                                      | Integer                           | 0x00..0x17 |
 | number_integer  | 24..255                                    | Unsigned integer (1 byte follow)  | 0x18       |
 | number_integer  | 256..65535                                 | Unsigned integer (2 bytes follow) | 0x19       |
 | number_integer  | 65536..4294967295                          | Unsigned integer (4 bytes follow) | 0x1A       |
-| number_integer  | 4294967296..9223372036854775807            | Unsigned integer (8 bytes follow) | 0x1B       |
+| number_integer  | 4294967296..18446744073709551615           | Unsigned integer (8 bytes follow) | 0x1B       |
 | number_unsigned | 0..23                                      | Integer                           | 0x00..0x17 |
 | number_unsigned | 24..255                                    | Unsigned integer (1 byte follow)  | 0x18       |
 | number_unsigned | 256..65535                                 | Unsigned integer (2 bytes follow) | 0x19       |
@@ -95,24 +95,7 @@ see "binary" cells in the table above.

 !!! success "Complete mapping"

-    The mapping is **complete** in the sense that any JSON value type can be converted to a CBOR value, with one
-    exception: an integer outside the range of CBOR integers (only possible with a number type wider than 64 bits)
-    throws [`out_of_range.407`](../../home/exceptions.md#jsonexceptionout_of_range407).
-
-!!! info "Not a deterministic encoding"
-
-    `to_cbor` does not produce the deterministic encoding of
-    [RFC 8949, Section 4.2.1](https://www.rfc-editor.org/rfc/rfc8949.html#section-4.2.1), and not always the preferred
-    serialization of Section 4.1:
-
-    - floating-point numbers are written as single precision (0xFA) when that is exact and as double precision (0xFB)
-      otherwise; half precision (0xF9) is used only for `NaN` and `Infinity`/`-Infinity` (`1.0` is `fa 3f 80 00 00`);
-    - a binary subtype 0..23 is written as `d8 xx` instead of the one-byte tag 0xC0..0xD7 (`d8 01 41 01`);
-    - object members are written in the order of `object_t` (for the default `std::map`: sorted by the key string, not
-      by the encoded key bytes).
-
-    Integers and lengths use the shortest form, and indefinite-length items are never written. Decoders must accept all
-    of this; the RFC does not require the preferred serialization.
+    The mapping is **complete** in the sense that any JSON value type can be converted to a CBOR value.

 !!! info "NaN/infinity handling"

@@ -122,8 +105,9 @@ see "binary" cells in the table above.

 !!! note

-    Prior to version 3.8.0, NaN and Infinity were instead serialized as a CBOR double-precision float (type 0xFB,
-    9 bytes total).
+    Prior to version 3.13.0, NaN and Infinity were instead serialized as a CBOR double-precision float (type 0xFB,
+    9 bytes total), because the check used to select a smaller encoding compared magnitudes with NaN, which is
+    always `false` and caused the intended half-precision path to be skipped.

 !!! info "Unused CBOR types"

@@ -140,6 +124,7 @@ see "binary" cells in the table above.
      - expected conversions (0xD5..0xD7)
      - simple values (0xE0..0xF3, 0xF8)
      - undefined (0xF7)
+      - half-precision floats (0xF9)
      - break (0xFF)

 !!! info "Tagged items"
@@ -241,20 +226,10 @@ string of the same major type, as required by [RFC 8949, Section 3.2.3](https://
    ([RFC 8392](https://www.rfc-editor.org/rfc/rfc8392.html)), cannot be read with this library and need a
    general-purpose CBOR library instead.

-    A tag in front of a text string key is handled like a tag in front of a value: it is skipped with
-    `cbor_tag_handler_t::ignore` or `cbor_tag_handler_t::store` and rejected with the default `error`. A tag does not
-    turn a key of another type into a text string.
-
-    A map with duplicate keys is well-formed, but invalid under
-    [RFC 8949, Section 5.3.1](https://www.rfc-editor.org/rfc/rfc8949.html#section-5.3.1). `from_cbor` accepts it and the
-    last value wins; the [SAX interface](../parsing/sax_interface.md) reports every occurrence.
-
 !!! warning "Ill-formed UTF-8 in text strings"

    [RFC 8949, Section 3.1](https://www.rfc-editor.org/rfc/rfc8949.html#section-3.1) requires CBOR text strings (major
-    type 3) to be valid UTF-8 (a string with ill-formed UTF-8 is "well-formed but invalid"), and
-    [Section 5.3.1](https://www.rfc-editor.org/rfc/rfc8949.html#section-5.3.1) leaves it up to the decoder whether to
-    verify this, so checking is opt-in: with the
+    type 3) to be valid UTF-8, but leaves it up to the decoder whether to enforce this, so checking is opt-in: with the
    [`error_handler`](../../api/basic_json/from_cbor.md) parameter left at `keep` (the default), `from_cbor()` accepts a
    text string (object keys included) whose bytes are not valid UTF-8 and hands them back unchanged. Passing
    `error_handler_t::strict` makes `from_cbor()` check and throw
@@ -267,14 +242,9 @@ string of the same major type, as required by [RFC 8949, Section 3.2.3](https://
    [`JSON_STRICT_BINARY_UTF8`](../../api/macros/json_strict_binary_utf8.md) is enabled), it throws the same exception
    instead. Byte strings (major type 2) are unaffected, since they are not required to hold text.

-    The chunks of an indefinite-length text string are checked one by one, as required by
-    [RFC 8949, Section 3.2.3](https://www.rfc-editor.org/rfc/rfc8949.html#section-3.2.3): a code point cannot be spread
-    over two chunks. With `error_handler_t::strict` such a string throws `parse_error.113`; `replace` and `ignore`
-    sanitize each chunk on its own.
-
 !!! warning "Tagged items"

-    Tagged items (0xC0..0xDB) will throw a parse error by default. They can be ignored by passing `cbor_tag_handler_t::ignore` to function `from_cbor`, in which case the tag is skipped and the enclosed data item is parsed on its own. Passing `cbor_tag_handler_t::store` to function `from_cbor` stores tagged byte strings (for bytes 0xd8..0xdb) as binary values with the tag as subtype; other tagged values are read as if the tag were ignored. If several tags precede a byte string, only the tag directly in front of it is stored, and only when it has the form 0xD8..0xDB (a tag 0..23 written as 0xC0..0xD7 is skipped); for example, `D8 05 C1 41 AA` yields subtype `null`, whereas `C1 D8 05 41 AA` yields 5. Tags in front of a text string map key are skipped as well. Note that no tag is ever interpreted: for instance, a text string tagged with tag 0 (date/time) stays a string.
+    Tagged items (0xC0..0xDB) will throw a parse error by default. They can be ignored by passing `cbor_tag_handler_t::ignore` to function `from_cbor`, in which case the tag is skipped and the enclosed data item is parsed on its own. Passing `cbor_tag_handler_t::store` to function `from_cbor` stores tagged byte strings (for bytes 0xd8..0xdb) as binary values with the tag as subtype; other tagged values are read as if the tag were ignored. If several tags precede a byte string, only the innermost one is stored. Note that no tag is ever interpreted: for instance, a text string tagged with tag 0 (date/time) stays a string.

 ??? example "Example: deserialize a JSON value from CBOR"

@@ -22,7 +22,7 @@ to efficiently encode JSON values to byte vectors and to decode such vectors.
 | BON8        | incomplete: no unsigned integers above int64  | complete                                     |
 | BSON        | incomplete: top-level value must be an object | incomplete, but all JSON types are supported |
 | CBOR        | complete                                      | incomplete, but all JSON types are supported |
-| MessagePack | complete                                      | incomplete, but all JSON types are supported |
+| MessagePack | complete                                      | complete                                     |
 | UBJSON      | complete                                      | complete                                     |

 ### Binary values
@@ -11,7 +11,7 @@ extra byte in addition to the strings themselves.

 ## Byte layout

-MessagePack encodes each value as a one-byte *format marker*, optionally followed by a length or the payload. The marker range is partitioned to keep common values small: small non-negative integers (0 to 127) and small negative integers (-32 to -1) are the marker byte itself; short strings, arrays, and maps (up to 31 bytes for strings and up to 15 elements for arrays and maps) hide their length in the marker's lower bits. Larger values use dedicated markers that announce the width (8, 16, 32, or 64 bits) of the length or number that follows, in big-endian byte order. Like CBOR, containers are length-prefixed, so no terminator is needed.
+MessagePack encodes each value as a one-byte *format marker*, optionally followed by a length or the payload. The marker range is partitioned to keep common values small: small non-negative integers (0 to 127) and small negative integers (-32 to -1) are the marker byte itself; short strings, arrays, and maps (up to 31 or 15 elements) hide their length in the marker's lower bits. Larger values use dedicated markers that announce the width (8, 16, 32, or 64 bits) of the length or number that follows, in big-endian byte order. Like CBOR, containers are length-prefixed, so no terminator is needed.

 Consider the following JSON value (shown formatted for readability):

@@ -109,9 +109,8 @@ specification:
 !!! note

    Prior to version 3.13.0, NaN and Infinity were instead serialized as a MessagePack float 64 (type 0xCB, 9 bytes
-    total), because the check used to select the smaller float 32 encoding failed for NaN (all comparisons are `false`)
-    and for Infinity and -Infinity (outside the float range checked against `std::numeric_limits<float>::lowest()` and
-    `max()`), so the float 32 path was skipped.
+    total), because the check used to select the smaller float 32 encoding compared magnitudes with NaN, which is
+    always `false` and caused the float 32 path to be skipped.

