How to calculate actual selection or group width and height with clipping

Question

If a group has clipping paths then the height and visible bounds will include the dimensions of all paths selected including the clipped or hidden paths.
However, the width and height shown on the width and height of the AI detail boxes for the selection shows just the visible dimensions.
I want to retrieve the same values that is shown on the screen when a group is selected.
It is somewhat counterintuitive that the "visiblebounds" property does not show just what is visible.
I'm scaling the visible content of a selection to fit within another path but I need to be able to calculate the dimensions of the only what is visible after clipping.
What do you have to do with scripting to get these values? Go through all of the path items in a group recursively and save the min and max values of any visiblebounds for clipping paths? Any better alternatives that I don't see?
Thanks for any help!

Answer 1

I ended up using the visible bounds for both scale and position as I also discovered that attempting to position a group with the position method would also result in missing the mark. The top and left (x,y) bounds I calculated without the clipped dimensions needed to be used to offset the actual position where I wanted the visible paths to be displayed.
The functions I used ended up looking like so:
function GetPoints(dblInch)
     var dpi = 72;
     return dblInch * dpi;
function getRealVisibleBounds(grp) {
     var outerBounds = [];
     for(var i = grp.pageItems.length - 1; i >= 0;i--)  {
          var bounds = [];
          if(grp.pageItems[i].typename == 'GroupItem') {
               bounds =  getRealVisibleBounds(grp.pageItems[i]);
          else if((grp.pageItems[i].typename == 'PathItem' || grp.pageItems[i].typename == 'CompoundPathItem')
               && (grp.pageItems[i].clipping || !grp.clipped)) {
               bounds = grp.pageItems[i].visibleBounds;
          if (bounds.length > 0) {
               outerBounds = maxBounds(outerBounds,bounds);
     return (outerBounds.length == 0) ? null : outerBounds;
function positionVisible(grp,x,y)
     var bounds = getRealVisibleBounds(grp);
     var newX = GetPoints(x) + (grp.left - bounds[0]);
     var newY = GetPoints(y) + (grp.top - bounds[1]);
     grp.position = [newX,newY];
function maxBounds(ary1,ary2) {
     var res = [];
     if(ary1.length == 0)
          res = ary2;
     else if(ary2.length == 0)
          res = ary1;
     else {
          res[0] = Math.min(ary1[0],ary2[0]);
          res[1] = Math.max(ary1[1],ary2[1]);
          res[2] = Math.max(ary1[2],ary2[2]);
          res[3] = Math.min(ary1[3],ary2[3]);
     return res;
and were called like so:
var doc = app.activeDocument;
if (doc.selection.length > 0) {
     var grpOriginal = null;
     if(doc.selection.length == 1 && doc.selection[0].typename == 'GroupItem') {
          grpOriginal = doc.selection[0];
     else {
          grpOriginal = doc.groupItems.add();
     doc.selection = null;
     if(grpOriginal.layer.name != 'approval') {
          grpOriginal.move(doc.layers['approval'],ElementPlacement.PLACEATEND);
     var artBounds = getRealVisibleBounds (grpOriginal);
     var artWidth = artBounds[2] - artBounds[0];
     var artHeight = artBounds[1] - artBounds[3];
     var imprintHeight = GetPoints(0.625);       // 0.625 is new height, hardcoded for demonstration
     var imprintRatio = imprintHeight / artHeight;
     var section = doc.layers['sectionLeft'];    // section where I placed the art
     var grpArt = grpOriginal.duplicate(section,ElementPlacement.PLACEATEND);
     grpArt.resize(100.0 * imprintRatio,100.0 * imprintRatio);
     grpArt.move(section,ElementPlacement.PLACEATBEGINNING);
     grpArt.selected = true;
     positionVisible(grpArt,0.0,16.75);          // x & y hardcoded for demonstration
else {
     alert('artwork needs to be selected');
I'm new at this so are there any better options to do these things?