Instead of making the `.dht` and `.pubsub` properties optional, use dummy implementations that throw exceptions if they are not configured.
This way we don't have to null guard everywhere they are accessed.
Addresses PR comments from #1172 - fixes syntax of examples in docs, adds the transport manager to the exports map and renames fault tolerance enum for consistency.
Converts this module to typescript.
- Ecosystem modules renamed from (e.g.) `libp2p-tcp` to `@libp2p/tcp`
- Ecosystem module now have named exports
- Configuration has been updated, now pass instances of modules instead of classes:
- Some configuration keys have been renamed to make them more descriptive. `transport` -> `transports`, `connEncryption` -> `connectionEncryption`. In general where we pass multiple things, the key is now plural, e.g. `streamMuxer` -> `streamMuxers`, `contentRouting` -> `contentRouters`, etc. Where we are configuring a singleton the config key is singular, e.g. `connProtector` -> `connectionProtector` etc.
- Properties of the `modules` config key have been moved to the root
- Properties of the `config` config key have been moved to the root
```js
// before
import Libp2p from 'libp2p'
import TCP from 'libp2p-tcp'
await Libp2p.create({
modules: {
transport: [
TCP
],
}
config: {
transport: {
[TCP.tag]: {
foo: 'bar'
}
},
relay: {
enabled: true,
hop: {
enabled: true,
active: true
}
}
}
})
```
```js
// after
import { createLibp2p } from 'libp2p'
import { TCP } from '@libp2p/tcp'
await createLibp2p({
transports: [
new TCP({ foo: 'bar' })
],
relay: {
enabled: true,
hop: {
enabled: true,
active: true
}
}
})
```
- Use of `enabled` flag has been reduced - previously you could pass a module but disable it with config. Now if you don't want a feature, just don't pass an implementation. Eg:
```js
// before
await Libp2p.create({
modules: {
transport: [
TCP
],
pubsub: Gossipsub
},
config: {
pubsub: {
enabled: false
}
}
})
```
```js
// after
await createLibp2p({
transports: [
new TCP()
]
})
```
- `.multiaddrs` renamed to `.getMultiaddrs()` because it's not a property accessor, work is done by that method to calculate announce addresses, observed addresses, etc
- `/p2p/${peerId}` is now appended to all addresses returned by `.getMultiaddrs()` so they can be used opaquely (every consumer has to append the peer ID to the address to actually use it otherwise). If you need low-level unadulterated addresses, call methods on the address manager.
BREAKING CHANGE: types are no longer hand crafted, this module is now ESM only
When the node is shutting down, new connections can still be received.
If this happens we can end up writing into the datastore when it's
been closed which throws an error.
Instead, if we're not running, have the connection manager close new
incoming connections.
To allow us to shut down cleanly, we must wait the unhandling of protocols - this is because they write the new list of protocols into the datastore which might also be in the process of shutting down.
The order of the addresses can affect our success rate in dialing a
relay - if it's a loopback address or similar it won't work.
Instead try dialing every address.
We have a peerstore that keeps all data for all observed peers in memory with no eviction.
This is fine when you don't discover many peers but when using the DHT you encounter a significant number of peers so our peer storage grows and grows over time.
We have a persistent peer store, but it just periodically writes peers into the datastore to be read at startup, still keeping them in memory.
It also means a restart doesn't give you any temporary reprieve from the memory leak as the previously observed peer data is read into memory at startup.
This change refactors the peerstore to use a datastore by default, reading and writing peer info as it arrives. It can be configured with a MemoryDatastore if desired.
It was necessary to change the peerstore and *book interfaces to be asynchronous since the datastore api is asynchronous.
BREAKING CHANGE: `libp2p.handle`, `libp2p.registrar.register` and the peerstore methods have become async
The `AbortController` class is supported by browsers and node 14+ - we only support node 16+ (e.g. LTS+Current) so the `abort-controller` module isn't needed any more.
When we've previously seen loads of peers and stored them in the
datastore we'll try to dial them as part of starting the autodial
component.
If we or our peers have bad network connections this can make
starting a libp2p node take ages so don't wait for a round of auto
dialing before considering the component started.
