Bespoke Hyphenation Library

Hi Steve, I was intriged by this challenge and decided to write a script to do it. I did not know you are a keen scripter, and you probably have a better approach than mine, but I will post my script here for what it's worth. Only read it if you don't want the challenge of solving for yourself!   🙂  Oh and of course it hasn't been tested much so will definitely break for some cases.

- Mark

If you decide to test it out, here's how to set it up:

1. Put the two script files together into a folder

      -  "Manually Hyphenate Selected Text.js" (first script below)

      - "Hyphenator.js" (second listing below)

2. Write out your word list and save as "wordlist.txt" in the same folder (see example below).

3. Select some text in Illustrator and run "Manually Hyphenate Selected Text.js" script.

Here are screen grabs of my example. Note that running the script again, will first remove the previous hyphenation.

Script file 1:

/**
 * Performs "manual" hyphenation on selected text(s),
 * using custom word list.
 * Word list contains one word per line, and includes
 * a hyphen at every viable position, for example:
 * -----------------------------------
 *    methyl-sulfonyl-methane
 *    phyto-sphingo-sine
 *    tri-hydroxy-stearin
 * -----------------------------------
 * @File Manually Hyphenate Selected Text.js
 * @7111211 m1b
 * @discussion https://community.adobe.com/t5/illustrator-discussions/bespoke-hyphenation-library/m-p/14342552
 */
//@include 'Hyphenator.js'
(function () {

    var words = loadLinesFromFile('wordlist.txt');

    var hyphenator = new Hyphenator({ words: words }),
        doc = app.activeDocument,
        sel = doc.selection,
        textsToHyphenate = getTexts(sel);

    // hyphenate all the texts
    for (var i = 0; i < textsToHyphenate.length; i++)
        hyphenator.hyphenateText(textsToHyphenate[i]);

})();


/**
 * Get array of text ranges from the supplied item(s).
 * @7111211 m1b
 * @version 2024-01-09
 * @9397041 {TextRange|TextFrame|Array|GroupItem} item - the item or items to get text from.
 * @Returns {?Array<TextRange}
 */
function getTexts(item) {

    var texts = [];

    if (item == undefined)
        return;

    if (item.hasOwnProperty('baselineShift'))
        // already text
        texts.push(item);

    else if (
        item.constructor.name == 'Story'
        || item.constructor.name == 'TextFrame'
    )
        // almost text
        texts.push(item.textRange);

    else if (
        (
            // to bypass bug in TextFrames object:
            item.typename
            && item.typename == 'TextFrames'
        )
        || item.constructor.name == 'Stories'
        || item.constructor.name == 'Array'
    )
        // handle array of text frames
        for (var j = item.length - 1; j >= 0; j--)
            texts = texts.concat(getTexts(item[j]) || []);

    else if (item.constructor.name == 'GroupItem')
        // all text in group
        for (var j = item.pageItems.length - 1; j >= 0; j--)
            texts = texts.concat(getTexts(item.pageItems[j]) || []);

    else if (item.constructor.name == 'Document')
        // all text in document
        texts = texts.concat(getTexts(item.stories));

    return texts;

};


/**
 * Returns array of strings by reading
 * each line of text file.
 * @9397041 {String} nameOrPath - the path or file name of the target file.
 * @Returns {Array<String>}
 */
function loadLinesFromFile(nameOrPath) {

    if (!/\/\\/.test(nameOrPath))
        nameOrPath = File($.fileName).parent + '/' + nameOrPath;

    var f = File(nameOrPath);

    if (!File(nameOrPath).exists)
        return alert('Error: could not find word list at "' + nameOrPath + '"');

    f.open('r');

    return f.read().split(/[\n\r]/g);

};

Script File 2:

/**
 * @File Hypenator.js
 * @7111211 m1b
 * Helper objects:
 *   - Hyphenator
 *   - HyphenatedWord
 *
 * Public methods of Hyphenator:
 *   - removePreviousHyphenation
 *   - hyphenateText
 */

/**
 * A hyphenation helper object.
 * @7111211 m1b
 * @version 2024-01-13
 * @constructor
 * @9397041 {Object} options
 * @9397041 {Array<String>} options.words - the hypenated words to use.
 */
function Hyphenator(options) {

    options = options || {};

    var self = this;
    // the hyphen character
    self.hyphen = options.hyphen || '-';
    // a character to break a line
    self.breaker = options.breaker || ' '; // ordinary space for now
    // the word list
    self.words = options.words || [];

    for (var i = 0; i < self.words.length; i++)
        if ('String' === self.words[i].constructor.name)
            self.words[i] = new HyphenatedWord(self.words[i], self);

};


/**
 * Removes manual hyphenations
 * by searching for hypen/breaks.
 * @7111211 m1b
 * @version 2024-01-13
 * @9397041 {TextRange} text - an Illutrator TextRange.
 */
Hyphenator.prototype.removePreviousHyphenation = function removePreviousHyphenation(text) {

    var self = this,
        contents = text.contents,
        matcher = new RegExp(self.hyphen + self.breaker, 'g'),
        match,
        found = [],
        removeMe;

    matcher.lastIndex = 0;

    while (match = matcher.exec(contents))
        found.push(match.index);

