Finding the syllable block with the most strokes in real Korean words

stroke-count.py

# To use, `sudo apt install hunspell-ko`.

import unicodedata

f = open("/usr/share/myspell/dicts/ko.dic", "r")

# Ensure the text is composed so that each syllable is a single
# character.
normalized_text = unicodedata.normalize('NFC', f.read())

stroke_counts = {
    'KIYEOK': 1,
    'SSANGKIYEOK': 2,
    'NIEUN': 1,
    'TIKEUT': 2,
    'SSANGTIKEUT': 4,
    'RIEUL': 3,
    'MIEUM': 3,
    'PIEUP': 4,
    'SSANGPIEUP': 8,
    'SIOS': 2,
    'SSANGSIOS': 4,
    'IEUNG': 1,
    'CIEUC': 2,
    'SSANGCIEUC': 4,
    'CHIEUCH': 3,
    'SSANGCHIEUCH': 6,
    'KHIEUKH': 2,
    'THIEUTH': 3,
    'PHIEUPH': 4,
    'HIEUH': 3,

    'A': 2,
    'AE': 3,
    'YA': 3,
    'YAE': 4,
    'EO': 2,
    'E': 3,
    'YEO': 3,
    'YE': 4,
    'O': 2,
    'WA': 4,
    'WAE': 5,
    'OE': 3,
    'YO': 3,
    'U': 2,
    'WEO': 4,
    'WE': 5,
    'WI': 3,
    'YU': 3,
    'EU': 1,
    'YI': 2,
    'I': 1,
}

best_strokes = 0
for line in normalized_text.splitlines():
    for char in line:
        try:
            name = unicodedata.name(char)
        except:
            pass

        if 'HANGUL SYLLABLE' in name:
            # Decompose the syllable into characters
            jamo = unicodedata.normalize('NFD', char)
            total_strokes = 0
            for j in jamo:
                try:
                    jamo_name = unicodedata.name(j).split(' ')[-1]
                except:
                    pass
                total_strokes += sum([stroke_counts[a] for a in jamo_name.split('-')])
            if total_strokes > best_strokes:
                print(char, total_strokes, line)
                best_strokes = total_strokes

# Results:
# To use, `sudo apt install hunspell-ko`.

import unicodedata

f = open("/usr/share/myspell/dicts/ko.dic", "r")

# Ensure the text is composed so that each syllable is a single
# character.
normalized_text = unicodedata.normalize('NFC', f.read())

stroke_counts = {
    'KIYEOK': 1,
    'SSANGKIYEOK': 2,
    'NIEUN': 1,
    'TIKEUT': 2,
    'SSANGTIKEUT': 4,
    'RIEUL': 3,
    'MIEUM': 3,
    'PIEUP': 4,
    'SSANGPIEUP': 8,
    'SIOS': 2,
    'SSANGSIOS': 4,
    'IEUNG': 1,
    'CIEUC': 2,
    'SSANGCIEUC': 4,
    'CHIEUCH': 3,
    'SSANGCHIEUCH': 6,
    'KHIEUKH': 2,
    'THIEUTH': 3,
    'PHIEUPH': 4,
    'HIEUH': 3,

    'A': 2,
    'AE': 3,
    'YA': 3,
    'YAE': 4,
    'EO': 2,
    'E': 3,
    'YEO': 3,
    'YE': 4,
    'O': 2,
    'WA': 4,
    'WAE': 5,
    'OE': 3,
    'YO': 3,
    'U': 2,
    'WEO': 4,
    'WE': 5,
    'WI': 3,
    'YU': 3,
    'EU': 1,
    'YI': 2,
    'I': 1,
}

best_strokes = 0
for line in normalized_text.splitlines():
    for char in line:
        try:
            name = unicodedata.name(char)
        except:
            pass

        if 'HANGUL SYLLABLE' in name:
            # Decompose the syllable into characters
            jamo = unicodedata.normalize('NFD', char)
            total_strokes = 0
            for j in jamo:
                try:
                    jamo_name = unicodedata.name(j).split(' ')[-1]
                except:
                    pass
                total_strokes += sum([stroke_counts[a] for a in jamo_name.split('-')])
            if total_strokes >= best_strokes:
                print(char, total_strokes, line)
                best_strokes = total_strokes

# Results:
# 뺄 14 내뺄/4
# 뽑 14 내뽑다/44
# 뽑 14 내뽑아/2
# 뽑 14 내뽑은/3
# 뽑 14 내뽑을/4
# 뺌 14 발뺌/10
# 뺌 14 발뺌하다/44
# 뺌 14 발뺌하여/2
# 뺌 14 발뺌한/3
# 뺌 14 발뺌할/4
# 뺌 14 발뺌해/2
# 뺄 14 뺄/4
# 뺄 14 뺄셈/10
# 뺨 14 뺨/10
# 뺨 14 뺨따귀/25
# 뺨 14 뺨쳐/2
# 뺨 14 뺨치다/44
# 뺨 14 뺨치어/2
# 뺨 14 뺨친/3
# 뺨 14 뺨칠/4
# 뼘 14 뼘/10
# 뽑 14 뽑다/44
# 뽑 14 뽑아/2
# 뽑 14 뽑아내다/44
# 뽑 14 뽑은/3
# 뽑 14 뽑을/4
# 뽑 14 뽑혀/2
# 뽑 14 뽑히다/44
# 뽑 14 뽑히어/2
# 뽑 14 뽑힌/3
# 뽑 14 뽑힐/4
# 뺄 14 용뺄/4
# 뽑 14 제비뽑기/25
# 뼘 14 한뼘바지/25