Use trusted publishing for pypi (#57)

* feat: use trusted publishing for pypi

* feat: switch to pyproject.toml

* fix: replace np.NAN with np.nan

* fix: flake8 configuration

* fix: remove numpy types with underscores

* fix

.flake8

@@ -0,0 +1,22 @@
+[flake8]
+show-source = True
+max-complexity = 10
+hang-closing = False
+max-line-length = 88
+extend-select =
+    B950
+
+extend-ignore =
+    E203
+    E501
+
+extend-exclude =
+    .git
+    __pycache__
+    build
+    dist
+    doc
+
+per-file-ignores =
+    __init__.py: F401
+    gutenTAG/api/*: F401

.github/workflows/build.yml

@@ -143,6 +143,10 @@ jobs:
     if: github.event_name == 'push' && startsWith(github.ref, 'refs/tags')
     needs: [build-source-distribution, build-manylinux-wheels]
     runs-on: ubuntu-latest
+    environment: release
+    permissions:
+      # IMPORTANT: this permission is mandatory for trusted publishing
+      id-token: write
 
     steps:
     - name: Download packages
@@ -154,7 +158,4 @@ jobs:
         find dist -type f -exec mv {} dist/ \;
         find dist/* -type d -exec rmdir {} \; || true
     - name: Publish package to PyPi
-      uses: pypa/[email protected]
-      with:
-        user: __token__
-        password: ${{ secrets.PYPI_API_TOKEN }}
+      uses: pypa/[email protected]

generation_configs/multivariate-test-cases.py

@@ -217,7 +217,7 @@ def gen_bo_diversity_series():
             for bo, times in zip(bo_tuple, times_mapping[i]):
                 bo_defs.extend(random_bo(bo) for _ in range(times))
             for anom_pos in range(i):
-                pos_begin = np.sum(times_mapping[i][:anom_pos], dtype=np.int_)
+                pos_begin = np.sum(times_mapping[i][:anom_pos], dtype=np.int32)
                 pos_end = pos_begin + times_mapping[i][anom_pos]
                 pos = np.random.randint(pos_begin, pos_end)
                 timeseries.append(

gutenTAG/_version.py

@@ -1 +1 @@
-__version__: str = "1.4.1"
+__version__: str = "1.4.2"

gutenTAG/addons/timeeval.py

@@ -129,7 +129,7 @@ def _process_timeseries(
         dataset["trend"] = (
             config.get(BASE_OSCILLATION, {})
             .get(PARAMETERS.TREND, {})
-            .get(PARAMETERS.KIND, np.NAN)
+            .get(PARAMETERS.KIND, np.nan)
         )
         dataset["period_size"] = TimeEvalAddOn._calc_period_size(
             config.get(BASE_OSCILLATION, config.get(BASE_OSCILLATIONS, [{}])),
@@ -142,8 +142,8 @@ def _set_global_vals(df: pd.DataFrame) -> None:
         df["collection_name"] = "GutenTAG"
         df["dataset_type"] = "synthetic"
         df["datetime_index"] = False
-        df["split_at"] = np.NAN
-        df["stationarity"] = np.NAN
+        df["split_at"] = np.nan
+        df["stationarity"] = np.nan
 
     @staticmethod
     def _calc_contamination(anomalies: Iterable[Dict], ts_length: int) -> float:
@@ -180,7 +180,7 @@ def _calc_period_size(
                 BASE_OSCILLATION_NAMES.DIRICHLET,
                 BASE_OSCILLATION_NAMES.MLS,
             ]:
-                periods.append(np.NAN)
+                periods.append(np.nan)
             elif kind in [
                 BASE_OSCILLATION_NAMES.SINE,
                 BASE_OSCILLATION_NAMES.COSINE,

gutenTAG/base_oscillations/custom_input.py

@@ -101,8 +101,8 @@ def generate_only_base(
                 "Number of rows in the input timeseries file is less than the desired length"
             )
         col_type = df.dtypes[0]
-        if col_type != np.float_:
-            df = df.astype(np.float_)
+        if col_type != np.float64:
+            df = df.astype(np.float64)
             warnings.warn(
                 f"Input data was of {col_type} type and has been automatically converted to float."
             )

gutenTAG/generator/overview.py

@@ -22,7 +22,7 @@ def _decode_numpy_types(self, obj: Any) -> Any:
         if isinstance(
             obj,
             (
-                np.int_,
+                int,
                 np.intc,
                 np.intp,
                 np.int8,
@@ -37,16 +37,16 @@ def _decode_numpy_types(self, obj: Any) -> Any:
         ):
             return int(obj)
 
-        elif isinstance(obj, (np.float_, np.float16, np.float32, np.float64)):
+        elif isinstance(obj, (float, np.float16, np.float32, np.float64)):
             return float(obj)
 
-        elif isinstance(obj, (np.complex_, np.complex64, np.complex128)):
+        elif isinstance(obj, (np.complex64, np.complex128, np.complex256)):
             return {"real": obj.real, "imag": obj.imag}
 
         elif isinstance(obj, (np.ndarray,)):
             return obj.tolist()
 
