A function for translating romaji into either hiragana or katana (but no way to do it reverse so far)

transliteration.py

prefixTable = {
    ""  : 0 ,  "k" : 5 ,  "s" : 10,  "t" : 15,  "n" : 20,
    "h" : 25,  "m" : 30,  "y" : 35,  "r" : 40,  "w" : 45,
    "g" : 50,  "z" : 55,  "d" : 60,  "p" : 65,  "b" : 70,
    "j" : 75,  "ky": 80,  "sh": 85,  "ny": 90,  "hy": 95,
    "py": 100, "by": 105, "my": 110, "ch": 115, "ry": 120, 
    "gy": 125, "f" : 130, "ts": 135, "v" : 140, 
    "elongations"  : 145,
    "nn" : 150, ".": 151, "," : 152, "smalltsu" : 153, " " : 154,
    "oh" : 155 
    }

# prefixTable.setdefault()
# Mostly for readability. I.E. if x in vowels:
vowels           = {"a": 0, "i": 1, "u": 2, "e": 3, "o": 4}
doubleConsonants = ["k", "p", "s", "t"]
beginningLetters = ["k", "s", "t", "n", "h", "m", "y", "r", "w",
                    "g", "z", "d", "p", "b", "j", "c", "f", "v"]
secondLetters    = ["y", "h", "s"] # No, there isn't a lot
pronouncination  = [".", ",", " "]

transliterable = " abcdefghijklmnopqrstuvwxyz,."

# ゃ ゅ ょ
hiraganaTable = [
    "あ", "い", "う", "え", "お", # No consonants
    "か", "き", "く", "け", "こ", # K 
    "さ", "?",  "す", "せ", "そ", # S
    "た", "?",  "?",  "て", "と", # T
    "な", "に", "ぬ", "ね", "の", # N
    "は", "ひ", "?",  "へ", "ほ", # H
    "ま", "み", "む", "め", "も", # M
    "や", "?",  "ゆ", "?",  "よ", # Y
    "ら", "り", "る", "れ", "ろ", # R
    "わ", "?",  "?",  "?", "?", # W 
    "が", "ぎ", "ぐ", "げ", "ご", # G
    "ざ", "?",  "ず", "ぜ", "ぞ", # Z
    "だ", "?",  "?",  "で", "ど", # D
    "ぱ", "ぴ", "ぷ", "ぺ", "ぽ", # P
    "ば", "び", "ぶ", "べ", "ぼ", # B

    "じゃ", "じ", "じゅ", "?", "じょ", # J
    "きゃ", "?",  "きゅ", "?", "きょ", # Ky
    "しゃ", "し", "しゅ", "?", "しょ", # Sh
    "にゃ", "?",  "にゅ", "?", "にょ", # Ny
    "ひゃ", "?",  "ひゅ", "?", "ひょ", # Hy
    "ぴゃ", "?",  "ぴゅ", "?", "ぴょ", # Py
    "びゃ", "?",  "びゅ", "?", "びょ", # By
    "みゃ", "?",  "みゅ", "?", "みょ", # My
    "ちゃ", "ち", "ちゅ", "?", "ちょ", # Ch
    "りゃ", "?",  "りゅ", "?", "りょ", # Ry
    "ぎゃ", "?",  "ぎゅ", "?", "ぎょ", # Gy

    "?", "?", "ふ", "?", "?", # F
    "?", "?", "つ", "?", "?", # Ts
    "?", "?", "?",  "?", "?", # V, which doesn't exist 
                              #  in Hiragana.


    "あ", "い", "う", "い", "う", # Elongations
    "ん", "。", "、", "っ", " ",
    "を", "?" # hiraganaTable[-1]
    ]

katakanaTable = [
    "ア", "イ", "ウ", "エ", "オ", # No consonants
    "カ", "キ", "ク", "ケ", "コ", # K 
    "サ", "?",  "ス", "セ", "ソ", # S
    "タ", "ティ","トゥ","テ","ト", # T
    "ナ", "ニ", "ヌ", "ネ", "ノ", # N
    "ハ", "ヒ", "?",  "ヘ", "ホ", # H
    "マ", "ミ", "ム", "メ", "モ", # M
    "ヤ", "?",  "ユ", "?",  "ヨ", # Y
    "ラ", "リ", "ル", "レ", "ロ", # R
    "ワ", "ウィ","?", "ウェ","ウォ", # W 
    "ガ", "ギ", "グ", "ゲ", "ゴ", # G
    "ザ", "?",  "ズ", "ゼ", "ゾ", # Z
    "ダ", "ディ","デュ","デ","ド", # D
    "パ", "ピ", "プ", "ペ", "ポ", # P
    "バ", "ビ", "ブ", "ベ", "ボ", # B

    "ジャ", "ジ", "ジュ", "ジェ","ジョ", # J
    "キャ", "?",  "きゅ", "?", "キョ",  # Ky
    "シャ", "シ", "シュ", "シェ","ショ", # Sh
    "ニャ", "?",  "ニュ", "?", "ニョ", # Ny
    "ヒャ", "?",  "ヒュ", "?", "ヒョ", # Hy
    "ピャ", "?",  "ピュ", "?", "ピョ", # Py
    "ビャ", "?",  "ビュ", "?", "ビョ", # By
    "ミャ", "?",  "ミュ", "?", "ミョ", # My
    "チャ", "チ", "チュ", "チェ", "チョ", # Ch
    "リャ", "?",  "リュ", "?", "リョ", # Ry
    "ギャ", "?",  "ギュ", "?", "ギョ", # Gy

