deep-profanity-filter
Version:
A thorough profanity filter that considers most common circumventions. Works with your custom list of blocked and whitelisted words and phrases. Identifies and/or replaces bad words. Works with *wildcards* at *start and/or end* of words.
111 lines • 4.9 kB
JavaScript
;
var __values = (this && this.__values) || function(o) {
var s = typeof Symbol === "function" && Symbol.iterator, m = s && o[s], i = 0;
if (m) return m.call(o);
if (o && typeof o.length === "number") return {
next: function () {
if (o && i >= o.length) o = void 0;
return { value: o && o[i++], done: !o };
}
};
throw new TypeError(s ? "Object is not iterable." : "Symbol.iterator is not defined.");
};
Object.defineProperty(exports, "__esModule", { value: true });
exports.reconstructLocations = exports.reduceInputString = void 0;
/**
* Reduce a string by removing any special characters that are not latin characters,
* numbers or whitespace. Keep track of the segment start indices and lengths so that
* we can reverse engineer where each segment really started in the original string.
* @param inputString - The string that needs to be reduced by removing all special characters.
* @returns The reduced string (with all special characters removed), as well as an array
* of start indices and lengths that indicate where each part of the string really started
* and ended in the original string.
*/
var reduceInputString = function (inputString) {
var e_1, _a;
var matches = inputString.matchAll(/[a-zA-Z0-9\s]+/g);
var outString = '';
var locationInfo = [];
try {
for (var matches_1 = __values(matches), matches_1_1 = matches_1.next(); !matches_1_1.done; matches_1_1 = matches_1.next()) {
var match = matches_1_1.value;
locationInfo.push({
index: outString.length,
offset: match.index - outString.length,
});
outString += match[0];
}
}
catch (e_1_1) { e_1 = { error: e_1_1 }; }
finally {
try {
if (matches_1_1 && !matches_1_1.done && (_a = matches_1.return)) _a.call(matches_1);
}
finally { if (e_1) throw e_1.error; }
}
return {
reducedInput: outString,
reducedLocations: locationInfo,
};
};
exports.reduceInputString = reduceInputString;
/**
* Reconstructs the original "Location Information" of a bad word in a string that got
* reduced and then matched against a regular expression. Uses the information that was
* gathered when reducing the string (by removing all special characters except whitespace).
* @param reducedInfos - Information generated by `reduceInputString` before matching the
* reduced string against a regular expression.
* @param matchInfo - Information on the matched words and locations, gotten when matching
* the string from `reduceInputString` against a regular expression.
* @returns BadWordMatchData with the locations and lengths of the matches projected back
* onto the original, unreduced string.
*/
var reconstructLocations = function (reducedInfos, matchInfo) {
var e_2, _a, e_3, _b;
var newMatchInfo = [];
try {
for (var matchInfo_1 = __values(matchInfo), matchInfo_1_1 = matchInfo_1.next(); !matchInfo_1_1.done; matchInfo_1_1 = matchInfo_1.next()) {
var match = matchInfo_1_1.value;
var matchEndIndex = match.startIndex + match.length;
var startOffset = 0;
var endOffset = 0;
try {
for (var reducedInfos_1 = (e_3 = void 0, __values(reducedInfos)), reducedInfos_1_1 = reducedInfos_1.next(); !reducedInfos_1_1.done; reducedInfos_1_1 = reducedInfos_1.next()) {
var reducedInfo = reducedInfos_1_1.value;
if (reducedInfo.index <= match.startIndex) {
startOffset = reducedInfo.offset;
endOffset = reducedInfo.offset;
}
else if (reducedInfo.index < matchEndIndex) {
endOffset = reducedInfo.offset;
}
else {
break;
}
}
}
catch (e_3_1) { e_3 = { error: e_3_1 }; }
finally {
try {
if (reducedInfos_1_1 && !reducedInfos_1_1.done && (_b = reducedInfos_1.return)) _b.call(reducedInfos_1);
}
finally { if (e_3) throw e_3.error; }
}
newMatchInfo.push({
word: match.word,
startIndex: match.startIndex + startOffset,
length: match.length + endOffset - startOffset,
});
}
}
catch (e_2_1) { e_2 = { error: e_2_1 }; }
finally {
try {
if (matchInfo_1_1 && !matchInfo_1_1.done && (_a = matchInfo_1.return)) _a.call(matchInfo_1);
}
finally { if (e_2) throw e_2.error; }
}
return newMatchInfo;
};
exports.reconstructLocations = reconstructLocations;
//# sourceMappingURL=reduce_input_string.js.map