Add ruff for linting, remove flake8, remove isort, remove pylint (#2069)

sdv-dev · Jun 17, 2024 · 407114c · 407114c
1 parent a8c45ac
commit 407114c
Show file tree

Hide file tree

Showing 111 changed files with 4,495 additions and 6,154 deletions.
diff --git a/.gitignore b/.gitignore
@@ -1,3 +1,5 @@
+.vscode/
+
 # Byte-compiled / optimized / DLL files
 __pycache__/
 *.py[cod]

diff --git a/Makefile b/Makefile
@@ -99,14 +99,13 @@ check-dependencies: ## test if there are any broken dependencies
 	pip check
 
 .PHONY: lint
-lint: ## check style with flake8 and isort
+lint:
 	invoke lint
 
 .PHONY: fix-lint
-fix-lint: ## fix lint issues using autoflake, autopep8, and isort
-	find sdv tests -name '*.py' | xargs autoflake --in-place --remove-all-unused-imports --remove-unused-variables
-	autopep8 --in-place --recursive --aggressive sdv tests
-	isort --apply --atomic sdv tests
+fix-lint:
+	ruff check --fix .
+	ruff format
 
 
 # TEST TARGETS

diff --git a/pyproject.toml b/pyproject.toml
@@ -86,32 +86,7 @@ dev = [
     'Jinja2>=2,<4',
 
     # style check
-    'flake8>=3.7.7,<8',
-    'flake8-absolute-import>=1.0,<2',
-    'flake8-builtins>=1.5.3,<3',
-    'flake8-comprehensions>=3.6.1,<4',
-    'flake8-debugger>=4.0.0,<5',
-    'flake8-docstrings>=1.5.0,<2',
-    'flake8-eradicate>=1.1.0,<2',
-    'flake8-fixme>=1.1.1,<1.2',
-    'flake8-mock>=0.3,<1',
-    'flake8-multiline-containers>=0.0.18,<0.1',
-    'flake8-mutable>=1.2.0,<1.3',
-    'flake8-expression-complexity>=0.0.9,<0.1',
-    'flake8-print>=4.0.0,<4.1',
-    'flake8-pytest-style>=2.0.0,<3',
-    'flake8-quotes>=3.3.0,<4',
-    'flake8-sfs>=0.0.3,<2',
-    'flake8-variables-names>=0.0.4,<0.1',
-    'dlint>=0.11.0,<1',
-    'isort>=5.13.2,<6',
-    'pandas-vet>=0.2.3,<2024',
-    'pep8-naming>=0.12.1,<1',
-    'pydocstyle>=6.1.1,<7',
-
-    # fix style issues
-    'autoflake>=1.1,<3',
-    'autopep8>=1.4.3,<3',
+    'ruff>=0.4.5,<1',
 
     # distribute on PyPI
     'twine>=1.10.0,<6',
@@ -190,19 +165,55 @@ filename = "sdv/__init__.py"
 search = "__version__ = '{current_version}'"
 replace = "__version__ = '{new_version}'"
 
-[tool.isort]
-line_length = 99
-lines_between_types = 0
-multi_line_output = 4
-use_parentheses = true
-
-[tool.pydocstyle]
-convention = 'google'
-add-ignore = ['D105', 'D107', 'D407', 'D417']
-
 [tool.pytest.ini_options]
 addopts = "--ignore=pyproject.toml"
 
 [build-system]
 requires = ['setuptools', 'wheel']
 build-backend = 'setuptools.build_meta'
+
+[tool.ruff]
+preview = true
+line-length = 100
+indent-width = 4
+src = ["sdv"]
+exclude = [
+    "docs",
+    ".tox",
+    ".git",
+    "__pycache__",
+    ".ipynb_checkpoints"
+]
+
+[tool.ruff.lint]
+select = [
+    # Pyflakes
+    "F",
+    # Pycodestyle
+    "E",
+    "W",
+    "D200",
+    # isort
+    "I001",
+]
+ignore = [
+    "E501",
+    "D107",  # Missing docstring in __init__
+    "D417",   # Missing argument descriptions in the docstring, this is a bug from pydocstyle: https://github.com/PyCQA/pydocstyle/issues/449
+]
+
+[tool.ruff.format]
+quote-style = "single"
+indent-style = "space"
+preview = true
+docstring-code-format = true
+docstring-code-line-length = "dynamic"
+
+[tool.ruff.lint.isort]
+known-first-party = ["sdv"]
+
+[tool.ruff.lint.per-file-ignores]
+"__init__.py" = ["F401", "E402", "F403", "F405", "E501", "I001"]
+
+[tool.ruff.lint.pydocstyle]
+convention = "google"
diff --git a/scripts/release_notes_generator.py b/scripts/release_notes_generator.py
@@ -13,15 +13,15 @@
     'maintenance': 'Maintenance',
     'customer success': 'Customer Success',
     'documentation': 'Documentation',
-    'misc': 'Miscellaneous'
+    'misc': 'Miscellaneous',
 }
 ISSUE_LABELS = [
     'documentation',
     'maintenance',
     'internal',
     'bug',
     'feature request',
-    'customer success'
+    'customer success',
 ]
 NEW_LINE = '\n'
 GITHUB_URL = 'https://api.github.com/repos/sdv-dev/sdv'
@@ -30,14 +30,8 @@
 
