User:Js/urldecoder.js: Difference between revisions

Revision as of 15:43, 25 April 2009

Revision as of 16:46, 21 August 2009

Line 5:

if (!tlb) return

var bt = document.createElement('input')

bt.type = 'button'; bt.onclick = urlDecoderRun; bt.id = 'urlDecoder'

bt.value = '→[\[]]'; bt.id = 'urlDecoder'

bt.title = 'Decode URL before cursor or all URLs in selected text'

bt.style.cssText = 'background:#adbede; height:22px; vertical-align:top; padding:0'

Line 16:

}

if (wgAction=='edit' || wgAction=='submit') addOnloadHook(urlDecoderButton)

Line 23 ⟶ 22:

var httpRegExp = '(https?:\\/\\/[^\\]\\[\\n\\r<>" ]+)' // except []<>"

var beforeCursor = new RegExp('(\\[{0,2})'+httpRegExp+'( +[^\\]\n]+)?\\]{0,2}$', 'i')

var localPrefix = getPrefixesWMPrefixes(wgServer.substringunSecure(7wgServer+wgScript))

if (wgServer == 'https://secure.wikimedia.org')

localPrefix = getPrefixes(wgScript.split('/')[2]+'.'+wgScript.split('/')[1]+'.org')

var newText, linkSize, txtarea = document.editform.wpTextbox1

var isBeforeCursor = false

Line 38 ⟶ 35:

}else { //no selection

if (!(rr=range.duplicate())) return

rr.moveStart('character', - 10001500)

linkSize = processBeforeCursor(rr.text)

if (!linkSize) return

Line 59 ⟶ 56:

newText = processSelText(txt.substring(startPos, endPos))

}else{ //no selection

linkSize = processBeforeCursor(txt.substring((endPos-10001500>0?endPos-10001500:0), endPos))

if (!linkSize) return

startPos = endPos - linkSize //select matched

Line 73 ⟶ 70:

function processBeforeCursor(str){ //looks forfinds http:.* insidein string, returns it'sits distancelength to cursor,and also sets newText var

isBeforeCursor = true

var pos = str.lastIndexOf('http://') //to cut out possible preceding http links

if (pos == -1) pos = str.lastIndexOf('https://')

if (pos == -1) return 0

else if (pos >= 102) str = str.substring(pos-2) //cut chars before http, leavingmove someleft roomto forinclude possibleleading [s

var ma = str.match(beforeCursor) // result: (whole string)' '[', 'http:...', ' name]'

if (!ma) return 0

if (ma[3]) //link with name: automatically add brackets just in case

newText = decodeMatchedLinksimplifyMatched(ma[0], '[', ma[2], ma[3]+']')

else //just url: add closing bracket only if there is leading bracket

newText = decodeMatchedLinksimplifyMatched(ma[0], ma[1], ma[2], ma[1]?']':'')

return ma[0].length

}

Line 90 ⟶ 87:

function processSelText(txt){

txt = txt.replace(RegExp('(\\[{0,2})' + httpRegExp + '([^\\]\\[\\n\\r]*?\\]\\]?)?', 'ig'),

simplifyMatched)

decodeMatchedLink)

if (window.urlDecoderIntLinks) txt = txt.replace(/\[\[[^\]\|\n]+/g, decodeWikiLink)

function(lnk){//skip user_talk

return /^\[\[user_talk:[^#]+$/i.test(lnk) ? lnk : decodeAnchor(lnk)

})

return txt

}

function decodeMatchedLink(str, bracket, url, rest){//gets: (whole string), '[', url, ' name]'

if (!bracket) return decodeLink(url) + str.substring(url.length) //just url

if (!rest) return str //no matching closing bracket

var name = rest.replace(/^ +|\]+$/g,'') //trim leading spaces and ending brackets

str = decodeLink(url, name)

if (str.charAt(0) != '[' ) str = '['+str+']' //not wikilink - restore single brackets

return str

}

function simplifyMatched(str, bracket, url, rest){//arguments: (whole string), '[', url, ' name]'; calls decodeUrl

var pos = url.indexOf("''")

if (pos != -1) url = url.substring(0, pos) // double ' is not allowed inside urls

if (!bracket){//no brackets, just url

var trail = url.match(RegExp('[,;\\\\\.:!\\?' //trailing punctuation, per Parser.php

+ (!/\(/.test(url) ? '\\)' : '') + ']+$' )) //trailing no-matching )

if (trail) url = url.substring(0, url.length-trail[0].length) //move these out of url

if (/(\}\}|\|)$/.test(url)) return str //trailing | or }} can be a part of template, skip to be safe

return decodeUrl(url) + str.substring(url.length)

