nwaku/docs/tutorial/chat2.md

8.2 KiB

Using the chat2 application

Background

The chat2 application is a basic command-line chat app using the Waku v2 suite of protocols. It optionally connects to a fleet of nodes to provide end-to-end p2p chat capabilities. Each fleet is a publicly accessible network of Waku v2 peers, providing a bootstrap connection point for new peers, historical message storage, etc. The Waku team is currently using this application on the sandbox fleet for internal testing. For more information on the available fleets, see Connecting to a Waku v2 fleet. If you want to try our protocols, or join the dogfooding fun, follow the instructions below.

Preparation

Ensure you have cloned the nim-waku repository and installed all prerequisites as per these instructions.

Make the chat2 target.

make chat2

Basic application usage

To start the chat2 application in its most basic form, run the following from the project directory

./build/chat2

You should be prompted to provide a nickname for the chat session.

Choose a nickname >>

After entering a nickname, the app will randomly select and connect to a peer from the sandbox fleet.

No static peers configured. Choosing one at random from sandbox fleet...

It will then attempt to download historical messages from a random peer in the sandbox fleet.

Store enabled, but no store nodes configured. Choosing one at random from sandbox fleet...

Wait for the chat prompt (>>) and chat away!

To gracefully exit the chat2 application, use the /exit in-chat option

>> /exit
quitting...

Retrieving historical messages

The chat2 application can retrieve historical chat messages from a node supporting and running the Waku v2 store protocol, and will attempt to do so by default. It's possible to query a specific store node by configuring its multiaddr as storenode when starting the app:

./build/chat2 --storenode:/dns4/node-01.do-ams3.waku.test.statusim.net/tcp/30303/p2p/16Uiu2HAkykgaECHswi3YKJ5dMLbq2kPVCo89fcyTd38UcQD6ej5W

Alternatively, the chat2 application will select a random storenode for you from the configured fleet (sandbox by default) if storenode is left unspecified.

./build/chat2

To disable historical message retrieval, use the --store:false option:

./build/chat2 --store:false

Specifying a static peer

In order to connect to a specific node as relay peer, define that node's multiaddr as a staticnode when starting the app:

./build/chat2 --staticnode:/dns4/node-01.do-ams3.waku.test.statusim.net/tcp/30303/p2p/16Uiu2HAkykgaECHswi3YKJ5dMLbq2kPVCo89fcyTd38UcQD6ej5W

This will bypass the random peer selection process and connect to the specified node.

Connecting to a Waku v2 fleet

It is possible to specify a specific Waku v2 fleet to connect to when starting the app by using the --fleet option:

./build/chat2 --fleet:test

There are currently two fleets to select from, namely sandbox (waku.sandbox) and test (waku.test). The test fleet is updated with each incremental change to the nim-waku codebase. As a result it may have more advanced and experimental features, but will be less stable than sandbox. The sandbox fleet is a deployed network of the latest released Waku v2 nodes. If no fleet is specified, chat2 will connect to the sandbox fleet by default. To start chat2 without connecting to a fleet, use the --fleet:none option or specify a static peer.

Specifying a content topic

To publish chat messages on a specific content topic, use the --content-topic option:

./build/chat2 --content-topic:/waku/2/my-content-topic/proto

NOTE: Currently (2021/05/26) the content topic defaults to /waku/2/huilong/proto if left unspecified, where huilong is the name of our latest testnet.

In-chat options

Command Effect
/help displays available in-chat commands
/connect interactively connect to a new peer
/nick change nickname for current chat session
/exit exits the current chat session

chat2 message protobuf format

Each chat2 message is encoded as follows

message Chat2Message {
  uint64 timestamp = 1;
  string nick = 2;
  bytes payload = 3;
}

where timestamp is the Unix timestamp of the message, nick is the relevant chat2 user's selected nickname and payload is the actual chat message being sent. The payload is the byte array representation of a UTF8 encoded string.

Bridge messages between chat2 and matterbridge

To facilitate chat2 use in a variety of contexts, a chat2bridge can be deployed to bridge messages between chat2 and any protocol supported by matterbridge.

Configure and run matterbridge

  1. Download and install matterbridge and configure an instance for the protocol(s) you want to bridge to. Basic configuration instructions here
  2. Configure the matterbridge API. This is used by the chat2bridge to relay chat2 messages to and from matterbridge. Configuration instructions for the matterbridge API can be found here. The full matterbridge API specification can be found here. The template below shows an example of a matterbridge.toml configuration file for bridging chat2 to Discord. Follow the matterbridge Discord instructions to configure your own Token and Server.
[discord.mydiscord]

# You can get your token by following the instructions on
# https://github.com/42wim/matterbridge/wiki/Discord-bot-setup.
# If you want roles/groups mentions to be shown with names instead of ID,
# you'll need to give your bot the "Manage Roles" permission.
Token="MTk4NjIyNDgzNDcdOTI1MjQ4.Cl2FMZ.ZnCjm1XVW7vRze4b7Cq4se7kKWs-abD"

Server="myserver" # picked from guilds the bot is connected to

RemoteNickFormat="{NICK}@chat2: "

[api.myapi]
BindAddress="127.0.0.1:4242"
Buffer=1000
RemoteNickFormat="{NICK}@{PROTOCOL}"

[[gateway]]
name="gateway1"
enable=true

[[gateway.inout]]
account="discord.mydiscord"
channel="general"

[[gateway.inout]]
account="api.myapi"
channel="api"
  1. Run matterbridge using the configuration file created in the previous step. Note the API listening address and port in the matterbridge logs (configured as the BindAddress in the previous step).
./matterbridge -conf matterbridge.toml
[0000]  INFO api:          Listening on 127.0.0.1:4242

Configure and run chat2bridge

  1. From the nim-waku project directory, make the chat2bridge target
make chat2bridge
  1. Run chat2bridge with the following configuration options:
--mb-host-address         Listening address of the Matterbridge host
--mb-host-port            Listening port of the Matterbridge host
--mb-gateway              Matterbridge gateway
./build/chat2bridge --mb-host-address=127.0.0.1 --mb-host-port=4242 --mb-gateway="gateway1"

Note that chat2bridge encompasses a full wakunode2 which can be configured with the normal configuration parameters. For a full list of configuration options, run --help.

./build/chat2bridge --help

Connect chat2bridge to a chat2 network

  1. To bridge messages on an existing chat2 network, connect to any relay peer(s) in that network from chat2bridge. This can be done by either specifying the peer(s) as a --staticnode when starting the chat2bridge or calling the post_waku_v2_admin_v1_peers method on the JSON-RPC API. Note that the latter requires the chat2bridge to be run with --rpc=true and --rpc-admin=true.
  2. To bridge from a new chat2 instance, simply specify the chat2bridge listening address as a chat2 static peer.