reorg

.gitignore

@@ -19,11 +19,7 @@ Cargo.lock
 /target
 
 # Data files
-/data/wiki-data
-/data/index_unit_test/index
-/data/test
-/data/small
-/data/illinois
+/data
 
 # TODO
 todo.md

Cargo.toml

@@ -1,17 +1,6 @@
-[package]
-name = "search-rs"
-version = "0.1.0"
-edition = "2021"
+[workspace]
 
-[lib]
-name = "search"
-path = "src/lib.rs"
-
-[dependencies]
-rand = "0.8"
-tokenizers = { version = "0.15.0", features = ["http"] }
-rust-stemmers = "1.2.0"
-rayon = "1.8.0"
-indicatif = {version = "0.17.0", features = ["rayon", "improved_unicode"]}
-fxhash = "0.2.1"
-tempdir = "0.3.7"
+members = [
+    "search", 
+    "client"
+]

README.md

@@ -15,7 +15,7 @@ Search engine written in Rust, based on an inverted index on disk.
 
 **Index construction**
 - [x] In-memory datasets index construction;
-- [ ] Proper vocabulary and paths on disk;
+- [x] Proper vocabulary and paths on disk;
 - [ ] Spelling correction index;
 - [ ] Disk-based partial index construction and merging;
 
@@ -30,6 +30,10 @@ Search engine written in Rust, based on an inverted index on disk.
 - [ ] Query quality; 
 - [ ] Disk overhead.
 
+**Client**
+- [x] CLI
+- [ ] Web interface
+
 ## Crates in use
 - [stemmer-rs](https://github.com/lise-henry/stemmer-rs)
 - [tokenizers](https://github.com/huggingface/tokenizers)

client/Cargo.toml

@@ -0,0 +1,10 @@
+[package]
+name = "client"
+version = "0.1.0"
+edition = "2021"
+
+# See more keys and their definitions at https://doc.rust-lang.org/cargo/reference/manifest.html
+
+[dependencies]
+actix-web = "4.4.1"
+search = { path = "../search" }

client/src/main.rs

@@ -0,0 +1,28 @@
+use actix_web::{get, post, web, App, HttpResponse, HttpServer, Responder};
+
+#[get("/")]
+async fn hello() -> impl Responder {
+    HttpResponse::Ok().body("Hello world!")
+}
+
+#[post("/echo")]
+async fn echo(req_body: String) -> impl Responder {
+    HttpResponse::Ok().body(req_body)
+}
+
+async fn manual_hello() -> impl Responder {
+    HttpResponse::Ok().body("Hey there!")
+}
+
+#[actix_web::main]
+async fn main() -> std::io::Result<()> {
+    HttpServer::new(|| {
+        App::new()
+            .service(hello)
+            .service(echo)
+            .route("/hey", web::get().to(manual_hello))
+    })
+    .bind(("127.0.0.1", 8080))?
+    .run()
+    .await
+}

search/Cargo.toml

@@ -0,0 +1,14 @@
+[package]
+name = "search"
+version = "0.1.0"
+edition = "2021"
+
+
+[dependencies]
+rand = "0.8"
+tokenizers = { version = "0.15.0", features = ["http"] }
+rust-stemmers = "1.2.0"
+rayon = "1.8.0"
+indicatif = {version = "0.17.0", features = ["rayon", "improved_unicode"]}
+fxhash = "0.2.1"
+tempdir = "0.3.7"

src/index/mod.rs → search/src/index/mod.rs

@@ -97,13 +97,9 @@ mod test {
     fn test_build() {
         let index_path = &create_temporary_dir_path();
 
-        Index::build_index(
-            "data/index_unit_test/docs",
-            index_path,
-            "data/index_unit_test/test_tokenizer",
-        );
+        Index::build_index("test_data/docs", index_path, "test_data/test_tokenizer");
 
-        let mut idx = Index::load_index(index_path, "data/index_unit_test/test_tokenizer");
+        let mut idx = Index::load_index(index_path, "test_data/test_tokenizer");
 
         for ele in ["hello", "man", "world"] {
             assert!(idx.term_to_index.contains_key(ele));
@@ -119,13 +115,7 @@ mod test {
 
         hello_docs.sort();
 
-        assert_eq!(
-            hello_docs,
-            [
-                "data/index_unit_test/docs/1.txt",
-                "data/index_unit_test/docs/2.txt"
-            ]
-        );
+        assert_eq!(hello_docs, ["test_data/docs/1.txt", "test_data/docs/2.txt"]);
 
         assert_eq!(pl.collection_frequency, 2);
 

src/main.rs → search/src/main.rs

@@ -1,6 +1,6 @@
 use std::env;
 use std::io::{self, Write};
-use std::process::Command;
+use std::process::{exit, Command};
 use std::time::{Duration, Instant};
 
 use search::index::Index;
@@ -53,10 +53,7 @@ fn main() {
     let args: Vec<String> = env::args().collect();
 
     if args.len() < 3 || args.len() > 5 {
-        println!(
-            "Usage: {} <base_path> <load_or_build> [build_num_threads]",
-            args[0]
-        );
+        println!("Usage: cargo run --bin search <base_path> <load_or_build> [build_num_threads]");
         return;
     }
 
@@ -87,9 +84,12 @@ fn main() {
         Index::build_index(&docs_path, &index_path, &tokenizer_path);
         let elapsed_time = start_time.elapsed();
         println!(
-            "Index built in {}.\n",
-            HumanDuration(Duration::from_secs(elapsed_time.as_secs()))
+            "Index built in {}.\n\nLoad options:\n- CLI: cargo run --release --bin search {} load",
+            HumanDuration(Duration::from_secs(elapsed_time.as_secs())),
+            base_path
         );
+
+        exit(0);
     }
 
     let mut q = QueryProcessor::build_query_processor(&index_path, &tokenizer_path);