Pulling all of the tables from a Word document into JSON files

docxtables2json.py

#!/usr/bin/env python3

from sys import stdout, stderr, argv
import os.path
import json

try:
    from docx import Document
except ImportError:
    print("Maybe you should $ pip install python-docx", file=stderr)
    raise

for path in argv[1:]:
    directory, filename = os.path.split(path)
    basename, extension = os.path.splitext(filename)

    with open(path, 'rb') as input_file:
        document = Document(input_file)
        for index, table in enumerate(document.tables):
            headers = [cell.text for cell in table.rows[0].cells]
            outpath = os.path.join(directory, f"{basename}table-{index}.json")
            numrows = len(table.rows) - 1 # Content rows

            print(outpath, file=stdout)
            with open(outpath, 'w') as outputfile:
                outputfile.write('[') # Instead of building a massive JSON

                for index, row in enumerate(table.rows[1:]):
                    values = [cell.text for cell in row.cells]
                    data = dict(zip(headers, values))
                    json.dump(data, outputfile)
                    if index != numrows - 1:
                        outputfile.write(',')

                outputfile.write(']')