refactor(lancedb): docs, make examples idempotent

refactor: clippy

rig-lancedb/README.md

@@ -17,6 +17,9 @@
 ## Rig-Lancedb
 This companion crate implements a Rig vector store based on Lancedb.
 
+## Pre-requisites
+If you are using `rig-lancedb` locally, you must ensure you have `protoc` (the [Protobuf Compiler](https://protobuf.dev/installation/)).
+
 ## Usage
 
 Add the companion crate to your `Cargo.toml`, along with the rig-core crate:

rig-lancedb/examples/vector_search_local_ann.rs

@@ -38,8 +38,15 @@ async fn main() -> Result<(), anyhow::Error> {
         .build()
         .await?;
 
-    let table = db
-        .create_table(
+    let table = if db
+        .table_names()
+        .execute()
+        .await?
+        .contains(&"definitions".to_string())
+    {
+        db.open_table("definitions").execute().await?
+    } else {
+        db.create_table(
             "definitions",
             RecordBatchIterator::new(
                 vec![as_record_batch(embeddings, model.ndims())],
@@ -47,9 +54,11 @@ async fn main() -> Result<(), anyhow::Error> {
             ),
         )
         .execute()
-        .await?;
+        .await?
+    };
 
     // See [LanceDB indexing](https://lancedb.github.io/lancedb/concepts/index_ivfpq/#product-quantization) for more information
+    if table.index_stats("embedding").await?.is_none() {
         table
             .create_index(
                 &["embedding"],
@@ -57,6 +66,7 @@ async fn main() -> Result<(), anyhow::Error> {
             )
             .execute()
             .await?;
+    }
 
     // Define search_params params that will be used by the vector store to perform the vector search.
     let search_params = SearchParams::default();

rig-lancedb/examples/vector_search_local_enn.rs

@@ -32,8 +32,15 @@ async fn main() -> Result<(), anyhow::Error> {
     // Initialize LanceDB locally.
     let db = lancedb::connect("data/lancedb-store").execute().await?;
 
-    let table = db
-        .create_table(
+    let table = if db
+        .table_names()
+        .execute()
+        .await?
+        .contains(&"definitions".to_string())
+    {
+        db.open_table("definitions").execute().await?
+    } else {
+        db.create_table(
             "definitions",
             RecordBatchIterator::new(
                 vec![as_record_batch(embeddings, model.ndims())],
@@ -41,7 +48,8 @@ async fn main() -> Result<(), anyhow::Error> {
             ),
         )
         .execute()
-        .await?;
+        .await?
+    };
 
     let vector_store = LanceDbVectorIndex::new(table, model, "id", search_params).await?;
 

rig-lancedb/examples/vector_search_s3_ann.rs

@@ -44,8 +44,15 @@ async fn main() -> Result<(), anyhow::Error> {
         .build()
         .await?;
 
-    let table = db
-        .create_table(
+    let table = if db
+        .table_names()
+        .execute()
+        .await?
+        .contains(&"definitions".to_string())
+    {
+        db.open_table("definitions").execute().await?
+    } else {
+        db.create_table(
             "definitions",
             RecordBatchIterator::new(
                 vec![as_record_batch(embeddings, model.ndims())],
@@ -53,9 +60,11 @@ async fn main() -> Result<(), anyhow::Error> {
             ),
         )
         .execute()
-        .await?;
+        .await?
+    };
 
     // See [LanceDB indexing](https://lancedb.github.io/lancedb/concepts/index_ivfpq/#product-quantization) for more information
+    if table.index_stats("embedding").await?.is_none() {
         table
             .create_index(
                 &["embedding"],
@@ -68,6 +77,7 @@ async fn main() -> Result<(), anyhow::Error> {
             )
             .execute()
             .await?;
+    }
 
     // Define search_params params that will be used by the vector store to perform the vector search.
     let search_params = SearchParams::default().distance_type(DistanceType::Cosine);