json_view: keep GCC's inlining of view walks with the diagnostics macros

Measured on GCC 13 (x86-64) with JSON_DIAGNOSTICS or JSON_DIAGNOSTIC_POSITIONS,
a recursive walk over a view took 2.7 to 2.9 times the instructions: GCC no
longer inlined the walk into itself. The cause was not the walk's code but
that the path and range code of errors.hpp was instantiated in every
translation unit (the non-template throw helpers referenced it), which
changed GCC's early inlining of unrelated view functions. diagnostics() and
the throw helpers that take an error_site are now templates, so that code is
instantiated only where an exception can be thrown; the walks have the
instruction counts of develop again.

Iterators no longer keep their container: with it, one walk was still 1.1 %
slower. key() of an iterator over an array or a primitive value
(invalid_iterator.207) is therefore thrown without a context, as documented.

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

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

@@ -68,9 +68,10 @@ Strong guarantee: there are no changes to the view or the document it refers to.
     - Throws [`out_of_range.404`](../../home/exceptions.md#jsonexceptionout_of_range404) if `ptr` cannot be resolved
       because a reference token is used on a primitive value.
 
-None of these exceptions carry a [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) path: the view has no
-`BasicJsonType` value to point at, so the exception is created without one, even if `BasicJsonType` was built with
-`JSON_DIAGNOSTICS` enabled.
+As for `BasicJsonType`, with [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) these exceptions name the JSON
+Pointer path of the value within the document, and with
+[`JSON_DIAGNOSTIC_POSITIONS`](../macros/json_diagnostic_positions.md) its bytes in the source text (where
+[`materialize()`](materialize.md) sets them).
 
 ## Complexity
 

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

@@ -20,9 +20,9 @@ Strong guarantee: there are no changes to the view or the document it refers to.
 Throws [`invalid_iterator.214`](../../home/exceptions.md#jsonexceptioninvalid_iterator214) if the view is
 [null](is_null.md) or [discarded](is_discarded.md), or if it is an empty array or object.
 
-This exception does not carry a [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) path: the view has no
-`BasicJsonType` value to point at, so the exception is created without one, even if `BasicJsonType` was built with
-`JSON_DIAGNOSTICS` enabled.
+As for `BasicJsonType`, with [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) the exception names the JSON Pointer
+path of the value within the document, and with [`JSON_DIAGNOSTIC_POSITIONS`](../macros/json_diagnostic_positions.md)
+its bytes in the source text (where [`materialize()`](materialize.md) sets them).
 
 ## Complexity
 

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

@@ -20,9 +20,9 @@ Strong guarantee: there are no changes to the view or the document it refers to.
 Throws [`invalid_iterator.214`](../../home/exceptions.md#jsonexceptioninvalid_iterator214) if the view is
 [null](is_null.md) or [discarded](is_discarded.md), or if it is an empty array or object.
 
-This exception does not carry a [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) path: the view has no
-`BasicJsonType` value to point at, so the exception is created without one, even if `BasicJsonType` was built with
-`JSON_DIAGNOSTICS` enabled.
+As for `BasicJsonType`, with [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) the exception names the JSON Pointer
+path of the value within the document, and with [`JSON_DIAGNOSTIC_POSITIONS`](../macros/json_diagnostic_positions.md)
+its bytes in the source text (where [`materialize()`](materialize.md) sets them).
 
 ## Complexity
 

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

@@ -80,14 +80,13 @@ Strong guarantee: there are no changes to the view or the document it refers to.
 - For every other `T`: whatever [`materialize().get<T>()`](../basic_json/get.md) throws — typically `type_error.302`,
   or whatever a user-provided `from_json()` throws.
 
-None of the exceptions thrown directly by this function or by a `from_json()` that reads the view (the first four
-bullets above) carry a
-[`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) path: the view has no `BasicJsonType` value to point at. An
-exception thrown while converting through `materialize()` (the last bullet) is different: it is thrown by a real
-`BasicJsonType` value, so it **does** carry a `JSON_DIAGNOSTICS` path if `BasicJsonType` was built with it enabled.
-That path is relative to the root of the view's subtree, because `materialize()` builds only the subtree (with parent
-pointers): it is empty if the failing value is the subtree's root itself, and it never is the path within the whole
-document.
+The exceptions thrown directly by this function or by a `from_json()` that reads the view (the first four bullets
+above) carry the context of those of `BasicJsonType`: with [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) the
+JSON Pointer path of the value within the document, and with
+[`JSON_DIAGNOSTIC_POSITIONS`](../macros/json_diagnostic_positions.md) its bytes in the source text. An exception
+thrown while converting through `materialize()` (the last bullet) is thrown by the materialized `BasicJsonType` value
+instead: its path is relative to the root of the view's subtree, because `materialize()` builds only the subtree
+(with parent pointers), so it is empty if the failing value is the subtree's root itself.
 
 ## Complexity
 

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

@@ -22,9 +22,9 @@ Strong guarantee: there are no changes to the view or the document it refers to.
 Throws [`type_error.302`](../../home/exceptions.md#jsonexceptiontype_error302) if the value is not a string; example:
 `"type must be string, but is array"`.
 
-This exception does not carry a [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) path: the view has no
-`BasicJsonType` value to point at, so the exception is created without one, even if `BasicJsonType` was built with
-`JSON_DIAGNOSTICS` enabled.
+As for `BasicJsonType`, with [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) the exception names the JSON Pointer
+path of the value within the document, and with [`JSON_DIAGNOSTIC_POSITIONS`](../macros/json_diagnostic_positions.md)
+its bytes in the source text (where [`materialize()`](materialize.md) sets them).
 
 ## Complexity
 

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

@@ -22,9 +22,9 @@ Strong guarantee: there are no changes to the view or the document it refers to.
 Throws [`type_error.302`](../../home/exceptions.md#jsonexceptiontype_error302) if the value is not a number; example:
 `"type must be number, but is string"`.
 
-This exception does not carry a [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) path: the view has no
-`BasicJsonType` value to point at, so the exception is created without one, even if `BasicJsonType` was built with
-`JSON_DIAGNOSTICS` enabled.
+As for `BasicJsonType`, with [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) the exception names the JSON Pointer
+path of the value within the document, and with [`JSON_DIAGNOSTIC_POSITIONS`](../macros/json_diagnostic_positions.md)
+its bytes in the source text (where [`materialize()`](materialize.md) sets them).
 
 ## Complexity
 

docs/mkdocs/docs/api/basic_json_view/operator[].md

@@ -73,9 +73,10 @@ Strong guarantee: there are no changes to the view or the document it refers to.
     - [`parse_error.109`](../../home/exceptions.md#jsonexceptionparse_error109) if an array index in `ptr` is not a
       number.
 
-None of these exceptions carry a [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) path: the view has no
-`BasicJsonType` value to point at, so the exception is created without one, even if `BasicJsonType` was built with
-`JSON_DIAGNOSTICS` enabled.
+As for `BasicJsonType`, with [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) these exceptions name the JSON
+Pointer path of the value within the document, and with
+[`JSON_DIAGNOSTIC_POSITIONS`](../macros/json_diagnostic_positions.md) its bytes in the source text (where
+[`materialize()`](materialize.md) sets them).
 
 ## Complexity
 

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

@@ -61,9 +61,10 @@ Strong guarantee: there are no changes to the view or the document it refers to.
    `default_value` instead of throwing, exactly as [`BasicJsonType::value`](../basic_json/value.md) catches
    `out_of_range` and returns `default_value`.
 
-None of these exceptions carry a [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) path: the view has no
-`BasicJsonType` value to point at, so the exception is created without one, even if `BasicJsonType` was built with
-`JSON_DIAGNOSTICS` enabled.
+As for `BasicJsonType`, with [`JSON_DIAGNOSTICS`](../macros/json_diagnostics.md) these exceptions name the JSON
+Pointer path of the value within the document, and with
+[`JSON_DIAGNOSTIC_POSITIONS`](../macros/json_diagnostic_positions.md) its bytes in the source text (where
+[`materialize()`](materialize.md) sets them).
 
 ## Complexity
 

docs/mkdocs/docs/features/json_view.md

@@ -134,9 +134,14 @@ document: `#!cpp auto v = json_document::parse(text).root();` does not compile.
 - **Macros.** [`JSON_STRICT_NUL_HANDLING`](../api/macros/json_strict_nul_handling.md) and
   [`JSON_NOEXCEPTION`](../api/macros/json_noexception.md)/[`JSON_THROW_USER`](../api/macros/json_throw_user.md)
   behave the same way they do for `<nlohmann/json.hpp>`. These are the macros that change what the parser accepts or
-  how it reports errors. The others either do not apply ([`JSON_DIAGNOSTICS`](../api/macros/json_diagnostics.md)
-  needs parent pointers in `basic_json` values) or only affect the values `materialize()` produces, through
-  `BasicJsonType`.
+  how it reports errors. With [`JSON_DIAGNOSTICS`](../api/macros/json_diagnostics.md) and
+  [`JSON_DIAGNOSTIC_POSITIONS`](../api/macros/json_diagnostic_positions.md), the exceptions of views name the path
+  and the bytes of the value as those of `#!cpp basic_json` do; as a view has neither parent pointers nor stored
+  positions, both are found from the document's index when the exception is thrown, which takes time linear in the
+  size of the document (only for an exception). One exception has neither: `key()` of an iterator over an array or a
+  primitive value ([`invalid_iterator.207`](../home/exceptions.md#jsonexceptioninvalid_iterator207)), as a view's
+  iterator does not know the value it iterates over. The other macros only affect the values `materialize()` produces,
+  through `BasicJsonType`.
 
 ## What is different
 
@@ -167,10 +172,6 @@ document: `#!cpp auto v = json_document::parse(text).root();` does not compile.
   [`at`](../api/basic_json_view/at.md) throws for every missing value.
 - **Duplicate keys: lookups find the first member, `parse()` keeps the last.** See [Duplicate keys](#duplicate-keys)
   below.
-- **No [`JSON_DIAGNOSTICS`](../api/macros/json_diagnostics.md) path.** Exceptions thrown by `basic_json_view`'s own
-  element access and lookup functions never carry the JSON Pointer path `JSON_DIAGNOSTICS` would otherwise add: the
-  view has no `basic_json` value to point at, so the exception is created without one, regardless of how
-  `BasicJsonType` was built.
 - **Ordering comparisons are not provided** by `basic_json_view` — there is no `#!cpp operator<`.
   [`operator==`](../api/basic_json_view/operator_eq.md) and [`operator!=`](../api/basic_json_view/operator_ne.md) are
   provided, though: two views, or a view and a `BasicJsonType` value, compare equal exactly when
@@ -329,14 +330,10 @@ Its body may only use what views and `basic_json` have in common: the type check
 a view does not have; such a type is converted through `materialize()`. A `from_json()` for `json` only is
 converted through `materialize()` as well.
 
-Two things differ when a `from_json()` reads the view:
-
-- **Duplicate keys.** Members looked up by name (`at()`, `value()`, ... and thus the macros) find the **first** member
-  with a repeated key, see [Duplicate keys](#duplicate-keys); `#!cpp materialize().get<T>()` would see the last one.
-  Containers built from an object, such as `#!cpp std::map`, keep the **last** value either way, as
-  [`parse()`](../api/basic_json/parse.md) does.
-- **[`JSON_DIAGNOSTICS`](../api/macros/json_diagnostics.md).** Exceptions thrown by the view carry no path to the
-  value.
+One thing differs when a `from_json()` reads the view: members looked up by name (`at()`, `value()`, ... and thus
+the macros) find the **first** member with a repeated key, see [Duplicate keys](#duplicate-keys);
+`#!cpp materialize().get<T>()` would see the last one. Containers built from an object, such as `#!cpp std::map`,
+keep the **last** value either way, as [`parse()`](../api/basic_json/parse.md) does.
 
 `to_json()` is unaffected: it writes `basic_json` values, never views.
 

include/nlohmann/detail/exceptions.hpp

@@ -431,6 +431,28 @@ std::nullptr_t exception_context(const View& /*unused*/) noexcept
     return nullptr;
 }
 
+namespace view
+{
+// the path and bytes of the value of a view (defined in <nlohmann/detail/view/errors.hpp>)
+template<typename View>
+std::string diagnostics_of(const View& v);
+}  // namespace view
+
+// what goes in front of the message of an exception about j: nothing for a
+// basic_json value (exception_context() passes the value itself), the context
+// basic_json would add (JSON_DIAGNOSTICS, JSON_DIAGNOSTIC_POSITIONS) for a view
+template<typename BasicJsonType, enable_if_t<is_basic_json<BasicJsonType>::value, int> = 0>
+const char* exception_prefix(const BasicJsonType& /*unused*/) noexcept
+{
+    return "";
+}
+
+template<typename View, enable_if_t<is_json_view<View>::value, int> = 0>
+std::string exception_prefix(const View& v)
+{
+    return view::diagnostics_of(v);
+}
+
 /*!
 @brief throws because @a j does not have the type a conversion expects
 @param[in] expected  the expected type(s), e.g. "array" or "binary or array"
@@ -442,7 +464,7 @@ JSON_HEDLEY_NO_RETURN JSON_HEDLEY_NEVER_INLINE void throw_type_must_be(const cha
 {
     static_cast<void>(expected); // unused when JSON_NOEXCEPTION is defined
     static_cast<void>(j);
-    JSON_THROW(type_error::create(exception_id::type_mismatch, concat("type must be ", expected, ", but is ", j.type_name()), exception_context(j)));
+    JSON_THROW(type_error::create(exception_id::type_mismatch, concat(exception_prefix(j), "type must be ", expected, ", but is ", j.type_name()), exception_context(j)));
 }
 
 /*!

include/nlohmann/detail/macro_scope.hpp

@@ -347,7 +347,7 @@
             return ej_pair.second == j;                                                         \
         });                                                                                     \
         if (it != std::end(m)) e = it->first;                                                   \
-        else ::nlohmann::detail::templated_json_throw<nlohmann::detail::out_of_range>(nlohmann::detail::out_of_range::create(nlohmann::detail::exception_id::value_out_of_range, nlohmann::detail::concat("enum value out of range for " #ENUM_TYPE ": ", nlohmann::detail::from_json_dump(j)), nlohmann::detail::exception_context(j))); \
+        else ::nlohmann::detail::templated_json_throw<nlohmann::detail::out_of_range>(nlohmann::detail::out_of_range::create(nlohmann::detail::exception_id::value_out_of_range, nlohmann::detail::concat(nlohmann::detail::exception_prefix(j), "enum value out of range for " #ENUM_TYPE ": ", nlohmann::detail::from_json_dump(j)), nlohmann::detail::exception_context(j))); \
     }
 
 // Ugly macros to avoid uglier copy-paste when specializing basic_json. They

include/nlohmann/detail/view/edit.hpp

@@ -157,7 +157,7 @@ class editor
         node* const o = own(object);
         if (o->kind != node_kind::object && o->kind != node_kind::null)
         {
-            throw_type_error(exception_id::subscript_wrong_type, "cannot use operator[] with a string argument with ", object.type_name());
+            throw_type_error(site_of(object), exception_id::subscript_wrong_type, "cannot use operator[] with a string argument with ", object.type_name());
         }
         check_utf8(key.data(), key.size());
         const encoded e = encode(std::forward<V>(value));
@@ -208,9 +208,9 @@ class editor
         node* const a = own(array);
         if (a->kind != node_kind::array)
         {
-            throw_type_error(exception_id::subscript_wrong_type, "cannot use operator[] with a numeric argument with ", array.type_name());
+            throw_type_error(site_of(array), exception_id::subscript_wrong_type, "cannot use operator[] with a numeric argument with ", array.type_name());
         }
-        check_index(idx, a->len);
+        check_index(a, idx, a->len);
         const encoded e = encode(std::forward<V>(value));
         node* const slot = const_cast<node*>(nav::value(element_at<true>(m_doc, a, idx))); // NOLINT(cppcoreguidelines-pro-type-const-cast)
         assign(slot, e, a, true);
@@ -225,7 +225,7 @@ class editor
         node* const a = own(array);
         if (a->kind != node_kind::array && a->kind != node_kind::null)
         {
-            throw_type_error(exception_id::push_back_wrong_type, "cannot use push_back() with ", array.type_name());
+            throw_type_error(site_of(array), exception_id::push_back_wrong_type, "cannot use push_back() with ", array.type_name());
         }
         const encoded e = encode(std::forward<V>(value));
         if (a->kind == node_kind::null)
@@ -249,9 +249,9 @@ class editor
         node* const a = own(array);
         if (a->kind != node_kind::array)
         {
-            throw_type_error(exception_id::insert_wrong_type, "cannot use insert() with ", array.type_name());
+            throw_type_error(site_of(array), exception_id::insert_wrong_type, "cannot use insert() with ", array.type_name());
         }
-        check_index(idx, a->len + 1);
+        check_index(a, idx, a->len + 1);
         const encoded e = encode(std::forward<V>(value));
         node* const slot = new_slot(e);
         node* const h = block_of(m_doc, a, 1);
@@ -269,7 +269,7 @@ class editor
         node* const o = own(object);
         if (o->kind != node_kind::object)
         {
-            throw_type_error(exception_id::erase_wrong_type, "cannot use erase() with ", object.type_name());
+            throw_type_error(site_of(object), exception_id::erase_wrong_type, "cannot use erase() with ", object.type_name());
         }
         for (const node* k = nav::first(m_doc, o), *end = nav::end(m_doc, o); k != end; k = document_data::after(k + 1))
         {
@@ -287,9 +287,9 @@ class editor
         node* const a = own(array);
         if (a->kind != node_kind::array)
         {
-            throw_type_error(exception_id::erase_wrong_type, "cannot use erase() with ", array.type_name());
+            throw_type_error(site_of(array), exception_id::erase_wrong_type, "cannot use erase() with ", array.type_name());
         }
-        check_index(idx, a->len);
+        check_index(a, idx, a->len);
         node* const h = block_of(m_doc, a, 0);
         std::memmove(h + 1 + idx, h + 2 + idx, (h->next - 2 - idx) * sizeof(node));
         --h->next;
@@ -316,11 +316,11 @@ class editor
         return const_cast<node*>(v.m_node); // NOLINT(cppcoreguidelines-pro-type-const-cast): the nodes belong to this document
     }
 
-    static void check_index(std::size_t idx, std::size_t limit)
+    void check_index(const node* array, std::size_t idx, std::size_t limit) const
     {
         if (idx >= limit)
         {
-            throw_array_index_out_of_range(idx);
+            throw_array_index_out_of_range(error_site(&m_doc, array), idx);
         }
     }
 

include/nlohmann/detail/view/errors.hpp

@@ -10,11 +10,15 @@
 
 #include <algorithm> // min
 #include <cstddef> // size_t
-#include <string> // string
+#include <string> // string, to_string
+#include <vector> // vector
 
 #include <nlohmann/json.hpp>
 #include <nlohmann/detail/view/builder.hpp>
+#include <nlohmann/detail/view/document_data.hpp>
 #include <nlohmann/detail/view/macro_scope.hpp>
+#include <nlohmann/detail/view/materialize.hpp>
+#include <nlohmann/detail/view/node.hpp>
 
 NLOHMANN_JSON_NAMESPACE_BEGIN
 namespace detail
@@ -22,8 +26,285 @@ namespace detail
 namespace view
 {
 
