6.7 KiB
Tutorial 0: Introduction and Common Patterns
Welcome to the first logos-libp2p-module tutorial!
Before creating nodes or connecting peers, it helps to understand the conventions used by every tutorial in this series.
The C++ wrapper (logos-libp2p-module) exposes a single main class, Libp2pModuleImpl.
Most methods on that class return the Logos result type. In C++ module code that
type is named StdLogosResult.
StdLogosResult has three fields:
| Field | Meaning |
|---|---|
success |
true when the call completed successfully |
value |
a nlohmann::json value returned by the call |
error |
a diagnostic string when success == false |
Always Check success First
Every tutorial follows the same rule: check unsuccessful outcomes before using returned data.
StdLogosResult info = node.peerInfo();
if (!info.success) {
fprintf(stderr, "Failed to get peer info: %s\n",
info.error.c_str());
return 1;
}
This keeps the code predictable: errors are handled immediately, and the rest of the step can assume the operation succeeded.
Tutorial Executables Fail Fast
These programs are examples, not long-running services. When a required
operation fails, they print a useful error message to stderr and return
1. Successful tutorials print progress to stdout and return 0.
Controlling libp2p Logs
The wrapped libp2p binding can emit runtime logs. Tutorials set the log
level to "fatal" by default so stdout only shows the tutorial's
own progress messages during normal runs. Choose another level such as
"info", "debug", or "trace" when you need more
detail while debugging:
setLogLevel("debug");
// ... or
Libp2pModuleImpl::setLogLevel("debug");
The level names are "none", "trace", "debug", "info", "notice",
"warn", "error" and "fatal"; any other name returns a failed result
and leaves the level as it was.
Log levels are inclusive minimum thresholds: "trace" emits trace
and above, "debug" emits debug and above, and so on.
"none" is the lowest threshold, so it emits all logs; it does not
disable logging. Use "fatal" for the quietest built-in threshold.
In your own application, "error" is often a useful default.
It keeps normal output quiet while still surfacing conditions
that may indicate libp2p is misbehaving or that your integration code needs an
adjustment. Some error logs describe remote-peer behavior, retries, or
recoverable internal state, so they may not require any action from your side.
Convert JSON Values Explicitly
The value field is JSON. Convert it to the C++ type you need at the
point where you use it:
std::string peerId = info.value["peerId"].get<std::string>();
for (const nlohmann::json& addr : info.value["addrs"]) {
printf(" %s\n", addr.get<std::string>().c_str());
}
Scalar values use get<T>(). Objects are accessed by key. Arrays are
iterated with range-for loops.
Tutorial code assumes returned JSON values have the documented type, so it uses and converts values directly instead of checking the JSON type first.
Binary Data May Be Encoded
Some APIs carry arbitrary bytes, such as stream reads, DHT values, public keys, or service discovery records. Those values may be returned as strings encoded for JSON transport. The tutorials decode them before comparing or printing human-readable payloads.
#include <cstdio>
#include <string>
#include "plugin.h"
int main()
{
printf("=== Tutorial 0: Introduction and Common Patterns ===\n\n");
// Silence logs from wrapped library.
setLogLevel("fatal");
Step 1: Create and start a node
Even the first real operation returns a result. Check it before moving on.
Libp2pModuleImpl node;
StdLogosResult startRes = node.start();
if (!startRes.success) {
fprintf(stderr, "Failed to start node: %s\n",
startRes.error.c_str());
return 1;
}
printf("Node started\n");
Step 2: Use scalar JSON values
Some calls return a single string, number, or boolean in value.
StdLogosResult version = node.getNodeInfo("Version");
if (!version.success) {
fprintf(stderr, "Failed to get module version: %s\n",
version.error.c_str());
return 1;
}
printf("Module version: %s\n",
version.value.get<std::string>().c_str());
Step 3: Use JSON objects and arrays
peerInfo() returns a JSON object. The peerId field is a string, while
addrs is an array because a node may listen on multiple addresses.
StdLogosResult info = node.peerInfo();
if (!info.success) {
fprintf(stderr, "Failed to get peer info: %s\n",
info.error.c_str());
return 1;
}
std::string peerId = info.value["peerId"].get<std::string>();
printf("Peer ID: %s\n", peerId.c_str());
printf("Listening addresses:\n");
for (const nlohmann::json& addr : info.value["addrs"]) {
printf(" %s\n", addr.get<std::string>().c_str());
}
Step 4: Stop cleanly
Production code should always attempt to close resources gracefully first.
The tutorials do not always process every cleanup outcome, such as failures while stopping a node or closing streams, in order to keep the example code focused and easy to follow.
StdLogosResult stopRes = node.stop();
if (!stopRes.success) {
fprintf(stderr, "Failed to stop node: %s\n",
stopRes.error.c_str());
return 1;
}
printf("\n=== Tutorial 0 Complete ===\n");
return 0;
}
Summary
In this tutorial you learned the conventions used everywhere else:
- Most calls return
StdLogosResult - Check
successbefore readingvalue - Print
errorwhen a call fails - Convert JSON values with
get<T>(), object keys, or array iteration - Return
1from tutorial executables when required operations fail
Run tutorial
Run this tutorial now to check that your environment is set up and the tutorial executable has already been built:
./build/tutorial/tutorial_0_introduction
If that command fails because the tutorial has not been built yet, build the tutorials with this command and then run it again:
nix --extra-experimental-features 'nix-command flakes' develop --command ./tutorial/build_tutorials.sh
Exercise: Enable libp2p debug logs
Change this tutorial's log level from "fatal" to "debug":
setLogLevel("debug");
Compile the tutorial binaries again, using the same command from above.