Ah, even simpler. Thanks Peter!

It would be a good idea to add some error checking in there, especially on the save file step. It currently overwrites without asking, and you probably don’t want that longer term. You could add a step to check for an existing file, and if it exists either ask the user if they want to overwrite. Or it could automatically create a backup copy of the existing file.

Ok, ok, I’ll quit being lazy. :)

This has been updated to attempt to find all number series using GREP and update.


(function(){
app.findGrepPreferences = NothingEnum.nothing;
app.findGrepPreferences.findWhat = "(d+, )((d)[, ]*)+";
var found = app.activeDocument.findGrep();

function removeCommonPrefix(numbers) {
var updatedNumbers =[];
var i = numbers.length -1;
while (i >= 1) {
var currentNumber = numbers[i];
var currentNumberPrefix = numbers[i].toString().slice(0, 3);
var preceedingNumberPrefix = numbers[i -1].toString().slice(0, 3);
var numberSuffix = numbers[i].toString().slice(3);
if (currentNumberPrefix == preceedingNumberPrefix) {
var number = numberSuffix;
} else {
var number = currentNumber;
}
updatedNumbers.unshift(number);
i--;
}
//the first number will always need the initial 3, so at the end of the loop add it to the start
updatedNumbers.unshift(numbers[0]);
return updatedNumbers;
}
})();


If you can find an entire series of numbers with GREP in a script, you should be able to split them into an array, loop through them to compare each number with the preceding number, then add them to a new array. So if the first 3 of the current number match the first 3 of the preceding number, only add the last 3 numbers to the new array. Otherwise, add the number as is (hopefully that makes sense).

I’m extra bored today, so I threw together a function to handle the number checking (I’m sure there’s a cooler way to do this, but this seems to work). If the function works as you expect, you could expand the script to loop through multiple number series in the document, feed them to this function, and use the returned array to update the document.


var sourceNumbers = [405354, 405358, 405359, 405369, 305046, 305047, 405360, 405381, 405382, 405378];

return updatedNumbers;
}


Hmm, the only other thing I can think of is to read the contents of EPS file (the metadata should be plain text). If you open a MathType file in Notepad, do you see a line like this near the top?
%%Creator: MathType

If that’s present in a MathType 6.7a file, then you should be able to read the EPS file with a script as discussed in this Adobe forum thread:
https://forums.adobe.com/thread/472724

Hope that helps!

I think Loic’s approach will be simpler and more efficient, but as a learning exercise I thought I’d point out a couple of reasons why your original code isn’t working. First, the curly brace {} nesting isn’t quite right. Everything inside the for loop needs to be in a single set of curly braces. Then, everything that depends on an if statement needs to be in its own set of braces. So you’d want something like this:
for (i=0;i<myTable.cells.length;i++){
if (myTable.cells[i].texts[0].contents[0] == “Air Conditioning & Heating”){
myTable.rows.add(LocationOptions.AT_END, myTable);
myTable.rows[-1].contents = “###Heating & Air Conditioning”;
}
[[[Additional if statements here]]]
}

In your existing code, the for loop ends at the first closing curly brace, then errors because the next line begins with an opening brace that it doesn’t know what to do with. Also, it’s always placing “Heating & Air Conditioning” because an if statement that doesn’t have a set of curly braces after it only applies to the next line. So myTable.rows.add depends on the if statement, but myTable.rows[-1].contents will always run.

Second, I think the if statement test might not come out quite right. I’m pretty sure .contents[0] will give you the first letter of the contents, so your first test winds up being something like “H” == “Air Conditioning & Heating”.

I can’t think of a reason why either of those settings would cause problems in an EPUB. Can they give you a reason why they don’t want these settings? (Note, I don’t really work with external clients, but I’m guessing it’s not that much fun to get them to change or explain their ways….)

For 1), InDesign does export keep option settings into the CSS, but I’ve never found an EPUB reading system that pays any attention to CSS keep options. If the settings are within paragraph styles, I’d think it would be very easy to remove those from the CSS (if they’re concerned about these settings I’m assuming they’re using the InDesign generated CSS and not writing their own).

If any keep options are set as a local setting and not in a style then they’ll show up in the CSS as an override. For example, if you’ve set keep options locally on a paragraph you’ll have an additional style declaration in the CSS (meaning there could be more settings to remove than just the one in the paragraph style).

Anyway, trying to typeset a book while not using the keep options sounds like a huge pain. That seems like it would add hours to typesetting, while removing those settings from the CSS could be done with a find and replace in a few minutes.

For 2), again, I’m not sure why this would cause an issue. There’s a CSS rule for transforming text to uppercase: text-transform:uppercase. InDesign will export this CSS rule if you’ve applied all caps in a paragraph style. The only difference I can think of is that most EPUB TOCs don’t allow styles to be applied, so if you have all caps applied by CSS they’ll still appear as title case in the navigation (to me this is preferable; all cap headings in the body can look ok, but all caps in an EPUB TOC don’t).

As far as scripts go, I’m sure there are options out there. It sounds like a common issue and a simple thing for a script. Here’s a GREP based option:
https://www.kahrel.plus.com/indesign/grep_change_case.html

I see. Thanks for the explanation!

I think this might be fairly simple (at least, simpler than I was originally making it). I was trying to think of was of being able to mark fonts as “used” and have the script skip them. But I think an even easier answer is to use script you originally posted with just two minor tweaks.

First, somewhere around the font array declaration (var font = [];), add a variable with the font that you only want to use once:
var uniqueFont = “Comic Sans MSBold”;

Second, at the very end of the script (after all the curly braces), add:
app.selection[0].characters[Math.floor(Math.random() * app.selection[0].characters.length)].appliedFont = uniqueFont;

The idea here is to let the original script work as intended. It will randomly apply a font to every character. Then, when that’s finished, it will apply your uniqueFont to a single, random character. That should keep things nice and simple!