1============================================= 2Nanopb: Protocol Buffers with small code size 3============================================= 4 5.. include :: menu.rst 6 7Nanopb is an ANSI-C library for encoding and decoding messages in Google's `Protocol Buffers`__ format with minimal requirements for RAM and code space. 8It is primarily suitable for 32-bit microcontrollers. 9 10__ http://code.google.com/apis/protocolbuffers/ 11 12Overall structure 13================= 14 15For the runtime program, you always need *pb.h* for type declarations. 16Depending on whether you want to encode, decode, or both, you also need *pb_encode.h/c* or *pb_decode.h/c*. 17 18The 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 script *nanopb_generator.py* accomplishes this. 19 20.. image:: generator_flow.png 21 22So a typical project might include these files: 23 241) Nanopb runtime library: 25 - pb.h 26 - pb_decode.h and pb_decode.c (needed for decoding messages) 27 - pb_encode.h and pb_encode.c (needed for encoding messages) 282) Protocol description (you can have many): 29 - person.proto (just an example) 30 - person.pb.c (autogenerated, contains initializers for const arrays) 31 - person.pb.h (autogenerated, contains type declarations) 32 33Features and limitations 34======================== 35 36**Features** 37 38#) Pure C runtime 39#) Small code size (2–10 kB depending on processor, plus any message definitions) 40#) Small ram usage (typically ~300 bytes, plus any message structs) 41#) Allows specifying maximum size for strings and arrays, so that they can be allocated statically. 42#) No malloc needed: everything can be allocated statically or on the stack. 43#) You can use either encoder or decoder alone to cut the code size in half. 44#) Support for most protobuf features, including: all data types, nested submessages, default values, repeated and optional fields, packed arrays, extension fields. 45#) Callback mechanism for handling messages larger than can fit in available RAM. 46#) Extensive set of tests. 47 48**Limitations** 49 50#) Some speed has been sacrificed for code size. 51#) Encoding is focused on writing to streams. For memory buffers only it could be made more efficient. 52#) The deprecated Protocol Buffers feature called "groups" is not supported. 53#) Fields in the generated structs are ordered by the tag number, instead of the natural ordering in .proto file. 54#) Unknown fields are not preserved when decoding and re-encoding a message. 55#) Reflection (runtime introspection) is not supported. E.g. you can't request a field by giving its name in a string. 56#) Numeric arrays are always encoded as packed, even if not marked as packed in .proto. This causes incompatibility with decoders that do not support packed format. 57#) Cyclic references between messages are supported only in callback mode. 58 59Getting started 60=============== 61 62For starters, consider this simple message:: 63 64 message Example { 65 required int32 value = 1; 66 } 67 68Save this in *message.proto* and compile it:: 69 70 user@host:~$ protoc -omessage.pb message.proto 71 user@host:~$ python nanopb/generator/nanopb_generator.py message.pb 72 73You should now have in *message.pb.h*:: 74 75 typedef struct { 76 int32_t value; 77 } Example; 78 79 extern const pb_field_t Example_fields[2]; 80 81Now in your main program do this to encode a message:: 82 83 Example mymessage = {42}; 84 uint8_t buffer[10]; 85 pb_ostream_t stream = pb_ostream_from_buffer(buffer, sizeof(buffer)); 86 pb_encode(&stream, Example_fields, &mymessage); 87 88After that, buffer will contain the encoded message. 89The number of bytes in the message is stored in *stream.bytes_written*. 90You can feed the message to *protoc --decode=Example message.proto* to verify its validity. 91 92For a complete example of the simple case, see *example/simple.c*. 93For a more complex example with network interface, see the *example/network_server* subdirectory. 94 95Compiler requirements 96===================== 97Nanopb should compile with most ansi-C compatible compilers. It however 98requires a few header files to be available: 99 100#) *string.h*, with these functions: *strlen*, *memcpy*, *memset* 101#) *stdint.h*, for definitions of *int32_t* etc. 102#) *stddef.h*, for definition of *size_t* 103#) *stdbool.h*, for definition of *bool* 104 105If these header files do not come with your compiler, you can use the 106file *extra/pb_syshdr.h* instead. It contains an example of how to provide 107the dependencies. You may have to edit it a bit to suit your custom platform. 108 109To use the pb_syshdr.h, define *PB_SYSTEM_HEADER* as *"pb_syshdr.h"* (including the quotes). 110Similarly, you can provide a custom include file, which should provide all the dependencies 111listed above. 112 113Running the test cases 114====================== 115Extensive unittests and test cases are included under the *tests* folder. 116 117To build the tests, you will need the `scons`__ build system. The tests should 118be runnable on most platforms. Windows and Linux builds are regularly tested. 119 120__ http://www.scons.org/ 121 122In addition to the build system, you will also need a working Google Protocol 123Buffers *protoc* compiler, and the Python bindings for Protocol Buffers. On 124Debian-based systems, install the following packages: *protobuf-compiler*, 125*python-protobuf* and *libprotobuf-dev*. 126 127
