Help with custom translator

Kagi77 · 11-09-2011, 12:44 PM

I am attempting to write a simple JS code for a translator I'm making. It is based on a language a friend has made up. My problem is I want to be able to translate from English to the gibberish and vice versa.

Code:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title></title>
<script type="text/javascript">
var wordPairs = [
["no", "il"],
["yes", "kra"],
["true", "k-kra"],
["action", "likt"],
["to be", "k"],
["to not be", "il’k"],
["to eat", "kuklu"],
["to fix", "aik"],
["to have", "kuru"],
["to jump", "likt-kikin-ik’haska"],
["to move", "likt"],
["to own", "kuru"],
["to see", "lkl-kasa-ik’kikn-ak"],
["to use", "k"],
["to use", "il’k"],
["to watch", "lkl-kasa-ik'kikin-ak"],

];
function checkText(str){
for(i=0; i < wordPairs.length; i++){
str = str.replace(wordPairs[i][0],wordPairs[i][1]);
}
document.getElementById('reults').innerHTML = str;
}
window.onload=function(){
document.getElementById('txtIn').onchange=function(){
checkText(this.value);
}
}
</script>
</head>
<body>
<input type="text" id="txtIn" />
<div id="reults"></div> 
</body>
</html>

I got the original code from some people here; however, I cannot get it to reverse as well. Thank you in advance for any help.

thesam101 · 11-09-2011, 02:00 PM

Hi Kagi77

something like this?:

Code:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title></title>
<script type="text/javascript">
var wordPairs = [
["no", "il"],
["yes", "kra"],
["true", "k-kra"],
["action", "likt"],
["to be", "k"],
["to not be", "il’k"],
["to eat", "kuklu"],
["to fix", "aik"],
["to have", "kuru"],
["to jump", "likt-kikin-ik’haska"],
["to move", "likt"],
["to own", "kuru"],
["to see", "lkl-kasa-ik’kikn-ak"],
["to use", "k"],
["to use", "il’k"],
["to watch", "lkl-kasa-ik'kikin-ak"],

];

var textInput;
var str;
var direction;

function checkText(str)
{
	str = document.getElementById('txtIn').value.toLowerCase();
	direction = document.getElementById('directionSelect').selectedIndex;
	if(direction==0)
	{
		for(i=0; i < wordPairs.length; i++)
		{
			str = str.replace(wordPairs[i][0],wordPairs[i][1]);
		}
	}
	else
	{
		for(i=0; i < wordPairs.length; i++)
		{
			str = str.replace(wordPairs[i][1],wordPairs[i][0]);
		}
	}
	
	document.getElementById('result').innerHTML = str;
	
}


window.onload=function()
{
	document.getElementById('convertBtn').onclick=checkText;
}
</script>
</head>
<body>
<input type="text" id="txtIn" />
<button id="convertBtn">Convert</button>
<select id="directionSelect">
<option>English to Giberish</option>
<option>Giberish to English</option>
</select>
<div id="result"></div> 
</body>
</html>

Kagi77 · 11-10-2011, 02:25 AM

Yes that is what I was looking for thank you! There is a problem with translating where it is picking up other words. If you type in k-kra on gibberish to english it gives you to-be, yes when it should be true. I know it was something to do with the fact it isn't looking for exact matches (which is what I want). How do I do that?

Old Pedant · 11-10-2011, 05:01 AM

Easy to fix: rearrange the ordering of your word pairs so that the longer gibberish words come first.

So that k-kra is found before either k or kra alone.

Kagi77 · 11-10-2011, 06:02 AM

See this is why I come here. I would have never thought to do that! Thank you so much!

Old Pedant · 11-10-2011, 11:57 PM

Here's a completely different approach.

It's not useful/better for this situation, but it shows the kinds of things you *could* consider doing if you go further.

Code:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
                      "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title></title>
<script type="text/javascript">
var eToG = {
    "no" : "il",
    "yes" : "kra",
    "true" : "k-kra",
    "action" : "likt",
    "to be" : "k",
    "not be" : "il’k",
    "to eat" : "kuklu",
    "to fix" : "aik",
    "to have" : "kuru",
    "to jump" : "likt-kikin-ik’haska",
    "to move" : "likt",
    "to own" : "kuru",
    "to see" : "lkl-kasa-ik’kikn-ak",
    "to use" : "nil’k",
    "to watch" : "lkl-kasa-ik'kikin-ak"
}