+/*!
+@brief the value an exception is about
+
+basic_json passes the value to the exception, which then names its path
+(JSON_DIAGNOSTICS) and its bytes in the source (JSON_DIAGNOSTIC_POSITIONS). A
+view's value has no parent pointer and no stored positions: both are found
+from the document and the node, when the exception is thrown (so only the
+error path pays for them). Without the two macros, a site is empty.
+*/
+class error_site
+{
+  public:
+    error_site() noexcept = default;
+
+    error_site(const document_data* d, const node* n) noexcept
+#if JSON_DIAGNOSTICS || JSON_DIAGNOSTIC_POSITIONS
+        : m_doc(d), m_node(n)
+#endif
+    {
+        static_cast<void>(d);
+        static_cast<void>(n);
+    }
+
+#if JSON_DIAGNOSTICS || JSON_DIAGNOSTIC_POSITIONS
+    const document_data* doc() const noexcept
+    {
+        return m_doc;
+    }
+    const node* value() const noexcept
+    {
+        return m_node;
+    }
+
+  private:
+    const document_data* m_doc = nullptr;
+    const node* m_node = nullptr;
+#endif
+};
+
+/// the site of a view's value (the views make this a friend)
+struct view_access
+{
+    template<typename View>
+    static error_site site(const View& v) noexcept
+    {
+        return error_site(v.m_doc, v.m_node);
+    }
+};
+
+template<typename View>
+error_site site_of(const View& v) noexcept
+{
+    return view_access::site(v);
+}
+
+#if JSON_DIAGNOSTICS
+/*!
+@brief the JSON pointer tokens from the root of document @a d to node @a target
+
+Nodes have no parent index, so the path is found by descending from the root:
+before the first edit, only into the subtree that contains the target (a
+container spans the next `next` nodes); after edits, elements can live in
+other sequences, so every container is searched. Iterative (the nesting depth
+is limited by memory only). False if the target is not found (then tokens is
+unspecified).
+*/
+template<bool Editable>
+bool value_path(const document_data& d, const node* target, std::vector<std::string>& tokens)
+{
+    using nav = navigation<Editable>;
+    struct frame
+    {
+        const node* pos;
+        const node* end;
+        std::size_t index;
+        bool object;
+    };
+    const node* const root = d.nodes;
+    if (target == root)
+    {
+        return true;
+    }
+    if (!is_container(*root))
+    {
+        return false;
+    }
+    std::vector<frame> stack;
+    stack.push_back(frame{nav::first(d, root), nav::end(d, root), 0, root->kind == node_kind::object});
+    // (tokens[i] leads from stack[i] to stack[i + 1], or to the target)
+    while (!stack.empty())
+    {
+        frame& f = stack.back();
+        if (f.pos == f.end)
+        {
+            stack.pop_back();
+            if (!tokens.empty())
+            {
+                tokens.pop_back();
+            }
+            continue;
+        }
+        const node* entry = f.pos;
+        std::string token;
+        if (f.object)
+        {
+            // (as basic_json: the key up to a null byte)
+            token.assign(d.str(*entry), entry->len);
+            const std::size_t nul = token.find('\0');
+            if (nul != std::string::npos)
+            {
+                token.resize(nul);
+            }
+            ++entry;
+        }
+        else
+        {
+            token = std::to_string(f.index);
+        }
+        const node* const child = nav::value(entry);
+        f.pos = document_data::after(entry);
+        ++f.index;
+        if (child == target)
+        {
+            tokens.push_back(token);
+            return true;
+        }
+        if (is_container(*child) && (nav::editable() || (target > child && target < child + child->next)))
+        {
+            tokens.push_back(token);
+            stack.push_back(frame{nav::first(d, child), nav::end(d, child), 0, child->kind == node_kind::object});
+        }
+    }
+    return false;
+}
+#endif
+
+#if JSON_DIAGNOSTIC_POSITIONS
+/*!
+@brief the bytes [start, end) of node @a n in the source, as materialize() sets them
+
+Scalars as position_source::set_scalar; an array or object from its opening
+bracket to its closing bracket, which is found by following the chain of last
+elements to a scalar (or an empty container) and scanning from there to the
+closing brackets. None (npos) for values written by edits, and for containers
+whose last element has none, as with materialize().
+*/
+template<bool Editable>
+void value_range(const document_data& d, const node* n, std::size_t& start, std::size_t& end)
+{
+    using nav = navigation<Editable>;
+    position_source p;
+    start = std::string::npos;
+    end = std::string::npos;
+    if (!is_container(*n))
+    {
+        p.set_scalar(d, *n);
+        start = p.start;
+        end = p.end;
+        return;
+    }
+    if ((n->flags & node_flags::is_new) != 0 || n->off >= d.size)
+    {
+        return;
+    }
+    // the open brackets: n, then the last element as long as it is a non-empty container
+    std::size_t closing = 0;
+    std::size_t pos = std::string::npos;
+    for (const node* c = n;;)
+    {
+        if (is_container(*c))
+        {
+            if ((c->flags & node_flags::is_new) != 0 || c->off >= d.size)
+            {
+                return;
+            }
+            ++closing;
+            const node* const first = nav::first(d, c);
+            const node* const last_end = nav::end(d, c);
+            if (first == last_end)
+            {
+                pos = std::size_t{c->off} + 1;
+                break;
+            }
+            const std::size_t value_offset = c->kind == node_kind::object ? 1 : 0;
+            const node* last = first;
+            for (const node* e = document_data::after(first + value_offset); e != last_end; e = document_data::after(e + value_offset))
+            {
+                last = e;
+            }
+            c = nav::value(last + value_offset);
+            continue;
+        }
+        p.set_scalar(d, *c);
+        pos = p.end;
+        break;
+    }
+    for (; closing != 0 && pos != std::string::npos; --closing)
+    {
+        pos = position_source::container_end(d.base[0], d.size, pos);
+    }
+    if (pos != std::string::npos)
+    {
+        start = n->off;
+        end = pos;
+    }
+}
+#endif
+
+/// the context basic_json puts in front of the message of an exception about
+/// a value: "(/path) " with JSON_DIAGNOSTICS (none for the root), then
+/// "(bytes a-b) " with JSON_DIAGNOSTIC_POSITIONS (with JSON_DIAGNOSTICS, only
+/// if there is a path)
+///
+/// This and the throw helpers that take an error_site are templates, so that
+/// the path and range code is instantiated only where an exception can be
+/// thrown: instantiated in every translation unit (as a non-template inline
+/// function makes it), it changed GCC's early inlining of unrelated view
+/// functions, so that GCC 13 no longer inlined a recursive walk over a view
+/// into itself (2.7 times the instructions).
+template<typename Unused = void>
+std::string diagnostics(const error_site site)
+{
+#if JSON_DIAGNOSTICS || JSON_DIAGNOSTIC_POSITIONS
+    if (site.doc() == nullptr || site.value() == nullptr)
+    {
+        return "";
+    }
+    const document_data& d = *site.doc();
+    const bool edited = d.edits != nullptr;
+    std::string bytes;
+#if JSON_DIAGNOSTIC_POSITIONS
+    std::size_t start = 0;
+    std::size_t end = 0;
+    if (edited)
+    {
+        value_range<true>(d, site.value(), start, end);
+    }
+    else
+    {
+        value_range<false>(d, site.value(), start, end);
+    }
+    if (start != std::string::npos && end != std::string::npos)
+    {
+        bytes = concat("(bytes ", std::to_string(start), "-", std::to_string(end), ") ");
+    }
+#endif
+#if JSON_DIAGNOSTICS
+    std::vector<std::string> tokens;
+    const bool found = edited ? value_path<true>(d, site.value(), tokens) : value_path<false>(d, site.value(), tokens);
+    if (!found || tokens.empty())
+    {
+        return "";
+    }
+    std::string path;
+    for (const std::string& token : tokens)
+    {
+        path.push_back('/');
+        path += detail::escape(token);
+    }
+    return concat('(', path, ") ", bytes);
+#else
+    return bytes;
+#endif
+#else
+    static_cast<void>(site);
+    return "";
+#endif
+}
+
+/// the context of an exception about view @a v (see exception_prefix())
+template<typename View>
+std::string diagnostics_of(const View& v)
+{
+    return diagnostics<>(site_of(v));
+}
+
 // Exceptions are thrown out of line, so that the accessors that may throw stay
-// small enough to be inlined.
+// small enough to be inlined. The overloads with an error_site put its context
+// in front of the message.
 
 [[noreturn]] NLOHMANN_VIEW_NOINLINE inline void throw_type_error(exception_id id, const char* prefix, const char* type)
 {
@@ -45,6 +326,24 @@ namespace view
     NLOHMANN_VIEW_THROW(invalid_iterator::create(id, msg, nullptr));
 }
 
+template<typename Unused = void>
+[[noreturn]] NLOHMANN_VIEW_NOINLINE void throw_type_error(const error_site site, exception_id id, const char* prefix, const char* type)
+{
+    NLOHMANN_VIEW_THROW(type_error::create(id, concat(diagnostics<>(site), prefix, type), nullptr));
+}
+
+template<typename Unused = void>
+[[noreturn]] NLOHMANN_VIEW_NOINLINE void throw_out_of_range(const error_site site, exception_id id, const std::string& msg)
+{
+    NLOHMANN_VIEW_THROW(out_of_range::create(id, concat(diagnostics<>(site), msg), nullptr));
+}
+
+template<typename Unused = void>
+[[noreturn]] NLOHMANN_VIEW_NOINLINE void throw_invalid_iterator(const error_site site, exception_id id, const char* msg)
+{
+    NLOHMANN_VIEW_THROW(invalid_iterator::create(id, concat(diagnostics<>(site), msg), nullptr));
+}
+
 /// a parse error without a position (as those of json_pointer)
 [[noreturn]] NLOHMANN_VIEW_NOINLINE inline void throw_parse_error(exception_id id, const std::string& msg)
 {
@@ -53,22 +352,23 @@ namespace view
 
 /// out_of_range.401: the array index @a idx is not below the size
 template<typename Index>
-[[noreturn]] NLOHMANN_VIEW_NOINLINE void throw_array_index_out_of_range(const Index idx)
+[[noreturn]] NLOHMANN_VIEW_NOINLINE void throw_array_index_out_of_range(const error_site site, const Index idx)
 {
-    throw_out_of_range(exception_id::array_index_out_of_range, concat("array index ", std::to_string(idx), " is out of range"));
+    throw_out_of_range(site, exception_id::array_index_out_of_range, concat("array index ", std::to_string(idx), " is out of range"));
 }
 
 /// out_of_range.402: the JSON Pointer token '-' of an array of @a size elements
-[[noreturn]] NLOHMANN_VIEW_NOINLINE inline void throw_array_index_past_end(std::size_t size)
+template<typename Unused = void>
+[[noreturn]] NLOHMANN_VIEW_NOINLINE void throw_array_index_past_end(const error_site site, std::size_t size)
 {
-    throw_out_of_range(exception_id::pointer_past_the_end_index, concat("array index '-' (", std::to_string(size), ") is out of range"));
+    throw_out_of_range(site, exception_id::pointer_past_the_end_index, concat("array index '-' (", std::to_string(size), ") is out of range"));
 }
 
 /// out_of_range.403: the object has no member @a key
 template<typename Key>
-[[noreturn]] NLOHMANN_VIEW_NOINLINE void throw_key_not_found(const Key& key)
+[[noreturn]] NLOHMANN_VIEW_NOINLINE void throw_key_not_found(const error_site site, const Key& key)
 {
-    throw_out_of_range(exception_id::key_not_found, concat("key '", key, "' not found"));
+    throw_out_of_range(site, exception_id::key_not_found, concat("key '", key, "' not found"));
 }
 
 /// SAX handler of the error replay: accepts every event without storing it,

include/nlohmann/detail/view/iterator.hpp

@@ -105,7 +105,10 @@ class view_iterator
     }
 
     /// the key of the current object member; throws invalid_iterator.207 for
-    /// other iterators, like basic_json's iterators
+    /// other iterators, like basic_json's iterators (but without the context
+    /// JSON_DIAGNOSTICS and JSON_DIAGNOSTIC_POSITIONS add: the iterator does not
+    /// know the value it iterates over, and keeping it would make iterators
+    /// larger, which keeps GCC from inlining recursive walks)
     string_view_t key() const
     {
         if (NLOHMANN_VIEW_UNLIKELY(m_value_offset == 0))

include/nlohmann/detail/view/number.hpp

@@ -19,7 +19,6 @@
 #include <nlohmann/detail/view/macro_scope.hpp>
 #include <nlohmann/detail/view/node.hpp>
 #include <nlohmann/detail/view/scan.hpp>
-#include <nlohmann/detail/view/simd.hpp>
 
 NLOHMANN_JSON_NAMESPACE_BEGIN
 namespace detail
@@ -104,56 +103,12 @@ NLOHMANN_VIEW_NOINLINE FloatType float_value(const char* first, const node& n)
     return convert_float<FloatType>(first, last, dot, mantissa_end);
 }
 
-#if NLOHMANN_VIEW_VECTOR
-/*!
-@brief the value of the first 16 digits of a run of digits with a point after
-int_digits of them (or none, if int_digits >= 16)
-
-The 17 bytes at p are read at once: the bytes after the point move down by one,
-then pairs, quadruples, and octets of digits are combined by multiplications
-(after simdjson). The bytes [p, p + 32) must be readable. Bytes that are not
-digits (only in a damaged image loaded with image_check::bounds) give a wrong
-value, but no undefined behavior.
-*/
-NLOHMANN_VIEW_ALWAYS_INLINE std::uint64_t sixteen_digits(const unsigned char* p, unsigned int_digits) noexcept
-{
-#if NLOHMANN_VIEW_NEON
-    const uint8x16_t low = vld1q_u8(p);
-    const uint8x16_t index = vcombine_u8(vcreate_u8(0x0706050403020100u), vcreate_u8(0x0F0E0D0C0B0A0908u));
-    const uint8x16_t before_point = vcltq_u8(index, vdupq_n_u8(static_cast<std::uint8_t>(int_digits)));
-    const uint8x16_t digits = vsubq_u8(vbslq_u8(before_point, low, vextq_u8(low, vld1q_u8(p + 16), 1)), vdupq_n_u8('0'));
-    const uint16x8_t twos = vpaddlq_u8(vmulq_u8(digits, vreinterpretq_u8_u16(vdupq_n_u16(0x010A))));
-    const uint32x4_t fours = vpaddlq_u16(vmulq_u16(twos, vreinterpretq_u16_u32(vdupq_n_u32(0x00010064))));
-    const uint64x2_t eights = vpaddlq_u32(vmulq_u32(fours, vreinterpretq_u32_u64(vdupq_n_u64(0x0000000100002710u))));
-    return (vgetq_lane_u64(eights, 0) * 100000000u) + vgetq_lane_u64(eights, 1);
-#else
-    // NOLINTBEGIN(portability-simd-intrinsics)
-    const __m128i low = _mm_loadu_si128(static_cast<const __m128i*>(static_cast<const void*>(p))); // NOLINT(bugprone-casting-through-void)
-    const __m128i high = _mm_loadu_si128(static_cast<const __m128i*>(static_cast<const void*>(p + 16))); // NOLINT(bugprone-casting-through-void)
-    const __m128i before_point = _mm_cmpgt_epi8(_mm_set1_epi8(static_cast<char>(int_digits)), _mm_setr_epi8(0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15));
-    const __m128i moved = _mm_or_si128(_mm_srli_si128(low, 1), _mm_slli_si128(high, 15));
-    const __m128i digits = _mm_sub_epi8(_mm_or_si128(_mm_and_si128(before_point, low), _mm_andnot_si128(before_point, moved)), _mm_set1_epi8('0'));
-    // (16-bit lanes: SSE2 has no multiply-add of bytes)
-    const __m128i tens = _mm_setr_epi16(10, 1, 10, 1, 10, 1, 10, 1);
-    const __m128i twos = _mm_packs_epi32(_mm_madd_epi16(_mm_unpacklo_epi8(digits, _mm_setzero_si128()), tens),
-                                         _mm_madd_epi16(_mm_unpackhi_epi8(digits, _mm_setzero_si128()), tens));
-    const __m128i fours = _mm_madd_epi16(twos, _mm_setr_epi16(100, 1, 100, 1, 100, 1, 100, 1));
-    const __m128i eights = _mm_madd_epi16(_mm_packs_epi32(fours, fours), _mm_setr_epi16(10000, 1, 10000, 1, 10000, 1, 10000, 1));
-    const auto upper = static_cast<std::uint32_t>(_mm_cvtsi128_si32(eights));
-    const auto lower = static_cast<std::uint32_t>(_mm_cvtsi128_si32(_mm_srli_si128(eights, 4)));
-    // NOLINTEND(portability-simd-intrinsics)
-    return (std::uint64_t{upper} * 100000000u) + lower;
-#endif
-}
-#endif
-
 /*!
 @brief the digits of a float token with at most 19 digits, from its layout
 
 The digit layout recorded while parsing says where the integer digits, the
 fraction digits, and the exponent are, so the digits are read eight at a
-time without scanning (with SIMD, a token of 16 to 19 digits sixteen at a
-time).
+time without scanning.
 
 @param[in] p  first character of the token
 @param[in] e  end of the token
@@ -163,34 +118,15 @@ NLOHMANN_VIEW_ALWAYS_INLINE float_significand layout_decimal(const unsigned char
 {
     const bool negative = *p == '-';
     p += negative ? 1 : 0;
-    std::uint64_t w = 0;
+    std::uint64_t w = parse_upto19(p, int_digits, limit);
+    p += int_digits;
     std::int64_t q = 0;
-#if NLOHMANN_VIEW_VECTOR
-    const unsigned count = int_digits + frac_digits;
-    if (count >= 16 && limit - p >= 32)
+    if (frac_digits != 0)
     {
-        // (the point, if any, follows the integer digits; the digits after
-        // the first 16 are at most three)
-        w = sixteen_digits(p, int_digits);
-        for (unsigned i = 16; i < count; ++i)
-        {
-            w = (w * 10) + static_cast<std::uint64_t>(p[i + (i >= int_digits ? 1 : 0)] - '0');
-        }
-        p += count + (frac_digits != 0 ? 1 : 0);
+        w = (w * int_pow10(frac_digits)) + parse_upto19(p + 1, frac_digits, limit);
+        p += 1 + frac_digits;
         q = -static_cast<std::int64_t>(frac_digits);
     }
-    else /* NOLINT(readability/braces) */
-#endif
-    {
-        w = parse_upto19(p, int_digits, limit);
-        p += int_digits;
-        if (frac_digits != 0)
-        {
-            w = (w * int_pow10(frac_digits)) + parse_upto19(p + 1, frac_digits, limit);
-            p += 1 + frac_digits;
-            q = -static_cast<std::int64_t>(frac_digits);
-        }
-    }
     if (p != e)
     {
         // [eE][+-]digits; huge exponents saturate (the parser rejected
@@ -225,22 +161,12 @@ NLOHMANN_VIEW_ALWAYS_INLINE float_significand layout_decimal(const unsigned char
 The result is correctly rounded by the lexer's conversion
 (detail::decimal_to_float(): Clinger's fast path where both operands are
 exact, else the Eisel-Lemire algorithm, which needs no fallback for up to 19
-digits), so it is the value parse() produces. Tokens of 16 or more digits go
-to Eisel-Lemire directly: the fast path takes few of them (none of 17 digits),
-and its check is a branch that data like GeoJSON coordinates mispredicts.
+digits), so it is the value parse() produces.
 */
 template<typename FloatType>
 NLOHMANN_VIEW_ALWAYS_INLINE FloatType layout_float(const unsigned char* p, const unsigned char* e, unsigned int_digits, unsigned frac_digits, const unsigned char* limit) noexcept
 {
-    const float_significand d = layout_decimal(p, e, int_digits, frac_digits, limit);
-    if (int_digits + frac_digits >= 16)
-    {
-        using format = ieee_binary_format<std::numeric_limits<FloatType>::digits>;
-        native_float_t<FloatType> value{};
-        float_from_bits(eisel_lemire<format>(d.exponent, d.w) | (d.negative ? (std::uint64_t{1} << format::sign_bit()) : 0u), value);
-        return value;
-    }
-    return decimal_to_float<FloatType>(d);
+    return decimal_to_float<FloatType>(layout_decimal(p, e, int_digits, frac_digits, limit));
 }
 
 /// the value of a float set by an edit: its token (the shortest round-trip
@@ -259,80 +185,6 @@ NLOHMANN_VIEW_NOINLINE FloatType edited_float(const char* token, const node& n)
     return float_value<FloatType>(token, n);
 }
 
-#if NLOHMANN_VIEW_VECTOR
-/// eisel_lemire() for the token w * 10^q of coordinate_float()
-template<typename Format>
-NLOHMANN_VIEW_ALWAYS_INLINE std::uint64_t coordinate_bits(std::int64_t q, std::uint64_t w, Format /*format*/) noexcept
-{
-    return eisel_lemire<Format>(q, w);
-}
-
-/*!
-@brief (binary64) eisel_lemire() without the cases that tokens of 16 or 17
-digits with 5 or more after the point cannot reach: a value of 0, a
-subnormal, an infinity, or a tie (q < -4)
-*/
-NLOHMANN_VIEW_ALWAYS_INLINE std::uint64_t coordinate_bits(std::int64_t q, std::uint64_t w, ieee_binary_format<53> format) noexcept
-{
-    if (NLOHMANN_VIEW_UNLIKELY(w == 0 || q > -5))
-    {
-        return eisel_lemire<ieee_binary_format<53>>(q, w);
-    }
-    static_cast<void>(format);
-    const int lz = count_leading_zeros(w);
-    const std::uint64_t normalized = w << static_cast<unsigned>(lz);
-    const auto index = static_cast<std::size_t>(2 * (q - pow5_128_smallest_power));
-    const uint128_parts product = full_multiplication(normalized, pow5_128()[index]);
-    std::uint64_t high = product.high;
-    if (NLOHMANN_VIEW_UNLIKELY((high & 0x1FFu) == 0x1FFu))
-    {
-        // (the lower bits may carry into the result)
-        const std::uint64_t second = full_multiplication(normalized, pow5_128()[index + 1]).high;
-        high += product.low + second < product.low ? 1u : 0u;
-    }
-    const auto upperbit = static_cast<unsigned>(high >> 63u);
-    std::uint64_t mantissa = high >> (upperbit + 9);
-    auto power2 = static_cast<std::uint64_t>(((217706 * q) >> 16) + 63 + static_cast<std::int64_t>(upperbit) - lz + 1023);
-    mantissa = (mantissa + (mantissa & 1u)) >> 1u;
-    const bool carry = mantissa >= (std::uint64_t{2} << 52u);
-    mantissa = carry ? (std::uint64_t{1} << 52u) : mantissa;
-    power2 += carry ? 1u : 0u;
-    return (mantissa & ~(std::uint64_t{1} << 52u)) | (power2 << 52u);
-}
-
-/*!
-@brief (binary32 and binary64) the value of a float token of the source with
-16 or 17 digits, a point, and no exponent, the shape of most coordinates
-written with full precision; whether the token of n has this shape
-
-These tokens skip the general layout of layout_decimal(): no digits beyond
-the 17th, no exponent (and Eisel-Lemire directly, as in layout_float()).
-*/
-template<typename FloatType>
-NLOHMANN_VIEW_ALWAYS_INLINE bool coordinate_float(const document_data& d, const node& n, FloatType& value) noexcept
-{
-    const unsigned int_digits = n.extra & 0xFFu;
-    const unsigned frac_digits = n.extra >> 8u;
-    const unsigned count = int_digits + frac_digits;
-    const auto* const first = reinterpret_cast<const unsigned char*>(d.src + n.off); // NOLINT(cppcoreguidelines-pro-type-reinterpret-cast)
-    const unsigned negative = first[0] == '-' ? 1u : 0u;
-    if (count - 16u > 1u || frac_digits == 0 || n.len != negative + count + 1 || d.size - n.off < 33)
-    {
-        return false;
-    }
-    const unsigned char* const digits = first + negative;
-    const std::uint64_t w = sixteen_digits(digits, int_digits);
-    // (the 17th digit follows the point)
-    const std::uint64_t w17 = (w * 10) + static_cast<std::uint64_t>(digits[17] - '0');
-    using format = ieee_binary_format<std::numeric_limits<FloatType>::digits>;
-    native_float_t<FloatType> result{};
-    float_from_bits(coordinate_bits(-static_cast<std::int64_t>(frac_digits), count == 17 ? w17 : w, format{}) | (static_cast<std::uint64_t>(negative) << format::sign_bit()), result);
-    value = result;
-    return true;
-}
-
-#endif
-
 /// the value of the float token of a node, as parse() converts it; floats and
 /// doubles with at most 19 digits are converted from the digit layout
 template<typename FloatType>
@@ -348,13 +200,6 @@ FloatType float_value(const document_data& d, const node& n)
 template<typename FloatType>
 FloatType float_value(const document_data& d, const node& n, std::true_type /*binary32 or binary64*/)
 {
-#if NLOHMANN_VIEW_VECTOR
-    FloatType value {};
-    if (coordinate_float(d, n, value))
-    {
-        return value;
-    }
-#endif
     const unsigned int_digits = n.extra & 0xFFu;
     const unsigned frac_digits = n.extra >> 8u;
     if (NLOHMANN_VIEW_LIKELY(int_digits + frac_digits <= 19)) // (255 marks "many")

include/nlohmann/detail/view/pointer.hpp

@@ -45,6 +45,7 @@ index_status array_index(const StringType& s, std::size_t& idx) noexcept
 
