Script for generating a Swedish word list for Wordle

saldo-5words.py

import xml.etree.ElementTree as ET

# Get saldo.xml at https://spraakbanken.gu.se/resurser/saldo
tree = ET.parse('saldo.xml')
root = tree.getroot()

words = set()
chars = set()
for fr in root.findall(".//FormRepresentation"):
    pos_el = fr.find("./feat[@att='partOfSpeech']")
    word_el = fr.find("./feat[@att='writtenForm']")
    # Find nouns, adjectives, verbs, adverbs and pronouns. (Most importantly, NOT proper nouns.)
    if pos_el is not None and pos_el.attrib.get('val') in ['nn', 'av', 'vb', 'ab', 'pp']:
        word = word_el.attrib.get('val')
        # Skip words with punctuation, remove diacritics.
        if len(word) == 5 and '-' not in word and ':' not in word:
            word = word.upper() \
                .replace('É', 'E').replace('È', 'E').replace('Ê', 'E') \
                .replace('Ñ', 'N') \
                .replace('Ü', 'U')
            words.add(word)
            chars.update(*word.lower())

words = sorted(words)
chars = sorted(chars)  # returns äåö in this wrong order, but it doesn't matter

with open('sv_5words.txt', 'w') as f:
    f.writelines(word + '\n' for word in words)

with open('sv_characters.txt', 'w') as f:
    f.writelines(char + '\n' for char in chars)