A53 This is the header for a function. It shows the explanation of each bit of the declaration. For normal functions, you have to put just before the xNAME xxx : CON(3) 0 If you were defining a command, you have to do: CON(1) 8 string tried to I problems at Of course, these commands just push the arguments on the stack so the system can use these as callbacks. It's different than with the HP48/49 where each application contains itself. On the HP39G everything works with callback, that's why you can switch between applications on the HP39 but not on the HP49. compiling the If you look at my statistic aplet, it contains a Makefile, using Cygnus for Windows, you can just run it. It will create all the required files. Then you have to use the HP39 connectivity kit to send the files, you can't ==== Sorry to take up a little more of your time. Can this be used to make a function at any point in the aplet, or does it have to come before or after the objects in the hash table (or somewhere else)? And does the function have to be in a library to run or can it just be in the regular part of the aplet? So I think I get how it functions in the aplet structure to allow swithcing between tasks, but I am still blank on a way to allow inputs into Inform to be either an id or a string? Can I stop it from being decompliled 39 style and go with the decompilation aka 48/49? I know it's a little more difficult. That's why I am having more problems and difficulties with this and why I appreciate your help even more. As I only have a month to get everything done I'd like, time is running out on me. TW ~The enemy's gate is down. of xNAME have It has to be the nibbles just before xNAME function like this: ASSEMBLE CON(3) 0 RPL xNAME thats_a_function ASSEMBLE CON(1) 8 RPL xNAME thats_a_command Functions can be used inside a symbolic object, command can't the No, functions must be part of a library (and so does commands). Actually, if you want to do anything more system oriented like plotting etc, you have to use a library. But it has a lot of advantages, you can very easily add new keyword in the HP39/40G (unlike calculator like TI83 etc) swithcing to be and go I haven't looked at that yet, once the argument are on the stack, it's the same as the HP48, how a field will be compiled depends on the compilation flags (and decompilation) has been set-up for this object. Check Jim Donnelly's book for more information. Somebody did a InputForm builder for the HP49, it's great and can be used for the HP39. Actually, if you have a look you will see that the input form of my statistics aplets have been made using this generator (including some code in the HP39 :) ==== I am in the right path. For some reason it won't run the create program. If I manually type it in one step at a time, it will work just fine. Running the script however, and it can't find the programs. So to run the first step, I can manually type in ./sasm -e -N SUPROM39 to run the first step. In the script it has sasm -e -N SUPROM39 . It gives back sasm: not found . If I change this to ./sasm -e -N SUPROM39 in the script, it returns ./sasm: not found. So I'm pretty sure it isn't recognizing the line as a command. If I echo it, then it is just echoed and not evaluated, so i need a command that will treat the line as a command entry instead of whatever it is treating it as now. TW ~The enemy's gate is down. It sounds like it is recognising it as a command to me. One problem may be that the current directory is changing in the script and so the location of sasm is no longer .. Try executing this command before running the script: export PATH=$PATH:$(pwd) That'll put the current directory in the path, like JYA's suggestion, ==== Recently I have started trying to learn SysRPL. I have a question about using Metakernel's MASD to compile programs, though: I think I need an entries table so it can recognize commands. What do I use for this? Also, I read something about a file called STARTEXT that tells the compiler where the extable is. Is the format of this file 2/extable.48 @ (with 2 being the port and extable.48 being the name of the extable file)? Scott Spillman ==== In some threads like http://groups.google.com/groups?selm=qzuhexr89ws.fsf%40sand.science.uva.nl the two stable areas 25EC - B3C7 and 25565 - 40000 were mentioned. Now if an entry is unsupported but in this area (e.g. the ML entries SINF 310B3 and COSF 310BA), can I assume it is stable and thus use it in my ==== CHOOSE command much more powerful, in about the INFORM command more powerful. Like InformBuilder, CHOOSEXT is mainly for those