Updated test infrastructure with latest versions (redis#3484)

* Updated latest image to 8.0-M03-pre

* Changed redis image for testing purposes

* Updated image version

* Updated redis server versions

* Updated test case

* Revert version restriction

* Updated redis versions

* Added tests for new default scorer

* Skipped test on 8.0

* Fixed query to match exact-match syntax

* Codestyle fixes

* Added condition for 8.0-M04-pre image

* Added test for INFO section

Author: vladvildanov

.github/actions/run-tests/action.yml

@@ -35,6 +35,10 @@ runs:
         CLIENT_LIBS_TEST_IMAGE: "redislabs/client-libs-test:${{ inputs.redis-version }}"
       run: |
         set -e
+        
+        if [ "${{inputs.redis-version}}" == "8.0-M04-pre" ]; then
+          export REDIS_IMAGE=redis:8.0-M03
+        fi
                 
         echo "::group::Installing dependencies"
         pip install -U setuptools wheel
@@ -56,9 +60,9 @@ runs:
         
           # Mapping of redis version to stack version  
           declare -A redis_stack_version_mapping=(
-            ["7.4.1"]="7.4.0-v1"
-            ["7.2.6"]="7.2.0-v13"
-            ["6.2.16"]="6.2.6-v17"
+            ["7.4.2"]="7.4.0-v2"
+            ["7.2.7"]="7.2.0-v14"
+            ["6.2.17"]="6.2.6-v18"
           )
                   
           if [[ -v redis_stack_version_mapping[$REDIS_VERSION] ]]; then

.github/workflows/integration.yaml

@@ -29,7 +29,7 @@ env:
   COVERAGE_CORE: sysmon
   REDIS_IMAGE: redis:latest
   REDIS_STACK_IMAGE: redis/redis-stack-server:latest
-  CURRENT_REDIS_VERSION: '7.4.1'
+  CURRENT_REDIS_VERSION: '7.4.2'
 
 jobs:
   dependency-audit:
@@ -74,7 +74,7 @@ jobs:
       max-parallel: 15
       fail-fast: false
       matrix:
-        redis-version: ['8.0-M02', '${{ needs.redis_version.outputs.CURRENT }}', '7.2.6', '6.2.16']
+        redis-version: ['8.0-M04-pre', '${{ needs.redis_version.outputs.CURRENT }}', '7.2.7', '6.2.17']
         python-version: ['3.8', '3.12']
         parser-backend: ['plain']
         event-loop: ['asyncio']

redis/commands/search/query.py

@@ -179,6 +179,8 @@ def scorer(self, scorer: str) -> "Query":
         Use a different scoring function to evaluate document relevance.
         Default is `TFIDF`.
 
+        Since Redis 8.0 default was changed to BM25STD.
+
         :param scorer: The scoring function to use
                        (e.g. `TFIDF.DOCNORM` or `BM25`)
         """

tests/test_asyncio/test_search.py

@@ -341,6 +341,7 @@ async def test_client(decoded_r: redis.Redis):
 
 @pytest.mark.redismod
 @pytest.mark.onlynoncluster
+@skip_if_server_version_gte("7.9.0")
 async def test_scores(decoded_r: redis.Redis):
     await decoded_r.ft().create_index((TextField("txt"),))
 
@@ -361,6 +362,29 @@ async def test_scores(decoded_r: redis.Redis):
         assert "doc1" == res["results"][1]["id"]
 
 
+@pytest.mark.redismod
+@pytest.mark.onlynoncluster
+@skip_if_server_version_lt("7.9.0")
+async def test_scores_with_new_default_scorer(decoded_r: redis.Redis):
+    await decoded_r.ft().create_index((TextField("txt"),))
+
+    await decoded_r.hset("doc1", mapping={"txt": "foo baz"})
+    await decoded_r.hset("doc2", mapping={"txt": "foo bar"})
+
+    q = Query("foo ~bar").with_scores()
+    res = await decoded_r.ft().search(q)
+    if is_resp2_connection(decoded_r):
+        assert 2 == res.total
+        assert "doc2" == res.docs[0].id
+        assert 0.87 == pytest.approx(res.docs[0].score, 0.01)
+        assert "doc1" == res.docs[1].id
+    else:
+        assert 2 == res["total_results"]
+        assert "doc2" == res["results"][0]["id"]
+        assert 0.87 == pytest.approx(res["results"][0]["score"], 0.01)
+        assert "doc1" == res["results"][1]["id"]
+
+
 @pytest.mark.redismod
 async def test_stopwords(decoded_r: redis.Redis):
     stopwords = ["foo", "bar", "baz"]
@@ -663,7 +687,7 @@ async def test_summarize(decoded_r: redis.Redis):
     await createIndex(decoded_r.ft())
     await waitForIndex(decoded_r, "idx")
 
