anemoi-utils 0.4.28__tar.gz → 0.4.29__tar.gz

{anemoi_utils-0.4.28 → anemoi_utils-0.4.29}/.pre-commit-config.yaml

@@ -41,7 +41,7 @@ repos:
     - --profile black
     - --project anemoi
 - repo: https://github.com/astral-sh/ruff-pre-commit
-  rev: v0.11.12
+  rev: v0.12.2
   hooks:
   - id: ruff
     args:
@@ -69,7 +69,7 @@ repos:
   hooks:
   - id: pyproject-fmt
 -   repo: https://github.com/jshwi/docsig # Check docstrings against function sig
-    rev: v0.69.3
+    rev: v0.70.0
     hooks:
     -   id: docsig
         args:

anemoi_utils-0.4.29/.release-please-manifest.json

@@ -0,0 +1,3 @@
+{
+    ".": "0.4.29"
+}

{anemoi_utils-0.4.28 → anemoi_utils-0.4.29}/CHANGELOG.md

@@ -8,6 +8,22 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 Please add your functional changes to the appropriate section in the PR.
 Keep it human-readable, your future self will thank you!
 
+## [0.4.29](https://github.com/ecmwf/anemoi-utils/compare/0.4.28...0.4.29) (2025-07-22)
+
+
+### Features
+
+* Better support for negative timedeltas ([#180](https://github.com/ecmwf/anemoi-utils/issues/180)) ([3f8041a](https://github.com/ecmwf/anemoi-utils/commit/3f8041a46b525b6fcbe6171cd8a8a40ec30b2c1f))
+* **deps:** Use mlflow-skinny instead of mlflow ([#184](https://github.com/ecmwf/anemoi-utils/issues/184)) ([82e5c30](https://github.com/ecmwf/anemoi-utils/commit/82e5c3053962cd8e1e8f6a1ea9e8f92492e497b4))
+* Protect mlflow token file ([#183](https://github.com/ecmwf/anemoi-utils/issues/183)) ([fdf0fc8](https://github.com/ecmwf/anemoi-utils/commit/fdf0fc84ee3e8076928f6c888374cd3aa008023b))
+* **sanitise:** Sanitation level ([#175](https://github.com/ecmwf/anemoi-utils/issues/175)) ([8d85d8f](https://github.com/ecmwf/anemoi-utils/commit/8d85d8fd889bf72b8066cc021d4d7b329a360848))
+* Support negative timedelta ([#178](https://github.com/ecmwf/anemoi-utils/issues/178)) ([546f6ec](https://github.com/ecmwf/anemoi-utils/commit/546f6ec76534cd39094957ce3b57b34f14f7a000))
+
+
+### Bug Fixes
+
+* Clean utils ([#185](https://github.com/ecmwf/anemoi-utils/issues/185)) ([de3c7a4](https://github.com/ecmwf/anemoi-utils/commit/de3c7a47f14c258997942564717c480caa124ee6))
+
 ## [0.4.28](https://github.com/ecmwf/anemoi-utils/compare/0.4.27...0.4.28) (2025-07-03)
 
 

{anemoi_utils-0.4.28/src/anemoi_utils.egg-info → anemoi_utils-0.4.29}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: anemoi-utils
-Version: 0.4.28
+Version: 0.4.29
 Summary: A package to hold various functions to support training of ML models on ECMWF data.
 Author-email: "European Centre for Medium-Range Weather Forecasts (ECMWF)" <software.support@ecmwf.int>
 License:                                  Apache License
@@ -251,7 +251,7 @@ Requires-Dist: termcolor; extra == "docs"
 Provides-Extra: grib
 Requires-Dist: requests; extra == "grib"
 Provides-Extra: mlflow
-Requires-Dist: mlflow>=2.11.1; extra == "mlflow"
+Requires-Dist: mlflow-skinny>=2.11.1; extra == "mlflow"
 Requires-Dist: requests; extra == "mlflow"
 Provides-Extra: provenance
 Requires-Dist: gitpython; extra == "provenance"
@@ -259,6 +259,7 @@ Requires-Dist: nvsmi; extra == "provenance"
 Provides-Extra: s3
 Requires-Dist: boto3>1.36; extra == "s3"
 Provides-Extra: tests
+Requires-Dist: anemoi-utils[mlflow]; extra == "tests"
 Requires-Dist: pytest; extra == "tests"
 Requires-Dist: pytest-mock>=3; extra == "tests"
 Provides-Extra: text