 ??? example "Example: serialize a JSON value to MessagePack"

@@ -136,7 +135,6 @@ The library maps MessagePack types to JSON value types as follows:
 | fixarray         | array           | 0x90..0x9F |
 | fixstr           | string          | 0xA0..0xBF |
 | nil              | `null`          | 0xC0       |
-| (never used)     | –               | 0xC1       |
 | false            | `false`         | 0xC2       |
 | true             | `true`          | 0xC3       |
 | float 32         | number_float    | 0xCA       |
@@ -169,30 +167,6 @@ The library maps MessagePack types to JSON value types as follows:
 | fixext 16        | binary          | 0xD8       |
 | negative fixint  | number_integer  | 0xE0-0xFF  |

-The byte 0xC1 is reserved as "never used" by the specification. As a value it is rejected with
-[`parse_error.112`](../../home/exceptions.md#jsonexceptionparse_error112) (`invalid byte: 0xC1`); as a map key it is
-rejected with [`parse_error.113`](../../home/exceptions.md#jsonexceptionparse_error113) and the generic "expected length
-specification" message, because the key is read as a string.
-
-Duplicate keys in a map are accepted; the last value wins.
-
-### Extension types (ext, fixext)
-
-MessagePack ext and fixext values are read as binary values, and the ext type byte is stored as the binary subtype.
-Although the specification defines the type as a signed 8-bit integer (0 to 127 for applications, -128 to -1 reserved,
-with -1 being the Timestamp extension), the library reads it as an unsigned byte: a negative ext type `n` becomes
-subtype `256+n`, so ext type -1 is read as subtype 255 and -128 as subtype 128. The Timestamp extension is not
-interpreted; it is returned as binary with subtype 255. Use
-[`has_subtype()`](../../api/byte_container_with_subtype/has_subtype.md) to tell a `bin` value (no subtype) from an ext
-value (subtype, even if it is 0).
-
-When serializing, a binary value with a subtype is always written as fixext (1, 2, 4, 8, or 16 bytes) or ext, even
-for subtype 0 (`d6 00 01 02 03 04`); a binary value without a subtype is written as bin. Subtypes 128 to 255 are
-written as the reserved negative ext types (255 as -1), so `json::binary({1,2,3,4}, 255)` becomes
-`d6 ff 01 02 03 04`, which other MessagePack libraries decode as a 32-bit timestamp. A subtype above 255 throws
-[`out_of_range.415`](../../home/exceptions.md#jsonexceptionout_of_range415). The payload length is limited to
-4294967295 bytes like all other lengths.
-
 !!! info

    Any MessagePack output created by `to_msgpack` can be successfully parsed by `from_msgpack`.
@@ -203,8 +177,7 @@ written as the reserved negative ext types (255 as -1), so `json::binary({1,2,3,
    JSON-compatible [profile](https://github.com/msgpack/msgpack/blob/master/spec.md#profile) sketched in the
    MessagePack specification, this library restricts map keys to `str` values. Maps with keys of any other type are
    rejected with a [`parse_error.113`](../../home/exceptions.md#jsonexceptionparse_error113) exception (or, with
-    `allow_exceptions` set to `false`, a discarded value). For most key types the message names the type that was
-    found, for instance:
+    `allow_exceptions` set to `false`, a discarded value) naming the type of the key that was found, for instance:

    ```
    [json.exception.parse_error.113] parse error at byte 2: syntax error while parsing MessagePack object key: only string keys are supported, but found nil; last byte: 0xC0
@@ -216,7 +189,7 @@ written as the reserved negative ext types (255 as -1), so `json::binary({1,2,3,
 !!! warning "Ill-formed UTF-8 in string values"

    The MessagePack specification explicitly allows a `str` value (`fixstr`, `str 8`, `str 16`, `str 32`) to contain
-    a byte sequence that is not valid UTF-8, and recommends that a deserializer provide a way to obtain the original bytes.
+    a byte sequence that is not valid UTF-8, and expects a deserializer to hand the original bytes back unchanged.
    This library follows that by default: with its
    [`error_handler`](../../api/basic_json/from_msgpack.md) parameter left at `keep` (the default),
    `from_msgpack()` reads `str` bytes (object keys included) as-is, without validating them, so such a value
@@ -301,8 +301,6 @@ representation using the length of the byte array.
    [`json::from_cbor`](../api/basic_json/from_cbor.md) (see
    [`cbor_tag_handler_t`](../api/basic_json/cbor_tag_handler_t.md)).

-    With `ignore`, the subtype is dropped (`"subtype": null`); with `store`, the tag is kept (`"subtype": 42`):
-
    ```json
    {
      "binary": {
@@ -312,21 +310,11 @@ representation using the length of the byte array.
    }
    ```

