Add README

Author: TanmayArya-1p

crates/grpc_server/.sample.env

@@ -3,7 +3,7 @@ GRPC_SERVER_DIMENSION=3 // required
 
 GRPC_SERVER_HOST=localhost // defaults to 127.0.0.1 aka localhost
 GRPC_SERVER_PORT=8080 // defaults to 8080
-GRPC_SERVER_STORAGE_TYPE=inmemory // (inmemory/rocksdb) defaults to inmemory
+GRPC_SERVER_STORAGE_TYPE=inmemory // (inmemory/rocksdb) defaults to 'inmemory'
 GRPC_SERVER_INDEX_TYPE=flat // defaults to flat
 GRPC_SERVER_DATA_PATH=data // defaults to a temporary directory
 GRPC_SERVER_LOGGING=true // defaults to true

crates/grpc_server/README.md

@@ -0,0 +1,31 @@
+### Build
+
+Clone the repository and run `cargo build --bin grpc_server` to build the binary of the gRPC server crate.
+
+You can than then start the gRPC server by running:
+
+```bash
+cargo run --bin grpc_server
+```
+
+### Configuration
+
+Use the [.sample.env](.sample.env) shown below as a reference to set your environment variables in a `.env` file.
+
+```bash
+GRPC_SERVER_ROOT_PASSWORD=123 // required
+GRPC_SERVER_DIMENSION=3 // required
+
+GRPC_SERVER_HOST=localhost // defaults to 127.0.0.1 aka localhost
+GRPC_SERVER_PORT=8080 // defaults to 8080
+GRPC_SERVER_STORAGE_TYPE=inmemory // (inmemory/rocksdb) defaults to inmemory
+GRPC_SERVER_INDEX_TYPE=flat // defaults to flat
+GRPC_SERVER_DATA_PATH=data // defaults to a temporary directory
+GRPC_SERVER_LOGGING=true // defaults to true
+
+```
+
+
+### Testing
+
+The [vector-db.proto](proto/vector-db.proto) can be imported into any gRPC client.

crates/grpc_server/src/main.rs

@@ -71,7 +71,6 @@ impl VectorDb for VectorDBService {
 
         let point = point_opt.unwrap();
 
-        // TODO: handle payload properly once we decide to define it
         Ok(Response::new(Point {
             id: Some(PointId { id: point.id }),
             vector: Some(DenseVector {
@@ -89,8 +88,6 @@ impl VectorDb for VectorDBService {
 
         let search_request = request.into_inner();
 
-        //TODO: distance function panics if different dimensions; no dimension enforcement
-
         // extract request arguments
         let query_vect = search_request.query_vector.unwrap();
         let similarity = SIMILARITY_PROTOBUFF_MAP[search_request.similarity as usize];