Yep. I read it and followed along. :-)
I also noticed that the GetNextID method was used for multiple cursors and expressed my uncertainty about Brett's solution in an earlier comment. Since the code inside each case is nearly identical, I'm wondering if that can be further optimized by passing in the array property to be updated and thereby eliminate the DO CASE. I'll let you know what I find.
One other small optimization that I have been trying out, and which does seem to get me back a few more seconds in my control dataset, is to use INDEXSEEK() in the GetCellRecord method instead of SEEK().
If you don't want to talk about m dots, how about tabs vs spaces... :-)
↧