-    ```json
-    {
-      "binary": {
-        "bytes": [202, 254, 186, 190],
-        "subtype": 42
-      }
-    }
-    ```
-
 ### MessagePack

 [MessagePack](binary_formats/messagepack.md) supports binary values and subtypes. If a subtype is given, the ext family
-is used. The library will choose the smallest representation among fixext1, fixext2, fixext4, fixext8, fixext16, ext8,
-ext16, and ext32. The subtype is then written as the signed 8-bit ext type: subtypes 128 to 255 map to the negative
-(reserved) ext types, and on reading the ext type is exposed as an unsigned subtype from 0 to 255.
+is used. The library will choose the smallest representation among fixext1, fixext2, fixext4, fixext8, ext8, ext16, and
+ext32. The subtype is then added as a signed 8-bit integer.

 If no subtype is given, the bin family (bin8, bin16, bin32) is used.

@@ -368,7 +368,8 @@ using array_t = ArrayType<basic_json, AllocatorType<basic_json>>;
  `#!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 by the binary readers. The library's internal string concatenation additionally has
+- `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+=`,
  `append(first, last)`, and `append(data, size)`.
 - The comparison operator `==` against another `StringType`, and `<` for use as a key of the chosen
@@ -677,7 +678,8 @@ using binary_t = nlohmann::byte_container_with_subtype<BinaryType>;
  [`StringType`](#stringtype), whose non-`#!cpp const` `data()` is C++17 and later only.
 - Default-constructible, copy-constructible, and move-constructible.
 - Member functions `size()`, `empty()`, `data()`, `resize()`, `operator[]`, `back()`, `begin()`, `end()`, `cbegin()`,
-  and `cend()` with random-access iterators. `push_back()` is **not** required.
+  and `cend()` with random-access iterators, and `insert(pos, first, last)`, which the CBOR reader uses to join the
+  chunks of an indefinite-length byte string. `push_back()` is **not** required.
 - Comparison operators: `==` is used by
  [`byte_container_with_subtype`](../../api/byte_container_with_subtype/index.md), the relational operators by
  [`basic_json`'s comparison operators](../../api/basic_json/operator_le.md).
@@ -1153,7 +1153,7 @@ A JSON Patch `move` operation's `"from"` location is a proper prefix of its `"pa
 ### json.exception.out_of_range.415

 MessagePack's ext type and BSON's binary subtype are each stored in a single byte. This exception is thrown when serializing a
-[`byte_container_with_subtype`](../api/byte_container_with_subtype/index.md) whose subtype exceeds 255, the largest value the library can store in that byte.
+[`byte_container_with_subtype`](../api/byte_container_with_subtype/index.md) whose subtype exceeds 255.

 !!! failure "Example message"

@@ -1218,10 +1218,6 @@ class binary_reader
        // level as required by RFC 8949, Section 3.2.3
        bool indefinite = false;

-        // the chunk currently being validated; only used for the chunks of an
-        // indefinite-length string when the UTF-8 check can do something
-        string_t chunk;
-
        while (true)
        {
            if (JSON_HEDLEY_UNLIKELY(!unexpect_eof("string")))
@@ -1241,82 +1237,26 @@ class binary_reader
            }

            // a break marker closes the indefinite-length string; outside
-            // of one it falls through to the error below. Every chunk has
-            // been checked on its own, so there is nothing left to check.
+            // of one it falls through to the error below
            if (indefinite && current == 0xFF)
            {
-                return true;
+                return check_string_utf8(result, context);
+            }
+
+            if (JSON_HEDLEY_UNLIKELY(!get_cbor_string_chunk(result, indefinite)))
+            {
+                return false;
            }

            if (!indefinite)
            {
-                return get_cbor_string_chunk(result, false) && check_string_utf8(result, context);
-            }
-
-            if (error_handler == error_handler_t::keep)
-            {
-                // nothing to check: read the chunk directly into the result
-                if (JSON_HEDLEY_UNLIKELY(!get_cbor_string_chunk(result, true)))
-                {
-                    return false;
-                }
-            }
-            else
-            {
-                // RFC 8949, Section 3.2.3: each chunk is a text string of its
-                // own and must be valid UTF-8 by itself, i.e., a code point
-                // cannot be spread over two chunks
-                chunk.clear();
-                if (JSON_HEDLEY_UNLIKELY(!get_cbor_string_chunk(chunk, true) || !check_string_utf8(chunk, context)))
-                {
-                    return false;
-                }
-                result.append(chunk.data(), chunk.size());
+                return check_string_utf8(result, context);
            }

            get();
        }
    }

-    /*!
-    @brief skips the tags in front of a CBOR object key
-
-    A tag may wrap any data item, so it may also wrap a key. Like in front of a
-    value, the tags are skipped for the handlers ignore and store; with the
-    handler error they are left to @ref get_cbor_object_key, which rejects the
-    key. The tag loop is iterative, so the number of tags is not limited by the
-    recursion depth.
-
-    @param[in] tag_handler  how CBOR tags should be treated
-
-    @return whether the tags could be skipped (false on EOF in a tag argument)
-
-    @pre @a current is the first byte of the key
-    */
-    bool skip_cbor_key_tags(const cbor_tag_handler_t tag_handler)
-    {
-        if (tag_handler == cbor_tag_handler_t::error)
-        {
-            return true;
-        }
-
-        // 0xC0..0xDB: tag with the tag number in the head (0xC0..0xD7) or in
-        // the 1, 2, 4, or 8 bytes that follow (0xD8..0xDB)
-        while (current >= 0xC0 && current <= 0xDB)
-        {
-            if (current >= 0xD8)
-            {
-                std::uint64_t tag_to_ignore{};
-                if (JSON_HEDLEY_UNLIKELY(!get_cbor_argument(tag_to_ignore)))
-                {
-                    return false;
-                }
-            }
-            get();
-        }
-        return true;
-    }
-
    /*!
    @brief reads a CBOR object key

@@ -1683,10 +1623,6 @@ class binary_reader

                if (top.is_object)
                {
-                    if (JSON_HEDLEY_UNLIKELY(!skip_cbor_key_tags(tag_handler)))
-                    {
-                        return false;
-                    }
                    if (JSON_HEDLEY_UNLIKELY(!read_object_key<&binary_reader::get_cbor_object_key>(key)))
                    {
                        return false;
@@ -4194,9 +4130,9 @@ class binary_reader
        // sequence), UBJSON, BJData, or BSON requires a decoder to reject
        // ill-formed UTF-8. Checking (and, with @ref error_handler_t::strict,
        // rejecting, or with `replace`/`ignore`, sanitizing) is opt-in via
-        // @ref error_handler, applied by @ref check_string_utf8 at the call
-        // site to the whole string or, for an indefinite-length CBOR string,
-        // to each chunk on its own (RFC 8949, Section 3.2.3).
+        // @ref error_handler, applied once the whole string (all chunks of
+        // an indefinite-length CBOR string included) has been assembled, by
+        // @ref check_string_utf8 at the call site.
        return get_bytes(len, "string", result);
    }

@@ -255,6 +255,81 @@ struct number_token
    bool is_float; ///< a fraction or an exponent
 };

+/// advance over ASCII digits, eight bytes at a time
+inline const unsigned char* digit_run(const unsigned char* p, const unsigned char* e) noexcept
+{
+    constexpr std::uint64_t ones = 0x0101010101010101ull;
+    while (e - p >= 8)
+    {
+        // (the digits become 0..9; a byte above 9, or with bit 7 set, is no digit. A carry
+        // out of a byte with bit 7 set only affects the bytes above, after the first one
+        // that count_trailing_zeros finds.)
+        const std::uint64_t x = read_eight_bytes(p) ^ (ones * 0x30u);
+        const std::uint64_t not_digit = ((x + (ones * 0x76u)) | x) & (ones * 0x80u);
+        if (not_digit != 0)
+        {
+            return p + (static_cast<std::size_t>(count_trailing_zeros(not_digit)) / 8);
+        }
+        p += 8;
+    }
+    while (p != e && is_digit(*p))
+    {
+        ++p;
+    }
+    return p;
+}
+
+/// whether the n bytes at p are ASCII digits (n > 0)
+inline bool all_digits(const unsigned char* p, std::size_t n) noexcept
+{
+    const auto digits = [](const unsigned char* q)
+    {
+        constexpr std::uint64_t ones = 0x0101010101010101ull;
+        const std::uint64_t x = read_eight_bytes(q) ^ (ones * 0x30u);
+        return (((x + (ones * 0x76u)) | x) & (ones * 0x80u)) == 0;
+    };
+    if (n >= 8)
+    {
+        for (std::size_t i = 0; i + 8 < n; i += 8)
+        {
+            if (!digits(p + i))
+            {
+                return false;
+            }
+        }
+        return digits(p + n - 8); // (overlapping the bytes before: they are digits)
+    }
+    for (std::size_t i = 0; i < n; ++i)
+    {
+        if (!is_digit(p[i]))
+        {
+            return false;
+        }
+    }
+    return true;
+}
+
+/// Whether a float token has the shape that the parser records for most of
+/// them: [-]digits.digits, with the numbers of digits of the node's layout
+/// (extra). Then the token is a number of that layout, and the general scan
+/// that follows is not needed.
+inline bool plain_float_token(const unsigned char* s, std::size_t len, std::uint16_t extra) noexcept
+{
+    const std::size_t int_digits = extra & 0xFFu;
+    const std::size_t frac_digits = extra >> 8u;
+    if (int_digits == 0 || int_digits >= 255 || frac_digits == 0 || frac_digits >= 255 || len == 0)
+    {
+        return false;
+    }
+    const std::size_t negative = *s == '-' ? 1 : 0;
+    if (len != negative + int_digits + 1 + frac_digits)
+    {
+        return false;
+    }
+    const unsigned char* const p = s + negative;
+    return !(int_digits > 1 && *p == '0') && p[int_digits] == '.' && all_digits(p, int_digits) && all_digits(p + int_digits + 1, frac_digits);
+}
+
 /// whether [s, s + len) is a JSON number (the grammar the parser accepts)
 inline bool scan_number_token(const unsigned char* s, std::size_t len, number_token& t)
 {
@@ -273,11 +348,7 @@ inline bool scan_number_token(const unsigned char* s, std::size_t len, number_to
    }
    else if (*p >= '1' && *p <= '9')
    {
-        do
-        {
-            ++p;
-        }
-        while (p != e && is_digit(*p));
+        p = digit_run(p + 1, e);
