User:PleaseStand/segregate-refs-dev.js
/* segregate-refs.js: A user script to simplify editing of articles
using inline ref tags with the Cite.php extension to MediaWiki.
Copyright (c) 2010, PleaseStand
This software is licensed under these licenses:
1. Creative Commons Attribution-Share Alike 3.0 Unported License
(see
2. GNU Free Documentation License, any published version.
(see
3. Permission to use, copy, modify, and/or distribute this software for any
purpose with or without fee is hereby granted, provided that the above
copyright notice and this permission notice appear in all copies.
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
ANY SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
You may select the license(s) of your choice if you wish to copy, modify, or
distribute this software. If you modify the software and do not wish to
license your changes under one or more of the licenses, please remove
the license(s) from the list above.
- /
/*global window, addOnloadHook, SegregateRefsJsL10n, SegregateRefsJsAllowConversion,
wikEdUseWikEd, WikEdUpdateTextarea, WikEdUpdateFrame*/
//
// Translate the right-hand side of these if necessary.
// Put translations in a separate file, changing the first line to:
// var SegregateRefsJsL10n = {
var SegregateRefsJsMsgs = {
version: 1.11,
buttonText: "Segregate refs for editing",
buttonStyle: "background: #dfd;",
buttonConvertText: "Migrate article to LDR",
buttonConvertStyle: "background: #fdd;",
autoWord: "Auto",
convertRefsWarning: "WARNING: You need consensus to migrate an article to list-defined references format (LDR) BEFORE you do so.\n\nClick Cancel now if consensus has not been established in favor of this migration. If there is consensus to make the conversion, click OK to do so.",
groupPrompt: "Please enter the name of a group (as it appears in the wikitext, including any quotes). Leave this blank if unsure.",
refsHeader: "Inline footnotes",
convertHeader: "Generated refs list",
refsCommentComplete: "\n\n",
convertSummary: "Converted footnotes to LDR format (using segregate-refs)",
convertFurther: "This script has done most of the work. However, you still need to do the following:\n\n* Insert the refs list in the new textbox into the proper place in the wikitext.\n* If converting a special group, optionally remove the group attributes.\n* Replace all autogenerated names with human-generated names.\n\nYou can do the above with the Find/Replace command in many text editors. (Always use the quoted form of the attributes.) Then, paste the text back into the edit form and save the page.",
integrateWarning: "The refs listed below are missing from the text. If you continue, they will be permanently deleted. Are you sure?\n\nUnused refs: "
};
( function ( $ ) {
var editForm, refsDiv, refsH2, mainTextbox, refsTextbox, randPrefix, messages,
refsButton, convertButton, unloadHandlerRegistered = false;
/**
* Unquote a wikitext tag attribute.
*
* @param string quotedValue
* @return string
*/
function htmlUnquote( quotedValue ) {
var d = document.createElement( 'div' );
d.innerHTML = '';
return d.firstChild.value;
}
/**
* Quote a wikitext tag attribute, choosing single quotes versus
* double quotes depending on which is shorter.
*
* @param string value
* @return string
*/
function htmlQuote( value ) {
var sQ, dQ;
value = value.replace( /\&/g, '&' );
sQ = "'" + value.replace( /'/g, ''' ) + "'",
dQ = '"' + value.replace( /"/g, '"' ) + '"';
return sQ.length < dQ.length ? sQ : dQ;
}
// Looks for ref tags in the text, skipping problematic extension tags.
// For example, "references" may contain out-of-line refs, which should be skipped.
function RefScanner( argWikiText ) {
this.wikiText = argWikiText;
this.refScanRegex = /(?:|<(nowiki|source|references|ref)(?:|\s(?:[^"']|"[^"]*"|'[^']*')*?)(?:\/>|(?:>[\s\S]*?<\/\1(?:|\s[^>]*)>)))/gi;
}
// Get the next ref from the text.
RefScanner.prototype.getRef = function () {
var results;
do {
results = this.refScanRegex.exec( this.wikiText );
if ( !results ) {
return null;
}
if ( results[1] === undefined ) {
results = [0, 0];
}
} while ( results[1].toString().toLowerCase() !== 'ref' );
return results[0];
};
// Extracts attributes from ref tags.
function RefParser( argWikiText ) {
// This is mostly a copy of refScanRegex, except that the whole string must be a ref,
// and no more, and two parts are extracted: $1=attributes, $2=remaining portion of ref
var refParseRegex = /^|(?:>[\s\S]*?<\/ref(?:|\s[^>]*)>))$/i;
this.wikiText = argWikiText;
this.parsedRef = refParseRegex.exec( this.wikiText );
if ( !this.parsedRef ) {
throw new Error( 'invalid ref' );
}
}
// Get all attributes of the tag.
