SBASIC Programming

[dilwyn]

I was offered a slot on this site many, many moons ago to revive Knoware.no, but despite repeated reminders, nothing seems to be happening..

I would give you words of encouragement on this in the sense that the service I've received since my site became a sub-domain of QL Forum has been impeccable.

One or two reports of site down to Rob has resulted in a quick support ticket being raised and site quickly getting back online.

That was also the case when my old site was hosted by Tony Firshman. The move away from Tony's servers had nothing to do with his excellent service, was totally down to the total uselessness of the domain name handling company which failed often and seemed to take ages and usually a real struggle to get anything done. In the end, when someone on the Forum suggested it, I did a clean break and moved everything to the QL Forum hosting. And I have never regretted doing so.

I'd have no worries adding your software to my site as a temporary measure - I have enough free space from QL Forum hosting. It wouldn't be a knoware.no or anything identifiably Per Witte from page or domain names. It'd be pages specifically for your software, rather like I've done for Peter and Maskenlos with Q68 and QIMSI, or like the "Chas Dillon Sources section" page - a "Per Witte software page" - something like www.dilwyn.theqlforum.com/knoware.html or /perwitte.html or dilwyn.theqlforum.com/perwitte/index.html and so on. I don't think I'd be able to give you personal access to upload etc (unless Rob says otherwise). Purely just a temporary measure until you were able to get your regular site online again.

No pressure, but the offer is there if you felt like taking me up on it.

[Derek_Stewart]

Hi Per,

I am interested in you software, I had forgot to look at the files from your shared area. I have been a little busy with non computer work and maybe guilty of being lazy.

I have moved the files onto my Q68 for a test run.

I like QOMBI0, and QWIRC to work Q68 would nice.

I will report back in due course.

[pjw]

Hey Dilwyn,

Thats very kind of you!

However, I must decline. The whole handwritten site is geared up to be at Knoware.no, I fear it will just become a mess.

Also, recently I decided to reorganise all my toolkits and assembler libraries in aid of a system to automatically generate ad hoc toolkits for S*BASIC programs. So until that project is complete (a couple of weeks, I reckon) I wouldnt want Knoware.no online anyway.

I havent been lazy; Ive added loads of new stuff, so once I get back online it will open with a bang, like a Mexican piñata!

I hope one or two people will enjoy it!

Thanks again,

[Andrew]

I was offered a slot on this site many, many moons ago to revive Knoware.no, but despite repeated reminders, nothing seems to be happening..

I hope to see Knoware.no back online soon! I learned a lot from your code.

As per your suggestion (long time ago!) I wanted to use cv2nat in my last iteration of QStar - it really works great. But I could not remember how to find the offsets to initialising code and to the set of definitions of procs/fns, to include cv2nat in compiled QStar. Can you please explain how to find the correct values?

. Ive added some 40 new titles and updates. ... If youd be willing to test some of my latest I could send you a link.

I am also willing to test them - if my limited QL knowledge is sufficient.

[pjw]

<>
As per your suggestion (long time ago!) I wanted to use cv2nat in my last iteration of QStar - it really works great. But I could not remember how to find the offsets to initialising code and to the set of definitions of procs/fns, to include cv2nat in compiled QStar. Can you please explain how to find the correct values?

Oops! It seems that vital information was missing from the "manual". As a general rule you can take it that all my toolkits will have the keyword table offset at 10, as in:

Code: Select all

REMark $$asmb=<path>CV2NAT_bin,0,10
REMark %%<path>CV2NAT_bin,0,10

with the top one for Qlib, and the bottom for Turbo. Change <path> to the location of the toolkit. Other people may have them offset at 12, so if one doesnt work, try the other. In rarer cases other offsets will be used because the toolkit has some initialisation code. This should always be documented.

Do you still have the complete CV2NAT toolkits?

I am also willing to test them - if my limited QL knowledge is sufficient.