var gToE = [];
for ( var word in eToG )
{
    gToE[ eToG[word] ] = word;
}

function translate(text,table)
{
    var where = document.getElementById("results");
    where.innerHTML = "";
    var words = text.value.split(" ");
    var w = 0;
    while ( w < words.length )
    {
        var word = words[w++];
        var xl = table[word];
        if ( xl == null && w < words.length ) 
        { 
            word += " " + words[w++];
            xl = table[word];
            if ( xl == null ) --w;
        }
        where.innerHTML += " " + ( xl == null ? "??" : xl );
    }
}

function setup( )
{
    var form  = document.forms[0];
    var xlate = document.getElementById("results");
    form.english.onchange = function() { translate(this,eToG); }
    form.gibberish.onchange = function() { translate(this,gToE); }
}

window.onload = setup;
</script>
</head>
<body>
<form>
    English: <input type="text" name="english" /><br/>
    Gibberish: <input type="text" name="gibberish" /><br/>
    
    Translation: <div id="results"></div> 
</body>
</html>

SouperAsylum · 12-26-2012, 02:33 AM

Hey I borrowed this code from you! (thank you by the way!) could someone explain to me why it's making up it's own words for me? i have "retwa" in there for "water", but its putting is wstayer and I can't figure out why... here's my code, i apologize for the amount of words, but i think it might be playing into this.

Code:

<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN"
"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html xmlns="http://www.w3.org/1999/xhtml">
<head>
<title>Ideenian Translator</title>
<p><H2><center><font color="#8306CC" face="Freestyle Script" size="20">
Ideenian Translator</center></H2></p>
</br>

<p><font color="D9270F" face="Gabriola" size="7">
<center>
<script type="text/javascript">
var wordPairs = [

["a", "sa"],
["agree", "disenro"],
["air", "anim"],
["all", "lal"],
["allow", "wolit"],
["also", "aka"],
["am", "wa"],
["and", "nec"],
["angry", "grapy"],
["animal", "nanan"],
["answer", "quoner"],
["apple", "plepa"],
["are", "sonar"],
["aside", "dina"],
["ask ", "peto"],
["at", "tay"],
["ate", "te"],
["away", "yawan"],
["band", "nabna"],
["bath", "thab"],
["bathe", "thabay"],
["be ", "eb"],
["become", "motus"],
["bed", "cubal"],
["bee ", "eeb"],
["begin", "incepto"],
["believe", "putov"],
["belt", "blehteh"],
["best", "bonta"],
["book", "bok"],
["borrow", "rotumu"],
["box", "blan"],
["break", "erf"],
["breath", "santin"],
["bring", "rinto"],
["brother", "soha"],
["but", "et"],
["buy", "chay"],
["can", "nac"],
["carry", "ray"],
["change ", "jetio"],
["cold", "doona"],
["color", "roloc"],
["come", "mec"],
["come", "meco"],
["cool", "looin"],
["correct", "rotsek"],
["crazy", "zycra"],
["creature", "turan"],
["creepy", "peecyr"],
["crystal ", "katyr"],
["day", "yad"],
["death", "thead"],
["do", "odi"],
["dress", "dreeah"],
["dry", "dren"],
["earth", "lethra"],
["eat", "ta"],
["egg", "geh"],
["eight", "tet"],
["end", "nedn"],
["evil", "mwaha"],
["father", "dira"],
["feel", "tago"],
["fine", "nifa"],
["fire", "rifelin"],
["fish", "shif"],
["five", "vif"],
["fix", "ify"],
["flower", "reflow"],
["food", "odof"],
["for", "ga"],
["four", "rof"],
["fresh", "shefeh"],
["Friday", "dirfy"],
["friend ", "nolaf"],
["from", "ab"],
["go", "ogg"],
["good", "dool"],
["goodbye", "tasen/tas/en"],
["grass", "snaf"],
["happy", "peeha"],
["hat", "tah"],
["hate", "tahee"],
["have", "vah"],
["he", "eh"],
["heal", "leah"],
["hear", "reah"],
["hello", "tasai/tas/ai"],
["here", "reh"],
["his", "sih"],
["hot", "toen"],
["house", "sohs"],
["how", "woni"],
["hungry ", "nuhryg"],
["I", "ni"],
["in", "nis"],
["is", "si"],
["it", "jina"],
["job", "obij"],
["just", "stuj"],
["know", "peco"],
["lay", "yali"],
["leaf", "flekin"],
["life", "tiva"],
["light", "gith"],
["like", "keyan"],
["little", "tili"],
["look", "okol"],
["loopy", "pyloo"],
["lot", "tolin"],
["love", "lolam"],
["Me", "mah"],
["Monday", "donmy"],
["moon", "noom"],
["morning", "minorg"],
["mother", "diha"],
["Myself", "glof"],
["nature", "renta"],
["my", "ni"],
["night", "tinta"],
["nine", "nin"],
["no", "na"],
["not", "dah"],
["of", "fon"],
["ok", "kona"],
["on", "la"],
["one", "na"],
["or", "vel"],
["our", "ro"],
["out", "tu"],
["pants", "saytah"],
["plant", "tanlep"],
["power", "repo"],
["pretty", "typra"],
["pull", "teef"],
["pyrimid", "primy"],
["right", "nogra"],
["royalty", "lortai"],
["sad", "dashinga"],
["same", "nill"],
["Saturday", "dusty"],
["say", "oros"],
["school", "skol"],
["seven", "vensa"],
["she", "esh"],
["shirt", "showana"],
["shorts", "showta"],
["sick", "keena"],
["silly", "silin"],
["sister", "soma"],
["six", "ix"],
["skirt", "skowana"],
["smell", "lesh"],
["so", "nas"],
["song ", "sara"],
["sorry", "trislus/tris"],
["start", "sar"],
["sun", "nus"],
["Sunday", "dunsy"],
["thankful", "nemus"],
["thankyou", "nemoy/nem"],
["the", "eth"],
["there", "rethe"],
["this", "ith"],
["three", "ret"],
["through", "ruh"],
["throw", "wot"],
["Thursday", "dusthy"],
["to", "ot/oto"],
["toward", " ervas"],
["true ", "resai"],
["Tuesday", "desty"],
["two", "twon"],
["until", "ad"],
["up", "puin"],
["very", "revy"],
["warm", "mar"],
["was", "aws"],
["Water", "retwa"],
["Wednesday", "denwy"],
["wet", "tweh"],
["when", "lin"],
["with", "nai"],
["word", "darin"],
["would", "delow"],
["write", "aiter"],
["yes", "weys"],
["you", "oy"],
["your", "ory"],
["zero", "roz"],
["call", "vor"],



