Correcção ortográfica e gramatical automática em português europeu.
// ==UserScript==
// @name NaziGramatical
// @namespace http://nazigramatical.x10.bz/
// @version 0.6
// @description Correcção ortográfica e gramatical automática em português europeu.
// @author NaziGramatical
// @match https://www.reddit.com/*
// @grant GM_setValue
// @grant GM_getValue
// @grant GM_xmlhttpRequest
// @grant GM_getResourceText
// @connect nazigramatical.x10.bz
// @resource pt.dic http://nazigramatical.x10.bz/pt.dic
// ==/UserScript==
(function() {
'use strict';
var parsedRules = [];
var lastReplace = {count: 0};
var dicWordsCache = {};
var dic = false;
var lastCheck = 0;
var undos = 0;
var disabled = false;
var log = function () { /*console.log.apply(this, arguments);*/ };
var loadRules = function () {
var savedRules = GM_getValue('rules', []);
for (var i = 0; i < savedRules.length; i++) {
var regs = savedRules[i][0].match(/\/(.+)\/([^/]+)/);
parsedRules.push([new RegExp(regs[1], regs[2]), savedRules[i][1]]);
}
};
var dicWordExistsCS = function (word) {
var time = Date.now();
var result;
if (dicWordsCache[word] !== undefined) result = dicWordsCache[word];
else {
result = (dic.indexOf("\n" + word + "\n") != -1);
}
log('Dic search: ' + word + ' Time: ' + (Date.now() - time) + ' ms. Result: ' + result);
return result;
};
var dicWordExists = function (word) {
if (dicWordExistsCS(word)) return true;
var wordLC = word.toLowerCase();
return word != wordLC && dicWordExistsCS(wordLC);
};
var dicCheckSentence = function (str) {
var words = str.match(/[a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ]{3,}/g);
if (words) {
for (var i = 0; i < words.length; i++) {
if (!dicWordExists(words[i])) return false;
}
}
return true;
};
var saveRules = function () {
var savedRules = [];
for (var i = 0; i < parsedRules.length; i++) {
if (typeof parsedRules[i][1] != 'string') continue;
savedRules.push([parsedRules[i][0].toString(), parsedRules[i][1]]);
}
GM_setValue('rules', savedRules);
};
var replaceMultiple = function (str, a, b) {
for (var i = 0; i < a.length; i++) str = str.replace(a[i], b[i]);
return str;
};
var regexGroups = function (regex) {
return regex.replace(/\\pL/g, '[a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ]')
.replace(/\\pBL/g, '(?<=^|[^a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ])')
.replace(/\\pBR/g, '(?=$|[^a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ])')
.replace(/\\pV/g, '[aeiouáâãéêíóôõúÁÂÃÉÊÍÓÔÕÚ]')
.replace(/\\pVE/g, '(?:e|é|ê)')
.replace(/\\pC/g, '[b-df-hj-np-tv-zçÇ]')
.replace(/\\pNUM/g, '(?:[0-9]+(?:[,.][0-9]+)?|uma?|dois|duas|tr[êe]s|quatro|cinco|seis|sete|oito|nove|dez|onze|doze|treze|catorze|quinze|dezasseis|dezassete|dezoito|dezanove|vinte|trinta|quarenta|cinquenta|sessenta|setenta|oitenta|noventa|cem|duzentos|trezentos|quatrocentos|quinhentos|seiscentos|setecentos|oitocentos|novezentos|mil)');
};
var parseRules = function (rules) {
var parsedRules = [];
for (var i = 0; i < rules.length; i++) {
var rule = regexGroups(rules[i][0].toString());
var replace = (typeof rules[i][1] == 'string') ?
