On Monday 12 December 2005 12:29, Jim Elwell wrote: > They DO support the extraordinarily complex job of managing material > and labor resources in a manufacturing environment. But the > programmers can sure seem short-sighted in some ways. For example, > our $80k software allows part descriptions to be only 40 ASCII > characters long -- entirely inadequate. We've complained for years, > there have been numerous new releases, but this hasn't changed. So I > don't have much hope for Unicode being introduced.
The reason for that is probably that the database uses a fixed width column, and increasing the width would require running a conversion program on the database. For the program I'm overhauling, I'm making all strings variable-width; the program internally uses fixed widths, but they can be increased if need be without modifying the database. I'm also storing all numeric data in integer or floating-point; currently they are stored as fixed-width strings, which resulted in garbage in those columns and some data being truncated. I have written a program to convert a data file to a format used by the South Carolina government. This file has all fields fixed-width, and it's mostly spaces but if something is too wide, it's truncated. I found this insane. phma
