new (may be not new) Sedoric bug found (all versions)

This is the best place to discuss about the various Oric operating systems like Sedoric, Randos, FT-Dos, and others, as well as serious software, utilities, word processors, disassemblers, etc... that runs on oric computers.
User avatar
waskol
Flight Lieutenant
Posts: 414
Joined: Wed Jun 13, 2007 8:20 pm
Location: FRANCE, Paris

new (may be not new) Sedoric bug found (all versions)

Post by waskol »

In "Sedoric à nu" of Andre Cheramy, I saw that there was a problem preventing to use the DIM commans between an OPEN and a CLOSE (file access).

Well, I think I saw another one...

Between an OPEN and a CLOSE, you cannot USE any multi-dimensional array of integer.

For instance :

Code: Select all

10 DIM T%(10,10)
20 OPEN S,"FILE.BIN",1:REWIND 1
30 FOR I=0 TO 10
40 FOR J=0 TO 10
50 TAKE 1,R%
60 T%(I,J)=R%:' <---BUG !!! (BAD RECORD NUMBER....)
70 NEXT J
80 NEXT I
90 CLOSE 1
With a one-dimensional array of integer, it's OK :

Code: Select all

10 DIM T%(10)
20 OPEN S,"FILE.BIN",1:REWIND 1
30 FOR I=0 TO 10
50 TAKE 1,T%(I)
80 NEXT I
90 CLOSE 1
With a multi-dimensionnal array of floating values, it works too :

Code: Select all

10 DIM T(10,10)
20 OPEN S,"FILE.BIN",1:REWIND 1
30 FOR I=0 TO 10
40 FOR J=0 TO 10
50 PUT 1,T(I,J)
70 NEXT J
80 NEXT I
90 CLOSE 1
It took me 3 days to understand what was going wrong with my array of integers !!!
User avatar
Dbug
Site Admin
Posts: 4438
Joined: Fri Jan 06, 2006 10:00 pm
Location: Oslo, Norway
Contact:

Post by Dbug »

I wonder if these bugs are comming from the ROM side of things, or in the Sedoric implementation.

This kind of things could motivate the creation of a new release of Sedoric :)
User avatar
waskol
Flight Lieutenant
Posts: 414
Joined: Wed Jun 13, 2007 8:20 pm
Location: FRANCE, Paris

Post by waskol »

Hello DBUG :P

In facts, :in "SEDOIC à nu", Andre points out very well that in the current and past implementations of the SED, that when the OPEN command is treated, a new array named FI is created along side those already existing in the Oric memory part that deals with arrays created in BASIC.
In fact, this new array is inserted (and not append) just before the other arrays, then removed once you have CLOSEd all the files.

Andre states also that the way it is managed (during a PUT or a a TAKE), or the way the other arrays are managed (elsewhere between the OPEN and the CLOSE) are done a little bit "à la hussarde" (excuse me for that french expression). For this reason, Andre tells to never use a DIM statment between an OPEN and a CLOSE.
I think that the same implementation brings some problems with multidimmensionnal arrays of integer values. :wink:

Andre, if you read me... may be I brought you the clue you needed to solve this for a SEDORIC v4 :wink:
Ray030471
2nd Star Corporal
Posts: 23
Joined: Mon Jan 25, 2021 9:18 pm

Re: new (may be not new) Sedoric bug found (all versions)

Post by Ray030471 »

It's been a while since this bug was reported and it's also been a while since I worked with my ORIC but I think that I have found the cause of this problem.
I am in the process of extending my Assembler to cover all SEDORIC commands and, in doing this, I realised that the (special) FI array is probably encoded incorrectly in the array data space of a Basic program. I looked in this forum to see if this had already been noticed, hence this posting here.
In the "SEDORIC A NU" disassembly listing, pages 89 and 206 are relevant to this issue. The FI array is encoded effectively as an integer array, ie both name characters inverted, whereas I believe that only the first character should be inverted (as for a function in the scalar data space of a Basic program).
So, currently, any sufficiently large integer array, with any number of dimensions, will cause this problem.

I don't have a copy of the very first version of SEDORIC and so I cannot determine when this bug was introduced. I would appreciate it if anyone could give me a ".dsk" file for this so that I can carry out research.

I have removed this bug in my own copy of SEDORIC V2 which now performs as it should in this respect.
User avatar
iss
Wing Commander
Posts: 1637
Joined: Sat Apr 03, 2010 5:43 pm
Location: Bulgaria
Contact:

Re: new (may be not new) Sedoric bug found (all versions)

Post by iss »

Ray030471 wrote: Sun Jan 07, 2024 12:45 pm I don't have a copy of the very first version of SEDORIC and so I cannot determine when this bug was introduced. I would appreciate it if anyone could give me a ".dsk" file for this so that I can carry out research.
Here are all available SEDORIC versions on oric.org: click!
(You need to be logged-in for downloads).

(PS: Any news about BD-DOS? :D)
User avatar
Chema
Game master
Posts: 3013
Joined: Tue Jan 17, 2006 10:55 am
Location: Gijón, SPAIN
Contact:

Re: new (may be not new) Sedoric bug found (all versions)

Post by Chema »

It would be nice to have an article on all the different versions, bugs and features. I did not know so many versions existed!

