glosser/transcriber.js

// General functions

function addDotsToWord(word, previous) {
    
    var preDot = "";
    var postDot = "";
    
    // if word begins with a vowel: dot before
    if (isVowel(word.charAt(0))) {
        preDot = ".";
    }
    
    // if word is a cmene: dot after
    if (!isVowel(word.charAt(word.length - 1))) {
        postDot = ".";
        
        // and except for preceding selma'o DOI or LA, also a dot before
        if (!previous || (previous !== "doi" && previous !== "la")) { // TODO: check if there are more of these
            preDot = ".";
        }
    }
    
    return preDot + word + postDot;
}

function isVowel(c) {
    return c === "a" || c === "e" || c === "i" || c === "o" || c === "u" || c === "y";
}

function arrayToString(array) {
    var result = "";
    
    for (var i = 0; i < array.length - 1; i++) {
        result += array[i] + " ";
    }
    
    result += array[array.length - 1];
    
    return result;
}

// Latin mode

function transcribeToLatin(text) {
    var result = [];
    
    for (var i in text) {
        result.push(addDotsToWord(text[i], text[i - 1]));
    }
    
    return result;
}

// Cyrillic mode

function transcribeToCyrillic(text) {
    var result = [];
    
    for (var i in text) {
        result.push(wordToCyrillic(addDotsToWord(text[i], text[i - 1])));
    }
    
    return result;
}

function wordToCyrillic(word) {
    
    var cyrillicWord = "";
    
    for (var i = 0; i < word.length; i++) {
        var letter = word.charAt(i);
        
        var cyrillicLetter = cyrillicTable[letter];
        if (!cyrillicLetter) {
            cyrillicLetter = letter;
        }
        cyrillicWord += cyrillicLetter;
    }
    
    return cyrillicWord;
}

var cyrillicTable = {};
cyrillicTable["b"] = "&#1073;";
cyrillicTable["c"] = "&#1096;";
cyrillicTable["d"] = "&#1076;";
cyrillicTable["f"] = "&#1092;";
cyrillicTable["g"] = "&#1075;";
cyrillicTable["j"] = "&#1078;";
cyrillicTable["k"] = "&#1082;";
cyrillicTable["l"] = "&#1083;";
cyrillicTable["m"] = "&#1084;";
cyrillicTable["n"] = "&#1085;";
cyrillicTable["p"] = "&#1087;";
cyrillicTable["r"] = "&#1088;";
cyrillicTable["s"] = "&#1089;";
cyrillicTable["t"] = "&#1090;";
cyrillicTable["v"] = "&#1074;";
cyrillicTable["x"] = "&#1093;";
cyrillicTable["z"] = "&#1079;";
cyrillicTable["a"] = "&#1072;";
cyrillicTable["e"] = "&#1077;";
cyrillicTable["i"] = "&#1080;";
cyrillicTable["o"] = "&#1086;";
cyrillicTable["u"] = "&#1091;";
cyrillicTable["y"] = "&#1098;";

// Tengwar mode

function transcribeToTengwar(text) {
    var result = [];
    
    for (var i in text) {
        result.push(wordToTengwar(addDotsToWord(text[i], text[i - 1])));
    }
    
    return result;
}

function wordToTengwar(word) {
    
    var tengwarWord = "";
    var canUseTehta = false;
    
    for (var i = 0; i < word.length; i++) {
        var letter = word.charAt(i);
        
        if (letter === "'") {
            tengwarWord += "&#57389;" // halla
            canUseTehta = false;
            continue;
        }
        
        if (letter === ".") {
            // It would have been enough to set canUseTehta to true to get ":" for ".i",
            // but Tengwar Unicode fonts do not have tehta placement rules for the "lowered pusta",
            // which I render as the ASCII "." (period).
            // Therefore, add an extra case for ".i".
            if (word.length > i + 1 && word.charAt(i + 1) === "i") {
                // And also prevent longer words.
                if (word.length === i + 2 || word.charAt(i + 2) === " ") {
                    tengwarWord += "&#57441;"; // double pusta
                    i++;
                    canUseTehta = false;
                    continue;
                }
            }
            
            tengwarWord += ".";
            canUseTehta = false;
            continue;
        }
        
