Merge pull request #335 from JamesParrott/pyshp-3.0.0-alpha-return-of-the-type-hints

Add Speed test.  Return of the type hints

Author: JamesParrott

.github/actions/test/action.yml

@@ -72,7 +72,7 @@ runs:
         name: PyShp_wheel_and_sdist
         path: dist
 
-    - name: Install PyShp from the wheel (built in prev step)
+    - name: Install PyShp from the wheel (downloaded in prev step)
       shell: bash
       working-directory: dist/
       run: |

.github/workflows/speed_test.yml

@@ -0,0 +1,70 @@
+
+name: Run the speed tests
+
+on:
+  push:
+  pull_request:
+  workflow_dispatch:
+
+jobs:
+
+
+  build_wheel_and_sdist:
+    runs-on: ubuntu-latest
+    steps:
+    - uses: actions/checkout@v4
+    - uses: actions/setup-python@v5
+    - name: Build wheel from the project repo
+      uses: ./.github/actions/build_wheel_and_sdist
+
+  run_speed_tests:
+    needs: build_wheel_and_sdist
+    strategy:
+      fail-fast: false
+      matrix:
+        python-version: [
+          "3.9",
+          "3.13",
+        ]
+        os: [
+          "windows-latest",
+          "ubuntu-24.04",
+        ]
+
+
+    runs-on: ${{ matrix.os }}
+    steps:
+    - uses: actions/setup-python@v5
+      with:
+        python-version: ${{ matrix.python-version }}
+
+    - name: Download wheel and sdist (built in previous job)
+      uses: actions/download-artifact@v4
+      with:
+        name: PyShp_wheel_and_sdist
+        path: dist
+
+    - name: Install PyShp + test deps from the wheel (downloaded in prev step)
+      shell: bash
+      working-directory: dist/
+      run: |
+        WHEEL_NAME=$(ls pyshp-*py3-none-any.whl)
+        python -m pip install $WHEEL_NAME[test]
+
+    - uses: actions/checkout@v4
+      with:
+        path: ./Pyshp
+
+    - name: Checkout shapefiles and zip file artefacts repo
+      uses: actions/checkout@v4
+      with:
+        repository: JamesParrott/PyShp_test_shapefile
+        path: ./PyShp_test_shapefile
+
+    - name: Run Speed tests.
+      env:
+        PYSHP_TEST_REPO: ./PyShp_test_shapefile
+      run: python ./Pyshp/run_benchmarks.py
+
+
+

run_benchmarks.py

@@ -0,0 +1,83 @@
+# Based on Taneli Hukkinen's https://github.com/hukkin/tomli-w/blob/master/benchmark/run.py
+
+from __future__ import annotations
+
+import functools
+import os
+import timeit
+from collections.abc import Callable
+from pathlib import Path
+from typing import Union
+
+import shapefile as shp
+
+# For shapefiles from https://github.com/JamesParrott/PyShp_test_shapefile
+DEFAULT_PYSHP_TEST_REPO = (
+    rf"{os.getenv('USERPROFILE')}\Coding\repos\PyShp_test_shapefile"
+)
+PYSHP_TEST_REPO = Path(os.getenv("PYSHP_TEST_REPO", DEFAULT_PYSHP_TEST_REPO))
+REPO_ROOT = Path(__file__).parent
+
+
+blockgroups_file = REPO_ROOT / "shapefiles" / "blockgroups.shp"
+edit_file = REPO_ROOT / "shapefiles" / "test" / "edit.shp"
+merge_file = REPO_ROOT / "shapefiles" / "test" / "merge.shp"
+states_provinces_file = PYSHP_TEST_REPO / "ne_10m_admin_1_states_provinces.shp"
+tiny_countries_file = PYSHP_TEST_REPO / "ne_110m_admin_0_tiny_countries.shp"
+gis_osm_natural_file = PYSHP_TEST_REPO / "gis_osm_natural_a_free_1.zip"
+
+
+def benchmark(
+    name: str,
+    run_count: int,
+    func: Callable,
+    col_width: tuple,
+    compare_to: float | None = None,
+) -> float:
+    placeholder = "Running..."
+    print(f"{name:>{col_width[0]}} | {placeholder}", end="", flush=True)
+    time_taken = timeit.timeit(func, number=run_count)
+    print("\b" * len(placeholder), end="")
+    time_suffix = " s"
+    print(f"{time_taken:{col_width[1]-len(time_suffix)}.3g}{time_suffix}", end="")
+    print()
+    return time_taken
+
+
+def open_shapefile_with_PyShp(target: Union[str, os.PathLike]):
+    with shp.Reader(target) as r:
+        for shapeRecord in r.iterShapeRecords():
+            pass
+
+
+READER_TESTS = {
+    "Blockgroups": blockgroups_file,
+    "Edit": edit_file,
+    "Merge": merge_file,
+    "States_35MB": states_provinces_file,
+    "Tiny Countries": tiny_countries_file,
+    "GIS_OSM_zip_10MB": gis_osm_natural_file,
+}
+
+
+def run(run_count: int) -> None:
+    col_width = (21, 10)
+    col_head = ("parser", "exec time", "performance (more is better)")
+    # Load files to avoid one off delays that only affect first disk seek
+    for file_path in READER_TESTS.values():
+        file_path.read_bytes()
+    print(f"Running benchmarks {run_count} times:")
+    print("-" * col_width[0] + "---" + "-" * col_width[1])
+    print(f"{col_head[0]:>{col_width[0]}} | {col_head[1]:>{col_width[1]}}")
+    print("-" * col_width[0] + "-+-" + "-" * col_width[1])
+    for test_name, target in READER_TESTS.items():
+        benchmark(
+            f"Read {test_name}",
+            run_count,
+            functools.partial(open_shapefile_with_PyShp, target=target),
+            col_width,
+        )
+
+
+if __name__ == "__main__":
+    run(1)