1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
|
# msgpack
For msgpack support, you must also include the header `<rfl/msgpack.hpp>` and link to the [msgpack-c](https://github.com/msgpack/msgpack-c) library.
Furthermore, when compiling reflect-cpp, you need to pass `-DREFLECTCPP_MSGPACK=ON` to cmake. If you are using vcpkg or Conan, there
should be an appropriate feature (vcpkg) or option (Conan) that will abstract this away for you.
msgpack is a JSON-like binary format with an emphasis on small binary sizes.
## Reading and writing
Suppose you have a struct like this:
```cpp
struct Person {
std::string first_name;
std::string last_name;
rfl::Timestamp<"%Y-%m-%d"> birthday;
std::vector<Person> children;
};
```
A `Person` can be serialized like this:
```cpp
const auto person = Person{...};
const std::vector<char> bytes = rfl::msgpack::write(person);
```
You can parse bytes like this:
```cpp
const rfl::Result<Person> result = rfl::msgpack::read<Person>(bytes);
```
## Loading and saving
You can also load and save to disc using a very similar syntax:
```cpp
const rfl::Result<Person> result = rfl::msgpack::load<Person>("/path/to/file.msgpack");
const auto person = Person{...};
rfl::msgpack::save("/path/to/file.msgpack", person);
```
## Reading from and writing into streams
You can also read from and write into any `std::istream` and `std::ostream` respectively.
```cpp
const rfl::Result<Person> result = rfl::msgpack::read<Person>(my_istream);
const auto person = Person{...};
rfl::msgpack::write(person, my_ostream);
```
Note that `std::cout` is also an ostream, so this works as well:
```cpp
rfl::msgpack::write(person, std::cout) << std::endl;
```
(Since msgpack is a binary format, the readability of this will be limited, but it might be useful for debugging).
## Custom constructors
One of the great things about C++ is that it gives you control over
when and how you code is compiled.
For large and complex systems of structs, it is often a good idea to split up
your code into smaller compilation units. You can do so using custom constructors.
For the msgpack format, these must be a static function on your struct or class called
`from_msgpack` that take a `rfl::msgpack::Reader::InputVarType` as input and return
the class or the class wrapped in `rfl::Result`.
In your header file you can write something like this:
```cpp
struct Person {
rfl::Rename<"firstName", std::string> first_name;
rfl::Rename<"lastName", std::string> last_name;
rfl::Timestamp<"%Y-%m-%d"> birthday;
using InputVarType = typename rfl::msgpack::Reader::InputVarType;
static rfl::Result<Person> from_msgpack(const InputVarType& _obj);
};
```
And in your source file, you implement `from_msgpack` as follows:
```cpp
rfl::Result<Person> Person::from_msgpack(const InputVarType& _obj) {
const auto from_nt = [](auto&& _nt) {
return rfl::from_named_tuple<Person>(std::move(_nt));
};
return rfl::msgpack::read<rfl::named_tuple_t<Person>>(_obj)
.transform(from_nt);
}
```
This will force the compiler to only compile the msgpack parsing when the source file is compiled.
|
