Setup CI test pipeline (#7)

* BLD:
- first test of a GitHub CI for testing

* FIX:
- fixed wrong pip install and python versions in CI ?

* FIX:
- fixed loading of test files that do not contain test but dependencies that cannot be loaded ?

* BUG:
- escaped imports of unaccessible dependencies for test-file generation in GitHub CI tests by making the `if __name__ == "__main__"` include these imports and function definitions

* TST:
- test whether Cython build can be removed from GitHub actions for testing ?

* TST:
- reverted removal of Cython for testing

* PKG:
- moved the full package back into an `src`-folder

BLD:
- updated `pyproject.toml` and `setup.py` to account for the movement to the `src`-folder
- removed dedicated Cython build from CI test pipeline

* BLD:
- changed CI pipeline target branches

.github/workflows/python-package.yml

@@ -0,0 +1,33 @@
+name: Build and Test
+
+on:
+  push:
+    branches:
+      - main
+  pull_request:
+    branches:
+      - main
+      - develop
+
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.9", "3.10", "3.11", "3.12"]
+
+    steps:
+      - uses: actions/checkout@v2
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Install dependencies and Build package
+        run: |
+          python -m pip install --upgrade pip setuptools wheel
+          pip install .["git_ci"]
+
+      - name: Run tests
+        run: |
+          pytest -n=auto -x

pyproject.toml

@@ -21,10 +21,10 @@ include-package-data = true
 package-data = {"*" = ["AUTHORS.txt", "VERSION.txt"]}
 
 [tool.setuptools.dynamic]
-version = {file = "robust_hermite_ft/VERSION.txt"}
+version = {file = "src/robust_hermite_ft/VERSION.txt"}
 readme = {file = ["README.rst"]}
 dependencies = {file = "requirements/base.txt"}
-optional-dependencies = {fast = {file = "requirements/fast.txt"}, dev = {file = "requirements/dev.txt"}, examples = {file = "requirements/examples.txt"}}
+optional-dependencies = {fast = {file = "requirements/fast.txt"}, dev = {file = "requirements/dev.txt"}, examples = {file = "requirements/examples.txt"}, git_ci = {file = "requirements/git_ci.txt"}}
 
 [tool.isort]
 profile = "black"

pytest.ini

@@ -0,0 +1,2 @@
+[pytest]
+norecursedirs = tests/reference_files

requirements/git_ci.txt

@@ -0,0 +1,9 @@
+black
+coverage
+cython>=3.0.10
+cython-lint>=0.16.0
+numba>=0.55.0
+pytest
+pytest-cov
+pytest-xdist
+ruff

setup.py

@@ -10,12 +10,12 @@
 import Cython.Compiler.Options
 import numpy as np
 from Cython.Build import cythonize
-from setuptools import Extension, setup
+from setuptools import Extension, find_packages, setup
 
 # === Constants ===
 
 SOURCES = [
-    "robust_hermite_ft/hermite_functions/_c_hermite.pyx",
+    "src/robust_hermite_ft/hermite_functions/_c_hermite.pyx",
 ]
 
 # === Setup ===
@@ -41,8 +41,10 @@
 ]
 
 setup(
+    package_dir={"": "src"},
+    packages=find_packages("src"),
     ext_modules=cythonize(CY_MODULES, nthreads=1, annotate=True),
     package_data={"robust_hermite_ft": ["*.pxd"]},  # include pxd files
-    include_package_data=False,  # ignore other files
+    include_package_data=True,
     zip_safe=False,
 )

tests/reference_files/generate_hermfunc_references.py

@@ -14,18 +14,9 @@
 import json
 import os
 from dataclasses import asdict, dataclass, field
-from functools import partial
-from multiprocessing import Pool
-from time import perf_counter
-from typing import Dict, Tuple
+from typing import Dict, List, Tuple
 
 import numpy as np
-from sympy import Symbol as sp_Symbol
-from sympy import exp as sp_exp
-from sympy import pi as sp_pi
-from sympy import sqrt as sp_sqrt
-from sympy import symbols as sp_symbols
-from tqdm import tqdm
 
 # === Constants ===
 
@@ -61,7 +52,7 @@ class HermiteFunctionsParameters:
 
     n: int
     alpha: float
-    ns_for_single_function: list[int] = field(default_factory=list)
+    ns_for_single_function: List[int] = field(default_factory=list)
 
 
 @dataclass
@@ -86,132 +77,152 @@ class ReferenceHermiteFunctionsMetadata:
     x_values: np.ndarray
 
 
-# === Functions ===
+# === Main code ===
 
+if __name__ == "__main__":
 