    // process each found, going backwards
    while (removeMe = found.pop()) {
        text.characters[removeMe].remove();
        text.characters[removeMe].remove();
    }

};


/**
 * Finds the first word in the given text,
 * and returns an object containing the word
 * and its start index.
 * If a `from` number is supplied, the search
 * will start from there.
 * @7111211 m1b
 * @version 2024-01-13
 * @9397041 {TextRange} text - an Illustrator Text Range.
 * @9397041 {Number} [from] - start searching from this character index (default: 0).
 * @Returns {Object} - {word: HypenatedWord, start: Number}
 */
Hyphenator.prototype.getFirstWord = function getFirstWord(text, from) {

    var self = this,
        contents = text.contents.slice(from || 0),
        firstWord = {
            start: Infinity,
            word: undefined
        };

    for (var i = 0, start; i < self.words.length; i++) {

        start = contents.search(new RegExp('\\b' + self.words[i].word + '\\b', 'i'));

        if (
            -1 !== start
            && start < firstWord.start
        ) {
            firstWord.start = start;
            firstWord.word = self.words[i];
        }

    }

    if (undefined !== firstWord.word) {
        // update offset
        firstWord.start += from;
        return firstWord;
    }

};


/**
 * Perform the hypenation on the given text.
 * @7111211 m1b
 * @version 2024-01-13
 * @9397041 {TextRange} text - an Illustrator TextRange.
 */
Hyphenator.prototype.hyphenateText = function hyphenateText(text) {

    var self = this;

    if (
        !text.hasOwnProperty('contents')
        || 0 === text.contents.length
    )
        return;

    self.removePreviousHyphenation(text);

    var currentWord,
        advance = 0,
        currentLineID,
        breaks = [];

    wordsLoop:
    while (currentWord = self.getFirstWord(text, advance)) {

        advance = currentWord.start;

        partsLoop:
        for (var i = 1, removeMe; i < currentWord.word.parts.length; i++) {

            // add the length of the preceding part
            advance += currentWord.word.parts[i - 1].contents.length;

            // the lineID is used to test whether linebreaks shifted
            currentLineID = getLineID(text.characters[advance]);

            // add the hypen/breaker to the contents of the character before
            text.characters[advance - 1].contents += self.hyphen + self.breaker;
            breaks.push(advance)
            advance += 2;

            // track the hyphen in
            text.characters[advance - 1].tracking = -250;

            // sanity check
            if (0 === text.characters[text.characters.length - 1].lines.length)
                return alert('Warning: hyphenating failed due to overset text.');

            // check if the line shifted
            if (getLineID(text.characters[advance - 2]) < currentLineID) {

                // the break has caused the line
                // to jump up to the _previous_ line
                currentLineID = getLineID(text.characters[advance - 2]);

                // see if previous breaks need removing
                breaksLoop:
                for (var j = 0, offset = 0; j < breaks.length; j++) {

                    // offset takes removals into account
                    breaks[j] += offset;

                    if (j >= breaks.length - 1)
                        // we never remove the most recent break
                        break breaksLoop;

                    if (getLineID(text.characters[breaks[j]]) === currentLineID) {

                        // remove this obsolete break
                        text.characters[breaks[j]].remove();
                        text.characters[breaks[j]].remove();
                        breaks.splice(j--, 1);

                        // make adjustments
                        offset -= 2;
                        advance -= 2;

                    }

                }

                continue partsLoop;

            }

            else if (getLineID(text.characters[advance + 2]) > currentLineID) {
                // the break is good, because the
                // breaker is at the end of the line
                continue partsLoop;
            }

            else {
                // break not needed, because there is room for more text after it
                removeMe = breaks.pop();
                text.characters[removeMe].remove();
                text.characters[removeMe].remove();
                advance -= 2;
            }

        }

    }

    function getLineID(text) {
        return text.lines[0].start;
    };

};


/**
 * A word object with hyphenation information.
 * @7111211 m1b
 * @version 2024-01-13
 * @constructor
 * @9397041 {String} hyphenatedWord - the word, maximally hyphenated.
 * @9397041 {Hyphenator} hyphenator - the hypenator to use.
 */
function HyphenatedWord(hyphenatedWord, hyphenator) {

    var self = this;

    self.hyphenator = hyphenator;
    self.hyphenatedWord = hyphenatedWord;
    self.matchHyphens = new RegExp(hyphenator.hyphen, 'g');
    self.word = hyphenatedWord.replace(self.matchHyphens, '');
    self.length = self.word.length;
    self.parts = [];

    var parts = hyphenatedWord.split(hyphenator.hyphen);

    for (var i = 0, index = 0; i < parts.length; i++, index += parts[i - 1].length)
        self.parts.push({ contents: parts[i], start: index });

};