14 files changed, 419 insertions, 22 deletions

diff --git a/Cargo.lock b/Cargo.lock
index a7cf8b56..10ad2b2b 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -425,6 +425,7 @@ dependencies = [
  "garage_rpc 0.1.0",
  "garage_table 0.1.1",
  "garage_util 0.1.0",
+ "garage_web",
  "hex",
  "log",
  "pretty_env_logger",
@@ -666,6 +667,37 @@ dependencies = [
 ]
 
 [[package]]
+name = "garage_web"
+version = "0.1.0"
+dependencies = [
+ "err-derive",
+ "futures",
+ "futures-util",
+ "garage_api",
+ "garage_model 0.1.1",
+ "garage_table 0.1.1",
+ "garage_util 0.1.0",
+ "hex",
+ "http",
+ "httpdate",
+ "hyper",
+ "idna",
+ "log",
+ "percent-encoding",
+ "rand",
+ "rmp-serde",
+ "roxmltree",
+ "rustls",
+ "serde",
+ "serde_json",
+ "sha2",
+ "sled",
+ "tokio",
+ "toml",
+ "webpki",
+]
+
+[[package]]
 name = "generator"
 version = "0.6.21"
 source = "registry+https://github.com/rust-lang/crates.io-index"
diff --git a/Cargo.toml b/Cargo.toml
index 7a8c74e9..739e698e 100644
--- a/Cargo.toml
+++ b/Cargo.toml
@@ -5,6 +5,7 @@ members = [
 	"src/table",
 	"src/model",
 	"src/api",
+	"src/web",
 	"src/garage",
 ]
 
diff --git a/README.md b/README.md
index 10dcc0cd..a33e2850 100644
--- a/README.md
+++ b/README.md
@@ -20,14 +20,15 @@ Our main use case is to provide a distributed storage layer for small-scale self
 
 We propose the following quickstart to setup a full dev. environment as quickly as possible:
 
-  1. Setup a rust/cargo environment and install s3cmd. eg. `dnf install rust cargo s3cmd`
-  2. Run `cargo build` to build the project
-  3. Run `./script/dev-cluster.sh` to launch a test cluster (feel free to read the script)
-  4. Run `./script/dev-configure.sh` to configure your test cluster with default values (same datacenter, 100 tokens)
-  5. Run `./script/dev-bucket.sh` to create a bucket named `éprouvette` and an API key that will be stored in `/tmp/garage.s3`
-  6. Run `source ./script/dev-env.sh` to configure your CLI environment
-  7. You can use `garage` to manage the cluster. Try `garage --help`.
-  8. You can use `s3grg` to add, remove, and delete files. Try `s3grg --help`, `s3grg put /proc/cpuinfo s3://éprouvette/cpuinfo.txt`, `s3grg ls s3://éprouvette`. `s3grg` is a wrapper on `s3cmd` configured with the previously generated API key (the one in `/tmp/garage.s3`).
+  1. Setup a rust/cargo environment. eg. `dnf install rust cargo`
+  2. Install awscli v2 by following the guide [here](https://docs.aws.amazon.com/cli/latest/userguide/install-cliv2.html).
+  3. Run `cargo build` to build the project
+  4. Run `./script/dev-cluster.sh` to launch a test cluster (feel free to read the script)
+  5. Run `./script/dev-configure.sh` to configure your test cluster with default values (same datacenter, 100 tokens)
+  6. Run `./script/dev-bucket.sh` to create a bucket named `eprouvette` and an API key that will be stored in `/tmp/garage.s3`
+  7. Run `source ./script/dev-env.sh` to configure your CLI environment
+  8. You can use `garage` to manage the cluster. Try `garage --help`.
+  9. You can use `s3grg` to add, remove, and delete files. Try `s3grg --help`, `s3grg cp /proc/cpuinfo s3://eprouvette/cpuinfo.txt`, `s3grg ls s3://eprouvette`. `s3grg` is a wrapper on the `aws s3` subcommand configured with the previously generated API key (the one in `/tmp/garage.s3`).
