Add the string testcase (#19287)

Signed-off-by: Njingkl <jingjing.jia@zilliz.com>
Signed-off-by: Njingkl <jingjing.jia@zilliz.com>

tests/python_client/testcases/test_delete.py

@@ -1745,3 +1745,36 @@ class TestDeleteString(TestcaseBase):
                            check_task=CheckTasks.check_query_results, check_items={'exp_res': res, 'primary_field': ct.default_string_field_name, 'with_vec': True})
         collection_w.search(data=[df_new[ct.default_float_vec_field_name][0]], anns_field=ct.default_float_vec_field_name,
                             param=default_search_params, limit=1)
+
+    @pytest.mark.tags(CaseLabel.L1)
+    def test_delete_with_string_field_is_empty(self):
+        """
+        target: test delete with string field is empty
+        method: 1.string field is PK, insert empty data
+                2.delete ids from collection
+                3.load and query with id
+        expected: No query result
+        """
+        # create collection, insert data without flush
+        schema = cf.gen_string_pk_default_collection_schema()
+        collection_w = self.init_collection_wrap(name=cf.gen_unique_str(prefix), schema=schema)
+
+        nb = 3000
+        df = cf.gen_default_list_data(nb)
+        df[2] = [""for _ in range(nb)] 
+
+        collection_w.insert(df)
+        collection_w.load()
+        assert collection_w.num_entities == nb
+
+        # delete
+        string_expr = "varchar in [\"\", \"\"]"
+        del_res, _ = collection_w.delete(string_expr)
+        assert del_res.delete_count == 2
+
+        # load and query with id
+        collection_w.load()
+        collection_w.query(string_expr, check_task=CheckTasks.check_query_empty)
+
+
+       

tests/python_client/testcases/test_index.py

@@ -1443,3 +1443,25 @@ class TestIndexString(TestcaseBase):
         collection_w.create_index(default_string_field_name, default_string_index_params, index_name=index_name2)
         collection_w.drop_index(index_name=index_name2)
         assert len(collection_w.indexes) == 0
+    
+    @pytest.mark.tags(CaseLabel.L1)
+    def test_index_with_string_field_empty(self):
+        """
+        target: test drop index with string field
+        method: 1.create collection and insert data
+                2.create index and uses collection.drop_index () drop index
+        expected: drop index successfully
+        """
+        c_name = cf.gen_unique_str(prefix)
+        collection_w = self.init_collection_wrap(name=c_name)
+
+        nb = 3000
+        data = cf.gen_default_list_data(nb)
+        data[2] = [""for _ in range(nb)] 
+        collection_w.insert(data=data)
+
+        collection_w.create_index(default_string_field_name, default_string_index_params, index_name=index_name2)
+        assert collection_w.has_index(index_name=index_name2)[0] == True
+
+        collection_w.drop_index(index_name=index_name2)
+        assert len(collection_w.indexes) == 0
\ No newline at end of file

tests/python_client/testcases/test_insert.py

@@ -1217,3 +1217,36 @@ class TestInsertString(TestcaseBase):
         data[2] = [" "for _ in range(nb)] 
         collection_w.insert(data)
         assert collection_w.num_entities == nb
+
+    @pytest.mark.tags(CaseLabel.L1)
+    def test_insert_string_field_empty(self):
+        """
+        target: test create collection with string field 
+        method: 1.create a collection  
+                2.Insert string field with empty
+        expected: Insert successfully
+        """
+        c_name = cf.gen_unique_str(prefix)
+        collection_w = self.init_collection_wrap(name=c_name)
+        nb = 1000
+        data = cf.gen_default_list_data(nb)
+        data[2] = [""for _ in range(nb)] 
+        collection_w.insert(data)
+        assert collection_w.num_entities == nb
+
+    @pytest.mark.tags(CaseLabel.L1)
+    def test_insert_string_field_is_pk_and_empty(self):
+        """
+        target: test create collection with string field is primary
+        method: 1.create a collection  
+                2.Insert string field with empty, string field is pk
+        expected: Insert successfully
+        """
+        c_name = cf.gen_unique_str(prefix)
+        schema = cf.gen_string_pk_default_collection_schema()
+        collection_w = self.init_collection_wrap(name=c_name, schema=schema)
+        nb = 1000
+        data = cf.gen_default_list_data(nb)
+        data[2] = [""for _ in range(nb)] 
+        collection_w.insert(data)
+        assert collection_w.num_entities == nb