 /// throws the exception json_pointer::array_index throws for this status
 template<typename StringType>
+/// (without a context, as json_pointer::array_index)
 [[noreturn]] NLOHMANN_VIEW_NOINLINE void throw_array_index_error(index_status status, const StringType& s)
 {
     switch (status)
@@ -83,7 +84,7 @@ View resolve_pointer(View cur, const Tokens& tokens, pointer_mode mode)
             {
                 if (mode == pointer_mode::checked)
                 {
-                    throw_key_not_found(token);
+                    throw_key_not_found(site_of(cur), token);
                 }
                 return View();
             }
@@ -95,7 +96,7 @@ View resolve_pointer(View cur, const Tokens& tokens, pointer_mode mode)
             {
                 if (throwing)
                 {
-                    throw_array_index_past_end(cur.size());
+                    throw_array_index_past_end(site_of(cur), cur.size());
                 }
                 return View();
             }
@@ -114,7 +115,7 @@ View resolve_pointer(View cur, const Tokens& tokens, pointer_mode mode)
             {
                 if (mode == pointer_mode::checked)
                 {
-                    throw_array_index_out_of_range(idx);
+                    throw_array_index_out_of_range(site_of(cur), idx);
                 }
                 return View();
             }
@@ -124,7 +125,7 @@ View resolve_pointer(View cur, const Tokens& tokens, pointer_mode mode)
         {
             if (throwing)
             {
-                throw_out_of_range(exception_id::pointer_unresolved, concat("unresolved reference token '", token, "'"));
+                throw_out_of_range(site_of(cur), exception_id::pointer_unresolved, concat("unresolved reference token '", token, "'"));
             }
             return View();
         }

include/nlohmann/detail/view/serializer.hpp

@@ -14,9 +14,7 @@
 #include <cstddef> // size_t
 #include <cstdint> // uint8_t, uint32_t
 #include <cstring> // memcpy, memset
-#include <functional> // less
 #include <limits> // numeric_limits
-#include <memory> // unique_ptr
 #include <type_traits> // integral_constant
 #include <vector> // vector
 
@@ -386,143 +384,6 @@ struct depth_counter<false>
     }
 };
 
-/*!
-@brief where the double x nearest to a token w * 10^q lies, for 10^15 <= w <
-10^17 and -17 <= q <= -1 (Eisel-Lemire, then a fixed-point position)
-
-dump() writes the shortest decimal in the rounding interval of x (the closest
-one if there are several, see zmij::to_shortest()). Its digits are those of a
-multiple of 10^g of the token's last digit, with g = dec_exp + 1 - q for x's
-decimal exponent dec_exp: the multiple of 10^g nearest to x if it lies in the
-interval (at most one does), else the multiple of 10^(g - 1) nearest to x
-(see token_digits_1() and token_digits_2()). For g <= 0, the token itself is
-in the interval.
-
-The position is exact in a 64-bit fixed-point number with 56 bits after the
-point (of the token's last digit): x / 10^q = 2c * 5^-q / 2^t for x = c * 2^e2
-and t = q + 1 - e2, with at most 56 bits after the point here, and |x / 10^q -
-w| is at most half the spacing of the doubles, below 12 units. So only the low
-64 bits of 2c * 5^-q are needed. c and e2 come from Eisel-Lemire's product
-directly (as eisel_lemire() rounds it: no tie is possible for q < -4, and its
-second product is added where it can carry into the result).
-
-@param[out] offset  x / 10^q - w + 32, with 56 bits after the point
-@param[out] radius  half the spacing of the doubles at x, likewise (+ 1 if
-                    the interval is closed)
-@return g; 3 if the digits are not decided here (a power of two, whose
-        interval is not symmetric, or out of range)
-*/
-NLOHMANN_VIEW_ALWAYS_INLINE int locate_token(std::uint64_t w, int q, std::uint64_t& offset, std::uint64_t& radius) noexcept
-{
-    const int lz = count_leading_zeros(w);
-    const std::uint64_t normalized = w << static_cast<unsigned>(lz);
-    const auto index = static_cast<std::size_t>(2 * (q - pow5_128_smallest_power));
-    const uint128_parts product = full_multiplication(normalized, pow5_128()[index]);
-    std::uint64_t high = product.high;
-    if (NLOHMANN_VIEW_UNLIKELY((high & 0x1FFu) == 0x1FFu))
-    {
-        // (as eisel_lemire(): the lower bits may carry into the result)
-        const std::uint64_t second = full_multiplication(normalized, pow5_128()[index + 1]).high;
-        high += product.low + second < product.low ? 1u : 0u;
-    }
-    const auto upperbit = static_cast<unsigned>(high >> 63u);
-    const std::uint64_t mantissa = high >> (upperbit + 9);
-    const std::uint64_t c = (mantissa + (mantissa & 1u)) >> 1u;
-    // e2 as eisel_lemire() computes the exponent (log2(10) ~ 217706 / 2^16)
-    const int e2 = static_cast<int>((217706 * static_cast<std::int64_t>(q)) >> 16) + 63 + static_cast<int>(upperbit) - lz - 52;
-    const int t = q + 1 - e2;
-    const int g = ((e2 * 315653) >> 20) + 1 - q; // (dec_exp as in zmij::to_shortest(), for a regular x)
-    static const std::array<std::uint64_t, 17> fives = {{1u, 5u, 25u, 125u, 625u, 3125u, 15625u, 78125u, 390625u, 1953125u, 9765625u, 48828125u, 244140625u, 1220703125u, 6103515625u, 30517578125u, 152587890625u}};
-    const std::uint64_t five = fives[static_cast<std::size_t>(-q)]; // 5^-q
-    const unsigned s = static_cast<unsigned>(56 - t) & 63u;
-    offset = ((2 * c * five) << s) - (w << 56u) + (std::uint64_t{32} << 56u);
-    radius = (five << s) + ((c & 1u) == 0 ? 1u : 0u);
-    // (c = 2^52: a power of two; c = 2^53: rounded up to one. With 8 to 16
-    // fraction digits, as the batch's tokens have, t is in [15, 40] and g at
-    // most 2. x lies halfway between two candidates only if its distance from
-    // the token is a whole unit (g == 2) or a half (g == 1): such tokens are
-    // left to the full conversion)
-    const bool decided = c - ((std::uint64_t{1} << 52u) + 1) < (std::uint64_t{1} << 52u) - 1
-                         && (offset & ((std::uint64_t{1} << 55u) - 1)) != 0;
-    return decided ? g : 3;
-}
-
-/// all bits set if b, else none
-NLOHMANN_VIEW_ALWAYS_INLINE std::uint64_t mask_if(bool b) noexcept
-{
-    return std::uint64_t{0} - static_cast<std::uint64_t>(b);
-}
-
-/// a if b, else c, without a branch (the conditions of token_digits_1/2() are
-/// as good as random, and GCC compiled their ?: as branches: one misprediction
-/// per float of canada.json on x86-64)
-NLOHMANN_VIEW_ALWAYS_INLINE std::uint64_t select_if(bool b, std::uint64_t a, std::uint64_t c) noexcept
-{
-    return c ^ ((a ^ c) & mask_if(b));
-}
-
-/// the distance of a position (see locate_token()) from the biased point 32
-NLOHMANN_VIEW_ALWAYS_INLINE std::uint64_t biased_distance(std::uint64_t position) noexcept
-{
-    constexpr std::uint64_t center = std::uint64_t{32} << 56u;
-    // |position - center|, without a branch: negated (complemented, plus 1) if below
-    const std::uint64_t below = mask_if(position < center);
-    return ((position - center) ^ below) - below;
-}
-
-/*!
-@brief (g == 1) the digits dump() writes for a token, from the position of its
-double (see locate_token()): the multiple of 10 nearest to x if it is in the
-interval, else w rounded to x (an exact tie between two candidates, like
-754570942.33984373, whose double is 754570942.33984375, is excluded by
-locate_token())
-
-With |x - w| at most the radius, which is below half the spacing of the
-candidates, the multiple of 10 nearest to x is w - r10 or the one above it.
-*/
-NLOHMANN_VIEW_ALWAYS_INLINE std::uint64_t token_digits_1(std::uint64_t w, std::uint64_t offset, std::uint64_t radius, std::uint64_t r10) noexcept
-{
-    constexpr std::uint64_t one = std::uint64_t{1} << 56u;
-    const std::uint64_t from = offset + (r10 * one); // x - (w - r10) (+ 32)
-    const bool up = from > 37 * one;                  // (nearer to the multiple above)
-    const bool shorter = biased_distance(from - (mask_if(up) & (10 * one))) < radius;
-    const std::uint64_t a = w - r10 + (mask_if(up) & 10u);
-    const std::uint64_t b = w + ((offset + (one / 2)) >> 56u) - 32;                // w rounded to x
-    return select_if(shorter, a, b);
-}
-
-/*!
-@brief (g == 2) the digits dump() writes for a token, as token_digits_1(): the
-multiple of 100 nearest to x if it is in the interval, else the multiple of 10
-nearest to x
-*/
-NLOHMANN_VIEW_ALWAYS_INLINE std::uint64_t token_digits_2(std::uint64_t w, std::uint64_t offset, std::uint64_t radius, std::uint64_t r10, std::uint64_t r100) noexcept
-{
-    constexpr std::uint64_t one = std::uint64_t{1} << 56u;
-    const std::uint64_t from = offset + (r100 * one);
-    const bool up = from > 82 * one;
-    const bool shorter = biased_distance(from - (mask_if(up) & (100 * one))) < radius;
-    const std::uint64_t a = w - r100 + (mask_if(up) & 100u);
-    // x - (w - r10), in [-12, 21]: the multiple of 10 nearest to x is that one,
-    // or one or two above, or one below
-    const std::uint64_t y = offset + (r10 * one);
-    const std::uint64_t steps = static_cast<std::uint64_t>(y > 37 * one) + static_cast<std::uint64_t>(y > 47 * one);
-    const bool down = y < 27 * one;
-    const std::uint64_t b = w - r10 + (10 * steps) - (mask_if(down) & 10u);
-    return select_if(shorter, a, b);
-}
-
-/// the four ASCII digits of x < 10^4, the first in the lowest byte (as
-/// little-endian memory holds them)
-NLOHMANN_VIEW_ALWAYS_INLINE std::uint32_t four_digit_text(std::uint32_t x) noexcept
-{
-    const std::uint32_t ab = (x * 5243u) >> 19u; // x / 100
-    std::uint32_t v = ab | ((x - (ab * 100u)) << 16u);
-    const std::uint32_t a = ((v * 103u) >> 10u) & 0x000F000Fu; // / 10 per 16-bit lane
-    v = a | ((v - (a * 10u)) << 8u);
-    return v + 0x30303030u;
-}
-
 /// how the view's dump() writes a value
 struct dump_style
 {
@@ -850,7 +711,6 @@ class view_serializer
         std::array<frame, 32> buffer; // NOLINT(cppcoreguidelines-pro-type-member-init,hicpp-member-init): written before read
         std::vector<frame> heap;
         small_stack<frame> stack(buffer.data(), buffer.size(), heap);
-        float_batch batch;
         const char* const src = m_doc.src;
         const char* const src_end = src + m_doc.size;
         char* w = m_out.cursor();
@@ -990,7 +850,7 @@ class view_serializer
                     else if (enabled(fast_double::value))
                     {
                         room(64);
-                        w = write_double_ahead(w, src, n, batch);
+                        w = write_double_at(w, *n);
                     }
                     else
                     {
@@ -1015,20 +875,6 @@ class view_serializer
                 case value_t::array:
                 {
                     const bool object = n->kind == node_kind::object;
-                    // (read-only documents, compact output) a pair of floats,
-                    // like the coordinates of GeoJSON, without a frame
-                    if (enabled(!Editable && !Pretty && !SourceNumbers && fast_double::value) && n->next == 3 && n->len == 2
-                            && n[1].kind == node_kind::number_float && n[2].kind == node_kind::number_float)
-                    {
-                        room(2 * 64);
-                        *w++ = '[';
-                        w = write_double_ahead(w, src, n + 1, batch);
-                        *w++ = ',';
-                        w = write_double_ahead(w, src, n + 2, batch);
-                        *w++ = ']';
-                        n += 2; // (the element before the next node)
-                        break;
-                    }
                     room(8);
                     if (n->len == 0)
                     {
@@ -1138,254 +984,6 @@ class view_serializer
         }
     }
 
-    /// (float_batch) the mark of digits whose rounding carried beyond the
-    /// token's last four (see mark())
-    static constexpr std::uint64_t carried() noexcept
-    {
-        return std::uint64_t{1} << 63u;
-    }
-
-    /// floats of 16 or 17 digits ahead of the walk, converted together (see
-    /// convert_ahead()), with their digits as mark() keeps them
-    struct float_batch
-    {
-        static constexpr std::size_t capacity() noexcept
-        {
-            return 256;
-        }
-        /// (on the heap, allocated by the first conversion: a frame of more
-        /// than a page made every dump of a small document slower)
-        struct entries
-        {
-            std::array<const node*, 256> at; // NOLINT(cppcoreguidelines-pro-type-member-init,hicpp-member-init): written before read
-            std::array<std::uint64_t, 256> digits; // NOLINT(cppcoreguidelines-pro-type-member-init,hicpp-member-init): written before read
-        };
-        std::unique_ptr<entries> storage{};
-        const node** at = nullptr;
-        std::uint64_t* digits = nullptr;
-        std::size_t count = 0;
-        std::size_t next = 0;
-        const node* scanned = nullptr; ///< where the last scan of the node array ended
-    };
-
-    /// (doubles) the float at n as dump() writes it, at w (64 bytes of room):
-    /// from the batch if it holds n
-    NLOHMANN_VIEW_ALWAYS_INLINE char* write_double_ahead(char* w, const char* src, const node* n, float_batch& batch)
-    {
-#if NLOHMANN_VIEW_VECTOR && NLOHMANN_VIEW_LITTLE_ENDIAN
-        // (digits 0: not written by the batch; carried(): rounding carried
-        // beyond the last four digits)
-        if (batch.next != batch.count && batch.at[batch.next] == n && batch.digits[batch.next] - 1 < carried() - 1)
-        {
-            return write_token_digits(w, src, *n, batch.digits[batch.next++]);
-        }
-        return write_double_beside(w, n, batch);
-#else
-        static_cast<void>(src);
-        static_cast<void>(batch);
-        return write_double_at(w, *n);
-#endif
-    }
-
-#if NLOHMANN_VIEW_VECTOR && NLOHMANN_VIEW_LITTLE_ENDIAN
-    /// write_double_ahead() for a float that is not the batch's next one (out
-    /// of line: the walk is faster with less code in its loop)
-    NLOHMANN_VIEW_NOINLINE char* write_double_beside(char* w, const node* n, float_batch& batch)
-    {
-        // (an editable document's walk can skip nodes of the array)
-        const std::less<const node*> before{}; // NOLINT(modernize-use-transparent-functors) C++11 has no std::less<>
-        while (batch.next != batch.count && before(batch.at[batch.next], n))
-        {
-            ++batch.next;
-        }
-        if ((batch.next != batch.count && batch.at[batch.next] == n) || convert_ahead(n, batch))
-        {
-            const std::uint64_t digits = batch.digits[batch.next++];
-            if ((digits & carried()) != 0)
-            {
-                return write_carried_digits(w, *n, digits & ~carried());
-            }
-            if (digits != 0)
-            {
-                return write_token_digits(w, m_doc.src, *n, digits);
-            }
-        }
-        return write_double_at(w, *n);
-    }
-
-    /*!
-    @brief convert the floats of 16 or 17 digits among the next nodes of the
-    array at once; whether n is the first of them
-
-    Such tokens (like the coordinates of GeoJSON, written with 17 digits) are
-    written as the token with its last four digits replaced
-    (write_token_digits()). Each step (the digits, locate_token(),
-    token_digits_1/2()) is a long chain of dependent operations; one loop per
-    step over many floats lets the processor overlap the floats, which it does
-    not across the steps of one float (on Apple M1, one loop for the digits and
-    locate_token() made the dump of canada.json 12 % slower). Each node of the array
-    is scanned once; a node that the walk does not reach in order (in an
-    editable document) is written without the batch.
-    */
-    NLOHMANN_VIEW_NOINLINE bool convert_ahead(const node* n, float_batch& batch)
-    {
-        const std::less<const node*> before{}; // NOLINT(modernize-use-transparent-functors) C++11 has no std::less<>
-        const node* const last = m_doc.nodes + m_doc.node_count;
-        if (before(n, batch.scanned) || before(n, m_doc.nodes) || !before(n, last))
-        {
-            return false;
-        }
-        const char* const src = m_doc.src;
-        const char* const src_end = src + m_doc.size;
-        const node* const stop = last - n > 1024 ? n + 1024 : last;
-        batch.count = 0;
-        batch.next = 0;
-        // the digits of each token (and its last four from the text)
-        std::array<std::uint64_t, 256> digits;     // NOLINT(cppcoreguidelines-pro-type-member-init,hicpp-member-init): written before read
-        std::array<int, 256> exponent;             // NOLINT(cppcoreguidelines-pro-type-member-init,hicpp-member-init): written before read
-        std::array<std::uint32_t, 256> last_four;  // NOLINT(cppcoreguidelines-pro-type-member-init,hicpp-member-init): written before read
-        const node* p = n;
-        if (!batch.storage)
-        {
-            batch.storage.reset(new typename float_batch::entries); // NOLINT(cppcoreguidelines-owning-memory,modernize-make-unique): C++11
-            batch.at = batch.storage->at.data();
-            batch.digits = batch.storage->digits.data();
-        }
-        for (; p != stop && batch.count != float_batch::capacity(); ++p)
-        {
-            // a token of the source: 16 or 17 digits, at least 8 after the
-            // point, no exponent, 32 bytes to read
-            const unsigned frac_digits = p->extra >> 8u;
-            const unsigned int_digits = p->extra & 0xFFu;
-            const unsigned count = int_digits + frac_digits;
-            if (p->kind == node_kind::number_float && (p->flags & node_flags::storage) == 0
-                    && count - 16u <= 1u && frac_digits >= 8 && src_end - (src + p->off) >= 33)
-            {
-                const auto* const token = reinterpret_cast<const unsigned char*>(src + p->off); // NOLINT(cppcoreguidelines-pro-type-reinterpret-cast)
-                const unsigned sign = token[0] == '-' ? 1 : 0;
-                const unsigned char* const first = token + sign;
-                const std::uint64_t w = sixteen_digits(first, int_digits);
-                // (the 17th digit follows the point; a token with an exponent
-                // gets digits outside the range of locate_token(), and is not
-                // written by the batch. A token 0.xxx with 16 significant
-                // digits is: its double is written as "0." and its digits too)
-                const std::uint64_t w17 = (w * 10) + static_cast<std::uint64_t>(first[17] - '0');
-                const std::size_t i = batch.count++;
-                batch.at[i] = p;
-                if (p->len != sign + count + 1)
-                {
-                    digits[i] = 0;
-                }
-                else
-                {
-                    digits[i] = count == 17 ? w17 : w;
-                }
-                exponent[i] = -static_cast<int>(frac_digits);
-                // (byte 0: the last digit, byte 1: the last two, bits 16-29: the last four)
-                std::uint32_t four = 0;
-                std::memcpy(&four, token + p->len - 4, 4);
-                four -= 0x30303030u;
-                const std::uint32_t pairs = ((four & 0x0F0F0F0Fu) * 2561u) >> 8u; // (byte 0, 2: the pairs)
-                last_four[i] = (four >> 24u) | ((pairs >> 16u & 0xFFu) << 8u) | ((((pairs & 0x00FF00FFu) * 6553601u) >> 16u) << 16u);
-            }
-        }
-        batch.scanned = p;
-        // the position of each double, and the floats of either kind
-        std::array<std::uint64_t, 256> offset; // NOLINT(cppcoreguidelines-pro-type-member-init,hicpp-member-init): written before read
-        std::array<std::uint64_t, 256> radius; // NOLINT(cppcoreguidelines-pro-type-member-init,hicpp-member-init): written before read
-        std::array<std::uint8_t, 256> kind_1;  // NOLINT(cppcoreguidelines-pro-type-member-init,hicpp-member-init): written before read
-        std::array<std::uint8_t, 256> kind_2;  // NOLINT(cppcoreguidelines-pro-type-member-init,hicpp-member-init): written before read
-        std::size_t count_1 = 0;
-        std::size_t count_2 = 0;
-        for (std::size_t i = 0; i != batch.count; ++i)
-        {
-            const std::uint64_t w = digits[i];
-            // (the range excludes the values of bytes that are not digits, only
-            // in a damaged image loaded with image_check::bounds)
-            const int g = w - 1000000000000000u < 99000000000000000u ? locate_token(w, exponent[i], offset[i], radius[i]) : 3;
-            batch.digits[i] = mask_if(g <= 0) & ((last_four[i] >> 16u) + 1);
-            kind_1[count_1] = static_cast<std::uint8_t>(i);
-            count_1 += static_cast<std::size_t>(g == 1);
-            kind_2[count_2] = static_cast<std::uint8_t>(i);
-            count_2 += static_cast<std::size_t>(g == 2);
-        }
-        // the digits to write, in one loop per kind (see mark())
-        for (std::size_t k = 0; k != count_1; ++k)
-        {
-            const std::size_t i = kind_1[k];
-            const std::uint64_t s = token_digits_1(digits[i], offset[i], radius[i], last_four[i] & 0xFFu);
-            batch.digits[i] = mark(digits[i], s, last_four[i] >> 16u);
-        }
-        for (std::size_t k = 0; k != count_2; ++k)
-        {
-            const std::size_t i = kind_2[k];
-            const std::uint64_t s = token_digits_2(digits[i], offset[i], radius[i], last_four[i] & 0xFFu, (last_four[i] >> 8u) & 0xFFu);
-            batch.digits[i] = mark(digits[i], s, last_four[i] >> 16u);
-        }
-        return batch.count != 0 && batch.at[0] == n;
-    }
-
-    /*!
-    @brief a float token of 16 or 17 digits as dump() writes it, at w (64 bytes
-    of room), from the digits of its double (see locate_token()) whose
-    rounding carried beyond the token's last four digits
-
-    (like 43.42027299999999 -> 43.420273): the digits go to the writer of
-    Zmij's results as its 15 or 16 leading digits and one more, which omits
-    the trailing zeros.
-    */
-    NLOHMANN_VIEW_NOINLINE char* write_carried_digits(char* w, const node& n, std::uint64_t digits) const noexcept
-    {
-        const int exponent = -(n.extra >> 8u);
-        const std::uint64_t leading = digits / 10;
-        const auto last = static_cast<unsigned char>(digits - (leading * 10));
-        const bool seventeen = digits >= 10000000000000000u;
-        *w = '-';
-        w += m_doc.src[n.off] == '-' ? 1 : 0;
-        return ::nlohmann::detail::dtoa_impl::write_shortest(w, seventeen ? ::nlohmann::detail::zmij::shortest_decimal{leading, exponent, last, last != 0}
-                : ::nlohmann::detail::zmij::shortest_decimal{digits, exponent - 1, 0, false});
-    }
-
-    /// the digits s of a token with the digits w and the last four digits
-    /// r4, as the batch keeps them: the new last four digits + 1 (only those
-    /// are written over the token's), or s marked if it differs from w beyond
-    /// them (0 stays for a float the batch does not write)
-    static std::uint64_t mark(std::uint64_t w, std::uint64_t s, std::uint64_t r4) noexcept
-    {
-        const std::uint64_t last = s - (w - r4);
-        return last < 10000 ? last + 1 : s | carried();
-    }
-
-    /*!
-    @brief a float token of 16 or 17 digits as dump() writes it, at w (64 bytes
-    of room): the token with its last four digits replaced by digits - 1 (see
-    mark(); the others are the token's), and without trailing zeros after the
-    point, but one
-    */
-    char* write_token_digits(char* w, const char* src, const node& n, std::uint64_t digits)
-    {
-        const char* const token = src + n.off;
-        std::memcpy(w, token, 32);
-        char* const end = w + n.len;
-        const std::uint32_t last = four_digit_text(static_cast<std::uint32_t>(digits - 1));
-        std::memcpy(end - 4, &last, 4);
-        // (the token's text before the new digits, from the source: reading
-        // the output back would wait for the stores)
-        std::uint64_t before = 0;
-        std::memcpy(&before, token + n.len - 12, 8);
-        const std::uint32_t last_zeros = last ^ 0x30303030u;
-        const std::uint64_t before_zeros = before ^ 0x3030303030303030u;
-        if (NLOHMANN_VIEW_UNLIKELY(last_zeros == 0 && before_zeros == 0))
-        {
-            return write_double_at(w, n); // (twelve or more trailing zeros)
-        }
-        const unsigned zeros = last_zeros != 0 ? (static_cast<unsigned>(count_leading_zeros(last_zeros)) / 8) - 4
-                               : 4 + (static_cast<unsigned>(count_leading_zeros(before_zeros)) / 8);
-        const unsigned frac_digits = n.extra >> 8u;
-        return end - (zeros < frac_digits ? zeros : frac_digits - 1);
-    }
-#endif
-
     /// a float node as dump() writes it
     void write_float_node(const node& n)
     {

include/nlohmann/detail/view/value.hpp

@@ -60,7 +60,7 @@ NLOHMANN_VIEW_ALWAYS_INLINE T arithmetic_value(const document_data& d, const nod
         case value_t::binary:
         case value_t::discarded:
         default:
-            throw_type_error(exception_id::type_mismatch, "type must be number, but is ", value_type_name(static_cast<value_t>(n.kind)));
+            throw_type_error(error_site(&d, &n), exception_id::type_mismatch, "type must be number, but is ", value_type_name(static_cast<value_t>(n.kind)));
     }
 }
 
