Jupyter Notebookのoutputsとexecution_countを消す（上書き＋バックアップ）

nb_remove_outputs.py

"""
Original: https://gist.github.com/damianavila/5305869
"""

import sys
import io
import os
import glob
from datetime import datetime
from os.path import dirname, basename
from shutil import move, copytree
from nbformat import read, write, NO_CONVERT

if __name__ == '__main__':
    TIMESTAMP=datetime.now().strftime('%Y%m%d-%H%M%S')

    nb_paths = glob.glob('books/**/*.ipynb', recursive=True)

    for nb_path in nb_paths:
        original_dir = dirname(nb_path)
        original_name = basename(nb_path)

        with io.open(nb_path, 'r', encoding='utf8') as f:
            nb = read(f, NO_CONVERT)

        # backup
        backup_dir = '/'.join(['backup', TIMESTAMP, original_dir])
        os.makedirs(backup_dir, exist_ok=True)
        move(nb_path, '{}/{}'.format(backup_dir, original_name))

        for cell in nb.cells:
            if cell.cell_type == 'code':
                cell.outputs = []
                cell.execution_count = None
        
        with io.open(nb_path, 'w', encoding='utf8') as f:
            write(nb, f, NO_CONVERT)