}else if (rest) //both brackets and possibly name

return decodeUrl(url, rest.replace(/\]+$|^ +| +$/g,'')) //trim ending brackets and spaces in 'name]'

else return str //probably broken wikicode in selected text

}

function decodeLinkdecodeUrl(url, name){ //url -> %-decoded -> [[link|name]] (if possible); name is optional

url = unSecure(url)

var parts, endingDots, result

if (nameurl.indexOf('%') name!= -1) try { url = name.replacedecodeURI(/^url) +|} +$/g,''catch(e){} //trimdecode spaces%

url = url.replace(/%(3B|2F|2C|3A)/g, function(s){return decodeURIComponent(s)}) //decode ;/,:

//secure Wikimedia link into non-secure

url = url.replace(/https:[ <>"\/\/secure.wikimedia.org[\]]/g, function(\w+s)\/{return encodeURIComponent(\w+s)})\ /([^\]\|\n\r ]+)/," disallowed 'http://$2.$1.org/$3')chars

//bugzilla

if (parts = url.match(/^https:\/\/bugzilla\.wikimedia.org\/show_bug\.cgi\?id=(\d+)/))

return '[\[mediazilla:'+parts[1]+ (name?'|'+name:'') + ']]'

//decode %

if (url.indexOf('%') != -1) try { url = decodeURI(url) } catch(e){}

url = url.replace(/%(3B|2F|2C|3A)/g, function(s){return decodeURIComponent(s)}) // ;/,:

//encode back some chars not allowed by MediaWiki

url = url.replace(/[ <>"\[\]]/g, function(s){return encodeURIComponent(s)}) //"

if (isBeforeCursor)

for (var n in window.urlDecoderEngNames) //to eng keywords

url = url.replace(RegExp('(title=|wiki\/)('+urlDecoderEngNames[n]+':)'), '$1' + n + ':')

var link = toWikilink(url)

//analyze

if (link){ //(below) usually no need for leading colon in articles (e.g. inserting interwiki or category)

parts = url.substring(7).split('/')

if (wgNamespaceNumber==0 || wgNamespaceNumber==14) link=link.replace(/^:/,'')

var ___domain = parts[0]

return '[\[' + link + (name?'|'+name:'') + ']]'

var linkPrefix = getPrefixes(___domain)

}else if (typeof name == 'string') return '[' + url + (name?' '+name:'') + ']' //empty name

//return unchanged if cannot convert into wiki link

else return url

if (!linkPrefix || parts[1] != 'wiki' || url.indexOf('?')!=-1)

}

return name ? '[' + url + ' ' + name + ']' : url

//convert to wiki link

function toWikilink(url){//url -> wikilink, otherwise null

//try bugzilla and user-defined prefixes

if (!window.urlDecoderPrefixes) urlDecoderPrefixes = {}

urlDecoderPrefixes['https://bugzilla.wikimedia.org/show_bug.cgi?id=']='mediazilla'

for (var key in urlDecoderPrefixes)

if (url.toLowerCase().indexOf(key)!=-1)

return urlDecoderPrefixes[key]+':'+ url.substring(url.indexOf(key)+key.length)

//try WM prefixes

var parts = url.substring(7).split('/')

if (parts[1]!='wiki' || url.indexOf('?')!=-1) return null

var linkPrefix = WMPrefixes(url.toLowerCase()), prefixes = ''

if (!linkPrefix) return null

var title = url.substring(parts[0].length + parts[1].length + 9) //get part after /wiki/

title = decodeWikiLinkdecodeAnchor(title)

if (linkPrefix[0] && (linkPrefix[0] != localPrefix[0])) prefixes = linkPrefix[0]

var prefixes = ''

if (linkPrefix[01] && (linkPrefix[01] != localPrefix[01])) prefixes += ':' + linkPrefix[01]

if (linkPrefix[1]prefixes &&|| isColonNeeded(linkPrefix[1] != localPrefix[1]title)) prefixes += ':'+linkPrefix[1] //dividing colon or cat/file leading colon

if (prefixes)return prefixes += ':'title

if (!prefixes && isColonNeeded(title)) prefixes = ':' //leading colon for cat/file

//return result

result = '[\[' + prefixes + title

if (name) result += '|' + name + ']]'

else result = (result + ']]').replace(/(\.+)\]\]/,']]$1') //url. -> [[w]]. not [[w.]]

return result

}

function decodeWikiLink(link){

function decodeAnchor(link){//simplify internal link: replace %20 and _ then decode anchor

link = link.replace(/_/g, ' ').replace(/^ +| +$/g, '')

link = link.replace(/(_|%20)/g, ' ').replace(/^ +| +$/g, '')

var parts = link.split('#')

if (parts.length != 2) return link //no anchor

var anchor = parts[1], hidIdx = -1, hidden = []

link = parts[0]

anchor = anchor.replace( //hide IPs

/(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)/,

function(s){ hidden[++hidIdx] = s; return '\x01' + hidIdx + '\x02' }

)