-    q = Query("king henry").paging(0, 1)
+    q = Query('"king henry"').paging(0, 1)
     q.highlight(fields=("play", "txt"), tags=("<b>", "</b>"))
     q.summarize("txt")
 
@@ -675,7 +699,7 @@ async def test_summarize(decoded_r: redis.Redis):
             == doc.txt
         )
 
-        q = Query("king henry").paging(0, 1).summarize().highlight()
+        q = Query('"king henry"').paging(0, 1).summarize().highlight()
 
         doc = sorted((await decoded_r.ft().search(q)).docs)[0]
         assert "<b>Henry</b> ... " == doc.play
@@ -691,7 +715,7 @@ async def test_summarize(decoded_r: redis.Redis):
             == doc["extra_attributes"]["txt"]
         )
 
-        q = Query("king henry").paging(0, 1).summarize().highlight()
+        q = Query('"king henry"').paging(0, 1).summarize().highlight()
 
         doc = sorted((await decoded_r.ft().search(q))["results"])[0]
         assert "<b>Henry</b> ... " == doc["extra_attributes"]["play"]
@@ -1029,6 +1053,7 @@ async def test_phonetic_matcher(decoded_r: redis.Redis):
 @pytest.mark.onlynoncluster
 # NOTE(imalinovskyi): This test contains hardcoded scores valid only for RediSearch 2.8+
 @skip_ifmodversion_lt("2.8.0", "search")
+@skip_if_server_version_gte("7.9.0")
 async def test_scorer(decoded_r: redis.Redis):
     await decoded_r.ft().create_index((TextField("description"),))
 
@@ -1087,6 +1112,69 @@ async def test_scorer(decoded_r: redis.Redis):
         assert 0.0 == res["results"][0]["score"]
 
 
+@pytest.mark.redismod
+@pytest.mark.onlynoncluster
+# NOTE(imalinovskyi): This test contains hardcoded scores valid only for RediSearch 2.8+
+@skip_ifmodversion_lt("2.8.0", "search")
+@skip_if_server_version_lt("7.9.0")
+async def test_scorer_with_new_default_scorer(decoded_r: redis.Redis):
+    await decoded_r.ft().create_index((TextField("description"),))
+
+    await decoded_r.hset(
+        "doc1", mapping={"description": "The quick brown fox jumps over the lazy dog"}
+    )
+    await decoded_r.hset(
+        "doc2",
+        mapping={
+            "description": "Quick alice was beginning to get very tired of sitting by her quick sister on the bank, and of having nothing to do."  # noqa
+        },
+    )
+
+    if is_resp2_connection(decoded_r):
+        # default scorer is BM25STD
+        res = await decoded_r.ft().search(Query("quick").with_scores())
+        assert 0.23 == pytest.approx(res.docs[0].score, 0.05)
+        res = await decoded_r.ft().search(Query("quick").scorer("TFIDF").with_scores())
+        assert 1.0 == res.docs[0].score
+        res = await decoded_r.ft().search(
+            Query("quick").scorer("TFIDF.DOCNORM").with_scores()
+        )
+        assert 0.14285714285714285 == res.docs[0].score
+        res = await decoded_r.ft().search(Query("quick").scorer("BM25").with_scores())
+        assert 0.22471909420069797 == res.docs[0].score
+        res = await decoded_r.ft().search(Query("quick").scorer("DISMAX").with_scores())
+        assert 2.0 == res.docs[0].score
+        res = await decoded_r.ft().search(
+            Query("quick").scorer("DOCSCORE").with_scores()
+        )
+        assert 1.0 == res.docs[0].score
+        res = await decoded_r.ft().search(
+            Query("quick").scorer("HAMMING").with_scores()
+        )
+        assert 0.0 == res.docs[0].score
+    else:
+        res = await decoded_r.ft().search(Query("quick").with_scores())
+        assert 0.23 == pytest.approx(res["results"][0]["score"], 0.05)
+        res = await decoded_r.ft().search(Query("quick").scorer("TFIDF").with_scores())
+        assert 1.0 == res["results"][0]["score"]
+        res = await decoded_r.ft().search(
+            Query("quick").scorer("TFIDF.DOCNORM").with_scores()
+        )
+        assert 0.14285714285714285 == res["results"][0]["score"]
+        res = await decoded_r.ft().search(Query("quick").scorer("BM25").with_scores())
+        assert 0.22471909420069797 == res["results"][0]["score"]
+        res = await decoded_r.ft().search(Query("quick").scorer("DISMAX").with_scores())
+        assert 2.0 == res["results"][0]["score"]
+        res = await decoded_r.ft().search(
+            Query("quick").scorer("DOCSCORE").with_scores()
+        )
+        assert 1.0 == res["results"][0]["score"]
+        res = await decoded_r.ft().search(
+            Query("quick").scorer("HAMMING").with_scores()
+        )
+        assert 0.0 == res["results"][0]["score"]
+
+
 @pytest.mark.redismod
 async def test_get(decoded_r: redis.Redis):
     await decoded_r.ft().create_index((TextField("f1"), TextField("f2")))