-def _eval_sym_hermite_worker(
-    row_index: int,
-    x: np.ndarray,
-    x_sym: sp_Symbol,
-    n: int,
-    alpha: float,
-    expressions: np.ndarray,
-    num_digits: int,
-) -> Tuple[int, np.ndarray]:
-    """
-    Worker function to evaluate the Hermite functions at the given points ``x``.
-
-    """
-
-    # the Hermite functions are evaluated at the given points
-    hermite_function_values = np.empty(shape=n + 1, dtype=np.float64)
-
-    # the Hermite functions are evaluated using the recurrence relation
-    for iter_j in range(0, n + 1):
-        # the expression for the Hermite function is evaluated
-        hermite_expression = expressions[iter_j]
-        hermite_function_values[iter_j] = hermite_expression.subs(
-            x_sym, x[row_index] / alpha
-        ).evalf(n=num_digits)
-
-    return row_index, hermite_function_values
-
-
-def _eval_sym_dilated_hermite_function_basis(
-    x: np.ndarray,
-    n: int,
-    alpha: float,
-    num_digits: int = 16,
-) -> np.ndarray:
-    """
-    Evaluates the first ``n + 1`` dilated Hermite functions at the given points ``x``.
-    They are defined as
-
-    .. image:: docs/hermite_functions/equations/DilatedHermiteFunctions.png
-
-    Parameters
-    ----------
-    x : :class:`np.ndarray` of shape (m,)
-        The points at which the Hermite functions are evaluated.
-    n : :class:`int`
-        The order of the Hermite functions.
-    alpha : :class:`float`
-        The scaling factor of the independent variable ``x``.
-    num_digits : :class:`int`, default=16
-        The number of digits used in the symbolic evaluation of the Hermite functions.
-        For orders ``n >= 50`` and high ``x / alpha``-values, the symbolic evaluation
-        might be inaccurate. In this case, going to quadruple precision
-        (``n_digits~=32``) or higher might be necessary.
-
-    Returns
-    -------
-    hermite_function_basis : :class:`np.ndarray` of shape (m, n + 1)
-        The values of the first ``n + 1`` dilated Hermite functions evaluated at the
-        points ``x``.
-
-    """
-
-    # the Hermite functions are evaluated using their recurrence relation given by
-    # h_{n+1}(x) = sqrt(2 / (n + 1)) * x * h_{n}(x) - sqrt(n / (n + 1)) * h_{n-1}(x)
-    # with the initial conditions h_{-1}(x) = 0 and
-    # h_{0}(x) = pi**(-1/4) * exp(-x**2 / 2)
-    x_sym = sp_symbols("x")
-    hermite_expressions = np.empty(shape=(n + 1), dtype=object)
-
-    # the first two Hermite function expressions are defined with the involved Gaussian
-    # function not multiplied in yet to avoid the build-up of large expressions
-    h_i_minus_1 = 0
-    h_i = sp_exp(-(x_sym**2) / 2) / sp_sqrt(sp_sqrt(sp_pi))  # type: ignore
-    hermite_expressions[0] = h_i
-
-    # the Hermite functions are evaluated using the recurrence relation
-    for iter_j in tqdm(range(0, n), desc="Generating Hermite expressions", leave=False):
-        h_i_plus_1 = (
-            sp_sqrt(2 / (iter_j + 1)) * x_sym * h_i
-            - sp_sqrt(iter_j / (iter_j + 1)) * h_i_minus_1  # type: ignore
-        )
-        h_i_minus_1, h_i = h_i, h_i_plus_1
-        hermite_expressions[iter_j + 1] = h_i
-
-    # the Hermite functions are evaluated at the given points
-    hermite_function_basis = np.empty(shape=(x.size, n + 1), dtype=np.float64)
-
-    # the evaluation is done in parallel to speed up the process but a progress bar is
-    # used to keep track of the progress
-    with Pool() as pool:
-        worker = partial(
-            _eval_sym_hermite_worker,
-            x=x,
-            x_sym=x_sym,
-            n=n,
-            alpha=alpha,
-            expressions=hermite_expressions,
-            num_digits=num_digits,
-        )
-        results = list(
-            tqdm(
-                pool.imap(worker, range(0, x.size)),
-                total=x.size,
-                desc="Evaluating Hermite functions",
-                leave=False,
+    # === Imports ===
+
+    from functools import partial
+    from multiprocessing import Pool
+    from time import perf_counter
+
+    from sympy import Symbol as sp_Symbol
+    from sympy import exp as sp_exp
+    from sympy import pi as sp_pi
+    from sympy import sqrt as sp_sqrt
+    from sympy import symbols as sp_symbols
+    from tqdm import tqdm
+
+    # === Functions ===
+
+    def _eval_sym_hermite_worker(
+        row_index: int,
+        x: np.ndarray,
+        x_sym: sp_Symbol,
+        n: int,
+        alpha: float,
+        expressions: np.ndarray,
+        num_digits: int,
+    ) -> Tuple[int, np.ndarray]:
+        """
+        Worker function to evaluate the Hermite functions at the given points ``x``.
+
+        """
+
+        # the Hermite functions are evaluated at the given points
+        hermite_function_values = np.empty(shape=n + 1, dtype=np.float64)
+
+        # the Hermite functions are evaluated using the recurrence relation
+        for iter_j in range(0, n + 1):
