Merge pull request #2312 from nlohmann/docs

Update documentation

doc/mkdocs/docs/features/binary_formats/cbor.md

@@ -64,7 +64,6 @@ binary          | *size*: 4294967296..18446744073709551615   | byte string (8 by
 
 	If NaN or Infinity are stored inside a JSON number, they are serialized properly. This behavior differs from the normal JSON serialization which serializes NaN or Infinity to `null`.
 
-
 !!! info "Unused CBOR types"
 
 	The following CBOR types are not used in the conversion:
@@ -77,13 +76,16 @@ binary          | *size*: 4294967296..18446744073709551615   | byte string (8 by
       - bignum (0xC2..0xC3)
       - decimal fraction (0xC4)
       - bigfloat (0xC5)
-      - tagged items (0xC6..0xD4, 0xD8..0xDB)
       - expected conversions (0xD5..0xD7)
       - simple values (0xE0..0xF3, 0xF8)
       - undefined (0xF7)
       - half-precision floats (0xF9)
       - break (0xFF)
 
+!!! info "Tagged items"
+
+    Binary subtypes will be serialized as tagged items. See [binary values](../binary_values.md#cbor) for an example.
+
 ??? example
 
     ```cpp
@@ -150,7 +152,6 @@ Double-Precision Float | number_float    | 0xFB
      - bignum (0xC2..0xC3)
      - decimal fraction (0xC4)
      - bigfloat (0xC5)
-     - tagged items (0xC6..0xD4, 0xD8..0xDB)
      - expected conversions (0xD5..0xD7)
      - simple values (0xE0..0xF3, 0xF8)
      - undefined (0xF7)
@@ -159,6 +160,10 @@ Double-Precision Float | number_float    | 0xFB
 
 	CBOR allows map keys of any type, whereas JSON only allows strings as keys in object values. Therefore, CBOR maps with keys other than UTF-8 strings are rejected.
 
+!!! warning "Tagged items"
+
+    Tagged items will throw a parse error by default. However, they can be ignored by passing `cbor_tag_handler_t::ignore` to function `from_cbor`.
+
 ??? example
 
     ```cpp

doc/mkdocs/docs/features/binary_values.md

@@ -158,14 +158,14 @@ JSON does not have a binary type, and this library does not introduce a new type
 
 ### CBOR
 
-[CBOR](binary_formats/cbor.md) supports binary values, but no subtypes. Any binary value will be serialized as byte strings. The library will choose the smallest representation using the length of the byte array.
+[CBOR](binary_formats/cbor.md) supports binary values, but no subtypes. Subtypes will be serialized as tags. Any binary value will be serialized as byte strings. The library will choose the smallest representation using the length of the byte array.
 
 ??? example
 
     Code:
     
     ```cpp
-    // create a binary value of subtype 42 (will be ignored by CBOR)
+    // create a binary value of subtype 42
     json j;
     j["binary"] = json::binary({0xCA, 0xFE, 0xBA, 0xBE}, 42);
 
@@ -173,17 +173,18 @@ JSON does not have a binary type, and this library does not introduce a new type
     auto v = json::to_cbor(j);      
     ```
             
-    `v` is a `std::vector<std::uint8t>` with the following 13 elements:
+    `v` is a `std::vector<std::uint8t>` with the following 15 elements:
     
     ```c
     0xA1                                   // map(1)
         0x66                               // text(6)
             0x62 0x69 0x6E 0x61 0x72 0x79  // "binary"
+        0xD8 0x2A                          // tag(42)
         0x44                               // bytes(4)
             0xCA 0xFE 0xBA 0xBE            // content
     ```
 
-    Note the subtype (42) is **not** serialized, and deserializing `v` would yield the following value:
+    Note that the subtype is serialized as tag. However, parsing tagged values yield a parse error unless `json::cbor_tag_handler_t::ignore` is passed to `json::from_cbor`.
 