Also for the rest of OSes in the Oric. There is little tech info on them...
Jonh
Pilot Officer
Posts: 69
Joined: Tue Jun 01, 2021 4:07 pm
Location: Cambridge, UK

Re: new (may be not new) Sedoric bug found (all versions)

Post by Jonh »

Lots of info (to V3.0) in 'Sedoric à Nu' by Andrê, particularly Annexes 17 (bugs) and 23 (different OSs). V4.0 simply (?) added multiple directories.
Available here: https://library.defence-force.org/index ... h=sedoric/

I believe both V3.0 and V4.0 are still in regular use, personally I get on so well with V3.0 that I've never used V4.0, I have no need of multiple directories... So Ray, please do both versions if you solve the bug! May I suggest eradicating the bug would not justify a V5.0, but rather a V3.1 and a V4.1 so that their separate identities are preserved?
User avatar
Chema
Game master
Posts: 3013
Joined: Tue Jan 17, 2006 10:55 am
Location: Gijón, SPAIN
Contact:

Re: new (may be not new) Sedoric bug found (all versions)

Post by Chema »

Yes, but a) it is in French and b) online version lacks formatting and, most importantly, the last annexes :(
Jonh
Pilot Officer
Posts: 69
Joined: Tue Jun 01, 2021 4:07 pm
Location: Cambridge, UK

Re: new (may be not new) Sedoric bug found (all versions)

Post by Jonh »

Hi Chema,

a) True!
b) this is odd, it downloads as a 1.6Mb PDF file, best to Acrobat Reader, where it is fully formatted and complete. Annex 17 is at page 569, Annex 23 at page 611.
User avatar
Chema
Game master
Posts: 3013
Joined: Tue Jan 17, 2006 10:55 am
Location: Gijón, SPAIN
Contact:

Re: new (may be not new) Sedoric bug found (all versions)

Post by Chema »

Jonh wrote: Thu Jan 11, 2024 2:16 am b) this is odd, it downloads as a 1.6Mb PDF file, best to Acrobat Reader, where it is fully formatted and complete. Annex 17 is at page 569, Annex 23 at page 611.
Uh... I made a mistake then I opened an html file. This is good for automatic translation, but it was not complete. I'll double check.
Ray030471
2nd Star Corporal
Posts: 23
Joined: Mon Jan 25, 2021 9:18 pm

Re: new (may be not new) Sedoric bug found (all versions)

Post by Ray030471 »

Thanks for the link to the CEO site. I have registered and had a good look at SEDORIC V1.7 which also has this FI% bug. However, this version is not the original (it has some of my modifications in it) and it would be nice to have the very first version of SEDORIC (which I hadn't modified).
I have also obtained a copy of OUM_Disc3 and that SEDORIC also has the FI% bug.
Jonh
Pilot Officer
Posts: 69
Joined: Tue Jun 01, 2021 4:07 pm
Location: Cambridge, UK

Re: new (may be not new) Sedoric bug found (all versions)

Post by Jonh »

This is interesting. Here in the UK we long used V1.7 before you produced V2.0 for expanded disk storage. However, it seems V1.7 arose from some early bug fixes by yourself. The original release in France appears to have been V1.006 :

To quote André in Sedoric 3.0 à Nu, p.616 (link above, my translation) - "SEDORIC has had several versions, the most notable being version 1.006 dated 01/01/86, followed much later by versions 2.x et 3.0." So France was on V1.006 while we in the UK benefitted from V1.7

I do have a number of early Sedoric formatted 3" disks, sent to me back in the 80's from France. I need to check if they are Master disks. Before I look at them, could you provide me with a DEEK or two to check whether your bug fix code in V1.7 is present or not?
Ray030471
2nd Star Corporal
Posts: 23
Joined: Mon Jan 25, 2021 9:18 pm

Re: new (may be not new) Sedoric bug found (all versions)

Post by Ray030471 »

I have V1.3 and it has the TAKE correction (for reading a real variable from a Sequential file) which I created.
Relevant bytes for checking are between $F5FE and $F609 inclusively.
The updated bytes (corrected TAKE) are
A5 02
A4 03
20 BA D2
4C 20 F6
EA
EA
whereas the original bytes are
A0 04
B1 02
99 D0 00
88
10 F8
30 16

I hope that someone can help me to determine when this FI% bug was first introduced.

By the way, my BD500 development project is going well and I just have a bit more testing to do before I release it.
In relation to this, I could do with another floppy disk interface so that I can make a few hardware adjustments to improve the running of this BD500 operating system on an ORIC/CUMANA disk interface. Is there a spare one out there?
Jonh
Pilot Officer
Posts: 69
Joined: Tue Jun 01, 2021 4:07 pm
Location: Cambridge, UK

Re: new (may be not new) Sedoric bug found (all versions)

Post by Jonh »

Ray, I've found a Sedoric disk that announces itself in the first sector as 'V1.006 01/01/86' described by André as the original release of Sedoric. It is a Master disk containing the first issue of the Club Oric International Journal from January 1987, so only a year on.

I've sent you a Private Message about getting a copy to you.
User avatar
coco.oric
Squad Leader
Posts: 720
Joined: Tue Aug 11, 2009 9:50 am
Location: North of France
Contact:

Re: new (may be not new) Sedoric bug found (all versions)

Post by coco.oric »

V4.0 simply (?) added multiple directories.
Hello,
Sedoric 4.0 not only add directories but some bugs were corrected too.
To be checked if the bug remains in distributed at https://www.oric.org/software/sedoric_v4_0-2338.html

It was in my long time todo list to maintain it. if the bug remains, may be my motivation will be higher :D
coco.oric as DidierV, CEO Member
Historic owner of Oric, Apple II, Atari ST, Amiga
Post Reply