Find Junk characters in Indesign, without using Glyphs

A full discussion of missing glyphs in InDesign must include Peter Kahrel's (link removed by moderator), itself the result of extensive discussions in the old forums. In my work, the AWOL items generally are not junk, but obscure forms: rare Chinese chars., diacritics for romanizing Sanskrit, IPA, etc.  The script takes time to run because it examines every char., which is what I usually need.

The red “tofu” box you’re seeing isn’t actually a specific character you can target with GREP it’s a missing glyph (the font can’t display that character).

That’s why your GREP is catching lots of “junk” but not specifically those boxes. GREP works on the character itself, not how it renders.

If you want to find those tofu boxes specifically, the most reliable way in InDesign is:

Use Preflight > enable “Missing Glyphs”
That will flag exactly those characters

If you still want a GREP approach, the closest you can do is narrow things down with something like

[^\x00-\x7F]

but that will also catch valid characters like accents, dashes, etc., so it’s more of a broad sweep than a precise solution.

If you need automation, a script is the only way to properly isolate missing glyphs, because it can check whether the font actually supports the character.

So in short:

GREP = good for general cleanup
Preflight = best for tofu/missing glyphs
Script = best if you want to automate it

Script idea would be something like 
 

var doc = app.activeDocument;

// --- Create highlight color ---
var color;
try {
    color = doc.colors.itemByName("TofuHighlight");
    color.name;
} catch(e) {
    color = doc.colors.add({
        name: "TofuHighlight",
        model: ColorModel.PROCESS,
        space: ColorSpace.RGB,
        colorValue: [255, 0, 0]
    });
}

// --- Create character style ---
var style;
try {
    style = doc.characterStyles.itemByName("TofuStyle");
    style.name;
} catch(e) {
    style = doc.characterStyles.add({
        name: "TofuStyle",
        fillColor: color
    });
}

// --- Loop through characters ---
var chars = doc.stories.everyItem().characters;

for (var i = 0; i < chars.length; i++) {
    var ch = chars[i];

    try {
        // Missing glyphs often return glyph ID = 0
        if (ch.glyphID === 0 && ch.contents !== " ") {
            ch.appliedCharacterStyle = style;
        }
    } catch(e) {}
}

alert("Tofu (missing glyphs) highlighted.");

At the end - you can then search for the colour in the document using the Edit>Find Change to see if there’s any remnants. 

Between the script and the Preflight it should catch them all 

Only this line does anything

ch.appliedCharacterStyle = style;

You can do other things with this line

Replace with a placeholder - I used a question mark - but you don’t have to you can use anything

ch.contents = "?";

Or if you want to remove them

ch.remove();

It works, but my requirment is to take only the tofu square, but it  highlight all the junk char can you gimme to find these type of junk. 

Depends what you want to do 

First thing I think of here is the Preflight - and to check for Missing Glyphs - that would pretty much do it. 

You mentioned scripting, but not what you’ve tried or what isn’t working or why it’s not working - or even what you want it to do.

Something like this would highlight them in red

Then you can use Find/Replace to find the colour and decide what you want to do

var doc = app.activeDocument;

// --- Create / get highlight color ---
var highlightColor;
try {
    highlightColor = doc.colors.itemByName("JunkHighlight");
    highlightColor.name; // forces error if not valid
} catch(e) {
    highlightColor = doc.colors.add({
        name: "JunkHighlight",
        model: ColorModel.PROCESS,
        space: ColorSpace.RGB,
        colorValue: [255, 0, 0] // bright red
    });
}

// --- Create / get character style ---
var highlightStyle;
try {
    highlightStyle = doc.characterStyles.itemByName("JunkCharStyle");
    highlightStyle.name;
} catch(e) {
    highlightStyle = doc.characterStyles.add({
        name: "JunkCharStyle",
        fillColor: highlightColor
    });
}

// --- Loop through characters ---
var chars = doc.stories.everyItem().characters;

for (var i = 0; i < chars.length; i++) {
    var c = chars[i].contents;

    if (c.charCodeAt(0) > 127) {
        chars[i].appliedCharacterStyle = highlightStyle;
    }
}

Similarly you could write a log to the desktop - but it has limitations at the moment but could be useful

var doc = app.activeDocument;

// Output file on Desktop
var file = new File(Folder.desktop + "/InDesign_Junk_Report.txt");
file.open("w");
file.writeln("Junk Character Report\n");

// Get all characters
var chars = doc.stories.everyItem().characters;

for (var i = 0; i < chars.length; i++) {
    var ch = chars[i];
    var c = ch.contents;

    if (c && c.length > 0 && c.charCodeAt(0) > 127) {

        var code = c.charCodeAt(0);

        // Page number
        var page = "Unknown";
        try {
            page = ch.parentTextFrames[0].parentPage.name;
        } catch(e) {}

        // Line number
        var line = "Unknown";
        try {
            line = ch.lines[0].index + 1;
        } catch(e) {}

        // Context (a few characters around it)
        var context = "";
        try {
            var start = Math.max(0, i - 5);
            var end = Math.min(chars.length - 1, i + 5);
            for (var j = start; j <= end; j++) {
                context += chars[j].contents;
            }
        } catch(e) {}

        file.writeln(
            "Char: [" + c + "] " +
            "Code: " + code +
            " | Page: " + page +
            " | Line: " + line +
            " | Context: " + context
        );
    }
}

file.close();

alert("Report saved to Desktop.");

Or another approach would be with GREP
 

~S|~s|~m|\x{200B}|\x{200C}|\x{200D}|\x{FEFF}

Or likes of

[^A-Za-z0-9.,;:!?'"()\-\s]

Or add other symbols you don’t want flagged
 

[^A-Za-z0-9.,;:!?'"()\-\s+\-×÷=]