@@ -70,7 +70,7 @@ std::vector<U, A> vector_value(const View& v)
 {
     if (NLOHMANN_VIEW_UNLIKELY(!v.is_array()))
     {
-        throw_type_error(exception_id::type_mismatch, "type must be array, but is ", v.type_name());
+        throw_type_error(site_of(v), exception_id::type_mismatch, "type must be array, but is ", v.type_name());
     }
     std::vector<U, A> r;
     r.reserve(v.size());
@@ -88,7 +88,7 @@ Map map_value(const View& v)
 {
     if (NLOHMANN_VIEW_UNLIKELY(!v.is_object()))
     {
-        throw_type_error(exception_id::type_mismatch, "type must be object, but is ", v.type_name());
+        throw_type_error(site_of(v), exception_id::type_mismatch, "type must be object, but is ", v.type_name());
     }
     Map r;
     for (auto it = v.begin(); it != v.end(); ++it)

include/nlohmann/json_view.hpp

@@ -285,7 +285,7 @@ class basic_json_view
             {
                 return {};
             }
-            detail::view::throw_type_error(detail::exception_id::subscript_wrong_type, "cannot use operator[] with a string argument with ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::subscript_wrong_type, "cannot use operator[] with a string argument with ", type_name());
         }
         return lookup(key);
     }
@@ -311,7 +311,7 @@ class basic_json_view
             {
                 return {};
             }
-            detail::view::throw_type_error(detail::exception_id::subscript_wrong_type, "cannot use operator[] with a numeric argument with ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::subscript_wrong_type, "cannot use operator[] with a numeric argument with ", type_name());
         }
         return idx < m_node->len ? element(idx) : basic_json_view();
     }
@@ -344,12 +344,12 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(!is_object()))
         {
-            detail::view::throw_type_error(detail::exception_id::at_wrong_type, "cannot use at() with ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::at_wrong_type, "cannot use at() with ", type_name());
         }
         const basic_json_view r = lookup(key);
         if (NLOHMANN_VIEW_UNLIKELY(r.is_discarded()))
         {
-            detail::view::throw_key_not_found(std::string(key.data(), key.size()));
+            detail::view::throw_key_not_found(site(), std::string(key.data(), key.size()));
         }
         return r;
     }
@@ -370,11 +370,11 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(!is_array()))
         {
-            detail::view::throw_type_error(detail::exception_id::at_wrong_type, "cannot use at() with ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::at_wrong_type, "cannot use at() with ", type_name());
         }
         if (NLOHMANN_VIEW_UNLIKELY(idx >= m_node->len))
         {
-            detail::view::throw_array_index_out_of_range(idx);
+            detail::view::throw_array_index_out_of_range(site(), idx);
         }
         return element(idx);
     }
@@ -402,7 +402,7 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(!is_object()))
         {
-            detail::view::throw_type_error(detail::exception_id::value_wrong_type, "cannot use value() with ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::value_wrong_type, "cannot use value() with ", type_name());
         }
         const basic_json_view r = lookup(key);
         return r.is_discarded() ? default_value : r.template get<T>();
@@ -412,7 +412,7 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(!is_object()))
         {
-            detail::view::throw_type_error(detail::exception_id::value_wrong_type, "cannot use value() with ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::value_wrong_type, "cannot use value() with ", type_name());
         }
         // (the default is made only if there is no such member)
         const basic_json_view r = lookup(key);
@@ -427,7 +427,7 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(!is_structured()))
         {
-            detail::view::throw_type_error(detail::exception_id::value_wrong_type, "cannot use value() with ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::value_wrong_type, "cannot use value() with ", type_name());
         }
         const basic_json_view r = detail::view::resolve_pointer(*this, detail::json_pointer_access::reference_tokens(ptr), detail::view::pointer_mode::value);
         return r.is_discarded() ? default_value : r.template get<T>();
@@ -437,7 +437,7 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(!is_structured()))
         {
-            detail::view::throw_type_error(detail::exception_id::value_wrong_type, "cannot use value() with ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::value_wrong_type, "cannot use value() with ", type_name());
         }
         const basic_json_view r = detail::view::resolve_pointer(*this, detail::json_pointer_access::reference_tokens(ptr), detail::view::pointer_mode::value);
         return r.is_discarded() ? string_t(default_value) : r.template get<string_t>();
@@ -450,7 +450,7 @@ class basic_json_view
         const iterator it = begin();
         if (NLOHMANN_VIEW_UNLIKELY(it == end()))
         {
-            detail::view::throw_invalid_iterator(detail::exception_id::iterator_value_unavailable, "cannot get value");
+            detail::view::throw_invalid_iterator(site(), detail::exception_id::iterator_value_unavailable, "cannot get value");
         }
         return *it;
     }
@@ -606,7 +606,7 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(!is_string()))
         {
-            detail::view::throw_type_error(detail::exception_id::type_mismatch, "type must be string, but is ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::type_mismatch, "type must be string, but is ", type_name());
         }
         return {m_doc->str(*m_node), m_node->len};
     }
@@ -618,7 +618,7 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(!is_number()))
         {
-            detail::view::throw_type_error(detail::exception_id::type_mismatch, "type must be number, but is ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::type_mismatch, "type must be number, but is ", type_name());
         }
         return {m_doc->str(*m_node), detail::view::number_length(*m_node)};
     }
@@ -764,12 +764,19 @@ class basic_json_view
     template<typename, view_mode> friend class basic_json_view;
     template<typename, typename> friend class detail::view::editor;
     template<typename, typename> friend class detail::view::view_side;
+    friend struct detail::view::view_access;
     friend iterator;
 
     basic_json_view(const document_data* d, const node* n) noexcept
         : m_doc(d), m_node(n)
     {}
 
+    /// this value as the context of an exception
+    detail::view::error_site site() const noexcept
+    {
+        return detail::view::error_site(m_doc, m_node);
+    }
+
     using json_side_t = detail::view::json_side<BasicJsonType, string_view_t>;
 
     static detail::view::view_side<BasicJsonType, basic_json_view> side(const basic_json_view& v) noexcept
@@ -876,7 +883,7 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(!is_boolean()))
         {
-            detail::view::throw_type_error(detail::exception_id::type_mismatch, "type must be boolean, but is ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::type_mismatch, "type must be boolean, but is ", type_name());
         }
         return (m_node->flags & detail::view::node_flags::is_true) != 0;
     }
@@ -886,7 +893,7 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(m_node == nullptr))
         {
-            detail::view::throw_type_error(detail::exception_id::type_mismatch, "type must be number, but is ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::type_mismatch, "type must be number, but is ", type_name());
         }
         return detail::view::arithmetic_value<T, BasicJsonType>(*m_doc, *m_node);
     }
@@ -895,7 +902,7 @@ class basic_json_view
     {
         if (NLOHMANN_VIEW_UNLIKELY(!is_null()))
         {
-            detail::view::throw_type_error(detail::exception_id::type_mismatch, "type must be null, but is ", type_name());
+            detail::view::throw_type_error(site(), detail::exception_id::type_mismatch, "type must be null, but is ", type_name());
         }
         return nullptr;
     }
@@ -1309,7 +1316,7 @@ class basic_json_document
     template<typename I, typename V, typename std::enable_if<detail::view::is_non_bool_integral<I>::value, int>::type = 0>
     view_type set(view_type array, I idx, V && value)
     {
-        return editor().set(array, index(idx), std::forward<V>(value));
+        return editor().set(array, index(array, idx), std::forward<V>(value));
     }
 
     /// set the value at a JSON pointer: its parent must exist; an object
@@ -1346,7 +1353,7 @@ class basic_json_document
                 const std::size_t idx = pointer_index(token);
                 if (idx >= 0xFFFFFFFFu)
                 {
-                    detail::view::throw_array_index_out_of_range(idx);
+                    detail::view::throw_array_index_out_of_range(detail::view::site_of(parent), idx);
                 }
                 for (std::size_t i = 0; i < idx; ++i)
                 {
@@ -1380,7 +1387,7 @@ class basic_json_document
     template<typename I, typename V, typename std::enable_if<detail::view::is_non_bool_integral<I>::value, int>::type = 0>
     view_type insert(view_type array, I idx, V && value)
     {
-        return editor().insert(array, index(idx), std::forward<V>(value));
+        return editor().insert(array, index(array, idx), std::forward<V>(value));
     }
 
     /// remove all members with this key; returns their number
@@ -1393,7 +1400,7 @@ class basic_json_document
     template<typename I, typename std::enable_if<detail::view::is_non_bool_integral<I>::value, int>::type = 0>
     void erase(view_type array, I idx)
     {
-        editor().erase(array, index(idx));
+        editor().erase(array, index(array, idx));
     }
 
     /// remove the value at a JSON pointer; returns the number of removed
@@ -1427,25 +1434,25 @@ class basic_json_document
         return detail::view::editor<BasicJsonType, view_type>(*m_data);
     }
 
-    /// an index (out_of_range.401 if negative)
+    /// an index into @a array (out_of_range.401 if negative)
     template<typename I>
-    static std::size_t index(I idx)
+    static std::size_t index(const view_type& array, I idx)
     {
-        return index(idx, std::is_signed<I> {});
+        return index(array, idx, std::is_signed<I> {});
     }
 
     template<typename I>
-    static std::size_t index(I idx, std::true_type /*signed*/)
+    static std::size_t index(const view_type& array, I idx, std::true_type /*signed*/)
     {
         if (idx < 0)
         {
-            detail::view::throw_array_index_out_of_range(idx);
+            detail::view::throw_array_index_out_of_range(detail::view::site_of(array), idx);
         }
         return static_cast<std::size_t>(idx);
     }
 
     template<typename I>
-    static std::size_t index(I idx, std::false_type /*unsigned*/)
+    static std::size_t index(const view_type& /*array*/, I idx, std::false_type /*unsigned*/)
     {
         return static_cast<std::size_t>(idx);
     }

single_include/nlohmann/json.hpp

@@ -2720,7 +2720,7 @@ JSON_HEDLEY_DIAGNOSTIC_POP
             return ej_pair.second == j;                                                         \
         });                                                                                     \
         if (it != std::end(m)) e = it->first;                                                   \
-        else ::nlohmann::detail::templated_json_throw<nlohmann::detail::out_of_range>(nlohmann::detail::out_of_range::create(nlohmann::detail::exception_id::value_out_of_range, nlohmann::detail::concat("enum value out of range for " #ENUM_TYPE ": ", nlohmann::detail::from_json_dump(j)), nlohmann::detail::exception_context(j))); \
+        else ::nlohmann::detail::templated_json_throw<nlohmann::detail::out_of_range>(nlohmann::detail::out_of_range::create(nlohmann::detail::exception_id::value_out_of_range, nlohmann::detail::concat(nlohmann::detail::exception_prefix(j), "enum value out of range for " #ENUM_TYPE ": ", nlohmann::detail::from_json_dump(j)), nlohmann::detail::exception_context(j))); \
     }
 
 // Ugly macros to avoid uglier copy-paste when specializing basic_json. They
@@ -5928,6 +5928,28 @@ std::nullptr_t exception_context(const View& /*unused*/) noexcept
     return nullptr;
 }
 
+namespace view
+{
+// the path and bytes of the value of a view (defined in <nlohmann/detail/view/errors.hpp>)
+template<typename View>
+std::string diagnostics_of(const View& v);
+}  // namespace view
+
+// what goes in front of the message of an exception about j: nothing for a
+// basic_json value (exception_context() passes the value itself), the context
+// basic_json would add (JSON_DIAGNOSTICS, JSON_DIAGNOSTIC_POSITIONS) for a view
+template<typename BasicJsonType, enable_if_t<is_basic_json<BasicJsonType>::value, int> = 0>
+const char* exception_prefix(const BasicJsonType& /*unused*/) noexcept
+{
+    return "";
+}
+
+template<typename View, enable_if_t<is_json_view<View>::value, int> = 0>
+std::string exception_prefix(const View& v)
+{
+    return view::diagnostics_of(v);
+}
+
 /*!
 @brief throws because @a j does not have the type a conversion expects
 @param[in] expected  the expected type(s), e.g. "array" or "binary or array"
@@ -5939,7 +5961,7 @@ JSON_HEDLEY_NO_RETURN JSON_HEDLEY_NEVER_INLINE void throw_type_must_be(const cha
 {
     static_cast<void>(expected); // unused when JSON_NOEXCEPTION is defined
     static_cast<void>(j);
-    JSON_THROW(type_error::create(exception_id::type_mismatch, concat("type must be ", expected, ", but is ", j.type_name()), exception_context(j)));
+    JSON_THROW(type_error::create(exception_id::type_mismatch, concat(exception_prefix(j), "type must be ", expected, ", but is ", j.type_name()), exception_context(j)));
 }
 
 /*!

tests/src/json_view_test_helpers.hpp

@@ -15,6 +15,7 @@
 #include <nlohmann/json_view.hpp>
 
 #include <algorithm> // any_of
+#include <cstddef> // size_t
 #include <functional> // function
 #include <random> // mt19937
 #include <string> // string
@@ -88,6 +89,26 @@ inline bool has_duplicate_keys(const nlohmann::ordered_json_view& v)
     });
 }
 
+// the message of an exception about the value at path (a JSON pointer, "" for
+// the root) whose source bytes are [start, end): the context basic_json adds
+// with JSON_DIAGNOSTICS and JSON_DIAGNOSTIC_POSITIONS, if enabled, then what
+inline std::string message(const std::string& id, const std::string& path, std::size_t start, std::size_t end, const std::string& what)
+{
+    std::string context;
+#if JSON_DIAGNOSTIC_POSITIONS
+    context = "(bytes " + std::to_string(start) + "-" + std::to_string(end) + ") ";
+#else
+    static_cast<void>(start);
+    static_cast<void>(end);
+#endif
+#if JSON_DIAGNOSTICS
+    context = path.empty() ? "" : "(" + path + ") " + context;
+#else
+    static_cast<void>(path);
+#endif
+    return "[json.exception." + id + "] " + context + what;
+}
+
 #if !defined(JSON_NOEXCEPTION)
 // the exception a call throws, or "" if it throws none
 inline std::string exception_of_call(const std::function<void()>& f)

tests/src/unit-json_view.cpp

@@ -225,7 +225,7 @@ TEST_CASE("json_view")
 #if JSON_DIAGNOSTICS
         // the parents are set, so errors name the path
         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&);
+        CHECK_THROWS_WITH_AS(m.at("a").at("b").at(0).at("x"), json_view_test::message("type_error.304", "/a/b/0", 11, 12, "cannot use at() with number").c_str(), json::type_error&);
 #endif
     }
 
@@ -750,19 +750,6 @@ std::string exception_of(F f)
     return "";
 }
 
-// the exception a call on a basic_json value throws, without the byte range of
-// the value (JSON_DIAGNOSTIC_POSITIONS), which the exceptions of views lack
-template<typename F>
-std::string json_exception_of(F f)
-{
-    std::string what = exception_of(f);
-    const std::size_t at = what.find("] (bytes ");
-    if (at != std::string::npos)
-    {
-        what.erase(at + 2, what.find(") ", at) - at);
-    }
-    return what;
-}
 #endif
 
 // compares a view with the ordered_json value materialize() gives for it:
@@ -971,24 +958,25 @@ TEST_CASE("json_view element access and iteration")
 #if !defined(JSON_NOEXCEPTION)
             if (!j.is_object())
             {
-                CHECK(exception_of([&] { static_cast<void>(v["a"]); }) == json_exception_of([&] { static_cast<void>(j["a"]); }));
+                CHECK(exception_of([&] { static_cast<void>(v["a"]); }) == exception_of([&] { static_cast<void>(j["a"]); }));
             }
             if (!j.is_array())
             {
-                CHECK(exception_of([&] { static_cast<void>(v[0]); }) == json_exception_of([&] { static_cast<void>(j[0]); }));
+                CHECK(exception_of([&] { static_cast<void>(v[0]); }) == exception_of([&] { static_cast<void>(j[0]); }));
             }
-            CHECK(exception_of([&] { static_cast<void>(v.at("a")); }) == json_exception_of([&] { static_cast<void>(j.at("a")); }));
-            CHECK(exception_of([&] { static_cast<void>(v.at("missing")); }) == json_exception_of([&] { static_cast<void>(j.at("missing")); }));
-            CHECK(exception_of([&] { static_cast<void>(v.at(0)); }) == json_exception_of([&] { static_cast<void>(j.at(0)); }));
-            CHECK(exception_of([&] { static_cast<void>(v.at(5)); }) == json_exception_of([&] { static_cast<void>(j.at(5)); }));
+            CHECK(exception_of([&] { static_cast<void>(v.at("a")); }) == exception_of([&] { static_cast<void>(j.at("a")); }));
+            CHECK(exception_of([&] { static_cast<void>(v.at("missing")); }) == exception_of([&] { static_cast<void>(j.at("missing")); }));
+            CHECK(exception_of([&] { static_cast<void>(v.at(0)); }) == exception_of([&] { static_cast<void>(j.at(0)); }));
+            CHECK(exception_of([&] { static_cast<void>(v.at(5)); }) == exception_of([&] { static_cast<void>(j.at(5)); }));
             if (!(j.is_object() && j.empty())) // (key() of an end iterator)
             {
-                CHECK(exception_of([&] { static_cast<void>(v.begin().key()); }) == json_exception_of([&] { static_cast<void>(j.begin().key()); }));
+                // (without the context of basic_json: view iterators do not know the value they iterate over)
+                CHECK(exception_of([&] { static_cast<void>(v.begin().key()); }) == (j.is_object() ? "" : "[json.exception.invalid_iterator.207] cannot use key() for non-object iterators"));
             }
             if (!j.empty() || j.is_null())
             {
-                CHECK(exception_of([&] { static_cast<void>(v.front()); }) == json_exception_of([&] { static_cast<void>(j.front()); }));
-                CHECK(exception_of([&] { static_cast<void>(v.back()); }) == json_exception_of([&] { static_cast<void>(j.back()); }));
+                CHECK(exception_of([&] { static_cast<void>(v.front()); }) == exception_of([&] { static_cast<void>(j.front()); }));
+                CHECK(exception_of([&] { static_cast<void>(v.back()); }) == exception_of([&] { static_cast<void>(j.back()); }));
             }
 #endif
             CHECK(v.contains("a") == j.contains("a"));
@@ -1000,8 +988,8 @@ TEST_CASE("json_view element access and iteration")
         const json_document d = json_document::parse(R"({"a":[]})");
         CHECK(d.root()["b"].is_discarded());
         CHECK(d.root()["a"][0].is_discarded());
-        CHECK_THROWS_WITH_AS(d.root()["a"].front(), "[json.exception.invalid_iterator.214] cannot get value", json::invalid_iterator&);
-        CHECK_THROWS_WITH_AS(d.root()["a"].back(), "[json.exception.invalid_iterator.214] cannot get value", json::invalid_iterator&);
+        CHECK_THROWS_WITH_AS(d.root()["a"].front(), json_view_test::message("invalid_iterator.214", "/a", 5, 7, "cannot get value").c_str(), json::invalid_iterator&);
+        CHECK_THROWS_WITH_AS(d.root()["a"].back(), json_view_test::message("invalid_iterator.214", "/a", 5, 7, "cannot get value").c_str(), json::invalid_iterator&);
         const json_view invalid{};
         CHECK(invalid.begin() == invalid.end());
         CHECK(std::string(invalid.type_name()) == "discarded");
@@ -1032,10 +1020,10 @@ TEST_CASE("json_view element access and iteration")
         CHECK(v[json_view::json_pointer("/a/b/5")][0].is_discarded());
 #if !defined(JSON_NOEXCEPTION)
         // type errors on values that are not discarded stay
-        CHECK_THROWS_WITH_AS(v[0], "[json.exception.type_error.305] cannot use operator[] with a numeric argument with object", json::type_error&);
-        CHECK_THROWS_WITH_AS(v["a"]["b"]["c"], "[json.exception.type_error.305] cannot use operator[] with a string argument with array", json::type_error&);
-        CHECK_THROWS_WITH_AS(v["s"]["c"], "[json.exception.type_error.305] cannot use operator[] with a string argument with string", json::type_error&);
-        CHECK_THROWS_WITH_AS(v["s"][0], "[json.exception.type_error.305] cannot use operator[] with a numeric argument with string", json::type_error&);
+        CHECK_THROWS_WITH_AS(v[0], json_view_test::message("type_error.305", "", 0, 29, "cannot use operator[] with a numeric argument with object").c_str(), json::type_error&);
+        CHECK_THROWS_WITH_AS(v["a"]["b"]["c"], json_view_test::message("type_error.305", "/a/b", 10, 17, "cannot use operator[] with a string argument with array").c_str(), json::type_error&);
+        CHECK_THROWS_WITH_AS(v["s"]["c"], json_view_test::message("type_error.305", "/s", 23, 28, "cannot use operator[] with a string argument with string").c_str(), json::type_error&);
+        CHECK_THROWS_WITH_AS(v["s"][0], json_view_test::message("type_error.305", "/s", 23, 28, "cannot use operator[] with a numeric argument with string").c_str(), json::type_error&);
         CHECK_THROWS_AS(v["a"]["b"][0]["c"], json::type_error&);
         CHECK_THROWS_AS(v["s"][json_view::json_pointer("/x")], json::out_of_range&);
         // at() keeps throwing on a discarded view
@@ -1110,9 +1098,9 @@ TEST_CASE("json_view element access and iteration")
         CHECK(v[(std::numeric_limits<std::size_t>::max)()].is_discarded());
         CHECK(v[std::numeric_limits<int>::max()].is_discarded());
 #if !defined(JSON_NOEXCEPTION)
-        CHECK_THROWS_WITH_AS(v.at(3), "[json.exception.out_of_range.401] array index 3 is out of range", json::out_of_range&);
-        CHECK_THROWS_WITH_AS(v.at(3u), "[json.exception.out_of_range.401] array index 3 is out of range", json::out_of_range&);
-        CHECK_THROWS_WITH_AS(v.at(std::int64_t(3)), "[json.exception.out_of_range.401] array index 3 is out of range", json::out_of_range&);
+        CHECK_THROWS_WITH_AS(v.at(3), json_view_test::message("out_of_range.401", "", 0, 10, "array index 3 is out of range").c_str(), json::out_of_range&);
+        CHECK_THROWS_WITH_AS(v.at(3u), json_view_test::message("out_of_range.401", "", 0, 10, "array index 3 is out of range").c_str(), json::out_of_range&);
+        CHECK_THROWS_WITH_AS(v.at(std::int64_t(3)), json_view_test::message("out_of_range.401", "", 0, 10, "array index 3 is out of range").c_str(), json::out_of_range&);
         CHECK_THROWS_AS(v.at(-1), json::out_of_range&);
         CHECK_THROWS_AS(v.at(-1L), json::out_of_range&);
         CHECK_THROWS_AS(v.at(std::int64_t(-1)), json::out_of_range&);
@@ -1246,26 +1234,6 @@ TEST_CASE("json_view element access and iteration")
 