tests/test_commands.py

@@ -4332,7 +4332,6 @@ def test_xgroup_create_mkstream(self, r):
         assert r.xinfo_groups(stream) == expected
 
     @skip_if_server_version_lt("7.0.0")
-    @skip_if_server_version_gte("7.9.0")
     def test_xgroup_create_entriesread(self, r: redis.Redis):
         stream = "stream"
         group = "group"
@@ -4341,28 +4340,6 @@ def test_xgroup_create_entriesread(self, r: redis.Redis):
         # no group is setup yet, no info to obtain
         assert r.xinfo_groups(stream) == []
 
-        assert r.xgroup_create(stream, group, 0, entries_read=7)
-        expected = [
-            {
-                "name": group.encode(),
-                "consumers": 0,
-                "pending": 0,
-                "last-delivered-id": b"0-0",
-                "entries-read": 7,
-                "lag": -6,
-            }
-        ]
-        assert r.xinfo_groups(stream) == expected
-
-    @skip_if_server_version_lt("7.9.0")
-    def test_xgroup_create_entriesread_with_fixed_lag_field(self, r: redis.Redis):
-        stream = "stream"
-        group = "group"
-        r.xadd(stream, {"foo": "bar"})
-
-        # no group is setup yet, no info to obtain
-        assert r.xinfo_groups(stream) == []
-
         assert r.xgroup_create(stream, group, 0, entries_read=7)
         expected = [
             {

tests/test_search.py

@@ -314,6 +314,7 @@ def test_client(client):
 
 @pytest.mark.redismod
 @pytest.mark.onlynoncluster
+@skip_if_server_version_gte("7.9.0")
 def test_scores(client):
     client.ft().create_index((TextField("txt"),))
 
@@ -334,6 +335,29 @@ def test_scores(client):
         assert "doc1" == res["results"][1]["id"]
 
 
+@pytest.mark.redismod
+@pytest.mark.onlynoncluster
+@skip_if_server_version_lt("7.9.0")
+def test_scores_with_new_default_scorer(client):
+    client.ft().create_index((TextField("txt"),))
+
+    client.hset("doc1", mapping={"txt": "foo baz"})
+    client.hset("doc2", mapping={"txt": "foo bar"})
+
+    q = Query("foo ~bar").with_scores()
+    res = client.ft().search(q)
+    if is_resp2_connection(client):
+        assert 2 == res.total
+        assert "doc2" == res.docs[0].id
+        assert 0.87 == pytest.approx(res.docs[0].score, 0.01)
+        assert "doc1" == res.docs[1].id
+    else:
+        assert 2 == res["total_results"]
+        assert "doc2" == res["results"][0]["id"]
+        assert 0.87 == pytest.approx(res["results"][0]["score"], 0.01)
+        assert "doc1" == res["results"][1]["id"]
+
+
 @pytest.mark.redismod
 def test_stopwords(client):
     client.ft().create_index((TextField("txt"),), stopwords=["foo", "bar", "baz"])
@@ -623,7 +647,7 @@ def test_summarize(client):
     createIndex(client.ft())
     waitForIndex(client, getattr(client.ft(), "index_name", "idx"))
 
-    q = Query("king henry").paging(0, 1)
+    q = Query('"king henry"').paging(0, 1)
     q.highlight(fields=("play", "txt"), tags=("<b>", "</b>"))
     q.summarize("txt")
 
@@ -635,7 +659,7 @@ def test_summarize(client):
             == doc.txt
         )
 
-        q = Query("king henry").paging(0, 1).summarize().highlight()
+        q = Query('"king henry"').paging(0, 1).summarize().highlight()
 
         doc = sorted(client.ft().search(q).docs)[0]
         assert "<b>Henry</b> ... " == doc.play
@@ -651,7 +675,7 @@ def test_summarize(client):
             == doc["extra_attributes"]["txt"]
         )
 
