From bf6e0001e00bff1a0933dd53588bbc45b312cc76 Mon Sep 17 00:00:00 2001
From: Satvik Mishra <112589278+satvshr@users.noreply.github.com>
Date: Tue, 30 Dec 2025 02:07:17 +0530
Subject: [PATCH 1/6] changes made

---
 openml/_api_calls.py |   4 +-
 openml/config.py     | 252 +++++++++++++++++++++----------------------
 2 files changed, 127 insertions(+), 129 deletions(-)

diff --git a/openml/_api_calls.py b/openml/_api_calls.py
index 81296b3da..12567ac7a 100644
--- a/openml/_api_calls.py
+++ b/openml/_api_calls.py
@@ -71,7 +71,7 @@ def resolve_env_proxies(url: str) -> str | None:
 
 
 def _create_url_from_endpoint(endpoint: str) -> str:
-    url = config.server
+    url = config._config.server
     if not url.endswith("/"):
         url += "/"
     url += endpoint
@@ -301,7 +301,7 @@ def _file_id_to_url(file_id: int, filename: str | None = None) -> str:
     Presents the URL how to download a given file id
     filename is optional
     """
-    openml_url = config.server.split("/api/")
+    openml_url = config._config.server.split("/api/")
     url = openml_url[0] + f"/data/download/{file_id!s}"
     if filename is not None:
         url += "/" + filename
diff --git a/openml/config.py b/openml/config.py
index cf66a6346..98a48a1c6 100644
--- a/openml/config.py
+++ b/openml/config.py
@@ -1,6 +1,7 @@
 """Store module level information like the API key, cache directory and the server"""
 
 # License: BSD 3-Clause
+# ruff: noqa: PLW0603
 from __future__ import annotations
 
 import configparser
@@ -11,10 +12,11 @@
 import shutil
 import warnings
 from contextlib import contextmanager
+from dataclasses import dataclass, replace
 from io import StringIO
 from pathlib import Path
-from typing import Any, Iterator, cast
-from typing_extensions import Literal, TypedDict
+from typing import Any, Iterator
+from typing_extensions import Literal
 from urllib.parse import urlparse
 
 logger = logging.getLogger(__name__)
@@ -27,19 +29,62 @@
 _TEST_SERVER_NORMAL_USER_KEY = "normaluser"
 
 
-class _Config(TypedDict):
-    apikey: str
-    server: str
-    cachedir: Path
-    avoid_duplicate_runs: bool
-    retry_policy: Literal["human", "robot"]
-    connection_n_retries: int
-    show_progress: bool
+# Default values (see also https://github.com/openml/OpenML/wiki/Client-API-Standards)
+_user_path = Path("~").expanduser().absolute()
+
+
+def _resolve_default_cache_dir() -> Path:
+    user_defined_cache_dir = os.environ.get(OPENML_CACHE_DIR_ENV_VAR)
+    if user_defined_cache_dir is not None:
+        return Path(user_defined_cache_dir)
+
+    if platform.system().lower() != "linux":
+        return _user_path / ".openml"
+
+    xdg_cache_home = os.environ.get("XDG_CACHE_HOME")
+    if xdg_cache_home is None:
+        return Path("~", ".cache", "openml")
+
+    # This is the proper XDG_CACHE_HOME directory, but
+    # we unfortunately had a problem where we used XDG_CACHE_HOME/org,
+    # we check heuristically if this old directory still exists and issue
+    # a warning if it does. There's too much data to move to do this for the user.
+
+    # The new cache directory exists
+    cache_dir = Path(xdg_cache_home) / "openml"
+    if cache_dir.exists():
+        return cache_dir
+
+    # The old cache directory *does not* exist
+    heuristic_dir_for_backwards_compat = Path(xdg_cache_home) / "org" / "openml"
+    if not heuristic_dir_for_backwards_compat.exists():
+        return cache_dir
+
+    root_dir_to_delete = Path(xdg_cache_home) / "org"
+    openml_logger.warning(
+        "An old cache directory was found at '%s'. This directory is no longer used by "
+        "OpenML-Python. To silence this warning you would need to delete the old cache "
+        "directory. The cached files will then be located in '%s'.",
+        root_dir_to_delete,
+        cache_dir,
+    )
+    return Path(xdg_cache_home)
+
+
+@dataclass(frozen=True)
+class OpenMLConfig:
+    apikey: str = ""
+    server: str = "https://www.openml.org/api/v1/xml"
+    cachedir: Path = _resolve_default_cache_dir()  # noqa: RUF009
+    avoid_duplicate_runs: bool = False
+    retry_policy: Literal["human", "robot"] = "human"
+    connection_n_retries: int = 5
+    show_progress: bool = False
 
 
 def _create_log_handlers(create_file_handler: bool = True) -> None:  # noqa: FBT001, FBT002
     """Creates but does not attach the log handlers."""
-    global console_handler, file_handler  # noqa: PLW0603
+    global console_handler, file_handler, _root_cache_directory  # noqa: PLW0602
     if console_handler is not None or file_handler is not None:
         logger.debug("Requested to create log handlers, but they are already created.")
         return
@@ -105,61 +150,22 @@ def set_file_log_level(file_output_level: int) -> None:
     _set_level_register_and_store(file_handler, file_output_level)
 