{anemoi_utils-0.4.28 → anemoi_utils-0.4.29}/pyproject.toml

@@ -69,7 +69,7 @@ optional-dependencies.docs = [
 
 optional-dependencies.grib = [ "requests" ]
 
-optional-dependencies.mlflow = [ "mlflow>=2.11.1", "requests" ]
+optional-dependencies.mlflow = [ "mlflow-skinny>=2.11.1", "requests" ]
 
 optional-dependencies.provenance = [ "gitpython", "nvsmi" ]
 
@@ -77,7 +77,7 @@ optional-dependencies.s3 = [
   "boto3>1.36",
 ]
 
-optional-dependencies.tests = [ "pytest", "pytest-mock>=3" ]
+optional-dependencies.tests = [ "anemoi-utils[mlflow]", "pytest", "pytest-mock>=3" ]
 
 optional-dependencies.text = [ "termcolor", "wcwidth" ]
 

{anemoi_utils-0.4.28 → anemoi_utils-0.4.29}/src/anemoi/utils/_version.py

@@ -17,5 +17,5 @@ __version__: str
 __version_tuple__: VERSION_TUPLE
 version_tuple: VERSION_TUPLE
 
-__version__ = version = '0.4.28'
-__version_tuple__ = version_tuple = (0, 4, 28)
+__version__ = version = '0.4.29'
+__version_tuple__ = version_tuple = (0, 4, 29)

{anemoi_utils-0.4.28 → anemoi_utils-0.4.29}/src/anemoi/utils/dates.py

@@ -199,6 +199,15 @@ def as_timedelta(frequency: Union[int, str, datetime.timedelta]) -> datetime.tim
     except ValueError:
         pass
 
+    if frequency.startswith(" ") or frequency.startswith(" "):
+        frequency = frequency.strip()
+
+    if frequency.startswith("-"):
+        return -as_timedelta(frequency[1:])
+
+    if frequency.startswith("+"):
+        return as_timedelta(frequency[1:])
+
     if re.match(r"^\d+[hdms]$", frequency, re.IGNORECASE):
         unit = frequency[-1].lower()
         v = int(frequency[:-1])
@@ -261,6 +270,8 @@ def frequency_to_string(frequency: datetime.timedelta) -> str:
     frequency = frequency_to_timedelta(frequency)
 
     total_seconds = frequency.total_seconds()
+    if total_seconds < 0:
+        return f"-{frequency_to_string(-frequency)}"
     assert int(total_seconds) == total_seconds, total_seconds
     total_seconds = int(total_seconds)
 

{anemoi_utils-0.4.28 → anemoi_utils-0.4.29}/src/anemoi/utils/mlflow/auth.py

@@ -22,6 +22,7 @@ from typing import TYPE_CHECKING
 import requests
 from requests.exceptions import HTTPError
 
+from ..config import config_path
 from ..config import load_config
 from ..config import save_config
 from ..remote import robust
@@ -87,6 +88,14 @@ class TokenAuth:
 