    }
    else
    {
@@ -289,10 +360,7 @@ inline bool scan_number_token(const unsigned char* s, std::size_t len, number_to
    if (p != e && *p == '.')
    {
        const unsigned char* const f0 = ++p;
-        while (p != e && is_digit(*p))
-        {
-            ++p;
-        }
+        p = digit_run(p, e);
        if (p == f0)
        {
            return false;
@@ -403,10 +471,15 @@ struct float_range
 template<typename Check>
 bool check_distinct_ranges(std::vector<byte_range>& ranges, Check check)
 {
-    std::sort(ranges.begin(), ranges.end(), [](const byte_range & a, const byte_range & b)
+    // (save() writes the ranges in the order of the nodes, which is sorted already)
+    const auto before = [](const byte_range & a, const byte_range & b)
    {
        return a.off != b.off ? a.off < b.off : a.len < b.len;
-    });
+    };
+    if (!std::is_sorted(ranges.begin(), ranges.end(), before))
+    {
+        std::sort(ranges.begin(), ranges.end(), before);
+    }
    std::size_t end = 0;
    for (std::size_t i = 0; i < ranges.size(); ++i)
    {
@@ -428,7 +501,8 @@ bool check_distinct_ranges(std::vector<byte_range>& ranges, Check check)
 /// record layouts that match it), tokens must not overlap.
 inline bool check_float_ranges(std::vector<float_range>& ranges, const unsigned char* text)
 {
-    std::sort(ranges.begin(), ranges.end(), [](const float_range & a, const float_range & b)
+    // (save() writes the tokens in the order of the nodes, which is sorted already)
+    const auto before = [](const float_range & a, const float_range & b)
    {
        if (a.off != b.off)
        {
@@ -439,7 +513,11 @@ inline bool check_float_ranges(std::vector<float_range>& ranges, const unsigned
            return a.len < b.len;
        }
        return a.extra < b.extra;
-    });
+    };
+    if (!std::is_sorted(ranges.begin(), ranges.end(), before))
+    {
+        std::sort(ranges.begin(), ranges.end(), before);
+    }
    std::size_t end = 0;
    std::size_t i = 0;
    while (i < ranges.size())
@@ -449,14 +527,18 @@ inline bool check_float_ranges(std::vector<float_range>& ranges, const unsigned
        {
            return false;
        }
-        number_token t{};
        const unsigned char* const s = text + r.off;
+        std::uint16_t layout = r.extra;
+        if (!plain_float_token(s, r.len, r.extra))
+        {
+            number_token t{};
            if (!scan_number_token(s, r.len, t) || !float_token_finite(s, r.len, t))
            {
                return false;
            }
            // the digit layout the parser records (or "many", as compaction writes it)
-        const std::uint16_t layout = float_layout(t);
+            layout = float_layout(t);
+        }
        for (; i < ranges.size() && ranges[i].off == r.off && ranges[i].len == r.len; ++i)
        {
            if (ranges[i].extra != layout && ranges[i].extra != 0xFFFFu)
@@ -743,8 +743,12 @@ class basic_json_view
    template<typename, bool> friend class basic_json_document;
    template<typename, bool> friend class basic_json_view;
    template<typename, typename> friend class detail::view::editor;
+    template<typename, typename> friend class detail::view::view_side;
    friend iterator;

+    /// whether the view is one of an editable document
+    static constexpr bool is_editable = Editable;
+
    basic_json_view(const document_data* d, const node* n) noexcept
        : m_doc(d), m_node(n)
    {}
@@ -18685,10 +18685,6 @@ class binary_reader
        // level as required by RFC 8949, Section 3.2.3
        bool indefinite = false;

-        // the chunk currently being validated; only used for the chunks of an
-        // indefinite-length string when the UTF-8 check can do something
-        string_t chunk;
-
        while (true)
        {
            if (JSON_HEDLEY_UNLIKELY(!unexpect_eof("string")))
@@ -18708,82 +18704,26 @@ class binary_reader
            }

            // a break marker closes the indefinite-length string; outside
-            // of one it falls through to the error below. Every chunk has
-            // been checked on its own, so there is nothing left to check.
+            // of one it falls through to the error below
            if (indefinite && current == 0xFF)
            {
-                return true;
+                return check_string_utf8(result, context);
+            }
+
+            if (JSON_HEDLEY_UNLIKELY(!get_cbor_string_chunk(result, indefinite)))
+            {
+                return false;
            }

            if (!indefinite)
            {
-                return get_cbor_string_chunk(result, false) && check_string_utf8(result, context);
-            }
-
-            if (error_handler == error_handler_t::keep)
-            {
-                // nothing to check: read the chunk directly into the result
-                if (JSON_HEDLEY_UNLIKELY(!get_cbor_string_chunk(result, true)))
-                {
-                    return false;
-                }
-            }
-            else
-            {
-                // RFC 8949, Section 3.2.3: each chunk is a text string of its
-                // own and must be valid UTF-8 by itself, i.e., a code point
-                // cannot be spread over two chunks
-                chunk.clear();
-                if (JSON_HEDLEY_UNLIKELY(!get_cbor_string_chunk(chunk, true) || !check_string_utf8(chunk, context)))
-                {
-                    return false;
-                }
-                result.append(chunk.data(), chunk.size());
+                return check_string_utf8(result, context);
            }

            get();
        }
    }

-    /*!
-    @brief skips the tags in front of a CBOR object key
-
-    A tag may wrap any data item, so it may also wrap a key. Like in front of a
-    value, the tags are skipped for the handlers ignore and store; with the
-    handler error they are left to @ref get_cbor_object_key, which rejects the
-    key. The tag loop is iterative, so the number of tags is not limited by the
-    recursion depth.
-
-    @param[in] tag_handler  how CBOR tags should be treated
-
-    @return whether the tags could be skipped (false on EOF in a tag argument)
-
-    @pre @a current is the first byte of the key
-    */
-    bool skip_cbor_key_tags(const cbor_tag_handler_t tag_handler)
-    {
-        if (tag_handler == cbor_tag_handler_t::error)
-        {
-            return true;
-        }
-
-        // 0xC0..0xDB: tag with the tag number in the head (0xC0..0xD7) or in
-        // the 1, 2, 4, or 8 bytes that follow (0xD8..0xDB)
-        while (current >= 0xC0 && current <= 0xDB)
-        {
-            if (current >= 0xD8)
-            {
-                std::uint64_t tag_to_ignore{};
-                if (JSON_HEDLEY_UNLIKELY(!get_cbor_argument(tag_to_ignore)))
-                {
-                    return false;
-                }
-            }
-            get();
-        }
-        return true;
-    }
-
    /*!
    @brief reads a CBOR object key

@@ -19150,10 +19090,6 @@ class binary_reader

                if (top.is_object)
                {
-                    if (JSON_HEDLEY_UNLIKELY(!skip_cbor_key_tags(tag_handler)))
-                    {
-                        return false;
-                    }
                    if (JSON_HEDLEY_UNLIKELY(!read_object_key<&binary_reader::get_cbor_object_key>(key)))
                    {
                        return false;
@@ -21661,9 +21597,9 @@ class binary_reader
        // sequence), UBJSON, BJData, or BSON requires a decoder to reject
        // ill-formed UTF-8. Checking (and, with @ref error_handler_t::strict,
        // rejecting, or with `replace`/`ignore`, sanitizing) is opt-in via
-        // @ref error_handler, applied by @ref check_string_utf8 at the call
-        // site to the whole string or, for an indefinite-length CBOR string,
-        // to each chunk on its own (RFC 8949, Section 3.2.3).
+        // @ref error_handler, applied once the whole string (all chunks of
+        // an indefinite-length CBOR string included) has been assembled, by
+        // @ref check_string_utf8 at the call site.
        return get_bytes(len, "string", result);
    }

@@ -1807,93 +1807,6 @@ TEST_CASE("CBOR")
            CHECK(json::from_cbor(std::vector<uint8_t>({0xa1, 0xff, 0x01}), true, false).is_discarded());
        }

-        SECTION("tag in front of a key")
-        {
-            const auto ignore = json::cbor_tag_handler_t::ignore;
-            const auto store = json::cbor_tag_handler_t::store;
-            const json expected = {{"a", 10}};
-
-            // the tags are skipped in front of a text string key, like in
-            // front of a value
-            const std::vector<std::vector<std::uint8_t>> tagged =
-            {
-                {0xA1, 0xC0, 0x61, 0x61, 0x0A},                         // tag in the head
-                {0xA1, 0xD8, 0x05, 0x61, 0x61, 0x0A},                   // 1-byte tag number
-                {0xA1, 0xD9, 0x01, 0x2C, 0x61, 0x61, 0x0A},             // 2-byte tag number
-                {0xA1, 0xDA, 0x00, 0x01, 0x00, 0x00, 0x61, 0x61, 0x0A}, // 4-byte tag number
-                {0xA1, 0xDB, 0x00, 0x00, 0x00, 0x01, 0x00, 0x00, 0x00, 0x00, 0x61, 0x61, 0x0A}, // 8-byte tag number
-                {0xA1, 0xC0, 0xD8, 0x05, 0xC1, 0x61, 0x61, 0x0A},       // chained tags
-                {0xBF, 0xC0, 0x61, 0x61, 0x0A, 0xFF},                   // indefinite-length map
-                {0xBF, 0x61, 0x62, 0x01, 0xC0, 0x61, 0x61, 0x0A, 0xFF}, // second key tagged
-                {0xA1, 0xC0, 0x7F, 0x61, 0x61, 0xFF, 0x0A},             // indefinite-length text key
-                {0xA1, 0xC0, 0x61, 0x61, 0xC0, 0x0A},                   // tagged key and value
-            };
-
-            for (const auto& input : tagged)
-            {
-                CAPTURE(input)
-                json const j_ignore = json::from_cbor(input, true, true, ignore);
-                json const j_store = json::from_cbor(input, true, true, store);
-                CHECK(j_ignore.at("a") == 10);
-                CHECK(j_store.at("a") == 10);
-                CHECK(j_ignore == j_store);
-            }
-            CHECK(json::from_cbor(tagged[0], true, true, ignore) == expected);
-            CHECK(json::from_cbor(tagged[1], true, true, store) == expected);
-
-            // nested maps
-            CHECK(json::from_cbor(std::vector<std::uint8_t>({0xA1, 0xC0, 0x61, 0x61, 0xA1, 0xD8, 0x05, 0x61, 0x62, 0x01}), true, true, ignore) == json{{"a", {{"b", 1}}}});
-
-            // the tag handler error still rejects every tag
-            for (const auto& input : tagged)
-            {
-                CAPTURE(input)