 
-# Default values (see also https://github.com/openml/OpenML/wiki/Client-API-Standards)
-_user_path = Path("~").expanduser().absolute()
-
-
-def _resolve_default_cache_dir() -> Path:
-    user_defined_cache_dir = os.environ.get(OPENML_CACHE_DIR_ENV_VAR)
-    if user_defined_cache_dir is not None:
-        return Path(user_defined_cache_dir)
-
-    if platform.system().lower() != "linux":
-        return _user_path / ".openml"
-
-    xdg_cache_home = os.environ.get("XDG_CACHE_HOME")
-    if xdg_cache_home is None:
-        return Path("~", ".cache", "openml")
+_config: OpenMLConfig = OpenMLConfig()
+_root_cache_directory: Path = _config.cachedir
 
-    # This is the proper XDG_CACHE_HOME directory, but
-    # we unfortunately had a problem where we used XDG_CACHE_HOME/org,
-    # we check heuristically if this old directory still exists and issue
-    # a warning if it does. There's too much data to move to do this for the user.
 
-    # The new cache directory exists
-    cache_dir = Path(xdg_cache_home) / "openml"
-    if cache_dir.exists():
-        return cache_dir
+def __getattr__(name: str) -> Any:
+    if hasattr(_config, name):
+        return getattr(_config, name)
+    raise AttributeError(f"module 'openml.config' has no attribute '{name}'")
 
-    # The old cache directory *does not* exist
-    heuristic_dir_for_backwards_compat = Path(xdg_cache_home) / "org" / "openml"
-    if not heuristic_dir_for_backwards_compat.exists():
-        return cache_dir
 
-    root_dir_to_delete = Path(xdg_cache_home) / "org"
-    openml_logger.warning(
-        "An old cache directory was found at '%s'. This directory is no longer used by "
-        "OpenML-Python. To silence this warning you would need to delete the old cache "
-        "directory. The cached files will then be located in '%s'.",
-        root_dir_to_delete,
-        cache_dir,
-    )
-    return Path(xdg_cache_home)
-
-
-_defaults: _Config = {
-    "apikey": "",
-    "server": "https://www.openml.org/api/v1/xml",
-    "cachedir": _resolve_default_cache_dir(),
-    "avoid_duplicate_runs": False,
-    "retry_policy": "human",
-    "connection_n_retries": 5,
-    "show_progress": False,
-}
-
-# Default values are actually added here in the _setup() function which is
-# called at the end of this module
-server = _defaults["server"]
+def __setattr__(name: str, value: Any) -> None:  # noqa: N807
+    global _config
+    if hasattr(_config, name):
+        _config = replace(_config, **{name: value})
+    else:
+        raise AttributeError(f"module 'openml.config' has no attribute '{name}'")
 
 
 def get_server_base_url() -> str:
@@ -172,23 +178,12 @@ def get_server_base_url() -> str:
     -------
     str
     """
-    domain, path = server.split("/api", maxsplit=1)
+    domain, _ = _config.server.split("/api", maxsplit=1)
     return domain.replace("api", "www")
 
 
-apikey: str = _defaults["apikey"]
-show_progress: bool = _defaults["show_progress"]
-# The current cache directory (without the server name)
-_root_cache_directory: Path = Path(_defaults["cachedir"])
-avoid_duplicate_runs = _defaults["avoid_duplicate_runs"]
-
-retry_policy: Literal["human", "robot"] = _defaults["retry_policy"]
-connection_n_retries: int = _defaults["connection_n_retries"]
-
-
 def set_retry_policy(value: Literal["human", "robot"], n_retries: int | None = None) -> None:
-    global retry_policy  # noqa: PLW0603
-    global connection_n_retries  # noqa: PLW0603
+    global _config
     default_retries_by_policy = {"human": 5, "robot": 50}
 
     if value not in default_retries_by_policy:
@@ -202,8 +197,11 @@ def set_retry_policy(value: Literal["human", "robot"], n_retries: int | None = N
     if isinstance(n_retries, int) and n_retries < 1:
         raise ValueError(f"`n_retries` is '{n_retries}' but must be positive.")
 
-    retry_policy = value
-    connection_n_retries = default_retries_by_policy[value] if n_retries is None else n_retries
+    _config = replace(
+        _config,
+        retry_policy=value,
+        connection_n_retries=(default_retries_by_policy[value] if n_retries is None else n_retries),
+    )
 
 
 class ConfigurationForExamples:
@@ -222,24 +220,30 @@ def start_using_configuration_for_example(cls) -> None:
         To configuration as was before this call is stored, and can be recovered
         by using the `stop_use_example_configuration` method.
         """
-        global server  # noqa: PLW0603
-        global apikey  # noqa: PLW0603
+        global _config
 
-        if cls._start_last_called and server == cls._test_server and apikey == cls._test_apikey:
+        if (
+            cls._start_last_called
+            and _config.server == cls._test_server
+            and _config.apikey == cls._test_apikey
+        ):
             # Method is called more than once in a row without modifying the server or apikey.
             # We don't want to save the current test configuration as a last used configuration.
             return
 
-        cls._last_used_server = server
-        cls._last_used_key = apikey
+        cls._last_used_server = _config.server
+        cls._last_used_key = _config.apikey
         cls._start_last_called = True
 
