revery ruff

Author: satvshr

openml/_api_calls.py

@@ -362,7 +362,7 @@ def _send_request(  # noqa: C901, PLR0912
     files: FILE_ELEMENTS_TYPE | None = None,
     md5_checksum: str | None = None,
 ) -> requests.Response:
-    n_retries = 1
+    n_retries = max(1, config.connection_n_retries)
 
     response: requests.Response | None = None
     delay_method = _human_delay if config.retry_policy == "human" else _robot_delay

openml/config.py

@@ -157,7 +157,7 @@ def _resolve_default_cache_dir() -> Path:
     "cachedir": _resolve_default_cache_dir(),
     "avoid_duplicate_runs": False,
     "retry_policy": "human",
-    "connection_n_retries": 1,
+    "connection_n_retries": 5,
     "show_progress": False,
 }
 

tests/conftest.py

@@ -297,8 +297,8 @@ def with_test_cache(test_files_directory, request):
     openml.config.set_root_cache_directory(_root_cache_directory)
     if tmp_cache.exists():
         shutil.rmtree(tmp_cache)
+        
 
- 
 @pytest.fixture
 def static_cache_dir():
     return Path(__file__).parent / "files" 
@@ -308,4 +308,4 @@ def workdir(tmp_path):
     original_cwd = Path.cwd()
     os.chdir(tmp_path)
     yield tmp_path
-    os.chdir(original_cwd)
+    os.chdir(original_cwd)

tests/test_datasets/test_dataset_functions.py

@@ -406,7 +406,6 @@ def test__download_minio_file_works_with_bucket_subdirectory(self):
             file_destination
         ), "_download_minio_file can download from subdirectories"
 
-
     @mock.patch("openml._api_calls._download_minio_file")
     @pytest.mark.test_server()
     def test__get_dataset_parquet_is_cached(self, patch):
@@ -530,7 +529,10 @@ def test_deletion_of_cache_dir_faulty_download(self, patch):
         datasets_cache_dir = os.path.join(openml.config.get_cache_directory(), "datasets")
         assert len(os.listdir(datasets_cache_dir)) == 0
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.test_server()
     def test_publish_dataset(self):
         arff_file_path = self.static_cache_dir / "org" / "openml" / "test" / "datasets" / "2" / "dataset.arff"
@@ -690,7 +692,10 @@ def test_attributes_arff_from_df_unknown_dtype(self):
             with pytest.raises(ValueError, match=err_msg):
                 attributes_arff_from_df(df)
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.test_server()
     def test_create_dataset_numpy(self):
         data = np.array([[1, 2, 3], [1.2, 2.5, 3.8], [2, 5, 8], [0, 1, 0]]).T
@@ -725,7 +730,10 @@ def test_create_dataset_numpy(self):
         ), "Uploaded arff does not match original one"
         assert _get_online_dataset_format(dataset.id) == "arff", "Wrong format for dataset"
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.test_server()
     def test_create_dataset_list(self):
         data = [
@@ -781,7 +789,10 @@ def test_create_dataset_list(self):
         ), "Uploaded ARFF does not match original one"
         assert _get_online_dataset_format(dataset.id) == "arff", "Wrong format for dataset"
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.test_server()
     def test_create_dataset_sparse(self):
         # test the scipy.sparse.coo_matrix
@@ -930,7 +941,10 @@ def test_get_online_dataset_format(self):
             dataset_id
         ), "The format of the ARFF files is different"
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.test_server()
     def test_create_dataset_pandas(self):
         data = [
@@ -1156,7 +1170,10 @@ def test_ignore_attributes_dataset(self):
                 paper_url=paper_url,
             )
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.test_server()
     def test_publish_fetch_ignore_attribute(self):
         """Test to upload and retrieve dataset and check ignore_attributes"""
@@ -1276,7 +1293,10 @@ def test_create_dataset_row_id_attribute_error(self):
                 paper_url=paper_url,
             )
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.test_server()
     def test_create_dataset_row_id_attribute_inference(self):
         # meta-information
@@ -1445,7 +1465,10 @@ def test_data_edit_non_critical_field(self):
         edited_dataset = openml.datasets.get_dataset(did)
         assert edited_dataset.description == desc
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.test_server()
     def test_data_edit_critical_field(self):
         # Case 2
@@ -1498,7 +1521,10 @@ def test_data_edit_requires_valid_dataset(self):
             description="xor operation dataset",
         )
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.test_server()
     def test_data_edit_cannot_edit_critical_field_if_dataset_has_task(self):
         # Need to own a dataset to be able to edit meta-data
@@ -1551,7 +1577,6 @@ def test_data_fork(self):
             data_id=999999,
         )
 