 namespace
 {
-#if !defined(JSON_NOEXCEPTION)
-// an exception message without the context that basic_json adds with
-// JSON_DIAGNOSTICS ("(/path) ") and JSON_DIAGNOSTIC_POSITIONS ("(bytes 1-2) ");
-// the view's exceptions have no such context
-std::string without_path(std::string msg)
-{
-    for (const char* prefix :
-            {"] (/", "] (bytes "
-            })
-    {
-        const std::size_t open = msg.find(prefix);
-        if (open != std::string::npos)
-        {
-            msg.erase(open + 2, msg.find(") ", open) + 2 - (open + 2));
-        }
-    }
-    return msg;
-}
-#endif
-
 // the bits of a float, to compare values bit for bit
 std::uint64_t bits(double x)
 {
@@ -1336,18 +1304,18 @@ void check_values(const ordered_json_view& v, const ordered_json& j, const std::
     // conversions to the wrong type throw what basic_json throws
     if (!j.is_number())
     {
-        CHECK(exception_of([&] { static_cast<void>(v.get<int>()); }) == without_path(exception_of([&] { static_cast<void>(j.get<int>()); })));
+        CHECK(exception_of([&] { static_cast<void>(v.get<int>()); }) == exception_of([&] { static_cast<void>(j.get<int>()); }));
     }
-    CHECK(exception_of([&] { static_cast<void>(v.get<bool>()); }) == without_path(exception_of([&] { static_cast<void>(j.get<bool>()); })));
-    CHECK(exception_of([&] { static_cast<void>(v.get<std::string>()); }) == without_path(exception_of([&] { static_cast<void>(j.get<std::string>()); })));
-    CHECK(exception_of([&] { static_cast<void>(v.get<std::nullptr_t>()); }) == without_path(exception_of([&] { static_cast<void>(j.get<std::nullptr_t>()); })));
+    CHECK(exception_of([&] { static_cast<void>(v.get<bool>()); }) == exception_of([&] { static_cast<void>(j.get<bool>()); }));
+    CHECK(exception_of([&] { static_cast<void>(v.get<std::string>()); }) == exception_of([&] { static_cast<void>(j.get<std::string>()); }));
+    CHECK(exception_of([&] { static_cast<void>(v.get<std::nullptr_t>()); }) == exception_of([&] { static_cast<void>(j.get<std::nullptr_t>()); }));
     if (!j.is_array())
     {
-        CHECK(exception_of([&] { static_cast<void>(v.get<std::vector<int>>()); }) == without_path(exception_of([&] { static_cast<void>(j.get<std::vector<int>>()); })));
+        CHECK(exception_of([&] { static_cast<void>(v.get<std::vector<int>>()); }) == exception_of([&] { static_cast<void>(j.get<std::vector<int>>()); }));
     }
     if (!j.is_object())
     {
-        CHECK(exception_of([&] { static_cast<void>(v.get<std::map<std::string, int>>()); }) == without_path(exception_of([&] { static_cast<void>(j.get<std::map<std::string, int>>()); })));
+        CHECK(exception_of([&] { static_cast<void>(v.get<std::map<std::string, int>>()); }) == exception_of([&] { static_cast<void>(j.get<std::map<std::string, int>>()); }));
     }
 #endif
 
@@ -1490,8 +1458,8 @@ TEST_CASE("json_view values")
         CHECK(v[3].number_token() == "123456789012345678901234567890");
         CHECK(v[4].number_token() == "-12");
         CHECK(v[5].number_token() == "7");
-        CHECK_THROWS_WITH_AS(v[6].number_token(), "[json.exception.type_error.302] type must be number, but is string", json::type_error&);
-        CHECK_THROWS_WITH_AS(v.get_string(), "[json.exception.type_error.302] type must be string, but is array", json::type_error&);
+        CHECK_THROWS_WITH_AS(v[6].number_token(), json_view_test::message("type_error.302", "/6", 56, 59, "type must be number, but is string").c_str(), json::type_error&);
+        CHECK_THROWS_WITH_AS(v.get_string(), json_view_test::message("type_error.302", "", 0, 60, "type must be string, but is array").c_str(), json::type_error&);
     }
 
     SECTION("conversions")
@@ -1565,15 +1533,15 @@ TEST_CASE("json_view values")
         CHECK(v["o"]["x"].value(json::json_pointer("/1"), 0) == j["o"]["x"].value(json::json_pointer("/1"), 0));
         CHECK(v["o"]["x"].value(json::json_pointer("/7"), 3) == j["o"]["x"].value(json::json_pointer("/7"), 3));
 #if !defined(JSON_NOEXCEPTION)
-        CHECK(exception_of([&] { static_cast<void>(v["o"]["x"].value("k", 0)); }) == without_path(exception_of([&] { static_cast<void>(j["o"]["x"].value("k", 0)); })));
-        CHECK(exception_of([&] { static_cast<void>(v.value("s", 0)); }) == without_path(exception_of([&] { static_cast<void>(j.value("s", 0)); })));
-        CHECK(exception_of([&] { static_cast<void>(v["n"].value("x", 0)); }) == without_path(exception_of([&] { static_cast<void>(j["n"].value("x", 0)); })));
-        CHECK(exception_of([&] { static_cast<void>(v["n"].value(json::json_pointer("/x"), 0)); }) == without_path(exception_of([&] { static_cast<void>(j["n"].value(json::json_pointer("/x"), 0)); })));
+        CHECK(exception_of([&] { static_cast<void>(v["o"]["x"].value("k", 0)); }) == exception_of([&] { static_cast<void>(j["o"]["x"].value("k", 0)); }));
+        CHECK(exception_of([&] { static_cast<void>(v.value("s", 0)); }) == exception_of([&] { static_cast<void>(j.value("s", 0)); }));
+        CHECK(exception_of([&] { static_cast<void>(v["n"].value("x", 0)); }) == exception_of([&] { static_cast<void>(j["n"].value("x", 0)); }));
+        CHECK(exception_of([&] { static_cast<void>(v["n"].value(json::json_pointer("/x"), 0)); }) == exception_of([&] { static_cast<void>(j["n"].value(json::json_pointer("/x"), 0)); }));
         // string defaults: the member is not a string, or the value is neither an object nor an array
-        CHECK(exception_of([&] { static_cast<void>(v.value("n", "default")); }) == without_path(exception_of([&] { static_cast<void>(j.value("n", "default")); })));
-        CHECK(exception_of([&] { static_cast<void>(v["n"].value("x", "default")); }) == without_path(exception_of([&] { static_cast<void>(j["n"].value("x", "default")); })));
-        CHECK(exception_of([&] { static_cast<void>(v.value(json::json_pointer("/n"), "default")); }) == without_path(exception_of([&] { static_cast<void>(j.value(json::json_pointer("/n"), "default")); })));
-        CHECK(exception_of([&] { static_cast<void>(v["s"].value(json::json_pointer("/x"), "default")); }) == without_path(exception_of([&] { static_cast<void>(j["s"].value(json::json_pointer("/x"), "default")); })));
+        CHECK(exception_of([&] { static_cast<void>(v.value("n", "default")); }) == exception_of([&] { static_cast<void>(j.value("n", "default")); }));
+        CHECK(exception_of([&] { static_cast<void>(v["n"].value("x", "default")); }) == exception_of([&] { static_cast<void>(j["n"].value("x", "default")); }));
+        CHECK(exception_of([&] { static_cast<void>(v.value(json::json_pointer("/n"), "default")); }) == exception_of([&] { static_cast<void>(j.value(json::json_pointer("/n"), "default")); }));
+        CHECK(exception_of([&] { static_cast<void>(v["s"].value(json::json_pointer("/x"), "default")); }) == exception_of([&] { static_cast<void>(j["s"].value(json::json_pointer("/x"), "default")); }));
 #endif
         // (defaults and members longer than the short string optimization)
         CHECK(v.value("s", "a default longer than fifteen characters") == "text");
@@ -1632,7 +1600,7 @@ TEST_CASE("json_view JSON pointers")
         {
             CAPTURE(pointer)
             const json::json_pointer p(pointer);
-            const std::string at_error = without_path(exception_of([&] { static_cast<void>(j.at(p)); }));
+            const std::string at_error = exception_of([&] { static_cast<void>(j.at(p)); });
             CHECK(exception_of([&] { static_cast<void>(v.at(p)); }) == at_error);
             if (at_error.empty())
             {
@@ -1646,7 +1614,7 @@ TEST_CASE("json_view JSON pointers")
             }
             else
             {
-                CHECK(exception_of([&] { static_cast<void>(v[p]); }) == without_path(exception_of([&] { static_cast<void>(j[p]); })));
+                CHECK(exception_of([&] { static_cast<void>(v[p]); }) == exception_of([&] { static_cast<void>(j[p]); }));
             }
             // (basic_json::contains() throws out_of_range.404 for an empty
             // array index token, although it is not meant to throw; the view
@@ -1657,7 +1625,7 @@ TEST_CASE("json_view JSON pointers")
                 static_cast<void>(found);
             });
             CHECK(v.contains(p) == (contains_error.empty() && j.contains(p)));
-            CHECK(exception_of([&] { static_cast<void>(v.value(p, 5)); }) == without_path(exception_of([&] { static_cast<void>(j.value(p, 5)); })));
+            CHECK(exception_of([&] { static_cast<void>(v.value(p, 5)); }) == exception_of([&] { static_cast<void>(j.value(p, 5)); }));
         }
     }
 #endif

tests/src/unit-json_view_dump.cpp

@@ -35,9 +35,9 @@ using json_view_test::has_duplicate_keys;
 #include <string>
 #include <vector>
 
-// These tests were split off unit-json_view.cpp, whose object file got too large for the MinGW
-// linker (see json_view_test_helpers.hpp). Like that file, this one is also built
-// with C++17, as it mentions JSON_HAS_CPP_17.
+// These tests were split off unit-json_view.cpp, whose object file got too large
+// for the MinGW linker (see json_view_test_helpers.hpp). Like that file, this one
+// is also built with C++17, as it mentions JSON_HAS_CPP_17.
 
 TEST_CASE("json_view dump")
 {
@@ -150,88 +150,6 @@ TEST_CASE("json_view dump")
             CHECK(many_doc.root().dump() == json::parse(many_tokens).dump());
         }
 
