update linter

.pre-commit-config.yaml

@@ -1,38 +1,27 @@
+exclude: scripts/
 repos:
 -   repo: https://github.com/pre-commit/pre-commit-hooks
     rev: v2.3.0
     hooks:
     -   id: check-yaml
     -   id: end-of-file-fixer
-    -   id: trailing-whitespace
+    -   id: check-merge-conflict
     -   id: mixed-line-ending
         args: ['--fix=lf']
--   repo: https://github.com/psf/black
-    rev: 22.6.0
+
+    # Ruff is a replacement for flake8 and many other linters (much faster too)
+-   repo: https://github.com/astral-sh/ruff-pre-commit
+    # Ruff version.
+    rev: v0.6.1
     hooks:
-    -   id: black
--   repo: https://github.com/PyCQA/flake8
-    rev: 3.8.3
+    -   id: ruff
+        args: ["--fix"]
+        # Run the formatter.
+    -   id: ruff-format
+
+    # ensures __future__ import annotations at top of files which require it
+    # for the typing features they are using.
+-   repo: https://github.com/frostming/fix-future-annotations
+    rev: 0.5.0
     hooks:
-    -   id: flake8
-        additional_dependencies:
-          - flake8-black
-          - flake8-breakpoint
-          - flake8-docstrings
--   repo: https://github.com/pycqa/isort
-    rev: 5.12.0
-    hooks:
-      - id: isort
-        name: isort (python)
-        args: ["--profile", "black"]
-      - id: isort
-        name: isort (cython)
-        types: [cython]
-      - id: isort
-        name: isort (pyi)
-        types: [pyi]
--   repo: https://github.com/kynan/nbstripout
-    rev: 0.3.9
-    hooks:
-    -   id: nbstripout
-        files: ".ipynb"
+    -   id: fix-future-annotations

pyproject.toml

@@ -64,3 +64,76 @@ dev = ["dbscan1d[test]"]
 "Bug Tracker" = "https://github.com/d-chambers/dbscan1d/issues"
 "Documentation" = "https://github.com/d-chambers/dbscan1d"
 "Homepage" = "https://github.com/d-chambers/dbscan1d"
+
+# --- formatting
+
+[tool.ruff]
+
+line-length = 88
+
+# enable certain types of linting
+lint.select = [
+    "E",
+    "F",
+    "UP",
+    "RUF",
+    "I001",
+    "D",
+    "FA",
+    "T",
+    "N",
+    "NPY",
+    "NPY201",
+]
+
+# Exclude a variety of commonly ignored directories.
+exclude = [
+    ".bzr",
+    ".direnv",
+    ".eggs",
+    ".git",
+    ".git-rewrite",
+    ".hg",
+    ".mypy_cache",
+    ".nox",
+    ".pants.d",
+    ".pytype",
+    ".ruff_cache",
+    ".svn",
+    ".tox",
+    ".venv",
+    "__pypackages__",
+    "_build",
+    "buck-out",
+    "build",
+    "dist",
+    "node_modules",
+    "venv",
+    "__init__.py"
+]
+
+# lowest python version supported
+target-version = "py310"
+
+lint.fixable = ["ALL"]
+
+# List of codes to ignore
+lint.ignore = ["D105", "D107", "D401", "D205", "D200", "D400", "N803", "N806"]
+
+[tool.ruff.lint.mccabe]
+# Unlike Flake8, default to a complexity level of 10.
+max-complexity = 10
+
+# config for docstring parsing
+[tool.ruff.lint.pydocstyle]
+convention = "numpy"
+
+[tool.pytest.ini_options]
+filterwarnings = [
+    # Ignore hdf5 warnings from pytables, See pytables #1035
+    'ignore::Warning:tables:'
+]
+
+[tool.ruff.format]
+# Use `\n` line endings for all files
+line-ending = "lf"

