Technical Note:

Text to dictionary conversions can be very slow if the text file is not sorted. Excalibur is in the sort phase when the progress indicator is between 50% and 75%. If you sorted your text file before initiating a text to dictionary conversion, and Excalibur is still very slow when the progress indicator is in this area, you may need to change the way you do the sort.

First, be sure that all the words are in lower case letters before doing the sort. This will have an effect on the sort.

Second, not all machines treat the characters in Apple's extended character set in the same way. If you sort your file on a Unix workstation, check to see where the words that begin with an accented character appear. If these words appear at the end of the file after a sort, then everything is fine. If they all appear at the top of the file, then there's a problem. Your machine is treating these characters as signed characters. Since they have their high order bit set, they have the smallest keys. This is not the same ordering that the Macintosh uses. See if your sort routine has a switch that lets you treat all characters as unsigned. If you're desperate, send some mail to zaccone@bucknell.edu. I've put together a quick and dirty sort routine that works properly on a Unix workstation. You will need an ANSI C compiler to compile it. (I use gcc.) A sorted file should take at most several minutes to convert to a dictionary. An unsorted file may take several hours.