-        // tokens of 16 and 17 digits without an exponent (like GeoJSON
-        // coordinates), which dump() converts in batches: doubles exactly
-        // halfway between two candidates for the shortest digits, random
-        // digits, and digits near a shorter decimal (...99999997,
-        // ...00000003), whose rounding carries; in pairs and among other
-        // values, so that the batches are interrupted; the last token near the
-        // end of the text
-        {
-            std::mt19937_64 tokens(5627); // NOLINT(cert-msc32-c,cert-msc51-cpp,bugprone-random-generator-seed)
-            const auto draw = [&tokens](std::size_t n)
-            {
-                const std::uint64_t r = tokens() % n;
-                return static_cast<std::size_t>(r);
-            };
-            std::string coordinates = "[754570942.33984373,257769445.56835938,858106901.94921875,97944964.876953125,"
-                                      "[0.5,1e-7],[1.5,-0.0,2.5],[1E2,12.5,0.1,3.25,4.5,5.75,6.125,7.0,8.5],[],[1.5],[1,2.5],[2.5,\"x\"],[[1.5],2.5],";
-            for (int i = 0; i < 20000; ++i)
-            {
-                const std::size_t length = 16 + draw(2);
-                std::string digits(1, static_cast<char>('1' + draw(9)));
-                const std::size_t kept = draw(3) == 0 ? length : 2 + draw(length - 3);
-                for (std::size_t k = 1; k < kept; ++k)
-                {
-                    digits += static_cast<char>('0' + draw(10));
-                }
-                digits += std::string(length - kept, draw(2) == 0 ? '9' : '0');
-                digits.back() = static_cast<char>('0' + draw(10));
-                const std::size_t point = 1 + draw(9);
-                const std::string token = (draw(2) == 0 ? "-" : "") + digits.substr(0, point) + "." + digits.substr(point);
-                if (i != 0)
-                {
-                    coordinates += ',';
-                }
-                if (draw(8) == 0)
-                {
-                    coordinates += R"({"x":)";
-                    coordinates += token;
-                    coordinates += R"(,"s":"a"})";
-                }
-                else
-                {
-                    coordinates += '[';
-                    coordinates += token;
-                    coordinates += ',';
-                    coordinates += token;
-                    coordinates += ']';
-                }
-            }
-            coordinates += ",-65.613616999999977]";
-            const json_document parsed = json_document::parse(coordinates);
-            const ordered_json j = ordered_json::parse(coordinates);
-            CHECK(parsed.root().dump() == j.dump());
-            CHECK(parsed.root().dump(1) == j.dump(1));
-            CHECK(parsed.root().dump(-1, ' ', true) == j.dump(-1, ' ', true));
-            CHECK(parsed.root().back().get<double>() == j.back().get<double>());
-            CHECK(parsed.root().back().dump() == "-65.61361699999998");
-            std::size_t i = 0;
-            for (const json_view v : parsed.root())
-            {
-                const ordered_json& e = j[i++];
-                CHECK(v.dump() == e.dump());
-                if (v.is_array() && !v.empty() && v[0].is_number())
-                {
-                    CHECK(v[0].get<double>() == e[0].get<double>());
-                }
-            }
-
-            // an editable document: edits move some arrays, so that the walk
-            // leaves the order of the parsed nodes
-            ordered_json_editable_document editable = ordered_json_editable_document::parse(coordinates);
-            ordered_json expected = j;
-            for (std::size_t k = 0; k < 20000; k += 97)
-            {
-                if (expected[k].is_array())
-                {
-                    editable.push_back(editable.root()[k], ordered_json(1.5));
-                    expected[k].push_back(1.5);
-                }
-            }
-            CHECK(editable.root().dump() == expected.dump());
-        }
-
         // random doubles, written as parse() and dump() would
         std::mt19937_64 rng(1170); // NOLINT(cert-msc32-c,cert-msc51-cpp,bugprone-random-generator-seed)
         std::string many = "[";
@@ -253,4 +171,673 @@ TEST_CASE("json_view dump")
         const nlohmann::basic_json_document<json_float> float_document = nlohmann::basic_json_document<json_float>::parse("[0.1, 1.5e10, 3.4028235e38]");
         CHECK(float_document.root().dump() == json_float::parse("[0.1, 1.5e10, 3.4028235e38]").dump());
     }
+
+    SECTION("members in document order, all of them")
+    {
+        const json_document d = json_document::parse(R"({"b": 1, "a": 2, "b": 3})");
+        CHECK(d.root().dump() == R"({"b":1,"a":2,"b":3})");
+        CHECK(d.root().dump(1) == "{\n \"b\": 1,\n \"a\": 2,\n \"b\": 3\n}");
+        // dump() writes all members, though the lookup finds the first
+        CHECK(d.root()["b"].dump() == "1");
+    }
+
+    SECTION("deep nesting")
+    {
+        const std::string deep = std::string(100000, '[') + std::string(100000, ']');
+        const json_document deep_document = json_document::parse(deep);
+        CHECK(deep_document.root().dump() == deep);
+    }
+
+    SECTION("ensure_ascii: a character that needs an escape at each position of a string")
+    {
+        // string lengths around the width of the vector steps, in a string value, a key, and a value near the end of the text
+        for (const char* const special :
+                {
+                    "\xC3\xA9", "\xE4\xB8\xAD", "\xF0\x9F\x98\x80", "\x7F", "\\n", "\\u0001", "\\\"", "\\\\"
+                })
+        {
+            for (std::size_t length = 0; length <= 40; ++length)
+            {
+                for (std::size_t position = 0; position <= length; ++position)
+                {
+                    std::string content(length, 'a');
+                    content.insert(position, special);
+                    const std::array<std::string, 3> texts =
+                    {
+                        {
+                            "\"" + content + "\"",
+                            "[\"" + content + "\"]",
+                            "{\"" + content + "\":1}",
+                        }
+                    };
+                    for (const std::string& text : texts)
+                    {
+                        CAPTURE(text)
+                        const ordered_json_document d = ordered_json_document::parse(text);
+                        const ordered_json j = ordered_json::parse(text);
+                        CHECK(d.root().dump(-1, ' ', true) == j.dump(-1, ' ', true));
+                        CHECK(d.root().dump(2, ' ', true) == j.dump(2, ' ', true));
+                        CHECK(d.root().dump() == j.dump());
+                    }
+                }
+            }
+        }
+    }
+
+    SECTION("ensure_ascii: every character of the Basic Multilingual Plane and some beyond")
+    {
+        // (the two- and three-byte sequences, including the ranges that need a check: E0, ED, EF BF BF)
+        std::string text = "[";
+        std::size_t in_chunk = 0;
+        const auto append = [&](std::uint32_t cp)
+        {
+            if (cp < 0x80)
+            {
+                if (cp < 0x20 || cp == '"' || cp == '\\')
+                {
+                    return;
+                }
+                text += static_cast<char>(cp);
+            }
+            else if (cp < 0x800)
+            {
+                text += static_cast<char>(0xC0 | (cp >> 6));
+                text += static_cast<char>(0x80 | (cp & 0x3F));
+            }
+            else if (cp < 0x10000)
+            {
+                text += static_cast<char>(0xE0 | (cp >> 12));
+                text += static_cast<char>(0x80 | ((cp >> 6) & 0x3F));
+                text += static_cast<char>(0x80 | (cp & 0x3F));
+            }
+            else
+            {
+                text += static_cast<char>(0xF0 | (cp >> 18));
+                text += static_cast<char>(0x80 | ((cp >> 12) & 0x3F));
+                text += static_cast<char>(0x80 | ((cp >> 6) & 0x3F));
+                text += static_cast<char>(0x80 | (cp & 0x3F));
+            }
+        };
+        for (std::uint32_t cp = 0x20; cp < 0x10400; ++cp)
+        {
+            if (cp >= 0xD800 && cp < 0xE000)
+            {
+                continue;
+            }
+            if (in_chunk == 0)
+            {
+                text += text.size() > 1 ? ",\"" : "\"";
+            }
+            append(cp);
+            if (++in_chunk == 37)
+            {
+                text += "\"";
+                in_chunk = 0;
+            }
+        }
+        text += in_chunk != 0 ? "\"]" : "]";
+        const ordered_json_document d = ordered_json_document::parse(text);
+        const ordered_json j = ordered_json::parse(text);
+        CHECK(d.root().dump(-1, ' ', true) == j.dump(-1, ' ', true));
+        CHECK(d.root().dump(2, ' ', true) == j.dump(2, ' ', true));
+        CHECK(d.root().dump() == j.dump());
+        CHECK(d.root().dump(2) == j.dump(2));
+    }
+
+    SECTION("deep nesting, indented")
+    {
+        // indentations of up to and beyond the width of the writer's fixed-size copy
+        for (const int depth :
+                {
+                    1, 5, 20, 70, 300
+                })
+        {
+            std::string text;
+            for (int i = 0; i < depth; ++i)
+            {
+                text += i % 2 == 0 ? "[1," : R"({"k":2,"\u00e9":)";
+            }
+            text += "[]";
+            for (int i = depth - 1; i >= 0; --i)
+            {
+                text += i % 2 == 0 ? "]" : "}";
+            }
+            const ordered_json_document d = ordered_json_document::parse(text);
+            const ordered_json j = ordered_json::parse(text);
+            for (const int indent :
+                    {
+                        0, 1, 2, 3, 8
+                    })
+            {
+                CHECK(d.root().dump(indent) == j.dump(indent));
+                CHECK(d.root().dump(indent, '\t') == j.dump(indent, '\t'));
+            }
+        }
+    }
+
+    SECTION("the output buffer of a small value is small")
+    {
+        // an escaped key after the value: its node lies in the arena, so the
+        // source extent of the value cannot be read from the next node
+        const std::string big(100000, 'a');
+        const std::string text = R"({"small":1,"list":[1,2,3],"k\n":")" + big + R"("})";
+        const json_document d = json_document::parse(text);
+        const auto small = d.root()["small"].dump();
+        CHECK(small == "1");
+        CHECK(small.capacity() < 4096);
+        const auto list = d.root()["list"].dump();
+        CHECK(list == "[1,2,3]");
+        CHECK(list.capacity() < 4096);
+        CHECK(d.root()["list"].dump(2).capacity() < 4096);
+
+        // the whole document and the large value are unaffected
+        CHECK(d.root().dump() == ordered_json::parse(text).dump());
+        CHECK(d.root()["k\n"].dump() == "\"" + big + "\"");
+    }
+
+    SECTION("output that outgrows the estimate")
+    {
+        // ensure_ascii writes six bytes for each two-byte character
+        std::string chars;
+        for (int i = 0; i < 5000; ++i)
+        {
+            chars += "\xC3\xA9";
+        }
+        const json_document d = json_document::parse(R"({"a":")" + chars + R"(","k\n":1})");
+        const json expected = json::parse("\"" + chars + "\"");
+        CHECK(d.root()["a"].dump(-1, ' ', true) == expected.dump(-1, ' ', true));
+        CHECK(d.root()["a"].dump(-1, ' ', true).size() == 2 + (5000 * 6));
+    }
+
+    SECTION("ensure_ascii with strings of the source, decoded strings, and edited strings")
+    {
+        // plain ASCII, non-ASCII, escaped (decoded), and mixed strings, as values and as keys;
+        // the second text has a DEL, which ensure_ascii writes as an escape
+        for (const char* const del :
+                {"", "\x7F"
+                })
+        {
+            const std::string text = std::string("{\"plain\":\"abc\",\"caf\xC3\xA9\":\"\xC3\xA9\xE2\x82\xAC\xF0\x9F\x98\x80\",")
+                                     + "\"esc\\n\\\"\":\"a\\tb\\u00e4\\ud83d\\ude00\",\"del\":\"a" + del + "b\",\"\":\"\"," // NOLINT(modernize-raw-string-literal)
+                                     + "\"list\":[\"x\",\"\xC3\xBC\",\"\\\\\",\"long ASCII string with more than thirty-two bytes in it\","
+                                     + "\"long string with more than thirty-two bytes and one non-ASCII byte \xC3\xB6\"],"
+                                     + "\"n\":[1,2.5,-0,true,null]}";
+            const ordered_json j = ordered_json::parse(text);
+            const ordered_json_document d = ordered_json_document::parse(text);
+            CHECK(d.root().dump(-1, ' ', true) == j.dump(-1, ' ', true));
+            CHECK(d.root().dump(4, ' ', true) == j.dump(4, ' ', true));
+            CHECK(d.root().dump() == j.dump());
+            CHECK(d.root()["list"].dump(-1, ' ', true) == j["list"].dump(-1, ' ', true));
+            CHECK(d.root()["del"].dump(-1, ' ', true) == j["del"].dump(-1, ' ', true));
+
+            // strings that an edit wrote (the source strings that they replace were plain)
+            ordered_json_editable_document e = ordered_json_editable_document::parse(text);
+            ordered_json k = j;
+            e.set(e.root()["plain"], std::string("\xC3\xA4gypten"));
+            k["plain"] = "\xC3\xA4gypten";
+            e.set(e.root()["list"][0], std::string("ascii after edit"));
+            k["list"][0] = "ascii after edit";
+            e.set(e.root()["list"][1], std::string("\xC3\xA4"));
+            k["list"][1] = "\xC3\xA4";
+            e.set(e.root()["del"], std::string("tab\there"));
+            k["del"] = "tab\there";
+            CHECK(e.root().dump(-1, ' ', true) == k.dump(-1, ' ', true));
+            CHECK(e.root().dump(2, ' ', true) == k.dump(2, ' ', true));
+            CHECK(e.root().dump() == k.dump());
+
+            // strings that were edited from or to ASCII text
+            e.set(e.root()["list"][2], std::string("\xC3\xA4"));
+            k["list"][2] = "\xC3\xA4";
+            e.set(e.root()["list"][4], std::string("plain again"));
+            k["list"][4] = "plain again";
+            CHECK(e.root().dump(-1, ' ', true) == k.dump(-1, ' ', true));
+        }
+    }
+
+    SECTION("ensure_ascii with DEL or a non-ASCII character at every position")
+    {
+        // the check whether a string of the source has no byte >= 0x7F reads
+        // eight bytes at a time and then the rest one by one
+        for (std::size_t length = 1; length <= 20; ++length)
+        {
+            for (std::size_t pos = 0; pos < length; ++pos)
+            {
+                for (const char* const c :
+                        {
+                            "\x7F", "\x7E", "\xC3\xA4", "\xC3\xBF" // NOLINT(modernize-raw-string-literal)
+                        })
+                {
+                    const std::string s = std::string(pos, 'a') + c + std::string(length - pos - 1, 'b');
+                    std::string text = "[\"";
+                    text.append(s).append("\",{\"").append(s).append("\":1}]");
+                    const json_document d = json_document::parse(text);
+                    const json j = json::parse(text);
+                    CHECK(d.root().dump(-1, ' ', true) == j.dump(-1, ' ', true));
+                    CHECK(d.root().dump(1, ' ', true) == j.dump(1, ' ', true));
+                }
+            }
+        }
+    }
+
+    SECTION("streams and discarded views")
+    {
+        const json_document d = json_document::parse(R"({"a": [1, 2]})");
+        std::ostringstream compact;
+        compact << d.root();
+        CHECK(compact.str() == R"({"a":[1,2]})");
+        std::ostringstream pretty;
+        pretty << std::setw(2) << std::setfill('.') << d.root() << d.root()["a"];
+        CHECK(pretty.str() == "{\n..\"a\": [\n....1,\n....2\n..]\n}[1,2]");
+        CHECK(json_view().dump() == json(json::value_t::discarded).dump());
+    }
+}
+
+TEST_CASE("json_view large objects")
+{
+    // objects with 128 members or more are looked up with a hash index
+    for (const std::size_t members :
+            {
+                127u, 128u, 129u, 10000u
+            })
+    {
+        CAPTURE(members)
+        std::string text = "{";
+        for (std::size_t i = 0; i < members; ++i)
+        {
+            text += (i != 0 ? ",\"" : "\"") + std::string(i % 23, 'k') + std::to_string(i) + (i % 7 == 0 ? "\\n" : "") + "\":" + std::to_string(i);
+        }
+        text += R"(,"":"empty key","k1":"a duplicate of an earlier key"})";
+        const json_document d = json_document::parse(text);
+        const json_view v = d.root();
+        const json j = json::parse(text);
+        for (std::size_t i = 0; i < members; ++i)
+        {
+            const std::string key = std::string(i % 23, 'k') + std::to_string(i) + (i % 7 == 0 ? "\n" : "");
+            CHECK(v[key].get<std::size_t>() == i);
+            CHECK(v.contains(key));
+            CHECK(v.find(key).key() == key);
+            CHECK(v.at(key).get<std::size_t>() == i);
+            CHECK(!v.contains(key + "x"));
+        }
+        CHECK(v[""].get_string() == "empty key");
+        CHECK(v["k1"].get<int>() == 1); // the first of duplicate keys, as for small objects
+        CHECK(!v.contains("missing"));
+        CHECK_THROWS_WITH_AS(v.at("missing"), json_view_test::message("out_of_range.403", "", 0, text.size(), "key 'missing' not found").c_str(), json::out_of_range&);
+        CHECK(v == j);
+        CHECK(v.materialize() == j);
+    }
+
+    SECTION("duplicate keys: the first member wins, with and without a table")
+    {
+        // an object of `total` members: the keys "k0".."k<n-1>" in order, then
+        // three keys repeated twice more (one copy in the middle, one at the
+        // end), and two keys repeated once; the value of a member is its
+        // position, so that the first member of a key can be told apart
+        struct member
+        {
+            std::string key;
+            std::size_t position;
+        };
+        const auto make_members = [](std::size_t total)
+        {
+            std::vector<std::string> keys;
+            for (std::size_t i = 0; i + 8 < total; ++i)
+            {
+                keys.push_back("k" + std::to_string(i));
+            }
+            const std::size_t n = keys.size();
+            const std::array<std::size_t, 3> triple = {{3, 17, n - 1}};
+            const std::array<std::size_t, 2> twice = {{5, n / 2}};
+            std::vector<std::string> ordered = keys;
+            for (const std::size_t i : triple)
+            {
+                ordered.insert(ordered.begin() + static_cast<std::ptrdiff_t>(ordered.size() / 2), keys[i]);
+            }
+            for (const std::size_t i : twice)
+            {
+                ordered.insert(ordered.begin() + static_cast<std::ptrdiff_t>(ordered.size() / 3), keys[i]);
+            }
+            for (const std::size_t i : triple)
+            {
+                ordered.push_back(keys[i]);
+            }
+            std::vector<member> result;
+            result.reserve(ordered.size());
+            for (std::size_t i = 0; i < ordered.size(); ++i)
+            {
+                result.push_back({ordered[i], i});
+            }
+            return result;
+        };
+
+        // 100 and 127 members: no table; 128 members and more: a table
+        for (const std::size_t total :
+                {
+                    100u, 127u, 128u, 200u, 5000u
+                })
+        {
+            CAPTURE(total)
+            const std::vector<member> members = make_members(total);
+            REQUIRE(members.size() >= total);
+            REQUIRE(members.size() >= 100);
+            std::string text = "{";
+            std::map<std::string, std::size_t> first;
+            std::map<std::string, std::size_t> last;
+            std::map<std::string, std::size_t> occurrences;
+            for (const member& m : members)
+            {
+                text += (text.size() > 1 ? ",\"" : "\"") + m.key + "\":" + std::to_string(m.position);
+                first.insert({m.key, m.position});
+                last[m.key] = m.position;
+                ++occurrences[m.key];
+            }
+            text += '}';
+            REQUIRE(first.size() < members.size());
+
+            const json_document d = json_document::parse(text);
+            const json_view v = d.root();
+            const json j = json::parse(text);
+            CHECK(v.size() == members.size()); // every occurrence is visited
+            const json m = v.materialize();
+            for (const auto& entry : first)
+            {
+                CAPTURE(entry.first)
+                const std::size_t expected = entry.second;
+                CHECK(v[entry.first].get<std::size_t>() == expected);
+                CHECK(v.at(entry.first).get<std::size_t>() == expected);
+                CHECK(v.find(entry.first).value().get<std::size_t>() == expected);
+                CHECK(v.value(entry.first, std::size_t{0}) == expected);
+                CHECK(v.contains(entry.first));
+                CHECK(v.count(entry.first) == occurrences[entry.first]);
+                const std::string pointer = "/" + entry.first;
+                CHECK(v[json::json_pointer(pointer)].get<std::size_t>() == expected);
+                CHECK(v.at(json::json_pointer(pointer)).get<std::size_t>() == expected);
+                CHECK(v.value(json::json_pointer(pointer), std::size_t{0}) == expected);
+                CHECK(v.contains(json::json_pointer(pointer)));
+                // materialize() and parse() keep the last value instead
+                CHECK(j[entry.first].get<std::size_t>() == last.at(entry.first));
+                CHECK(m[entry.first].get<std::size_t>() == last.at(entry.first));
+            }
+            CHECK(!v.contains("k"));
+            CHECK(v["missing"].is_discarded());
+            CHECK(v.materialize() == j);
+            CHECK(v == j);
+        }
+    }
+
+    SECTION("colliding keys")
+    {
+        // the hash is not seeded, so keys that all land in one place must not
+        // make the table build quadratic: such an object gets no table and is
+        // searched linearly
+        constexpr std::size_t members = 300;
+        constexpr std::size_t table_size = 1024; // the table size for 300 members: the next power of two >= 600
+        std::vector<std::string> colliding;
+        std::vector<std::string> spread;
+        for (std::uint64_t counter = 0; colliding.size() < members || spread.size() < members; ++counter)
+        {
+            std::string key(8, 'a');
+            std::uint64_t x = counter;
+            for (std::size_t i = 0; i < 8; ++i, x /= 26)
+            {
+                key[i] = static_cast<char>('a' + (x % 26));
+            }
+            const bool lands_in_slot_zero = (nlohmann::detail::view::key_hash(key.data(), key.size()) & (table_size - 1)) == 0;
+            if (lands_in_slot_zero && colliding.size() < members)
+            {
+                colliding.push_back(key);
+            }
+            else if (!lands_in_slot_zero && spread.size() < members)
+            {
+                spread.push_back(key);
+            }
+        }
+
+        const auto make_text = [](const std::vector<std::string>& keys)
+        {
+            std::string text = "{";
+            for (std::size_t i = 0; i < keys.size(); ++i)
+            {
+                text += (i != 0 ? ",\"" : "\"") + keys[i] + "\":" + std::to_string(i % 10);
+            }
+            return text + "}";
+        };
+        const std::string colliding_text = make_text(colliding);
+        const std::string spread_text = make_text(spread);
+
+        json_document with_collisions = json_document::parse(colliding_text);
+        json_document without_collisions = json_document::parse(spread_text);
+        for (const auto* pair :
+                {
+                    &colliding, &spread
+                })
+        {
+            const json_view v = (pair == &colliding ? with_collisions : without_collisions).root();
+            for (std::size_t i = 0; i < members; ++i)
+            {
+                CAPTURE(i)
+                CHECK(v[(*pair)[i]].get<std::size_t>() == i % 10);
+                CHECK(v.at((*pair)[i]).get<std::size_t>() == i % 10);
+                CHECK(v.find((*pair)[i]).key() == (*pair)[i]);
+                CHECK(!v.contains((*pair)[i] + "x"));
+            }
+            CHECK(!v.contains("missing"));
+        }
+        CHECK(with_collisions.root() == json::parse(colliding_text));
+
+        // only the object with the spread keys got a table (both texts have the
+        // same length, so the tables are the only difference)
+        with_collisions.shrink_to_fit();
+        without_collisions.shrink_to_fit();
+        CHECK(without_collisions.memory_usage() >= with_collisions.memory_usage() + (table_size * sizeof(std::uint32_t)));
+    }
+
+    SECTION("keys that differ in their last bytes are spread over the table")
+    {
+        // The hash must spread the usual families of keys (numbered names, hex
+        // strings, dates, short names): a key farther than index_max_displacement
+        // from its home slot would leave the object without a table. The keys
+        // are placed with linear probing, as the table does.
+        const auto max_displacement = [](const std::vector<std::string>& keys)
+        {
+            std::size_t slots = 16;
+            while (slots < 2 * keys.size())
+            {
+                slots *= 2;
+            }
+            std::vector<bool> used(slots, false);
+            std::size_t farthest = 0;
+            for (const std::string& key : keys)
+            {
+                std::size_t i = nlohmann::detail::conditional_static_cast<std::size_t>(nlohmann::detail::view::key_hash(key.data(), key.size())) & (slots - 1);
+                std::size_t distance = 0;
+                while (used[i])
+                {
+                    i = (i + 1) & (slots - 1);
+                    ++distance;
+                }
+                used[i] = true;
+                farthest = (std::max)(farthest, distance);
+            }
+            return farthest;
+        };
+        // the keys of a printf-like pattern with one %d, %0Nd or %0Nx, formatted with a stream
+        // (a format string that is not a literal is a Flawfinder error)
+        const auto numbered = [](const char* format, int count, int step)
+        {
+            const std::string pattern(format);
+            const std::size_t percent = pattern.find('%');
+            std::size_t conversion = percent + 1;
+            std::size_t width = 0;
+            while (pattern[conversion] >= '0' && pattern[conversion] <= '9')
+            {
+                width = (width * 10) + static_cast<std::size_t>(pattern[conversion] - '0');
+                ++conversion;
+            }
+            const bool hex = pattern[conversion] == 'x';
+            std::vector<std::string> keys;
+            for (int i = 0; i < count; ++i)
+            {
+                std::ostringstream key;
+                key << pattern.substr(0, percent) << std::setfill('0') << std::setw(static_cast<int>(width))
+                    << (hex ? std::hex : std::dec) << (i * step) << pattern.substr(conversion + 1);
+                keys.push_back(key.str());
+            }
+            return keys;
+        };
+        for (const char* format :
+                {"key%d", "%d", "item_%04d", "a%05d", "%03dxyz", "id%d", "user_%d", "%d_suffix", "0x%04x", "A%07d", "12345678%d", "1234567%d", "123456789%d",
+                 "configuration_parameter_value_%04d_extra", "2026-01-%02dT00:00:00Z"
+                })
+        {
+            CAPTURE(format)
+            CHECK(max_displacement(numbered(format, std::string(format).find("%02d") != std::string::npos ? 60 : 1000, 1)) < nlohmann::detail::view::index_max_displacement); // NOLINT(abseil-string-find-str-contains)
+            CHECK(max_displacement(numbered(format, 600, 7)) < nlohmann::detail::view::index_max_displacement);
+        }
+        std::mt19937_64 rng(7); // NOLINT(cert-msc32-c,cert-msc51-cpp,bugprone-random-generator-seed)
+        for (const std::size_t length :
+                {
+                    std::size_t{1}, std::size_t{2}, std::size_t{3}, std::size_t{5}, std::size_t{7}, std::size_t{8}, std::size_t{9}, std::size_t{12}, std::size_t{16}, std::size_t{17}, std::size_t{33}
+                })
+        {
+            CAPTURE(length)
+            std::set<std::string> distinct;
+            // (there are 26 and 676 keys of one and two letters)
+            std::size_t wanted = 700;
+            if (length == 1)
+            {
+                wanted = 26;
+            }
+            else if (length == 2)
+            {
+                wanted = 500;
+            }
+            while (distinct.size() < wanted)
+            {
+                std::string key(length, 'a');
+                for (char& c : key)
+                {
+                    c = static_cast<char>('a' + (rng() % 26));
+                }
+                distinct.insert(key);
+            }
+            CHECK(max_displacement(std::vector<std::string>(distinct.begin(), distinct.end())) < nlohmann::detail::view::index_max_displacement);
+        }
+    }
+
+    SECTION("keys of every length that differ in one byte")
+    {
+        // In a large object, a key is found by comparing its bytes with those of the member the
+        // table points to: up to 16 bytes as two overlapping words, longer ones with memcmp.
+        // Keys of every length from 0 to 40, and keys that differ from them in one byte.
+        std::mt19937_64 rng(2026); // NOLINT(cert-msc32-c,cert-msc51-cpp,bugprone-random-generator-seed)
+        std::vector<std::string> keys;
+        std::set<std::string> distinct;
+        for (std::size_t length = 0; length <= 40; ++length)
+        {
+            for (int copy = 0; copy < 5; ++copy)
+            {
+                std::string key(length, 'a');
+                for (char& c : key)
+                {
+                    c = static_cast<char>('a' + (rng() % 26));
+                }
+                if (distinct.insert(key).second)
+                {
+                    keys.push_back(key);
+                }
+            }
+        }
+        std::string text = "{";
+        for (std::size_t i = 0; i < keys.size(); ++i)
+        {
+            text += (i != 0 ? ",\"" : "\"") + keys[i] + "\":" + std::to_string(i);
+        }
+        text += "}";
+        REQUIRE(keys.size() >= 128); // (a table is built for 128 members)
+        const json_document d = json_document::parse(text);
+        const json_view v = d.root();
+        for (std::size_t i = 0; i < keys.size(); ++i)
+        {
+            CAPTURE(keys[i])
+            CHECK(v[keys[i]].get<std::size_t>() == i);
+            CHECK(v.at(keys[i]).get<std::size_t>() == i);
+            CHECK(v.contains(keys[i]));
+            for (std::size_t at = 0; at < keys[i].size(); ++at)
+            {
+                std::string other = keys[i];
+                other[at] = other[at] == 'z' ? 'y' : 'z';
+                CAPTURE(other)
+                CHECK(v.contains(other) == (distinct.count(other) != 0));
+            }
+            CHECK(v.contains(keys[i] + "a") == (distinct.count(keys[i] + "a") != 0));
+            if (!keys[i].empty())
+            {
+                CHECK(v.contains(keys[i].substr(1)) == (distinct.count(keys[i].substr(1)) != 0));
+            }
+        }
+    }
+
+    SECTION("shrink_to_fit releases the tables' spare capacity")
+    {
+        const auto make_text = [](int objects, int members)
+        {
+            std::string text = "[";
+            for (int object = 0; object < objects; ++object)
+            {
+                text += object != 0 ? ",{" : "{";
+                for (int i = 0, n = members + object; i < n; ++i)
+                {
+                    text += (i != 0 ? ",\"" : "\"") + std::to_string(i) + "\":" + std::to_string(i);
+                }
+                text += '}';
+            }
+            return text + "]";
+        };
+        const std::string small_text = make_text(5, 150);
+        const std::string big_text = make_text(40, 400);
+
+        // reading a big text, and then a small one, leaves the spare capacity
+        // of the big one: shrink_to_fit() brings the document to the size of
+        // one parsed from the small text alone
+        json_document d = json_document::parse(big_text);
+        const std::size_t big = d.memory_usage();
+        d.read(small_text);
+        CHECK(d.memory_usage() >= big);
+        d.shrink_to_fit();
+        json_document fresh = json_document::parse(small_text);
+        fresh.shrink_to_fit();
+        CHECK(d.memory_usage() == fresh.memory_usage());
+        CHECK(d.memory_usage() < big / 2);
+        CHECK(d.root() == json::parse(small_text));
+        for (int object = 0; object < 5; ++object)
+        {
+            const json_view v = d.root()[static_cast<std::size_t>(object)];
+            for (int i = 0, n = 150 + object; i < n; ++i)
+            {
+                CHECK(v[std::to_string(i)].get<int>() == i);
+            }
+        }
+    }
+
+    SECTION("nested, reused, and in arrays")
+    {
+        std::string inner = "{";
+        for (int i = 0; i < 300; ++i)
+        {
+            inner += (i != 0 ? ",\"m" : "\"m") + std::to_string(i) + "\":" + std::to_string(i);
+        }
+        inner += '}';
+        const std::string text = "[" + inner + ",{\"x\":" + inner + "}," + inner + "]";
+        json_document d = json_document::parse(text);
+        CHECK(d.root()[0]["m299"].get<int>() == 299);
+        CHECK(d.root()[1]["x"]["m150"].get<int>() == 150);
+        CHECK(d.root()[2]["m0"].get<int>() == 0);
+        const std::size_t with_index = d.memory_usage();
+        d.read(std::string("{\"small\": 1}"));
+        CHECK(d.root()["small"].get<int>() == 1);
+        d.read(text);
+        CHECK(d.root()[2]["m7"].get<int>() == 7);
+        CHECK(d.memory_usage() >= with_index / 2);
+    }
 }