 
 Now you should be ready to start hacking on garage!
 
@@ -85,7 +86,9 @@ api_bind_addr = "[::1]:3900"	# the S3 API port, HTTP without TLS. Add a reverse
 s3_region = "garage"				# set this to anything. S3 API calls will fail if they are not made against the region set here.
 
 [s3_web]
-web_bind_addr = "[::1]:3902"
+bind_addr = "[::1]:3902"
+root_domain = ".garage.tld"
+index = "index.html"
 ```
 
 Build Garage using `cargo build --release`.
diff --git a/config.dev.toml b/config.dev.toml
index 966bee4b..215bc50c 100644
--- a/config.dev.toml
+++ b/config.dev.toml
@@ -17,4 +17,6 @@ api_bind_addr = "[::1]:3900"	# the S3 API port, HTTP without TLS. Add a reverse
 s3_region = "garage"				# set this to anything. S3 API calls will fail if they are not made against the region set here.
 
 [s3_web]
-web_bind_addr = "[::1]:3902"
+bind_addr = "[::1]:3902"
+root_domain = ".garage.tld"
+index = "index.html"
diff --git a/script/dev-cluster.sh b/script/dev-cluster.sh
index 101a18bd..64c17cfc 100755
--- a/script/dev-cluster.sh
+++ b/script/dev-cluster.sh
@@ -41,6 +41,8 @@ s3_region = "garage"				# set this to anything. S3 API calls will fail if they a
 
 [s3_web]
 bind_addr = "[::]:$((3920+$count))"
+root_domain = ".garage.tld"
+index = "index.html"
 EOF
 
 echo -en "$LABEL configuration written to $CONF_PATH\n"
diff --git a/script/dev-env.sh b/script/dev-env.sh
index 88d2941f..15f08e2e 100755
--- a/script/dev-env.sh
+++ b/script/dev-env.sh
@@ -6,14 +6,9 @@ GARAGE_DEBUG="${REPO_FOLDER}/target/debug/"
 GARAGE_RELEASE="${REPO_FOLDER}/target/release/"
 PATH="${GARAGE_DEBUG}:${GARAGE_RELEASE}:$PATH"
 
-ACCESS_KEY=`cat /tmp/garage.s3 |cut -d' ' -f1`
-SECRET_KEY=`cat /tmp/garage.s3 |cut -d' ' -f2`
-
-alias s3grg="s3cmd \
-  --host 127.0.0.1:3911 \
-  --host-bucket 127.0.0.1:3911 \
-  --access_key=$ACCESS_KEY \
-  --secret_key=$SECRET_KEY \
-  --region=garage \
-  --no-ssl"
+export AWS_ACCESS_KEY_ID=`cat /tmp/garage.s3 |cut -d' ' -f1`
+export AWS_SECRET_ACCESS_KEY=`cat /tmp/garage.s3 |cut -d' ' -f2`
+export AWS_DEFAULT_REGION='garage'
 
+alias s3grg="aws s3 \
+  --endpoint-url http://127.0.0.1:3911"
diff --git a/script/test-smoke.sh b/script/test-smoke.sh
index 7b462b00..7faf2a07 100755
--- a/script/test-smoke.sh
+++ b/script/test-smoke.sh
@@ -15,8 +15,8 @@ source ${SCRIPT_FOLDER}/dev-env.sh
 
 dd if=/dev/urandom of=/tmp/garage.rnd bs=1M count=10
 
-s3grg put /tmp/garage.rnd s3://eprouvette/
+s3grg cp /tmp/garage.rnd s3://eprouvette/
 s3grg ls s3://eprouvette
-s3grg get s3://eprouvette/garage.rnd /tmp/garage.dl
+s3grg cp s3://eprouvette/garage.rnd /tmp/garage.dl
 
 diff /tmp/garage.rnd /tmp/garage.dl
diff --git a/src/garage/Cargo.toml b/src/garage/Cargo.toml
index cb16bcd4..39288f40 100644
--- a/src/garage/Cargo.toml
+++ b/src/garage/Cargo.toml
@@ -19,6 +19,7 @@ garage_rpc = { version = "0.1", path = "../rpc" }
 garage_table = { version = "0.1.1", path = "../table" }
 garage_model = { version = "0.1.1", path = "../model" }
 garage_api = { version = "0.1.1", path = "../api" }
+garage_web = { version = "0.1", path = "../web" }
 
 bytes = "0.4"