         # Test server key for examples
-        server = cls._test_server
-        apikey = cls._test_apikey
+        _config = replace(
+            _config,
+            server=cls._test_server,
+            apikey=cls._test_apikey,
+        )
         warnings.warn(
-            f"Switching to the test server {server} to not upload results to the live server. "
-            "Using the test server may result in reduced performance of the API!",
+            f"Switching to the test server {_config.server} to not upload results to "
+            "the live server. Using the test server may result in reduced performance of the API!",
             stacklevel=2,
         )
 
@@ -254,11 +258,9 @@ def stop_using_configuration_for_example(cls) -> None:
                 "`start_use_example_configuration` must be called first.",
             )
 
-        global server  # noqa: PLW0603
-        global apikey  # noqa: PLW0603
+        global _config
+        _config = replace(_config, server=cls._test_server, apikey=cls._test_apikey)
 
-        server = cast(str, cls._last_used_server)
-        apikey = cast(str, cls._last_used_key)
         cls._start_last_called = False
 
 
@@ -327,7 +329,7 @@ def determine_config_file_path() -> Path:
     return config_dir / "config"
 
 
-def _setup(config: _Config | None = None) -> None:
+def _setup(config: dict[str, Any] | None = None) -> None:
     """Setup openml package. Called on first import.
 
     Reads the config file and sets up apikey, server, cache appropriately.
@@ -336,11 +338,8 @@ def _setup(config: _Config | None = None) -> None:
     openml.config.server = SOMESERVER
     We could also make it a property but that's less clear.
     """
-    global apikey  # noqa: PLW0603
-    global server  # noqa: PLW0603
-    global _root_cache_directory  # noqa: PLW0603
-    global avoid_duplicate_runs  # noqa: PLW0603
-    global show_progress  # noqa: PLW0603
+    global _config
+    global _root_cache_directory
 
     config_file = determine_config_file_path()
     config_dir = config_file.parent
@@ -358,19 +357,24 @@ def _setup(config: _Config | None = None) -> None:
     if config is None:
         config = _parse_config(config_file)
 
-    avoid_duplicate_runs = config["avoid_duplicate_runs"]
-    apikey = config["apikey"]
-    server = config["server"]
-    show_progress = config["show_progress"]
-    n_retries = int(config["connection_n_retries"])
+    _config = replace(
+        _config,
+        apikey=config["apikey"],
+        server=config["server"],
+        show_progress=config["show_progress"],
+        avoid_duplicate_runs=config["avoid_duplicate_runs"],
+        retry_policy=config["retry_policy"],
+        connection_n_retries=int(config["connection_n_retries"]),
+    )
 
-    set_retry_policy(config["retry_policy"], n_retries)
+    set_retry_policy(config["retry_policy"], _config.connection_n_retries)
 
     user_defined_cache_dir = os.environ.get(OPENML_CACHE_DIR_ENV_VAR)
     if user_defined_cache_dir is not None:
         short_cache_dir = Path(user_defined_cache_dir)
     else:
         short_cache_dir = Path(config["cachedir"])
+
     _root_cache_directory = short_cache_dir.expanduser().resolve()
 
     try:
@@ -389,29 +393,31 @@ def _setup(config: _Config | None = None) -> None:
 
 def set_field_in_config_file(field: str, value: Any) -> None:
     """Overwrites the `field` in the configuration file with the new `value`."""
-    if field not in _defaults:
-        raise ValueError(f"Field '{field}' is not valid and must be one of '{_defaults.keys()}'.")
+    global _config
+    if not hasattr(_config, field):
+        raise ValueError(
+            f"Field '{field}' is not valid and must be one of '{_config.__dict__.keys()}'."
+        )
 
-    # TODO(eddiebergman): This use of globals has gone too far
-    globals()[field] = value
+    _config = replace(_config, **{field: value})
     config_file = determine_config_file_path()
-    config = _parse_config(config_file)
+    existing = _parse_config(config_file)
     with config_file.open("w") as fh:
-        for f in _defaults:
+        for f in _config.__dict__:
             # We can't blindly set all values based on globals() because when the user
             # sets it through config.FIELD it should not be stored to file.
             # There doesn't seem to be a way to avoid writing defaults to file with configparser,
             # because it is impossible to distinguish from an explicitly set value that matches
             # the default value, to one that was set to its default because it was omitted.
-            value = globals()[f] if f == field else config.get(f)  # type: ignore
-            if value is not None:
-                fh.write(f"{f} = {value}\n")
+            v = value if f == field else existing.get(f)
+            if v is not None:
+                fh.write(f"{f} = {v}\n")
 
 
-def _parse_config(config_file: str | Path) -> _Config:
+def _parse_config(config_file: str | Path) -> dict[str, Any]:
     """Parse the config file, set up defaults."""
     config_file = Path(config_file)
-    config = configparser.RawConfigParser(defaults=_defaults)  # type: ignore
+    config = configparser.RawConfigParser(defaults=_config.__dict__)  # type: ignore
 
     # The ConfigParser requires a [SECTION_HEADER], which we do not expect in our config file.
     # Cheat the ConfigParser module by adding a fake section header
@@ -434,16 +440,8 @@ def _parse_config(config_file: str | Path) -> _Config:
     return configuration  # type: ignore
 
 
-def get_config_as_dict() -> _Config:
-    return {
-        "apikey": apikey,
-        "server": server,
-        "cachedir": _root_cache_directory,
-        "avoid_duplicate_runs": avoid_duplicate_runs,
-        "connection_n_retries": connection_n_retries,
-        "retry_policy": retry_policy,
-        "show_progress": show_progress,
-    }
+def get_config_as_dict() -> dict[str, Any]:
+    return _config.__dict__.copy()
 
 
 # NOTE: For backwards compatibility, we keep the `str`
@@ -467,7 +465,7 @@ def get_cache_directory() -> str:
         The current cache directory.
 
