doing something like

 cd /Net/cvs/cvs.gedasymbols.org/www/user/
 find . -type f -name \*.fp > list
 .. removeing a breadboard file names with about 1800 pins in it from list
 cat list | while read a; do echo "$a"; grep -i pin "$a"; done > all_pins

and then running

 ./pins.pl < all_pins

I get the files

 file.txt
 data.txt
 cnt.txt

I use file.txt to identify fp files with a lot of pins in it, so that
I can remove thoose file names from list above.

data.txt have columns:
  drill/hole dia, thickness, number of pins found in all_pins

cnt.txt have columns:
  drill/hole dia, total number of pins found in all_pins

Running:

 ./stat.pl < data.txt > stat.txt

I get lines as in cnt.txt, plus colums for: min mean max std.dev
plus last in the file I get parameters for linear regression lines
for the min mean max and mean+std.dev (mpe) columns/values.

I can then enter thoose four sets of linear koefficients in ../gen.pl