-        elif isinstance(obj, (np.bool_)):
+        elif isinstance(obj, (bool, np.bool8)):
             return bool(obj)
 
         elif isinstance(obj, (np.void)):

pyproject.toml

@@ -0,0 +1,66 @@
+[project]
+name = "timeeval-GutenTAG"
+description = "A good Timeseries Anomaly Generator."
+license = {file="LICENSE"}
+authors= [
+    {"name" = "Phillip Wenig", email = "[email protected]"},
+    {"name" = "Sebastian Schmidl", email = "[email protected]"}
+]
+classifiers=[
+    "License :: OSI Approved :: MIT License",
+    "Programming Language :: Python",
+    "Programming Language :: Python :: 3",
+    "Programming Language :: Python :: 3.7",
+    "Programming Language :: Python :: 3.8",
+    "Programming Language :: Python :: 3.9",
+    "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Typing :: Typed",
+    "Topic :: Software Development",
+    "Topic :: Scientific/Engineering",
+    "Intended Audience :: Science/Research",
+    "Intended Audience :: Developers",
+]
+dynamic = ["readme", "version", "requires-python", "scripts", "dependencies"]
+
+[build-system]
+requires = ["setuptools"]
+build-backend = "setuptools.build_meta"
+
+[tool.coverage.run]
+branch = false
+parallel = true
+omit = ["**/__main__.py"]
+
+[tool.coverage.report]
+exclude_lines = [
+    "pragma: no cover",
+    "raise NotImplementedError",
+    "\\.\\.\\.",
+    "def __repr__",
+    "pass",
+    "if __name__ == .__main__.:",
+]
+fail_under = 80
+skip_covered = true
+sort = "Cover"
+
+[tool.mypy]
+python_version = "3.9"
+show_error_codes = true
+
+[[tool.mypy.overrides]]
+module = [
+    "sklearn.*",
+    "matplotlib.*",
+    "pandas.*",
+    "tqdm.*",
+    "psutil.*",
+    "prts.*",
+    "statsmodels.*",
+    "joblib.*",
+    "numpyencoder.*",
+    "scipy.*",
+    "neurokit2.*"
+]
+ignore_missing_imports = true

setup.py

@@ -108,28 +108,9 @@ def run(self):
     setup(
         name=f"timeeval-{PYTHON_NAME}",
         version=VERSION,
-        description="A good Timeseries Anomaly Generator.",
         long_description=README,
         long_description_content_type="text/markdown",
-        author="Phillip Wenig and Sebastian Schmidl",
-        author_email="[email protected]",
         url="https://github.com/TimeEval/gutentag",
-        license="MIT",
-        classifiers=[
-            "License :: OSI Approved :: MIT License",
-            "Programming Language :: Python",
-            "Programming Language :: Python :: 3",
-            "Programming Language :: Python :: 3.7",
-            "Programming Language :: Python :: 3.8",
-            "Programming Language :: Python :: 3.9",
-            "Programming Language :: Python :: 3.10",
-            "Programming Language :: Python :: 3.11",
-            "Typing :: Typed",
-            "Topic :: Software Development",
-            "Topic :: Scientific/Engineering",
-            "Intended Audience :: Science/Research",
-            "Intended Audience :: Developers",
-        ],
         packages=find_packages(exclude=("tests", "tests.*")),
         package_data={"gutenTAG": ["py.typed", "config/schema/*"]},
         install_requires=REQUIRED,

tests/test_base_oscillations/test_custom_input.py

@@ -154,7 +154,7 @@ def test_read_floats(self):
                     use_column_test=tpe,
                 )
             #  data is properly converted
-            self.assertEqual(timeseries.dtype, np.float_)
+            self.assertEqual(timeseries.dtype, np.float64)
 
     def test_convert_to_float_with_warning(self):
         for tpe in ["ints", "bools"]:
@@ -168,7 +168,7 @@ def test_convert_to_float_with_warning(self):
             # warning is raised
             self.assertRegex(str(w.warning), "automatically converted to float")
             #  data is properly converted
-            self.assertEqual(timeseries.dtype, np.float_)
+            self.assertEqual(timeseries.dtype, np.float64)
 
     def test_error_on_string_type(self):
         with self.assertRaises(ValueError) as e:

tests/test_generator/test_overview_sanitizer.py

@@ -12,18 +12,18 @@ def setUp(self) -> None:
             "timeseries": [
                 {
                     "name": "ecg",
-                    "length": np.int_(100),
-                    "base-oscillations": [{"kind": "ecg", "frequency": np.float_(2)}],
+                    "length": np.int32(100),
+                    "base-oscillations": [{"kind": "ecg", "frequency": np.float64(2)}],
                     "anomalies": [
                         {
                             "position": "middle",
-                            "length": np.int_(40),
-                            "channel": np.int_(0),
+                            "length": np.int32(40),
+                            "channel": np.int32(0),
                             "kinds": [
                                 {
                                     "kind": "pattern-shift",
-                                    "shift_by": np.int_(5),
-                                    "transition_window": np.int_(10),
+                                    "shift_by": np.int32(5),
+                                    "transition_window": np.int32(10),
                                     "some_array": np.array([1.2, 2.3]),
                                 }
                             ],