-
     @pytest.mark.production_server()
     def test_list_datasets_with_high_size_parameter(self):
         # Testing on prod since concurrent deletion of uploded datasets make the test fail

tests/test_flows/test_flow.py

@@ -5,6 +5,7 @@
 import copy
 import hashlib
 import re
+import os
 import time
 from packaging.version import Version
 from unittest import mock
@@ -33,7 +34,6 @@
 from openml.testing import SimpleImputer, TestBase
 
 
-
 class TestFlow(TestBase):
     _multiprocess_can_split_ = True
 
@@ -180,7 +180,10 @@ def test_to_xml_from_xml(self):
         openml.flows.functions.assert_flows_equal(new_flow, flow)
         assert new_flow is not flow
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_publish_flow(self):
@@ -223,7 +226,10 @@ def test_publish_existing_flow(self, flow_exists_mock):
             f"collected from {__file__.split('/')[-1]}: {flow.flow_id}",
         )
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_publish_flow_with_similar_components(self):
@@ -275,7 +281,10 @@ def test_publish_flow_with_similar_components(self):
         TestBase._mark_entity_for_removal("flow", flow3.flow_id, flow3.name)
         TestBase.logger.info(f"collected from {__file__.split('/')[-1]}: {flow3.flow_id}")
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_semi_legal_flow(self):
@@ -386,7 +395,10 @@ def get_sentinel():
         flow_id = openml.flows.flow_exists(name, version)
         assert not flow_id
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_existing_flow_exists(self):
@@ -428,7 +440,10 @@ def test_existing_flow_exists(self):
             )
             assert downloaded_flow_id == flow.flow_id
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_sklearn_to_upload_to_flow(self):

tests/test_flows/test_flow_functions.py

@@ -12,6 +12,7 @@
 from unittest import mock
 from unittest.mock import patch
 
+import os
 import pandas as pd
 import pytest
 import requests
@@ -309,7 +310,10 @@ def test_get_flow1(self):
         flow = openml.flows.get_flow(1)
         assert flow.external_version is None
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_get_flow_reinstantiate_model(self):
@@ -393,7 +397,10 @@ def test_get_flow_reinstantiate_flow_not_strict_pre_023(self):
         assert flow.flow_id is None
         assert "sklearn==0.19.1" not in flow.dependencies
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_get_flow_id(self):

tests/test_openml/test_api_calls.py

@@ -7,6 +7,7 @@
 
 import minio
 import pytest
+import os
 
 import openml
 from openml.config import ConfigurationForExamples
@@ -20,7 +21,10 @@ def test_too_long_uri(self):
         with pytest.raises(openml.exceptions.OpenMLServerError, match="URI too long!"):
             openml.datasets.list_datasets(data_id=list(range(10000)))
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @unittest.mock.patch("time.sleep")
     @unittest.mock.patch("requests.Session")
     @pytest.mark.test_server()

tests/test_runs/test_run.py

@@ -118,7 +118,10 @@ def _check_array(array, type_):
         else:
             assert run_prime_trace_content is None
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_to_from_filesystem_vanilla(self):
@@ -154,7 +157,10 @@ def test_to_from_filesystem_vanilla(self):
             f"collected from {__file__.split('/')[-1]}: {run_prime.run_id}",
         )
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.flaky()
     @pytest.mark.test_server()
@@ -191,7 +197,10 @@ def test_to_from_filesystem_search(self):
             f"collected from {__file__.split('/')[-1]}: {run_prime.run_id}",
         )
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_to_from_filesystem_no_model(self):
@@ -298,7 +307,10 @@ def assert_run_prediction_data(task, run, model):
             assert_method(y_pred, saved_y_pred)
             assert_method(y_test, saved_y_test)
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_publish_with_local_loaded_flow(self):
@@ -343,7 +355,10 @@ def test_publish_with_local_loaded_flow(self):
             assert openml.flows.flow_exists(flow.name, flow.external_version)
             openml.runs.get_run(loaded_run.run_id)
 
-    @pytest.mark.skip(reason="Pending resolution of #1657")
+    @pytest.mark.skipif(
+        os.getenv("OPENML_USE_LOCAL_SERVICES") == "true",
+        reason="Pending resolution of #1657",
+    )
     @pytest.mark.sklearn()
     @pytest.mark.test_server()
     def test_offline_and_online_run_identical(self):