Expected behavior

Censoring strategy should generate a replacement string that itself does not contain any profanity. However, since grawlixCensorStrategy returns a string of random characters that include "@" and "$", sometimes it generates "@$$", which is considered profanity by the default englishDataset/englishRecommendedTransformers matcher recommended in the guide.

Actual behavior

This causes TextCensor with the default settings to sometimes replace "ass" or "@$$" with "@$$".

Minimal reproducible example

const {
	RegExpMatcher,
	TextCensor,
	englishDataset,
	englishRecommendedTransformers,
} = require('obscenity');

const matcher = new RegExpMatcher({
	...englishDataset.build(),
	...englishRecommendedTransformers,
});

const censor = new TextCensor();
const input = '@$$';
const matches = matcher.getAllMatches(input);
for (let i = 0; i < 100; i++) {
    if (censor.applyTo(input, matches) === input)
        console.log("input did not change after censoring");
}

Steps to reproduce

No response

Additional context

No response

Node.js version

v20

Obscenity version

v0.4.0

Priority

• Low
• Medium
• High

Terms

• I agree to follow the project's Code of Conduct.
• I have searched existing issues for similar reports.