Create hostmeta and nodeinfo crates (#1)

Reviewed-on: #1

.editorconfig

@@ -0,0 +1,21 @@
+root = true
+
+[*]
+charset = utf-8
+end_of_line = lf
+indent_size = 4
+indent_style = tab
+insert_final_newline = true
+tab_width = 4
+
+[*.yml]
+indent_size = 2
+indent_style = space
+
+[*.md]
+indent_size = 3
+indent_style = space
+
+[*.rst]
+indent_size = 3
+indent_style = space

Cargo.toml

@@ -1,3 +1,3 @@
 [workspace]
 resolver = "2"
-members = ["acrate-core", "acrate-nodeinfo"]
+members = ["acrate-core", "acrate-hostmeta", "acrate-nodeinfo"]

acrate-core/Cargo.toml

@@ -6,3 +6,8 @@ edition = "2021"
 [dependencies]
 diesel = "2.2.4"
 diesel_migrations = "2.2.0"
+acrate-hostmeta = { path = "../acrate-hostmeta" }
+acrate-nodeinfo = { path = "../acrate-nodeinfo" }
+
+[lints.clippy]
+tabs-in-doc-comments = "allow"

acrate-core/migrations/2024-11-09-084403_Add nodeinfo table/down.sql

@@ -0,0 +1 @@
+DROP TABLE nodeinfo;

acrate-core/migrations/2024-11-09-084403_Add nodeinfo table/up.sql

@@ -0,0 +1,9 @@
+CREATE TABLE nodeinfo (
+	nodeinfo_schema		VARCHAR		NOT NULL,
+	nodeinfo_href		VARCHAR		NOT NULL,
+	nodeinfo_data		JSON		NOT NULL,
+
+	last_updated		TIMESTAMP	NOT NULL,
+
+	PRIMARY KEY(nodeinfo_href)
+);

acrate-core/src/lib.rs

@@ -1,14 +1,4 @@
-pub fn add(left: u64, right: u64) -> u64 {
+//! Core crate of the `acrate` project.
-    left + right
-}
 
-#[cfg(test)]
+pub use acrate_nodeinfo as nodeinfo;
-mod tests {
+pub use acrate_hostmeta as hostmeta;
-    use super::*;
-
-    #[test]
-    fn it_works() {
-        let result = add(2, 2);
-        assert_eq!(result, 4);
-    }
-}

acrate-hostmeta/Cargo.toml

@@ -0,0 +1,19 @@
+[package]
+name = "acrate-hostmeta"
+version = "0.1.0"
+edition = "2021"
+
+[dependencies]
+log = "0.4.22"
+quick-xml = { version = "0.37.0", features = ["overlapped-lists", "serialize"] }
+reqwest = { version = "0.12.9", features = ["json", "stream"] }
+serde = { version = "1.0.214", features = ["derive"] }
+serde_json = "1.0.132"
+
+[dev-dependencies]
+pretty_env_logger = "0.5.0"
+tokio = { version = "1.41.1", features = ["macros", "rt-multi-thread"] }
+tokio-test = "0.4.4"
+
+[lints.clippy]
+tabs-in-doc-comments = "allow"

