Threaded Mode | Linear Mode

DavidM · 08-14-2016, 01:34 PM

(08-14-2016 05:55 AM)JDW Wrote: David, I typed in the following code on my 48GX:

Code:

<< #21d #8d BLANK "" 2 →LIST DUP DUP DUP DUP DUP 6 →LIST TMENU >>

(To input that right-arrow to the left of LIST, I pressed the blue/green right-shift key followed by the 0 key.)

When I pressed ENTER the 21 and 8 decimal values changed to binary (10101b and 1000b).

The 21 and 8 values changed because your current base display mode is set to binary. The values compiled into your program are still the same, they are simply being displayed using base 2 instead of base 10 (or hex, or octal).

(08-14-2016 05:55 AM)JDW Wrote: I then saved it as "MENu" using STO and then tapped that new softmenu to run it, but that results in the following error:

Code:

→LIST Error: Too Few Arguments { Graphic 21 x 8 "" } { Graphic 21 x 8 "" } { Graphic 21 x 8 "" } 6

Please advise.

My first guess is that you may have missed a DUP (there should be 5 of them). The evidence would suggest that there are less than 6 of the '{ Graphic 21 x 8 "" }' objects on the stack at that point in the running code, which shouldn't have happened if all the DUPs are in place.

I don't have a 48GX to test this on, but I have successfully run the code as supplied on both a 48SX and 50g with no problems. It also ran on an emulated 48GX without issue, so I'd look for a typo as the culprit.

(08-14-2016 07:19 AM)JDW Wrote: Accutron,

Thank you for the specifics. I have the newest "R" revision ROM in my 48GX, ...

The menus flickered off and back on in perhaps 100ms. They didn't stay off.

I then tried #4E347h, but with the menus already showing that did nothing at all.

Looks like a no-go.

That's exactly what should have happened. TURNMENUOFF only works within the context of a running SysRPL program; the menu disappears, and the ABUFF "text grob" is expanded into that area for your program to be able to draw into it as a contiguous part of the screen. All of that is reset as soon as the program stops, however. At your program's conclusion, the calculator's normal stack display is resumed.

Attempting to use that area for stack display for normal calculator operations really is a no-go. The 48-50 series systems don't provide any built-in support for that, and you'd really have to hack the system to get that to work.

Also, please be aware that the SYSEVAL command is potentially very dangerous. You are essentially telling your calculator to start running whatever code happens to be at a specific memory address, which it will do without consideration for what's there. Data loss and corruption can result if you supply an address that isn't what you originally intended (such as what happens when you use an address meant for a different platform). It sounds like you may have run into this situation. If you plan on using that command, make frequent backups and be prepared to restore your calculator after problems occur.