User:Gerbrant/edit/autoReplace/default.js

Code that you insert on this page could contain malicious content capable of compromising your account. If you import a script from another page with "importScript", "mw.loader.load", "iusc", or "lusc", take note that this causes you to dynamically load a remote script, which could be changed by others. Editors are responsible for all edits and actions they perform, including by scripts. User scripts are not centrally supported and may malfunction or become inoperable due to software changes. A guide to help you find broken scripts is available. If you are unsure whether code you are adding to this page is safe, you can ask at the appropriate village pump.
This code will be executed when previewing this page.

Documentation for this user script can be added at User:Gerbrant/edit/autoReplace/default.

Note: After saving, you have to bypass your browser's cache to see the changes. Google Chrome, Firefox, Microsoft Edge and Safari: Hold down the ⇧ Shift key and click the Reload toolbar button. For details and instructions about other browsers, see Wikipedia:Bypass your cache.

new function()
{

wloRE = "\\[\\[";
wlcRE = "\\]\\]";
monthRE = "([Jj]anuary|[Ff]ebruary|[Mm]arch|[Aa]pril|[Mm]ay|[Jj]une|[Jj]uly|[Aa]ugust|[Ss]eptember|[Oo]ctober|[Nn]ovember|[Dd]ecember)";
dayRE = "(\\d{1,2})(?:th)?";
yearRE = "(\\d{4})";

monthDayRE = monthRE + "\\s*" + dayRE;
dayMonthRE = dayRE + "\\s*" + monthRE;
wlMonthDayRE = wloRE + monthDayRE + wlcRE;
wlDayMonthRE = wloRE + dayMonthRE + wlcRE;
wlYearRE = wloRE + yearRE + wlcRE;

csRE1 = "\\s*(?:,\\s*|\\s+)";
csRE2 = "\\s*,?\\s*";

unitPRE = "([YZEPTGMkKhdcmµnpfazy]|da|)";
unitRE = "([gsmAKNJWCVΩFTHSM]|mol|cd|rad|sr|Hz|Pa|lm|lx|Wb|Bq|Gy|Sv|kat|°C)";

function fixCaps(a){  return a.slice(0,1) + a.slice(1).toLowerCase();  }

function monthDayYear(a, m, d, y){return "[[" + fixCaps(m) + " " + d + "]], [[" + y + "]]";}
function dayMonthYear(a, d, m, y){return "[[" + d + " " + fixCaps(m) + "]] [[" + y + "]]";}

function binPrefCap(p){  return p == "k" || p == "K" ? "k" : p.toUpperCase();  }

module("Gerbrant.edit.autoReplace.default", [

//Leave the following alone:
{re: "\\[\\[\\d{4}-\\d{2}-\\d{2}\\]\\]|\\|\\s*accessdate\\s*=\\s*\\d{4}-\\d{2}-\\d{2}|ISBN [-0-9]+|\\|\\s*url\\s*=.*(?:\\||\\}\\})|http://[-A-Za-z0-9._\\\\/~%+&#?!=@\\u0028\\u0029]+|\\s*\\n\\s*----|\\s*[\\n\\r]\\|-|\\|\\s*isbn\\s*=\\s*[-0-9]+|\\{\\{convert\\s*\\|.*?\\}\\}", hf: function(a){return a;}},