tests/src/unit-json_view_dump_escape.cpp

@@ -1,298 +0,0 @@
-//     __ _____ _____ _____
-//  __|  |   __|     |   | |  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;
-using nlohmann::ordered_json_view;
-
-#include <algorithm>
-#include <array>
-#include <cmath>
-#include <cstddef>
-#include <cstdint>
-#include <cstring>
-#include <iomanip>
-#include <map>
-#include <random>
-#include <set>
-#include <sstream>
-#include <string>
-#include <vector>
-
-// These tests were split off unit-json_view_dump.cpp, whose object file got too large for the MinGW
-// linker (see json_view_test_helpers.hpp). Like that file, this one is also built
-// with C++17, as it mentions JSON_HAS_CPP_17.
-
-TEST_CASE("json_view dump: escaping, nesting, and output buffers")
-{
-    SECTION("members in document order, all of them")
-    {
-        const json_document d = json_document::parse(R"({"b": 1, "a": 2, "b": 3})");
-        CHECK(d.root().dump() == R"({"b":1,"a":2,"b":3})");
-        CHECK(d.root().dump(1) == "{\n \"b\": 1,\n \"a\": 2,\n \"b\": 3\n}");
-        // dump() writes all members, though the lookup finds the first
-        CHECK(d.root()["b"].dump() == "1");
-    }
-
-    SECTION("deep nesting")
-    {
-        const std::string deep = std::string(100000, '[') + std::string(100000, ']');
-        const json_document deep_document = json_document::parse(deep);
-        CHECK(deep_document.root().dump() == deep);
-    }
-
-    SECTION("ensure_ascii: a character that needs an escape at each position of a string")
-    {
-        // string lengths around the width of the vector steps, in a string value, a key, and a value near the end of the text
-        for (const char* const special :
-                {
-                    "\xC3\xA9", "\xE4\xB8\xAD", "\xF0\x9F\x98\x80", "\x7F", "\\n", "\\u0001", "\\\"", "\\\\"
-                })
-        {
-            for (std::size_t length = 0; length <= 40; ++length)
-            {
-                for (std::size_t position = 0; position <= length; ++position)
-                {
-                    std::string content(length, 'a');
-                    content.insert(position, special);
-                    const std::array<std::string, 3> texts =
-                    {
-                        {
-                            "\"" + content + "\"",
-                            "[\"" + content + "\"]",
-                            "{\"" + content + "\":1}",
-                        }
-                    };
-                    for (const std::string& text : texts)
-                    {
-                        CAPTURE(text)
-                        const ordered_json_document d = ordered_json_document::parse(text);
-                        const ordered_json j = ordered_json::parse(text);
-                        CHECK(d.root().dump(-1, ' ', true) == j.dump(-1, ' ', true));
-                        CHECK(d.root().dump(2, ' ', true) == j.dump(2, ' ', true));
-                        CHECK(d.root().dump() == j.dump());
-                    }
-                }
-            }
-        }
-    }
-
-    SECTION("ensure_ascii: every character of the Basic Multilingual Plane and some beyond")
-    {
-        // (the two- and three-byte sequences, including the ranges that need a check: E0, ED, EF BF BF)
-        std::string text = "[";
-        std::size_t in_chunk = 0;
-        const auto append = [&](std::uint32_t cp)
-        {
-            if (cp < 0x80)
-            {
-                if (cp < 0x20 || cp == '"' || cp == '\\')
-                {
-                    return;
-                }
-                text += static_cast<char>(cp);
-            }
-            else if (cp < 0x800)
-            {
-                text += static_cast<char>(0xC0 | (cp >> 6));
-                text += static_cast<char>(0x80 | (cp & 0x3F));
-            }
-            else if (cp < 0x10000)
-            {
-                text += static_cast<char>(0xE0 | (cp >> 12));
-                text += static_cast<char>(0x80 | ((cp >> 6) & 0x3F));
-                text += static_cast<char>(0x80 | (cp & 0x3F));
-            }
-            else
-            {
-                text += static_cast<char>(0xF0 | (cp >> 18));
-                text += static_cast<char>(0x80 | ((cp >> 12) & 0x3F));
-                text += static_cast<char>(0x80 | ((cp >> 6) & 0x3F));
-                text += static_cast<char>(0x80 | (cp & 0x3F));
-            }
-        };
-        for (std::uint32_t cp = 0x20; cp < 0x10400; ++cp)
-        {
-            if (cp >= 0xD800 && cp < 0xE000)
-            {
-                continue;
-            }
-            if (in_chunk == 0)
-            {
-                text += text.size() > 1 ? ",\"" : "\"";
-            }
-            append(cp);
-            if (++in_chunk == 37)
-            {
-                text += "\"";
-                in_chunk = 0;
-            }
-        }
-        text += in_chunk != 0 ? "\"]" : "]";
-        const ordered_json_document d = ordered_json_document::parse(text);
-        const ordered_json j = ordered_json::parse(text);
-        CHECK(d.root().dump(-1, ' ', true) == j.dump(-1, ' ', true));
-        CHECK(d.root().dump(2, ' ', true) == j.dump(2, ' ', true));
-        CHECK(d.root().dump() == j.dump());
-        CHECK(d.root().dump(2) == j.dump(2));
-    }
-
-    SECTION("deep nesting, indented")
-    {
-        // indentations of up to and beyond the width of the writer's fixed-size copy
-        for (const int depth :
-                {
-                    1, 5, 20, 70, 300
-                })
-        {
-            std::string text;
-            for (int i = 0; i < depth; ++i)
-            {
-                text += i % 2 == 0 ? "[1," : R"({"k":2,"\u00e9":)";
-            }
-            text += "[]";
-            for (int i = depth - 1; i >= 0; --i)
-            {
-                text += i % 2 == 0 ? "]" : "}";
-            }
-            const ordered_json_document d = ordered_json_document::parse(text);
-            const ordered_json j = ordered_json::parse(text);
-            for (const int indent :
-                    {
-                        0, 1, 2, 3, 8
-                    })
-            {
-                CHECK(d.root().dump(indent) == j.dump(indent));
-                CHECK(d.root().dump(indent, '\t') == j.dump(indent, '\t'));
-            }
-        }
-    }
-
-    SECTION("the output buffer of a small value is small")
-    {
-        // an escaped key after the value: its node lies in the arena, so the
-        // source extent of the value cannot be read from the next node
-        const std::string big(100000, 'a');
-        const std::string text = R"({"small":1,"list":[1,2,3],"k\n":")" + big + R"("})";
-        const json_document d = json_document::parse(text);
-        const auto small = d.root()["small"].dump();
-        CHECK(small == "1");
-        CHECK(small.capacity() < 4096);
-        const auto list = d.root()["list"].dump();
-        CHECK(list == "[1,2,3]");
-        CHECK(list.capacity() < 4096);
-        CHECK(d.root()["list"].dump(2).capacity() < 4096);
-
-        // the whole document and the large value are unaffected
-        CHECK(d.root().dump() == ordered_json::parse(text).dump());
-        CHECK(d.root()["k\n"].dump() == "\"" + big + "\"");
-    }
-
-    SECTION("output that outgrows the estimate")
-    {
-        // ensure_ascii writes six bytes for each two-byte character
-        std::string chars;
-        for (int i = 0; i < 5000; ++i)
-        {
-            chars += "\xC3\xA9";
-        }
-        const json_document d = json_document::parse(R"({"a":")" + chars + R"(","k\n":1})");
-        const json expected = json::parse("\"" + chars + "\"");
-        CHECK(d.root()["a"].dump(-1, ' ', true) == expected.dump(-1, ' ', true));
-        CHECK(d.root()["a"].dump(-1, ' ', true).size() == 2 + (5000 * 6));
-    }
-
-    SECTION("ensure_ascii with strings of the source, decoded strings, and edited strings")
-    {
-        // plain ASCII, non-ASCII, escaped (decoded), and mixed strings, as values and as keys;
-        // the second text has a DEL, which ensure_ascii writes as an escape
-        for (const char* const del :
-                {"", "\x7F"
-                })
-        {
-            const std::string text = std::string("{\"plain\":\"abc\",\"caf\xC3\xA9\":\"\xC3\xA9\xE2\x82\xAC\xF0\x9F\x98\x80\",")
-                                     + "\"esc\\n\\\"\":\"a\\tb\\u00e4\\ud83d\\ude00\",\"del\":\"a" + del + "b\",\"\":\"\"," // NOLINT(modernize-raw-string-literal)
-                                     + "\"list\":[\"x\",\"\xC3\xBC\",\"\\\\\",\"long ASCII string with more than thirty-two bytes in it\","
-                                     + "\"long string with more than thirty-two bytes and one non-ASCII byte \xC3\xB6\"],"
-                                     + "\"n\":[1,2.5,-0,true,null]}";
-            const ordered_json j = ordered_json::parse(text);
-            const ordered_json_document d = ordered_json_document::parse(text);
-            CHECK(d.root().dump(-1, ' ', true) == j.dump(-1, ' ', true));
-            CHECK(d.root().dump(4, ' ', true) == j.dump(4, ' ', true));
-            CHECK(d.root().dump() == j.dump());
-            CHECK(d.root()["list"].dump(-1, ' ', true) == j["list"].dump(-1, ' ', true));
-            CHECK(d.root()["del"].dump(-1, ' ', true) == j["del"].dump(-1, ' ', true));
-
-            // strings that an edit wrote (the source strings that they replace were plain)
-            ordered_json_editable_document e = ordered_json_editable_document::parse(text);
-            ordered_json k = j;
-            e.set(e.root()["plain"], std::string("\xC3\xA4gypten"));
-            k["plain"] = "\xC3\xA4gypten";
-            e.set(e.root()["list"][0], std::string("ascii after edit"));
-            k["list"][0] = "ascii after edit";
-            e.set(e.root()["list"][1], std::string("\xC3\xA4"));
-            k["list"][1] = "\xC3\xA4";
-            e.set(e.root()["del"], std::string("tab\there"));
-            k["del"] = "tab\there";
-            CHECK(e.root().dump(-1, ' ', true) == k.dump(-1, ' ', true));
-            CHECK(e.root().dump(2, ' ', true) == k.dump(2, ' ', true));
-            CHECK(e.root().dump() == k.dump());
-
-            // strings that were edited from or to ASCII text
-            e.set(e.root()["list"][2], std::string("\xC3\xA4"));
-            k["list"][2] = "\xC3\xA4";
-            e.set(e.root()["list"][4], std::string("plain again"));
-            k["list"][4] = "plain again";
-            CHECK(e.root().dump(-1, ' ', true) == k.dump(-1, ' ', true));
-        }
-    }
-
-    SECTION("ensure_ascii with DEL or a non-ASCII character at every position")
-    {
-        // the check whether a string of the source has no byte >= 0x7F reads
-        // eight bytes at a time and then the rest one by one
-        for (std::size_t length = 1; length <= 20; ++length)
-        {
-            for (std::size_t pos = 0; pos < length; ++pos)
-            {
-                for (const char* const c :
-                        {
-                            "\x7F", "\x7E", "\xC3\xA4", "\xC3\xBF" // NOLINT(modernize-raw-string-literal)
-                        })
-                {
-                    const std::string s = std::string(pos, 'a') + c + std::string(length - pos - 1, 'b');
-                    std::string text = "[\"";
-                    text.append(s).append("\",{\"").append(s).append("\":1}]");
-                    const json_document d = json_document::parse(text);
-                    const json j = json::parse(text);
-                    CHECK(d.root().dump(-1, ' ', true) == j.dump(-1, ' ', true));
-                    CHECK(d.root().dump(1, ' ', true) == j.dump(1, ' ', true));
-                }
-            }
-        }
-    }
-
-    SECTION("streams and discarded views")
-    {
-        const json_document d = json_document::parse(R"({"a": [1, 2]})");
-        std::ostringstream compact;
-        compact << d.root();
-        CHECK(compact.str() == R"({"a":[1,2]})");
-        std::ostringstream pretty;
-        pretty << std::setw(2) << std::setfill('.') << d.root() << d.root()["a"];
-        CHECK(pretty.str() == "{\n..\"a\": [\n....1,\n....2\n..]\n}[1,2]");
-        CHECK(json_view().dump() == json(json::value_t::discarded).dump());
-    }
-}

tests/src/unit-json_view_edit.cpp

@@ -532,18 +532,18 @@ TEST_CASE("json_view edits: errors")
 
     CHECK_THROWS_WITH_AS(d.set(other_editable.root(), 1), "[json.exception.invalid_iterator.202] view does not belong to this document", json::invalid_iterator&);
     CHECK_THROWS_WITH_AS(d.set(json_editable_view(), 1), "[json.exception.invalid_iterator.202] view does not belong to this document", json::invalid_iterator&);
-    CHECK_THROWS_WITH_AS(d.set(root["a"], "k", 1), "[json.exception.type_error.305] cannot use operator[] with a string argument with array", json::type_error&);
-    CHECK_THROWS_WITH_AS(d.set(root["o"], 0, 1), "[json.exception.type_error.305] cannot use operator[] with a numeric argument with object", json::type_error&);
-    CHECK_THROWS_WITH_AS(d.set(root["a"], 2, 1), "[json.exception.out_of_range.401] array index 2 is out of range", json::out_of_range&);
-    CHECK_THROWS_WITH_AS(d.set(root["a"], -1, 1), "[json.exception.out_of_range.401] array index -1 is out of range", json::out_of_range&);
-    CHECK_THROWS_WITH_AS(d.push_back(root["o"], 1), "[json.exception.type_error.308] cannot use push_back() with object", json::type_error&);
-    CHECK_THROWS_WITH_AS(d.insert(root["n"], 0, 1), "[json.exception.type_error.309] cannot use insert() with number", json::type_error&);
-    CHECK_THROWS_WITH_AS(d.insert(root["a"], 3, 1), "[json.exception.out_of_range.401] array index 3 is out of range", json::out_of_range&);
-    CHECK_THROWS_WITH_AS(d.erase(root["n"], "k"), "[json.exception.type_error.307] cannot use erase() with number", json::type_error&);
-    CHECK_THROWS_WITH_AS(d.erase(root["o"], 0), "[json.exception.type_error.307] cannot use erase() with object", json::type_error&);
-    CHECK_THROWS_WITH_AS(d.erase(root["a"], 2), "[json.exception.out_of_range.401] array index 2 is out of range", json::out_of_range&);
+    CHECK_THROWS_WITH_AS(d.set(root["a"], "k", 1), json_view_test::message("type_error.305", "/a", 21, 27, "cannot use operator[] with a string argument with array").c_str(), json::type_error&);
+    CHECK_THROWS_WITH_AS(d.set(root["o"], 0, 1), json_view_test::message("type_error.305", "/o", 6, 14, "cannot use operator[] with a numeric argument with object").c_str(), json::type_error&);
+    CHECK_THROWS_WITH_AS(d.set(root["a"], 2, 1), json_view_test::message("out_of_range.401", "/a", 21, 27, "array index 2 is out of range").c_str(), json::out_of_range&);
+    CHECK_THROWS_WITH_AS(d.set(root["a"], -1, 1), json_view_test::message("out_of_range.401", "/a", 21, 27, "array index -1 is out of range").c_str(), json::out_of_range&);
+    CHECK_THROWS_WITH_AS(d.push_back(root["o"], 1), json_view_test::message("type_error.308", "/o", 6, 14, "cannot use push_back() with object").c_str(), json::type_error&);
+    CHECK_THROWS_WITH_AS(d.insert(root["n"], 0, 1), json_view_test::message("type_error.309", "/n", 34, 35, "cannot use insert() with number").c_str(), json::type_error&);
+    CHECK_THROWS_WITH_AS(d.insert(root["a"], 3, 1), json_view_test::message("out_of_range.401", "/a", 21, 27, "array index 3 is out of range").c_str(), json::out_of_range&);
+    CHECK_THROWS_WITH_AS(d.erase(root["n"], "k"), json_view_test::message("type_error.307", "/n", 34, 35, "cannot use erase() with number").c_str(), json::type_error&);
+    CHECK_THROWS_WITH_AS(d.erase(root["o"], 0), json_view_test::message("type_error.307", "/o", 6, 14, "cannot use erase() with object").c_str(), json::type_error&);
+    CHECK_THROWS_WITH_AS(d.erase(root["a"], 2), json_view_test::message("out_of_range.401", "/a", 21, 27, "array index 2 is out of range").c_str(), json::out_of_range&);
     CHECK_THROWS_WITH_AS(d.erase(json::json_pointer("")), "[json.exception.out_of_range.405] JSON pointer has no parent", json::out_of_range&);
-    CHECK_THROWS_WITH_AS(d.erase(json::json_pointer("/missing/x")), "[json.exception.out_of_range.403] key 'missing' not found", json::out_of_range&);
+    CHECK_THROWS_WITH_AS(d.erase(json::json_pointer("/missing/x")), json_view_test::message("out_of_range.403", "", 0, 47, "key 'missing' not found").c_str(), json::out_of_range&);
     CHECK_THROWS_WITH_AS(d.set(json::json_pointer("/a/01"), 1), "[json.exception.parse_error.106] parse error: array index '01' must not begin with '0'", json::parse_error&);
     CHECK_THROWS_WITH_AS(d.set(root, json_editable_view()), "[json.exception.type_error.302] type must be a value, but is discarded", json::type_error&);
     CHECK_THROWS_WITH_AS(d.set(root, json::binary({1, 2})), "[json.exception.type_error.319] cannot store a binary value in a json_document", json::type_error&);

tests/src/unit-json_view_from_json.cpp

@@ -337,8 +337,7 @@ TEST_CASE("json_view from_json")
         CHECK(d.root().get<std::unordered_map<std::string, int>>().at("a") == 3);
     }
 
-    // (the exceptions of views have neither the path nor the byte range of the value)
-#if !defined(JSON_NOEXCEPTION) && !JSON_DIAGNOSTICS && !JSON_DIAGNOSTIC_POSITIONS
+#if !defined(JSON_NOEXCEPTION)
     SECTION("std types: the same errors as from the materialized value")
     {
         const json_document d = json_document::parse(R"([[1, "x"], {"a": 1}, "text", [1], 5, null])");
@@ -403,10 +402,10 @@ TEST_CASE("json_view from_json")
         CHECK(d.root()[4].get<strict_color>() == strict_color::red);
 #if !defined(JSON_NOEXCEPTION)
         const json j = d.root().materialize();
-        CHECK(exception_of_call([&] { static_cast<void>(d.root()[2].get<strict_color>()); }) == "[json.exception.out_of_range.410] enum value out of range for strict_color: \"purple\"");
 #if !JSON_DIAGNOSTICS && !JSON_DIAGNOSTIC_POSITIONS
-        CHECK(exception_of_call([&] { static_cast<void>(d.root()[2].get<strict_color>()); }) == exception_of_call([&] { static_cast<void>(j[2].get<strict_color>()); }));
+        CHECK(exception_of_call([&] { static_cast<void>(d.root()[2].get<strict_color>()); }) == "[json.exception.out_of_range.410] enum value out of range for strict_color: \"purple\"");
 #endif
+        CHECK(exception_of_call([&] { static_cast<void>(d.root()[2].get<strict_color>()); }) == exception_of_call([&] { static_cast<void>(j[2].get<strict_color>()); }));
 #endif
     }
 
@@ -452,11 +451,9 @@ TEST_CASE("json_view from_json")
         CHECK(o.root().get<shape>().center.y == 2);
     }
 
-#if !defined(JSON_NOEXCEPTION) && !JSON_DIAGNOSTICS && !JSON_DIAGNOSTIC_POSITIONS
+#if !defined(JSON_NOEXCEPTION)
     SECTION("errors are those of the same conversion of a json value")
     {
-        // (with JSON_DIAGNOSTICS, those of json name the path of the value,
-        // and with JSON_DIAGNOSTIC_POSITIONS its bytes)
         const json_document d = json_document::parse(R"([{"x": 1}, {"x": "1", "y": 2}, [1, 2], {"id": "7", "c": "red"}])");
         const json j = d.root().materialize();
         for (std::size_t i = 0; i < 3; ++i)