     """
-    url_suffix = urlparse(server).netloc
+    url_suffix = urlparse(_config.server).netloc
     reversed_url_suffix = os.sep.join(url_suffix.split(".")[::-1])  # noqa: PTH118
     return os.path.join(_root_cache_directory, reversed_url_suffix)  # noqa: PTH118
 
@@ -491,7 +489,7 @@ def set_root_cache_directory(root_cache_directory: str | Path) -> None:
     --------
     get_cache_directory
     """
-    global _root_cache_directory  # noqa: PLW0603
+    global _root_cache_directory
     _root_cache_directory = Path(root_cache_directory)
 
 
@@ -502,7 +500,7 @@ def set_root_cache_directory(root_cache_directory: str | Path) -> None:
 
 
 @contextmanager
-def overwrite_config_context(config: dict[str, Any]) -> Iterator[_Config]:
+def overwrite_config_context(config: dict[str, Any]) -> Iterator[dict[str, Any]]:
     """A context manager to temporarily override variables in the configuration."""
     existing_config = get_config_as_dict()
     merged_config = {**existing_config, **config}
@@ -515,10 +513,10 @@ def overwrite_config_context(config: dict[str, Any]) -> Iterator[_Config]:
 
 __all__ = [
     "get_cache_directory",
+    "get_config_as_dict",
     "set_root_cache_directory",
     "start_using_configuration_for_example",
     "stop_using_configuration_for_example",
-    "get_config_as_dict",
 ]
 
 _setup()

From 834782c105b5244095e20f17059c081b88634640 Mon Sep 17 00:00:00 2001
From: Satvik Mishra <112589278+satvshr@users.noreply.github.com>
Date: Tue, 30 Dec 2025 12:31:52 +0530
Subject: [PATCH 2/6] bug fixing

---
 examples/Advanced/datasets_tutorial.py        |  2 +-
 .../benchmark_with_optunahub.py               |  4 +--
 .../flow_id_tutorial.py                       |  2 +-
 openml/_api_calls.py                          | 12 ++++-----
 openml/cli.py                                 |  2 +-
 openml/config.py                              | 16 +++++++-----
 openml/runs/functions.py                      |  2 +-
 openml/testing.py                             | 14 +++++-----
 tests/conftest.py                             | 16 ++++++------
 tests/test_datasets/test_dataset_functions.py | 14 +++++-----
 tests/test_openml/test_config.py              | 26 +++++++++----------
 tests/test_utils/test_utils.py                |  2 +-
 12 files changed, 58 insertions(+), 54 deletions(-)

diff --git a/examples/Advanced/datasets_tutorial.py b/examples/Advanced/datasets_tutorial.py
index cc57686d0..3a4833206 100644
--- a/examples/Advanced/datasets_tutorial.py
+++ b/examples/Advanced/datasets_tutorial.py
@@ -139,7 +139,7 @@
 # only for the dataset owner. Further, critical fields cannot be edited if the dataset has any
 # tasks associated with it. To edit critical fields of a dataset (without tasks) owned by you,
 # configure the API key:
-# openml.config.apikey = 'FILL_IN_OPENML_API_KEY'
+# openml.config._config.apikey = 'FILL_IN_OPENML_API_KEY'
 # This example here only shows a failure when trying to work on a dataset not owned by you:
 
 # %%
diff --git a/examples/_external_or_deprecated/benchmark_with_optunahub.py b/examples/_external_or_deprecated/benchmark_with_optunahub.py
index ece3e7c40..c8f5f7b0c 100644
--- a/examples/_external_or_deprecated/benchmark_with_optunahub.py
+++ b/examples/_external_or_deprecated/benchmark_with_optunahub.py
@@ -44,7 +44,7 @@
 # account (you don't need one for anything else, just to upload your results),
 # go to your profile and select the API-KEY.
 # Or log in, and navigate to https://www.openml.org/auth/api-key
-openml.config.apikey = ""
+openml.config._config.apikey = ""
 ############################################################################
 # Prepare for preprocessors and an OpenML task
 # ============================================
@@ -95,7 +95,7 @@ def objective(trial: optuna.Trial) -> Pipeline:
     run = openml.runs.run_model_on_task(pipe, task=task_id, avoid_duplicate_runs=False)
 
     logger.log(1, f"Model has been trained - {run}")
-    if openml.config.apikey != "":
+    if openml.config._config.apikey != "":
         try:
             run.publish()
 
diff --git a/examples/_external_or_deprecated/flow_id_tutorial.py b/examples/_external_or_deprecated/flow_id_tutorial.py
index e813655fc..c533cfd9f 100644
--- a/examples/_external_or_deprecated/flow_id_tutorial.py
+++ b/examples/_external_or_deprecated/flow_id_tutorial.py
@@ -16,7 +16,7 @@
 
 # %%
 openml.config.start_using_configuration_for_example()
-openml.config.server = "https://api.openml.org/api/v1/xml"
+openml.config._configserver = "https://api.openml.org/api/v1/xml"
 