acrate-hostmeta/src/lib.rs

@@ -0,0 +1,548 @@
+//! Resource descriptior deserializer.
+//! 
+//! # Specification
+//! 
+//! - <https://datatracker.ietf.org/doc/html/rfc6415>
+//! - <https://datatracker.ietf.org/doc/html/rfc7033>
+
+use std::collections::HashMap;
+
+use serde::Deserialize;
+
+/// A resource descriptor object.
+/// 
+/// # Specification
+/// 
+/// - <https://datatracker.ietf.org/doc/html/rfc6415#section-3>
+/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4>
+/// 
+#[derive(Debug, Clone, Deserialize)]
+pub struct ResourceDescriptor {
+	/// The resource this document refers to.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4.1>
+	/// 
+	#[serde(alias = "Subject")]
+	pub subject: Option<String>,
+
+	/// Other names the resource described by this document can be referred to.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4.2>
+	/// 
+	#[serde(alias = "Alias")]
+	pub aliases: Option<Vec<String>>,
+
+	/// Additional information about the resource described by this document.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4.3>
+	/// 
+	#[serde(alias = "Property")]
+	pub properties: Option<Vec<ResourceDescriptorProperty>>,
+
+	/// Links established between the [`Self::subject`] and other resources.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc6415#section-3.1.1>
+	/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4.4>
+	/// 
+	#[serde(alias = "Link")]
+	pub links: Option<Vec<ResourceDescriptorLink>>,
+}
+
+/// A link element, which puts the subject resource in relation with another.
+/// 
+/// # Specification
+/// 
+/// - <https://datatracker.ietf.org/doc/html/rfc6415#section-3.1.1>
+/// 
+#[derive(Debug, Clone, Deserialize)]
+pub struct ResourceDescriptorLink {
+	/// The kind of relation established by the subject with the attached resource.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4.4.1>
+	/// 
+  	#[serde(alias = "@rel")]
+	pub rel: String,
+
+	/// The media type of the resource put in relation.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4.4.2>
+	/// 
+  	#[serde(alias = "@type")]
+	pub r#type: Option<String>,
+
+	/// URI to the resource put in relation.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4.4.3>
+	/// 
+  	#[serde(alias = "@href")]
+	pub href: Option<String>,
+
+	/// Titles of the resource put in relation in various languages.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4.4.4>
+	/// 
+	pub titles: Option<Vec<HashMap<String, String>>>,
+
+	/// Additional information about the resource put in relation.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4.4.5>
+	/// 
+	pub properties: Option<Vec<ResourceDescriptorProperty>>,
+
+	/// Template to fill to get the URL to resource-specific information.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc6415#section-4.2>
+	/// 
+  	#[serde(alias = "@template")]
+	pub template: Option<String>,
+}
+
+/// A property element, which describes a certain aspect of the subject resource.
+/// 
+/// # Specification
+/// 
+/// - <https://datatracker.ietf.org/doc/html/rfc7033#section-4.4.3>
+/// 
+#[derive(Debug, Clone, Deserialize)]
+pub struct ResourceDescriptorProperty {
+	/// The property identifier, or type.
+	#[serde(alias = "@type")]
+	pub r#type: String,
+
+	/// The property value.
+	pub value: Option<String>,
+}
+
+impl ResourceDescriptor {
+	/// Get a JRD (JSON [`ResourceDescriptor`]).
+	/// 
+	/// # Notes
+	/// 
+	/// This follows redirects until the redirect chain is 10 hops; see [`reqwest::redirect`] for more info.
+	/// 
+	/// # Examples
+	/// 
+	/// ```
+	/// # tokio_test::block_on(async {
+	/// use acrate_hostmeta::ResourceDescriptor;
+	/// 
+	/// let client = reqwest::Client::new();
+	/// let url: reqwest::Url = "https://junimo.party/.well-known/nodeinfo".parse()
+	/// 	.expect("URL to be valid");
+	/// 
+	/// let rd = ResourceDescriptor::get_jrd(&client, url)
+	/// 	.await
+	/// 	.expect("JRD to be processed correctly");
+	/// # })
+	/// ```
+	/// 
+	pub async fn get_jrd(client: &reqwest::Client, url: reqwest::Url) -> Result<Self, GetJRDError> {
+		use GetJRDError::*;
+
+		log::debug!("Getting JRD document at: {url}");
+
+		log::trace!("Building request...");
+		let request = {
+			log::trace!("Creating new request...");
+			let mut request = reqwest::Request::new(reqwest::Method::GET, url);
+
+			log::trace!("Setting request headers...");
+			let headers = request.headers_mut();
+
+			log::trace!("Setting `Accept: application/json`...");
+			let _ = headers.insert(reqwest::header::ACCEPT, "application/json".parse().unwrap());
+
+			request
+		};
+
+		log::trace!("Sending request...");
+		let response = client.execute(request)
+			.await
+			.map_err(Request)?;
+
+		log::trace!("Checking `Content-Type` of the response...");
+		let content_type = response
+			.headers()
+			.get(reqwest::header::CONTENT_TYPE)
+			.ok_or(ContentTypeMissing)?;
+
+		log::trace!("Extracting MIME type from the `Content-Type` header...");
+		let mime_type = extract_mime_from_content_type(content_type)
+			.ok_or(ContentTypeInvalid)?;
+
+		log::trace!("Ensuring MIME type is acceptable for JRD parsing...");
+		if !(mime_type == "application/json" || mime_type == "application/jrd+json") {
