Generate diacritic fingerspelling dictionary for Plover

diacritics.py

# Save as diacritics.py, then run `python3 diacritics.py`

import unicodedata
import json

characters = [
    [  # Acute
        'KRAO*UT', '\u0301', 'á', 'é', 'í', 'ó', 'ú'
    ],
    [  # Grave
        'TKPWRA*F', '\u0300', 'à', 'è', 'ì', 'ò', 'ù'
    ],
    [  # Circumflex
        'KREURBG', '\u0302', 'â', 'ê', 'î', 'ô', 'û'
    ],
    [  # Umlaut
        '*UPLT', '\u0308', 'ä', 'ë', 'ï', 'ï', 'ö', 'ü'
    ],
    [  # Tilde
        '*LTD', '\u0303', 'ã', 'ñ', 'õ'
    ],
    [  # Ring
        '*RNG', '\u030A', 'å'
    ],
    [  # Cedilla
        'STKEUL', '\u0327', 'ç'
    ]
]
fingerspelling = {
    'a': 'A*',
    'c': 'KR*',
    'e': '*E',
    'i': '*EU',
    'n': 'TPH*',
    'o': 'O*',
    'u': '*U',
    'y': 'KWR*'
}
uppercase_fingerspelling_stroke_fragment = 'P'
unhandled = [
    'æ', 'ø', 'ß'
]

if __name__ == '__main__':
    the_dict = {}
    for character_set in characters:
        (diacritic_stroke, combining_unicode, *characters) = character_set

        the_dict[diacritic_stroke] = f'{{&{combining_unicode}}}'

        for character in characters:
            # é → e
            letter = unicodedata.normalize('NFKD', character).encode('ASCII', 'ignore')
            letter = str(letter, 'utf-8')

            # e → *E
            fingerspelling_stroke = fingerspelling[letter]

            # É
            uppercase_character = character.upper()

            # *E → *EP
            uppercase_fingerspelling_stroke = fingerspelling_stroke + \
                uppercase_fingerspelling_stroke_fragment

            the_dict[
                f'{fingerspelling_stroke}/{diacritic_stroke}'
            ] = f'{{>}}{{&{character}}}'
            the_dict[
                f'{uppercase_fingerspelling_stroke}/{diacritic_stroke}'
            ] = f'{{&{uppercase_character}}}'

    print(json.dumps(the_dict, ensure_ascii=False, indent=4))

Output.json

{
    "KRAO*UT": "{&́}",
    "A*/KRAO*UT": "{>}{&á}",
    "A*P/KRAO*UT": "{&Á}",
    "*E/KRAO*UT": "{>}{&é}",
    "*EP/KRAO*UT": "{&É}",
    "*EU/KRAO*UT": "{>}{&í}",
    "*EUP/KRAO*UT": "{&Í}",
    "O*/KRAO*UT": "{>}{&ó}",
    "O*P/KRAO*UT": "{&Ó}",
    "*U/KRAO*UT": "{>}{&ú}",
    "*UP/KRAO*UT": "{&Ú}",
    "TKPWRA*F": "{&̀}",
    "A*/TKPWRA*F": "{>}{&à}",
    "A*P/TKPWRA*F": "{&À}",
    "*E/TKPWRA*F": "{>}{&è}",
    "*EP/TKPWRA*F": "{&È}",
    "*EU/TKPWRA*F": "{>}{&ì}",
    "*EUP/TKPWRA*F": "{&Ì}",
    "O*/TKPWRA*F": "{>}{&ò}",
    "O*P/TKPWRA*F": "{&Ò}",
    "*U/TKPWRA*F": "{>}{&ù}",
    "*UP/TKPWRA*F": "{&Ù}",
    "KREURBG": "{&̂}",
    "A*/KREURBG": "{>}{&â}",
    "A*P/KREURBG": "{&Â}",
    "*E/KREURBG": "{>}{&ê}",
    "*EP/KREURBG": "{&Ê}",
    "*EU/KREURBG": "{>}{&î}",
    "*EUP/KREURBG": "{&Î}",
    "O*/KREURBG": "{>}{&ô}",
    "O*P/KREURBG": "{&Ô}",
    "*U/KREURBG": "{>}{&û}",
    "*UP/KREURBG": "{&Û}",
    "*UPLT": "{&̈}",
    "A*/*UPLT": "{>}{&ä}",
    "A*P/*UPLT": "{&Ä}",
    "*E/*UPLT": "{>}{&ë}",
    "*EP/*UPLT": "{&Ë}",
    "*EU/*UPLT": "{>}{&ï}",
    "*EUP/*UPLT": "{&Ï}",
    "O*/*UPLT": "{>}{&ö}",
    "O*P/*UPLT": "{&Ö}",
    "*U/*UPLT": "{>}{&ü}",
    "*UP/*UPLT": "{&Ü}",
    "*LTD": "{&̃}",
    "A*/*LTD": "{>}{&ã}",
    "A*P/*LTD": "{&Ã}",
    "TPH*/*LTD": "{>}{&ñ}",
    "TPH*P/*LTD": "{&Ñ}",
    "O*/*LTD": "{>}{&õ}",
    "O*P/*LTD": "{&Õ}",
    "*RNG": "{&̊}",
    "A*/*RNG": "{>}{&å}",
    "A*P/*RNG": "{&Å}",
    "STKEUL": "{&̧}",
    "KR*/STKEUL": "{>}{&ç}",
    "KR*P/STKEUL": "{&Ç}"
}