 rand = "0.7"
diff --git a/src/garage/server.rs b/src/garage/server.rs
index 6caea5eb..ec78c067 100644
--- a/src/garage/server.rs
+++ b/src/garage/server.rs
@@ -11,6 +11,7 @@ use garage_util::error::Error;
 use garage_api::api_server;
 use garage_model::garage::Garage;
 use garage_rpc::rpc_server::RpcServer;
+use garage_web::web_server;
 
 use crate::admin_rpc::*;
 
@@ -56,6 +57,7 @@ pub async fn run_server(config_file: PathBuf) -> Result<(), Error> {
 	info!("Initializing RPC and API servers...");
 	let run_rpc_server = Arc::new(rpc_server).run(wait_from(watch_cancel.clone()));
 	let api_server = api_server::run_api_server(garage.clone(), wait_from(watch_cancel.clone()));
+	let web_server = web_server::run_web_server(garage.clone(), wait_from(watch_cancel.clone()));
 
 	futures::try_join!(
 		garage
@@ -78,6 +80,10 @@ pub async fn run_server(config_file: PathBuf) -> Result<(), Error> {
 			info!("API server exited");
 			rv
 		}),
+		web_server.map(|rv| {
+			info!("Web server exited");
+			rv
+		}),
 		background.run().map(|rv| {
 			info!("Background runner exited");
 			Ok(rv)
diff --git a/src/util/config.rs b/src/util/config.rs
index b985114d..f4c841b7 100644
--- a/src/util/config.rs
+++ b/src/util/config.rs
@@ -35,6 +35,8 @@ pub struct Config {
 	pub rpc_tls: Option<TlsConfig>,
 
 	pub s3_api: ApiConfig,
+
+	pub s3_web: WebConfig,
 }
 
 #[derive(Deserialize, Debug, Clone)]
@@ -50,6 +52,13 @@ pub struct ApiConfig {
 	pub s3_region: String,
 }
 
+#[derive(Deserialize, Debug, Clone)]
+pub struct WebConfig {
+	pub bind_addr: SocketAddr,
+	pub root_domain: String,
+	pub index: String,
+}
+
 fn default_max_concurrent_rpc_requests() -> usize {
 	12
 }
diff --git a/src/web/Cargo.toml b/src/web/Cargo.toml
new file mode 100644
index 00000000..0d08fdbf
--- /dev/null
+++ b/src/web/Cargo.toml
@@ -0,0 +1,47 @@
+[package]
+name = "garage_web"
+version = "0.1.0"
+authors = ["Alex Auvolat <alex@adnab.me>", "Quentin Dufour <quentin@dufour.io>"]
+edition = "2018"
+license = "GPL-3.0"
+description = "S3-like website endpoint crate for the Garage object store"
+repository = "https://git.deuxfleurs.fr/Deuxfleurs/garage"
+
+[lib]
+path = "lib.rs"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+
+[dependencies]
+garage_util = { version = "0.1", path = "../util" }
+garage_table = { version = "0.1.1", path = "../table" }
+garage_model = { version = "0.1.1", path = "../model" }
+garage_api = { version = "0.1.1", path = "../api" }
+
+rand = "0.7"
+hex = "0.3"
+sha2 = "0.8"
+err-derive = "0.2.3"
+log = "0.4"
+
+sled = "0.31"
+
+toml = "0.5"
+rmp-serde = "0.14.3"
+serde = { version = "1.0", default-features = false, features = ["derive", "rc"] }
+serde_json = "1.0"
+
+futures = "0.3"
+futures-util = "0.3"
+tokio = { version = "0.2", default-features = false, features = ["rt-core", "rt-threaded", "io-driver", "net", "tcp", "time", "macros", "sync", "signal", "fs"] }
+
+http = "0.2"
+hyper = "0.13"
+percent-encoding = "2.1.0"
+rustls = "0.17"
+webpki = "0.21"
+
+roxmltree = "0.11"
+idna = "0.2"
+
+httpdate = "0.3"
diff --git a/src/web/error.rs b/src/web/error.rs