+			log::error!("MIME type `{mime_type}` is not acceptable for JRD parsing.");
+			return Err(ContentTypeInvalid)
+		}
+
+		log::trace!("Attempting to parse response as JSON...");
+		let data = response.json::<Self>()
+			.await
+			.map_err(Parse)?;
+
+		Ok(data)
+	}
+
+	/// Get a XRD (Extensible [`ResourceDescriptor`]).
+	/// 
+	/// # Notes
+	/// 
+	/// This follows redirects until the redirect chain is 10 hops; see [`reqwest::redirect`] for more info.
+	/// 
+	/// # Examples
+	/// 
+	/// ```
+	/// # tokio_test::block_on(async {
+	/// use acrate_hostmeta::ResourceDescriptor;
+	/// 
+	/// let client = reqwest::Client::new();
+	/// let url: reqwest::Url = "https://junimo.party/.well-known/host-meta".parse()
+	/// 	.expect("URL to be valid");
+	/// 
+	/// let rd = ResourceDescriptor::get_xrd(&client, url)
+	/// 	.await
+	/// 	.expect("XRD to be processed correctly");
+	/// # })
+	/// ```
+	/// 
+	pub async fn get_xrd(client: &reqwest::Client, url: reqwest::Url) -> Result<Self, GetXRDError> {
+		use GetXRDError::*;
+
+		log::debug!("Getting host-meta XRD document at: {url}");
+
+		log::trace!("Building request...");
+		let request = {
+			log::trace!("Creating new request...");
+			let mut request = reqwest::Request::new(reqwest::Method::GET, url);
+
+			log::trace!("Setting request headers...");
+			let headers = request.headers_mut();
+
+			log::trace!("Setting `Accept: application/xrd+xml`...");
+			let _ = headers.insert(reqwest::header::ACCEPT, "application/xrd+xml".parse().unwrap());
+
+			request
+		};
+
+		log::trace!("Sending request...");
+		let response = client.execute(request)
+			.await
+			.map_err(Request)?;
+
+		log::trace!("Checking `Content-Type` of the response...");
+		let content_type = response
+			.headers()
+			.get(reqwest::header::CONTENT_TYPE)
+			.ok_or(ContentTypeMissing)?;
+
+		log::trace!("Extracting MIME type from the `Content-Type` header...");
+		let mime_type = extract_mime_from_content_type(content_type)
+			.ok_or(ContentTypeInvalid)?;
+
+		log::trace!("Ensuring MIME type is acceptable for XRD parsing...");
+		if mime_type != "application/xrd+xml" {
+			log::error!("MIME type `{mime_type}` is not acceptable for XRD parsing.");
+			return Err(ContentTypeInvalid)
+		}
+
+		log::trace!("Attempting to parse response as text...");
+		let data = response.text()
+			.await
+			.map_err(Decode)?;
+
+		log::trace!("Parsing response as XML...");
+		let data = quick_xml::de::from_str::<Self>(&data)
+			.map_err(Parse)?;
+
+		Ok(data)
+	}
+	
+	/// Attempt to discover a [`ResourceDescriptor`] at the given URL in various ways.
+	/// 
+	/// In order, this method attempts:
+	/// 
+	/// 1. HTTPS [XRD](Self::get_xrd)
+	/// 2. HTTPS [JRD](Self::get_jrd)
+	/// 3. HTTPS [JRD](Self::get_jrd) with .json path extension
+	/// 4. HTTP [XRD](Self::get_xrd)
+	/// 5. HTTP [JRD](Self::get_jrd)
+	/// 6. HTTP [JRD](Self::get_jrd) with .json path extension
+	/// 
+	/// # Notes
+	/// 
+	/// This follows redirects until the redirect chain is 10 hops; see [`reqwest::redirect`] for more info.
+	/// 
+	/// # Examples
+	/// 
+	/// ```
+	/// # tokio_test::block_on(async {
+	/// use acrate_hostmeta::ResourceDescriptor;
+	/// 
+	/// let client = reqwest::Client::new();
+	/// let url: reqwest::Url = "https://junimo.party/.well-known/host-meta".parse()
+	/// 	.expect("URL to be valid");
+	/// 
+	/// let rd = ResourceDescriptor::discover(&client, url)
+	/// 	.await
+	/// 	.expect("resource descriptor to be discovered correctly");
+	/// # })
+	/// ```
+	/// 
+	pub async fn discover(client: &reqwest::Client, mut url: reqwest::Url) -> Result<Self, ResourceDescriptorDiscoveryError> {
+		use ResourceDescriptorDiscoveryError::*;
+
+		log::debug!("Discovering resource descriptor document at: {url}");
+
+		log::trace!("Unsetting URL query...");
+		url.set_query(None);
+
+		log::trace!("Unsetting URL fragment...");
+		url.set_fragment(None);
+
+		log::trace!("Setting URL scheme to HTTPS...");
+		url.set_scheme("https")
+			.map_err(UrlManipulation)?;
+
+		log::trace!("Cloning URL for HTTPS XRD retrieval...");
+		let https_xrd_url = url.clone();
+		
+		log::trace!("Attempting HTTPS XRD retrieval...");
+		let https_xrd = match Self::get_xrd(client, https_xrd_url).await {
+			Ok(data) => {
+				log::trace!("HTTPS XRD retrieval was successful, returning...");
+				return Ok(data)
+			}
+			Err(err) => {
+				log::warn!("HTTPS XRD retrieval failed.");
+				err
+			}
+		};
+
+		log::trace!("Cloning URL for HTTPS JRD retrieval...");
+		let https_jrd_url = url.clone();
+
+		log::trace!("Attempting HTTPS JRD retrieval...");
+		let https_jrd = match Self::get_jrd(client, https_jrd_url).await {
+			Ok(data) => {
+				log::trace!("HTTPS JRD retrieval was successful, returning...");
+				return Ok(data)
+			}
+			Err(err) => {
+				log::warn!("HTTPS JRD retrieval failed.");
+				err
+			}
+		};
+
+		log::trace!("Cloning URL for HTTPS JRD .json retrieval...");
+		let mut https_jrdj_url = url.clone();
+
+		log::trace!("Altering URL path for HTTPS JRD .json retrieval...");
+		https_jrdj_url.set_path(
+			&format!("{}.json", https_jrdj_url.path())
+		);
+
+		log::trace!("Attempting HTTPS JRD .json retrieval...");