];

var textInput;
var str;
var direction;

function checkText(str)
{
	str = document.getElementById('txtIn').value.toLowerCase();
	direction = document.getElementById('directionSelect').selectedIndex;
	if(direction==0)
	{
		for(i=0; i < wordPairs.length; i++)
		{
			str = str.replace(wordPairs[i][0],wordPairs[i][1]);
		}
	}
	else
	{
		for(i=0; i < wordPairs.length; i++)
		{
			str = str.replace(wordPairs[i][1],wordPairs[i][0]);
		}
	}
	
	document.getElementById('result').innerHTML = str;
	
}


window.onload=function()
{
	document.getElementById('convertBtn').onclick=checkText;
}
</script>
<body background="Magic Forest.jpg">
<input type="text" id="txtIn" />
<button id="convertBtn">Translate</button>
<select id="directionSelect">
<option>English to Ideenian</option>
<option>Ideenian to English</option>
</select>
<div id="result"></div> 
</center></p>
</body>
</html>

Philip M · 12-26-2012, 10:20 AM

You are using replace incorrectly. If you replace (part of) the value of str then that new value of str becomes the text to replace on the next iteration of the loop. water > wsater > wstayer. Do it like this (you want to find/replace whole words only):-

Code:

function checkText() {
var str = document.getElementById('txtIn').value.toLowerCase();
var direction = document.getElementById('directionSelect').selectedIndex;
var newstr = "Not Found";

if (direction==0)  {	
for(i=0; i < wordPairs.length; i++) {
if (str == wordPairs[i][0].toLowerCase()) {
newstr = wordPairs[i][1];
}
}
}

else {	
for(i=0; i < wordPairs.length; i++) {
if (str == wordPairs[i][1].toLowerCase()) {
newstr = wordPairs[i][0];
}
}
}
	
document.getElementById('result').innerHTML = newstr;
	
}

This translates just a single word (but only the noun singular, not the plural e.g. eggs). If you want to translate whole sentences you will need to use something like Old Pedant's script. But like your script that only translates the infinitive of the verb. If "to see" = "lkl-kasa-ik’kikn-ak", what is "I saw" or "I have seen"?

BTW, ask and bee (and quite a few others) return "Not Found". I will leave you to work out why.

Roger Boyes probes Vienna Boys’ Choir - Times OnLine