     @staticmethod
     def load_config() -> dict:
+        path = config_path(TokenAuth.config_file)
+
+        if not os.path.exists(path):
+            save_config(TokenAuth.config_file, {})
+
+        if os.path.exists(path) and os.stat(path).st_mode & 0o777 != 0o600:
+            os.chmod(path, 0o600)
+
         return load_config(TokenAuth.config_file)
 
     def enabled(fn: Callable) -> Callable:  # noqa: N805

anemoi_utils-0.4.29/src/anemoi/utils/mlflow/utils.py

@@ -0,0 +1,44 @@
+# (C) Copyright 2024 Anemoi contributors.
+#
+# This software is licensed under the terms of the Apache Licence Version 2.0
+# which can be obtained at http://www.apache.org/licenses/LICENSE-2.0.
+#
+# In applying this licence, ECMWF does not waive the privileges and immunities
+# granted to it by virtue of its status as an intergovernmental organisation
+# nor does it submit to any jurisdiction.
+from __future__ import annotations
+
+import os
+
+import requests
+
+from ..remote import robust
+
+
+def health_check(tracking_uri: str) -> None:
+    """Query the health endpoint of an MLflow server.
+
+    If the server is not reachable, raise an error and remind the user that authentication may be required.
+
+    Raises
+    ------
+    ConnectionError
+        If the server is not reachable.
+
+    """
+    token = os.getenv("MLFLOW_TRACKING_TOKEN")
+
+    headers = {"Authorization": f"Bearer {token}"}
+    response = robust(requests.get, retry_after=30, maximum_tries=10)(
+        f"{tracking_uri}/health",
+        headers=headers,
+        timeout=60,
+    )
+
+    if response.text == "OK":
+        return
+
+    error_msg = f"Could not connect to MLflow server at {tracking_uri}. "
+    if not token:
+        error_msg += "The server may require authentication, did you forget to turn it on?"
+    raise ConnectionError(error_msg)

{anemoi_utils-0.4.28 → anemoi_utils-0.4.29}/src/anemoi/utils/sanitise.py

@@ -17,19 +17,24 @@ from urllib.parse import urlencode
 from urllib.parse import urlparse
 from urllib.parse import urlunparse
 
-# Patterns used but earthkit-data for url-patterns and path-patterns
+# Patterns used by earthkit-data for url-patterns and path-patterns
 
-RE1 = re.compile(r"{([^}]*)}")
-RE2 = re.compile(r"\(([^}]*)\)")
+RE1 = re.compile(r"{([^}]*)}")  # {*}
+RE2 = re.compile(r"\(([^}]*)\)")  # (*)
 
 
-def sanitise(obj: Any) -> Any:
-    """Sanitise an object by replacing all full paths with shortened versions and URL passwords with '***'.
+def sanitise(obj: Any, level=1) -> Any:
+    """Sanitise an object by replacing all full paths with shortened versions and URL credentials with '***'.
 
     Parameters
     ----------
     obj : Any
         The object to sanitise.
+    level : int, optional
+        The level of sanitation. The higher levels will also apply the levels below it.
+        - 1: Shorten file paths to file name and hide credentials in URLs (default).
+        - 2: Hide hostnames in URLs.
+        - 3: Hide full file paths and URLs.
 
     Returns
     -------
@@ -37,6 +42,8 @@ def sanitise(obj: Any) -> Any:
         The sanitised object.
     """
 
+    assert level in (1, 2, 3), "level must be 1, 2 or 3"
+
     if isinstance(obj, dict):
         return {sanitise(k): sanitise(v) for k, v in obj.items()}
 
@@ -47,29 +54,21 @@ def sanitise(obj: Any) -> Any:
         return tuple(sanitise(v) for v in obj)
 
     if isinstance(obj, str):
-        return _sanitise_string(obj)
+        return _sanitise_string(obj, level)
 
     return obj
 
 
-def _sanitise_string(obj: str) -> str:
-    """Sanitise a string by replacing full paths and URL passwords.
-
-    Parameters
-    ----------
-    obj : str
-        The string to sanitise.
-
-    Returns
-    -------
-    str
-        The sanitised string.
-    """
+def _sanitise_string(obj: str, level=1) -> str:
+    """Sanitise a string by replacing full paths and URL passwords."""
 