+		let https_jrdj = match Self::get_jrd(client, https_jrdj_url).await {
+			Ok(data) => {
+				log::trace!("HTTPS JRD .json retrieval was successful, returning...");
+				return Ok(data)
+			}
+			Err(err) => {
+				log::warn!("HTTPS JRD .json retrieval failed.");
+				err
+			}
+		};
+
+		log::trace!("Setting URL scheme to HTTP...");
+		url.set_scheme("http")
+			.map_err(UrlManipulation)?;
+
+		log::trace!("Cloning URL for HTTP XRD retrieval...");
+		let http_xrd_url = url.clone();
+		
+		log::trace!("Attempting HTTP XRD retrieval...");
+		let http_xrd = match Self::get_xrd(client, http_xrd_url).await {
+			Ok(data) => {
+				log::trace!("HTTP XRD retrieval was successful, returning...");
+				return Ok(data)
+			}
+			Err(err) => {
+				log::warn!("HTTP XRD retrieval failed.");
+				err
+			}
+		};
+
+		log::trace!("Cloning URL for HTTP JRD retrieval...");
+		let http_jrd_url = url.clone();
+
+		log::trace!("Attempting HTTP JRD retrieval...");
+		let http_jrd = match Self::get_jrd(client, http_jrd_url).await {
+			Ok(data) => {
+				log::trace!("HTTP JRD retrieval was successful, returning...");
+				return Ok(data)
+			}
+			Err(err) => {
+				log::warn!("HTTP JRD retrieval failed.");
+				err
+			}
+		};
+
+		log::trace!("Cloning URL for HTTP JRD .json retrieval...");
+		let mut http_jrdj_url = url.clone();
+
+		log::trace!("Altering URL path for HTTPS JRD .json retrieval...");
+		http_jrdj_url.set_path(
+			&format!("{}.json", http_jrdj_url.path())
+		);
+
+		log::trace!("Attempting HTTP JRD .json retrieval...");
+		let http_jrdj = match Self::get_jrd(client, http_jrdj_url).await {
+			Ok(data) => {
+				log::trace!("HTTP JRD .json retrieval was successful, returning...");
+				return Ok(data)
+			}
+			Err(err) => {
+				log::warn!("HTTP JRD .json retrieval failed.");
+				err
+			}
+		};
+
+		Err(
+			ResourceDescriptorDiscoveryError::Fetch(
+				ResourceDescriptorDiscoveryFailures { 
+					https_xrd,
+					https_jrd,
+					https_jrdj,
+					http_xrd,
+					http_jrd,
+					http_jrdj,
+				}
+			)
+		)
+	}
+
+	/// Well-known path for host-meta documents.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://datatracker.ietf.org/doc/html/rfc6415#section-2>
+	/// 
+	pub const WELLKNOWN_HOSTMETA_PATH: &str = "/.well-known/host-meta";
+
+	/// Attempt to discover a host-meta document at the given base URL.
+	/// 
+	/// # Examples
+	/// 
+	/// ```
+	/// # tokio_test::block_on(async {
+	/// use acrate_hostmeta::ResourceDescriptor;
+	/// 
+	/// let client = reqwest::Client::new();
+	/// let base: reqwest::Url = "https://junimo.party".parse()
+	/// 	.expect("URL to be valid");
+	/// 
+	/// let rd = ResourceDescriptor::discover_hostmeta(&client, base)
+	/// 	.await
+	/// 	.expect("host-meta to be discovered correctly");
+	/// })
+	/// ```
+	/// 
+	pub async fn discover_hostmeta(client: &reqwest::Client, mut base: reqwest::Url) -> Result<Self, ResourceDescriptorDiscoveryError> {
+		base.set_path(Self::WELLKNOWN_HOSTMETA_PATH);
+
+		Self::discover(client, base)
+			.await
+	}
+}
+
+/// Error occurred during [`ResourceDescriptor::discover`].
+#[derive(Debug)]
+pub enum ResourceDescriptorDiscoveryError {
+	/// Manipulation of the provided base [`reqwest::Url`] failed.
+	/// 
+	/// See [reqwest::Url::set_scheme] for possible causes.
+	UrlManipulation(()),
+
+	/// All attempts of fetching a resource descriptor document failed.
+	Fetch(ResourceDescriptorDiscoveryFailures),
+}
+
+/// Request errors occurred during [`ResourceDescriptor::discover`].
+#[derive(Debug)]
+pub struct ResourceDescriptorDiscoveryFailures {
+	/// HTTPS XRD retrieval.
+	pub https_xrd: GetXRDError,
+
+	/// HTTPS JRD retrieval.
+	pub https_jrd: GetJRDError,
+
+	/// HTTPS JRD with .json extension retrieval.
+	pub https_jrdj: GetJRDError,
+
+	/// HTTPS XRD retrieval.
+	pub http_xrd: GetXRDError,
+
+	/// HTTP JRD retrieval.
+	pub http_jrd: GetJRDError,
+
+	/// HTTP JRD with .json extension retrieval.
+	pub http_jrdj: GetJRDError,
+}
+
+/// Error occurred during [`ResourceDescriptor::get_xrd`].
+#[derive(Debug)]
+pub enum GetXRDError {
+	/// The HTTP request failed.
+	Request(reqwest::Error),
+	/// The `Content-Type` header of the response is missing.
+	ContentTypeMissing,
+	/// The `Content-Type` header of the response is invalid.
+	ContentTypeInvalid,
+	/// The document failed to be read as text.
+	Decode(reqwest::Error),
+	/// The document failed to be parsed as XML by [`quick_xml`].
+	Parse(quick_xml::DeError),
+}
+
+/// Error occurred during [`ResourceDescriptor::get_jrd`].
+#[derive(Debug)]
+pub enum GetJRDError {
+	/// The HTTP request failed.
+	Request(reqwest::Error),
+	/// The `Content-Type` header of the response is missing.
+	ContentTypeMissing,
+	/// The `Content-Type` header of the response is invalid.
+	ContentTypeInvalid,
+	/// The document failed to be parsed as JSON by [`reqwest`].
+	Parse(reqwest::Error),
+}
+
+/// Extract the MIME type from the value of the `Content-Type` header.
+fn extract_mime_from_content_type(value: &reqwest::header::HeaderValue) -> Option<String> {
+	let value = value.to_str().ok()?;
+	match value.split_once("; ") {
+		None => Some(value.to_string()),
+		Some((mime, _)) => Some(mime.to_string()),
+	}
+}