 # %%
 # Defining a classifier
diff --git a/openml/_api_calls.py b/openml/_api_calls.py
index 12567ac7a..c3f6d285f 100644
--- a/openml/_api_calls.py
+++ b/openml/_api_calls.py
@@ -172,7 +172,7 @@ def _download_minio_file(
             bucket_name=bucket,
             object_name=object_name,
             file_path=str(destination),
-            progress=ProgressBar() if config.show_progress else None,
+            progress=ProgressBar() if config._config.show_progress else None,
             request_headers=_HEADERS,
         )
         if destination.is_file() and destination.suffix == ".zip":
@@ -317,7 +317,7 @@ def _read_url_files(
     and sending file_elements as files
     """
     data = {} if data is None else data
-    data["api_key"] = config.apikey
+    data["api_key"] = config._config.apikey
     if file_elements is None:
         file_elements = {}
     # Using requests.post sets header 'Accept-encoding' automatically to
@@ -337,8 +337,8 @@ def __read_url(
     md5_checksum: str | None = None,
 ) -> requests.Response:
     data = {} if data is None else data
-    if config.apikey:
-        data["api_key"] = config.apikey
+    if config._config.apikey:
+        data["api_key"] = config._config.apikey
     return _send_request(
         request_method=request_method,
         url=url,
@@ -363,10 +363,10 @@ def _send_request(  # noqa: C901, PLR0912
     files: FILE_ELEMENTS_TYPE | None = None,
     md5_checksum: str | None = None,
 ) -> requests.Response:
-    n_retries = max(1, config.connection_n_retries)
+    n_retries = max(1, config._config.connection_n_retries)
 
     response: requests.Response | None = None
-    delay_method = _human_delay if config.retry_policy == "human" else _robot_delay
+    delay_method = _human_delay if config._config.retry_policy == "human" else _robot_delay
 
     # Error to raise in case of retrying too often. Will be set to the last observed exception.
     retry_raise_e: Exception | None = None
diff --git a/openml/cli.py b/openml/cli.py
index d0a46e498..fb39afe97 100644
--- a/openml/cli.py
+++ b/openml/cli.py
@@ -339,7 +339,7 @@ def main() -> None:
         "'https://openml.github.io/openml-python/main/usage.html#configuration'.",
     )
 
-    configurable_fields = [f for f in config._defaults if f not in ["max_retries"]]
+    configurable_fields = [f for f in config.get_config_as_dict() if f not in ["max_retries"]]
 
     parser_configure.add_argument(
         "field",
diff --git a/openml/config.py b/openml/config.py
index 98a48a1c6..20825463e 100644
--- a/openml/config.py
+++ b/openml/config.py
@@ -15,7 +15,7 @@
 from dataclasses import dataclass, replace
 from io import StringIO
 from pathlib import Path
-from typing import Any, Iterator
+from typing import Any, Iterator, cast
 from typing_extensions import Literal
 from urllib.parse import urlparse
 
@@ -71,7 +71,7 @@ def _resolve_default_cache_dir() -> Path:
     return Path(xdg_cache_home)
 
 
-@dataclass(frozen=True)
+@dataclass
 class OpenMLConfig:
     apikey: str = ""
     server: str = "https://www.openml.org/api/v1/xml"
@@ -259,8 +259,11 @@ def stop_using_configuration_for_example(cls) -> None:
             )
 
         global _config
-        _config = replace(_config, server=cls._test_server, apikey=cls._test_apikey)
-
+        _config = replace(
+            _config,
+            server=cast(str, cls._last_used_server),
+            apikey=cast(str, cls._last_used_key),
+        )
         cls._start_last_called = False
 
 
@@ -334,8 +337,8 @@ def _setup(config: dict[str, Any] | None = None) -> None:
 
     Reads the config file and sets up apikey, server, cache appropriately.
     key and server can be set by the user simply using
-    openml.config.apikey = THEIRKEY
-    openml.config.server = SOMESERVER
+    openml.config._config.apikey = THEIRKEY
+    openml.config._config.server = SOMESERVER
     We could also make it a property but that's less clear.
     """
     global _config
@@ -376,6 +379,7 @@ def _setup(config: dict[str, Any] | None = None) -> None:
         short_cache_dir = Path(config["cachedir"])
 
     _root_cache_directory = short_cache_dir.expanduser().resolve()
+    _config = replace(_config, cachedir=_root_cache_directory)
 
     try:
         cache_exists = _root_cache_directory.exists()
diff --git a/openml/runs/functions.py b/openml/runs/functions.py
index 666b75c37..7fa560833 100644
--- a/openml/runs/functions.py
+++ b/openml/runs/functions.py
@@ -226,7 +226,7 @@ def run_flow_on_task(  # noqa: C901, PLR0912, PLR0915, PLR0913
         raise ValueError("flow_tags should be a list")
 
     if avoid_duplicate_runs is None:
-        avoid_duplicate_runs = openml.config.avoid_duplicate_runs
+        avoid_duplicate_runs = openml.config._config.avoid_duplicate_runs
 