who know some UsrRPL but no SysRPL. Programming is based on Carsten's analysis of the involved flashpointer 2 72 - but the user need nothing to know about it. Downloading the library from ftp://ftp.math.fu-berlin.de/pub/usr/raut/HP49/tools/ to the calculator will last but 5 seconds because the library has only 0.4 KB :-) The main advantage is that you can now create choose boxes not only with the standare menu CANCEL and OK, but with *any* application menu. For instance, a menu looking like the menu of the rightshift ALG choose box. Here you have a HELP option in the menu (if flag -117 is clear). You may determine in advance whether a menu option discards or preserves the choose box. There is also a library command, Index, which yields the index of the highlightened item in the choose list. What is NSUB to DOLIST is Index to the new choose-command. Suggestions for improvement are wellcome. Have fun, Wolfgang no, it doesn't. The user is completely free to display his own help-messages. But the application menu of Choose from CHOOSEXT does not only serve to add a help-option. E.g., the ==== Just to put oil in the fire ;-) (Only a joke) Congratulations for this excellent idea. I just downloaded the software. Today afternoon I am going to test it. Now that you made it, I wonder why it was not built-in. It is exactly what is needed for extending the usage of choose boxes. With CHOOSEXT and InFormBuilder Not at all, I just didn't understand your + which seems to mean hold (I always write rightshift-hold in keyman.txt to avoid any misunderstanding). Keyman assigns the rightshift-hold menu keys in such a way that all kind of names are quoted as long as these are known to the system. This is also done in edit-mode, e.g., if some string or other object is currently edited. But it's done *outside* the editor, i.e., the menu name is put into the stack, not put into the command line ! If you don't like this behaviour, just discard the first command in the assignments, the TakeOver. Then the rightshift-hold menu keys would just beep in edit mode. How to modify the assignment? Get it to the stack with RclK from Keyman. There are two possibilities to procede. With extable, just edit the assignment with Sys~ from OT49, eliminate the TakeOver, toggle Sys~ again and reassign with A?D. At the first key should be assigned :: BINT1 GETPROC ; etc. If you don't have extable, explode the assignment with 3tog from OT49 which makes a metaobject with count %3 in Level 1. Replace %3 by %2 and toggle ==== does there exist varibles that are executed when you turn the calculator on and off?? ==== Maybe you could reassign RS-ON to a program like << action-on-turnoff OFF action-on-turnon HTH Thomas -- Thomas Rast If you cannot convince them, t.rast@iname.com confuse them. What's *supposed* to happen is that any program in STARTOFF is run when the calculator turns off with an automatic shutdown, whether the default 5 minute shutdown or the shutdown set by the reserved variable TOFF. Note that STARTOFF would normally have OFF as the last command in the program. To take care of manual turnoffs, I expect that you could assign STARTOFF to the RightShift OFF key. Unfortunately, this doesn't seem to be working for me using ROM 1.19-5; does anyone else experience this? I don't know of anything similar for when the calculator is turned on, but suggest that for most purposes you could just do whatever needs to be done just before the OFF command. Excellent Thomas! Using a calculator with ROM 1.18, I stored in STARTOFF, and it worked as it should, recording the (automatic) turnoff tine and the turn on time. Presumably it could be assigned to the key as well. Anybody know why STARTOFF isn't working for me in my ROM 1.19-5 calculator? Is it affected by flags? Was this feature removed in 1.19-5? James 1.19-5? Belay the last, it didn't work because (blush) I stored my program in STARTOEF instead of STARTOFF. And yes, if you have the program stored in STARTOFF, you can do 'STARTOFF' 101.3 ASN to have it execute when you do a manual turnoff as well as an automatic turnoff, as long as you're in user mode. James ==== Yes. The object STARTUP will be evaluated after a warmstart (ON+F3) and the object STARTOFF will be evaluated when then calc turns off I want to execute subroutine written in SysRPL every one second (any time) - like TIMER in DELPHI. Is it posible in SysRPL (maybe parametrized outer loop) or only in ==== Actually I am (was) working on a 49 port of exAL but I got side tracked. I have considered picking the project back up lately. I would also like to write a set