tests/python_client/testcases/test_query.py

@@ -1384,3 +1384,65 @@ class  TestqueryString(TestcaseBase):
                            check_items={exp_res: df_dict_list,
                                         "primary_field": default_int_field_name,
                                         "with_vec": True})
+
+    @pytest.mark.tags(CaseLabel.L2)
+    def test_query_string_field_pk_is_empty(self):
+        """
+        target: test query with string expr and string field is primary
+        method: create collection , string field is primary
+                collection load and insert empty data with string field
+                collection query uses string expr in string field
+        expected: query successfully
+        """
+        # 1. create a collection
+        schema = cf.gen_string_pk_default_collection_schema()
+        collection_w = self.init_collection_wrap(cf.gen_unique_str(prefix), schema=schema)
+        
+        collection_w.load()
+        
+        nb = 3000
+        df = cf.gen_default_list_data(nb)
+        df[2] = [""for _ in range(nb)] 
+
+        collection_w.insert(df)
+        assert collection_w.num_entities == nb
+
+        
+        string_exp = "varchar >= \"\""
+        output_fields = [default_int_field_name, default_float_field_name, default_string_field_name]
+        res, _ = collection_w.query(string_exp, output_fields=output_fields)
+
+        assert len(res) == 1
+
+
+    @pytest.mark.tags(CaseLabel.L2)
+    def test_query_string_field_not_primary_is_empty(self):
+        """
+        target: test query with string expr and string field is not primary
+        method: create collection , string field is primary
+                collection load and insert empty data with string field
+                collection query uses string expr in string field
+        expected: query successfully
+        """
+        # 1.  create a collection
+        collection_w, vectors = self.init_collection_general(prefix, insert_data=False)[0:2]
+        
+        nb = 3000
+        df = cf.gen_default_list_data(nb)
+        df[2] = [""for _ in range(nb)] 
+
+        collection_w.insert(df)
+        assert collection_w.num_entities == nb
+        collection_w.load()
+        
+        collection_w.create_index(ct.default_float_vec_field_name, default_index_params)
+        assert collection_w.has_index()[0]
+
+        
+        output_fields = [default_int_field_name, default_float_field_name, default_string_field_name]
+        
+        expr = "varchar == \"\""
+        res, _ = collection_w.query(expr, output_fields=output_fields)
+
+        assert len(res) == nb
+        

tests/python_client/testcases/test_search.py

@@ -3550,3 +3550,88 @@ class  TestsearchString(TestcaseBase):
                                          "limit": default_limit,
                                          "_async": _async}
                             )
+
+    @pytest.mark.tags(CaseLabel.L2)
+    def test_search_string_field_is_primary_insert_empty(self, _async):
+        """
+        target: test search with string expr and string field is primary
+        method: create collection ,string field is primary
+                collection load and insert data
+                collection search uses string expr in string field 
+        expected: Search successfully
+        """
+        # 1. initialize with data
+        collection_w, _, _, _ = \
+            self.init_collection_general(prefix, False, primary_field=ct.default_string_field_name)[0:4]
+
+        nb = 3000
+        data = cf.gen_default_list_data(nb)
+        data[2] = [""for _ in range(nb)] 
+        collection_w.insert(data=data)
+        
+        collection_w.load()
+         
+        
+        search_string_exp = "varchar >= \"\""
+        limit =1
+
+        # 2. search
+        log.info("test_search_string_field_is_primary_true: searching collection %s" % collection_w.name)
+        vectors = [[random.random() for _ in range(default_dim)] for _ in range(default_nq)]
+        output_fields = [default_string_field_name, default_float_field_name]
+        collection_w.search(vectors[:default_nq], default_search_field,
+                            default_search_params, limit,
+                            search_string_exp, 
+                            output_fields=output_fields,
+                            _async=_async,
+                            travel_timestamp=0,
+                            check_task=CheckTasks.check_search_results,
+                            check_items={"nq": default_nq,
+                                         "limit": limit,
+                                         "_async": _async})
+
+
+    @pytest.mark.tags(CaseLabel.L2)
+    def test_search_string_field_not_primary_is_empty(self, _async):
+        """
+        target: test search with string expr and string field is not primary
+        method: create collection and insert data 
+                create index and collection load
+                collection search uses string expr in string field, string field is not primary
+        expected: Search successfully
+        """
+        # 1. initialize with data
+        collection_w, _, _, insert_ids = \
+            self.init_collection_general(prefix, False, primary_field=ct.default_int64_field_name)[0:4]
+
+        nb = 3000
+        data = cf.gen_default_list_data(nb)
+        data[2] = [""for _ in range(nb)] 
+
+        collection_w.insert(data)
+        assert collection_w.num_entities == nb
+
+        # 2. create index
+        index_param = {"index_type": "IVF_FLAT", "metric_type": "L2", "params": {"nlist": 100}}
+        collection_w.create_index("float_vector", index_param)
+        collection_w.load()
+
+
+        search_string_exp = "varchar >= \"\""
+        
+        # 3. search
+        log.info("test_search_string_field_not_primary: searching collection %s" % collection_w.name)
+        vectors = [[random.random() for _ in range(default_dim)] for _ in range(default_nq)]
+        output_fields = [default_string_field_name, default_float_field_name]
+        collection_w.search(vectors[:default_nq], default_search_field,
+                            default_search_params, default_limit, 
+                            search_string_exp,
+                            output_fields=output_fields,
+                            _async=_async,
+                            travel_timestamp=0,
+                            check_task=CheckTasks.check_search_results,
+                            check_items={"nq": default_nq,
+                                         "ids": insert_ids,
+                                         "limit": default_limit,
+                                         "_async": _async})
+