Remove by_text parm (#85)

langchain_weaviate/vectorstores.py

@@ -94,23 +94,21 @@ def __init__(
         relevance_score_fn: Optional[
             Callable[[float], float]
         ] = _default_score_normalizer,
-        by_text: bool = True,
         use_multi_tenancy: bool = False,
     ):
         """Initialize with Weaviate client."""
 
         if not isinstance(client, weaviate.WeaviateClient):
             raise ValueError(
-                "client should be an instance of "
-                "weaviate.WeaviateClient, got {type(client)}"
+                "client should be an instance of"
+                f" weaviate.WeaviateClient, got {type(client)}"
             )
         self._client = client
         self._index_name = index_name or f"LangChain_{uuid4().hex}"
         self._embedding = embedding
         self._text_key = text_key
         self._query_attrs = [self._text_key]
         self.relevance_score_fn = relevance_score_fn
-        self._by_text = by_text
         if attributes is not None:
             self._query_attrs.extend(attributes)
 
@@ -410,7 +408,6 @@ def from_texts(
         *,
         index_name: Optional[str] = None,
         text_key: str = "text",
-        by_text: bool = False,
         relevance_score_fn: Optional[
             Callable[[float], float]
         ] = _default_score_normalizer,
@@ -433,7 +430,6 @@ def from_texts(
             tenant: The tenant name. Defaults to None.
             index_name: Index name.
             text_key: Key to use for uploading/retrieving text to/from vectorstore.
-            by_text: Whether to search by text or by embedding.
             relevance_score_fn: Function for converting whatever distance function the
                 vector store uses to a relevance score, which is a normalized similarity
                 score (0 means dissimilar, 1 means similar).
@@ -462,7 +458,6 @@ def from_texts(
             embedding=embedding,
             attributes=attributes,
             relevance_score_fn=relevance_score_fn,
-            by_text=by_text,
             use_multi_tenancy=tenant is not None,
         )
 

tests/integration_tests/test_vectorstores.py

@@ -165,7 +165,9 @@ def test_similarity_search_by_text(
     """Test end to end construction and search by text."""
 
     docsearch = WeaviateVectorStore.from_texts(
-        texts, embedding_openai, client=weaviate_client, by_text=True
+        texts,
+        embedding_openai,
+        client=weaviate_client,
     )
 
     output = docsearch.similarity_search("foo", k=1)
@@ -361,7 +363,7 @@ def test_similarity_search_with_score(
 
     # now create an instance with an embedding
     docsearch = WeaviateVectorStore.from_texts(
-        texts, embedding_openai, client=weaviate_client, by_text=False
+        texts, embedding_openai, client=weaviate_client
     )
 
     results = docsearch.similarity_search_with_score("kitty", k=1)
@@ -588,3 +590,35 @@ def test_search_with_multi_tenancy(
         ValueError, match="has multi-tenancy enabled, but request was without tenant"
     ):
         docsearch.similarity_search("foo", k=1)
+
+
+def test_invalid_client_type():
+    with pytest.raises(ValueError) as excinfo:
+        invalid_client = "invalid_client"
+        index_name = "test_index"
+        text_key = "text"
+
+        WeaviateVectorStore(
+            client=invalid_client,
+            index_name=index_name,
+            text_key=text_key,
+        )
+
+    assert (
+        str(excinfo.value)
+        == "client should be an instance of weaviate.WeaviateClient, got <class 'str'>"
+    )
+
+
+def test_embedding_property(weaviate_client, embedding_openai):
+    index_name = "test_index"
+    text_key = "text"
+
+    docsearch = WeaviateVectorStore(
+        client=weaviate_client,
+        index_name=index_name,
+        text_key=text_key,
+        embedding=embedding_openai,
+    )
+
+    assert type(docsearch.embeddings) == OpenAIEmbeddings