     parsed = urlparse(obj, allow_fragments=True)
 
     if parsed.scheme and parsed.scheme[0].isalpha():
-        return _sanitise_url(parsed)
+        return _sanitise_url(parsed, level)
+
+    if level > 2:
+        return "hidden"
 
     if obj.startswith("/") or obj.startswith("~"):
         return _sanitise_path(obj)
@@ -77,19 +76,8 @@ def _sanitise_string(obj: str) -> str:
     return obj
 
 
-def _sanitise_url(parsed: Any) -> str:
-    """Sanitise a URL by replacing passwords with '***'.
-
-    Parameters
-    ----------
-    parsed : Any
-        The parsed URL.
-
-    Returns
-    -------
-    str
-        The sanitised URL.
-    """
+def _sanitise_url(parsed: Any, level=1) -> str:
+    """Sanitise a URL by replacing passwords with '***'."""
 
     LIST = [
         "pass",
@@ -107,6 +95,9 @@ def _sanitise_url(parsed: Any) -> str:
         "_api_key",
         "username",
         "login",
+        "auth",
+        "auth_token",
+        "auth_key",
     ]
 
     scheme, netloc, path, params, query, fragment = parsed
@@ -130,26 +121,25 @@ def _sanitise_url(parsed: Any) -> str:
                 qs[k] = "hidden"
         params = urlencode(qs, doseq=True)
 
-    return urlunparse([scheme, netloc, path, params, query, fragment])
+    if level > 1:
+        if (bits := netloc.split("@")) and len(bits) > 1:
+            netloc = f"{bits[0]}@hidden"
+        else:
+            netloc = "hidden"
 
+    if level > 2:
+        return urlunparse([scheme, netloc, "", "", "", ""])
 
-def _sanitise_path(path: str) -> str:
-    """Sanitise a file path by shortening it.
+    return urlunparse([scheme, netloc, path, params, query, fragment])
 
-    Parameters
-    ----------
-    path : str
-        The file path to sanitise.
 
-    Returns
-    -------
-    str
-        The sanitised file path.
-    """
+def _sanitise_path(path: str) -> str:
+    """Sanitise a file path by shortening it."""
     bits = list(reversed(Path(path).parts))
     result = [bits.pop(0)]
     for bit in bits:
         if RE1.match(bit) or RE2.match(bit):
+            # keep earthkit-data folder patterns
             result.append(bit)
             continue
         if result[-1] == "...":

{anemoi_utils-0.4.28 → anemoi_utils-0.4.29/src/anemoi_utils.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: anemoi-utils
-Version: 0.4.28
+Version: 0.4.29
 Summary: A package to hold various functions to support training of ML models on ECMWF data.
 Author-email: "European Centre for Medium-Range Weather Forecasts (ECMWF)" <software.support@ecmwf.int>
 License:                                  Apache License
@@ -251,7 +251,7 @@ Requires-Dist: termcolor; extra == "docs"
 Provides-Extra: grib
 Requires-Dist: requests; extra == "grib"
 Provides-Extra: mlflow
-Requires-Dist: mlflow>=2.11.1; extra == "mlflow"
+Requires-Dist: mlflow-skinny>=2.11.1; extra == "mlflow"
 Requires-Dist: requests; extra == "mlflow"
 Provides-Extra: provenance
 Requires-Dist: gitpython; extra == "provenance"
@@ -259,6 +259,7 @@ Requires-Dist: nvsmi; extra == "provenance"
 Provides-Extra: s3
 Requires-Dist: boto3>1.36; extra == "s3"
 Provides-Extra: tests
+Requires-Dist: anemoi-utils[mlflow]; extra == "tests"
 Requires-Dist: pytest; extra == "tests"
 Requires-Dist: pytest-mock>=3; extra == "tests"
 Provides-Extra: text

{anemoi_utils-0.4.28 → anemoi_utils-0.4.29}/src/anemoi_utils.egg-info/SOURCES.txt