of routines to aid in my new found hobby of astronomy. Seems there is always so much I *want* to do. Anyway, I will begin working again on the spreadsheet program as time permits. Of course anyone is welcome to contribute and/or otherwise assist. One of the main things that needs to be done before anything else is a fast, small and configurable input routine similar to the one implemented in the 48 version. It was written by Arnold Moy and was quite good but since it was done in assembly I have not had the time to port it to 49. Also, if anyone knows of available online resources discussing the ==== Yeahh - I'd like to assist/contribute but I'm not that experienced. . . ==== Terrorism is a scalar, not a boolean. It ranges from mere murder, like the brave and fearless souls who shoved Leon Klinghoffer's wheelchair overboard on the Achille Lauro, thereby striking a powerful blow for Palestinian freedom, up to Atta and his merry men, sacrificing their own lives in order to punish the crusading United States by murdering 6000 or more people who had nothing more to do with Islam than Klinghoffer did Palestine. Do not become too entranced with the magnitude of the September 11 tragedy, Thomas. It differs only in scale, not substance. Terrorism sprouts from tiny seeds, the harvest, large or small, is the same. But it is far easier to detect and kill the seed, before the harvest. -- .-. .-. .---. .---. .-..-. | Wild Open Source Inc. `----'`-^-'`-'`-'`-'`-' `-' | little more commonplace. home: www.smith-house.org | work: www.wildopensource.com ==== It may be that the terrorists thought they murdered as a retaliation and for their religion or their god. But is this violence any good? Now the US will (or plan to) strike back. What then? Other terrorists will probably do a new attack. I mean, we all know what that leads to, Israel and Palestine still don't have a working peace agreement. It's not only the magnitute of the attacks itself. It's also the impact they have, or will have, on our life. They've destroyed a part of our dream of freedom. Now all people from certain countries may need visa again to travel to the US, for example. Similiarly, the man running amok has endangered a part of the Swiss tradition, being that politicians have a close relation to the people, travel to work like everybody else, walk around on their own (i.e. no bodyguards, no armoured cars), and everyone can go and watch the sessions (?) etc. Fortunately in most cases there are only extra security checks at the entrance, no limitations on who may visit the buildings and sessions (? again). I'm not sure whether I got you right but... do you really want to arrest all alt.flame posters?? Would you call everyone who doesn't always show the expected or appropriate respect a terrorist?? Yes, it's in most cases easier to kill the seed than the plant. BUT: In this special case I still don't understand how anybody can even *dare* to draw a line from Nick to terrorism. Especially not after the recent attacks. ==== To terrorize others has a wide range of meanings. I merely wanted to say that I feel terrorized in a newsgroup if somebody sends dozens of messages a day without essential substance. Nick seemingly did not understand the point. He continued today to post a lot of things without substantical information. ==== First of all, your choice of wording was at least very inappropriate, since we had just suffered that terrible terrorist attack on NY. Your meaning would of course be misunderstood - by calling Nick a terrorist, what would people associate with? You cannot deny this - your post was very agressive and insulting towards Nick, and they still are. Secondly, who decides what is essential substance? What if I do fancy more hearing Nicks small post, than about Keyman or something else twice a day? Personllay, I do not use Keyman, and therefore have no use of your posts regarding this. I do even not care if the HP49G will eventualy have 15 or 43 key levels. See? Do I put you in a kill filter? Neither did I then. Or maybe I did understand after all. Are you moderating Nicks posts? You do not have that previlige - thank God. Please stop being so egoistic - I have no problem filtering this group with my brain. It certainly shouldn't be a problem for you either. It's not seen as a kind feature surpressing other people. Well, Wolfgang, you just do that. Who else should we put there? I remember you writing I think this newsgroup is a community, when someone posted right after september 11th. Are you sure you believe that, or were you just howling with the wolves? ====