CompuServe Thread

Mike, when I returned from Cleveland, I took a shot at changing my version of FLTIBM along the lines sugested by your messages, partly to understand WS and partly as a learning experience. After something of a struggle, my version is working. I found it necessary to convert all characters >80h ( subtract 80h from them), to avoid graphics on the Model 100 screen; WS sets a number of characters high, apparently for the purpose of justification ( i.e., most spaces must be soft to allow expansion in the case of right justification — or however they do it). I also found it necessary to: replace all soft cr's by nulls; replace all hard & soft lf's by nulls; and to strip ^B, ^V, ^S, ^D, & ^T (this is a function of which print control characters you use in WS; there are others). There are two kinds of soft hyphens, by the way — internal to the line and at the end of a line. And there was a "non-break space" to be dealt with. COM file is now 1080 bytes, up from 822! FLTIBM.ASM = FLTWS.ASM weighs in at 19.1Kbytes. This includes some reductions in the code, and moving some things around a bit ( e.g., all DB etal at the start of the program). My guess is that the market for this one is not going to be that large ( usually go up to the IBM for adding to a WS file, not the other direction- – and not all IBM users are WS users). But it was a fun thing to do. And I learned a few things (like don't try to make long conditional jumps)! I'd like to say I've learned "if it works, don't fix it" — but I have not. I'm anxious to compare notes with you when you can get back on the SIG. We should compare implementations before uploading either (if an upload is the way to go). I sent you an earlier version (shorter!) of this message on Thursday, but it's scrolled off — so this is a somewhat expanded repeat of it! Regards, Phil