{re: "ISBN([-0-9]+)", hf: function(a, n){return "ISBN " + n;}},

{re: " --(?!>) ?| —(?! )", hf: function(){return " — ";}},
{re: "(?!<!)([^ ]{2})--(?!>) ?", hf: function(a,s){return s + " — ";}},
{re: "([^ ])— ?", hf: function(a,l){return l + " — ";}},
{re: " -(?!->) ?| –(?! )", hf: function(){return " – ";}},
{re: "(?!<!-)([^ ]{3})- ", hf: function(a,s){return s + " – ";}},
{re: "([^ ])– ", hf: function(a,l){return l + " – ";}},
{re: "(€[0-9,]+\\.)--?", hf: function(a,v){return v + "—";}},
{re: "(\\d)-(?=\\d)", hf: function(a,l){return l + "–";}},
{re: "\\.\\.\\.", hf: function(){return "…";}},
// Convert the word ohm(s) or the html entity into the actual Ω symbol (Omega, not the actual ohm symbol &#8486;) and make sure it's spaced
{re: "(\\d)\\s?" + unitPRE + "(?:\\s|-|)(?:&Omega;|ohm|Ohm)s?([\\s,.])", hf: function(a,d,p,t){return d + " " + p + "Ω" + t;}},
// Convert various micro symbols into the actual micro symbol, make sure it's spaced
{re: "(\\d)\\s?(?:&mu;|μ|&micro;)" + unitRE + "([\\s,.])", hf: function(a,d,u,t){return d + " µ" + u + t;}},
// Convert capital K to lowercase k in units
{re: "(\\d)\\s?K" + unitRE + "([\\s,.])", hf: function(a,d,u,t){return d + " k" + u + t;}},
// Capitalize units correctly (kHz is already covered)
{re: "(\\d)\\s?(?!MHz)(?:[Mm][Hh][Zz])([\\s,.])", hf: function(a,d,t){return d + " MHz" + t;}},
{re: "(\\d)\\s?(?!GHz)(?:[Gg][Hh][Zz])([\\s,.])", hf: function(a,d,t){return d + " GHz" + t;}},
{re: "(\\d)\\s?" + unitPRE + "(?:hz|HZ)([\\s,.])", hf: function(a,d,p,t){return d + " " + p + "Hz" + t;}},
{re: "(\\d)\\s?" + unitPRE + "(?:pa|PA)([\\s,.])", hf: function(a,d,p,t){return d + " " + p + "Pa" + t;}},
// Add a space before dB or B
{re: "(\\d)dB\\b", hf: function(a,d){return d + " dB";}},
{re: "(\\d)B\\b", hf: function(a,d){return d + " B";}},
// Add a space before any units that were missed before
// Separate one for seconds since they give a lot of false positives like "1970s ".  Only difference is mandatory prefix.
// (changed the way that works – now uses a negative lookahead)
{re: "((?:\\D|^)(?!\\d\\d(?:\\d\\d)?s)[0-9,.]*\\d)" + unitPRE + unitRE + "([\\s,.])", hf: function(a,d,p,u,t){return d + " " + p + u + t;}},
// bps or b/s or bits/s --> bit/s
{re: "([KkMmGgTtPpEeYyZz]|)(?:bps|bits\\/s|b\\/s)", hf: function(a,p){return p + "bit/s";}},
// Bps or byte/s or bytes/s --> B/s
{re: "([KkMmGgTtPpEeYyZz]|)(?:Bps|bytes?\\/s)", hf: function(a,p){return binPrefCap(p) + "B/s";}},
// (After that, – would req. multiple runs, so I integrated this in the above) make capitalization correct
{re: "([Kmgtpeyz])(bit|B)\\/s", hf: function(a,p,u){return binPrefCap(p) + u + "/s";}},
// Common error
{re: "\\bmibi(bit|byte)\\b", hf: function(a,u){return "mebi" + u;}},
// Convert hyphen next to a number into a minus sign character
{re: "([^a-zA-Z0-9\\,\\_\\{])-(\\d)", hf: function(a,l,r){return l + "−" + r;}},
// Changes 2x3 to 2×3
{re: "((?:\\D|^)(?!0x[0-9A-Fa-f])[0-9,.]*\\d\\s?)x(\\s?\\d)", hf: function(a,l,r){return l + "×" + r;}},
// Changes 10^3 to 10<sup>3</sup>
// Changes x^3 to x<sup>3</sup>
{re: "([0-9a-zA-Z])\\^(\\d+)(?!(?:(?!<math>).)*</math>)", hf: function(a,l,r){return l + "<sup>" + r + "</sup>";}},
// Changes <sup> tags inside <math> tags back into carets
// (don't know of a way to exclude them from the above statement)
//(but I do, see the RE above)
// Changes 2 +/- 3 to 2±3
{re: "(\\s|\\d)\\+\\/?(?:-|−|-)(\\s|\\d)", hf: function(a,l,r){return l + "±" + r;}},
// Remove all extra newlines?
{re: "(?:\\n\\s*){3,}", hf: function(){return "\n\n";}},

{re: "(?:\\n\\s*){2,}([*#;:])", hf: function(a,s){return "\n" + s;}},
{re: "\\[\\[(.*?)\\|\\1\\]\\]", hf: function(a,b){return "[[" + b + "]]";}},
{re: "\\{\\{[Cc]quote\\|((?:.|\\n)*?)\\}\\}", hf: function(a,b)
{  var r = /(\s*\n){2,}/g;
   if(r.test(b)) b = "<p>" + b.replace(r, "</p>\n<p>") + "</p>";
   return "<blockquote>" + b + "</blockquote>";  }},
{re: "\\{\\{(?:[Rr]quote\\|.*?|[Cc]quote2)\\|((?:.|\\n)*?)\\|((?:.|\\n)*?)\\}\\}", hf: function(a,b,c)
{  var r = /(\s*\n){2,}/g;
   if(r.test(b)) b = "<p>" + b.replace(r, "</p>\n<p>") + " – " + c + "</p>";
   else b += " – " + c;
   return "<blockquote>" + b + "</blockquote>";  }},

{re: monthDayRE + csRE1 + yearRE, hf: monthDayYear},
{re: wlMonthDayRE + csRE2 + yearRE, hf: monthDayYear},
{re: monthDayRE + csRE2 + wlYearRE, hf: monthDayYear},

{re: dayMonthRE + csRE2 + yearRE, hf: dayMonthYear},
{re: wlDayMonthRE + csRE2 + yearRE, hf: dayMonthYear},
{re: dayMonthRE + csRE2 + wlYearRE, hf: dayMonthYear},
{re: wloRE + monthRE + "\\s*(\\d{1,2})th" + wlcRE + csRE2 + wlYearRE, hf: monthDayYear},

//Leave all other links alone to prevent accidentally breaking them:
{re: "\\[\\[.*?(?:\\]\\]|\\|)", hf: function(a){return a;}},

{re: "\\d{4}-\\d{2}-\\d{2}(?!\\]\\])", hf: function(a){return "[[" + a + "]]";}},
{re: "(\\d{4})\\.(\\d{2})\\.(\\d{2})", hf: function(a, y, m, d){return "[[" + y + "-" + m + "-" + d + "]]";}},

{re: "\\s*(<ref(?:>|\\s+name)(?:(?!<ref(?:>|\\s+name)).)*(?:</ref>|/>))\\s*([.,:;!?])", hf: function(a,b,c){return c + b;}},
{re: "\\s*<ref(?=>|\\s)", hf: function(){return "<ref";}},

{re: "…\\.", hf: function(){return "…";}}

]);

}