-                json _;
-                CHECK_THROWS_AS(_ = json::from_cbor(input), json::parse_error&);
-            }
-            {
-                json _;
-                CHECK_THROWS_WITH_AS(_ = json::from_cbor(std::vector<std::uint8_t>({0xA1, 0xD8, 0x05, 0x61, 0x61, 0x0A})), "[json.exception.parse_error.113] parse error at byte 2: syntax error while parsing CBOR object key: only string keys are supported, but found a tag; last byte: 0xD8", json::parse_error&);
-            }
-
-            // a tag does not turn a key of another type into a text string
-            {
-                json _;
-                CHECK_THROWS_WITH_AS(_ = json::from_cbor(std::vector<std::uint8_t>({0xA1, 0xC0, 0x01, 0x01}), true, true, ignore), "[json.exception.parse_error.113] parse error at byte 3: syntax error while parsing CBOR object key: only string keys are supported, but found an unsigned integer; last byte: 0x01", json::parse_error&);
-                CHECK_THROWS_WITH_AS(_ = json::from_cbor(std::vector<std::uint8_t>({0xA1, 0xD8, 0x05, 0x41, 0x61, 0x01}), true, true, store), "[json.exception.parse_error.113] parse error at byte 4: syntax error while parsing CBOR object key: only string keys are supported, but found a byte string; last byte: 0x41", json::parse_error&);
-                CHECK(json::from_cbor(std::vector<std::uint8_t>({0xA1, 0xC0, 0x01, 0x01}), true, false, ignore).is_discarded());
-            }
-
-            // end of input after the tag, or inside the tag number
-            {
-                json _;
-                CHECK_THROWS_WITH_AS(_ = json::from_cbor(std::vector<std::uint8_t>({0xA1, 0xC0}), true, true, ignore), "[json.exception.parse_error.110] parse error at byte 3: syntax error while parsing CBOR string: unexpected end of input", json::parse_error&);
-                CHECK_THROWS_AS(_ = json::from_cbor(std::vector<std::uint8_t>({0xA1, 0xD8}), true, true, ignore), json::parse_error&);
-                CHECK_THROWS_AS(_ = json::from_cbor(std::vector<std::uint8_t>({0xA1, 0xDB, 0x00, 0x00}), true, true, store), json::parse_error&);
-                CHECK_THROWS_AS(_ = json::from_cbor(std::vector<std::uint8_t>({0xBF, 0xC0}), true, true, store), json::parse_error&);
-            }
-
-            // SAX: the key event is reported without the tag (start_object,
-            // key, number_unsigned, end_object are four events)
-            {
-                const std::vector<std::uint8_t> input = {0xA1, 0xC0, 0x61, 0x61, 0x0A};
-                SaxCountdown four(4);
-                CHECK(json::sax_parse(input, &four, json::input_format_t::cbor, true, false, false, ignore));
-                SaxCountdown three(3);
-                CHECK(!json::sax_parse(input, &three, json::input_format_t::cbor, true, false, false, ignore));
-                SaxCountdown error_handler(4);
-                CHECK(!json::sax_parse(input, &error_handler, json::input_format_t::cbor));
-            }
-
-            // a long chain of tags needs no recursion
-            {
-                std::vector<std::uint8_t> input = {0xA1};
-                input.insert(input.end(), 100000, 0xC0);
-                input.insert(input.end(), {0x61, 0x61, 0x0A});
-                CHECK(json::from_cbor(input, true, true, ignore) == expected);
-            }
-        }
-
        SECTION("non-string key (see #2766 and #3381)")
        {
            // only text strings map to JSON object keys; any other key is
@@ -2002,10 +1915,9 @@ TEST_CASE("CBOR")
        {
            json _;

-            // with the default error handler keep, the chunks are concatenated
-            // as is, without checking that each chunk is valid UTF-8 on its
-            // own (RFC 8949, Section 3.2.3), so a code point split across two
-            // chunks yields a valid string
+            // the chunks are concatenated as is, without checking that each
+            // chunk is valid UTF-8 on its own (RFC 8949, Section 3.2.3), so
+            // a code point split across two chunks yields a valid string
            CHECK_NOTHROW(_ = json::from_cbor(std::vector<uint8_t>({0x7f, 0x61, 0xc3, 0x61, 0xa9, 0xff})));
            CHECK(_ == "\xc3\xa9");
            CHECK(_.dump() == "\"\xc3\xa9\"");
@@ -2025,63 +1937,6 @@ TEST_CASE("CBOR")
            CHECK(json::from_cbor(std::vector<uint8_t>({0x7f, 0x62, 0xc3, 0xa9, 0x62, 0xc3, 0xb6, 0xff})) == "\xc3\xa9\xc3\xb6");
        }

-        SECTION("ill-formed UTF-8 in a chunk of an indefinite-length string (RFC 8949, Section 3.2.3)")
-        {
-            const auto strict = json::error_handler_t::strict;
-            const auto replace = json::error_handler_t::replace;
-            const auto ignore = json::error_handler_t::ignore;
-            const auto keep = json::error_handler_t::keep;
-            const auto tags = json::cbor_tag_handler_t::error;
-
-            // a code point split over two chunks
-            const std::vector<std::uint8_t> split = {0x7f, 0x61, 0xc3, 0x61, 0xa9, 0xff};
-            // a 4-byte code point split over two chunks
-            const std::vector<std::uint8_t> split_emoji = {0x7f, 0x62, 0xf0, 0x9f, 0x62, 0x98, 0x80, 0xff};
-            // the same in an object key
-            const std::vector<std::uint8_t> split_key = {0xa1, 0x7f, 0x61, 0xc3, 0x61, 0xa9, 0xff, 0x01};
-            // a valid chunk followed by one with a truncated code point
-            const std::vector<std::uint8_t> truncated = {0x7f, 0x62, 0x41, 0xc3, 0x61, 0xa9, 0xff};
-
-            // strict: the string is invalid
-            for (const auto& input :
-                    {
-                        split, split_emoji, split_key, truncated
-                    })
-            {
-                CAPTURE(input)
-                json _;
-                CHECK_THROWS_AS(_ = json::from_cbor(input, true, true, tags, strict), json::parse_error&);
-                CHECK(json::from_cbor(input, true, false, tags, strict).is_discarded());
-            }
-            {
-                json _;
-                CHECK_THROWS_WITH_AS(_ = json::from_cbor(split, true, true, tags, strict), "[json.exception.parse_error.113] parse error at byte 3: syntax error while parsing CBOR string: invalid string: ill-formed UTF-8 byte", json::parse_error&);
-                CHECK_THROWS_WITH_AS(_ = json::from_cbor(split_key, true, true, tags, strict), "[json.exception.parse_error.113] parse error at byte 4: syntax error while parsing CBOR key: invalid string: ill-formed UTF-8 byte", json::parse_error&);
-            }
-
-            // replace and ignore: every chunk is sanitized on its own
-            CHECK(json::from_cbor(split, true, true, tags, replace) == "\xEF\xBF\xBD\xEF\xBF\xBD");
-            CHECK(json::from_cbor(split, true, true, tags, ignore) == "");
-            CHECK(json::from_cbor(truncated, true, true, tags, replace) == "A\xEF\xBF\xBD\xEF\xBF\xBD");
-            CHECK(json::from_cbor(truncated, true, true, tags, ignore) == "A");
-            CHECK(json::from_cbor(split_key, true, true, tags, replace) == json{{"\xEF\xBF\xBD\xEF\xBF\xBD", 1}});
-
-            // keep: unchanged, the chunks are concatenated as is
-            CHECK(json::from_cbor(split, true, true, tags, keep) == "\xc3\xa9");
-            CHECK(json::from_cbor(split_emoji, true, true, tags, keep) == "\xf0\x9f\x98\x80");
-
-            // conforming input is unaffected by the error handler
-            const std::vector<std::uint8_t> valid = {0x7f, 0x62, 0xc3, 0xa9, 0x64, 0xf0, 0x9f, 0x98, 0x80, 0x60, 0xff};
-            for (const auto handler :
-                    {
-                        strict, replace, ignore, keep
-                    })
-            {
-                CHECK(json::from_cbor(valid, true, true, tags, handler) == "\xc3\xa9\xf0\x9f\x98\x80");
-                CHECK(json::from_cbor(std::vector<std::uint8_t>({0x7f, 0xff}), true, true, tags, handler) == "");
-            }
-        }
-
        SECTION("many chunks in indefinite-length string")
        {
            constexpr std::size_t chunks = 100000;
@@ -0,0 +1,364 @@
+//     __ _____ _____ _____
+//  __|  |   __|     |   | |  JSON for Modern C++ (supporting code)
+// |  |  |__   |  |  | | | |  version 3.12.0
+// |_____|_____|_____|_|___|  https://github.com/nlohmann/json
+//
+// SPDX-FileCopyrightText: 2013-2026 Niels Lohmann <https://nlohmann.me>
+// SPDX-License-Identifier: MIT
+
+#include "doctest_compatibility.h"
+
+#include <nlohmann/json_view.hpp>
+using nlohmann::json;
+using nlohmann::ordered_json;
+using nlohmann::json_document;
+using nlohmann::ordered_json_editable_document;
+using nlohmann::json_view;
+using nlohmann::ordered_json_document;
+
+#include "json_view_test_helpers.hpp"
+using json_view_test::generator;
+
+#include <array>
+#include <cstddef>
+#include <string>
+#include <vector>
+
+// These tests were split off unit-json_view_dump.cpp, whose object file got too
+// large for the MinGW assembler ("file too big"). Like that file, this one is
+// also built with C++17, as it mentions JSON_HAS_CPP_17.
+
+TEST_CASE("json_view comparison")
+{
+    SECTION("equality of the values parse() produces")
+    {
+        generator g;
+        std::vector<std::string> texts;
+        for (int i = 0; i < 600; ++i)
+        {
+            std::string text;
+            g.value(text, 0);
+            texts.push_back(text);
+            // the same value written differently: sorted keys, canonical numbers
+            texts.push_back(json::parse(text).dump(1));
+        }
+        for (std::size_t i = 0; i + 2 < texts.size(); ++i)
+        {
+            for (std::size_t k = i; k < i + 3; ++k)
+            {
+                CAPTURE(texts[i])
+                CAPTURE(texts[k])
+                const json_document a = json_document::parse(texts[i]);
+                const json_document b = json_document::parse(texts[k]);
+                const json ja = json::parse(texts[i]);
+                const json jb = json::parse(texts[k]);
+                CHECK((a.root() == b.root()) == (ja == jb));
+                CHECK((a.root() != b.root()) == (ja != jb));
+                CHECK((a.root() == jb) == (ja == jb));
+                CHECK((jb == a.root()) == (ja == jb));
+                CHECK((a.root() != jb) == (ja != jb));
+                CHECK((jb != a.root()) == (ja != jb));
+
+                // ordered_json compares members in order
+                const ordered_json_document oa = ordered_json_document::parse(texts[i]);
+                const ordered_json_document ob = ordered_json_document::parse(texts[k]);
+                const ordered_json oja = ordered_json::parse(texts[i]);
+                const ordered_json ojb = ordered_json::parse(texts[k]);
