> When HASH.HELP became robust enough I quit using my program as I couldn't > justify the development time/effort. Instead I wrote a front-end which > generated paragraphs consisting primarily of HASH.HELP %filename% > statements. The paragraphs included COMO statements, and the final > statement ran a program to interpret the output from the HASH.HELP and > execute an appropriate RESIZE command.
Many users do this. But beware that HASH.HELP is not perfect. There is one bug, for example, that we exploit in an example in the UK UniVerse training courses, where HASH.HELP gets it hopelessly wrong but which is easy to fix by hand. Personally, I am a great believer in dynamic files. About five years ago I did a migration from PI/open to UniVerse for a site with over 13000 files. There was absolutely no way that I was going to do file calculations for all of those so I made the lot dynamic and said that I would tune up any that presented performance problems. I visited the site last week. They are still overjoyed with performance, have spent absolutely NO time on file tuning, and have experienced no problems with these files. Of course, I have lost a good source of income providing a regular file tuning service! Martin Phillips Ladybridge Systems 17b Coldstream Lane, Hardingstone, Northampton NN4 6DB +44-(0)1604-709200 -- u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users
