🚀 go-pugleaf

Michael,

You can optimise two things in your code. First, avoid referring to collections (such as myTable.rows and myTable.columns) repeatedly. Instead, create arrays from collections. Earlier this week this was discussed here: Harbs, "[JS] :: [CSX] - PageItem Stacking Order" #15, 6 Jan 2009 11:23 am </webx?14@@.3bbf275d.59b767cf/14> . see especially the reference to Kris Coppetier's article in Harbs's post, who explains that each time you refer to a collection, InDesign rebuilds it -- so you can imagine that rebuilding a big table such as yours causes the script to run for hours.

In your case, instead of referring to myTable.rows.length-1, do this:

myRows = myTable.rows.everyItem().getElements();
myRows_length = myRows.length-1;
for (var f = myRows; . . .

Secondly, you needn't refer to myRows[f].columns[...].cells[...]: myRows[...].cells[...] will do too. So your script coulkd look like this:

myRows = myTable.rows.everyItem().getElements(); 
myRows_length = myRows.length-1; 
for (var f = myRows_length; f >= 0; f--) 
   myRows[f].cells[3].merge (myRows[f].cells[4]);

This will probably speed things up.

Peter

I did some timing experiments a while back on referencing cells in tables and I was amazed to discover that the fastest way to reference a cell is:

myTable.cells.item("c:r")

where c is the column number and r is the row number. This was significantly faster (in my tests, probably in early CS3 days but might have been earlier) than any other method I could think of.

Dave

Let me correct that a bit. What I actually used was:

myTable.cells.item(c + ":" + r);

where c and r and numbers and so can be derived from loop variables.

Dave

Interesting -- thanks.

Peter

Thanks to both of you. I'll try both of your suggestions and see how much faster I can make my script.

One more question - in your experience, do scripts seem to get bogged down with large tables? For example, my table script was running on my 120 page table for around 12 hours and still wasn't finished. But, when I cut the table down to 25 pages, the script completed in under an hour.

I thought the change in the time it would take to run would be proportional, but it doesn't seem to be. I don't think the loops get stuck running indefinitely, but I could be wrong.

Thanks again!

Are you doing anything about the undo queue in your script? As that gets longer and longer it can really bog down a script.

I usually have something like this in my main loop:

if (j % 100 == 0) { 
  myDoc.save(File(myFullName)); 
}

where j is the main loop counter.

Dave