From fc89f6ea74307b163f3282afd5ed612f5ce60eb0 Mon Sep 17 00:00:00 2001
From: Sara Zan <sara.zanzottera@deepset.ai>
Date: Tue, 6 Dec 2022 14:48:58 +0100
Subject: [PATCH] fix: revert Weaviate query with filters and improve tests
 (#3646)

* revert weaviate query with filters and improve tests

* pylint

* upgrade weaviate container

* use latest docker tag

* fix text

* fix text
---
 .github/workflows/tests.yml           |  2 +-
 haystack/document_stores/weaviate.py  | 34 ++++++++++++++-------------
 test/document_stores/test_weaviate.py |  5 ++--
 3 files changed, 22 insertions(+), 19 deletions(-)

diff --git a/.github/workflows/tests.yml b/.github/workflows/tests.yml
index 79712e486..bd57a42a7 100644
--- a/.github/workflows/tests.yml
+++ b/.github/workflows/tests.yml
@@ -335,7 +335,7 @@ jobs:
     runs-on: ${{ matrix.os }}
     services:
       weaviate:
-        image: semitechnologies/weaviate:1.16.0
+        image: semitechnologies/weaviate:latest
         env:
           AUTHENTICATION_ANONYMOUS_ACCESS_ENABLED: "true"
           PERSISTENCE_DATA_PATH: "/var/lib/weaviate"
diff --git a/haystack/document_stores/weaviate.py b/haystack/document_stores/weaviate.py
index f3dab95ec..c16d4b071 100644
--- a/haystack/document_stores/weaviate.py
+++ b/haystack/document_stores/weaviate.py
@@ -985,28 +985,30 @@ class WeaviateDocumentStore(BaseDocumentStore):
             )
 
             # Retrieval with BM25 AND filtering
-            if filters:
-
-                # Once Weaviate starts supporting filters with BM25:
-                filter_dict = LogicalFilterClause.parse(filters).convert_to_weaviate()
+            if filters:  # pylint: disable=no-else-raise
+                raise NotImplementedError(
+                    "Weaviate currently does not support filters WITH inverted index text query (eg BM25)!"
+                )
+                # # Once Weaviate starts supporting filters with BM25:
+                # filter_dict = LogicalFilterClause.parse(filters).convert_to_weaviate()
+                # gql_query = (
+                #     weaviate.gql.get.GetBuilder(
+                #         class_name=index, properties=properties, connection=self.weaviate_client
+                #     )
+                #     .with_near_vector({"vector": [0, 0]})
+                #     .with_where(filter_dict)
+                #     .with_limit(top_k)
+                #     .build()
+                # )
+            else:
+                # BM25 retrieval without filtering
                 gql_query = (
-                    weaviate.gql.get.GetBuilder(
-                        class_name=index, properties=properties, connection=self.weaviate_client
-                    )
+                    gql.get.GetBuilder(class_name=index, properties=properties, connection=self.weaviate_client)
                     .with_near_vector({"vector": [0, 0]})
-                    .with_where(filter_dict)
                     .with_limit(top_k)
                     .build()
                 )
 
-            # BM25 retrieval without filtering
-            gql_query = (
-                gql.get.GetBuilder(class_name=index, properties=properties, connection=self.weaviate_client)
-                .with_near_vector({"vector": [0, 0]})
-                .with_limit(top_k)
-                .build()
-            )
-
             # Build the BM25 part of the GQL manually.
             # Currently the GetBuilder of the Weaviate-client (v3.6.0)
             # does not support the BM25 part of GQL building, so
diff --git a/test/document_stores/test_weaviate.py b/test/document_stores/test_weaviate.py
index a92ad9846..80c206345 100644
--- a/test/document_stores/test_weaviate.py
+++ b/test/document_stores/test_weaviate.py
@@ -180,8 +180,9 @@ class TestWeaviateDocumentStore(DocumentStoreBaseTestAbstract):
         assert len(docs) == 3
 
         # BM25 retrieval WITH filters is not yet supported as of Weaviate v1.14.1
-        # with pytest.raises(Exception):
-        docs = ds.query(query_text, filters={"name": ["filename2"]})
+        # Should be from 1.18: https://github.com/semi-technologies/weaviate/issues/2393
+        # docs = ds.query(query_text, filters={"name": ["name_1"]})
+        # assert len(docs) == 1
 
         docs = ds.query(filters={"name": ["name_0"]})
         assert len(docs) == 3