Script modification: delete row if cell contains specific text

[Benjamin39010368auuu]

Hi guys,

I am a newbie and and trying to using the great resources out there and not be needy, but this one is stumping me.

I have created a table for a data merge.  Each cell contains some text as a heading (in this case "SUPER OBLATA") and then a data merge target underneath it.  Sometimes the table does not have data for that target so it is blank leaving the cell with the heading "SUPER OBLATA" and nothing underneath it.  I used AI to create a script to examine the first column, remove hidden spaces/breaks/characters, and then check if the cell contains exactly the header.  If found, delete the row.

The problem is, that the script doesn't find any such rows even though there are many. 

If I go into the cell, delete everything, and then type the term in with a bunch of spaces and breaks and rerun the script, it works fine.

What gives?

// Ensure a document is open
if (app.documents.length > 0) {
    var doc = app.activeDocument;
    var tablesProcessed = 0;

    // Normalize the search term by removing spaces, line breaks, and converting to lowercase
    var searchTerm = "SUPER OBLATA".replace(/\s+/g, '').toLowerCase();

    // Regular expression to match hidden characters
    var hiddenCharRegex = /[\u00A0\u00AD\u200B]/g;

    // Iterate over all stories in the document
    for (var i = 0; i < doc.stories.length; i++) {
        var story = doc.stories[i];
        for (var j = 0; j < story.tables.length; j++) {
            var table = story.tables[j];
            var rowsDeleted = 0;

            // Iterate over rows from bottom to top to avoid skipping rows after deletion
            for (var k = table.rows.length - 1; k >= 0; k--) {
                // Get the text content of the first cell in the row
                var cellText = table.rows[k].cells[0].contents;

                // Normalize the cell text by removing spaces, line breaks, hidden characters, and converting to lowercase
                var normalizedCellText = cellText.replace(/\s+/g, '').replace(hiddenCharRegex, '').toLowerCase();

                // Check if the normalized cell text matches the search term
                if (normalizedCellText === searchTerm) {
                    // Delete the entire row
                    table.rows[k].remove();
                    rowsDeleted++;
                }
            }

            if (rowsDeleted > 0) {
                tablesProcessed++;
            }
        }
    }

    if (tablesProcessed > 0) {
        alert("Processed " + tablesProcessed + " table(s). Rows where the first cell contains 'SUPER OBLATA' (ignoring spaces, line breaks, and hidden characters) have been deleted.");
    } else {
        alert("No rows where the first cell contains 'SUPER OBLATA' (ignoring spaces, line breaks, and hidden characters) were found.");
    }
} else {
    alert("No document is open.");
}

Thanks for your help!

Also, if using AI script is some sort of faux pas here, my bad.

-Ben

[rob day]

Hi @Benjamin39010368auuu , can you share an ID doc with a table the script doesn't work  with? Attach it to a reply

[Benjamin39010368auuu]

Here is an example where one cell is empty and another cell has content.

Thank you!

[rob day · Accepted solution]

I’m not sure the script has to be as complex.

Is there always a soft return before the  SUPER OBLATA string? This might work:

// Ensure a document is open
if (app.documents.length > 0) {
    var doc = app.activeDocument;
    var tablesProcessed = 0;
    var res = getGrepSearch("\\nSUPER OBLATA")
    for (var i = 0; i < res.length; i++){
        if (res[i].parent.constructor.name == "Cell" && res[i].parent.lines.length == 2) {
            res[i].parent.parentRow.remove();
            tablesProcessed++;
        }
    };  

} else {
    alert("No document is open.");
}


/**
* Gets results of a text search as an array 
* @ param text to search for 
* @ return result as an array 
*/
function getGrepSearch(fp){
    app.findGrepPreferences = app.changeGrepPreferences = app.findChangeGrepOptions = null;
    app.findChangeGrepOptions.properties = {includeHiddenLayers:true, includeLockedLayersForFind:true, includeLockedStoriesForFind:true, includeMasterPages:true} 
    app.findGrepPreferences.findWhat = fp;
    return app.findGrep()
}

[Benjamin39010368auuu]

This worked great!  It's flexible so that I can modify it for other circumstances.  Thanks for your help.  

[brian_p_dts]

Maybe a simple question, but are you running this script after you execute the merge? Not during preview? Because nothing would happen during preview. Also, are any of the cells in the table merged? As @rob day mentions, a sample executed merge doc would help. 

[Benjamin39010368auuu]

Simple questions aren't always simple for me, so better to ask.  I ran the script in a merged document, not through the preview function.  I attached the first two pages of the script in the previous reply.

[rob day]

Also i don’t think you can assume the cell’s contents will be a text string. For example this:

Get cell 1 contents:

var t=app.activeDocument.textFrames[0].parentStory.tables[0]

$.writeln(t.cells[0].contents)
//returns nothing

$.writeln(t.cells[0].allPageItems[0])
//returns [object TextFrame]

$.writeln(t.cells[0].allPageItems[0].contents)
//returns Yo

$.writeln(t.cells[1].contents)
//returns hi

$.writeln(t.cells[2].contents)
//returns [object Rectangle]