     # TODO: At some point in the future do not allow for arguments in old order (changed 6-2018).
     # Flexibility currently still allowed due to code-snippet in OpenML100 paper (3-2019).
diff --git a/openml/testing.py b/openml/testing.py
index d1da16876..fbf7edf44 100644
--- a/openml/testing.py
+++ b/openml/testing.py
@@ -99,13 +99,13 @@ def setUp(self, n_levels: int = 1, tmpdir_suffix: str = "") -> None:
         os.chdir(self.workdir)
 
         self.cached = True
-        openml.config.apikey = TestBase.user_key
+        openml.config._config.apikey = TestBase.user_key
         self.production_server = "https://www.openml.org/api/v1/xml"
         openml.config.set_root_cache_directory(str(self.workdir))
 
         # Increase the number of retries to avoid spurious server failures
-        self.retry_policy = openml.config.retry_policy
-        self.connection_n_retries = openml.config.connection_n_retries
+        self.retry_policy = openml.config._config.retry_policy
+        self.connection_n_retries = openml.config._config.connection_n_retries
         openml.config.set_retry_policy("robot", n_retries=20)
 
     def use_production_server(self) -> None:
@@ -114,8 +114,8 @@ def use_production_server(self) -> None:
 
         Please use this sparingly - it is better to use the test server.
         """
-        openml.config.server = self.production_server
-        openml.config.apikey = ""
+        openml.config._config.server = self.production_server
+        openml.config._config.apikey = ""
 
     def tearDown(self) -> None:
         """Tear down the test"""
@@ -127,8 +127,8 @@ def tearDown(self) -> None:
                 # one of the files may still be used by another process
                 raise e
 
-        openml.config.connection_n_retries = self.connection_n_retries
-        openml.config.retry_policy = self.retry_policy
+        openml.config._config.connection_n_retries = self.connection_n_retries
+        openml.config._config.retry_policy = self.retry_policy
 
     @classmethod
     def _mark_entity_for_removal(
diff --git a/tests/conftest.py b/tests/conftest.py
index bd974f3f3..ba7c65813 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -97,8 +97,8 @@ def delete_remote_files(tracker, flow_names) -> None:
     :param tracker: Dict
     :return: None
     """
-    openml.config.server = TestBase.test_server
-    openml.config.apikey = TestBase.user_key
+    openml.config._config.server = TestBase.test_server
+    openml.config._config.apikey = TestBase.user_key
 
     # reordering to delete sub flows at the end of flows
     # sub-flows have shorter names, hence, sorting by descending order of flow name length
@@ -263,8 +263,8 @@ def verify_cache_state(test_files_directory) -> Iterator[None]:
 
 @pytest.fixture(autouse=True, scope="session")
 def as_robot() -> Iterator[None]:
-    policy = openml.config.retry_policy
-    n_retries = openml.config.connection_n_retries
+    policy = openml.config._config.retry_policy
+    n_retries = openml.config._config.connection_n_retries
     openml.config.set_retry_policy("robot", n_retries=20)
     yield
     openml.config.set_retry_policy(policy, n_retries)
@@ -273,12 +273,12 @@ def as_robot() -> Iterator[None]:
 @pytest.fixture(autouse=True)
 def with_server(request):
     if "production" in request.keywords:
-        openml.config.server = "https://www.openml.org/api/v1/xml"
-        openml.config.apikey = None
+        openml.config._config.server = "https://www.openml.org/api/v1/xml"
+        openml.config._config.apikey = None
         yield
         return
-    openml.config.server = "https://test.openml.org/api/v1/xml"
-    openml.config.apikey = TestBase.user_key
+    openml.config._config.server = "https://test.openml.org/api/v1/xml"
+    openml.config._config.apikey = TestBase.user_key
     yield
 
 
diff --git a/tests/test_datasets/test_dataset_functions.py b/tests/test_datasets/test_dataset_functions.py
index 266a6f6f7..ab5a4d8b8 100644
--- a/tests/test_datasets/test_dataset_functions.py
+++ b/tests/test_datasets/test_dataset_functions.py
@@ -153,7 +153,7 @@ def test_check_datasets_active(self):
             openml.datasets.check_datasets_active,
             [79],
         )
-        openml.config.server = self.test_server
+        openml.config._config.server = self.test_server
 
     def test_illegal_character_tag(self):
         dataset = openml.datasets.get_dataset(1)
@@ -179,7 +179,7 @@ def test__name_to_id_with_deactivated(self):
         self.use_production_server()
         # /d/1 was deactivated
         assert openml.datasets.functions._name_to_id("anneal") == 2
-        openml.config.server = self.test_server
+        openml.config._config.server = self.test_server
 
     @pytest.mark.production()
     def test__name_to_id_with_multiple_active(self):
@@ -417,8 +417,8 @@ def test__getarff_md5_issue(self):
             "oml:md5_checksum": "abc",
             "oml:url": "https://www.openml.org/data/download/61",
         }
-        n = openml.config.connection_n_retries
-        openml.config.connection_n_retries = 1
+        n = openml.config._config.connection_n_retries
+        openml.config._config.connection_n_retries = 1
 
         self.assertRaisesRegex(
             OpenMLHashException,
@@ -428,7 +428,7 @@ def test__getarff_md5_issue(self):
             description,
         )
 
-        openml.config.connection_n_retries = n
+        openml.config._config.connection_n_retries = n
 
     def test__get_dataset_features(self):
         features_file = _get_dataset_features_file(self.workdir, 2)
@@ -588,7 +588,7 @@ def test_data_status(self):
 