new file mode 100644
index 00000000..220bacfe
--- /dev/null
+++ b/src/web/error.rs
@@ -0,0 +1,55 @@
+use err_derive::Error;
+use hyper::StatusCode;
+
+use garage_util::error::Error as GarageError;
+
+#[derive(Debug, Error)]
+pub enum Error {
+	#[error(display = "API error: {}", _0)]
+	ApiError(#[error(source)] garage_api::error::Error),
+
+	// Category: internal error
+	#[error(display = "Internal error: {}", _0)]
+	InternalError(#[error(source)] GarageError),
+
+	#[error(display = "Internal error (Hyper error): {}", _0)]
+	Hyper(#[error(source)] hyper::Error),
+
+	#[error(display = "Internal error (HTTP error): {}", _0)]
+	HTTP(#[error(source)] http::Error),
+
+	// Category: cannot process
+	#[error(display = "Forbidden: {}", _0)]
+	Forbidden(String),
+
+	#[error(display = "Not found")]
+	NotFound,
+
+	// Category: bad request
+	#[error(display = "Invalid UTF-8: {}", _0)]
+	InvalidUTF8(#[error(source)] std::str::Utf8Error),
+
+	#[error(display = "Invalid XML: {}", _0)]
+	InvalidXML(#[error(source)] roxmltree::Error),
+
+	#[error(display = "Invalid header value: {}", _0)]
+	InvalidHeader(#[error(source)] hyper::header::ToStrError),
+
+	#[error(display = "Bad request: {}", _0)]
+	BadRequest(String),
+}
+
+impl Error {
+	pub fn http_status_code(&self) -> StatusCode {
+		match self {
+			Error::NotFound => StatusCode::NOT_FOUND,
+			Error::ApiError(e) => e.http_status_code(),
+			Error::Forbidden(_) => StatusCode::FORBIDDEN,
+			Error::InternalError(GarageError::RPC(_)) => StatusCode::SERVICE_UNAVAILABLE,
+			Error::InternalError(_) | Error::Hyper(_) | Error::HTTP(_) => {
+				StatusCode::INTERNAL_SERVER_ERROR
+			}
+			_ => StatusCode::BAD_REQUEST,
+		}
+	}
+}
diff --git a/src/web/lib.rs b/src/web/lib.rs
new file mode 100644
index 00000000..f28937b9
--- /dev/null
+++ b/src/web/lib.rs
@@ -0,0 +1,6 @@
+#[macro_use]
+extern crate log;
+
+pub mod error;
+
+pub mod web_server;
diff --git a/src/web/web_server.rs b/src/web/web_server.rs
new file mode 100644
index 00000000..f8a5cd14
--- /dev/null
+++ b/src/web/web_server.rs
@@ -0,0 +1,238 @@
+use std::{borrow::Cow, convert::Infallible, net::SocketAddr, sync::Arc};
+
+use futures::future::Future;
+
+use hyper::{
+	header::HOST,
+	server::conn::AddrStream,
+	service::{make_service_fn, service_fn},
+	Body, Method, Request, Response, Server,
+};
+
+use idna::domain_to_unicode;
+
+use crate::error::*;
+use garage_api::s3_get::{handle_get, handle_head};
+use garage_model::garage::Garage;
+use garage_util::error::Error as GarageError;
+
+pub async fn run_web_server(
+	garage: Arc<Garage>,
+	shutdown_signal: impl Future<Output = ()>,
+) -> Result<(), GarageError> {
+	let addr = &garage.config.s3_web.bind_addr;
+
+	let service = make_service_fn(|conn: &AddrStream| {
+		let garage = garage.clone();
+		let client_addr = conn.remote_addr();
+		async move {
+			Ok::<_, Error>(service_fn(move |req: Request<Body>| {
+				let garage = garage.clone();
+				handle_request(garage, req, client_addr)
+			}))
+		}
+	});
+
+	let server = Server::bind(&addr).serve(service);
