Fix: failing p3 test for SDK/HTTP APIs (#13062)

### What problem does this PR solve?

Adjust highlight parsing, add row-count SQL override, tweak retrieval
thresholding, and update tests with engine-aware skips/utilities.

### Type of change

- [x] Bug Fix (non-breaking change which fixes an issue)

test/testcases/test_sdk_api/test_chunk_management_within_dataset/test_list_chunks.py

@@ -18,6 +18,7 @@ from concurrent.futures import ThreadPoolExecutor, as_completed
 
 import pytest
 from common import batch_add_chunks
+from utils.engine_utils import get_doc_engine
 
 
 class TestChunksList:
@@ -84,6 +85,12 @@ class TestChunksList:
     )
     def test_keywords(self, add_chunks, params, expected_page_size):
         _, document, _ = add_chunks
+        if params.get("keywords") == "ragflow":
+            doc_engine = get_doc_engine(document.rag)
+            if doc_engine == "infinity" and expected_page_size == 1:
+                pytest.skip("issues/6509")
+            if doc_engine != "infinity" and expected_page_size == 5:
+                pytest.skip("issues/6509")
         chunks = document.list_chunks(**params)
         assert len(chunks) == expected_page_size, str(chunks)
 
@@ -99,6 +106,8 @@ class TestChunksList:
     )
     def test_id(self, add_chunks, chunk_id, expected_page_size, expected_message):
         _, document, chunks = add_chunks
+        if callable(chunk_id) and get_doc_engine(document.rag) == "infinity":
+            pytest.skip("issues/6499")
         chunk_ids = [chunk.id for chunk in chunks]
         if callable(chunk_id):
             params = {"id": chunk_id(chunk_ids)}

test/testcases/test_sdk_api/test_chunk_management_within_dataset/test_retrieval_chunks.py

@@ -18,6 +18,8 @@ from concurrent.futures import ThreadPoolExecutor, as_completed
 
 import pytest
 
+DOC_ENGINE = (os.getenv("DOC_ENGINE") or "").lower()
+
 
 class TestChunksRetrieval:
     @pytest.mark.p1
@@ -159,25 +161,25 @@ class TestChunksRetrieval:
                 {"top_k": 1},
                 4,
                 "",
-                marks=pytest.mark.skipif(os.getenv("DOC_ENGINE") in ["infinity", "opensearch"], reason="Infinity"),
+                marks=pytest.mark.skipif(DOC_ENGINE in ["infinity", "opensearch"], reason="Infinity"),
             ),
             pytest.param(
                 {"top_k": 1},
                 1,
                 "",
-                marks=pytest.mark.skipif(os.getenv("DOC_ENGINE") in [None, "opensearch", "elasticsearch"], reason="elasticsearch"),
+                marks=pytest.mark.skipif(DOC_ENGINE in ["", "opensearch", "elasticsearch"], reason="elasticsearch"),
             ),
             pytest.param(
                 {"top_k": -1},
                 4,
                 "must be greater than 0",
-                marks=pytest.mark.skipif(os.getenv("DOC_ENGINE") in ["infinity", "opensearch"], reason="Infinity"),
+                marks=pytest.mark.skipif(DOC_ENGINE in ["infinity", "opensearch"], reason="Infinity"),
             ),
             pytest.param(
                 {"top_k": -1},
                 4,
                 "3014",
-                marks=pytest.mark.skipif(os.getenv("DOC_ENGINE") in [None, "opensearch", "elasticsearch"], reason="elasticsearch"),
+                marks=pytest.mark.skipif(DOC_ENGINE in ["", "opensearch", "elasticsearch"], reason="elasticsearch"),
             ),
             pytest.param(
                 {"top_k": "a"},

test/testcases/test_sdk_api/test_dataset_mangement/test_update_dataset.py

@@ -25,6 +25,7 @@ from utils import encode_avatar
 from utils.file_utils import create_image_file
 from utils.hypothesis_utils import valid_names
 from configs import DEFAULT_PARSER_CONFIG
+from utils.engine_utils import get_doc_engine
 
 class TestRquest:
     @pytest.mark.p2
@@ -332,6 +333,8 @@ class TestDatasetUpdate:
     @pytest.mark.p2
     @pytest.mark.parametrize("pagerank", [0, 50, 100], ids=["min", "mid", "max"])
     def test_pagerank(self, client, add_dataset_func, pagerank):
+        if get_doc_engine(client) == "infinity":
+            pytest.skip("#8208")
         dataset = add_dataset_func
         dataset.update({"pagerank": pagerank})
         assert dataset.pagerank == pagerank, str(dataset)
@@ -342,6 +345,8 @@ class TestDatasetUpdate:
     @pytest.mark.skipif(os.getenv("DOC_ENGINE") == "infinity", reason="#8208")
     @pytest.mark.p2
     def test_pagerank_set_to_0(self, client, add_dataset_func):
+        if get_doc_engine(client) == "infinity":
+            pytest.skip("#8208")
         dataset = add_dataset_func
         dataset.update({"pagerank": 50})
         assert dataset.pagerank == 50, str(dataset)
@@ -358,6 +363,8 @@ class TestDatasetUpdate:
     @pytest.mark.skipif(os.getenv("DOC_ENGINE") != "infinity", reason="#8208")
     @pytest.mark.p2
     def test_pagerank_infinity(self, client, add_dataset_func):
+        if get_doc_engine(client) != "infinity":
+            pytest.skip("#8208")
         dataset = add_dataset_func
         with pytest.raises(Exception) as exception_info:
             dataset.update({"pagerank": 50})

test/testcases/test_sdk_api/test_memory_management/test_create_memory.py

@@ -81,6 +81,7 @@ class TestMemoryCreate:
 
     @pytest.mark.p2
     @given(name=valid_names())
+    @settings(deadline=None)
     def test_type_invalid(self, client, name):
         payload = {
             "name": name,

test/testcases/test_sdk_api/test_message_management/test_list_message.py

@@ -19,6 +19,7 @@ import random
 import pytest
 from ragflow_sdk import RAGFlow, Memory
 from configs import INVALID_API_TOKEN, HOST_ADDRESS
+from utils.engine_utils import get_doc_engine
 
 
 class TestAuthorization:
@@ -88,6 +89,8 @@ class TestMessageList:
     @pytest.mark.p2
     @pytest.mark.skipif(os.getenv("DOC_ENGINE") == "infinity", reason="Not support.")
     def test_search_keyword(self, client):
+        if get_doc_engine(client) == "infinity":
+            pytest.skip("Not support.")
         memory_id = self.memory_id
         session_ids = self.session_ids
         session_id = random.choice(session_ids)