feature : add export to py feature

recursively convert all ipynb files
exclude utils/ and tests/ folders

Author: beachdweller

utils/export_to_py.py

@@ -0,0 +1,96 @@
+'''
+Utility to export a Python script from a Jupyter notebook.
+Convert one ipynb or all recursively.
+'''
+
+import argparse
+import multiprocessing as mp
+import pathlib
+import subprocess
+import sys
+
+from typing import List
+
+
+import nbformat
+import nbconvert
+
+
+def parse_argv(argv: List[str]) -> argparse.Namespace:
+    parser = argparse.ArgumentParser(description='Export Jupyter notebooks to Python scripts.')
+    parser.add_argument('notebook', type=str, nargs='*', help='Path to the notebook(s) to convert.')
+    parser.add_argument('--destination', type=pathlib.Path, default=get_proj_folder() / 'temp', help='Destination directory for the converted scripts.')
+    parser.add_argument('--recursive', action='store_true', help='Recursively convert all notebooks in subdirectories.')
+    return parser.parse_args(argv[1:])
+
+
+def get_proj_folder() -> pathlib.Path:
+    """Get the project root folder."""
+    return pathlib.Path(__file__).parent.parent.resolve()
+
+
+def main(argv:List[str]):
+    args = parse_argv(argv)
+    notebooks = args.notebook
+
+    # get list of notebooks if not specified
+    if not notebooks:
+        notebooks = [
+            path for path in get_proj_folder().rglob('*.ipynb')
+        ] if args.recursive else [
+            path for path in get_proj_folder().glob('*.ipynb')
+        ]
+
+    # exclude notebooks in the utils and tests folders
+    notebooks = [
+        notebook for notebook in notebooks
+        if (
+            ('utils' not in notebook.parts)
+        and
+            ('tests' not in notebook.parts)
+        )
+    ]
+
+    converter = nbconvert.PythonExporter()
+
+    print(f'Using {mp.cpu_count()} processes for conversion.')
+    pool = mp.Pool(mp.cpu_count())
+    results = pool.starmap(
+        convert,
+        zip(
+            notebooks,
+            [args.destination]*len(notebooks),
+            [converter]*len(notebooks),
+        )
+    )
+    pool.close()
+    pool.join()
+
+    print(f'Converted {len(results) - results.count(None)} / {len(results)} notebooks to Python scripts.')
+
+
+def convert(notebook:pathlib.Path, destination:pathlib.Path, converter:nbconvert.Exporter) -> subprocess.CompletedProcess:
+    """Convert a single notebook to a Python script."""
+    if not notebook.is_file() or not notebook.suffix == '.ipynb':
+        print(f'Skipping non-notebook file: {notebook.relative_to(get_proj_folder())}')
+        result = None
+    else:
+        nb = nbformat.read(notebook, as_version=nbformat.NO_CONVERT)
+        try:
+            nbformat.validate(nb)
+        except nbformat.ValidationError as e:
+            print(f'Skipping invalid notebook: {notebook.relative_to(get_proj_folder())}')
+            result = None
+        else:
+            script_path = destination / ((notebook.relative_to(get_proj_folder())).with_suffix('.py'))
+            script_path.parent.mkdir(exist_ok=True)
+            print(f'Converting {notebook.relative_to(get_proj_folder())} to {script_path.relative_to(get_proj_folder())}')
+            output, resources_dict = nbconvert.export(converter, nb, output=script_path)
+            script_path.write_text(output, encoding='utf-8')
+            result = output
+    return result
+
+
+if "__main__" == __name__:
+    # Run the main function with command line arguments
+    main(sys.argv)