+	let graceful = server.with_graceful_shutdown(shutdown_signal);
+	info!("Web server listening on http://{}", addr);
+
+	graceful.await?;
+	Ok(())
+}
+
+async fn handle_request(
+	garage: Arc<Garage>,
+	req: Request<Body>,
+	addr: SocketAddr,
+) -> Result<Response<Body>, Infallible> {
+	info!("{} {} {}", addr, req.method(), req.uri());
+	let res = serve_file(garage, req).await;
+	match &res {
+		Ok(r) => debug!("{} {:?}", r.status(), r.headers()),
+		Err(e) => warn!("Response: error {}, {}", e.http_status_code(), e),
+	}
+
+	Ok(res.unwrap_or_else(error_to_res))
+}
+
+fn error_to_res(e: Error) -> Response<Body> {
+	let body: Body = Body::from(format!("{}\n", e));
+	let mut http_error = Response::new(body);
+	*http_error.status_mut() = e.http_status_code();
+	http_error
+}
+
+async fn serve_file(garage: Arc<Garage>, req: Request<Body>) -> Result<Response<Body>, Error> {
+	// Get http authority string (eg. [::1]:3902 or garage.tld:80)
+	let authority = req
+		.headers()
+		.get(HOST)
+		.ok_or(Error::BadRequest(format!("HOST header required")))?
+		.to_str()?;
+
+	// Get bucket
+	let (host, _) = domain_to_unicode(authority_to_host(authority)?);
+	let root = &garage.config.s3_web.root_domain;
+	let bucket = host_to_bucket(&host, root);
+
+	// Get path
+	let path = req.uri().path().to_string();
+	let index = &garage.config.s3_web.index;
+	let key = path_to_key(&path, &index)?;
+
+	info!("Selected bucket: \"{}\", selected key: \"{}\"", bucket, key);
+
+	let res = match req.method() {
+		&Method::HEAD => handle_head(garage, &bucket, &key).await?,
+		&Method::GET => handle_get(garage, &req, bucket, &key).await?,
+		_ => return Err(Error::BadRequest(format!("HTTP method not supported"))),
+	};
+
+	Ok(res)
+}
+
+/// Extract host from the authority section given by the HTTP host header
+///
+/// The HTTP host contains both a host and a port.
+/// Extracting the port is more complex than just finding the colon (:) symbol due to IPv6
+/// We do not use the collect pattern as there is no way in std rust to collect over a stack allocated value
+/// check here: https://docs.rs/collect_slice/1.2.0/collect_slice/
+fn authority_to_host(authority: &str) -> Result<&str, Error> {
+	let mut iter = authority.chars().enumerate();
+	let (_, first_char) = iter
+		.next()
+		.ok_or(Error::BadRequest(format!("Authority is empty")))?;
+
+	let split = match first_char {
+		'[' => {
+			let mut iter = iter.skip_while(|(_, c)| c != &']');
+			iter.next().expect("Authority parsing logic error");
+			iter.next()
+		}
+		_ => iter.skip_while(|(_, c)| c != &':').next(),
+	};
+
+	match split {
+		Some((i, ':')) => Ok(&authority[..i]),
+		None => Ok(authority),
+		Some((_, _)) => Err(Error::BadRequest(format!(
+			"Authority {} has an illegal format",
+			authority
+		))),
+	}
+}
+
+/// Host to bucket
+///
+/// Convert a host, like "bucket.garage-site.tld" or "john.doe.com"
+/// to the corresponding bucket, resp. "bucket" and "john.doe.com"
+/// considering that ".garage-site.tld" is the "root domain".
+/// This behavior has been chosen to follow AWS S3 semantic.
