ncmr-songs/wip/274-romaji.py

#!/usr/bin/env python3
"""Fill the Romaji column of 274-dev.md from its Kana column.

Kana is the source of truth. Romaji is generated wāpuro-style — no macrons,
since they are awkward to type: おう→ou, ゅう→yuu, っ doubles the following
consonant, ー repeats the preceding vowel, ん→n' before a vowel or y.

The trailing generic is split off and lowercased (Kawabata hashi, Houshin jinja,
Takinoue kosenkyou); 丁目 take a hyphen (Shinmachi 3-choume). Compound proper
stems stay as one token, because kana carries no word boundary — add them to
MANUAL below if you want a break.

    python3 274-romaji.py            # rewrite the Romaji column
    python3 274-romaji.py --check    # report differences, change nothing

Rows with no kana are left with an empty Romaji cell rather than a guess.
"""
import io, os, re, sys

HERE = os.path.dirname(os.path.abspath(__file__))
DOC = os.path.join(HERE, '274-dev.md')

# stems that must not have a trailing 川 split off — it is part of the name
NOSPLIT = {'のぼりかわ', 'にいかわ', 'しんかわ', 'みかわ', 'たきさわがわ'}
# hand-set forms that the mechanical rules cannot derive
MANUAL = {}


import re
Y={'きゃ':'kya','きゅ':'kyu','きょ':'kyo','しゃ':'sha','しゅ':'shu','しょ':'sho',
'ちゃ':'cha','ちゅ':'chu','ちょ':'cho','にゃ':'nya','にゅ':'nyu','にょ':'nyo',
'ひゃ':'hya','ひゅ':'hyu','ひょ':'hyo','みゃ':'mya','みゅ':'myu','みょ':'myo',
'りゃ':'rya','りゅ':'ryu','りょ':'ryo','ぎゃ':'gya','ぎゅ':'gyu','ぎょ':'gyo',
'じゃ':'ja','じゅ':'ju','じょ':'jo','びゃ':'bya','びゅ':'byu','びょ':'byo',
'ぴゃ':'pya','ぴゅ':'pyu','ぴょ':'pyo','ぢゃ':'ja','ぢゅ':'ju','ぢょ':'jo',
'しぇ':'she','ちぇ':'che','じぇ':'je','てぃ':'ti','でぃ':'di','ふぁ':'fa',
'ふぃ':'fi','ふぇ':'fe','ふぉ':'fo','うぃ':'wi','うぇ':'we','うぉ':'wo','とぅ':'tu'}
S={'あ':'a','い':'i','う':'u','え':'e','お':'o','か':'ka','き':'ki','く':'ku','け':'ke','こ':'ko',
'さ':'sa','し':'shi','す':'su','せ':'se','そ':'so','た':'ta','ち':'chi','つ':'tsu','て':'te','と':'to',
'な':'na','に':'ni','ぬ':'nu','ね':'ne','の':'no','は':'ha','ひ':'hi','ふ':'fu','へ':'he','ほ':'ho',
'ま':'ma','み':'mi','む':'mu','め':'me','も':'mo','や':'ya','ゆ':'yu','よ':'yo',
'ら':'ra','り':'ri','る':'ru','れ':'re','ろ':'ro','わ':'wa','を':'o','ん':'n',
'が':'ga','ぎ':'gi','ぐ':'gu','げ':'ge','ご':'go','ざ':'za','じ':'ji','ず':'zu','ぜ':'ze','ぞ':'zo',
'だ':'da','ぢ':'ji','づ':'zu','で':'de','ど':'do','ば':'ba','び':'bi','ぶ':'bu','べ':'be','ぼ':'bo',
'ぱ':'pa','ぴ':'pi','ぷ':'pu','ぺ':'pe','ぽ':'po',
'ぁ':'a','ぃ':'i','ぅ':'u','ぇ':'e','ぉ':'o','ゃ':'ya','ゅ':'yu','ょ':'yo'}
VOW='aiueo'

def kana2romaji(s):
    out=[];i=0
    while i<len(s):
        c=s[i]
        if c.isdigit(): out.append(c); i+=1; continue
        if c=='ー':
            if out and out[-1] and out[-1][-1] in VOW: out.append(out[-1][-1])
            i+=1; continue
        if c=='っ':
            j=i+1
            if j<len(s):
                nxt=Y.get(s[j:j+2]) or S.get(s[j],'')
                if nxt and nxt[0] not in VOW: out.append(nxt[0])
            i+=1; continue
        two=s[i:i+2]
        if two in Y: out.append(Y[two]); i+=2; continue
        if c in S:
            r=S[c]
            if c=='ん':
                nx=s[i+1:i+2]
                nr=Y.get(s[i+1:i+3]) or S.get(nx,'')
                if nr and (nr[0] in VOW or nr[0]=='y'): r="n'"
            out.append(r); i+=1; continue
        out.append(c); i+=1
    return ''.join(out)

GENERIC=['しんごうじょう','こせんきょう','りくきょう','ほどうきょう','ちゅうがっこう','しょうがっこう','ゆうびんきょく','はつでんしょ',
 'へんでんしょ','ちゅうしゃじょう','しょうぼうしょ','としょかん','きねんかん','こうみんかん',
 'きゃんぷじょう','しぇるたー','とんねる','じんじゃ','だいばし','おおはし','ばし','はし',
 'がわ','かわ','だむ','やくば','こうこう','こうえん','かいかん','ほーる','えき','てん','ちょうめ']


def split_generic(kana):
    if kana in NOSPLIT: return kana,''
    for g in GENERIC:
        if kana.endswith(g) and len(kana)>len(g):
            return kana[:-len(g)], g
    return kana, ''

import re as _re
def romanise(kana):
    m=_re.match(r'^(.*?)(\d+)ちょうめ$',kana)
    if m:
        a=kana2romaji(m.group(1)); a=a[:1].upper()+a[1:]
        return '%s %s-choume'%(a,m.group(2))
    stem,gen=split_generic(kana)
    a=kana2romaji(stem); b=kana2romaji(gen) if gen else ''
    a=a[:1].upper()+a[1:]
    return (a+' '+b).strip() if b else a


def process(text):
    out, filled, changed = [], 0, 0
    for line in text.split('\n'):
        if line.startswith('| ') and line.count('|') >= 13:
            c = [x.strip() for x in line.strip('|').split('|')]
            if re.match(r'^\d+\.\d+$', c[0]):
                kana = c[3]
                new = MANUAL.get(kana, romanise(kana)) if kana else ''
                if new != c[4]:
                    if not c[4]:
                        filled += 1
                    else:
                        changed += 1
                    c[4] = new
                    line = '| ' + ' | '.join(c) + ' |'
        out.append(line)
    return '\n'.join(out), filled, changed


def main():
    text = io.open(DOC, encoding='utf-8').read()
    new, filled, changed = process(text)
    if '--check' in sys.argv:
        print('%d blank cells would be filled, %d existing would change' % (filled, changed))
        return
    io.open(DOC, 'w', encoding='utf-8').write(new)
    print('%d filled, %d normalised' % (filled, changed))


if __name__ == '__main__':
    main()