@@ -94,10 +94,9 @@ tests/test_dates.py
 tests/test_frequency.py
 tests/test_mlflow_auth.py
 tests/test_mlflow_client.py
-tests/test_mlflow_utils.py
 tests/test_provenance.py
 tests/test_remote.py
-tests/test_sanetise.py
+tests/test_sanitise.py
 tests/test_utils.py
 tests/test-transfer-data/file
 tests/test-transfer-data/directory/exotic filename ;^"'[=.,#]()êüçò✅.txt

{anemoi_utils-0.4.28 → anemoi_utils-0.4.29}/src/anemoi_utils.egg-info/requires.txt

@@ -34,7 +34,7 @@ termcolor
 requests
 
 [mlflow]
-mlflow>=2.11.1
+mlflow-skinny>=2.11.1
 requests
 
 [provenance]
@@ -45,6 +45,7 @@ nvsmi
 boto3>1.36
 
 [tests]
+anemoi-utils[mlflow]
 pytest
 pytest-mock>=3
 

anemoi_utils-0.4.28/tests/test_sanetise.py → anemoi_utils-0.4.29/tests/test_sanitise.py

@@ -47,6 +47,12 @@ def test_sanitise_urls() -> None:
     assert sanitise("http://www.example.com/path;username=secret") == "http://www.example.com/path;username=hidden"
     assert sanitise("http://www.example.com/path;login=secret") == "http://www.example.com/path;login=hidden"
 
+    assert sanitise("http://www.example.com/path;_api_token=secret", level=2) == "http://hidden/path;_api_token=hidden"
+    assert sanitise("http://johndoe:password@host:port/path", level=2) == "http://user:***@hidden/path"
+    assert sanitise("http://host:port/path", level=2) == "http://hidden/path"
+
+    assert sanitise("http://www.example.com/path;_api_token=secret", level=3) == "http://hidden"
+
 
 def test_sanitise_paths() -> None:
     """Test the sanitise function for sanitizing file paths."""
@@ -65,6 +71,11 @@ def test_sanitise_paths() -> None:
     assert sanitise("sub/folder/test.grib") == "sub/folder/test.grib"
     assert sanitise("./folder/test.grib") == "./folder/test.grib"
 
+    assert sanitise("./folder/test.grib", level=2) == "./folder/test.grib"
+
+    assert sanitise("./folder/test.grib", level=3) == "hidden"
+    assert sanitise("/home/johndoe/.ssh/id_rsa", level=3) == "hidden"
+
 
 if __name__ == "__main__":
     for name, obj in list(globals().items()):