Thanks for the offer! I'll send you a PM with a link shortly

[dilwyn]

Hey Dilwyn,

Thats very kind of you!

However, I must decline. The whole handwritten site is geared up to be at Knoware.no, I fear it will just become a mess.

Also, recently I decided to reorganise all my toolkits and assembler libraries in aid of a system to automatically generate ad hoc toolkits for S*BASIC programs. So until that project is complete (a couple of weeks, I reckon) I wouldnt want Knoware.no online anyway.

I havent been lazy; Ive added loads of new stuff, so once I get back online it will open with a bang, like a Mexican piñata!

I hope one or two people will enjoy it!

Thanks again,

No worries - the offer remains open if you change your mind in time when you are ready to go.

If you did, the most obvious way would be to send me the ready-to-go HTML pages including relative links to the downloads, and the download packages, which I just upload to a folder name of your choice (within reason and as long as it didn't already exist).

Thus if you did go with QL Forum hosting eventually like I did, Rob could quickly and easily copy them to their new home with hopefully the relative links ensuring no changes to any links needed in the new home.

[dilwyn]

Oops! It seems that vital information was missing from the "manual". As a general rule you can take it that all my toolkits will have the keyword table offset at 10, as in:

Code: Select all

REMark $$asmb=<path>CV2NAT_bin,0,10
REMark %%<path>CV2NAT_bin,0,10

with the top one for Qlib, and the bottom for Turbo. Change <path> to the location of the toolkit. Other people may have them offset at 12, so if one doesn't work, try the other. In rarer cases other offsets will be used because the toolkit has some initialisation code. This should always be documented.

Here's a very quick and very dirty way to check if the last value, the offset, should be 12 or 10 for most simple toolkits or extensions you want included with QLib or Turbo compilations when there is no documented initialising offset (i.e. 2nd parameter is zero), in case anyone finds it useful when you don't have access to the sources to check what it should be and don't want to go through disassembling to find out.

It really is "dirty" and doesn't handle anything except simple cases of 12 or 10 (i.e. not where the value points much further in), but in most cases it'll work OK just to differentiate between 10 and 12 offsets if you're unsure. Hopefully the REM statements will explain it all. Just works by looking for LEA offset,a1 word values in the first 100 bytes (or rather 50 words) of the code. LEA offset,a1 is how most simple extensions files point to the appropriate part of the code.

Code: Select all

100 REMark find procedure definition table in an extensions file
110 REMark looks for LEA offset,a1
120 REMark or '43FA'<offset.w>
130 REMark dec 67,250 <offset_w>
140 REMark dec. 17402.w <offset.w>
150 REMark offset.w is offset from itself to get start of procblock
160 REMark
170 INPUT'Extensions file?';ef$
180 OPEN_IN #3,ef$
190 byte% = 0
200 FOR a = 0 TO 100 STEP 2
210   GET #3,byte%
220   IF byte% = 17402 THEN
230     GET #3,offset%
240     PRINT'Procblock at ';a+2+offset%
250     EXIT a
260   END IF
270 NEXT a
280   PRINT'Not found.'
290 END FOR a
300 CLOSE #3

[robheaton]

I was offered a slot on this site many, many moons ago to revive Knoware.no, but despite repeated reminders, nothing seems to be happening..

This was completely my fault, Per, I thought I had sent you an email with some info, but I can't find that at all.
I've sent you a PM.

[dilwyn]

Ha, Rob's signature made me laugh!

[stevepoole]

Hi Folks,
Looking down through the list of sections, there seems to be a major design problem with the forum setup. Some sections have not been accessed since 1921, whereas other ones, such as hardware or Software get snowed under with postings in less than a day... So unless we scroll through and across each section's postings, we can easily miss out on something new!
Would it not be more helpfull to keep all one day's section's postings listed on the main page, so as to be able to hit on them with just one click ? New software, posted say, just one week ago may therefore go unnoticed.... It is not everyone that keeps regularly up to date ? Steve.
________________