         # admin key for test server (only admins can activate datasets.
         # all users can deactivate their own datasets)
-        openml.config.apikey = TestBase.admin_key
+        openml.config._config.apikey = TestBase.admin_key
 
         openml.datasets.status_update(did, "active")
         self._assert_status_of_dataset(did=did, status="active")
@@ -1507,7 +1507,7 @@ def test_list_datasets_with_high_size_parameter(self):
         datasets_b = openml.datasets.list_datasets(size=np.inf)
 
         # Reverting to test server
-        openml.config.server = self.test_server
+        openml.config._config.server = self.test_server
         assert len(datasets_a) == len(datasets_b)
 
 
diff --git a/tests/test_openml/test_config.py b/tests/test_openml/test_config.py
index 7ef223504..3ff4bcb00 100644
--- a/tests/test_openml/test_config.py
+++ b/tests/test_openml/test_config.py
@@ -46,7 +46,7 @@ class TestConfig(openml.testing.TestBase):
     def test_non_writable_home(self, log_handler_mock, warnings_mock):
         with tempfile.TemporaryDirectory(dir=self.workdir) as td:
             os.chmod(td, 0o444)
-            _dd = copy(openml.config._defaults)
+            _dd = copy(openml.config.get_config_as_dict())
             _dd["cachedir"] = Path(td) / "something-else"
             openml.config._setup(_dd)
 
@@ -110,26 +110,26 @@ class TestConfigurationForExamples(openml.testing.TestBase):
     def test_switch_to_example_configuration(self):
         """Verifies the test configuration is loaded properly."""
         # Below is the default test key which would be used anyway, but just for clarity:
-        openml.config.apikey = TestBase.admin_key
-        openml.config.server = self.production_server
+        openml.config._config.apikey = TestBase.admin_key
+        openml.config._config.server = self.production_server
 
         openml.config.start_using_configuration_for_example()
 
-        assert openml.config.apikey == TestBase.user_key
-        assert openml.config.server == self.test_server
+        assert openml.config._config.apikey == TestBase.user_key
+        assert openml.config._config.server == self.test_server
 
     @pytest.mark.production()
     def test_switch_from_example_configuration(self):
         """Verifies the previous configuration is loaded after stopping."""
         # Below is the default test key which would be used anyway, but just for clarity:
-        openml.config.apikey = TestBase.user_key
-        openml.config.server = self.production_server
+        openml.config._config.apikey = TestBase.user_key
+        openml.config._config.server = self.production_server
 
         openml.config.start_using_configuration_for_example()
         openml.config.stop_using_configuration_for_example()
 
-        assert openml.config.apikey == TestBase.user_key
-        assert openml.config.server == self.production_server
+        assert openml.config._config.apikey == TestBase.user_key
+        assert openml.config._config.server == self.production_server
 
     def test_example_configuration_stop_before_start(self):
         """Verifies an error is raised if `stop_...` is called before `start_...`."""
@@ -146,15 +146,15 @@ def test_example_configuration_stop_before_start(self):
     @pytest.mark.production()
     def test_example_configuration_start_twice(self):
         """Checks that the original config can be returned to if `start..` is called twice."""
-        openml.config.apikey = TestBase.user_key
-        openml.config.server = self.production_server
+        openml.config._config.apikey = TestBase.user_key
+        openml.config._config.server = self.production_server
 
         openml.config.start_using_configuration_for_example()
         openml.config.start_using_configuration_for_example()
         openml.config.stop_using_configuration_for_example()
 
-        assert openml.config.apikey == TestBase.user_key
-        assert openml.config.server == self.production_server
+        assert openml.config._config.apikey == TestBase.user_key
+        assert openml.config._config.server == self.production_server
 
 
 def test_configuration_file_not_overwritten_on_load():
diff --git a/tests/test_utils/test_utils.py b/tests/test_utils/test_utils.py
index 35be84903..1c0b50fe5 100644
--- a/tests/test_utils/test_utils.py
+++ b/tests/test_utils/test_utils.py
@@ -44,7 +44,7 @@ def min_number_evaluations_on_test_server() -> int:
 
 
 def _mocked_perform_api_call(call, request_method):
-    url = openml.config.server + "/" + call
+    url = openml.config._config.server + "/" + call
     return openml._api_calls._download_text_file(url)
 
 

From 38ae9beb47122c54df2122e113ac8a4727bb2eb7 Mon Sep 17 00:00:00 2001
From: Satvik Mishra <112589278+satvshr@users.noreply.github.com>
Date: Wed, 31 Dec 2025 00:07:30 +0530
Subject: [PATCH 3/6] test failures fix

---
 examples/Basics/introduction_tutorial.py | 2 +-
 openml/config.py                         | 9 +++++----
 2 files changed, 6 insertions(+), 5 deletions(-)

diff --git a/examples/Basics/introduction_tutorial.py b/examples/Basics/introduction_tutorial.py
index c864772f5..648bc90ed 100644
--- a/examples/Basics/introduction_tutorial.py
+++ b/examples/Basics/introduction_tutorial.py
@@ -35,7 +35,7 @@
 # %%
 import openml
 
-openml.config.apikey = "YOURKEY"
+openml.config._config.apikey = "YOURKEY"
 