     ```json
     {

doc/mkdocs/docs/features/comments.md

@@ -0,0 +1,83 @@
+# Comments
+
+This library does not support comments *by default*. It does so for three reasons:
+
+1. Comments are not part of the [JSON specification](https://tools.ietf.org/html/rfc8259). You may argue that `//` or `/* */` are allowed in JavaScript, but JSON is not JavaScript.
+2. This was not an oversight: Douglas Crockford [wrote on this](https://plus.google.com/118095276221607585885/posts/RK8qyGVaGSr) in May 2012:
+
+	> 	I removed comments from JSON because I saw people were using them to hold parsing directives, a practice which would have destroyed interoperability.  I know that the lack of comments makes some people sad, but it shouldn't. 
+
+	> 	Suppose you are using JSON to keep configuration files, which you would like to annotate. Go ahead and insert all the comments you like. Then pipe it through JSMin before handing it to your JSON parser.
+
+3. It is dangerous for interoperability if some libraries would add comment support while others don't. Please check [The Harmful Consequences of the Robustness Principle](https://tools.ietf.org/html/draft-iab-protocol-maintenance-01) on this.
+
+However, you can pass set parameter `ignore_comments` to `#!c true` in the parse function to ignore `//` or `/* */` comments. Comments will then be treated as whitespace.
+
+!!! example
+
+    Consider the following JSON with comments.
+
+    ```json
+    {
+        // update in 2006: removed Pluto
+        "planets": ["Mercury", "Venus", "Earth", "Mars",
+                    "Jupiter", "Uranus", "Neptune" /*, "Pluto" */]
+    }
+    ```
+    
+    When calling `parse` without additional argument, a parse error exception is thrown. If `skip_comments` is set to `#! true`, the comments are skipped during parsing:
+
+    ```cpp
+    #include <iostream>
+    #include "json.hpp"
+    
+    using json = nlohmann::json;
+    
+    int main()
+    {
+        std::string s = R"(
+        {
+            // update in 2006: removed Pluto
+            "planets": ["Mercury", "Venus", "Earth", "Mars",
+                        "Jupiter", "Uranus", "Neptune" /*, "Pluto" */]
+        }
+        )";
+        
+        try
+        {
+            json j = json::parse(s);
+        }
+        catch (json::exception &e)
+        {
+            std::cout << e.what() << std::endl;
+        }
+        
+        json j = json::parse(s,
+                             /* callback */ nullptr,
+                             /* allow exceptions */ true,
+                             /* skip_comments */ true);
+        std::cout << j.dump(2) << '\n';
+    }
+    ```
+
+    Output:
+    
+    ```
+    [json.exception.parse_error.101] parse error at line 3, column 9:
+    syntax error while parsing object key - invalid literal;
+    last read: '<U+000A>    {<U+000A>        /'; expected string literal
+    ```
+    
+    ```json
+    {
+      "planets": [
+        "Mercury",
+        "Venus",
+        "Earth",
+        "Mars",
+        "Jupiter",
+        "Uranus",
+        "Neptune"
+      ]
+    }
+    ```

doc/mkdocs/docs/features/element_access/checked_access.md

@@ -0,0 +1,77 @@
+# Checked access: at
+
+## Overview
+
+The `#!cpp at()` member function performs checked access; that is, it returns a reference to the desired value if it exists and throws a [`basic_json::out_of_range` exception](../../home/exceptions.md#out-of-range) otherwise.
+
+??? example
+
+    Consider the following JSON value:
+    
+    ```json
+    {
+        "name": "Mary Smith",
+        "age": 42,
+        "hobbies": ["hiking", "reading"]
+    }
+    ```
+    
+    Assume the value is parsed to a `json` variable `j`.
+
+    | expression | value |
+    | ---------- | ----- |
+    | `#!cpp j`  | `#!json {"name": "Mary Smith", "age": 42, "hobbies": ["hiking", "reading"]}` |
+    | `#!cpp j.at("name")`  | `#!json "Mary Smith"` |
+    | `#!cpp j.at("age")`  | `#!json 42` |
+    | `#!cpp j.at("hobbies")`  | `#!json ["hiking", "reading"]` |
+    | `#!cpp j.at("hobbies").at(0)`  | `#!json "hiking"` |
+    | `#!cpp j.at("hobbies").at(1)`  | `#!json "reading"` |
+
+The return value is a reference, so it can be modify the original value.
+
+??? example
+
+    ```cpp
+    j.at("name") = "John Smith";
+    ```
+    
+    This code produces the following JSON value:
+    
+    ```json
+    {
+        "name": "John Smith",
+        "age": 42,
+        "hobbies": ["hiking", "reading"]
+    }
+    ```
+
+When accessing an invalid index (i.e., and index greater than or equal to the array size) or the passed object key is non-existing, an exception is thrown.
+
+??? example
+
+    ```cpp
+    j.at("hobbies").at(3) = "cooking";
+    ```
+    
+    This code produces the following exception:
+    
+    ```
+    [json.exception.out_of_range.401] array index 3 is out of range
+    ```
+
+## Notes
+
+
+!!! failure "Exceptions"
+
+    - `at` can only be used with objects (with a string argument) or with arrays (with a numeric argument). For other types, a [`basic_json::type_error`](../../home/exceptions.md#jsonexceptiontype_error304) is thrown.
+    - [`basic_json::out_of_range` exception](../../home/exceptions.md#out-of-range) exceptions are thrown if the provided key is not found in an object or the provided index is invalid.
+
+## Summary
+
+| scenario | non-const value | const value |
+| -------- | ------------- | ----------- |
+| access to existing object key | reference to existing value is returned | const reference to existing value is returned |
+| access to valid array index | reference to existing value is returned | const reference to existing value is returned |
+| access to non-existing object key | `basic_json::out_of_range` exception is thrown | `basic_json::out_of_range` exception is thrown |
+| access to invalid array index | `basic_json::out_of_range` exception is thrown | `basic_json::out_of_range` exception is thrown |