'$1' + rules[i][1].replace(/\$(\d+)/g, function (match, p1) { return '$' + (Number(p1) + 1); }) :
(function (cb) {
return (function () {
var args = [];
for (var i = 0; i < arguments.length; i++) args.push(arguments[i] ? arguments[i] : '');
var r = cb([args[1] ? args[0].substr(1) : args[0]].concat(args.slice(2, -2)));
return r ? args[1] + r : '';
});
})(rules[i][1]);
var flags = '';
var regs = rule.match(/^\/(.+)\/(.*?)$/);
if (regs) {
rule = regs[1];
flags = regs[2];
}
var regs = rule.match(/^\(\?<([!=])(.+?)\)(.+)/);
if (regs) {
//log('Lookbehind rule:', rule);
if (regs[1] == '=') rule = '(^|[^a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ]|'+regs[2]+')' + regs[3] + '(?=[^]$)';
if (regs[1] == '!') {
if (/[^a-zA-Z0-9 ]/.test(regs[2])) {
log('Ignored imparsable rule with lookbehind:', rule);
continue;
}
rule = '(?!'+ regs[2] +')(.{'+regs[2].length+'}|^.{0,'+(regs[2].length-1)+'})' + regs[3] + '(?=[^]$)';
}
//log('Lookbehind rule parsed:', rule);
}
else if (/\(\?</.test(rule)) {
//log('Ignored rule with lookbehind:', rule);
continue;
}
else rule = '(^|[^a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ])' + rule + '(?=[^]$)';
if (!flags) flags = 'i';
parsedRules.push([new RegExp(rule, flags), replace]);
}
//log("ParsedRules: ", parsedRules);
return parsedRules;
};
var isEnglish = function (text) {
var regs = text.match(/(?:^| )(the|is|was|are|be|were|been|did|to|of|and|in|that|have|had|i|it|not|on|with|he|you|at|this|but|his|by|from|they|we|say|her|she|or|an|will|my|one|all|would|there|their|what|up|out|if|about|who|get|which|go|when|make|can|like|time|just|him|know|take|people|into|year|your|good|some|could|them|see|other|than|then|now|look|only|its|over|think|also|back|after|two|how|our|work|first|well|way|even|new|want|because|any|these|give|day|most|thanks?)(?=$|[^a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ])/ig);
var count = regs ? regs.length : 0;
if (count) {
var words = text.match(/[a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ\'-]+/g);
log("English words: "+count+" Total words: "+words.length+" Ratio: "+(count/words.length));
return count / words.length > 0.1;
}
log("No English words found.");
return false;
};
var isPortuguese = function (text) {
var words = text.match(/[a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ]{2,}/g);
var errors = 0;
if (words) {
words = words.slice(-10);
for (var i = 0; i < words.length; i++) {
if (!dicWordExists(words[i])) errors++;
}
log("Portuguese words: "+(words.length - errors)+" Total words: "+words.length+" Non Portuguese Ratio: "+(errors / words.length));
return errors / words.length <= 0.25;
}
return false;
};
var checkLanguage = function (text) {
text = text.replace(/https?:\/\/[^ )]+/g, '');
text = text.replace(/[ru]\/[a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ0-9_-]+/g, '');
text = text.replace(/«.+?»/g, '');
text = text.replace(/".+?"/g, '');
text = text.replace(/\*\*(.+?)\*\*/g, '$1'); // remove bold
text = text.replace(/\*.+?\*/g, '');
text = text.replace(/[a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ-]+[^]$/, ''); // remove last word (including compound words)
// get last 10 words longer than 2 chars
var match = text.match(/([a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ]+[^a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ]+([a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ]{1,1}[^a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ]+)?){0,10}$/);
if (!match) log('Error: no match on checkLanguage:', text);
else text = match[0];
log('checkLanguage', text);
return isPortuguese(text);
//return !isEnglish(text) && isPortuguese(text);
};
var scanLine = function (line) {
log(`scanline: "${line}"`);
var result = '';
for (var i = 0; i < parsedRules.length; i++) {
result = checkWord(line, parsedRules[i][0], parsedRules[i][1]);
if (result) break;
}
log(`result: "${result}"`);
//hyphenated word