    "ファ", "フィ", "フ", "フェ", "フォ", # F
    "ツァ", "ツィ", "ツ", "ツェ", "ツォ", # Ts
    "ヴァ", "ヴィ", "?",  "ヴェ", "ヴォ", # V

    "ー", "ー", "ー", "ー", "ー", # Elongations
    "ン", "。", "、", "ッ", " ",  # etc.
    "ヲ", "?" # katakanaTable[-1]
    ]

hir = hiraganaTable
kat = katakanaTable

def parseJ(romaji, writing):
    japanese = "\0"
    romaji   = romaji.lower()
    c        = 0

    while c < len(romaji):
        vowel    = ''
        posbe    = ''
        basiclen = 0

        # If its a space, ,/., or a non-letter character,
        #  do that first
 
        if romaji[c] in pronouncination:
            japanese += writing[prefixTable[romaji[c]]]
            c += 1
            continue
        if romaji[c] not in transliterable:
            japanese += romaji[c]
            c += 1
            continue

        # Special Particles:
        # These particles must be surrounded by a space, a comma,
        #  a period, or something that is not a letter (whitespace
        #  for example)
        # Wa (spelled as ha)
        if (
            c + 2 < len(romaji) and
            (japanese[-1] in pronouncination 
                or japanese[-1] not in transliterable)  and
            (romaji[c+2]  in pronouncination 
                or romaji[c+2] not in transliterable)   and
            (romaji[c] + romaji[c+1]) == "wa" 
        ):
            japanese += writing[25] 
            c += 2
            continue
        # (w)o
        elif (
            c + 1 < len(romaji) and
            (japanese[-1] in pronouncination 
                or japanese[-1] not in transliterable)  and
            (romaji[c+1]  in pronouncination 
                or romaji[c+1] not in transliterable)   and
            romaji[c]    == "o"
        ):
            japanese += writing[155]
            c += 1
            continue


        # Vowel only
        if romaji[c] in vowels:
            basiclen = 1
            japanese += writing[vowels[romaji[c]]]
        # one or two consonsant then a vowel
        elif romaji[c] in beginningLetters:
            posbe = romaji[c]
            basiclen = 1

            # Test to see if another consonsant follows it
            if c + 1 < len(romaji) and romaji[c+1] in secondLetters:
                basiclen = 2
                posbe   += romaji[c+1]

            # Then look for a vowel
            if (c + basiclen < len(romaji) and 
                romaji[c + basiclen] in vowels
            ):
                append = writing[prefixTable[posbe] 
                               + vowels[romaji [c + basiclen]]]
                japanese += append
                #c += 1 + basiclen
                basiclen += 1
            else:
                japanese += "?"
                c += 1
                continue  
        else:
            japanese += "?"
            c += 1
            continue

        # elongation (only one allowed - maybe allow more for onomotopoiea?)
        if (c + basiclen < len(romaji) and 
            romaji[c + basiclen] in vowels and
            romaji[c + basiclen - 1] == romaji[c + basiclen]
        ):
            index  = (prefixTable["elongations"] 
                      + vowels[romaji[c + basiclen]])
            append = writing[index]
            japanese += append
            basiclen += 1

        # Is there an n after it and there's no vowel?
        if (c + basiclen + 1 < len(romaji) and
            romaji[c + basiclen] == "n" and
            romaji[c + basiclen + 1] not in vowels
        ):
            n = writing[150]
            japanese += n
            basiclen += 1
        # Is there an n /and/ its the last thing there is?
        elif (c + basiclen == len(romaji) - 1 and
            romaji[c + basiclen] == "n"
        ):
            n = writing[150]
            japanese += n
            basiclen += 1
        # Is it a double consonant?
        elif (c + basiclen + 1 < len(romaji) and
              romaji[c + basiclen] == romaji[c + basiclen + 1] and
              romaji[c + basiclen] in doubleConsonants
        ):
            japanese += writing[153]
            basiclen += 1
 
        c += basiclen        
    # while loop end is here

    return japanese

def tests():
    print (hiraganaTable[prefixTable["ky"] + vowels["o"]])
    print (hiraganaTable[prefixTable["m"]  + vowels["u"]]) 
    print (parseJ(" wa o aieuo", hiraganaTable))
    print (parseJ(" wa o aieuo", katakanaTable))
    print (parseJ("wa o aieuo ha sa hasa hya tsu tsi", hiraganaTable))
    print (parseJ("wa o aieuo ha sa hasa hya tsu tsi", katakanaTable))
    print (parseJ("myaa kyoo", hiraganaTable))
    print (parseJ("myaa kyoo", katakanaTable))
    print (parseJ("kyuushu", hir))
    print (parseJ("kyuushu", kat))
    print (parseJ("konohana sakuya ninja nippon", hir))
    print (parseJ("konohana sakuya ninja nippon", kat))

    print (parseJ(" o\n", hir))

    # My name is Mike Solomon
    print (
        parseJ("Watashi no onamae wa ", hir) +
        parseJ("Maiku soromoun ", kat) +
        parseJ("desu.", hir)
    )

    # My hobby is ninjutsu
    print (parseJ("\nWatashi no shumi wa ninjutsu desu.", hir))

    # One plus 2 is...\n ...3!
    print (parseJ("1 + 2 wa...\n ...san desu!", hir))

if __name__ == "__main__":
    tests()