docs: add doc strings

rig-lancedb/src/lib.rs

@@ -22,15 +22,19 @@ fn serde_to_rig_error(e: serde_json::Error) -> VectorStoreError {
 }
 
 pub struct LanceDbVectorStore<M: EmbeddingModel> {
-    /// Defines which model is used to generate embeddings for the vector store
+    /// Defines which model is used to generate embeddings for the vector store.
     model: M,
+    /// LanceDB table containing embeddings.
     table: lancedb::Table,
+    /// Column name in `table` that contains the id of a record.
     id_field: String,
     /// Vector search params that are used during vector search operations.
     search_params: SearchParams,
 }
 
 impl<M: EmbeddingModel> LanceDbVectorStore<M> {
+    /// Apply the search_params to the vector query.
+    /// This is a helper function used by the methods `top_n` and `top_n_ids` of the `VectorStoreIndex` trait.
     fn build_query(&self, mut query: VectorQuery) -> VectorQuery {
         let SearchParams {
             distance_type,
@@ -136,7 +140,8 @@ impl<M: EmbeddingModel> LanceDbVectorStore<M> {
         })
     }
 
-    /// Define index on document table `id` field for search optimization.
+    /// Define an index on the specified fields of the lanceDB table for search optimization.
+    /// Note: it is required to add an index on the column containing the embeddings when performing an ANN type vector search.
     pub async fn create_index(
         &self,
         index: Index,

rig-lancedb/src/utils/deserializer.rs

@@ -25,6 +25,8 @@ fn arrow_to_rig_error(e: ArrowError) -> VectorStoreError {
     VectorStoreError::DatastoreError(Box::new(e))
 }
 
+/// Trait used to deserialize data returned from LanceDB queries into a serde_json::Value vector.
+/// Data returned by LanceDB is a vector of `RecordBatch` items.
 pub trait RecordBatchDeserializer {
     fn deserialize(&self) -> Result<Vec<serde_json::Value>, VectorStoreError>;
 }
@@ -43,6 +45,7 @@ impl RecordBatchDeserializer for Vec<RecordBatch> {
 
 impl RecordBatchDeserializer for RecordBatch {
     fn deserialize(&self) -> Result<Vec<serde_json::Value>, VectorStoreError> {
+        /// Recursive function that matches all possible data types store in LanceDB and converts them to serde_json::Value.
         fn type_matcher(column: &Arc<dyn Array>) -> Result<Vec<Value>, VectorStoreError> {
             match column.data_type() {
                 DataType::Null => Ok(vec![serde_json::Value::Null]),

rig-lancedb/src/utils/mod.rs

@@ -7,16 +7,12 @@ use rig::vector_store::VectorStoreError;
 
 use crate::lancedb_to_rig_error;
 
-/// Trait that facilitates the conversion of columnar data returned by a lanceDb query to the desired struct.
+/// Trait that facilitates the conversion of columnar data returned by a lanceDb query to serde_json::Value.
 /// Used whenever a lanceDb table is queried.
-/// First, execute the query and get the result as a list of RecordBatches (columnar data).
-/// Then, convert the record batches to the desired type using the try_from trait.
 pub trait Query {
     async fn execute_query(&self) -> Result<Vec<serde_json::Value>, VectorStoreError>;
 }
 
-/// Same as the above trait but for the VectorQuery type.
-/// Used whenever a lanceDb table vector search is executed.
 impl Query for lancedb::query::VectorQuery {
     async fn execute_query(&self) -> Result<Vec<serde_json::Value>, VectorStoreError> {
         let record_batches = self