scripts/profile_dbscan1d.ipynb

@@ -18,13 +18,20 @@
     "\n",
     "import matplotlib.pyplot as plt\n",
     "import pandas as pd\n",
-    "\n",
-    "from dbscan1d import DBSCAN1D\n",
     "from sklearn.cluster import DBSCAN\n",
     "from sklearn.datasets import make_blobs\n",
     "\n",
+    "from dbscan1d import DBSCAN1D\n",
     "\n",
-    "n_points = [10, 100, 1_000, 10_000, 20_000, 30_000, 40_000,]\n",
+    "n_points = [\n",
+    "    10,\n",
+    "    100,\n",
+    "    1_000,\n",
+    "    10_000,\n",
+    "    20_000,\n",
+    "    30_000,\n",
+    "    40_000,\n",
+    "]\n",
     "centers = 2"
    ]
   },
@@ -46,8 +53,8 @@
    "outputs": [],
    "source": [
     "# Profile\n",
-    "db1 = DBSCAN1D(.5, 4)\n",
-    "db2 = DBSCAN(.5, 4)"
+    "db1 = DBSCAN1D(0.5, 4)\n",
+    "db2 = DBSCAN(0.5, 4)"
    ]
   },
   {
@@ -57,16 +64,16 @@
    "outputs": [],
    "source": [
     "# profile each stream type with each function\n",
-    "df = pd.DataFrame(columns=['dbscan', 'dbscan1d'], index=n_points)\n",
+    "df = pd.DataFrame(columns=[\"dbscan\", \"dbscan1d\"], index=n_points)\n",
     "for n_point in n_points:\n",
-    "    print(f'on {n_point}')\n",
+    "    print(f\"on {n_point}\")\n",
     "    X = create_blobs(n_point, centers)\n",
-    "    print('starting dbscan1d')\n",
+    "    print(\"starting dbscan1d\")\n",
     "    ti1 = %timeit -o db1.fit_predict(X)\n",
-    "    df.loc[n_point, 'dbscan1d'] = ti1.best\n",
-    "    print('starting dbscan')\n",
+    "    df.loc[n_point, \"dbscan1d\"] = ti1.best\n",
+    "    print(\"starting dbscan\")\n",
     "    ti2 = %timeit -o db2.fit_predict(X)\n",
-    "    df.loc[n_point, 'dbscan'] = ti2.best\n",
+    "    df.loc[n_point, \"dbscan\"] = ti2.best\n",
     "    print()\n",
     "    print()"
    ]
@@ -89,20 +96,20 @@
    },
    "outputs": [],
    "source": [
-    "out_path = Path(__file__).parent / 'profile_results.png'\n",
+    "out_path = Path(__file__).parent / \"profile_results.png\"\n",
     "\n",
     "x = df.index.values\n",
-    "plt.loglog(x, df['dbscan'].values, label='dbscan', color='r')\n",
-    "plt.loglog(x, df['dbscan1d'].values, label='dbscan1d', color='b')\n",
+    "plt.loglog(x, df[\"dbscan\"].values, label=\"dbscan\", color=\"r\")\n",
+    "plt.loglog(x, df[\"dbscan1d\"].values, label=\"dbscan1d\", color=\"b\")\n",
     "\n",
-    "plt.xlabel('number of points')\n",
-    "plt.ylabel('run time (s)')\n",
+    "plt.xlabel(\"number of points\")\n",
+    "plt.ylabel(\"run time (s)\")\n",
     "\n",
     "plt.legend()\n",
     "\n",
     "plt.savefig(out_path)\n",
     "\n",
-    "plt.show()\n"
+    "plt.show()"
    ]
   }
  ],

src/dbscan1d/core.py

@@ -3,7 +3,8 @@
 
 It should be *much* more efficient for large datasets.
 """
-from typing import Optional
+
+from __future__ import annotations
 
 import numpy as np
 
@@ -17,9 +18,9 @@ class DBSCAN1D:
     """
 
     # params that change upon fit/training
-    core_sample_indices_: Optional[np.ndarray] = None
-    components_: Optional[np.ndarray] = None
-    labels_: Optional[np.ndarray] = None
+    core_sample_indices_: np.ndarray | None = None
+    components_: np.ndarray | None = None
+    labels_: np.ndarray | None = None
 
     def __init__(self, eps: float = 0.5, min_samples: int = 5, metric="euclidean"):
         self.eps = eps

src/dbscan1d/version.py

@@ -1,8 +1,9 @@
 """Module for reporting the version of dbscan1d."""
+
 from importlib.metadata import PackageNotFoundError, version
 
 try:
     __version__ = version("dbscan1d")
 # package is not installed
-except PackageNotFoundError:  # NOQA
-    __version__ = "0.0.0"  # NOQA
+except PackageNotFoundError:
+    __version__ = "0.0.0"

tests/test_dbscan1d.py

@@ -3,6 +3,7 @@
 
 Requires sklearn.
 """
+
 import copy
 from itertools import product
 from pathlib import Path
@@ -122,7 +123,8 @@ def generate_test_data(num_points, centers=None):
         num_points, n_features=1, centers=centers, random_state=13
     )
     X = blobs.flatten()
-    np.random.shuffle(X)
+    rng = np.random.default_rng()
+    rng.shuffle(X)
     return X, blob_labels
 
 
@@ -139,18 +141,18 @@ class TestSKleanEquivalent:
 
     # define a small range of dbscan input params over which tests will
     # be parametrized
-    eps_values = [0.0001, 0.1, 0.5, 1, 2]
-    min_samples_values = [1, 2, 5, 15]
-    db_params = list(product(eps_values, min_samples_values))
+    eps_values = (0.0001, 0.1, 0.5, 1, 2)
+    min_samples_values = (1, 2, 5, 15)
+    db_params = tuple(product(eps_values, min_samples_values))
 
-    centers = [
+    centers = (
         np.array([0, 5, 10]),
         np.arange(10),
         np.array([1, 2, 3, 4, 5, 10]),
         np.array([1, 1.1, 1.2, 1.3, 1.4, 1.5]),
         2,
         7,
-    ]
+    )
 
     @pytest.fixture(scope="class", params=centers)
     def blobs(self, request):