+fn host_to_bucket<'a>(host: &'a str, root: &str) -> &'a str {
+	if root.len() >= host.len() || !host.ends_with(root) {
+		return host;
+	}
+
+	let len_diff = host.len() - root.len();
+	let missing_starting_dot = root.chars().next() != Some('.');
+	let cursor = if missing_starting_dot {
+		len_diff - 1
+	} else {
+		len_diff
+	};
+	&host[..cursor]
+}
+
+/// Path to key
+///
+/// Convert the provided path to the internal key
+/// When a path ends with "/", we append the index name to match traditional web server behavior
+/// which is also AWS S3 behavior.
+fn path_to_key<'a>(path: &'a str, index: &str) -> Result<Cow<'a, str>, Error> {
+	let path_utf8 = percent_encoding::percent_decode_str(&path).decode_utf8()?;
+
+	if path_utf8.chars().next() != Some('/') {
+		return Err(Error::BadRequest(format!(
+			"Path must start with a / (slash)"
+		)));
+	}
+
+	match path_utf8.chars().last() {
+		None => Err(Error::BadRequest(format!(
+			"Path must have at least a character"
+		))),
+		Some('/') => {
+			let mut key = String::with_capacity(path_utf8.len() + index.len());
+			key.push_str(&path_utf8[1..]);
+			key.push_str(index);
+			Ok(key.into())
+		}
+		Some(_) => match path_utf8 {
+			Cow::Borrowed(pu8) => Ok((&pu8[1..]).into()),
+			Cow::Owned(pu8) => Ok((&pu8[1..]).to_string().into()),
+		},
+	}
+}
+
+#[cfg(test)]
+mod tests {
+	use super::*;
+
+	#[test]
+	fn authority_to_host_with_port() -> Result<(), Error> {
+		let domain = authority_to_host("[::1]:3902")?;
+		assert_eq!(domain, "[::1]");
+		let domain2 = authority_to_host("garage.tld:65200")?;
+		assert_eq!(domain2, "garage.tld");
+		let domain3 = authority_to_host("127.0.0.1:80")?;
+		assert_eq!(domain3, "127.0.0.1");
+		Ok(())
+	}
+
+	#[test]
+	fn authority_to_host_without_port() -> Result<(), Error> {
+		let domain = authority_to_host("[::1]")?;
+		assert_eq!(domain, "[::1]");
+		let domain2 = authority_to_host("garage.tld")?;
+		assert_eq!(domain2, "garage.tld");
+		let domain3 = authority_to_host("127.0.0.1")?;
+		assert_eq!(domain3, "127.0.0.1");
+		Ok(())
+	}
+
+	#[test]
+	fn host_to_bucket_test() {
+		assert_eq!(
+			host_to_bucket("john.doe.garage.tld", ".garage.tld"),
+			"john.doe"
+		);
+
+		assert_eq!(
+			host_to_bucket("john.doe.garage.tld", "garage.tld"),
+			"john.doe"
+		);
+
+		assert_eq!(host_to_bucket("john.doe.com", "garage.tld"), "john.doe.com");
+
+		assert_eq!(
+			host_to_bucket("john.doe.com", ".garage.tld"),
+			"john.doe.com"
+		);
+
+		assert_eq!(host_to_bucket("garage.tld", "garage.tld"), "garage.tld");
+
+		assert_eq!(host_to_bucket("garage.tld", ".garage.tld"), "garage.tld");
+	}
+
+	#[test]
+	fn path_to_key_test() -> Result<(), Error> {
+		assert_eq!(path_to_key("/file%20.jpg", "index.html")?, "file .jpg");
+		assert_eq!(path_to_key("/%20t/", "index.html")?, " t/index.html");
+		assert_eq!(path_to_key("/", "index.html")?, "index.html");
+		assert_eq!(path_to_key("/hello", "index.html")?, "hello");
+		assert!(path_to_key("", "index.html").is_err());
+		assert!(path_to_key("i/am/relative", "index.html").is_err());
+		Ok(())
+	}
+}