| from src.rule_based_system.Rule import Rule | |
| from src.rule_based_system.TextLengthRule import TEXT_SIZE_LIMIT | |
| from src.rule_based_system.Verdict import Verdict | |
| class BadWordRule(Rule): | |
| """ | |
| Bad words obtained from corners of the internet you do not want to visit: | |
| - https://www.ensie.nl/scheldwoordenboek# | |
| - https://scheldwoorden.goedbegin.nl/ | |
| - https://nl.wiktionary.org/wiki/Categorie:Scheldwoord_in_het_Nederlands | |
| - https://www.lannoo.be/sites/default/files/books/issuu/9789401453417.pdf | |
| - https://www.dutchmultimedia.nl/meest-verschrikkelijke-engelse-scheldwoorden/ | |
| - https://www.dutchmultimedia.nl/scheldwoordenboek-1-000-den-nederlandse-scheldwoorden/ | |
| - https://www.henkyspapiamento.com/10-papiaments-scheldwoorden-die-we-liever-niet-horen/ | |
| - https://volkabulaire.nl/tag/scheldwoorden/ | |
| - https://data.world/wordlists/dirty-naughty-obscene-and-otherwise-bad-words-in-dutch | |
| """ | |
| bad_words = None | |
| def __init__(self, bad_words: list, strict: bool): | |
| self.bad_words = bad_words | |
| self.strict = strict | |
| def get_verdict(self, comment_text: str) -> Verdict: | |
| comment_text = comment_text[0:TEXT_SIZE_LIMIT] | |
| bad_words = self.find_bad_words(comment_text.split()) | |
| return Verdict(len(bad_words) == 0, bad_words) | |
| def find_bad_words(self, text: list) -> list: | |
| detected_bad_words = [] | |
| for word in text: | |
| if word in self.bad_words: | |
| detected_bad_words.append(word) | |
| return detected_bad_words | |
| def is_strict(self) -> bool: | |
| return self.strict | |
| def get_rule_description(self) -> str: | |
| return "Comment text contained %s inappropriate words" % \ | |
| ('strictly' if self.is_strict() else 'ambiguous') | |