-        q = Query("king henry").paging(0, 1).summarize().highlight()
+        q = Query('"king henry"').paging(0, 1).summarize().highlight()
 
         doc = sorted(client.ft().search(q)["results"])[0]
         assert "<b>Henry</b> ... " == doc["extra_attributes"]["play"]
@@ -936,6 +960,7 @@ def test_phonetic_matcher(client):
 @pytest.mark.onlynoncluster
 # NOTE(imalinovskyi): This test contains hardcoded scores valid only for RediSearch 2.8+
 @skip_ifmodversion_lt("2.8.0", "search")
+@skip_if_server_version_gte("7.9.0")
 def test_scorer(client):
     client.ft().create_index((TextField("description"),))
 
@@ -982,6 +1007,55 @@ def test_scorer(client):
         assert 0.0 == res["results"][0]["score"]
 
 
+@pytest.mark.redismod
+@pytest.mark.onlynoncluster
+@skip_if_server_version_lt("7.9.0")
+def test_scorer_with_new_default_scorer(client):
+    client.ft().create_index((TextField("description"),))
+
+    client.hset(
+        "doc1", mapping={"description": "The quick brown fox jumps over the lazy dog"}
+    )
+    client.hset(
+        "doc2",
+        mapping={
+            "description": "Quick alice was beginning to get very tired of sitting by her quick sister on the bank, and of having nothing to do."  # noqa
+        },
+    )
+
+    # default scorer is BM25STD
+    if is_resp2_connection(client):
+        res = client.ft().search(Query("quick").with_scores())
+        assert 0.23 == pytest.approx(res.docs[0].score, 0.05)
+        res = client.ft().search(Query("quick").scorer("TFIDF").with_scores())
+        assert 1.0 == res.docs[0].score
+        res = client.ft().search(Query("quick").scorer("TFIDF.DOCNORM").with_scores())
+        assert 0.14285714285714285 == res.docs[0].score
+        res = client.ft().search(Query("quick").scorer("BM25").with_scores())
+        assert 0.22471909420069797 == res.docs[0].score
+        res = client.ft().search(Query("quick").scorer("DISMAX").with_scores())
+        assert 2.0 == res.docs[0].score
+        res = client.ft().search(Query("quick").scorer("DOCSCORE").with_scores())
+        assert 1.0 == res.docs[0].score
+        res = client.ft().search(Query("quick").scorer("HAMMING").with_scores())
+        assert 0.0 == res.docs[0].score
+    else:
+        res = client.ft().search(Query("quick").with_scores())
+        assert 0.23 == pytest.approx(res["results"][0]["score"], 0.05)
+        res = client.ft().search(Query("quick").scorer("TFIDF").with_scores())
+        assert 1.0 == res["results"][0]["score"]
+        res = client.ft().search(Query("quick").scorer("TFIDF.DOCNORM").with_scores())
+        assert 0.14285714285714285 == res["results"][0]["score"]
+        res = client.ft().search(Query("quick").scorer("BM25").with_scores())
+        assert 0.22471909420069797 == res["results"][0]["score"]
+        res = client.ft().search(Query("quick").scorer("DISMAX").with_scores())
+        assert 2.0 == res["results"][0]["score"]
+        res = client.ft().search(Query("quick").scorer("DOCSCORE").with_scores())
+        assert 1.0 == res["results"][0]["score"]
+        res = client.ft().search(Query("quick").scorer("HAMMING").with_scores())
+        assert 0.0 == res["results"][0]["score"]
+
+
 @pytest.mark.redismod
 def test_get(client):
     client.ft().create_index((TextField("f1"), TextField("f2")))
@@ -2605,9 +2679,8 @@ def test_search_missing_fields(client):
         },
     )
 
-    with pytest.raises(redis.exceptions.ResponseError) as e:
+    with pytest.raises(redis.exceptions.ResponseError):
         client.ft().search(Query("ismissing(@title)").return_field("id").no_content())
-    assert "to be defined with 'INDEXMISSING'" in e.value.args[0]
 
     res = client.ft().search(
         Query("ismissing(@features)").return_field("id").no_content()
@@ -2813,6 +2886,12 @@ def test_search_query_with_different_dialects(client):
         assert res["total_results"] == 0
 
 
+@pytest.mark.redismod
+@skip_if_server_version_lt("7.9.0")
+def test_info_exposes_search_info(client):
+    assert len(client.info("search")) > 0
+
+
 def _assert_search_result(client, result, expected_doc_ids):
     """
     Make sure the result of a geo search is as expected, taking into account the RESP