doc/mkdocs/docs/features/element_access/default_value.md

@@ -0,0 +1,32 @@
+# Access with default value: value
+
+## Overview
+
+In many situations such as configuration files, missing values are not exceptional, but may be treated as if a default value was present.
+
+??? example
+
+    Consider the following JSON value:
+    
+    ```json
+    {
+        "logOutput": "result.log",
+        "append": true
+    }
+    ```
+    
+    Assume the value is parsed to a `json` variable `j`.
+
+    | expression | value |
+    | ---------- | ----- |
+    | `#!cpp j`  | `#!json {"logOutput": "result.log", "append": true}` |
+    | `#!cpp j.value("logOutput", "logfile.log")`  | `#!json "result.log"` |
+    | `#!cpp j.value("append", true)`  | `#!json true` |
+    | `#!cpp j.value("append", false)`  | `#!json true` |
+    | `#!cpp j.value("logLevel", "verbose")`  | `#!json "verbose"` |
+
+## Note
+
+!!! failure "Exceptions"
+
+    - `value` can only be used with objects. For other types, a [`basic_json::type_error`](../../home/exceptions.md#jsonexceptiontype_error306) is thrown.

doc/mkdocs/docs/features/element_access/index.md

@@ -0,0 +1,9 @@
+# Overview
+
+There are many ways elements in a JSON value can be accessed:
+
+- unchecked access via [`operator[]`](unchecked_access.md)
+- checked access via [`at`](checked_access.md)
+- access with default value via [`value`](default_value.md)
+- iterators
+- JSON pointers

doc/mkdocs/docs/features/element_access/unchecked_access.md

@@ -0,0 +1,102 @@
+# Unchecked access: operator[]
+
+## Overview
+
+Elements in a JSON object and a JSON array can be accessed via `#!cpp operator[]` similar to a `#!cpp std::map` and a `#!cpp std::vector`, respectively.
+
+??? example
+
+    Consider the following JSON value:
+    
+    ```json
+    {
+        "name": "Mary Smith",
+        "age": 42,
+        "hobbies": ["hiking", "reading"]
+    }
+    ```
+    
+    Assume the value is parsed to a `json` variable `j`.
+
+    | expression | value |
+    | ---------- | ----- |
+    | `#!cpp j`  | `#!json {"name": "Mary Smith", "age": 42, "hobbies": ["hiking", "reading"]}` |
+    | `#!cpp j["name"]`  | `#!json "Mary Smith"` |
+    | `#!cpp j["age"]`  | `#!json 42` |
+    | `#!cpp j["hobbies"]`  | `#!json ["hiking", "reading"]` |
+    | `#!cpp j["hobbies"][0]`  | `#!json "hiking"` |
+    | `#!cpp j["hobbies"][1]`  | `#!json "reading"` |
+
+The return value is a reference, so it can be modify the original value. In case the passed object key is non-existing, a `#!json null` value is inserted which can be immediately be overwritten.
+
+??? example
+
+    ```cpp
+    j["name"] = "John Smith";
+    j["maidenName"] = "Jones";
+    ```
+    
+    This code produces the following JSON value:
+    
+    ```json
+    {
+        "name": "John Smith",
+        "maidenName": "Jones",
+        "age": 42,
+        "hobbies": ["hiking", "reading"]
+    }
+    ```
+
+When accessing an invalid index (i.e., and index greater than or equal to the array size), the JSON array is resized such that the passed index is the new maximal index. Intermediate values are filled with `#!json null`.
+
+??? example
+
+    ```cpp
+    j["hobbies"][0] = "running";
+    j["hobbies"][3] = "cooking";
+    ```
+    
+    This code produces the following JSON value:
+    
+    ```json
+    {
+        "name": "John Smith",
+        "maidenName": "Jones",
+        "age": 42,
+        "hobbies": ["running", "reading", null, "cooking"]
+    }
+    ```
+
+## Notes
+
+!!! info "Design rationale"
+
+    The library behaves differently to `#!cpp std::vector` and `#!cpp std::map`:
+    
+    - `#!cpp std::vector::operator[]` never inserts a new element.
+    - `#!cpp std::map::operator[]` is not available for const values.
+    
+    The type `#!cpp json` wraps all JSON value types. It would be impossible to remove `operator[]` for const objects. At the same time, inserting elements for non-const objects is really convenient as it avoids awkward `insert` calls. To this end, we decided to have an inserting non-const behavior for both arrays and objects.
+
+!!! info
+
+    The access is unchecked. In case the passed object key does not exist or the passed array index is invalid, no exception is thrown.
+
+!!! danger
+
+    - It is **undefined behavior** to access a const object with a non-existing key.
+    - It is **undefined behavior** to access a const array with an invalid index.
+    - In debug mode, an **assertion** will fire in both cases. You can disable assertions by defining the preprocessor symbol `#!cpp NDEBUG` or redefine the macro [`JSON_ASSERT(x)`](../macros.md#json_assertx).
+
+!!! failure "Exceptions"
+
+    `operator[]` can only be used with objects (with a string argument) or with arrays (with a numeric argument). For other types, a [`basic_json::type_error`](../../home/exceptions.md#jsonexceptiontype_error305) is thrown.
+
+## Summary
+
+| scenario | non-const value | const value |
+| -------- | ------------- | ----------- |
+| access to existing object key | reference to existing value is returned | const reference to existing value is returned |
+| access to valid array index | reference to existing value is returned | const reference to existing value is returned |
+| access to non-existing object key | reference to newly inserted `#!json null` value is returned | **undefined behavior**; assertion in debug mode |
+| access to invalid array index | reference to newly inserted `#!json null` value is returned; any index between previous maximal index and passed index are filled with `#!json null` | **undefined behavior**; assertion in debug mode |