RefParser.prototype.getAttributes = function () {
// In this regex, we need to extract a single name-value pair at a time.
var attParseRegex = /\s([^\s=>]+)\s*=\s*("[^"]*"|'[^']*'|[^\s"']*)/g;
if ( !this.parsedRef ) {
return null;
}
var attributes = Object.create( null ), results;
while ( ( results = attParseRegex.exec( this.parsedRef[1] ) ) ) {
attributes[results[1].toLowerCase()] = htmlUnquote( results[2] );
}
return attributes;
};
/**
* Segregate refs from content.
*
* @param string argWikiText The original wikitext
* @param string group The name of the ref group to process (default group is '')
* @param bool caseCues Mark the original ref code locations using capitalization?
* @return Object
*/
function segregateRefs( argWikiText, group, caseCues ) {
var prefixChars, randNo, randPrefix, refPreferred, scanner, ref, parser, attributes,
refGroup, refName, refStored, refEmpty, refLong, unnamedRefs = 0,
refNames = Object.create( null ), refCodes = [], refShort, outWikiText = '', offset = 0;
// Create a random prefix for autogenerated ref names
prefixChars = '0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZ';
randNo = Math.floor( Math.random() * ( prefixChars.length * prefixChars.length ) );
randPrefix = messages.autoWord
+ prefixChars.charAt( Math.floor( randNo / prefixChars.length ) )
+ prefixChars.charAt(randNo % prefixChars.length) +
'-';
// Create the beginning of the code for a preferred ref location
refPreferred = caseCues ? '
scanner = new RefScanner( argWikiText ); while ( ( ref = scanner.getRef() ) ) { parser = new RefParser( ref ); attributes = parser.getAttributes(); refGroup = attributes.group || ''; if ( group != refGroup ) { // The ref is in a different group continue; } if ( attributes.name !== undefined ) { // The ref already has a name (possibly the empty string) refName = attributes.name; refStored = refName in refNames; refEmpty = parser.parsedRef[2].slice( -2 ) == '/>' || parser.parsedRef[2].slice( 0, 3 ) == '>'; refLong = ref; } else { // We have to autogenerate one refName = randPrefix + ( ++unnamedRefs ).toString(10); refStored = false; refEmpty = false; refLong = '
refShort = refPreferred + 'name=' + htmlQuote( refName ) + '/>'; } else { refShort = refPreferred + 'name=' + htmlQuote( refName ) + ' group=' + htmlQuote( refGroup ) + '/>'; } } else if ( !refEmpty && refNames[refName].empty ) { // Already found an empty ref under this name, yet this one is non-empty // Fill in the long code for the existing entry refCodes[refNames[refName].code] = refLong; refNames[refName].empty = false; // Make a short code for the ref refShort = refPreferred + 'name=' + htmlQuote( refName ); if ( refGroup !== '' ) { refShort += ' group=' + htmlQuote( refGroup ) + '/>'; } refShort += '/>'; } else { // Leave the ref as-is refShort = caseCues ? refLong.replace( /^
} // Replace the long code with the short code outWikiText += argWikiText.slice( offset, scanner.refScanRegex.lastIndex - ref.length ); outWikiText += refShort; offset = scanner.refScanRegex.lastIndex; } outWikiText += argWikiText.slice( offset ); return { wikiText: outWikiText, refCodes: refCodes, randPrefix: randPrefix }; } /** * Insert ref contents back into the text. * * @param string argWikiText The wikitext without ref contents * @param string argRefText The ref contents * @param string randPrefix The randPrefix value returned by segregateRefs() * @param string caseCues The caseCues argument passed to segregateRefs() */ function integrateRefs(argWikiText, argRefText, randPrefix, caseCues) { // Remove an autogenerated ref name if possible. function cleanRefLong(dirtyRef) { var cleanRegex = /^<(ref) name=(?:"[^"]*"|'[^']*'|[^\s"']*)/i; return dirtyRef.replace(cleanRegex, '<$1'); } var scanner, ref, parser, attributes, refCodes = Object.create( null ), usageFreq = Object.create( null ), preferredRef = Object.create( null ), refLong, outWikiText = '', offset = 0; // First, we build an associative array of all the ref codes // that we might need to put back into the text. scanner = new RefScanner( argRefText ); while ( ( ref = scanner.getRef() ) ) { parser = new RefParser( ref ); attributes = parser.getAttributes(); if ( attributes.name !== undefined ) { // Only use the first ref having each name if ( !( attributes.name in refCodes ) ) { refCodes[attributes.name] = ref; } } } // Next, we build an associative array that holds the usage frequency // of every ref name used in text, and whether there is a preferred ref, // if caseCues are enabled. scanner = new RefScanner( argWikiText ); while ( ( ref = scanner.getRef() ) ) { parser = new RefParser( ref ); attributes = parser.getAttributes(); if ( attributes.name !== undefined ) { if ( !( attributes.name in usageFreq ) ) { usageFreq[attributes.name] = 1; } else { usageFreq[attributes.name]++; }