Script to find the area of shapes in Adobe Illustrator

GetShapeArea.jsx

/* Save this file with a jsx extension and place in your
Illustrator/Presets/en_US/Scripts folder. You can then
access it from the File > Scripts menu */

var decimalPlaces = 3;

if (app.documents.length > 0) {

	if (app.activeDocument.selection.length < 1) {
		alert('Select a path');
	} else if (app.activeDocument.selection[0].area) {
		// Individual Items
		var objects =  app.activeDocument.selection;
	} else if (app.activeDocument.selection[0].pathItems) {
		// Group/Compound Shape
		var objects = app.activeDocument.selection[0].pathItems;
	} else {
		alert('Please select a path or group.');
	}

	// Collect info
	var totalArea = 0;
	for (var i=0; i<objects.length; i++) {
		if (objects[i].area) {
			var totalArea = totalArea + objects[i].area;
		}
	}

	// Conversions
	var ppi = 72;
	var areaIn = totalArea / ppi / ppi;
	if (areaIn < 0) var areaIn = -areaIn;
	var areaCm = areaIn * 6.4516;

	// Display
	alert('Shape Area\
	' + areaIn.toFixed(decimalPlaces) + ' in² \
	' + areaCm.toFixed(decimalPlaces) + ' cm² \n\
	' + i + ' shapes');

}

I got discrepancies even when the polygon was a regular shape with a whole number as an area, like a square or rectangle. Here's a funny thing I just noticed: if the polygon (regular or irregular) is not snapped to the pixel grid, there's a chance that the script's pixel result will be slightly off because Illustrator gives a fractional pixel value when it isn't aligned to the grid. I would suggest going to Edit>Preferences>General and change Keyboard Increment to 1. Then go to ...>Units and change General and Stroke to Pixels (I think the script doesn't mind whether a polygon has a stroke or not). Then, go to ...>Guides & Grid and change "Gridline every" to 1 px. Go to View>Snap to Grid, and make sure it is checked. Then, in the document, make sure your polygons do not have pixel fractions by looking at the X,Y coordinates and the Width, Height dimensions in the Transform panel.

This fixes regular polygon area calculations, but if your polygon is crazy, like some of mine, I think the script may still throw fractions of pixels, and I think that's understandable.

Another suggestion: there's a chance that totalArea will return negative, so you may want to include a line to fix that up. Again, please let me know if I am way off base, just a graphic designer here!

I still get a discrepancy after re-calculating all the plates with this new method, only this time the results are less consistent, with a 12.17% difference for pixels, 14.25% difference in miles, and a 10.17% difference in kilometers. I think it's just the nature of the beast of rounding numbers, eh?

I'm working with numbers in the millions, so would there be a way to convert the results into numbers with proper commas?

@whitehorn799
Well I'm not sure how the area is calculated. It's an internal function, so perhaps something weird is going on there. Though I've never,not to my knowledge seen faulty outcome.
If I would place a 4x3 cm or pixel format randomly on the canvas, the outcome is the same
Are you willing to share your document?

https://www.dropbox.com/scl/fi/satcipmjk8u7vcenqfbc3/Tectonics.eps?rlkey=gu2t4arwv664vk78v0ev8i48c&dl=0 You can take a look if you like. I put two white squares on there and turned off the snap to pixel grid. The area of the smaller square is about 2 hundredths off, and the square has fractional pixel dimensions. If you turn on the grid, and you give the square whole numbers as WxH dimensions, the result is correct. Like I said, though, it's quite easy and close enough, I think!

@whitehorn799

thanks for highlighting this. It seems there is indeed and issue somewhere. I check the raw output of the area. Thats the negative number in the screengrab. Right is the number from that big white square. Which does have its awkward not rounded number 717 x x716.9996 px

I need to go over the code and see whats cause this.
Whats weird is that i checked it with lower values and it always seemed to return proper area values. like says 10x10 or 100x100
I think its an issue with using pixels and the conversion. If i use mm as ruler input and use those values of the square. I get 252.942 mm x 252.942 mm = 63979.655364 from a calculator and 63979.57105 from the script. Still not correct, but its closer

I think its due to my hot mess of using different decimals. i should look into this and make this neater code.

Yeah, there seems to be a narrow margin of error that gets exponentially more noticeable. Glad to help where I can!

@whitehorn799
Im still working on some other issues inbumpedninto. I noticed when a selection consists of multiple compoundpathitems, it would return the shape count as 1. Currently working on that. I believe the calculations are now fixed. As far as I can do that. I've noticed that still sometimes it returns a none round number even when the length and with are round numbers.. the pixel area is now as close as I can get it

Cool, thanks for updating this and helping out! I appreciate all the work you've done. Will you update the code when you iron out those bumps you were talking about?

@whitehorn799

Of course I will

For some reasons, it got negative area for some paths, so for multiple paths it substracted them. not added. Fixed with adding math.abs:
var totalArea = totalArea + Math.abs(objects[i].area);

But unfortunately, now compound shape calculation doesn't work correctly ¯_(ツ)_/¯