        var tehta = tehtaTable[letter];
        if (tehta) {
            if (i < word.length - 1 && tehtaTable[word.charAt(i + 1)]) {
                // diphtong
                tengwarWord += tengwaTable[letter];
                tengwarWord += tehtaTable[word.charAt(i + 1)];
                i++;
                canUseTehta = false;
                continue;
            } else if (!canUseTehta) {
                tengwarWord += "&#57390;"; // short carrier
                tengwarWord += tehta;
                canUseTehta = true;
                continue;
            } else {
                tengwarWord += tehta;
                canUseTehta = false;
                continue;
            }
        } 
        
        var tengwa = tengwaTable[letter];
        if (tengwa) {
            tengwarWord += tengwa;
            canUseTehta = true;
            continue;
        }
        
        tengwarWord += "?";
    }
    
    return tengwarWord;
}

var tengwaTable = {};
tengwaTable["b"] = "&#57349;";
tengwaTable["c"] = "&#57354;";
tengwaTable["d"] = "&#57348;";
tengwaTable["f"] = "&#57353;";
tengwaTable["g"] = "&#57351;";
tengwaTable["j"] = "&#57358;";
tengwaTable["k"] = "&#57347;";
tengwaTable["l"] = "&#57378;";
tengwaTable["m"] = "&#57361;";
tengwaTable["n"] = "&#57360;";
tengwaTable["p"] = "&#57345;";
tengwaTable["r"] = "&#57376;";
tengwaTable["s"] = "&#57380;";
tengwaTable["t"] = "&#57344;";
tengwaTable["v"] = "&#57357;";
tengwaTable["x"] = "&#57355;";
tengwaTable["z"] = "&#57382;";

tengwaTable["a"] = "&#57394;";
tengwaTable["e"] = "&#57386;";
tengwaTable["i"] = "&#57390;";
tengwaTable["o"] = "&#57366;";
tengwaTable["u"] = "&#57365;";
tengwaTable["y"] = "&#57388;";

var tehtaTable = {};
tehtaTable["a"] = "&#57408;";
tehtaTable["e"] = "&#57414;";
tehtaTable["i"] = "&#57412;";
tehtaTable["o"] = "&#57418;";
tehtaTable["u"] = "&#57420;";
tehtaTable["y"] = "&#57413;";

// Hiragana mode

function transcribeToHiragana(text) {
    var result = [];
    
    for (var i in text) {
        result.push(wordToHiragana(addDotsToWord(text[i], text[i - 1])));
    }
    
    return result;
}

function wordToHiragana(word) {
    
    var hiraganaWord = "";
    
    for (var i = 0; i < word.length; i++) {
        var next = word.charAt(i);
        var nextTwo = word.substring(i, i + 2);
        
        console.log(hiraganaTable["co"]);
        
        if (hiraganaTable[nextTwo]) {
            hiraganaWord += hiraganaTable[nextTwo];
            i++;
            continue;
        }
        if (hiraganaTable[next]) {
            hiraganaWord += hiraganaTable[next];
            continue;
        }
        hiraganaWord += next; // TODO this should not happen, of course
    }
    
    return hiraganaWord;
}

var hiraganaTable = {};
hiraganaTable["a"] = "&#12354;";
hiraganaTable["e"] = "&#12360;";
hiraganaTable["i"] = "&#12356;";
hiraganaTable["o"] = "&#12362;";
hiraganaTable["u"] = "&#12358;";

hiraganaTable["ca"] = "&#12375;&#12354;";
hiraganaTable["ce"] = "&#12375;&#12360;";
hiraganaTable["ci"] = "&#12375;";
hiraganaTable["co"] = "&#12375;&#12362;";
hiraganaTable["cu"] = "&#12375;&#12358;";
hiraganaTable["c"] = "&#12375;";

hiraganaTable["ca"] = "";
hiraganaTable["ce"] = "";
hiraganaTable["ci"] = "";
hiraganaTable["co"] = "";
hiraganaTable["cu"] = "";
hiraganaTable["c"] = "";

hiraganaTable["ka"] = "&#12363;";
hiraganaTable["ke"] = "&#12369;";
hiraganaTable["ki"] = "&#12365;";
hiraganaTable["ko"] = "&#12371;";
hiraganaTable["ku"] = "&#12367;";
hiraganaTable["k"] = "&#12367;";