anemoi_utils-0.4.28/.release-please-manifest.json

@@ -1,3 +0,0 @@
-{
-    ".": "0.4.28"
-}

anemoi_utils-0.4.28/src/anemoi/utils/mlflow/utils.py

@@ -1,159 +0,0 @@
-# (C) Copyright 2024 Anemoi contributors.
-#
-# This software is licensed under the terms of the Apache Licence Version 2.0
-# which can be obtained at http://www.apache.org/licenses/LICENSE-2.0.
-#
-# In applying this licence, ECMWF does not waive the privileges and immunities
-# granted to it by virtue of its status as an intergovernmental organisation
-# nor does it submit to any jurisdiction.
-from __future__ import annotations
-
-import functools
-import os
-from typing import Any
-
-import requests
-
-from ..remote import robust
-
-
-def health_check(tracking_uri: str) -> None:
-    """Query the health endpoint of an MLflow server.
-
-    If the server is not reachable, raise an error and remind the user that authentication may be required.
-
-    Raises
-    ------
-    ConnectionError
-        If the server is not reachable.
-
-    """
-    token = os.getenv("MLFLOW_TRACKING_TOKEN")
-
-    headers = {"Authorization": f"Bearer {token}"}
-    response = robust(requests.get, retry_after=30, maximum_tries=10)(
-        f"{tracking_uri}/health",
-        headers=headers,
-        timeout=60,
-    )
-
-    if response.text == "OK":
-        return
-
-    error_msg = f"Could not connect to MLflow server at {tracking_uri}. "
-    if not token:
-        error_msg += "The server may require authentication, did you forget to turn it on?"
-    raise ConnectionError(error_msg)
-
-
-def expand_iterables(
-    params: dict[str, Any],
-    *,
-    size_threshold: int | None = None,
-    recursive: bool = True,
-    delimiter: str = ".",
-) -> dict[str, Any]:
-    """Expand any iterable values to the form {key.i: value_i}.
-
-    If expanded will also add {key.all: [value_0, value_1, ...], key.length: len([value_0, value_1, ...])}.
-
-    If `size_threshold` is not None, expand the iterable only if the length of str(value) is
-    greater than `size_threshold`.
-
-    Parameters
-    ----------
-    params : dict[str, Any]
-        Parameters to be expanded.
-    size_threshold : int | None, optional
-        Threshold of str(value) to expand iterable at.
-        Default is None.
-    recursive : bool, optional
-        Expand nested dictionaries.
-        Default is True.
-    delimiter: str, optional
-        Delimiter to use for keys.
-        Default is ".".
-
-    Returns
-    -------
-    dict[str, Any]
-        Dictionary with all iterable values expanded.
-
-    Examples
-    --------
-        >>> expand_iterables({'a': ['a', 'b', 'c']})
-        {'a.0': 'a', 'a.1': 'b', 'a.2': 'c', 'a.all': ['a', 'b', 'c'], 'a.length': 3}
-        >>> expand_iterables({'a': {'b': ['a', 'b', 'c']}})
-        {'a': {'b.0': 'a', 'b.1': 'b', 'b.2': 'c', 'b.all': ['a', 'b', 'c'], 'b.length': 3}}
-        >>> expand_iterables({'a': ['a', 'b', 'c']}, size_threshold=100)
-        {'a': ['a', 'b', 'c']}
-        >>> expand_iterables({'a': [[0,1,2], 'b', 'c']})
-        {'a.0': {0: 0, 1: 1, 2: 2}, 'a.1': 'b', 'a.2': 'c', 'a.all': [[0, 1, 2], 'b', 'c'], 'a.length': 3}
-    """
-
-    def should_be_expanded(x: Any) -> bool:
-        return size_threshold is None or len(str(x)) > size_threshold
-
-    nested_func = functools.partial(expand_iterables, size_threshold=size_threshold, recursive=recursive)
-
-    def expand(val: dict | list) -> dict[str, Any]:
-        if not recursive:
-            return val
-        if isinstance(val, dict):
-            return nested_func(val)
-        if isinstance(val, list):
-            return nested_func(dict(enumerate(val)))
-        return val
-
-    expanded_params = {}
-
-    for key, value in params.items():
-        if isinstance(value, (list, tuple)):
-            if should_be_expanded(value):
-                for i, v in enumerate(value):
-                    expanded_params[f"{key}{delimiter}{i}"] = expand(v)
-
-                expanded_params[f"{key}{delimiter}all"] = value
-                expanded_params[f"{key}{delimiter}length"] = len(value)
-            else:
-                expanded_params[key] = value
-        else:
-            expanded_params[key] = expand(value)
-    return expanded_params
-
-
-def clean_config_params(params: dict[str, Any]) -> dict[str, Any]:
-    """Clean up params to avoid issues with mlflow.
-
-    Too many logged params will make the server take longer to render the
-    experiment.
-
-    Parameters
-    ----------
-    params : dict[str, Any]
-        Parameters to clean up.
-
-    Returns
-    -------
-    dict[str, Any]
-        Cleaned up params ready for MlFlow.
-    """
-    prefixes_to_remove = [
-        "hardware",
-        "data",
-        "dataloader",
-        "model",
-        "training",
-        "diagnostics",
-        "graph",
-        "metadata.config",
-        "config.dataset.sourcesmetadata.dataset.variables_metadata",
-        "metadata.dataset.sources",
-        "metadata.dataset.specific",
-        "metadata.dataset.variables_metadata",
-    ]
-
-    keys_to_remove = [key for key in params if any(key.startswith(prefix) for prefix in prefixes_to_remove)]
-    for key in keys_to_remove:
-        del params[key]
-    return params