doc/mkdocs/docs/features/macros.md

@@ -2,6 +2,10 @@
 
 Some aspects of the library can be configured by defining preprocessor macros before including the `json.hpp` header.
 
+## `JSON_ASSERT(x)`
+
+The default value is `#!cpp assert(x)`.
+
 ## `JSON_CATCH_USER(exception)`
 
 This macro overrides `#!cpp catch` calls inside the library. The argument is the type of the exception to catch. As of version 3.8.0, the library only catches `std::out_of_range` exceptions internally to rethrow them as [`json::out_of_range`](../home/exceptions.md#out-of-range) exceptions. The macro is always followed by a scope.

doc/mkdocs/docs/features/object_order.md

@@ -0,0 +1,67 @@
+# Object Order
+
+The [JSON standard](https://tools.ietf.org/html/rfc8259.html) defines objects as "an unordered collection of zero or more name/value pairs". As such, an implementation does not need to preserve any specific order of object keys.
+
+The default type `nlohmann::json` uses a `std::map` to store JSON objects, and thus stores object keys **sorted alphabetically**.
+
+??? example
+
+    ```cpp
+    #include <iostream>
+    #include "json.hpp"
+    
+    using json = nlohmann::json;
+    
+    int main()
+    {
+        json j;
+        j["one"] = 1;
+        j["two"] = 2;
+        j["three"] = 3;
+        
+        std::cout << j.dump(2) << '\n';
+    }
+    ```
+    
+    Output:
+
+    ```json
+    {
+      "one": 1,
+      "three": 3,
+      "two": 2
+    }
+    ```
+
+If you do want to preserve the **insertion order**, you can try the type [`nlohmann::ordered_json`](https://github.com/nlohmann/json/issues/2179).
+
+??? example
+
+    ```cpp
+    #include <iostream>
+    #include <nlohmann/json.hpp>
+    
+    using ordered_json = nlohmann::ordered_json;
+    
+    int main()
+    {
+        ordered_json j;
+        j["one"] = 1;
+        j["two"] = 2;
+        j["three"] = 3;
+        
+        std::cout << j.dump(2) << '\n';
+    }
+    ```
+    
+    Output:
+    
+    ```json
+    {
+      "one": 1,
+      "two": 2,
+      "three": 3
+    }
+    ```
+
+Alternatively, you can use a more sophisticated ordered map like [`tsl::ordered_map`](https://github.com/Tessil/ordered-map) ([integration](https://github.com/nlohmann/json/issues/546#issuecomment-304447518)) or [`nlohmann::fifo_map`](https://github.com/nlohmann/fifo_map) ([integration](https://github.com/nlohmann/json/issues/485#issuecomment-333652309)).

doc/mkdocs/docs/features/parsing/sax_interface.md

@@ -14,6 +14,7 @@ interface json::sax_t {
     + {abstract} bool number_float(number_float_t val, const string_t& s)
 
     + {abstract} bool string(string_t& val)
+    + {abstract} bool binary(binary_t& val)
 
     + {abstract} bool start_object(std::size_t elements)
     + {abstract} bool end_object()
@@ -41,6 +42,8 @@ bool number_float(number_float_t val, const string_t& s);
 
 // called when a string is parsed; value is passed and can be safely moved away
 bool string(string_t& val);
+// called when a binary value is parsed; value is passed and can be safely moved away
+bool binary(binary& val);
 
 // called when an object or array begins or ends, resp. The number of elements is passed (or -1 if not known)
 bool start_object(std::size_t elements);