//decode 4, 3 and 2-byte: http://en.wikipedia.org/wiki/UTF-8

anchor = anchor.replace(/\.F[0-4]\.[89AB][\dA-F]\.[89AB][\dA-F]\.[89AB][\dA-F]/g, deChar)

anchor = anchor.replace(/\.E[\dA-F]\.[89AB][\dA-F]\.[89AB][\dA-F]/g, deChar)

anchor = anchor.replace(/\.[CD][\dA-F]\.[89AB][\dA-F]/g, deChar)

anchor = anchor.replace( //hide IPs

//decode reserved 1-byte chars: http://meta.wikimedia.org/wiki/Url , http://en.wikipedia.org/wiki/Percent_encoding

/(?:^|[^0-9A-F\.])(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\.(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)/,

anchor = anchor.replace(/\.[2-5][0-9A-F]/g, function(str){

function(s){ hidden[++hidIdx] = s; return '\x01' + hidIdx + '\x02' }

var ch = deChar(str) // ;:@&=+$,/?%#[]<>

)

if ((ch != ':') &&

//decode 1-byte chars: all symbols except -.:_ and []{} prohibited in links

((encodeURIComponent(ch).replace(/%/,'.') == str) || ('!*\'()'.indexOf(ch) != -1)))

anchor = anchor.replace(/\.[2-7][0-9A-F]/g, function(hhh){

return ch

var ch = deChar(hhh)

else

if ('!"#$%&\'()*+,/;<=>?@\\^`~'.indexOf(ch) >= 0) return ch; else return hhh

return str //do not decode e.g. %35 which is '5' and not supposed to be encoded

})

//unhide IPs and return

function deChar(str){ return tryDecode(str.replace(/\.([0-9A-F][0-9A-F])/g, '%$1')) }

function tryDecode(s){ try {s = decodeURIComponent(s)} catch(e){}; return s }

//unhide IPs

for (var i=hidIdx; i>=0; i--) anchor = anchor.replace('\x01'+i+'\x02', hidden[i])

if (anchor.indexOf("''") != -1) return link //cannot have double '' in link

//return

else return linkparts[0] + '#' + anchor

}

function deChar(ss){

function getPrefixes(___domain){ //en.wikipedia.org -> [ 'w', 'en'] , using projPrefix{}

try{ss = decodeURIComponent(ss.replace(/\.([0-9A-F][0-9A-F])/g, '%$1'))} catch(e){}

var projPrefix = {

return ss

'test.wikipedia.org':'testwiki',

'wikipedia.org':'w',

'wikibooks.org':'b',

'wikinews.org':'n',

'wikiquote.org':'q',

'wikisource.org':'s',

'wikiversity.org':'v',

'wiktionary.org':'wikt',

'mediawiki.org':'mw',

'www.wikimedia.org':'foundation',

'wikimedia.org':'*',

'wikimediafoundation.org':'foundation'

}

}

___domain = ___domain.toLowerCase()

var proj, isFound

function WMPrefixes(url){ // htp://en.wikipedia.org/wiki/... -> [ 'w', 'en']

for (proj in projPrefix) if (___domain.indexOf(proj) != -1) {isFound=true; break}

var dd = url.substring(7).split('/')[0].split('.') // -> ['en','wikipedia','org']

if (!isFound) return null

if (dd.pop() != 'org') return null

var prefix = projPrefix[proj], lang = ''

var proj='', lang = '', part = dd.pop()

if (prefix == '*') prefix = ___domain.split('.')[0] // .wikimedia.org project

if (proj = {'mediawiki':'mw','wikimediafoundation':'foundation'}[part]);

else if ((___domain=___domain.replace(proj, '')) && (___domain != 'www.')) lang = ___domain.split('.')[0] //multi-lang project

else if (proj = {'wikipedia':'w','wikibooks':'b','wikinews':'n','wikiquote':'q',

//else if (prefix.charAt(0) == ':'){ //multi-lang project if (pp.length == 3) lang = pp[0]

'wikisource':'s','wikiversity':'v','wiktionary':'wikt'}[part]){

return [prefix, lang]

lang = dd.pop()

if (!lang || lang=='www') lang = ''

else if (lang=='test') {lang=''; proj='testwiki'}

}else if (part == 'wikimedia'){

part = dd.pop()

if (!part || part=='www') proj = 'foundation'

else if (/^(meta|commons|incubator|species|strategy)$/.test(part)) proj = part

else return null

}else return null

return [proj, lang]

}

function unSecure(url){

return url.replace(/https:\/\/secure\.wikimedia\.org\/(\w+)\/(\w+)\/([^\]\|\n\r ]+)/,

'http://$2.$1.org/$3')

}