TEST-modin-project#2290: Apply suggestions from code review

Co-authored-by: Anatoly Myachev <45976948+anmyachev@users.noreply.github.com>

TEST-modin-project#2290: addressing review comments

Signed-off-by: Alexander Myskov <alexander.myskov@intel.com>

.github/workflows/ci.yml

@@ -297,7 +297,7 @@ jobs:
       - shell: bash -l {0}
         run: bash <(curl -s https://codecov.io/bash)
 
-  test-all-except-io:
+  test-ubuntu-all-except-io:
     needs: [lint-commit, lint-flake8, lint-black, test-api, test-headers]
     runs-on: ubuntu-latest
     strategy:
@@ -364,7 +364,7 @@ jobs:
       - shell: bash -l {0}
         run: bash <(curl -s https://codecov.io/bash)
 
-  test-io:
+  test-ubuntu-io:
     needs: [lint-commit, lint-flake8, lint-black, test-api, test-headers]
     runs-on: ubuntu-latest
     strategy:

.github/workflows/push.yml

@@ -168,7 +168,7 @@ jobs:
       - shell: bash -l {0}
         run: bash <(curl -s https://codecov.io/bash)
 
-  test-all-except-io:
+  test-ubuntu-all-except-io:
     needs: prepare-cache
     runs-on: ubuntu-latest
     strategy:
@@ -235,7 +235,7 @@ jobs:
       - shell: bash -l {0}
         run: bash <(curl -s https://codecov.io/bash)
 
-  test-io:
+  test-ubuntu-io:
     needs: [lint-commit, lint-flake8, lint-black, test-api, test-headers]
     runs-on: ubuntu-latest
     strategy:

modin/pandas/test/test_io.py

@@ -71,12 +71,6 @@
 if not os.path.exists(IO_OPS_DATA_DIR):
     os.mkdir(IO_OPS_DATA_DIR)
 
-str_initial_spaces = """col1,col2,col3,col4
-five,  six,  seven,  eight
-    five,    six,    seven,    eight
-five, six,  seven,   eight
-"""
-
 
 @pytest.fixture
 def make_parquet_file():
@@ -600,36 +594,69 @@ def test_read_csv_col_handling(
             },
         ],
     )
+    @pytest.mark.parametrize("skipfooter", [0, 10])
+    def test_read_csv_parsing_1(
+        self,
+        make_csv_file,
+        request,
+        dtype,
+        engine,
+        converters,
+        skipfooter,
+    ):
+        if request.config.getoption("--simulate-cloud").lower() != "off":
+            pytest.xfail(
+                "The reason of tests fail in `cloud` mode is unknown for now - issue #2340"
+            )
+        kwargs = {
+            "dtype": dtype,
+            "engine": engine,
+            "converters": converters,
+            "skipfooter": skipfooter,
+        }
+
+        unique_name = get_unique_filename("test_read_csv_parsing", kwargs)
+        make_csv_file(
+            filename=unique_name,
+        )
+        if kwargs["dtype"]:
+            kwargs["dtype"] = {
+                col: "object" for col in pandas.read_csv(unique_name, nrows=1).columns
+            }
+
+        eval_io(
+            filepath_or_buffer=unique_name,
+            fn_name="read_csv",
+            check_exception_type=None,  # issue #2320
+            raising_exceptions=None,
+            check_kwargs_callable=not callable(converters),
+            **kwargs,
+        )
+
     @pytest.mark.parametrize("true_values", [["Yes"], ["Yes", "true"], None])
     @pytest.mark.parametrize("false_values", [["No"], ["No", "false"], None])
     @pytest.mark.parametrize("skiprows", [2, lambda x: x % 2])
     @pytest.mark.parametrize("skipfooter", [0, 10])
     @pytest.mark.parametrize("nrows", [123, None])
     @pytest.mark.parametrize("names", [["c1", "c2", "c3", "c4"], None])
-    def test_read_csv_parsing(
+    def test_read_csv_parsing_2(
         self,
         make_csv_file,
         request,
-        dtype,
-        engine,
-        converters,
         true_values,
         false_values,
         skiprows,
         skipfooter,
         nrows,
         names,
     ):
-        if nrows and (false_values or true_values):
+        if nrows and (false_values or true_values) and Engine.get() != "Python":
             pytest.xfail("modin and pandas dataframes differs - issue #2446")
         if request.config.getoption("--simulate-cloud").lower() != "off":
             pytest.xfail(
                 "The reason of tests fail in `cloud` mode is unknown for now - issue #2340"
             )
         kwargs = {
-            "dtype": dtype,
-            "engine": engine,
-            "converters": converters,
             "true_values": true_values,
             "false_values": false_values,
             "skiprows": skiprows,
@@ -645,22 +672,24 @@ def test_read_csv_parsing(
             if true_values or false_values
             else None,
         )
-        if kwargs["dtype"]:
-            kwargs["dtype"] = {
-                col: "object" for col in pandas.read_csv(unique_name, nrows=1).columns
-            }
 
         eval_io(
             filepath_or_buffer=unique_name,
             fn_name="read_csv",
             check_exception_type=None,  # issue #2320
             raising_exceptions=None,
-            check_kwargs_callable=not (callable(skiprows) or callable(converters)),
+            check_kwargs_callable=not callable(skiprows),
             **kwargs,
         )
 
     def test_read_csv_skipinitialspace(self, make_csv_file):
         unique_filename = get_unique_filename("test_read_csv_skipinitialspace")
+        str_initial_spaces = (
+            "col1,col2,col3,col4\n"
+            "five,  six,  seven,  eight\n"
+            "    five,    six,    seven,    eight\n"
+            "five, six,  seven,   eight\n"
+        )
 
         eval_io_from_str(str_initial_spaces, unique_filename, skipinitialspace=True)
 

modin/pandas/test/utils.py

@@ -868,36 +868,6 @@ def generate_none_dfs():
     return df, df2
 
 
-def make_dict_hash(dict_to_hash):
-    """Makes a hash from a dictionary considering nested
-    dictionaries, lists, sets or callables.
-
-    Parameters
-    ----------
-    dict_to_hash: dict
-        Dictionary that should be hashed.
-
-    Returns
-    -------
-        Dictionary hash.
-    """
-
-    def _make_hash(obj):
-        if isinstance(obj, (set, list, dict)):
-            object_hash = hash(frozenset(obj))
-        elif callable(obj):
-            object_hash = hash(obj.__name__)
-        else:
-            # conversion to str is done because hash(5) == hash(5.0)
-            object_hash = hash(obj if isinstance(obj, str) else str(obj))
-
-        return object_hash
-
-    # process unhashed values from dict_to_hash
-    new_dict = {key: _make_hash(value) for key, value in dict_to_hash.items()}
-    return hash(frozenset(new_dict))
-
-
 def get_unique_filename(
     test_name: str,
     kwargs: dict = {},
@@ -954,9 +924,10 @@ def get_unique_filename(
         )
         return os.path.join(data_dir, parameters_values + suffix_part + f".{extension}")
     else:
-        unique_id = make_dict_hash(kwargs)
+        import uuid
+
         return os.path.join(
-            data_dir, (test_name + f"_{str(unique_id)}" + suffix_part + f".{extension}")
+            data_dir, (uuid.uuid1().hex + suffix_part + f".{extension}")
         )