feat: fuzon-http implementation

fuzon-http/Cargo.toml

@@ -6,6 +6,7 @@ edition.workspace = true
 
 [dependencies]
 actix-web = "4.9.0"
+clap = { version = "4.5.18", features = ["derive"] }
 fuzon = { version = "0.2.2", path = "../fuzon" }
 serde = { version = "1.0.210", features = ["derive"] }
 serde_json = "1.0.128"

fuzon-http/README.md

@@ -0,0 +1,23 @@
+# fuzon-http
+
+This is a web-server to deploy fuzon as a web-service.
+All ontologies are loaded once on server startup, and the indices are kept in memory.
+
+## Configuration
+
+The server takes a configuration file as input to determine what ontologies to load, and which collections to load them into. Collections are individual matchers which can be queried independently.
+
+## Installation
+
+```shell
+cd fuzon-http
+cargo build --release
+../target/release/fuzon-http --config config/example.json
+```
+
+## Usage
+
+Fuzzy matching queries should use `GET /top?collection={collection}&top={top}&query={query}`.
+
+To discover available collections, use `GET /list`.
+

fuzon-http/config/example.json

@@ -0,0 +1,7 @@
+{
+  "collections": {
+    "cell_type": "https://purl.obolibrary.org/obo/cl.owl",
+    "source_material": "https://purl.obolibrary.org/obo/uberon.owl",
+    "taxon_id": "https://purl.obolibrary.org/obo/ncbitaxon/subsets/taxslim.owl"
+  }
+}

fuzon-http/src/main.rs

@@ -1,9 +1,11 @@
 use std::collections::HashMap;
 use actix_web::{get, web, App, HttpServer, Responder, Result};
+use clap::Parser;
 use fuzon::{TermMatcher};
 use serde::{Deserialize, Serialize};
 use serde_json;
 use std::sync::Arc;
+use std::fs::File;
 
 // URL query parameters when requesting matching codes
 #[derive(Debug, Deserialize)]
@@ -28,6 +30,7 @@ struct Config {
 }
 
 // Shared app state built from config and used by services
+#[derive(Clone, Debug)]
 struct AppState {
     collections: Arc<HashMap<String, TermMatcher>>,
 }
@@ -37,16 +40,13 @@ impl AppState {
         let collections = data
             .collections
             .into_iter()
-            .map(|(k, v)| (k, TermMatcher::from_paths(vec![&v]).unwrap())).collect();
+            .inspect(|(k, _)| println!("Loading {}...", k))
+            .map(|(k, v)| (k, TermMatcher::from_paths(vec![&v]).unwrap()))
+            .collect();
         AppState { collections: Arc::new(collections) }
     }
 }
 
-// Used for debugging
-#[get("/hello/{name}")]
-async fn greet(name: web::Path<String>) -> impl Responder {
-    format!("Hello {name}!")
-}
 
 // list collections: /list
 #[get("/list")]
@@ -75,28 +75,34 @@ async fn top(data: web::Data<AppState>, req: web::Query<CodeRequest>) -> Result<
     Ok(web::Json(top_terms))
 }
 
+/// http server to serve the fuzon terminology matching api
+#[derive(Parser, Debug)]
+#[command(version, about, long_about = None)]
+struct Args {
+    /// Path to the configuration file.
+    #[clap(short, long)]
+    config: String,
+}
+
+
 #[actix_web::main] // or #[tokio::main]
 async fn main() -> std::io::Result<()> {
+    let args = Args::parse();
+    let config_path = args.config;
+
+    let config: Config = serde_json::from_reader(
+        File::open(config_path).expect("Failed to open config file.")
+    ).expect("Failed to parse config.");
 
-    // NOTE: config as inline json for debugging, later this will be a config file
-    let config_data = r#"
-        {
-            "collections": {
-                "cell_type": "https://purl.obolibrary.org/obo/cl.owl",
-                "source_material": "https://purl.obolibrary.org/obo/uberon.owl",
-                "taxon_id": "https://purl.obolibrary.org/obo/ncbitaxon/subsets/taxslim.owl"
-            }
-        }"#;
     let data = web::block(move || 
-        serde_json::from_str::<Config>(config_data).unwrap()
+        AppState::from_config(config)
     )
         .await
-        .expect("Failed to parse config");
+        .expect("Failed to initialize state from config.");
 
     HttpServer::new(move || {
         App::new()
-            .app_data(web::Data::new(AppState::from_config(data.clone())))
-            .service(greet)
+            .app_data(web::Data::new(data.clone()))
             .service(list)
             .service(top)
     })

fuzon/src/lib.rs

@@ -33,6 +33,7 @@ lazy_static! {
     };
 }
 
+#[derive(Debug, Clone)]
 pub struct TermMatcher {
     pub terms: Vec<Term>,
 }