// ==UserScript== // @name OldSpeak // @namespace http://www.fourmilab.ch/webtools/greasemonkey/OldSpeak/ // @description Translates legacy media Newspeak to plain English // @include * // @version 1.1 // @homepage http://www.fourmilab.ch/webtools/greasemonkey/ /* OldSpeak.user.js by John Walker -- http://www.fourmilab.ch/ This Firefox Greasemonkey (http://www.greasespot.net/) user script (also usable with other browsers with a compatible user script facility) allows you to translate words or phrases in the body text of HTML/XHTML documents to alternative words or phrases, optionally with automatic modification for plural or other variant forms. This allows you to convert Orwellian Newspeak in legacy media and other documents to old-fashioned plain English. You can add any words you wish translated to the "deftrans" call below. If you require additional variant forms, add their expansions to the pluralise() function below it in the code. By default, the script is applied to all sites. If you'd like to restrict it specific sites (for example, one or more legacy media outlets), use the Tools/Greasemonkey/Manage User Scripts menu item and edit the Included Pages and Excluded Pages accordingly. You can debug patterns and the behaviour of this script by setting the variables "brackets" and/or "showchanges" nonzero below. If "brackets" is nonzero, all changed words will be enclosed in {braces}. If "showchanges" is nonzero, the original text and its replacement will be shown, separated by a Unicode right arrow character. */ /* This script is derived from the Profanity Filter: http://userscripts.org/scripts/show/7286 which in turn is based upon the Jmaxxz Vulgar Word Blocker http://userscripts.org/scripts/show/2287 and the "Dumb Quotes" script: http://diveintogreasemonkey.org/casestudy/dumbquotes.html in Mark Pilgrim's "Dive into Greasemonkey": http://diveintogreasemonkey.org/ */ // ==/UserScript== (function() { // Configuration parameters /* Set "brackets" nonzero if you wish changed phrases to be enclosed in {brackets}. */ var brackets = 0; /* Set "showchanges" nonzero to explicitly show transformation from the original word to the replacement. */ var showchanges = 0; var pattern = [], matches = [], translations = []; /* Define the transformations. The transformations are defined as key:value pairs. The key is a regular expression which matches the word or phrase. If the first character of the key is "=", the match will be case-sensitive, otherwise it will be case-insensitive. The value is the replacement word. If the pattern contains matching for plural or other variant forms, you must add code to the pluralise() function to recognise the variant match pattern and transform the replacement. Any grouping within these patterns *must* be done with a (?: ... ) specification to avoid interfering with the capture from the text string. You must be careful not to define a transformation whose replacement is matched by the pattern, for example: "sense": "nonsense", // DON'T DO THIS! as this will cause a CPU loop in the script. */ deftrans({ // Appeals to authority "experts?": "idiot", "analysts?": "moron", "politicians?": "crook", "=(?:Study|Report|Analysis)\\:": "Propaganda:", /* The following terms are predefined for those who may wish to include them in the words and phrases translated, and also serve as a template for creating new replacements. Note that the key values are strings from which regular expressions will be formed, not regular expressions themselves. Consequently, when you use the regular expression escape character "\", you must use a double backslash ("\\") to force the character in the string literal. To enable one or more of these transformations, simply remove the "//" from the start of the line containing it. */ // Issue euphemism // "affirmative\\s+action": "group quotas", // "global\\s+warming": "climate change", // "gun control": "victim disarmament", // "public\\s+assistance": "welfare", // "public\\s+schools?": "government school", // Political spin // "liberals?": "collectivist", // "right[\\s\\-]?wing": "individualist", // Political correctness // "african[\\s\\-]+americans?": "black", // "anti[\\s\\-]+choice": "anti-abortion", // "disabilit(?:y|ies)": "handicap", // "disabled": "handicapped", // "fetus": "unborn child", // "fetuses": "unborn children", // "gays?": "homosexual", // "(?:government|public)[\\s\\-]+funds?": "taxpayer fund", // "(?:government|public)[\\s\\-]+funded": "taxpayer funded", // "hearing[\\s\\-]+impaired": "deaf", // "homeless\\s+persons?": "bum", // "homeless\\s+people": "bums", // "native\\s+americans?": "Indian", // "pro[\\s\\-]+choice": "pro-abortion", // "pro[\\s\\-]+life": "anti-abortion", // "progressives?": "collectivist", // "rainforests?": "jungle", // "reproductive\\s+rights": "right to abortion", // "root\\s+causes?": "excuse", // "trial\\s+lawyers?": "ambulance chaser", // "undocumented\\s+workers?": "illegal alien", // "(?:vision|visually)[\\s\\-]+impaired": "blind", // "wetlands?": "swamp", }); // Build the table of replacements function deftrans(replacements) { for (var word in replacements) { pattern.push(word); var w = word.replace(/^=/, ""); matches.push(new RegExp("([^a-zA-Z])(" + w + ")([^a-zA-Z\u2192])", word.match(/^=/) ? "g" : "gi")); translations.push(replacements[word]); } } // Perform the replacements var curpat; // Hidden argument to repmatch() /* If pattern has a plural variant and the word matched conforms to it, pluralise the replacement word. Note that the replacement word need not used the same pluralisation convention as the matched word. */ function pluralise(word, replacement) { if (((pattern[curpat]).match(/\(\?:y\|ies\)$/) && word.match(/ies$/i)) || ((pattern[curpat]).match(/s\(\?:es\)\?$/) && word.match(/ses$/i)) || ((pattern[curpat]).match(/s\?$/) && word.match(/s$/i))) { if (replacement.match(/y$/)) { replacement = replacement.substr(0, replacement.length - 1) + "ies"; } else if (replacement.match(/s$/)) { replacement += "es"; } else { replacement += "s"; } } return replacement; } /* Replace matched sequence with possibly pluralised and case conforming replacement. */ function repmatch(matched, before, word, after) { var replacement = pluralise(word, translations[curpat]); if (word.match(/^[A-Z]/)) { // Word replacement has an initial capital. Capitalise replacement if (word.match(/^[^a-z]+$/)) { // Word contains no lower case letters. All-caps replacement replacement = replacement.toUpperCase(); } else { /* Capitalise first letter of replacement and, if more than one word, the first letter of each subsequent word in the replacement if the match is a single word or if each word of a multi-word match is capitalised. */ if (word.match(/^[A-Z]([a-z]*\s[A-Z])+/)) { replacement = replacement.replace(/(?:^.|\s[a-z])/g, function(m) { return m.toUpperCase(); }); } else { replacement = replacement.replace(/^./, function(m) { return m.toUpperCase(); }); } } } return before + (brackets ? "{" : "") + (showchanges ? (word + "\u2192") : "") + replacement + (brackets ? "}" : "") + after; } function transform(s) { for (curpat = 0; curpat < matches.length; curpat++) { s = s.replace(matches[curpat], repmatch); } return s; } // We only modify HTML/XHTML documents if (document.contentType && (!(document.contentType.match(/html/i)))) { return; } // We only modify documents retrieved with HTTP/HTTPS if (document.URL && (!(document.URL.match(/^https?:/i)))) { return; } // Replace in document title if (document.title) { var t = transform(" " + document.title + " "); document.title = t.substring(1, t.length - 1) } // Replace in body copy var textnodes = document.evaluate("//body//text()", document, null, XPathResult.UNORDERED_NODE_SNAPSHOT_TYPE, null); for (var i = 0; i < textnodes.snapshotLength; i++) { node = textnodes.snapshotItem(i); /* Test whether this text node appears within a