How can i delete documents from a vector store (AzureCosmosDBVectorSearch) based on a metadata field?

[yyueda]

Example Code

def add_user_metadata(username, chunks):
    for chunk in chunks:
        chunk.metadata["username"] = username
    
    return chunks

def embed_documents(chunks):
    try: 
        vectorstore = AzureCosmosDBVectorSearch.from_documents(
            chunks,
            embedding_model,
            collection=collection,
            index_name = INDEX_NAME,
            embedding_key=embedding_key,
        )
        
        # Create index if not yet created
        if vectorstore.index_exists() is False:
            num_lists = 100
            dimensions = 1536
            similarity_algorithm = CosmosDBSimilarityType.COS
            kind = CosmosDBVectorSearchType.VECTOR_IVF
            m = 16
            ef_construction = 64

            vectorstore.create_index(
                num_lists, dimensions, similarity_algorithm, kind, m, ef_construction
            )
        
        return True
    except Exception as e:
        print(f"Error embedding documents: {e}")
        return False

Description

I am storing each document chunk in my db, with the a username field in the metadata referring to the user. I want to be able to delete all the documents belonging to a specific user? Do i use the .delete() method for the vectorstore? How can i retrieve the document ids?