+                CHECK((oa.root() == ob.root()) == (oja == ojb));
+                CHECK((oa.root() == ojb) == (oja == ojb));
+            }
+        }
+    }
+
+    SECTION("objects whose first keys differ or repeat")
+    {
+        const std::array<const char*, 17> texts =
+        {
+            {
+                R"({"a":1,"b":2})", R"({"a":1,"c":2})", R"({"b":1,"a":2})", R"({"a":1,"a":2,"b":3})", R"({"a":2,"b":3})",
+                R"({"a":1,"b":1,"a":2})", R"({"a":2,"b":1})", R"({"a":1,"a":2})", R"({"b":1})", R"({"a":1,"b":2,"c":3,"d":4,"e":5})",
+                R"({"a":1,"b":2,"c":3,"d":4,"f":5})", R"({"a":1,"b":2,"c":3,"a":4,"e":5})", R"({"a":4,"b":2,"c":3,"e":5})",
+                R"({"\u0061":1,"b":2})", R"({})", R"([{"a":1,"b":2}])", R"([{"a":1,"c":2}])"
+            }
+        };
+        for (const char* const x : texts)
+        {
+            for (const char* const y : texts)
+            {
+                CAPTURE(x)
+                CAPTURE(y)
+                const ordered_json_document dx = ordered_json_document::parse(x);
+                const ordered_json_document dy = ordered_json_document::parse(y);
+                const ordered_json_editable_document ex = ordered_json_editable_document::parse(x);
+                const ordered_json jx = ordered_json::parse(x);
+                const ordered_json jy = ordered_json::parse(y);
+                CHECK((dx.root() == dy.root()) == (jx == jy));
+                CHECK((dx.root() == jy) == (jx == jy));
+                CHECK((ex.root() == dy.root()) == (jx == jy));
+            }
+        }
+    }
+
+    SECTION("objects with duplicate keys, small and large")
+    {
+        // n members with keys in descending order; with duplicates, some keys
+        // repeat later (the first after unique ones in key order); in reverse,
+        // the members are in ascending key order
+        const auto object = [](int n, bool duplicates, bool reverse, int value)
+        {
+            std::string s = "{";
+            for (int i = 0; i < n; ++i)
+            {
+                const int k = reverse ? i : n - 1 - i;
+                s += (i != 0 ? ",\"k" : "\"k") + std::to_string(100 + k) + "\":" + std::to_string(value + k);
+            }
+            if (duplicates)
+            {
+                for (int k = n / 2; k < n; k += 3)
+                {
+                    s += ",\"k" + std::to_string(100 + k) + "\":" + std::to_string(value + k);
+                }
+            }
+            return s + "}";
+        };
+        std::vector<std::string> texts;
+        for (const int n :
+                {
+                    3, 16, 17, 40
+                })
+        {
+            for (const int flags :
+                    {
+                        0, 1, 2, 3
+                    })
+            {
+                texts.push_back(object(n, (flags & 1) != 0, (flags & 2) != 0, 0));
+            }
+            texts.push_back(object(n, true, false, 1)); // other values
+        }
+        // many objects in one comparison (working memory reused across objects)
+        std::string array = "[";
+        std::string array_other = "[";
+        for (std::size_t i = 0; i < texts.size(); ++i)
+        {
+            array += (i != 0 ? "," : "") + texts[i];
+            array_other += (i != 0 ? "," : "") + texts[texts.size() - 1 - i];
+        }
+        texts.push_back(array + "]");
+        texts.push_back(array_other + "]");
+        for (const std::string& x : texts)
+        {
+            for (const std::string& y : texts)
+            {
+                CAPTURE(x)
+                CAPTURE(y)
+                const ordered_json_document dx = ordered_json_document::parse(x);
+                const ordered_json_document dy = ordered_json_document::parse(y);
+                const ordered_json jx = ordered_json::parse(x);
+                const ordered_json jy = ordered_json::parse(y);
+                CHECK((dx.root() == dy.root()) == (jx == jy));
+                CHECK((dx.root() == jy) == (jx == jy));
+                const json_document ux = json_document::parse(x);
+                const json uy = json::parse(y);
+                CHECK((ux.root() == uy) == (json::parse(x) == uy));
+            }
+        }
+    }
+
+    SECTION("numbers, duplicate keys, member order")
+    {
+        const auto same = [](const char* x, const char* y)
+        {
+            const json_document dx = json_document::parse(x);
+            const json_document dy = json_document::parse(y);
+            return dx.root() == dy.root();
+        };
+        CHECK(same("1", "1.0"));
+        CHECK(same("[1, -1, 2.5]", "[1.0, -1.0, 25e-1]"));
+        CHECK(!same("1", "1.5"));
+        CHECK(same("18446744073709551615", "18446744073709551615"));
+        CHECK(same(R"({"a": 1, "a": 2})", R"({"a": 2})"));
+        CHECK(!same(R"({"a": 1, "a": 2})", R"({"a": 1})"));
+        CHECK(same(R"({"a": 1, "b": 2})", R"({"b": 2, "a": 1})"));
+        CHECK(!same(R"({"a": 1})", R"({"a": 1, "b": 2})"));
+        CHECK(!same("[1, 2]", "[2, 1]"));
+        CHECK(!same("\"a\"", "\"b\""));
+        CHECK(same("\"\\u00e9\"", "\"\xc3\xa9\""));
+        CHECK(!same("null", "false"));
+        CHECK(!same("[]", "{}"));
+        const ordered_json_document dup = ordered_json_document::parse(R"({"a": 1, "b": 2, "a": 3})");
+        const ordered_json_document last = ordered_json_document::parse(R"({"a": 3, "b": 2})");
+        CHECK(dup.root() == last.root());
+        // (== compares as parse() resolves duplicates, a lookup finds the first)
+        CHECK(dup.root()["a"].materialize() == 1);
+        CHECK(dup.root()["a"] != last.root()["a"]);
+        const ordered_json_document ab = ordered_json_document::parse(R"({"a": 1, "b": 2})");
+        const ordered_json_document ba = ordered_json_document::parse(R"({"b": 2, "a": 1})");
+        CHECK(ab.root() != ba.root());
+
+        // discarded values compare as basic_json's do
+        const json discarded(json::value_t::discarded);
+        CHECK((json_view() == json_view()) == (discarded == discarded)); // NOLINT(readability-container-size-empty): operator== is tested
+        CHECK((json_view() == discarded) == (discarded == discarded)); // NOLINT(readability-container-size-empty)
+        const json_document null_document = json_document::parse("null");
+        CHECK(!(json_view() == null_document.root())); // NOLINT(readability-container-size-empty)
+        CHECK(!(null_document.root() == discarded));
+    }
+
+    SECTION("values that are the same text, or the same value written differently, at any depth")
+    {
+        // nested arrays and objects around depths where the walk over equal text gives up
+        for (const int depth :
+                {
+                    1, 5, 47, 48, 49, 50, 60, 300
+                })
+        {
+            CAPTURE(depth)
+            std::string head;
+            std::string tail;
+            for (int i = 0; i < depth; ++i)
+            {
+                head += i % 2 == 0 ? R"({"k":[1,"s",{"x":true},)" : R"([null,2.5,{"y":)";
+                tail.insert(0, i % 2 == 0 ? "]}" : "}]");
+            }
+            // the same text, one number changed, and one number written differently
+            const auto with_number = [&](const char* number)
+            {
+                std::string result = head;
+                result += number;
+                result += tail;
+                return result;
+            };
+            const std::string text = with_number("7");
+            const std::string changed = with_number("8");
+            const std::string spelled = with_number("7.0");
+            const json_document a = json_document::parse(text);
+            const json_document b = json_document::parse(text);
+            const json_document c = json_document::parse(changed);
+            const json_document d = json_document::parse(spelled);
+            CHECK(a.root() == b.root());
+            CHECK(a.root() != c.root());
+            CHECK(c.root() != a.root());
+            CHECK(a.root() == d.root());
+            CHECK(d.root() == a.root());
+            CHECK(a.root() == json::parse(text));
+            CHECK(a.root() != json::parse(changed));
+            CHECK(json::parse(spelled) == a.root());
+            const nlohmann::json_editable_document e = nlohmann::json_editable_document::parse(text);
+            CHECK(a.root() == e.root());
+            CHECK(e.root() == a.root());
+            CHECK(c.root() != e.root());
+            CHECK(e.root() == e.root());
+            const ordered_json_document oa = ordered_json_document::parse(text);
+            const ordered_json_document oc = ordered_json_document::parse(changed);
+            CHECK(oa.root() != oc.root());
+            CHECK(oa.root() == ordered_json::parse(text));
+            CHECK(oc.root() != ordered_json::parse(text));
+        }
+
+        // not the same text, but the same value
+        const auto same = [](const char* x, const char* y)
+        {
+            const json_document dx = json_document::parse(x);
+            const json_document dy = json_document::parse(y);
+            const bool equal = dx.root() == dy.root();
+            CHECK(equal == (json::parse(x) == json::parse(y)));
+            CHECK(equal == (dx.root() == json::parse(y)));
+            return equal;
+        };
+        CHECK(same(R"({"a":1,"a":2})", R"({"a":3,"a":2})"));
+        CHECK(same(R"({"a":1,"b":[1,{"c":2,"c":3}]})", R"({"b":[1,{"c":9,"c":3}],"a":1})"));
+        CHECK(!same(R"({"a":1,"a":2})", R"({"a":2,"a":3})"));
+        CHECK(same("[1.0,-0,0,1e2]", "[1.00,0,-0,100.0]"));
+        CHECK(same(R"(["\u00e9","\n"])", "[\"\xc3\xa9\",\"\\n\"]"));
+        CHECK(!same(R"({"a":[1,2,{"b":null}]})", R"({"a":[1,2,{"b":false}]})"));
+        CHECK(!same(R"({"a":[1,2,{"b":null}]})", R"({"a":[1,2,{"c":null}]})"));
+        CHECK(!same(R"({"a":[1,2,{"b":null}]})", R"({"a":[1,2,{"b":null,"c":1}]})"));
+
+        // edited documents: moved sequences, links, and values of new nodes
+        nlohmann::json_editable_document edited = nlohmann::json_editable_document::parse(R"({"a":[1,2,3],"b":{"c":"d"},"e":0.5})");
+        const json_document plain = json_document::parse(R"({"a":[1,2,3,4],"b":{"c":"x","y":null},"e":0.5,"z":[]})");
+        edited.push_back(edited.root()["a"], 4);
+        edited.set(edited.root()["b"], "c", "x");
+        edited.set(edited.root()["b"], "y", nullptr);
+        edited.set(edited.root(), "z", json::array());
+        CHECK(edited.root() == plain.root());
+        CHECK(plain.root() == edited.root());
+        edited.erase(edited.root()["a"], 0);
