activitypub-federation-rust/src/config.rs

321 lines
11 KiB
Rust

//! Configuration for this library, with various federation settings
//!
//! Use [FederationConfig::builder](crate::config::FederationConfig::builder) to initialize it.
//!
//! ```
//! # use activitypub_federation::config::FederationConfig;
//! # tokio::runtime::Runtime::new().unwrap().block_on(async {
//! let settings = FederationConfig::builder()
//! .domain("example.com")
//! .app_data(())
//! .http_fetch_limit(50)
//! .build().await?;
//! # Ok::<(), anyhow::Error>(())
//! # }).unwrap()
//! ```
use crate::{
error::Error,
protocol::verification::verify_domains_match,
traits::{ActivityHandler, Actor},
};
use async_trait::async_trait;
use derive_builder::Builder;
use dyn_clone::{clone_trait_object, DynClone};
use moka::future::Cache;
use openssl::pkey::{PKey, Private};
use reqwest_middleware::ClientWithMiddleware;
use serde::de::DeserializeOwned;
use std::{
ops::Deref,
sync::{
atomic::{AtomicU32, Ordering},
Arc,
},
time::Duration,
};
use url::Url;
/// Configuration for this library, with various federation related settings
#[derive(Builder, Clone)]
#[builder(build_fn(private, name = "partial_build"))]
pub struct FederationConfig<T: Clone> {
/// The domain where this federated instance is running
#[builder(setter(into))]
pub(crate) domain: String,
/// Data which the application requires in handlers, such as database connection
/// or configuration.
pub(crate) app_data: T,
/// Maximum number of outgoing HTTP requests per incoming HTTP request. See
/// [crate::fetch::object_id::ObjectId] for more details.
#[builder(default = "20")]
pub(crate) http_fetch_limit: u32,
#[builder(default = "reqwest::Client::default().into()")]
/// HTTP client used for all outgoing requests. Middleware can be used to add functionality
/// like log tracing or retry of failed requests.
pub(crate) client: ClientWithMiddleware,
/// Run library in debug mode. This allows usage of http and localhost urls. It also sends
/// outgoing activities synchronously, not in background thread. This helps to make tests
/// more consistent. Do not use for production.
#[builder(default = "false")]
pub(crate) debug: bool,
/// Allow HTTP urls even in production mode
#[builder(default = "self.debug.unwrap_or(false)")]
pub(crate) allow_http_urls: bool,
/// Timeout for all HTTP requests. HTTP signatures are valid for 10s, so it makes sense to
/// use the same as timeout when sending
#[builder(default = "Duration::from_secs(10)")]
pub(crate) request_timeout: Duration,
/// Function used to verify that urls are valid, See [UrlVerifier] for details.
#[builder(default = "Box::new(DefaultUrlVerifier())")]
pub(crate) url_verifier: Box<dyn UrlVerifier + Sync>,
/// Enable to sign HTTP signatures according to draft 10, which does not include (created) and
/// (expires) fields. This is required for compatibility with some software like Pleroma.
/// <https://datatracker.ietf.org/doc/html/draft-cavage-http-signatures-10>
/// <https://git.pleroma.social/pleroma/pleroma/-/issues/2939>
#[builder(default = "false")]
pub(crate) http_signature_compat: bool,
/// Actor Id and private key to use to sign all federated fetch requests.
/// This can be used to implement secure mode federation.
/// <https://docs.joinmastodon.org/spec/activitypub/#secure-mode>
#[builder(default = "None", setter(custom))]
pub(crate) signed_fetch_actor: Option<Arc<(Url, PKey<Private>)>>,
#[builder(
default = "Cache::builder().max_capacity(10000).build()",
setter(custom)
)]
pub(crate) actor_pkey_cache: Cache<Url, PKey<Private>>,
}
impl<T: Clone> FederationConfig<T> {
/// Returns a new config builder with default values.
pub fn builder() -> FederationConfigBuilder<T> {
FederationConfigBuilder::default()
}
pub(crate) async fn verify_url_and_domain<Activity, Datatype>(
&self,
activity: &Activity,
) -> Result<(), Error>
where
Activity: ActivityHandler<DataType = Datatype> + DeserializeOwned + Send + 'static,
{
verify_domains_match(activity.id(), activity.actor())?;
self.verify_url_valid(activity.id()).await?;
if self.is_local_url(activity.id()) {
return Err(Error::UrlVerificationError(
"Activity was sent from local instance",
));
}
Ok(())
}
/// Create new [Data] from this. You should prefer to use a middleware if possible.
pub fn to_request_data(&self) -> Data<T> {
Data {
config: self.clone(),
request_counter: Default::default(),
}
}
/// Perform some security checks on URLs as mentioned in activitypub spec, and call user-supplied
/// [`InstanceSettings.verify_url_function`].