anemoi_utils-0.4.28/tests/test_mlflow_utils.py

@@ -1,76 +0,0 @@
-# (C) Copyright 2024 European Centre for Medium-Range Weather Forecasts.
-# This software is licensed under the terms of the Apache Licence Version 2.0
-# which can be obtained at http://www.apache.org/licenses/LICENSE-2.0.
-# In applying this licence, ECMWF does not waive the privileges and immunities
-# granted to it by virtue of its status as an intergovernmental organisation
-# nor does it submit to any jurisdiction.
-
-from anemoi.utils.mlflow.utils import clean_config_params
-from anemoi.utils.mlflow.utils import expand_iterables
-
-
-def test_clean_config_params() -> None:
-    params = {
-        "config.dataset.format": None,
-        "config.model.num_channels": None,
-        "model.num_channels": None,
-        "data.frequency": None,
-        "diagnostics.plot": None,
-        "hardware.num_gpus": None,
-        "metadata.config.dataset": None,
-        "metadata.dataset.sources/1.specific.forward.forward.attrs.variables_metadata.z_500.mars.expver": None,
-        "metadata.dataset.specific.forward.forward.attrs.variables_metadata.z_500.mars.expver": None,
-        "config.data.normalizer.default": None,
-        "config.data.normalizer.std": None,
-        "config.data.normalizer.min-max": None,
-        "config.data.normalizer.max": None,
-    }
-
-    cleaned = clean_config_params(params)
-    result = {
-        "config.dataset.format": None,
-        "config.model.num_channels": None,
-        "config.data.normalizer.default": None,
-        "config.data.normalizer.std": None,
-        "config.data.normalizer.min-max": None,
-        "config.data.normalizer.max": None,
-    }
-    assert cleaned == result
-
-
-def test_expand_iterables_single_iterable() -> None:
-    # Test case with a single iterable
-    dictionary = {"a": ["a", "b", "c"]}
-    expanded = expand_iterables(dictionary)
-    assert expanded == {"a.0": "a", "a.1": "b", "a.2": "c", "a.all": ["a", "b", "c"], "a.length": 3}
-
-
-def test_expand_iterables_size_threshold() -> None:
-    # Test case with a single iterable
-    dictionary = {"a": ["a", "b", "c"]}
-    expanded = expand_iterables(dictionary, size_threshold=100)
-    assert expanded == dictionary
-
-
-def test_expand_iterables_with_nested_dict() -> None:
-    dictionary = {"a": {"b": ["a", "b", "c"]}}
-    expanded = expand_iterables(dictionary)
-    assert expanded == {"a": {"b.0": "a", "b.1": "b", "b.2": "c", "b.all": ["a", "b", "c"], "b.length": 3}}
-
-
-def test_expand_iterables_with_nested_dict_thresholded() -> None:
-    dictionary = {"a": {"b": ["a", "b", "c"]}, "c": ["d"]}
-    expanded = expand_iterables(dictionary, size_threshold=5)
-    assert expanded == {"a": {"b.0": "a", "b.1": "b", "b.2": "c", "b.all": ["a", "b", "c"], "b.length": 3}, "c": ["d"]}
-
-
-def test_expand_iterables_with_nested_list() -> None:
-    dictionary = {"a": [[0, 1, 2], "b", "c"]}
-    expanded = expand_iterables(dictionary)
-    assert expanded == {
-        "a.0": {0: 0, 1: 1, 2: 2},
-        "a.1": "b",
-        "a.2": "c",
-        "a.all": [[0, 1, 2], "b", "c"],
-        "a.length": 3,
-    }