 # %% [markdown]
 # ## Caching
diff --git a/openml/config.py b/openml/config.py
index 20825463e..f2020b8c6 100644
--- a/openml/config.py
+++ b/openml/config.py
@@ -261,8 +261,8 @@ def stop_using_configuration_for_example(cls) -> None:
         global _config
         _config = replace(
             _config,
-            server=cast(str, cls._last_used_server),
-            apikey=cast(str, cls._last_used_key),
+            server=cast("str", cls._last_used_server),
+            apikey=cast("str", cls._last_used_key),
         )
         cls._start_last_called = False
 
@@ -421,7 +421,7 @@ def set_field_in_config_file(field: str, value: Any) -> None:
 def _parse_config(config_file: str | Path) -> dict[str, Any]:
     """Parse the config file, set up defaults."""
     config_file = Path(config_file)
-    config = configparser.RawConfigParser(defaults=_config.__dict__)  # type: ignore
+    config = configparser.RawConfigParser(defaults=OpenMLConfig().__dict__)  # type: ignore
 
     # The ConfigParser requires a [SECTION_HEADER], which we do not expect in our config file.
     # Cheat the ConfigParser module by adding a fake section header
@@ -493,8 +493,9 @@ def set_root_cache_directory(root_cache_directory: str | Path) -> None:
     --------
     get_cache_directory
     """
-    global _root_cache_directory
+    global _root_cache_directory, _config
     _root_cache_directory = Path(root_cache_directory)
+    _config = replace(_config, cachedir=_root_cache_directory)
 
 
 start_using_configuration_for_example = (

From 93ab9c21ce0dcd307666f98766b924e5bc1c09ba Mon Sep 17 00:00:00 2001
From: Satvik Mishra <112589278+satvshr@users.noreply.github.com>
Date: Wed, 31 Dec 2025 00:13:37 +0530
Subject: [PATCH 4/6] Update flow_id_tutorial.py

---
 examples/_external_or_deprecated/flow_id_tutorial.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/examples/_external_or_deprecated/flow_id_tutorial.py b/examples/_external_or_deprecated/flow_id_tutorial.py
index c533cfd9f..496102085 100644
--- a/examples/_external_or_deprecated/flow_id_tutorial.py
+++ b/examples/_external_or_deprecated/flow_id_tutorial.py
@@ -16,7 +16,7 @@
 
 # %%
 openml.config.start_using_configuration_for_example()
-openml.config._configserver = "https://api.openml.org/api/v1/xml"
+openml.config._config.server = "https://api.openml.org/api/v1/xml"
 
 # %%
 # Defining a classifier

From aa25dd69aa2a8b08f17a3bd2d411a1829fd6eccf Mon Sep 17 00:00:00 2001
From: Satvik Mishra <112589278+satvshr@users.noreply.github.com>
Date: Wed, 31 Dec 2025 00:24:36 +0530
Subject: [PATCH 5/6] _defaults bug fixing

---
 openml/cli.py                    | 6 +++++-
 tests/test_openml/test_config.py | 2 +-
 2 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/openml/cli.py b/openml/cli.py
index fb39afe97..c1363ea74 100644
--- a/openml/cli.py
+++ b/openml/cli.py
@@ -9,6 +9,8 @@
 from typing import Callable
 from urllib.parse import urlparse
 
+from attr import fields
+
 from openml import config
 
 
@@ -339,7 +341,9 @@ def main() -> None:
         "'https://openml.github.io/openml-python/main/usage.html#configuration'.",
     )
 
-    configurable_fields = [f for f in config.get_config_as_dict() if f not in ["max_retries"]]
+    configurable_fields = [
+        f.name for f in fields(config.OpenMLConfig) if f.name not in ["max_retries"]
+    ]
 
     parser_configure.add_argument(
         "field",
diff --git a/tests/test_openml/test_config.py b/tests/test_openml/test_config.py
index 3ff4bcb00..104639460 100644
--- a/tests/test_openml/test_config.py
+++ b/tests/test_openml/test_config.py
@@ -46,7 +46,7 @@ class TestConfig(openml.testing.TestBase):
     def test_non_writable_home(self, log_handler_mock, warnings_mock):
         with tempfile.TemporaryDirectory(dir=self.workdir) as td:
             os.chmod(td, 0o444)
-            _dd = copy(openml.config.get_config_as_dict())
+            _dd = copy(openml.config.OpenMLConfig().__dict__)
             _dd["cachedir"] = Path(td) / "something-else"
             openml.config._setup(_dd)
 

From a98b6b1c7753dbf02d8d6a2dc552abff8e8c60bb Mon Sep 17 00:00:00 2001
From: Satvik Mishra <112589278+satvshr@users.noreply.github.com>
Date: Wed, 31 Dec 2025 19:10:58 +0530
Subject: [PATCH 6/6] removed __setattr__ given it is not supported

---
 openml/config.py | 8 --------
 1 file changed, 8 deletions(-)

diff --git a/openml/config.py b/openml/config.py
index f2020b8c6..ad8060e7d 100644
--- a/openml/config.py
+++ b/openml/config.py
@@ -160,14 +160,6 @@ def __getattr__(name: str) -> Any:
     raise AttributeError(f"module 'openml.config' has no attribute '{name}'")
 
 
-def __setattr__(name: str, value: Any) -> None:  # noqa: N807
-    global _config
-    if hasattr(_config, name):
-        _config = replace(_config, **{name: value})
-    else:
-        raise AttributeError(f"module 'openml.config' has no attribute '{name}'")
-
-
 def get_server_base_url() -> str:
     """Return the base URL of the currently configured server.