///
/// https://www.w3.org/TR/activitypub/#security-considerations
pub(crate) async fn verify_url_valid(&self, url: &Url) -> Result<(), Error> {
match url.scheme() {
"https" => {}
"http" => {
if !self.allow_http_urls {
return Err(Error::UrlVerificationError(
"Http urls are only allowed in debug mode",
));
}
}
_ => return Err(Error::UrlVerificationError("Invalid url scheme")),
};
// Urls which use our local domain are not a security risk, no further verification needed
if self.is_local_url(url) {
return Ok(());
}
if url.domain().is_none() {
return Err(Error::UrlVerificationError("Url must have a domain"));
}
if url.domain() == Some("localhost") && !self.debug {
return Err(Error::UrlVerificationError(
"Localhost is only allowed in debug mode",
));
}
self.url_verifier.verify(url).await?;
Ok(())
}
/// Returns true if the url refers to this instance. Handles hostnames like `localhost:8540` for
/// local debugging.
pub(crate) fn is_local_url(&self, url: &Url) -> bool {
let mut domain = url.host_str().expect("id has domain").to_string();
if let Some(port) = url.port() {
domain = format!("{}:{}", domain, port);
}
domain == self.domain
}
/// Returns the local domain
pub fn domain(&self) -> &str {
&self.domain
}
}
impl<T: Clone> FederationConfigBuilder<T> {
/// Sets an actor to use to sign all federated fetch requests
pub fn signed_fetch_actor<A: Actor>(&mut self, actor: &A) -> &mut Self {
let private_key_pem = actor
.private_key_pem()
.expect("actor does not have a private key to sign with");
let private_key = PKey::private_key_from_pem(private_key_pem.as_bytes())
.expect("Could not decode PEM data");
self.signed_fetch_actor = Some(Some(Arc::new((actor.id(), private_key))));
self
}
/// sets the number of parsed actor private keys to keep in memory
pub fn actor_pkey_cache(&mut self, cache_size: u64) -> &mut Self {
self.actor_pkey_cache = Some(Cache::builder().max_capacity(cache_size).build());
self
}
/// Constructs a new config instance with the values supplied to builder.
///
/// Values which are not explicitly specified use the defaults. Also initializes the
/// queue for outgoing activities, which is stored internally in the config struct.
/// Requires a tokio runtime for the background queue.
pub async fn build(&mut self) -> Result<FederationConfig<T>, FederationConfigBuilderError> {
let config = self.partial_build()?;
Ok(config)
}
}
impl<T: Clone> Deref for FederationConfig<T> {
type Target = T;
fn deref(&self) -> &Self::Target {
&self.app_data
}
}
/// Handler for validating URLs.
///
/// This is used for implementing domain blocklists and similar functionality. It is called
/// with the ID of newly received activities, when fetching remote data from a given URL
/// and before sending an activity to a given inbox URL. If processing for this domain/URL should
/// be aborted, return an error. In case of `Ok(())`, processing continues.
///
/// ```
/// # use async_trait::async_trait;
/// # use url::Url;
/// # use activitypub_federation::config::UrlVerifier;
/// # use activitypub_federation::error::Error;
/// # #[derive(Clone)]
/// # struct DatabaseConnection();
/// # async fn get_blocklist(_: &DatabaseConnection) -> Vec<String> {
/// # vec![]
/// # }
/// #[derive(Clone)]
/// struct Verifier {
/// db_connection: DatabaseConnection,
/// }
///
/// #[async_trait]
/// impl UrlVerifier for Verifier {
/// async fn verify(&self, url: &Url) -> Result<(), Error> {
/// let blocklist = get_blocklist(&self.db_connection).await;
/// let domain = url.domain().unwrap().to_string();
/// if blocklist.contains(&domain) {
/// Err(Error::Other("Domain is blocked".into()))
/// } else {
/// Ok(())
/// }
/// }
/// }
/// ```
#[async_trait]
pub trait UrlVerifier: DynClone + Send {
/// Should return Ok iff the given url is valid for processing.
async fn verify(&self, url: &Url) -> Result<(), Error>;
}
/// Default URL verifier which does nothing.
#[derive(Clone)]
struct DefaultUrlVerifier();
#[async_trait]
impl UrlVerifier for DefaultUrlVerifier {
async fn verify(&self, _url: &Url) -> Result<(), Error> {
Ok(())
}
}
clone_trait_object!(UrlVerifier);
/// Stores data for handling one specific HTTP request.
///
/// It gives acess to the `app_data` which was passed to [FederationConfig::builder].
///
/// Additionally it contains a counter for outgoing HTTP requests. This is necessary to
/// prevent denial of service attacks, where an attacker triggers fetching of recursive objects.
///
/// <https://www.w3.org/TR/activitypub/#security-recursive-objects>
pub struct Data<T: Clone> {
pub(crate) config: FederationConfig<T>,
pub(crate) request_counter: AtomicU32,
}
impl<T: Clone> Data<T> {
/// Returns the data which was stored in [FederationConfigBuilder::app_data]
pub fn app_data(&self) -> &T {
&self.config.app_data
}
/// The domain that was configured in [FederationConfig].
pub fn domain(&self) -> &str {
&self.config.domain
}
/// Returns a new instance of `Data` with request counter set to 0.
pub fn reset_request_count(&self) -> Self {
Data {
config: self.config.clone(),
request_counter: Default::default(),
}
}
/// Total number of outgoing HTTP requests made with this data.
pub fn request_count(&self) -> u32 {
self.request_counter.load(Ordering::Relaxed)
}
}
impl<T: Clone> Deref for Data<T> {
type Target = T;
fn deref(&self) -> &T {
&self.config.app_data
}
}
/// Middleware for HTTP handlers which provides access to [Data]
#[derive(Clone)]
pub struct FederationMiddleware<T: Clone>(pub(crate) FederationConfig<T>);
impl<T: Clone> FederationMiddleware<T> {
/// Construct a new middleware instance
pub fn new(config: FederationConfig<T>) -> Self {
FederationMiddleware(config)
}
}