Clarify embedding facilities in LanceDB docs

Signed-off-by: Marcel Coetzee <[email protected]>

docs/examples/custom_destination_lancedb/custom_destination_lancedb.py

@@ -38,7 +38,9 @@
 from dlt.sources.helpers.rest_client import RESTClient, AuthConfigBase
 
 # access secrets to get openai key and instantiate embedding function
-openai_api_key: str = dlt.secrets.get("destination.lancedb.credentials.embedding_model_provider_api_key")
+openai_api_key: str = dlt.secrets.get(
+    "destination.lancedb.credentials.embedding_model_provider_api_key"
+)
 func = get_registry().get("openai").create(name="text-embedding-3-small", api_key=openai_api_key)
 
 

docs/website/docs/dlt-ecosystem/destinations/lancedb.md

@@ -120,6 +120,8 @@ function.
 
 ## Using an Adapter to Specify Columns to Vectorise
 
+Out of the box, LanceDB will act as a normal database. To use LanceDB's embedding facilities, you'll need to specify which fields you'd like to embed in your dlt resource.
+
 The `lancedb_adapter` is a helper function that configures the resource for the LanceDB destination:
 
 ```py
@@ -142,6 +144,8 @@ lancedb_adapter(
 )
 ```
 
+Bear in mind that you can't use an adapter on a [dlt source](../../general-usage/source.md), only a [dlt resource](../../general-usage/resource.md).
+
 ## Write disposition
 
 All [write dispositions](../../general-usage/incremental-loading.md#choosing-a-write-disposition) are supported by the LanceDB destination.