 def _get_milestone_number(milestone_title):
     url = f'{GITHUB_URL}/milestones'
-    headers = {
-        'Authorization': f'Bearer {GITHUB_TOKEN}'
-    }
-    query_params = {
-        'milestone': milestone_title,
-        'state': 'all',
-        'per_page': 100
-    }
+    headers = {'Authorization': f'Bearer {GITHUB_TOKEN}'}
+    query_params = {'milestone': milestone_title, 'state': 'all', 'per_page': 100}
     response = requests.get(url, headers=headers, params=query_params)
     body = response.json()
     if response.status_code != 200:
@@ -52,17 +46,12 @@ def _get_milestone_number(milestone_title):
 
 
 def _get_issues_by_milestone(milestone):
-    headers = {
-        'Authorization': f'Bearer {GITHUB_TOKEN}'
-    }
+    headers = {'Authorization': f'Bearer {GITHUB_TOKEN}'}
     # get milestone number
     milestone_number = _get_milestone_number(milestone)
     url = f'{GITHUB_URL}/issues'
     page = 1
-    query_params = {
-        'milestone': milestone_number,
-        'state': 'all'
-    }
+    query_params = {'milestone': milestone_number, 'state': 'all'}
     issues = []
     while True:
         query_params['page'] = page

diff --git a/sdv/__init__.py b/sdv/__init__.py
@@ -16,8 +16,21 @@
 from types import ModuleType
 
 from sdv import (
-    constraints, data_processing, datasets, evaluation, io, lite, logging, metadata, metrics,
-    multi_table, sampling, sequential, single_table, version)
+    constraints,
+    data_processing,
+    datasets,
+    evaluation,
+    io,
+    lite,
+    logging,
+    metadata,
+    metrics,
+    multi_table,
+    sampling,
+    sequential,
+    single_table,
+    version,
+)
 
 __all__ = [
     'constraints',
@@ -33,7 +46,7 @@
     'sampling',
     'sequential',
     'single_table',
-    'version'
+    'version',
 ]
 
 

diff --git a/sdv/_utils.py b/sdv/_utils.py
@@ -1,4 +1,5 @@
 """Miscellaneous utility functions."""
+
 import operator
 import uuid
 import warnings
@@ -122,11 +123,7 @@ def _validate_datetime_format(column, datetime_format):
             Series of booleans, with True if the value matches the format, False if not.
     """
     pandas_datetime_format = datetime_format.replace('%-', '%')
-    datetime_column = pd.to_datetime(
-        column,
-        errors='coerce',
-        format=pandas_datetime_format
-    )
+    datetime_column = pd.to_datetime(column, errors='coerce', format=pandas_datetime_format)
     valid = pd.isna(column) | ~pd.isna(datetime_column)
 
     return set(column[~valid])
@@ -253,7 +250,7 @@ def check_sdv_versions_and_warn(synthesizer):
         public_missmatch = current_public_version != fitted_public_version
         enterprise_missmatch = current_enterprise_version != fitted_enterprise_version
 
-        if (public_missmatch or enterprise_missmatch):
+        if public_missmatch or enterprise_missmatch:
             static_message = (
                 'The latest bug fixes and features may not be available for this synthesizer. '
                 'To see these enhancements, create and train a new synthesizer on this version.'
@@ -346,23 +343,18 @@ def check_synthesizer_version(synthesizer, is_fit_method=False, compare_operator
     static_message = 'Downgrading your SDV version is not supported.'
     if is_fit_method:
         static_message = (
-            'Fitting this synthesizer again is not supported. '
-            'Please create a new synthesizer.'
+            'Fitting this synthesizer again is not supported. ' 'Please create a new synthesizer.'
         )
 
     fit_public_version = getattr(synthesizer, '_fitted_sdv_version', None)
     fit_enterprise_version = getattr(synthesizer, '_fitted_sdv_enterprise_version', None)
 
     is_public_lower = _compare_versions(
-        current_public_version,
-        fit_public_version,
-        compare_operator
+        current_public_version, fit_public_version, compare_operator
     )
 
     is_enterprise_lower = _compare_versions(
-        current_enterprise_version,
-        fit_enterprise_version,
-        compare_operator
+        current_enterprise_version, fit_enterprise_version, compare_operator
     )
 
     if is_public_lower and is_enterprise_lower:

diff --git a/sdv/constraints/__init__.py b/sdv/constraints/__init__.py
@@ -1,8 +1,19 @@
 """SDV Constraints module."""
+
 from sdv.constraints.base import Constraint
 from sdv.constraints.tabular import (
-    FixedCombinations, FixedIncrements, Inequality, Negative, OneHotEncoding, Positive, Range,
-    ScalarInequality, ScalarRange, Unique, create_custom_constraint_class)
+    FixedCombinations,
+    FixedIncrements,
+    Inequality,
+    Negative,
+    OneHotEncoding,
+    Positive,
+    Range,
+    ScalarInequality,
+    ScalarRange,
+    Unique,
+    create_custom_constraint_class,
+)
 
 __all__ = [
     'create_custom_constraint_class',
@@ -16,5 +27,5 @@
     'Negative',
     'Positive',
     'OneHotEncoding',
-    'Unique'
+    'Unique',
 ]