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
|
=============================================
Nanopb: Protocol Buffers with small code size
=============================================
.. include :: menu.rst
Nanopb is an ANSI-C library for encoding and decoding messages in Google's `Protocol Buffers`__ format with minimal requirements for RAM and code space.
It is primarily suitable for 32-bit microcontrollers.
__ http://code.google.com/apis/protocolbuffers/
Overall structure
=================
For the runtime program, you always need *pb.h* for type declarations.
Depending on whether you want to encode, decode or both, you also need *pb_encode.h/c* or *pb_decode.h/c*.
The high-level encoding and decoding functions take an array of *pb_field_t* structures, which describes the fields of a message structure. Usually you want these autogenerated from a *.proto* file. The tool string *nanopb_generator.py* accomplishes this.
So a typical project might include these files:
1) Nanopb runtime library:
- pb.h
- pb_decode.h and pb_decode.c
- pb_encode.h and pb_encode.c
2) Protocol description (you can have many):
- person.proto
- person.c (autogenerated, contains initializers for const arrays)
- person.h (autogenerated, contains type declarations)
Features and limitations
========================
**Features**
#) Pure C runtime
#) Small code size (2–10 kB depending on processor)
#) Small ram usage (typically 200 bytes)
#) Allows specifying maximum size for strings and arrays, so that they can be allocated statically.
#) No malloc needed: everything is stored on the stack.
#) You can use either encoder or decoder alone to cut the code size in half.
**Limitations**
#) User must provide callbacks when decoding arrays or strings without maximum size.
#) Some speed has been sacrificed for code size. For example varint calculations are always done in 64 bits.
#) Encoding is focused on writing to streams. For memory buffers only it could be made more efficient.
#) The deprecated Protocol Buffers feature called "groups" is not supported.
#) Fields in the generated structs are ordered by the tag number, instead of the natural ordering in .proto file.
Getting started
===============
For starters, consider this simple message::
message Example {
required int32 value = 1;
}
Save this in *example.proto* and compile it::
user@host:~$ protoc -omessage.pb message.proto
user@host:~$ python ../generator/nanopb_generator.py message.pb
You should now have in *example.h*::
typedef struct {
int32_t value;
} Example;
extern const pb_field_t Example_fields[2];
Now in your main program do this to encode a message::
Example mymessage = {42};
uint8_t buffer[10];
pb_ostream_t stream = pb_ostream_from_buffer(buffer, sizeof(buffer));
pb_encode(&stream, Example_fields, &mymessage);
After that, buffer will contain the encoded message.
The number of bytes in the message is stored in *stream.bytes_written*.
You can feed the message to *protoc --decode=Example example.proto* to verify its validity.
Wishlist
========
#) A specialized encoder for encoding to a memory buffer. Should serialize in reverse order to avoid having to determine submessage size beforehand.
#) A cleaner rewrite of the Python-based source generator.
#) Better performance for 16- and 8-bit platforms: use smaller datatypes where possible.
|