tests/src/unit-json_view_large.cpp

@@ -1,544 +0,0 @@
-//     __ _____ _____ _____
-//  __|  |   __|     |   | |  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;
-using nlohmann::ordered_json_view;
-
-#include "json_view_test_helpers.hpp"
-using json_view_test::generator;
-
-#include <algorithm>
-#include <array>
-#include <cmath>
-#include <cstddef>
-#include <cstdint>
-#include <cstring>
-#include <iomanip>
-#include <map>
-#include <random>
-#include <set>
-#include <sstream>
-#include <string>
-#include <vector>
-
-// These tests were split off unit-json_view_dump.cpp, whose object file got too large for the MinGW
-// linker (see json_view_test_helpers.hpp). 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("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
-    for (const std::size_t members :
-            {
-                127u, 128u, 129u, 10000u
-            })
-    {
-        CAPTURE(members)
-        std::string text = "{";
-        for (std::size_t i = 0; i < members; ++i)
-        {
-            text += (i != 0 ? ",\"" : "\"") + std::string(i % 23, 'k') + std::to_string(i) + (i % 7 == 0 ? "\\n" : "") + "\":" + std::to_string(i);
-        }
-        text += R"(,"":"empty key","k1":"a duplicate of an earlier key"})";
-        const json_document d = json_document::parse(text);
-        const json_view v = d.root();
-        const json j = json::parse(text);
-        for (std::size_t i = 0; i < members; ++i)
-        {
-            const std::string key = std::string(i % 23, 'k') + std::to_string(i) + (i % 7 == 0 ? "\n" : "");
-            CHECK(v[key].get<std::size_t>() == i);
-            CHECK(v.contains(key));
-            CHECK(v.find(key).key() == key);
-            CHECK(v.at(key).get<std::size_t>() == i);
-            CHECK(!v.contains(key + "x"));
-        }
-        CHECK(v[""].get_string() == "empty key");
-        CHECK(v["k1"].get<int>() == 1); // the first of duplicate keys, as for small objects
-        CHECK(!v.contains("missing"));
-        CHECK_THROWS_WITH_AS(v.at("missing"), "[json.exception.out_of_range.403] key 'missing' not found", json::out_of_range&);
-        CHECK(v == j);
-        CHECK(v.materialize() == j);
-    }
-
-    SECTION("duplicate keys: the first member wins, with and without a table")
-    {
-        // an object of `total` members: the keys "k0".."k<n-1>" in order, then
-        // three keys repeated twice more (one copy in the middle, one at the
-        // end), and two keys repeated once; the value of a member is its
-        // position, so that the first member of a key can be told apart
-        struct member
-        {
-            std::string key;
-            std::size_t position;
-        };
-        const auto make_members = [](std::size_t total)
-        {
-            std::vector<std::string> keys;
-            for (std::size_t i = 0; i + 8 < total; ++i)
-            {
-                keys.push_back("k" + std::to_string(i));
-            }
-            const std::size_t n = keys.size();
-            const std::array<std::size_t, 3> triple = {{3, 17, n - 1}};
-            const std::array<std::size_t, 2> twice = {{5, n / 2}};
-            std::vector<std::string> ordered = keys;
-            for (const std::size_t i : triple)
-            {
-                ordered.insert(ordered.begin() + static_cast<std::ptrdiff_t>(ordered.size() / 2), keys[i]);
-            }
-            for (const std::size_t i : twice)
-            {
-                ordered.insert(ordered.begin() + static_cast<std::ptrdiff_t>(ordered.size() / 3), keys[i]);
-            }
-            for (const std::size_t i : triple)
-            {
-                ordered.push_back(keys[i]);
-            }
-            std::vector<member> result;
-            result.reserve(ordered.size());
-            for (std::size_t i = 0; i < ordered.size(); ++i)
-            {
-                result.push_back({ordered[i], i});
-            }
-            return result;
-        };
-
-        // 100 and 127 members: no table; 128 members and more: a table
-        for (const std::size_t total :
-                {
-                    100u, 127u, 128u, 200u, 5000u
-                })
-        {
-            CAPTURE(total)
-            const std::vector<member> members = make_members(total);
-            REQUIRE(members.size() >= total);
-            REQUIRE(members.size() >= 100);
-            std::string text = "{";
-            std::map<std::string, std::size_t> first;
-            std::map<std::string, std::size_t> last;
-            std::map<std::string, std::size_t> occurrences;
-            for (const member& m : members)
-            {
-                text += (text.size() > 1 ? ",\"" : "\"") + m.key + "\":" + std::to_string(m.position);
-                first.insert({m.key, m.position});
-                last[m.key] = m.position;
-                ++occurrences[m.key];
-            }
-            text += '}';
-            REQUIRE(first.size() < members.size());
-
-            const json_document d = json_document::parse(text);
-            const json_view v = d.root();
-            const json j = json::parse(text);
-            CHECK(v.size() == members.size()); // every occurrence is visited
-            const json m = v.materialize();
-            for (const auto& entry : first)
-            {
-                CAPTURE(entry.first)
-                const std::size_t expected = entry.second;
-                CHECK(v[entry.first].get<std::size_t>() == expected);
-                CHECK(v.at(entry.first).get<std::size_t>() == expected);
-                CHECK(v.find(entry.first).value().get<std::size_t>() == expected);
-                CHECK(v.value(entry.first, std::size_t{0}) == expected);
-                CHECK(v.contains(entry.first));
-                CHECK(v.count(entry.first) == occurrences[entry.first]);
-                const std::string pointer = "/" + entry.first;
-                CHECK(v[json::json_pointer(pointer)].get<std::size_t>() == expected);
-                CHECK(v.at(json::json_pointer(pointer)).get<std::size_t>() == expected);
-                CHECK(v.value(json::json_pointer(pointer), std::size_t{0}) == expected);
-                CHECK(v.contains(json::json_pointer(pointer)));
-                // materialize() and parse() keep the last value instead
-                CHECK(j[entry.first].get<std::size_t>() == last.at(entry.first));
-                CHECK(m[entry.first].get<std::size_t>() == last.at(entry.first));
-            }
-            CHECK(!v.contains("k"));
-            CHECK(v["missing"].is_discarded());
-            CHECK(v.materialize() == j);
-            CHECK(v == j);
-        }
-    }
-
-    SECTION("colliding keys")
-    {
-        // the hash is not seeded, so keys that all land in one place must not
-        // make the table build quadratic: such an object gets no table and is
-        // searched linearly
-        constexpr std::size_t members = 300;
-        constexpr std::size_t table_size = 1024; // the table size for 300 members: the next power of two >= 600
-        std::vector<std::string> colliding;
-        std::vector<std::string> spread;
-        for (std::uint64_t counter = 0; colliding.size() < members || spread.size() < members; ++counter)
-        {
-            std::string key(8, 'a');
-            std::uint64_t x = counter;
-            for (std::size_t i = 0; i < 8; ++i, x /= 26)
-            {
-                key[i] = static_cast<char>('a' + (x % 26));
-            }
-            const bool lands_in_slot_zero = (nlohmann::detail::view::key_hash(key.data(), key.size()) & (table_size - 1)) == 0;
-            if (lands_in_slot_zero && colliding.size() < members)
-            {
-                colliding.push_back(key);
-            }
-            else if (!lands_in_slot_zero && spread.size() < members)
-            {
-                spread.push_back(key);
-            }
-        }
-
-        const auto make_text = [](const std::vector<std::string>& keys)
-        {
-            std::string text = "{";
-            for (std::size_t i = 0; i < keys.size(); ++i)
-            {
-                text += (i != 0 ? ",\"" : "\"") + keys[i] + "\":" + std::to_string(i % 10);
-            }
-            return text + "}";
-        };
-        const std::string colliding_text = make_text(colliding);
-        const std::string spread_text = make_text(spread);
-
-        json_document with_collisions = json_document::parse(colliding_text);
-        json_document without_collisions = json_document::parse(spread_text);
-        for (const auto* pair :
-                {
-                    &colliding, &spread
-                })
-        {
-            const json_view v = (pair == &colliding ? with_collisions : without_collisions).root();
-            for (std::size_t i = 0; i < members; ++i)
-            {
-                CAPTURE(i)
-                CHECK(v[(*pair)[i]].get<std::size_t>() == i % 10);
-                CHECK(v.at((*pair)[i]).get<std::size_t>() == i % 10);
-                CHECK(v.find((*pair)[i]).key() == (*pair)[i]);
-                CHECK(!v.contains((*pair)[i] + "x"));
-            }
-            CHECK(!v.contains("missing"));
-        }
-        CHECK(with_collisions.root() == json::parse(colliding_text));
-
-        // only the object with the spread keys got a table (both texts have the
-        // same length, so the tables are the only difference)
-        with_collisions.shrink_to_fit();
-        without_collisions.shrink_to_fit();
-        CHECK(without_collisions.memory_usage() >= with_collisions.memory_usage() + (table_size * sizeof(std::uint32_t)));
-    }
-
-    SECTION("keys that differ in their last bytes are spread over the table")
-    {
-        // The hash must spread the usual families of keys (numbered names, hex
-        // strings, dates, short names): a key farther than index_max_displacement
-        // from its home slot would leave the object without a table. The keys
-        // are placed with linear probing, as the table does.
-        const auto max_displacement = [](const std::vector<std::string>& keys)
-        {
-            std::size_t slots = 16;
-            while (slots < 2 * keys.size())
-            {
-                slots *= 2;
-            }
-            std::vector<bool> used(slots, false);
-            std::size_t farthest = 0;
-            for (const std::string& key : keys)
-            {
-                std::size_t i = nlohmann::detail::conditional_static_cast<std::size_t>(nlohmann::detail::view::key_hash(key.data(), key.size())) & (slots - 1);
-                std::size_t distance = 0;
-                while (used[i])
-                {
-                    i = (i + 1) & (slots - 1);
-                    ++distance;
-                }
-                used[i] = true;
-                farthest = (std::max)(farthest, distance);
-            }
-            return farthest;
-        };
-        // the keys of a printf-like pattern with one %d, %0Nd or %0Nx, formatted with a stream
-        // (a format string that is not a literal is a Flawfinder error)
-        const auto numbered = [](const char* format, int count, int step)
-        {
-            const std::string pattern(format);
-            const std::size_t percent = pattern.find('%');
-            std::size_t conversion = percent + 1;
-            std::size_t width = 0;
-            while (pattern[conversion] >= '0' && pattern[conversion] <= '9')
-            {
-                width = (width * 10) + static_cast<std::size_t>(pattern[conversion] - '0');
-                ++conversion;
-            }
-            const bool hex = pattern[conversion] == 'x';
-            std::vector<std::string> keys;
-            for (int i = 0; i < count; ++i)
-            {
-                std::ostringstream key;
-                key << pattern.substr(0, percent) << std::setfill('0') << std::setw(static_cast<int>(width))
-                    << (hex ? std::hex : std::dec) << (i * step) << pattern.substr(conversion + 1);
-                keys.push_back(key.str());
-            }
-            return keys;
-        };
-        for (const char* format :
-                {"key%d", "%d", "item_%04d", "a%05d", "%03dxyz", "id%d", "user_%d", "%d_suffix", "0x%04x", "A%07d", "12345678%d", "1234567%d", "123456789%d",
-                 "configuration_parameter_value_%04d_extra", "2026-01-%02dT00:00:00Z"
-                })
-        {
-            CAPTURE(format)
-            CHECK(max_displacement(numbered(format, std::string(format).find("%02d") != std::string::npos ? 60 : 1000, 1)) < nlohmann::detail::view::index_max_displacement); // NOLINT(abseil-string-find-str-contains)
-            CHECK(max_displacement(numbered(format, 600, 7)) < nlohmann::detail::view::index_max_displacement);
-        }
-        std::mt19937_64 rng(7); // NOLINT(cert-msc32-c,cert-msc51-cpp,bugprone-random-generator-seed)
-        for (const std::size_t length :
-                {
-                    std::size_t{1}, std::size_t{2}, std::size_t{3}, std::size_t{5}, std::size_t{7}, std::size_t{8}, std::size_t{9}, std::size_t{12}, std::size_t{16}, std::size_t{17}, std::size_t{33}
-                })
-        {
-            CAPTURE(length)
-            std::set<std::string> distinct;
-            // (there are 26 and 676 keys of one and two letters)
-            std::size_t wanted = 700;
-            if (length == 1)
-            {
-                wanted = 26;
-            }
-            else if (length == 2)
-            {
-                wanted = 500;
-            }
-            while (distinct.size() < wanted)
-            {
-                std::string key(length, 'a');
-                for (char& c : key)
-                {
-                    c = static_cast<char>('a' + (rng() % 26));
-                }
-                distinct.insert(key);
-            }
-            CHECK(max_displacement(std::vector<std::string>(distinct.begin(), distinct.end())) < nlohmann::detail::view::index_max_displacement);
-        }
-    }
-
-    SECTION("keys of every length that differ in one byte")
-    {
-        // In a large object, a key is found by comparing its bytes with those of the member the
-        // table points to: up to 16 bytes as two overlapping words, longer ones with memcmp.
-        // Keys of every length from 0 to 40, and keys that differ from them in one byte.
-        std::mt19937_64 rng(2026); // NOLINT(cert-msc32-c,cert-msc51-cpp,bugprone-random-generator-seed)
-        std::vector<std::string> keys;
-        std::set<std::string> distinct;
-        for (std::size_t length = 0; length <= 40; ++length)
-        {
-            for (int copy = 0; copy < 5; ++copy)
-            {
-                std::string key(length, 'a');
-                for (char& c : key)
-                {
-                    c = static_cast<char>('a' + (rng() % 26));
-                }
-                if (distinct.insert(key).second)
-                {
-                    keys.push_back(key);
-                }
-            }
-        }
-        std::string text = "{";
-        for (std::size_t i = 0; i < keys.size(); ++i)
-        {
-            text += (i != 0 ? ",\"" : "\"") + keys[i] + "\":" + std::to_string(i);
-        }
-        text += "}";
-        REQUIRE(keys.size() >= 128); // (a table is built for 128 members)
-        const json_document d = json_document::parse(text);
-        const json_view v = d.root();
-        for (std::size_t i = 0; i < keys.size(); ++i)
-        {
-            CAPTURE(keys[i])
-            CHECK(v[keys[i]].get<std::size_t>() == i);
-            CHECK(v.at(keys[i]).get<std::size_t>() == i);
-            CHECK(v.contains(keys[i]));
-            for (std::size_t at = 0; at < keys[i].size(); ++at)
-            {
-                std::string other = keys[i];
-                other[at] = other[at] == 'z' ? 'y' : 'z';
-                CAPTURE(other)
-                CHECK(v.contains(other) == (distinct.count(other) != 0));
-            }
-            CHECK(v.contains(keys[i] + "a") == (distinct.count(keys[i] + "a") != 0));
-            if (!keys[i].empty())
-            {
-                CHECK(v.contains(keys[i].substr(1)) == (distinct.count(keys[i].substr(1)) != 0));
-            }
-        }
-    }
-
-    SECTION("shrink_to_fit releases the tables' spare capacity")
-    {
-        const auto make_text = [](int objects, int members)
-        {
-            std::string text = "[";
-            for (int object = 0; object < objects; ++object)
-            {
-                text += object != 0 ? ",{" : "{";
-                for (int i = 0, n = members + object; i < n; ++i)
-                {
-                    text += (i != 0 ? ",\"" : "\"") + std::to_string(i) + "\":" + std::to_string(i);
-                }
-                text += '}';
-            }
-            return text + "]";
-        };
-        const std::string small_text = make_text(5, 150);
-        const std::string big_text = make_text(40, 400);
-
-        // reading a big text, and then a small one, leaves the spare capacity
-        // of the big one: shrink_to_fit() brings the document to the size of
-        // one parsed from the small text alone
-        json_document d = json_document::parse(big_text);
-        const std::size_t big = d.memory_usage();
-        d.read(small_text);
-        CHECK(d.memory_usage() >= big);
-        d.shrink_to_fit();
-        json_document fresh = json_document::parse(small_text);
-        fresh.shrink_to_fit();
-        CHECK(d.memory_usage() == fresh.memory_usage());
-        CHECK(d.memory_usage() < big / 2);
-        CHECK(d.root() == json::parse(small_text));
-        for (int object = 0; object < 5; ++object)
-        {
-            const json_view v = d.root()[static_cast<std::size_t>(object)];
-            for (int i = 0, n = 150 + object; i < n; ++i)
-            {
-                CHECK(v[std::to_string(i)].get<int>() == i);
-            }
-        }
-    }
-
-    SECTION("nested, reused, and in arrays")
-    {
-        std::string inner = "{";
-        for (int i = 0; i < 300; ++i)
-        {
-            inner += (i != 0 ? ",\"m" : "\"m") + std::to_string(i) + "\":" + std::to_string(i);
-        }
-        inner += '}';
-        const std::string text = "[" + inner + ",{\"x\":" + inner + "}," + inner + "]";
-        json_document d = json_document::parse(text);
-        CHECK(d.root()[0]["m299"].get<int>() == 299);
-        CHECK(d.root()[1]["x"]["m150"].get<int>() == 150);
-        CHECK(d.root()[2]["m0"].get<int>() == 0);
-        const std::size_t with_index = d.memory_usage();
-        d.read(std::string("{\"small\": 1}"));
-        CHECK(d.root()["small"].get<int>() == 1);
-        d.read(text);
-        CHECK(d.root()[2]["m7"].get<int>() == 7);
-        CHECK(d.memory_usage() >= with_index / 2);
-    }
-}

tests/src/unit-json_view_positions.cpp

@@ -81,6 +81,112 @@ std::string range(const json& j)
 {
     return j.start_pos() == npos && j.end_pos() == npos ? "none" : std::to_string(j.start_pos()) + "-" + std::to_string(j.end_pos());
 }
+
+#if !defined(JSON_NOEXCEPTION)
+using json_view_test::exception_of_call;
+
+// the exceptions about every value of v are those about the value at the same
+// place in j, a value of the whole document from parse() or materialize() (so
+// that its parents and positions are set); members whose key repeats are
+// skipped (the view finds the first one, j holds the last)
+template<typename View>
+void check_exceptions(const View& v, const json& j)
+{
+    CAPTURE(j.dump())
+    if (!v.is_number())
+    {
+        CHECK(exception_of_call([&] { static_cast<void>(v.template get<int>()); }) == exception_of_call([&] { static_cast<void>(j.get<int>()); }));
+    }
+    if (!v.is_string())
+    {
+        CHECK(exception_of_call([&] { static_cast<void>(v.template get<std::string>()); }) == exception_of_call([&] { static_cast<void>(j.get<std::string>()); }));
+    }
+    if (!v.is_array())
+    {
+        CHECK(exception_of_call([&] { static_cast<void>(v.template get<std::vector<int>>()); }) == exception_of_call([&] { static_cast<void>(j.get<std::vector<int>>()); }));
+    }
+    CHECK(exception_of_call([&] { static_cast<void>(v.at("missing")); }) == exception_of_call([&] { static_cast<void>(j.at("missing")); }));
+    CHECK(exception_of_call([&] { static_cast<void>(v.at(1000)); }) == exception_of_call([&] { static_cast<void>(j.at(1000)); }));
+    CHECK(exception_of_call([&] { static_cast<void>(v.at(json::json_pointer("/missing"))); }) == exception_of_call([&] { static_cast<void>(j.at(json::json_pointer("/missing"))); }));
+    CHECK(exception_of_call([&] { static_cast<void>(v.at(json::json_pointer("/-"))); }) == exception_of_call([&] { static_cast<void>(j.at(json::json_pointer("/-"))); }));
+    CHECK(exception_of_call([&] { static_cast<void>(v.value("k", 0)); }) == exception_of_call([&] { static_cast<void>(j.value("k", 0)); }));
+    if (!v.is_object())
+    {
+        CHECK(exception_of_call([&] { static_cast<void>(v["k"]); }) == exception_of_call([&] { static_cast<void>(j["k"]); }));
+        // (without the context of basic_json: view iterators do not know the value they iterate over)
+        CHECK(exception_of_call([&] { static_cast<void>(v.begin().key()); }) == "[json.exception.invalid_iterator.207] cannot use key() for non-object iterators");
+    }
+    if (!v.is_array())
+    {
+        CHECK(exception_of_call([&] { static_cast<void>(v[0]); }) == exception_of_call([&] { static_cast<void>(j[0]); }));
+    }
+    if (v.is_null())
+    {
+        CHECK(exception_of_call([&] { static_cast<void>(v.front()); }) == exception_of_call([&] { static_cast<void>(j.front()); }));
+    }
+    if (v.is_object())
+    {
+        for (auto it = v.begin(); it != v.end(); ++it)
+        {
+            const std::string key(it.key().data(), it.key().size());
+            if (v.count(key) == 1)
+            {
+                check_exceptions(it.value(), j.at(key));
+            }
+        }
+    }
+    else if (v.is_array())
+    {
+        for (std::size_t i = 0; i < v.size(); ++i)
+        {
+            check_exceptions(v[i], j[i]);
+        }
+    }
+}
+
+// as check_exceptions, for edits that fail (they change neither d nor j)
+void check_edit_exceptions(json_editable_document& d, const json_editable_document::view_type& v, json& j)
+{
+    CAPTURE(j.dump())
+    if (!v.is_object() && !v.is_null())
+    {
+        CHECK(exception_of_call([&] { d.set(v, "k", 1); }) == exception_of_call([&] { j["k"] = 1; }));
+        CHECK(exception_of_call([&] { d.erase(v, "k"); }) == exception_of_call([&] { j.erase("k"); }));
+    }
+    if (!v.is_array() && !v.is_null())
+    {
+        CHECK(exception_of_call([&] { d.set(v, 0, 1); }) == exception_of_call([&] { j[0] = 1; }));
+        CHECK(exception_of_call([&] { d.push_back(v, 1); }) == exception_of_call([&] { j.push_back(1); }));
+    }
+    if (!v.is_array())
+    {
+        CHECK(exception_of_call([&] { d.insert(v, 0, 1); }) == exception_of_call([&] { j.insert(j.cbegin(), 1); }));
+        CHECK(exception_of_call([&] { d.erase(v, 0); }) == exception_of_call([&] { j.erase(0); }));
+    }
+    else
+    {
+        CHECK(exception_of_call([&] { d.erase(v, 1000); }) == exception_of_call([&] { j.erase(1000); }));
+    }
+    if (v.is_object())
+    {
+        for (auto it = v.begin(); it != v.end(); ++it)
+        {
+            const std::string key(it.key().data(), it.key().size());
+            if (v.count(key) == 1)
+            {
+                check_edit_exceptions(d, it.value(), j.at(key));
+            }
+        }
+    }
+    else if (v.is_array())
+    {
+        for (std::size_t i = 0; i < v.size(); ++i)
+        {
+            check_edit_exceptions(d, v[i], j[i]);
+        }
+    }
+}
+#endif
 } // namespace
 
 TEST_CASE("json_view materialize() source positions")
@@ -345,3 +451,96 @@ TEST_CASE("json_view materialize() source positions")
     }
 #endif
 }
+
+#if !defined(JSON_NOEXCEPTION)
+TEST_CASE("json_view exceptions name the path and the bytes of the value")
+{
+    SECTION("as those of basic_json")
+    {
+        const std::vector<std::string> texts =
+        {
+            "null", "true", "42", "\"s\"", "[]", "{}", "[1, [2, [3, []]]]", R"({"a": {"b": [1, {"c": null}]}, "d": "e"})",
+            R"({"a/b": {"~": [1]}, "\u0041\n": {"x": "\t"}, "": [{}]})",
+            R"([1 /* ] */, {"a": [] // }
+, "b": {} } , ])",
+            "\n\t[\r\n  1 ,\n  {\n    \"a\" : [ ]\n  }\n]\n",
+        };
+        for (const auto& text : texts)
+        {
+            CAPTURE(text)
+            const json j = json::parse(text, nullptr, true, true, true);
+            const json_document d = json_document::parse(text, true, true, true);
+            check_exceptions(d.root(), j);
+            json_editable_document e = json_editable_document::parse(text, true, true, true);
+            check_exceptions(e.root(), j);
+            json m = e.root().materialize();
+            check_edit_exceptions(e, e.root(), m);
+        }
+    }
+
+    SECTION("generated documents")
+    {
+        json_view_test::generator gen;
+        for (int i = 0; i < 100; ++i)
+        {
+            std::string text;
+            gen.value(text, 0);
+            CAPTURE(text)
+            const json_document d = json_document::parse(text);
+            check_exceptions(d.root(), json::parse(text));
+        }
+    }
+
+    SECTION("edited documents: as those of materialize()")
+    {
+        const std::string text = R"({"a": [1, 2], "b": {"c": "\n", "d": true}, "e": [3, [4]], "f": "x"})";
+        json_editable_document d = json_editable_document::parse(text);
+        d.set(d.root()["f"], 5);
+        d.push_back(d.root()["a"], json::parse(R"({"n": [null]})"));
+        d.erase(d.root()["e"], 0);
+        d.set(d.root()["b"], "c2", d.root()["e"]);
+        d.insert(d.root()["e"], 0, "new");
+        json m = d.root().materialize();
+        check_exceptions(d.root(), m);
+        check_edit_exceptions(d, d.root(), m);
+        // (a value inside a new container: a path, but no bytes)
+        CHECK(exception_of_call([&] { static_cast<void>(d.root()["a"][2]["n"][0].get<int>()); }) == "[json.exception.type_error.302] "
+#if JSON_DIAGNOSTICS
+              "(/a/2/n/0) "
+#endif
+              "type must be number, but is null");
+    }
+
+    SECTION("examples")
+    {
+        const std::string text = R"({"people": [{"name": "a", "age": 1}, {"name": "c", "age": "x"}], "o": {"k": 1}})";
+        const json_document d = json_document::parse(text);
+        CHECK(exception_of_call([&] { static_cast<void>(d.root()["people"][1]["age"].get<int>()); }) == "[json.exception.type_error.302] "
+#if JSON_DIAGNOSTICS
+              "(/people/1/age) "
+#endif
+              "(bytes 58-61) type must be number, but is string");
+        CHECK(exception_of_call([&] { static_cast<void>(d.root().at(json::json_pointer("/people/0/zz"))); }) == "[json.exception.out_of_range.403] "
+#if JSON_DIAGNOSTICS
+              "(/people/0) "
+#endif
+              "(bytes 12-35) key 'zz' not found");
+        // the root has no path, and with JSON_DIAGNOSTICS then no bytes either (as basic_json)
+        CHECK(exception_of_call([&] { static_cast<void>(d.root().at("missing")); }) == "[json.exception.out_of_range.403] "
+#if !JSON_DIAGNOSTICS
+              "(bytes 0-79) "
+#endif
+              "key 'missing' not found");
+    }
+
+    SECTION("duplicate keys: the value the view finds")
+    {
+        const json_document d = json_document::parse(R"({"x": "1", "x": "22"})");
+        CHECK(exception_of_call([&] { static_cast<void>(d.root()["x"].get<int>()); }) == "[json.exception.type_error.302] "
+#if JSON_DIAGNOSTICS
+              "(/x) "
+#endif
+              "(bytes 6-9) type must be number, but is string");
+    }
+}
+#endif