var match = (result || line).match(/^(.+-)([a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ]+[^])$/);
if (match) {
var result2 = scanLine(match[1]);
if (result2) result = result2 + match[2];
}
return result;
};
var fixCase = function (str1, str2) {
//log(`fixCase: "${str1}" "${str2}"`);
var isUpperCase = function (s) { return (s == s.toUpperCase()); };
var isLowerCase = function (s) { return (s == s.toLowerCase()); };
var toTitleCase = function (s) { return s.charAt(0).toUpperCase() + s.substr(1); };
var isTitleCase = function (s) { return (s == toTitleCase(s)); };
if (isLowerCase(str1)) return str2;
if (isUpperCase(str1)) return str2.toUpperCase();
if (isTitleCase(str1)) return toTitleCase(str2);
return str2;
};
var checkWord = function (line, regex, replace) {
//log("Line: '"+line+"' Regex: "+regex+" Replace: "+replace);
if (regex.test(line)) {
var match = line.match(regex);
var index = match.index ? match.index + 1 : 0;
var replaced = line.replace(regex, replace).slice(index, -1);
//log(`replaced: '${replaced}'`);
if (replaced) {
log("Line: '"+line+"'");
log("Regex: "+regex+" Replace: "+replace+" Match:", match);
replaced = fixCase(line.slice(index, -1), replaced);
//log(`replaced: '${replaced}'`);
//log('check correction:', dicCheckSentence(replaced));
return line.slice(0, index) + replaced + line.slice(-1);
}
}
};
var removeDiacritics = function (str) {
(function(a, b) { for (var i = 0; i < a.length; i++) str = str.replace(new RegExp(a[i], 'g'), b[i]); })(
['á', 'â', 'é', 'ê', 'í', 'ó', 'ô', 'ú', 'Á', 'Â', 'É', 'Ê', 'Í', 'Ó', 'Ô', 'Ú'],
['a', 'a', 'e', 'e', 'i', 'o', 'o', 'u', 'A', 'A', 'E', 'E', 'I', 'O', 'O', 'U']);
return str;
};
var tryCorrection = function (w1, w2) {
log(`tryCorrection "${w1}" "${w2}"`);
if (!dicWordExists(w1) && dicWordExists(w2)) return w2;
};
document.addEventListener('input', function (e) {
if (disabled) return;
if (!dic) {
dic = {};
setTimeout(function () {
var time = Date.now();
dic = "\n"+GM_getResourceText('pt.dic')+"\n";
log('Loaded dic in ' + (Date.now() - time) + ' ms.');
}, 100);
return;
}
lastReplace.count++;
var input = e.target;
var text = input.value;
if (!text) return; // no text
var sep = text.substr(input.selectionEnd - 1, 1);
if (/[a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ-]/.test(sep)) return;
var left = text.substr(0, input.selectionEnd);
var leftLines = left.match(/.+(\n+|$)/g);
var line = leftLines.pop();
var right = text.substring(input.selectionEnd, text.length);
if (/^ *>/.test(line)) return; // quotes
if (/https?:\/\/[^ )]+[^]$/.test(line)) return; // urls
if (/[ru]\/[a-zA-ZáâãéêíóôõúçÁÂÃÉÊÍÓÔÕÚÇ0-9_-]+[^]$/.test(line)) return; // users/subs
if (/«[^»]+[^]$/.test(line)) return; // quoted
if (/^[^"]*"[^"]*([^"]*"[^"]*"[^"]*)*[^]$/.test(line)) return; // quoted
if (/^[^*]*\*[^*]*([^*]*\*[^*]*\*[^*]*)*[^]$/.test(line.replace(/\*\*.+?\*\*/g, ''))) return; // italic
if (!checkLanguage(left)) return;
lastCheck = Date.now(); // prevent cleanup
log(`Checking: "${line}"`);
var result = scanLine(line);
if (!result) return;
lastReplace = {input: input, value: input.value, selectionEnd: input.selectionEnd, count: 0};
log('Replacing "'+line+'" with "'+result+'"');
line = result;
leftLines.push(line);
left = leftLines.join('');
input.value = left + right;
input.selectionEnd = left.length;
});
document.addEventListener('keydown', function(e) {
if (e.ctrlKey && e.keyCode == 90) {
if (!(lastReplace.count--)) {
log("NaziGramatical: Undo.");
lastReplace.input.value = lastReplace.value;
lastReplace.input.selectionEnd = lastReplace.selectionEnd;
if (++undos == 2 && confirm('The spellchecker will be disabled in this page.')) disabled = true;
e.preventDefault();
}
lastReplace.count--;
}
});
// update rules
if (Date.now() - GM_getValue('rulesTime', 0) > GM_getValue('updateInterval', 0) * 1000) {
log("NaziGramatical: Updating rules.");
GM_xmlhttpRequest({
method: 'GET',