acrate-hostmeta/tests/hostmeta_tests.rs

@@ -0,0 +1,62 @@
+const CARGO_PKG_NAME: &str = env!("CARGO_PKG_NAME");
+const CARGO_PKG_VERSION: &str = env!("CARGO_PKG_VERSION");
+const CARGO_PKG_REPOSITORY: &str = env!("CARGO_PKG_REPOSITORY");
+
+
+fn init_log() {
+	let mut builder = pretty_env_logger::formatted_builder();
+	builder.target(pretty_env_logger::env_logger::Target::Stdout);
+	builder.filter_level(log::LevelFilter::max());
+	builder.is_test(true);
+
+	if builder.try_init().is_ok() {
+		log::debug!("Initialized logging!");
+	}
+}
+
+fn make_client() -> reqwest::Client {
+	let user_agent = format!("{CARGO_PKG_NAME}/{CARGO_PKG_VERSION} ({CARGO_PKG_REPOSITORY})");
+	
+	reqwest::Client::builder()
+		.user_agent(user_agent)
+		.build()
+		.expect("reqwest client to build")
+}
+
+macro_rules! test_discover_hostmeta {
+	($id:ident, $url:literal) => {
+		test_discover_hostmeta!($id, $url,);
+	};
+	($id:ident, $url:literal, $($tag:meta),*) => {
+		mod $id {
+			use acrate_hostmeta::*;
+			use super::*;
+
+			#[tokio::test]
+            $(#[$tag])*
+			async fn test() {
+				init_log();
+				let client = make_client();
+
+				let base: reqwest::Url = $url.parse()
+					.expect("a valid URL");
+
+				let doc = ResourceDescriptor::discover_hostmeta(&client, base)
+					.await
+					.expect("host-meta discovery to succeed");
+
+				log::info!("Parsed host-meta document: {doc:#?}");
+			}
+		}
+	};
+}
+
+test_discover_hostmeta!(akkoma, "https://junimo.party");
+test_discover_hostmeta!(mastodon, "https://mastodon.social");
+test_discover_hostmeta!(misskey, "https://misskey.io");
+test_discover_hostmeta!(iceshrimpnet, "https://ice.frieren.quest");
+test_discover_hostmeta!(gotosocial, "https://alpha.polymaths.social");
+test_discover_hostmeta!(bridgyfed, "https://fed.brid.gy");
+test_discover_hostmeta!(threads, "https://threads.net", ignore = "does not support host-meta");
+test_discover_hostmeta!(snac, "https://ngoa.giao.loan", ignore = "does not support host-meta");
+test_discover_hostmeta!(hollo, "https://hollo.social", ignore = "does not support host-meta");

acrate-nodeinfo/Cargo.toml

@@ -4,3 +4,16 @@ version = "0.1.0"
 edition = "2021"
 
 [dependencies]
+acrate-hostmeta = { version = "0.1.0", path = "../acrate-hostmeta" }
+log = "0.4.22"
+reqwest = { version = "0.12.9", features = ["json", "stream"] }
+serde = { version = "1.0.214", features = ["derive"] }
+serde_json = "1.0.132"
+
+[dev-dependencies]
+pretty_env_logger = "0.5.0"
+tokio = { version = "1.41.1", features = ["macros", "rt-multi-thread"] }
+tokio-test = "0.4.4"
+
+[lints.clippy]
+tabs-in-doc-comments = "allow"

acrate-nodeinfo/src/lib.rs

@@ -1,14 +1,469 @@
-pub fn add(left: u64, right: u64) -> u64 {
+//! Serde-based NodeInfo fetcher and loose parser.
-    left + right
+//! 
+//! > NodeInfo is an effort to create a standardized way of exposing metadata about a server running one of the distributed social networks.
+//! 
+//! # Specification
+//! 
+//! - <https://github.com/jhass/nodeinfo/blob/main/PROTOCOL.md>
+
+use serde::Deserialize;
+
+/// A variant of a NodeInfo document.
+/// 
+/// # Specification
+/// 
+/// - <https://github.com/jhass/nodeinfo/blob/main/PROTOCOL.md>
+#[derive(Debug, Clone)]
+pub enum NodeInfo {
+	V1(NodeInfo1),
+	V2(NodeInfo2),
 }
 
-#[cfg(test)]
+/// A NodeInfo document at version 1.X.
-mod tests {
+/// 
-    use super::*;
+/// # Specification
+/// 
+/// - <https://github.com/jhass/nodeinfo/blob/main/schemas/1.0/schema.json>
+/// - <https://github.com/jhass/nodeinfo/blob/main/schemas/1.1/schema.json>
+/// 
+#[derive(Debug, Clone, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct NodeInfo1 {
+	/// The schema version.
+	pub version: String,
 
