1
Fork 0

hostmeta: Move things to different submodules

This commit is contained in:
Steffo 2024-11-14 01:39:33 +01:00
parent d2b765a5b4
commit 428bed8a4a
Signed by: steffo
GPG key ID: 5ADA3868646C3FC0
4 changed files with 564 additions and 0 deletions

View file

@ -0,0 +1,475 @@
use serde::Deserialize;
use thiserror::Error;
use crate::link::ResourceDescriptorLink;
use crate::property::ResourceDescriptorProperty;
/// 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>>,
}
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 = crate::utils::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 = crate::utils::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, Error)]
pub enum ResourceDescriptorDiscoveryError {
/// Manipulation of the provided base [`reqwest::Url`] failed.
///
/// See [reqwest::Url::set_scheme] for possible causes.
#[error("manipulation of the provided URL failed")]
UrlManipulation(()),
/// All attempts of fetching a resource descriptor document failed.
#[error("fetchign the resource descriptor document failed")]
Fetch(ResourceDescriptorDiscoveryFailures),
}
/// Request errors occurred during [`ResourceDescriptor::discover`].
#[derive(Debug, Error)]
#[error("all attempts of fetching the resource descriptor document failed")]
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, Error)]
pub enum GetXRDError {
/// The HTTP request failed.
#[error("the HTTP request failed")]
Request(reqwest::Error),
/// The `Content-Type` header of the response is missing.
#[error("the Content-Type header of the response is missing")]
ContentTypeMissing,
/// The `Content-Type` header of the response is invalid.
#[error("the Content-Type header of the response is invalid")]
ContentTypeInvalid,
/// The document failed to be decoded as text.
#[error("the document failed to be decoded as text")]
Decode(reqwest::Error),
/// The document failed to be parsed as XML by [`quick_xml`].
#[error("the document failed to be parsed as XML")]
Parse(quick_xml::DeError),
}
/// Error occurred during [`ResourceDescriptor::get_jrd`].
#[derive(Debug, Error)]
pub enum GetJRDError {
/// The HTTP request failed.
#[error("the HTTP request failed")]
Request(reqwest::Error),
/// The `Content-Type` header of the response is missing.
#[error("the Content-Type header of the response is missing")]
ContentTypeMissing,
/// The `Content-Type` header of the response is invalid.
#[error("the Content-Type header of the response is invalid")]
ContentTypeInvalid,
/// The document failed to be parsed as JSON by [`reqwest`].
#[error("the document failed to be parsed as JSON")]
Parse(reqwest::Error),
}

View file

@ -0,0 +1,64 @@
use serde::Deserialize;
use std::collections::HashMap;
use crate::property::ResourceDescriptorProperty;
/// 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>,
}

View file

@ -0,0 +1,17 @@
use serde::Deserialize;
/// 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>,
}

View file

@ -0,0 +1,8 @@
/// Extract the MIME type from the value of the `Content-Type` header.
pub 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()),
}
}