+            # the expression for the Hermite function is evaluated
+            hermite_expression = expressions[iter_j]
+            hermite_function_values[iter_j] = hermite_expression.subs(
+                x_sym, x[row_index] / alpha
+            ).evalf(n=num_digits)
+
+        return row_index, hermite_function_values
+
+    def _eval_sym_dilated_hermite_function_basis(
+        x: np.ndarray,
+        n: int,
+        alpha: float,
+        num_digits: int = 16,
+    ) -> np.ndarray:
+        """
+        Evaluates the first ``n + 1`` dilated Hermite functions at the given points
+        ``x``.
+        They are defined as
+
+        .. image:: docs/hermite_functions/equations/DilatedHermiteFunctions.png
+
+        Parameters
+        ----------
+        x : :class:`np.ndarray` of shape (m,)
+            The points at which the Hermite functions are evaluated.
+        n : :class:`int`
+            The order of the Hermite functions.
+        alpha : :class:`float`
+            The scaling factor of the independent variable ``x``.
+        num_digits : :class:`int`, default=16
+            The number of digits used in the symbolic evaluation of the Hermite
+            functions.
+            For orders ``n >= 50`` and high ``x / alpha``-values, the symbolic
+            evaluation might be inaccurate. In this case, going to quadruple precision
+            (``n_digits~=32``) or higher might be necessary.
+
+        Returns
+        -------
+        hermite_function_basis : :class:`np.ndarray` of shape (m, n + 1)
+            The values of the first ``n + 1`` dilated Hermite functions evaluated at the
+            points ``x``.
+
+        """
+
+        # the Hermite functions are evaluated using their recurrence relation given by
+        # h_{n+1}(x) = sqrt(2 / (n + 1)) * x * h_{n}(x) - sqrt(n / (n + 1)) * h_{n-1}(x)
+        # with the initial conditions h_{-1}(x) = 0 and
+        # h_{0}(x) = pi**(-1/4) * exp(-x**2 / 2)
+        x_sym = sp_symbols("x")
+        hermite_expressions = np.empty(shape=(n + 1), dtype=object)
+
+        # the first two Hermite function expressions are defined with the involved
+        # Gaussian function not multiplied in yet to avoid the build-up of large
+        # expressions
+        h_i_minus_1 = 0
+        h_i = sp_exp(-(x_sym**2) / 2) / sp_sqrt(sp_sqrt(sp_pi))  # type: ignore
+        hermite_expressions[0] = h_i
+
+        # the Hermite functions are evaluated using the recurrence relation
+        for iter_j in tqdm(
+            range(0, n),
+            desc="Generating Hermite expressions",
+            leave=False,
+        ):
+            h_i_plus_1 = (
+                sp_sqrt(2 / (iter_j + 1)) * x_sym * h_i
+                - sp_sqrt(iter_j / (iter_j + 1)) * h_i_minus_1  # type: ignore
+            )
+            h_i_minus_1, h_i = h_i, h_i_plus_1
+            hermite_expressions[iter_j + 1] = h_i
+
+        # the Hermite functions are evaluated at the given points
+        hermite_function_basis = np.empty(shape=(x.size, n + 1), dtype=np.float64)
+
+        # the evaluation is done in parallel to speed up the process but a progress bar
+        # is used to keep track of the progress
+        with Pool() as pool:
+            worker = partial(
+                _eval_sym_hermite_worker,
+                x=x,
+                x_sym=x_sym,
+                n=n,
+                alpha=alpha,
+                expressions=hermite_expressions,
+                num_digits=num_digits,
+            )
+            results = list(
+                tqdm(
+                    pool.imap(worker, range(0, x.size)),
+                    total=x.size,
+                    desc="Evaluating Hermite functions",
+                    leave=False,
+                )
             )
-        )
-
-    # the results are stored in the matrix
-    for row_idx, row_values in results:
-        hermite_function_basis[row_idx, ::] = row_values
-
-    return hermite_function_basis / np.sqrt(alpha)
 
+        # the results are stored in the matrix
+        for row_idx, row_values in results:
+            hermite_function_basis[row_idx, ::] = row_values
 
-# === Main ===
+        return hermite_function_basis / np.sqrt(alpha)
 
-# this part generates NumPy binary files for the first 250 dilated Hermite functions
-# with different scaling factors evaluated at high precision for a series of 501 points
-# in the range [-45, 45]
-# NOTE: it is important that the number of points is odd to have a point at exactly 0
+    # === Test file generation ===
 
-if __name__ == "__main__":
+    # this part generates NumPy binary files for the first 250 dilated Hermite functions
+    # with different scaling factors evaluated at high precision for a series of 501
+    # points in the range [-45, 45]
+    # NOTE: it is important that the number of points is odd to have a point at
+    #       exactly 0
 
     # --- Setup ---