NanoPB: How to handle nested messages in C++
See also: C version: How to handle nested messages in C
NanoPB is a code-size optimized Protocol Buffers implementation for embedded systems. This post shows how to handle nested messages in C++ with NanoPB.
Proto definition
First, create a .proto file with nested messages:
syntax = "proto3";
package example;
message Address {
string street = 1;
string city = 2;
string country = 3;
}
message Person {
string name = 1;
uint32 age = 2;
Address address = 3;
}Generate NanoPB code
Generate the NanoPB code with a .options file to specify string buffer sizes:
Create nested.options:
example.Address.street max_size:64
example.Address.city max_size:32
example.Address.country max_size:32
example.Person.name max_size:64Then generate:
protoc --nanopb_out=. nested.protoThis will generate nested.pb.h and nested.pb.c.
C++ example
Here’s a complete C++ example handling nested messages:
#include <stdio.h>
#include <string.h>
#include "nested.pb.h"
#include "pb_encode.h"
#include "pb_decode.h"
int main() {
// Buffer for encoded message
uint8_t buffer[256];
size_t message_length;
// --- ENCODING ---
example_Person person = example_Person_init_zero;
// Set nested message values
strncpy(person.name, "John Doe", sizeof(person.name) - 1);
person.has_address = true;
strncpy(person.address.street, "123 Main St", sizeof(person.address.street) - 1);
strncpy(person.address.city, "Springfield", sizeof(person.address.city) - 1);
strncpy(person.address.country, "USA", sizeof(person.address.country) - 1);
// Create stream for encoding
pb_ostream_t ostream = pb_ostream_from_buffer(buffer, sizeof(buffer));
// Encode the message
if (!pb_encode(&ostream, example_Person_fields, &person)) {
printf("Encoding failed: %s\n", PB_GET_ERROR(&ostream));
return 1;
}
message_length = ostream.bytes_written;
printf("Encoded %zu bytes\n", message_length);
// Print hex dump of encoded data
printf("Encoded data: ");
for (size_t i = 0; i < message_length; i++) {
printf("%02x ", buffer[i]);
}
printf("\n");
// --- DECODING ---
example_Person decoded = example_Person_init_zero;
// Create stream for decoding
pb_istream_t istream = pb_istream_from_buffer(buffer, message_length);
// Decode the message
if (!pb_decode(&istream, example_Person_fields, &decoded)) {
printf("Decoding failed: %s\n", PB_GET_ERROR(&istream));
return 1;
}
// Print decoded values
printf("Decoded values:\n");
printf(" name: %s\n", decoded.name);
printf(" age: %u\n", decoded.age);
printf(" address:\n");
printf(" street: %s\n", decoded.address.street);
printf(" city: %s\n", decoded.address.city);
printf(" country: %s\n", decoded.address.country);
return 0;
}Compile command
Compile the example with nanopb. NanoPB is typically used by including the source files directly in your project:
g++ -o nested_example nested_example.cpp nested.pb.c pb_common.c pb_encode.c pb_decode.c -I.Note: NanoPB source files (pb_common.c, pb_encode.c, pb_decode.c) need to be compiled directly with your project. You can obtain these from the NanoPB GitHub repository.
Python test script
To verify the encoding, you can use Python’s protobuf library:
import nested_pb2
# Read the binary data
with open('encoded.bin', 'rb') as f:
data = f.read()
# Decode
msg = nested_pb2.Person()
msg.ParseFromString(data)
print("Python decoded values:")
print(f" name: {msg.name}")
print(f" age: {msg.age}")
print(f" address:")
print(f" street: {msg.address.street}")
print(f" city: {msg.address.city}")
print(f" country: {msg.address.country}")First, compile the Python protobuf definitions:
protoc --python_out=. nested.protoThen modify the C++ example to save the encoded data to a file:
// After encoding, add this:
FILE *f = fopen("encoded.bin", "wb");
fwrite(buffer, 1, message_length, f);
fclose(f);Example with optional nested message
Here’s an example with an optional nested message:
syntax = "proto3";
package example;
message Address {
string street = 1;
string city = 2;
}
message Person {
string name = 1;
optional Address address = 2;
}Create nested_optional.options:
example.Address.street max_size:64
example.Address.city max_size:32
example.Person.name max_size:64#include <stdio.h>
#include <string.h>
#include "nested_optional.pb.h"
#include "pb_encode.h"
#include "pb_decode.h"
int main() {
uint8_t buffer[256];
size_t message_length;
// --- ENCODING ---
example_Person person = example_Person_init_zero;
strncpy(person.name, "Jane Doe", sizeof(person.name) - 1);
// Set optional nested message
strncpy(person.address.street, "456 Oak Ave", sizeof(person.address.street) - 1);
strncpy(person.address.city, "Boston", sizeof(person.address.city) - 1);
person.has_address = true; // Important: set has_* flag
pb_ostream_t ostream = pb_ostream_from_buffer(buffer, sizeof(buffer));
if (!pb_encode(&ostream, example_Person_fields, &person)) {
printf("Encoding failed: %s\n", PB_GET_ERROR(&ostream));
return 1;
}
message_length = ostream.bytes_written;
printf("Encoded %zu bytes\n", message_length);
// --- DECODING ---
example_Person decoded = example_Person_init_zero;
pb_istream_t istream = pb_istream_from_buffer(buffer, message_length);
if (!pb_decode(&istream, example_Person_fields, &decoded)) {
printf("Decoding failed: %s\n", PB_GET_ERROR(&istream));
return 1;
}
printf("Decoded values:\n");
printf(" name: %s\n", decoded.name);
if (decoded.has_address) {
printf(" address:\n");
printf(" street: %s\n", decoded.address.street);
printf(" city: %s\n", decoded.address.city);
} else {
printf(" address: (not set)\n");
}
return 0;
}Key points
- Nested messages: Messages can contain other messages as fields
- Initialization: Use
*_init_zeroto initialize both outer and nested messages - Encoding: Set nested message fields directly, then encode the outer message
- Decoding: Decode the outer message, nested messages are decoded automatically
- Optional nested: Use
optionalkeyword and sethas_*flag - Structure: Nested messages provide hierarchical data organization
- Buffer sizes: Specify buffer sizes for all string fields in .options file
When to use nested messages
- When you have hierarchical or grouped data
- When you want to reuse message definitions
- When you need to organize related fields together
- When you want to improve message readability
- When you have complex data structures
Expected output
Encoded 35 bytes
Encoded data: 0a 08 4a 6f 68 6e 20 44 6f 65 10 1e 1a 1b 0a 0b 31 32 33 20 4d 61 69 6e 20 53 74 12 08 4e 65 77 20 59 6f 72 6b 1a 03 55 53 41
Decoded values:
name: John Doe
age: 30
address:
street: 123 Main St
city: New York
country: USAExpected output (optional nested)
Encoded 27 bytes
Decoded values:
name: Jane Doe
address:
street: 456 Oak Ave
city: BostonMore NanoPB posts
- Basic scalar types in C++
- Basic scalar types in C
- String types in C++
- String types in C
- Bytes types in C++
- Bytes types in C
- Optional fields in C++
- Optional fields in C
- Repeated fields/arrays in C++
- Repeated fields/arrays in C
- Enums in C++
- Enums in C
- Nested messages in C
- Oneof/union types in C++
- Oneof/union types in C
- Custom array converters in C++
- Custom array converters in C