Page 5 of 6

Re: COLON algorithm

Posted: Mon Jan 20, 2025 9:22 am
by pjw
tofro wrote: Mon Jan 20, 2025 12:58 am <>
And, in my personal opinion, config blocks are even better than the HOME Thing.
How CAN you say that?! ;)

Re: COLON algorithm

Posted: Mon Jan 20, 2025 9:35 am
by RalfR
Anyway, the fact is that we don't have a Home thing for QDOS/Minerva. Wolfgang wrote to me that the version that is at Dilwyn's site wasn't really finished and he didn't even know how it got there. Which is strange, because in principle this ZIP is complete with a description and an assembled file.

Re: COLON algorithm

Posted: Mon Jan 20, 2025 10:19 am
by Derek_Stewart
bwinkel67 wrote: Mon Jan 20, 2025 8:43 am
Derek_Stewart wrote: Mon Jan 20, 2025 7:03 am Minerva is QDOS based and supports Multitasking Superbasic
I think it's easier, for clarity, when referring to QDOS, to focus on the original Sinclair released versions. Minerva and SMSQ/E are reimplemented QDOS compatible operating systems. The differnece between Minerva and SMSQ/E is that the latter won't run on a BBQL.
No Minerva is a JS ROM with bug fixes.

Wikipedia entries are sometimes not totally correct

Re: COLON algorithm

Posted: Mon Jan 20, 2025 2:01 pm
by bwinkel67
Derek_Stewart wrote: Mon Jan 20, 2025 10:19 am No Minerva is a JS ROM with bug fixes.

Wikipedia entries are sometimes not totally correct
Laurence Reeves himself wrote that Wikipedia entry so those are the authors own words. Look at history and you'll see Lauwr, and you can see exactly the text he added, which is almost all that shows now. Most edits just seem to add punctiation. Click on his name and you'll get to here: https://en.wikipedia.org/wiki/User:Lauwr

Re: COLON algorithm

Posted: Mon Jan 20, 2025 2:20 pm
by bwinkel67
swensont wrote: Mon Jan 20, 2025 6:07 am The other is "Structured SuperBasic". SSB can be found here:

http://swensont.epizy.com/ssb272.zip

With documentation in a PDF here:

http://swensont.epizy.com/SSB.pdf
This looks really interesting. Plus it allows syntax highlighting in microEmacs. I haven't used it but if Popopo wants to develop BASIC code on the QL, perhaps this is the way to go.

Re: COLON algorithm

Posted: Mon Jan 20, 2025 2:25 pm
by Popopo
tofro wrote: Mon Jan 20, 2025 12:58 am Config blocks aren't limited to compiled programs or programs in machine code - You can use them in interpreted S*Basic programs as well. Tools to support this can be found here: https://ia801404.us.archive.org/0/items ... index.html

And, in my personal opinion, config blocks are even better than the HOME Thing.
Hummm... very interesting.
It is my first time I heard about config blocks.
Yesterday I understood that my approximation to implement the algorithm with Basic was a way out. Since it cannot manage in a good way (or any way) errors from failing access to a device or a file. The exceptions are caught by the OS and returned the code in a printed text way on the screen, but also breaking out the running program from the basic interpreter.

So, I thought that I have to go directly to C in order to able to catch those exceptions and managing them in order to skip the access to one device and proceed with the next one.

So perhaps those config blocks could be a good way if I want to persist in my wish of using Basic.
Please, could you tell me where I could find some general information about what they are and how they work with the CPU, OS or the system?
(I would like to avoid long readings that are too technical to understand at first or second reading, just to catch the concept.

Thank you a lot tofro

Re: COLON algorithm

Posted: Mon Jan 20, 2025 2:26 pm
by Popopo
Andrew wrote: Mon Jan 20, 2025 1:13 am
tofro wrote: Mon Jan 20, 2025 12:58 am Config blocks aren't limited to compiled programs or programs in machine code - You can use them in interpreted S*Basic programs as well. Tools to support this can be found here: https://ia801404.us.archive.org/0/items ... index.html
I didn't knew that! I learned something new today!
Do you see my dear teacher?
:) another thing to explore and enjoy

Re: COLON algorithm

Posted: Mon Jan 20, 2025 2:27 pm
by Popopo
swensont wrote: Mon Jan 20, 2025 6:07 am Popopo,

There are a couple of tools that will allow you to write SuperBasic programs with no line numbers. "Basic Linker" is one. The other is "Structured SuperBasic". SSB can be found here:

http://swensont.epizy.com/ssb272.zip

With documentation in a PDF here:

http://swensont.epizy.com/SSB.pdf

Tim
Thank you very much Tim. I will try them out

Re: COLON algorithm

Posted: Mon Jan 20, 2025 2:29 pm
by Popopo
bwinkel67 wrote: Mon Jan 20, 2025 6:55 am
Popopo wrote: Mon Jan 20, 2025 12:57 am Without using an external editor (from my PC), is it possible to do into the QL? Or do a need another tool?
Thanks
There are a bunch of editors on the QL that you can use. Heck, you could even use Quill and export as text file. I prefer Jan Bredenbeek's QED, which, in my opinion, is the best. You can find it on Dilwyn's site. It seems to follow the Amiga's ED key sequences.

However, for bigger things I prefer to use a PC text editor (I personally prefer Notepad) and just use an emulator to stick it into an MDV file and then read it via vDrive. That's about the easiest way.
Nice!
I will try QED too.
About vDrive, I have not. But programing with the emulator and the need to test it in my real QL... moves me to finish the project of the bluetooth cartridge for the MDP... uh uh... how many lines opened! :) QL has so many things to improve nowadays that it is the perfect machine for learning and practicing (after ZX 48K)

Re: COLON algorithm

Posted: Mon Jan 20, 2025 3:00 pm
by Popopo
Derek_Stewart wrote: Mon Jan 20, 2025 7:10 am Use a text editor, like QED to write your program withou line number and save.

Then LRUN <program> will run the program.

When I use to use QDOS: JS or Minerva, I used numberless boot program to start the system up and not have the boot file retained in memory.
Hi Derek,
nice to read you.
I "cannot" use LRUN, since when it is executed, the process where was working and variables get lost.
MERGE allows me to combine the running process with new code (in this case reading the value into another basic file).
The idea what I am exploring is to don't give a number line into the second bas file to merge, in order to exceute the assignation to the variable, get it into the memory and then continue with the main BAS program that merge it, using the new variable and its value loaded for a comparative.

In few lines...
1. We are in the fist Bas program
2. We do MRUN or MERGE to the second bas program that contains a line like: LET TOKEN = 1234
3. If the merge is right and without problem we have now in memory the new var called TOKEN with its value.
4. an IF make the comparative between TOKEN and another inner var.

That is the idea. But what I've try to explain above, is the basic interpreter (by now) delete some vars when I merge another file or even worse, it destroy the PC of the basic line of execution (it is a little confuse for me now because sometimes it seems that not).

The goal I seek is to be able in a very easy way to load values from another file and use it. Preferly from a BAS program that everyone can read, modify and use.
It seems that I would need to move to C