url: 'http://nazigramatical.x10.bz/rules.php',
onload: function (response) {
var data = JSON.parse(response.responseText);
parsedRules = parsedRules.concat(parseRules(data.rules));
GM_setValue('updateInterval', Math.min(604800, data.updateInterval));
GM_setValue('dic.index', data.dic);
saveRules();
log("NaziGramatical: Rules updated. Next update: "+data.updateInterval+" secs.");
}
});
GM_setValue('rulesTime', Date.now());
}
// load stored rules
else loadRules();
// static rules
(function () {
var rules = [
[/(\pL*[áâéêíóôú]\pL*)(mente|zinh[ao]s?)/, function (r) { return removeDiacritics(r[1]) + r[2]; }], //sózinho, sómente
[/(in)?d([ei])s(\pL{3,})/, function (r) { return tryCorrection(r[0], r[1] + (r[2] == 'e' ? 'dis' : 'des') + r[3]); }], //destraído, distoar
[/(\pL+i)ss(es?)/, function (r) { return tryCorrection(r[0], r[1]+'c'+r[2]); }], //chatisse
[/(\pL*[aeiou])([íú])([zlr])/, function (r) { return tryCorrection(r[0], r[1]+removeDiacritics(r[2])+r[3]); }], //saír
[/(\pL*)([áâéêíóôú])([zlr])/, function (r) { return tryCorrection(r[0], r[1]+removeDiacritics(r[2])+r[3]); }], //metêr, cristál
[/(\pL*[^aeiou])([áéíóúâêô])(\pC+[aeo](?:s|m|ns)?)/, function (r) { return tryCorrection(r[0], r[1]+removeDiacritics(r[2])+r[3]); }], //fála, páras
[/(\pL*[^aeiou])([íú])(s|m|ns)?/, function (r) { return tryCorrection(r[0], r[1]+removeDiacritics(r[2])+r[3]); }], //perú, patíns
//[/([aeiou])([iu]?\pC+[aeiou][oea]s?)/, function (r) { return tryCorrection(r[0], r[1]+removeDiacritics(r[2])+r[3]); }],
[/(\pL+)a([eo]s?)/, function (r) { return tryCorrection(r[0], r[1]+'ã'+r[2]); }], //mao, mae, maos, maes
[/(\pL+)o(es?)/, function (r) { return tryCorrection(r[0], r[1]+'õ'+r[2]); }], //poe, poes
[/(\pL+)c[aã]([eo]s?)/, function (r) { return tryCorrection(r[0], r[1]+'çã'+r[2]); }], //bêncao, bêncão, bêncaos, bêncãos, opcoes, opcões
[/(\pL+)c[oõ](es?)/, function (r) { return tryCorrection(r[0], r[1]+'çõ'+r[2]); }], //licoes, licões,
[/(\pL*\pV)[cp]([tcç]\pV\pL*)/, function (r) { return tryCorrection(r[0], (r[1] + r[2]).replace(/m(?=[cçt])/i, 'n')); }], // inflacção
// enclise
[/(((\pL+)([aeiouô]))([srz]))-([oa]s?)(?!-)/, function (r) {
if (dicWordExists(r[1])) {
var replace = '';
if (r[1] == 'quer') replace = 'quere-' + r[6];
else if (r[5] == 's') {
replace = r[2] + '-l' + r[6];
}
else {
replace = r[3] + r[4].replace('a', 'á').replace('e', 'ê').replace('o', 'ô') + '-l' + r[6];
}
return replace;
}
}],
// hyphen
[/(\pL{2,}?)(s?)-(\pL+)/, function (r) {
if (!dicWordExists(r[1] + r[2]) || !dicWordExists(r[3])) {
var left = r[1].slice(0, -1) + removeDiacritics(r[1].slice(-1)) + r[2];
var word = left + (/[aeiou]$/i.test(left) && /^[sr][aeiouáâãéêíóôõú]/i.test(r[3]) ? r[3].replace(/^([sr])/, '$1$1') : r[3]);
if (dicWordExists(word)) return word;
}
else if (
(/^(contra|extra|auto|neo|proto|pseudo)$/i.test(r[1]+r[2]) && !/^[aeiouáâãéêíóôõúhrs]/i.test(r[3])) ||
(/^(anti|arqui|semi)$/i.test(r[1]+r[2]) && !/^[iíhrs]/i.test(r[3])) ||
(/^(ante|entre|sobre)$/i.test(r[1]+r[2]) && !/^h/i.test(r[3])) ||
(/^(hiper|inter|super)$/i.test(r[1]+r[2]) && !/^[hr]/i.test(r[3])) ||
(/^(com|mal)$/i.test(r[1]+r[2]) && !/^[aeiouáâãéêíóôõúh]/i.test(r[3]))
) {
var word = r[1]+r[2]+r[3];
if (dicWordExists(word)) return word;
}
}],
// brute force
[/(\pL+)/, function (r) {
if (!dicWordExists(r[0])) {
var time = Date.now();
var rule = replaceMultiple(r[0], [/a/gi, /e/gi, /i/gi, /o/gi, /u/gi, /c/gi], ['[aáâã]', '[eéê]', '[ií]', '[oóôõ]', '[uú]', '[cç]']);
var regex = new RegExp('\\n' + rule + '\\n', 'gi');
var match = dic.match(regex);
log('Possible matches:', match, 'Time took:', Date.now() - time);
if (match && match.length == 1) return match[0].replace(/\n/g, '');
}
}]
];
parsedRules = parsedRules.concat(parseRules(rules));
})();
// clean dictionary when it's not needed anymore
setInterval(function () {
if (Date.now() - lastCheck > 25000) {
dic = false;
}
}, 10000);
})();