From 01a57198b8eb4c0c136eb8af77f3fd16359e4112 Mon Sep 17 00:00:00 2001
From: Tim Asp <707699+timothyasp@users.noreply.github.com>
Date: Fri, 10 Mar 2023 15:14:35 -0800
Subject: [PATCH] [bugfix] Fix persisted chromadb vectorstore (#1444)

If a `persist_directory` param was set, chromadb would throw a warning
that ""No embedding_function provided, using default embedding function:
SentenceTransformerEmbeddingFunction". and would error with a `Illegal
instruction: 4` error.

This is on a MBP M1 13.2.1, python 3.9.

I'm not entirely sure why that error happened, but when using
`get_or_create_collection` instead of `list_collection` on our end, the
error and warning goes away and chroma works as expected.

Added bonus this is cleaner and likely more efficient.
`list_collections` builds a new `Collection` instance for each collect,
then `Chroma` would just use the `name` field to tell if the collection
existed.
---
 langchain/vectorstores/chroma.py | 25 ++++++++-----------------
 1 file changed, 8 insertions(+), 17 deletions(-)

diff --git a/langchain/vectorstores/chroma.py b/langchain/vectorstores/chroma.py
index 0699b089..d4a8d63d 100644
--- a/langchain/vectorstores/chroma.py
+++ b/langchain/vectorstores/chroma.py
@@ -78,22 +78,12 @@ class Chroma(VectorStore):
         self._client = chromadb.Client(self._client_settings)
         self._embedding_function = embedding_function
         self._persist_directory = persist_directory
-
-        # Check if the collection exists, create it if not
-        if collection_name in [col.name for col in self._client.list_collections()]:
-            self._collection = self._client.get_collection(name=collection_name)
-            # TODO: Persist the user's embedding function
-            logger.warning(
-                f"Collection {collection_name} already exists,"
-                " Do you have the right embedding function?"
-            )
-        else:
-            self._collection = self._client.create_collection(
-                name=collection_name,
-                embedding_function=self._embedding_function.embed_documents
-                if self._embedding_function is not None
-                else None,
-            )
+        self._collection = self._client.get_or_create_collection(
+            name=collection_name,
+            embedding_function=self._embedding_function.embed_documents
+            if self._embedding_function is not None
+            else None,
+        )
 
     def add_texts(
         self,
@@ -224,9 +214,9 @@ class Chroma(VectorStore):
         Otherwise, the data will be ephemeral in-memory.
 
         Args:
+            texts (List[str]): List of texts to add to the collection.
             collection_name (str): Name of the collection to create.
             persist_directory (Optional[str]): Directory to persist the collection.
-            documents (List[Document]): List of documents to add.
             embedding (Optional[Embeddings]): Embedding function. Defaults to None.
             metadatas (Optional[List[dict]]): List of metadatas. Defaults to None.
             ids (Optional[List[str]]): List of document IDs. Defaults to None.
@@ -263,6 +253,7 @@ class Chroma(VectorStore):
         Args:
             collection_name (str): Name of the collection to create.
             persist_directory (Optional[str]): Directory to persist the collection.
+            ids (Optional[List[str]]): List of document IDs. Defaults to None.
             documents (List[Document]): List of documents to add to the vectorstore.
             embedding (Optional[Embeddings]): Embedding function. Defaults to None.
             client_settings (Optional[chromadb.config.Settings]): Chroma client settings