-    #[test]
+	/// Metadata about server software in use.
-    fn it_works() {
+	pub software: NodeInfo1Software,
-        let result = add(2, 2);
+
-        assert_eq!(result, 4);
+	/// The protocols supported on this server.
+	pub protocols: NodeInfo1Protocols,
+
+	/// The third party sites this server can connect to via their application API.
+	pub services: NodeInfo1Services,
+	
+	/// Whether this server allows open self-registration.
+	pub open_registrations: bool,
+
+	/// Usage statistics for this server.
+	pub usage: NodeInfo1Usage,
+
+	/// Free form key value pairs for software specific values. 
+	/// 
+	/// Clients should not rely on any specific key present.
+	pub metadata: serde_json::Value,
+}
+
+/// A NodeInfo document at version 2.X.
+/// 
+/// # Specification
+/// 
+/// - <https://github.com/jhass/nodeinfo/blob/main/schemas/2.0/schema.json>
+/// - <https://github.com/jhass/nodeinfo/blob/main/schemas/2.1/schema.json>
+/// - <https://github.com/jhass/nodeinfo/blob/main/schemas/2.2/schema.json>
+/// 
+#[derive(Debug, Clone, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct NodeInfo2 {
+	/// The schema version.
+	pub version: String,
+
+	/// Metadata specific to the instance. An instance is a the concrete installation of a software running on a server.
+	pub instance: Option<NodeInfo2Instance>,
+
+	/// Metadata about server software in use.
+	pub software: NodeInfo1Software,
+
+	/// The protocols supported on this server.
+	pub protocols: Vec<String>,
+
+	/// The third party sites this server can connect to via their application API.
+	pub services: Option<NodeInfo1Services>,
+
+	/// Whether this server allows open self-registration.
+	pub open_registrations: bool,
+
+	/// Usage statistics for this server.
+	pub usage: Option<NodeInfo1Usage>,
+
+	/// Free form key value pairs for software specific values. 
+	/// 
+	/// Clients should not rely on any specific key present.
+	pub metadata: serde_json::Value,
+}
+
+#[derive(Debug, Clone, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct NodeInfo1Software {	
+	/// The canonical name of this server software.
+	pub name: String,
+
+	/// The version of this server software.
+	pub version: String,
+
+	pub repository: Option<String>,
+
+	pub homepage: Option<String>,
+}
+
+#[derive(Debug, Clone, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct NodeInfo1Protocols {
+	/// The protocols this server can receive traffic for.
+	pub inbound: Vec<String>,
+
+	/// The protocols this server can generate traffic for.
+	pub outbound: Vec<String>,
+}
+
+#[derive(Debug, Clone, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct NodeInfo1Services {
+	/// The third party sites this server can retrieve messages from for combined display with regular traffic.
+	pub inbound: Vec<String>,
+	
+	/// The third party sites this server can publish messages to on the behalf of a user.
+	pub outbound: Vec<String>,
+}
+
+#[derive(Debug, Clone, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct NodeInfo1Usage {
+	/// Statistics about the users of this server.
+	pub users: Option<NodeInfo1UsageUsers>,
+
+	/// The amount of posts that were made by users that are registered on this server.
+	pub local_posts: Option<i32>,
+
+	/// The amount of comments that were made by users that are registered on this server.
+	pub local_comments: Option<i32>,
+}
+
+#[derive(Debug, Clone, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct NodeInfo1UsageUsers {
+	/// The total amount of on this server registered users.
+	pub total: Option<i32>,
+
+	/// The amount of users that signed in at least once in the last 180 days.
+	pub active_halfyear: Option<i32>,
+
+	/// The amount of users that signed in at least once in the last 30 days.
+	pub active_month: Option<i32>,
+
+	/// The amount of users that signed in at least once in the last 7 days.
+	pub active_week: Option<i32>,
+}
+
+#[derive(Debug, Clone, Deserialize)]
+#[serde(rename_all = "camelCase")]
+pub struct NodeInfo2Instance {
+	/// If supported by the software, the administrator-configured name of this instance.
+	pub name: String,
+
+	/// If supported by the software, the administrator-configured long form description of this instance.
+	pub description: String,
+}
+
+impl NodeInfo {
+	/// Well-known path for NodeInfo documents.
+	/// 
+	/// # Specification
+	/// 
+	/// - <https://github.com/jhass/nodeinfo/blob/main/PROTOCOL.md#discovery>
+	pub const WELLKNOWN_NODEINFO_PATH: &str = "/.well-known/nodeinfo";
+
+	/// Discover and get the latest NodeInfo version available given a certain base URL.
+	/// 
+	/// # Examples
+	/// 
+	/// ```
+	/// # tokio_test::block_on(async {
+	/// use acrate_nodeinfo::NodeInfo;
+	/// 
+	/// let client = reqwest::Client::new();
+	/// let url: reqwest::Url = "https://mastodon.online".parse()
+	/// 	.expect("URL to be valid");
+	/// 
+	/// let ni = NodeInfo::get_latest_wellknown(&client, url)
+	/// 	.await
+	/// 	.expect("NodeInfo to be processed correctly");
+	/// 
+	/// let version = match ni {
+	/// 	NodeInfo::V1(ni) => ni.version,
+	/// 	NodeInfo::V2(ni) => ni.version,
+	/// };
+	/// 
+	/// assert_eq!(version, "2.0");
+	/// # })
+	pub async fn get_latest_wellknown(client: &reqwest::Client, mut base: reqwest::Url) -> Result<Self, NodeInfoGetWellknownError> {
+		use NodeInfoGetWellknownError::*;
+
+		log::debug!("Getting well-known NodeInfo document at base: {base}");
+
+		log::trace!("Setting URL path to the well-known NodeInfo value...");
+		base.set_path(Self::WELLKNOWN_NODEINFO_PATH);
+
+		log::trace!("Discovering NodeInfo document locations...");
+		let discovery = acrate_hostmeta::ResourceDescriptor::discover(client, base)
+			.await
+			.map_err(Discovery)?;
+
+		log::trace!("Getting a list of NodeInfo document links...");
+		let mut links = discovery.links.unwrap_or_default();
+		links.sort_unstable_by_key(|o| o.rel.clone());  // TODO: Performance can be improved.
+		links.reverse();
+
+		for link in links.into_iter() {
+
+			log::trace!("Checking discovered link href...");
+			let url = match link.href {
+				None => {
+					log::warn!("Discovered link does not have an href, skipping...");
+					continue
+				},
+				Some(href) => {
+					log::trace!("Discovered link has an href, processing...");
+					href
+				},
+			};
+
+			log::trace!("Parsing discovered link href...");
+			let url: reqwest::Url = match url.parse() {
+				Err(e) => {
+					log::warn!("Discovered link has an invalid URL as href, skipping: {e:#?}");
+					continue
+				},
+				Ok(url) => {
+					log::trace!("Discovered link has a valid URL, processing...");
+					url
+				},
+			};
+
+			let rel = link.rel;
+
+			let nodeinfo = match rel.as_str() {
+				"http://nodeinfo.diaspora.software/ns/schema/1.0" => match NodeInfo1::get(client, url).await {
+					Err(e) => {
+						log::warn!("Failed to get NodeInfo v1.0 document, skipping: {e:#?}");
+						continue;
+					},
