User:Gimmetrow/fixRefs.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:Gimmetrow/fixRefs.

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.

//<pre><code>
// This code requires addLink from: [[User:Omegatron/monobook.js/addlink.js]]
//
// [[User:Gimmetrow/fixRefs.js]]
function formatFix() {
   // from [[User:AndyZ/peerreviewer.js]]
   var txt = document.editform.wpTextbox1;
   txt.value = txt.value               //whitespace
    .replace(/\t/g, " ")
    .replace(/^ ? ? \n/gm, "\n")
    .replace(/(\n\n)\n+/g, "$1")
    .replace(/== ? ?\n\n==/g, "==\n==")
    .replace(/\n\n(\* ?\[?http)/g, "\n$1")
    .replace(/^ ? ? \n/gm, "\n")
    .replace(/\n\n\*/g, "\n*")
    .replace(/[ \t][ \t]+/g, " ")
    .replace(/([=\n]\n)\n+/g, "$1")
    .replace(/ \n/g, "\n")              //ending sections 
    .replace(/(== ?)(external links:?|outside links|web ?links:?|exterior links:?)( ?==)/gi, "$1External links$3")
    .replace(/(== ?)(external link:?|web ?link:?|exterior link:?)( ?==)/gi, "$1External link$3")
    .replace(/(== ?)(reference:?)(s? ?==)/gi, "$1Reference$3")
    .replace(/(== ?)(source:?)(s? ?==)/gi, "$1Source$3")
    .replace(/(== ?)(further readings?:?)( ?==)/gi, "$1Further reading$3")
    .replace(/\[\[ ?[Cc]ategory ?: ?/g, "[[Category:")  //categories 
   //repair bad internal links
    .replace(/([^\[]|^)\[?\[([^\]]*?)\]\]?([^\]]|$)/gm, "$1[[$2]]$3")
    .replace(/\[\[ ?([^\]]*?) ?\]\]/g, "[[$1]]")
    .replace(/\[\[([^\]]*?)( |_)#([^\]]*?)\]\]/g, "[[$1#$3]]")
   //repair bad external links
    .replace(/\[?\[http:\/\/([^\]]*?)\]\]?/gi, "[http://$1]")
    .replace(/\[http:\/\/([^\]]*?)\|([^\]]*?)\]/gi, "[http://$1 $2]")
    .replace(/\[\[([^\]\|]+)\|([^\]\|]+)\]\]([A-Za-z\'][A-Za-z]*)([\.\,\;\:\"\!\?\s\n])/g
}

function delinkJustYears() {
   var txt = document.editform.wpTextbox1;
   txt.value = txt.value.replace(/([^,\]] |[^,\] ])\[\[([0-9][0-9][0-9][0-9])\]\]/g, "$1$2$3");

   var summary = document.editform.wpSummary;
   summary.value = summary.value + " delinking [[MOS:DATE|years]] ";
}

function fixRefMarks() {
   var qt = String.fromCharCode(34)
   // var hb = String.fromCharCode(124)
   var txt = document.editform.wpTextbox1;

   // fact and other templates (geographic reference template, ref/note)
   var FactTag = "({{[ ]*fact[ ]*}}|{{[ ]*fact[ ]*[\|][^}]*}}|{{[ ]*facts[ ]*}}|{{[ ]*citequote[ ]*}}|{{[ ]*citation needed[ ]*}}|{{[ ]*cn[ ]*}}|{{[ ]*verification needed[ ]*}}|{{[ ]*verify source[ ]*}}|{{[ ]*verify credibility[ ]*}}|{{[ ]*who[ ]*}}|{{[ ]*failed verification[ ]*}}|{{[ ]*nonspecific[ ]*}}|{{[ ]*dubious[ ]*}}|{{[ ]*or[ ]*}}|{{[ ]*lopsided[ ]*}}|{{[ ]*GR[ ]*[\|][ ]*[^ ]+[ ]*}}|{{[ ]*[c]?r[e]?f[ ]*[\|][^}]*}}|{{[ ]*ref[ _]label[ ]*[\|][^}]*}}|{{[ ]*ref[ _]num[ ]*[\|][^}]*}})";
   txt.value = txt.value.replace(new RegExp("[\n\r\f\t ]+?"+FactTag, "gi"), "$1");

   // one+ space/linefeed, avoid matching references tag as a named ref
   txt.value = txt.value.replace(new RegExp("[\n\r\f\t ]+?<ref([ >])", "gi"), "<ref$1");

   // remove trailing spaces from named refs
   txt.value = txt.value.replace(new RegExp("<ref ([^>]*[^> ])[ ]*>", "gi"), "<ref $1>");

   // remove some superscripted punctuation between refs (is between necessary?)
   txt.value = txt.value.replace(new RegExp("(</ref>|<ref[^>]*?/>)<sup>[ ]*[,;-]?[ ]*</sup><ref", "gi"), "$1<ref");
   txt.value = txt.value.replace(new RegExp("(</ref>|<ref[^>]*?/>)[ ]*[,;-]?[ ]*<ref", "gi"), "$1<ref");

   var LacksPunctuation = "([^\.,;:!\?"+qt+"'’])";
   var QuestionOrExclam = "([!\?])";
   var MinorPunctuation = "([\.,;:])";
   var Blockquote       = "(</blockquote>)";

   var Quote = "([" + qt + "'’]*)";
   //var Space = "[\n\r\f\t ]*"; // to handle  http://en.wikipedia.org/w/index.php?title=Cat&diff=99697351&oldid=99697026
   var Space = "[ ]*";

   var AnyPunctuation = "([\.,;:!\?])";
   var MajorPunctuation = "([,;:!\?])";
   var Period = "([\.])";