+        CHECK(edited.root() != plain.root());
+        CHECK(edited.root()["a"] == json::parse("[2,3,4]"));
+        const json_document tail = json_document::parse("[2,3,4]");
+        CHECK(edited.root()["a"] == tail.root());
+    }
+
+    SECTION("large documents, with a difference at any place")
+    {
+        const auto make = [](std::size_t count, std::size_t changed, bool swap_members)
+        {
+            std::string text = R"({"b":1,"list":[)";
+            for (std::size_t i = 0; i < count; ++i)
+            {
+                text += (i != 0 ? "," : "");
+                const std::string id = std::to_string(i == changed ? i + 1 : i);
+                if (swap_members && i % 3 == 0)
+                {
+                    text += R"({"v":[1.5,"s",{"k":)";
+                    text += id;
+                    text += R"(}],"id":)";
+                    text += id;
+                    text += "}";
+                }
+                else
+                {
+                    text += R"({"id":)";
+                    text += id;
+                    text += R"(,"v":[1.5,"s",{"k":)";
+                    text += id;
+                    text += "}]}";
+                }
+            }
+            return text + R"(],"a":{"y":2,"x":3}})";
+        };
+        const std::size_t count = 3000;
+        for (const std::size_t changed :
+                {
+                    count, std::size_t{0}, std::size_t{1}, count / 2, count - 1
+                })
+        {
+            for (const bool swap :
+                    {
+                        false, true
+                    })
+            {
+                CAPTURE(changed)
+                CAPTURE(swap)
+                const std::string text_a = make(count, count, false);
+                const std::string text_b = make(count, changed, swap);
+                const json_document a = json_document::parse(text_a);
+                const json_document b = json_document::parse(text_b);
+                const bool equal = changed == count;
+                CHECK((a.root() == b.root()) == equal);
+                CHECK((b.root() == a.root()) == equal);
+                CHECK((a.root() == json::parse(text_b)) == equal);
+                CHECK((json::parse(text_b) == a.root()) == equal);
+                CHECK((a.root()["list"] == b.root()["list"]) == equal);
+                const ordered_json_document oa = ordered_json_document::parse(text_a);
+                const ordered_json_document ob = ordered_json_document::parse(text_b);
+                CHECK((oa.root() == ob.root()) == (equal && !swap));
+                CHECK((oa.root() == ordered_json::parse(text_b)) == (equal && !swap));
+            }
+        }
+    }
+
+    SECTION("deep nesting")
+    {
+        const std::string deep = std::string(100000, '[') + std::string(100000, ']');
+        const json_document a = json_document::parse(deep);
+        const json_document b = json_document::parse(deep);
+        CHECK(a.root() == b.root());
+        CHECK(a.root() == json::parse(deep));
+        const std::string other = std::string(100000, '[') + "1" + std::string(100000, ']');
+        const json_document c = json_document::parse(other);
+        CHECK(a.root() != c.root());
+    }
+}
@@ -432,101 +432,6 @@ TEST_CASE("json_view dump")
    }
 }

-TEST_CASE("json_view comparison")
-{
-    SECTION("equality of the values parse() produces")
-    {
-        generator g;
-        std::vector<std::string> texts;
-        for (int i = 0; i < 600; ++i)
-        {
-            std::string text;
-            g.value(text, 0);
-            texts.push_back(text);
-            // the same value written differently: sorted keys, canonical numbers
-            texts.push_back(json::parse(text).dump(1));
-        }
-        for (std::size_t i = 0; i + 2 < texts.size(); ++i)
-        {
-            for (std::size_t k = i; k < i + 3; ++k)
-            {
-                CAPTURE(texts[i])
-                CAPTURE(texts[k])
-                const json_document a = json_document::parse(texts[i]);
-                const json_document b = json_document::parse(texts[k]);
-                const json ja = json::parse(texts[i]);
-                const json jb = json::parse(texts[k]);
-                CHECK((a.root() == b.root()) == (ja == jb));
-                CHECK((a.root() != b.root()) == (ja != jb));
-                CHECK((a.root() == jb) == (ja == jb));
-                CHECK((jb == a.root()) == (ja == jb));
-                CHECK((a.root() != jb) == (ja != jb));
-                CHECK((jb != a.root()) == (ja != jb));
-
-                // ordered_json compares members in order
-                const ordered_json_document oa = ordered_json_document::parse(texts[i]);
-                const ordered_json_document ob = ordered_json_document::parse(texts[k]);
-                const ordered_json oja = ordered_json::parse(texts[i]);
-                const ordered_json ojb = ordered_json::parse(texts[k]);
-                CHECK((oa.root() == ob.root()) == (oja == ojb));
-                CHECK((oa.root() == ojb) == (oja == ojb));
-            }
-        }
-    }
-
-    SECTION("numbers, duplicate keys, member order")
-    {
-        const auto same = [](const char* x, const char* y)
-        {
-            const json_document dx = json_document::parse(x);
-            const json_document dy = json_document::parse(y);
-            return dx.root() == dy.root();
-        };
-        CHECK(same("1", "1.0"));
-        CHECK(same("[1, -1, 2.5]", "[1.0, -1.0, 25e-1]"));
-        CHECK(!same("1", "1.5"));
-        CHECK(same("18446744073709551615", "18446744073709551615"));
-        CHECK(same(R"({"a": 1, "a": 2})", R"({"a": 2})"));
-        CHECK(!same(R"({"a": 1, "a": 2})", R"({"a": 1})"));
-        CHECK(same(R"({"a": 1, "b": 2})", R"({"b": 2, "a": 1})"));
-        CHECK(!same(R"({"a": 1})", R"({"a": 1, "b": 2})"));
-        CHECK(!same("[1, 2]", "[2, 1]"));
-        CHECK(!same("\"a\"", "\"b\""));
-        CHECK(same("\"\\u00e9\"", "\"\xc3\xa9\""));
-        CHECK(!same("null", "false"));
-        CHECK(!same("[]", "{}"));
-        const ordered_json_document dup = ordered_json_document::parse(R"({"a": 1, "b": 2, "a": 3})");
-        const ordered_json_document last = ordered_json_document::parse(R"({"a": 3, "b": 2})");
-        CHECK(dup.root() == last.root());
-        // (== compares as parse() resolves duplicates, a lookup finds the first)
-        CHECK(dup.root()["a"].materialize() == 1);
-        CHECK(dup.root()["a"] != last.root()["a"]);
-        const ordered_json_document ab = ordered_json_document::parse(R"({"a": 1, "b": 2})");
-        const ordered_json_document ba = ordered_json_document::parse(R"({"b": 2, "a": 1})");
-        CHECK(ab.root() != ba.root());
-
-        // discarded values compare as basic_json's do
-        const json discarded(json::value_t::discarded);
-        CHECK((json_view() == json_view()) == (discarded == discarded)); // NOLINT(readability-container-size-empty): operator== is tested
-        CHECK((json_view() == discarded) == (discarded == discarded)); // NOLINT(readability-container-size-empty)
-        const json_document null_document = json_document::parse("null");
-        CHECK(!(json_view() == null_document.root())); // NOLINT(readability-container-size-empty)
-        CHECK(!(null_document.root() == discarded));
-    }
-
-    SECTION("deep nesting")
-    {
-        const std::string deep = std::string(100000, '[') + std::string(100000, ']');
-        const json_document a = json_document::parse(deep);
-        const json_document b = json_document::parse(deep);
-        CHECK(a.root() == b.root());
-        CHECK(a.root() == json::parse(deep));
-        const std::string other = std::string(100000, '[') + "1" + std::string(100000, ']');
-        const json_document c = json_document::parse(other);
-        CHECK(a.root() != c.root());
-    }
-}
-
 TEST_CASE("json_view large objects")
 {
    // objects with 128 members or more are looked up with a hash index
@@ -682,6 +682,64 @@ TEST_CASE("json_view images: ownership")
 #if !defined(JSON_NOEXCEPTION)
 TEST_CASE("json_view images: errors")
 {
+    SECTION("float tokens: a changed byte is accepted if the text is still a number with the digits the node records")
+    {
+        for (const char* const token :
+                {
+                    "-12.375", "0.5", "123456.0", "-0.0", "9999999999999999999.5", "0.000000000000000000000001", "1.25e10", "5E-3"
+                })
+        {
+            CAPTURE(token)
+            const std::string text = std::string("[") + token + "]";
+            const json_document d = json_document::parse(text);
+            const std::vector<std::uint8_t> image = d.save();
+            const std::size_t at = text_at(image) + 1; // the first byte of the token
+            const std::uint16_t recorded = node_at(image, 1).extra;
+            CHECK(loaded_dump(image) == json::parse(text).dump());
+            for (std::size_t pos = 0; token[pos] != '\0'; ++pos)
+            {
+                for (unsigned int byte = 0; byte < 256; ++byte)
+                {
+                    std::vector<std::uint8_t> changed = image;
+                    changed[at + pos] = static_cast<std::uint8_t>(byte);
+                    std::string t = token;
+                    t[pos] = static_cast<char>(byte);
+                    // the digits of the integer part and of the fraction of the changed token
+                    std::size_t integer = 0;
+                    std::size_t fraction = 0;
+                    std::size_t k = (t[0] == '-') ? 1 : 0;
+                    for (; k < t.size() && t[k] >= '0' && t[k] <= '9'; ++k)
+                    {
+                        ++integer;
+                    }
+                    if (k < t.size() && t[k] == '.')
+                    {
+                        for (++k; k < t.size() && t[k] >= '0' && t[k] <= '9'; ++k)
+                        {
+                            ++fraction;
+                        }
+                    }
+                    // (as a JSON text, the token may have white space around it)
+                    const bool float_token = json::accept(t) && t.find_first_of(std::string(" \t\r\n\0", 5)) == std::string::npos && (t.find('.') != std::string::npos || t.find_first_of("eE") != std::string::npos) // NOLINT(abseil-string-find-str-contains)
+                                             && ((integer | (fraction << 8U)) == recorded);
+                    CAPTURE(pos)
+                    CAPTURE(byte)
+                    std::string result;
+                    try
+                    {
+                        result = loaded_dump(changed);