+					Ok(nodeinfo) => {
+						log::trace!("Successfully processed NodeInfo v1.0 document!");
+						Self::V1(nodeinfo)
+					}
+				}
+				"http://nodeinfo.diaspora.software/ns/schema/1.1" => match NodeInfo1::get(client, url).await {
+					Err(e) => {
+						log::warn!("Failed to get NodeInfo v1.1 document, skipping: {e:#?}");
+						continue;
+					},
+					Ok(nodeinfo) => {
+						log::trace!("Successfully processed NodeInfo v1.1 document!");
+						Self::V1(nodeinfo)
+					}
+				}
+				"http://nodeinfo.diaspora.software/ns/schema/2.0" => match NodeInfo2::get(client, url).await {
+					Err(e) => {
+						log::warn!("Failed to get NodeInfo v2.0 document, skipping: {e:#?}");
+						continue;
+					},
+					Ok(nodeinfo) => {
+						log::trace!("Successfully processed NodeInfo v2.0 document!");
+						Self::V2(nodeinfo)
+					}
+				}
+				"http://nodeinfo.diaspora.software/ns/schema/2.1" => match NodeInfo2::get(client, url).await {
+					Err(e) => {
+						log::warn!("Failed to get NodeInfo v2.1 document, skipping: {e:#?}");
+						continue;
+					},
+					Ok(nodeinfo) => {
+						log::trace!("Successfully processed NodeInfo v2.1 document!");
+						Self::V2(nodeinfo)
+					}
+				}
+				"http://nodeinfo.diaspora.software/ns/schema/2.2" => match NodeInfo2::get(client, url).await {
+					Err(e) => {
+						log::warn!("Failed to get NodeInfo v2.2 document, skipping: {e:#?}");
+						continue;
+					},
+					Ok(nodeinfo) => {
+						log::trace!("Successfully processed NodeInfo v2.2 document!");
+						Self::V2(nodeinfo)
+					}
+				}
+				_ => {
+					log::warn!("Discovered link has unknown rel `{rel}`, skipping.");
+					continue;
+				},
+			};
+
+			log::trace!("Successfully retrieved latest NodeInfo: {nodeinfo:#?}");
+			return Ok(nodeinfo);
+		}
+
+		log::warn!("Ran out of possible NodeInfo sources, returning an Unsupported error.");
+		Err(Unsupported)
+	}
+}
+
+/// An error occurred during [`NodeInfo::get_latest_wellknown`].
+#[derive(Debug)]
+pub enum NodeInfoGetWellknownError {
+	/// The discovery of possible locations for NodeInfo documents failed.
+	Discovery(acrate_hostmeta::ResourceDescriptorDiscoveryError),
+	/// No compatible NodeInfo documents were detected at the given URL.
+	Unsupported,
+}
+
+impl NodeInfo1 {
+	/// Get a NodeInfo v1.X document.
+	pub async fn get(client: &reqwest::Client, url: reqwest::Url) -> Result<Self, NodeInfoGetError> {
+		use NodeInfoGetError::*;
+
+		log::debug!("Getting NodeInfo v1.X document at: {url}");
+
+		log::trace!("Building request...");
+		let request = {
+			log::trace!("Creating new request...");
+			let mut request = reqwest::Request::new(reqwest::Method::GET, url);
+
+			log::trace!("Setting request headers...");
+			let headers = request.headers_mut();
+
+			log::trace!("Setting `Accept: application/json`...");
+			let _ = headers.insert(reqwest::header::ACCEPT, "application/json".parse().unwrap());
+
+			request
+		};
+
+		log::trace!("Sending request...");
+		let response = client.execute(request)
+			.await
+			.map_err(Request)?;
+
+		log::trace!("Checking `Content-Type` of the response...");
+		let content_type = response
+			.headers()
+			.get(reqwest::header::CONTENT_TYPE)
+			.ok_or(ContentTypeMissing)?;
+
+		log::trace!("Extracting MIME type from the `Content-Type` header...");
+		let mime_type = extract_mime_from_content_type(content_type)
+			.ok_or(ContentTypeInvalid)?;
+
+		log::trace!("Ensuring MIME type is acceptable for NodeInfo documents...");
+		if mime_type != "application/json" {
+			log::error!("MIME type `{mime_type}` is not acceptable for NodeInfo documents.");
+			return Err(ContentTypeInvalid)
+		}
+
+		log::trace!("Attempting to parse response as JSON...");
+		let data = response.json::<Self>()
+			.await
+			.map_err(Parse)?;
+
+		log::trace!("Making sure version is compatible with 1.X...");
+		if !data.version.starts_with("1.") {
+			return Err(Version);
+		}
+
+		Ok(data)
+	}
+}
+
+impl NodeInfo2 {
+	/// Get a NodeInfo v2.X document.
+	/// 
+	/// # Examples
+	/// 
+	/// ```
+	/// # tokio_test::block_on(async {
+	/// use acrate_nodeinfo::NodeInfo2;
+	/// 
+	/// let client = reqwest::Client::new();
+	/// let url: reqwest::Url = "https://junimo.party/nodeinfo/2.1.json".parse()
+	/// 	.expect("URL to be valid");
+	/// 
+	/// let rd = NodeInfo2::get(&client, url)
+	/// 	.await
+	/// 	.expect("NodeInfo to be obtained correctly");
+	/// # })
+	/// ```
+	pub async fn get(client: &reqwest::Client, url: reqwest::Url) -> Result<Self, NodeInfoGetError> {
+		use NodeInfoGetError::*;
+
+		log::debug!("Getting NodeInfo v2.X document at: {url}");
+
+		log::trace!("Building request...");
+		let request = {
+			log::trace!("Creating new request...");
+			let mut request = reqwest::Request::new(reqwest::Method::GET, url);
+
+			log::trace!("Setting request headers...");
+			let headers = request.headers_mut();
+
+			log::trace!("Setting `Accept: application/json`...");
+			let _ = headers.insert(reqwest::header::ACCEPT, "application/json".parse().unwrap());
+
+			request
+		};
+
+		log::trace!("Sending request...");
+		let response = client.execute(request)
+			.await
+			.map_err(Request)?;
+
+		log::trace!("Checking `Content-Type` of the response...");
+		let content_type = response
+			.headers()
+			.get(reqwest::header::CONTENT_TYPE)
+			.ok_or(ContentTypeMissing)?;
+
+		log::trace!("Extracting MIME type from the `Content-Type` header...");
+		let mime_type = extract_mime_from_content_type(content_type)
+			.ok_or(ContentTypeInvalid)?;
+
+		log::trace!("Ensuring MIME type is acceptable for NodeInfo documents...");
+		if mime_type != "application/json" {
+			log::error!("MIME type `{mime_type}` is not acceptable for NodeInfo documents.");
+			return Err(ContentTypeInvalid)
+		}
+
+		log::trace!("Attempting to parse response as JSON...");
+		let data = response.json::<Self>()
+			.await
+			.map_err(Parse)?;
+
+		log::trace!("Making sure version is compatible with 2.X...");
+		if !data.version.starts_with("2.") {
+			return Err(Version)
+		}
+
+		Ok(data)
+	}
+}
+
+/// An error encountered during [`NodeInfo1::get`] or [`NodeInfo2::get`].
+#[derive(Debug)]
+pub enum NodeInfoGetError {
+	/// The HTTP request failed.
+	Request(reqwest::Error),
+	/// The `Content-Type` header of the response is missing.
+	ContentTypeMissing,
+	/// The `Content-Type` header of the response is invalid.
+	ContentTypeInvalid,
+	/// The document failed to be parsed as JSON by [`reqwest`].
+	Parse(reqwest::Error),
+	/// The returned version does not match the version of the created struct.
+	Version,
+}
+
+/// Extract the MIME type from the value of the `Content-Type` header.
+fn extract_mime_from_content_type(value: &reqwest::header::HeaderValue) -> Option<String> {
+	let value = value.to_str().ok()?;
+	match value.split_once("; ") {
+		None => Some(value.to_string()),
+		Some((mime, _)) => Some(mime.to_string()),
 	}
 }

