RE: Contiguous memory and the *INKx indicators

["Joe Pluta" <joepluta@xxxxxxxxxxxxxxxxx>]

> From: Scott Klement
>
> Couldn't you base them seperately, to avoid potential problems with
> non-contiguous memory?
>
> Another possibility is to ignore the indicators being returned by the
> display file altogether, and instead check the AID byte to see what key
> was pressed.

Scott, both suggestions are good but they don't quite fit my need.  My
PSC400 product automatically converts RPG400 and RPGIV to run on the web,
and to do so, I need to emulate command key indicators.

I don't want to change the program logic, so that rules out using the AID
byte.  For that, I would have to find every place a command key indicator is
used and replace it with a comparison to an AID byte, which would be
particularly painful for left-handed indicators.  It gets even more
confusing when someone uses something like CF03(65), in which case both
*INKC and *IN65 are set on when the users hits F3 (and both are set off if
any other key is pressed).

In RPG400, I do 24 moves to set the command key indicators.  I wanted to
reduce the lines of code, and the based array does that quite nicely.
Basing each indicator individually would actually require MORE lines of
code, because I'd have to do not only the 24 moves, but also the 24 field
definitions and the 24 pointer definitions.  If it turns out that the *INKx
fields aren't contiguous anymore, I'll simply revert to the RPG400 method of
24 moves.

Joe