+                    }
+                    catch (const json::parse_error& e)
+                    {
+                        CHECK(std::string(e.what()) == check_failed);
+                        result = "rejected";
+                    }
+                    CHECK((result != "rejected") == float_token);
+                }
+            }
+        }
+    }
+
    SECTION("invalid UTF-8 of a decoded string: the compact dump writes it, the others throw")
    {
        const json_document d = json_document::parse(R"({"k\n":"\u00e9x"})");
Author	SHA1	Message	Date
Niels Lohmann	eed5792dbd	json_view: no per-object allocations in ordered_json comparisons For an object type that keeps the order of its members (ordered_json), resolve_indices() allocated a vector per compared object, and resolve_members() two. They now use one working vector owned by equal(), whose capacity is reused, and resolve_indices() needs none when the keys are unique. resolve_members() sorts small objects by insertion, as the other paths do. Instructions per comparison (callgrind, GCC 13 and clang 18, x86-64): twitter, ordered_json view == ordered_json: 0.939 / 0.941; 20000 small objects with a duplicate key: 0.741-0.768; all other rows (json, citm_catalog, canada, view == view) within 0.988-1.003. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-25 21:56:14 +01:00
Niels Lohmann	d4faace387	Merge develop into perf/view-compare Develop fixed the VS 2015 build of the DOM parser itself (#5916): its json_sax.hpp is kept, so this branch no longer changes the file. single_include is a fresh amalgamation. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-25 12:13:26 +01:00
Niels Lohmann	165f91ed3c	Build the DOM parser with Visual Studio 2015 again Since #5882, the three VS 2015 jobs of AppVeyor fail on develop: - the detection of keys that can be indexed instantiated std::hash for every string type, which VS 2015 rejects with a static_assert for a type it does not hash (alt_string): std::hash is now looked at only for objects in a vector - the frame stack copied its frames with std::copy to a raw pointer, which VS 2015 reports as unsafe (C4996, an error with /WX): a loop now Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-25 12:04:43 +01:00
Niels Lohmann	7833865dbd	Merge develop into perf/view-compare single_include is a fresh amalgamation. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-25 12:00:01 +01:00
Niels Lohmann	9707519b24	Move the json_view comparison tests into their own file With the tests this branch adds, the object file of unit-json_view_dump.cpp got too large for the MinGW x64 assembler ("file too big"). The test case "json_view comparison" is now unit-json_view_compare.cpp, unchanged and also built with C++17. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-24 20:03:58 +02:00
Niels Lohmann	2d040c89d0	Merge develop into perf/view-compare Develop silenced the same Infer false positive in materialize() (#5905): its version is kept. single_include is a fresh amalgamation. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-24 17:14:02 +02:00
Niels Lohmann	182b31f902	Mark Infer's null dereference in materialize() as a false positive Infer (ci_infer, also red on develop since #5882) reports a null dereference in json_sax_dom_parser::null() called from materialize(). The member pointer the parser stores a value of an object in is set by the key() before it, so it is not null there. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-24 15:40:31 +02:00
Niels Lohmann	6775994761	Merge develop into perf/view-compare Resolves the conflict in unit-json_view_image.cpp, where develop (#5889) and this branch each added a section to "json_view images: errors": both are kept. single_include is a fresh amalgamation. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-24 07:48:54 +02:00
Niels Lohmann	70d1baa57a	Remove a nested duplicate guard from the json_view image test clang-tidy reported the inner #if !defined(JSON_NOEXCEPTION) as redundant (the test case is in the same #if) and a find() != npos, which gets the NOLINT the other tests use. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-23 18:53:34 +02:00
Niels Lohmann	536672e927	Dispatch on the compile-time conditions of the json_view comparison MSVC (/W4 /WX) reported C4127 (conditional expression is constant) for the template parameter EA and the object type's ordering in an if condition; both are now tag-dispatched. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-23 09:44:39 +02:00
Niels Lohmann	55cdc93ff8	Keep the byte comparison helpers in compare.hpp Address review comments: bytes_equal() and bytes_less() move from node.hpp to compare.hpp, their only user; key_less() no longer splits the comment of resolve_indices() from its function; and the loop counters of bytes_equal() and all_digits() are scoped to their loops. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-23 06:47:22 +02:00
Niels Lohmann	db5bd7fbcc	Fix the compiler and lint failures of the json_view comparison - all_digits(): the lambda declares its constant itself; MSVC does not let a lambda without captures read a constexpr local of the function - equal(): the frame's "in document order" index is a namespace-scope constant (document_order); the local constexpr npos in the default member initializers of the local struct crashed GCC 9 and MinGW GCC and was not found by MSVC - keys_differ_early() keeps the first keys in std::arrays (cpplint: variable-length arrays; clang-tidy: C arrays), and the basic_json side's first_keys() is not noexcept (key() may throw) - the numbers of one type are compared in an if chain instead of a nested conditional operator - the dump test avoids C arrays and temporary string concatenations View comparison instructions are unchanged (Apple clang, twitter, citm, canada: -0.1 % to +0.1 %). Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-22 20:13:22 +02:00
Niels Lohmann	89bd5d63e1	Rename names that Flawfinder takes for the file function open() The comparison cursors' open() is now enter(), and a test's open and close strings are head and tail. Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-22 13:27:54 +02:00
Niels Lohmann	3fb7d9c710	Check the float tokens of a json_document image without a sort and a scan of every character load(image_check::full) checked every float token of an image by sorting the token ranges, scanning the token character by character, and comparing the digit layout with the node's. Now: - the ranges are sorted only if they are not sorted already (save() writes them in the order of the nodes, so they are); - a token of the shape [-]digits.digits whose digit counts are the ones the node records (nearly all of them) is accepted by checking its digits eight at a time, without the general scan; - the general scan skips runs of digits eight at a time. Damaged tokens are still rejected as before: the check accepts exactly the tokens that the general scan accepts. Instructions of load(image_check::full) (develop -> now): macOS arm64 canada 44.6M -> 30.9M (-31 %), jeopardy 11.3M -> 10.7M, citm 10.4M -> 10.0M, twitter 5.06M -> 4.94M Linux GCC 13 canada -50 %, jeopardy -43 %, twitter -31 %, citm -28 % Linux Clang 18 canada -42 %, jeopardy -48 %, twitter -35 %, citm -32 % image_check::bounds is unchanged (within 0.4 %). Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-22 09:30:44 +02:00
Niels Lohmann	fa8134ae85	Compare json_view values without a vector per container and a basic_json per number operator== of views (view/view, view/json, and editable documents) built two vectors of members for every object it entered, sorted them with stable_sort for every object, and converted each number to a basic_json value. It now: - keeps the members of all open objects in one vector (reserved with geometric growth, so that nested objects do not reallocate), and orders them by insertion sort for up to 16 members, or not at all when the keys ascend already; - compares numbers of one type directly, and builds a basic_json only for numbers of different types; - compares short texts (keys, numbers, most strings) with two overlapping loads (bytes_equal(), in node.hpp); - first walks both documents in lockstep, in the order of their text, which finds most pairs of equal values without the work of the comparison. The walk only enters an object if the comparison would visit its members in the same order (the object type keeps the order, the keys ascend, or all values are scalars), so a difference is never found later than before, and it records the containers it found different, so that the comparison does not walk them again; - rejects values of different types and arrays of different sizes before any of this, and, for an object type that keeps the order, objects whose first (up to four) members have different keys, which ends a comparison that differs near the start without preparing the members. Results are unchanged: numbers compare by value across types, duplicate keys resolve as parse() resolves them, and objects compare in document order for ordered_json and by key otherwise. Instructions of one comparison of a document with an equal copy, macOS arm64 (develop -> now), json_document / ordered_json_document / editable: twitter 16.1M -> 7.0M 17.7M -> 1.5M 16.6M -> 7.3M citm 45.0M -> 3.4M 47.0M -> 3.2M 45.7M -> 4.3M canada 111.4M -> 13.7M 111.4M -> 13.7M 113.3M -> 15.0M jeopardy 26.2M -> 3.7M 27.3M -> 3.7M 26.6M -> 12.1M and of a view with a basic_json value: twitter 11.4M -> 5.9M 11.2M -> 7.3M 11.6M -> 5.9M citm 33.9M -> 9.7M 32.6M -> 9.1M 34.6M -> 9.8M canada 92.9M -> 32.1M 92.9M -> 32.1M 93.9M -> 32.8M and Linux instruction counts (GCC 13 / Clang 18) of the same comparisons: document with an equal copy json_document ordered_json_document editable twitter -49 % / -52 % -89 % / -89 % -48 % / -50 % citm -90 % / -89 % -91 % / -90 % -88 % / -87 % jeopardy (4,000 questions) -85 % / -85 % -86 % / -86 % -46 % / -49 % view with a basic_json value twitter -41 % / -45 % -28 % / -35 % -45 % / -44 % citm -63 % / -65 % -63 % / -66 % -65 % / -64 % jeopardy -39 % / -43 % -14 % / -24 % -44 % / -43 % A sweep that places one difference at about 2,100 positions of eight documents (a token at an even spacing, the first and last tokens, the first, middle and last member of the root removed, one appended, the document wrapped) in five kinds of comparison (view/view of read-only and of editable documents, view/json, json_document, editable/view) finds the same results as before and no placement slower on the Mac (Clang); on Linux no placement is slower than 3 % (Clang and GCC). Signed-off-by: Niels Lohmann <mail@nlohmann.me>	2026-10-22 09:30:42 +02:00