acrate-nodeinfo/tests/nodeinfo_tests.rs

@@ -0,0 +1,70 @@
+const CARGO_PKG_NAME: &str = env!("CARGO_PKG_NAME");
+const CARGO_PKG_VERSION: &str = env!("CARGO_PKG_VERSION");
+const CARGO_PKG_REPOSITORY: &str = env!("CARGO_PKG_REPOSITORY");
+
+
+fn init_log() {
+	let mut builder = pretty_env_logger::formatted_builder();
+	builder.target(pretty_env_logger::env_logger::Target::Stdout);
+	builder.filter_level(log::LevelFilter::max());
+	builder.is_test(true);
+
+	if builder.try_init().is_ok() {
+		log::debug!("Initialized logging!");
+	}
+}
+
+fn make_client() -> reqwest::Client {
+	let user_agent = format!("{CARGO_PKG_NAME}/{CARGO_PKG_VERSION} ({CARGO_PKG_REPOSITORY})");
+	
+	reqwest::Client::builder()
+		.user_agent(user_agent)
+		.build()
+		.expect("reqwest client to build")
+}
+
+
+macro_rules! test {
+	($id:ident, $url:literal, $version:literal) => {
+		test!($id, $url, $version,);
+	};
+	($id:ident, $url:literal, $version:literal, $($tag:meta),*) => {
+		mod $id {
+			use acrate_nodeinfo::*;
+			use super::*;
+
+			#[tokio::test]
+            $(#[$tag])*
+			async fn test_version() {
+				init_log();
+				let client = make_client();
+
+				let base: reqwest::Url = $url.parse()
+					.expect("a valid URL");
+
+				let doc = NodeInfo::get_latest_wellknown(&client, base)
+					.await
+					.expect("NodeInfo discovery to succeed");
+
+				log::info!("Parsed NodeInfo document: {doc:#?}");
+
+				let version = match doc {
+					NodeInfo::V1(d) => d.version,
+					NodeInfo::V2(d) => d.version,
+				};
+
+				assert_eq!(version, $version);
+			}
+		}
+	};
+}
+
+test!(akkoma, "https://junimo.party", "2.1");
+test!(mastodon, "https://mastodon.social", "2.0");
+test!(misskey, "https://misskey.io", "2.1");
+test!(iceshrimpnet, "https://ice.frieren.quest", "2.1");
+test!(gotosocial, "https://alpha.polymaths.social", "2.0");
+test!(bridgyfed, "https://fed.brid.gy", "2.1");
+test!(threads, "https://threads.net", "", ignore = "does not support NodeInfo");
+test!(snac, "https://ngoa.giao.loan", "2.0");
+test!(hollo, "https://hollo.social", "2.1");