Script to copy KanjiVG characters and half the canvas size.

halfwidthify.py

#! /usr/bin/env python3
# -*- coding: utf-8 ; mode: python -*-
#
# © Copyright 2015 Roland Sieker <ospalh@gmail.com>
#
# License: GNU AGPL, version 3 or later;
# http://www.gnu.org/licenses/agpl.html

"""
"""

import argparse
import re
import unicodedata as ucd

fw_box = '<svg xmlns="http://www.w3.org/2000/svg" width="109" height="109" viewBox="0 0 109 109">'
hw_box = '<svg xmlns="http://www.w3.org/2000/svg" width="54.5" height="109" viewBox="27.25 0 54.5 109">'

def munge_file(args):
    fw_num = ord(args.kana)
    fw_code = '{0:05x}'.format(fw_num)
    fw_char = chr(fw_num)
    fw_name = ucd.name(fw_char)
    fw_file_name = '{}.svg'.format(fw_code)
    # N.B.: There is no fixed offset between full-width and half-width
    # katakana. Somehow we need a list or have to do this one by one.
    hw_num = int(args.hw_code, 16)
    hw_code = '{0:05x}'.format(hw_num)
    hw_char = chr(hw_num)
    hw_name = ucd.name(hw_char)
    hw_file_name = '{}.svg'.format(hw_code)
    if fw_name not in hw_name:
        raise RuntimeError('name miss match for {}'.format(fw_name))
        # I had to take this test out because we don’t have characters
        # 03099 　゙ and 0309a, ゚, and i just used バ and パ instead
        # and cut out the ハ.
    print('Full width file: {}'.format(fw_file_name))
    with open(fw_file_name) as fw_file:
        with open(hw_file_name, 'w') as hw_file:
            for line in fw_file.readlines():
                line = line.replace(fw_box, hw_box)
                hw_file.write(
                    re.sub(fw_code, hw_code, line))


if __name__ == '__main__':
    parser = argparse.ArgumentParser(
        description=u"""Copy files to half-width kana . """)
    parser.add_argument('--kana', '-k', type=str)
    parser.add_argument('--hw_code', '-c', type=str)
    args = parser.parse_args()
    munge_file(args)