This script converts a Microsoft Excel spreadsheet to a UTF-8 CSV file.

convert_xls_to_utf8_csv.py

#!/usr/bin/env python
"""Convert a Microsoft Excel spreadsheet to a UTF-8 csv.

Usage:

    # Make sure requrirements are installed.
    $ sudo pip install xlrd backports.csv

    # Run script.
    $ python convert_xls_to_utf8_csv.py <spreadsheet>

The CSV will be saved to the same path as the original file, but with
'.csv' appended. For example, '~/spreadsheets/foo.xlsx' would be saved
to '~/spreadsheets/foo.xlsx.csv'.
"""
import sys
import csv
import json
import io

import xlrd
import backports.csv as csv



def convert_xls_to_dict(filepath, sheet_index=0):
    """Convert a Microsoft Word spreadsheet to a python dict."""
    workbook = xlrd.open_workbook(filepath)
    worksheet = workbook.sheet_by_index(sheet_index)
    first_row = list()
    for col in range(worksheet.ncols):
        first_row.append(worksheet.cell_value(0, col))
    data = list()
    for row in range(1, worksheet.nrows):
        elm = dict()
        for col in range(worksheet.ncols):
            elm_key = first_row[col].split(':', 1)[-1].lower()
            elm[elm_key] = worksheet.cell_value(row, col)
        data.append(elm)
    return data


if __name__ == '__main__':
    with io.open('{}.csv'.format(sys.argv[-1]), 'w', newline='', encoding='utf-8') as fh:
        writer = csv.writer(fh)
        for i, row in enumerate(convert_xls_to_dict(sys.argv[-1])):
            # Write header if first row.
            if i == 0:
                writer.writerow(row.keys())

            writer.writerow(row.values())