   // can this complicated close tag be simplified?
   //var closetag = "(([^<]|<[^/]|</[^r]|</r[^e]|</re[^f]|</ref[^>])*?)</ref>";
   //var closetag = "((.*?))</ref>";
   var RefTag1 = "(<ref>([^<]|<[^/]|</[^r]|</r[^e]|</re[^f]|</ref[^>])*?</ref>)";
   var RefTag2 = "(<ref[^>]*?[^/]>([^<]|<[^/]|</[^r]|</r[^e]|</re[^f]|</ref[^>])*?</ref>)";
   var RefTag3 = "(<ref[^>]*?/>)";
   //var Note = "("+RefTag1 + "|(" + RefTag2 + ")|(" + FactTag + "))";

   var match0a = LacksPunctuation+Quote+FactTag+Space+AnyPunctuation;
   var match0b = QuestionOrExclam+Quote+FactTag+Space+MajorPunctuation;
   var match0c = MinorPunctuation+Quote+FactTag+Space+AnyPunctuation;
   var match0d = QuestionOrExclam+Quote+FactTag+Space+Period;

   var match1a = LacksPunctuation+Quote+RefTag1+Space+AnyPunctuation;
   var match1b = QuestionOrExclam+Quote+RefTag1+Space+MajorPunctuation;
   var match1c = MinorPunctuation+Quote+RefTag1+Space+AnyPunctuation;
   var match1d = QuestionOrExclam+Quote+RefTag1+Space+Period;

   var match2a = LacksPunctuation+Quote+RefTag2+Space+AnyPunctuation;
   var match2b = QuestionOrExclam+Quote+RefTag2+Space+MajorPunctuation;
   var match2c = MinorPunctuation+Quote+RefTag2+Space+AnyPunctuation;
   var match2d = QuestionOrExclam+Quote+RefTag2+Space+Period;

   var match3a = LacksPunctuation+Quote+RefTag3+Space+AnyPunctuation;
   var match3b = QuestionOrExclam+Quote+RefTag3+Space+MajorPunctuation;
   var match3c = MinorPunctuation+Quote+RefTag3+Space+AnyPunctuation;
   var match3d = QuestionOrExclam+Quote+RefTag3+Space+Period;

   for (var j=0; j<5; j++)  // repeat for multiple refs together
   {
      txt.value = txt.value.replace(new RegExp(match0a, "gi"), "$1$2$4$3");
      txt.value = txt.value.replace(new RegExp(match0b, "gi"), "$1$2$4$3");
      txt.value = txt.value.replace(new RegExp(match0c, "gi"), "$2$4$3");
      txt.value = txt.value.replace(new RegExp(match0d, "gi"), "$1$2$3");

      txt.value = txt.value.replace(new RegExp(match1a, "gi"), "$1$2$5$3");
      txt.value = txt.value.replace(new RegExp(match1b, "gi"), "$1$2$5$3");
      txt.value = txt.value.replace(new RegExp(match1c, "gi"), "$2$5$3");
      txt.value = txt.value.replace(new RegExp(match1d, "gi"), "$1$2$3");

      txt.value = txt.value.replace(new RegExp(match2a, "gi"), "$1$2$5$3");
      txt.value = txt.value.replace(new RegExp(match2b, "gi"), "$1$2$5$3");
      txt.value = txt.value.replace(new RegExp(match2c, "gi"), "$2$5$3");
      txt.value = txt.value.replace(new RegExp(match2d, "gi"), "$1$2$3");

      txt.value = txt.value.replace(new RegExp(match3a, "gi"), "$1$2$4$3");
      txt.value = txt.value.replace(new RegExp(match3b, "gi"), "$1$2$4$3");
      txt.value = txt.value.replace(new RegExp(match3c, "gi"), "$2$4$3");
      txt.value = txt.value.replace(new RegExp(match3d, "gi"), "$1$2$3");
   }
   // blockquotes (if trapped punctuation exists, remove it, and move blockquote past at least one ref)
   txt.value = txt.value.replace(new RegExp(Blockquote + AnyPunctuation + FactTag, "gi"), "$3$1");
   txt.value = txt.value.replace(new RegExp(Blockquote + AnyPunctuation + RefTag1, "gi"), "$3$1");
   txt.value = txt.value.replace(new RegExp(Blockquote + AnyPunctuation + RefTag2, "gi"), "$3$1");
   txt.value = txt.value.replace(new RegExp(Blockquote + AnyPunctuation + RefTag3, "gi"), "$3$1");
   txt.value = txt.value.replace(new RegExp(Blockquote + FactTag, "gi"), "$2$1");
   txt.value = txt.value.replace(new RegExp(Blockquote + RefTag1, "gi"), "$2$1");
   txt.value = txt.value.replace(new RegExp(Blockquote + RefTag2, "gi"), "$2$1");
   txt.value = txt.value.replace(new RegExp(Blockquote + RefTag3, "gi"), "$2$1");

   var summary = document.editform.wpSummary;
   //summary.value = summary.value + " [[WP:FN]] [[User:Gimmetrow/fixRefs.js|fixes]] ";
   summary.value = summary.value + " [[WP:FN]] fixes ";
}

addOnloadHook(function () {
    if(document.forms.editform) {
        // addLink('p-cactions', 'javascript:fixRefMarks()', 'fix refs', 'ca-reffix', 'Ref fix for the edit window', '', 'ca-history');
        addPortletLink('p-tb', 'javascript:delinkJustYears()', 'Delink years only', 't-delyrs', 'Delink years', '', '');
        addPortletLink('p-tb', 'javascript:formatFix()', 'Fix format', 't-forfix', 'Format fix', '', '');
        addPortletLink('p-tb', 'javascript:fixRefMarks()', 'Fix reference marks', 't-reffix', 'Ref fix', '', '');
    }
});
//</code></pre>