June 2021 - Oberon - mailman3.inf.ethz.ch

Re: [Oberon] UEFI
by peter＠easthope.ca 15 Jun '21

15 Jun '21

Liam Proven lproven at gmail.com Tue Nov 24 17:21:59 CET 2020 > As such, the fact that UEFI works so badly with non-MS OSes seems > likely to be quite intentional to me, and that it only cooperates with > big enterprise server OS vendors. The situation is difficult for small > FOSS players and not materially improving. > ... > The Oberon community needs to try to ensure that Native Oberon > and/or A2 continue to work on vanilla PC hardware. Liam, does UEFI being proprietary and undocumented mean that Native Oberon or A2 on contemporary machines is impossible without Herculean reverse engineering? Even if UEFI is navigated, the backdoor problem remains. Another motivation to put Oberon on the Skulski machines? Andreas suggested RISC-V. Can it also have a backdoor? Regards, ... P.L. -- tel: +1 604 670 0140

2 1

(small) bug in Fractions.Generate (in SmallPrograms.Mod)
by Hans Klaver 13 Jun '21

13 Jun '21

Hi all, Some time ago I spotted a small but pesky inaccuracy in the output of the command Fractions.Generate (part of SmallPrograms.Mod in the online sources of Project Oberon). Until then I only knew the versions of this program in the books 'Programming in Oberon' and 'Programming, a Tutorial', which give the correct output. Last December I reported this by e-mail to prof. Wirth, but I didn't get any reply. I hope he is doing well. The online version that is part of Project Oberon gives the following output (with => pointing to the correct output): 2 .5'0 3 .3'3 => .'3 4 .25'0 5 .2'0 6 .1'6 7 .1'428571 => .'142857 8 .125'0 9 .1'1 => .'1 10 .1'0 11 .0'90 => .'09 12 .08'3 13 .0'769230 => .'076923 14 .0'714285 15 .0'6 16 .0625'0 I corrected the Oberon-07 source code as follows to let it produce the correct output (corrected lines indicated by (**) ). ORP.Compile @/s Fractions.Generate 16 MODULE Fractions; (*NW 10.10.07; Tabulate fractions 1/n*) IMPORT Texts, Oberon; CONST Base = 10; N = 32; VAR W: Texts.Writer; (* (* Original version, with bug *) PROCEDURE Generate*; VAR i, j, m, r: INTEGER; d: ARRAY N OF INTEGER; (*digits*) x: ARRAY N OF INTEGER; (*index*) S: Texts.Scanner; BEGIN Texts.OpenScanner(S, Oberon.Par.text, Oberon.Par.pos); Texts.Scan(S); IF (S.class = Texts.Int) & (S.i < N) THEN i := 2; WHILE i <= S.i DO j := 0; WHILE j < i DO x[j] := 0; INC(j) END ; m := 0; r := 1; WHILE x[r] = 0 DO x[r] := m; r := Base*r; d[m] := r DIV i; r := r MOD i; INC(m) END ; Texts.WriteInt(W, i, 5); Texts.Write(W, 9X); Texts.Write(W, "."); j := 0; WHILE j < x[r] DO Texts.Write(W, CHR(d[j] + 48)); INC(j) END ; Texts.Write(W, "'"); WHILE j < m DO Texts.Write(W, CHR(d[j] + 48)); INC(j) END ; Texts.WriteLn(W); Texts.Append(Oberon.Log, W.buf); INC(i) END END END Generate; *) (* Corrected version *) PROCEDURE Generate*; VAR i, j, m, r: INTEGER; d: ARRAY N OF INTEGER; (*digits*) x: ARRAY N OF INTEGER; (*index*) S: Texts.Scanner; BEGIN Texts.OpenScanner(S, Oberon.Par.text, Oberon.Par.pos); Texts.Scan(S); IF (S.class = Texts.Int) & (S.i <= N) THEN i := 2; WHILE i <= S.i DO j := 0; WHILE j < i DO x[j] := 0; INC(j) END ; m := 0; r := 1; WHILE x[r] = 0 DO INC(m); (**) x[r] := m; r := Base*r; d[m] := r DIV i; r := r MOD i (**) END ; Texts.WriteInt(W, i, 5); Texts.Write(W, 9X); Texts.Write(W, "."); j := 1; (**) WHILE j < x[r] DO Texts.Write(W, CHR(d[j] + 48)); INC(j) END ; Texts.Write(W, "'"); j := x[r]; (**) WHILE j <= m DO Texts.Write(W, CHR(d[j] + 48)); INC(j) END ; Texts.WriteLn(W); Texts.Append(Oberon.Log, W.buf); INC(i) END END END Generate; BEGIN Texts.OpenWriter(W) END Fractions. -- Hans Klaver

1 0

Re (2): Oberon Text and new systems;
by peter＠easthope.ca 12 Jun '21

12 Jun '21

Michael, Our preceding two emails were mostly at crossed purposes. Two use cases can be visualized. One, a system where the programming content of computer science can be taught. For that, I understand that V5, Text included, works beautifully. Two, a simple workstation for engineering and science. That's what I understand interests Wojciech. V4 and ETH Oberon have been used in that capacity for over two decades already. Update of V4 and ETH Oberon to run on a FPGA machine is conceivable. From: Michael Schierl <schierlm(a)gmx.de> Date: Tue, 8 Jun 2021 23:02:57 +0200 > So, I would not try to provide any (badly compatible) HTML5 support to > a system like Oberon which tries to be simple and tries to remove what > can be removed. Acknowledged. I don't imagine a Web browser. Rather an alternative to Oberon Text. It exists already in V4 as Web.Open <something>.html and in ETH Oberon as Desktops.OpenDoc <something>.html . Capabilities are very limited but comparable to Text. A few revisions for consistency with HTML5 are visible in https://en.wikibooks.org/wiki/Oberon/A2#Sources,_Tools_and_Configuration_Te… . Sources Oberon.HTML.Mod and Oberon.HTMLDocs.Mod . > By the way, V5 text does not have vertical offsets, so no > subscripts/superscripts. Acknowledged. Vertical offset has no significance in Oberon source Text. Not needed in V5. A few observations. * As ETHO is distributed, the Oberon.Text configuration file, is Oberon Text. The system boots and runs just as well if the file is stored as plain ASCII. * The configuration file in A2 is Configuration.XML. It can be edited with most editors in a contemporary Linux, MacOS or MS Windows system. * Most of the A2 sources in the repository at https://svn-dept.inf.ethz.ch/svn/lecturers/a2/trunk/source/ are plain ASCII. Once authenticated, sources in the left column at https://en.wikibooks.org/wiki/Oberon/A2#Sources,_Tools_and_Configuration_Te… can be displayed with a single mouse click. Simplification of a workstation is possible and worthwhile. Regards, ... P.L. -- tel: +1 604 670 0140 Bcc: peter at easthope. ca

4 5

Re: [Oberon] FPGA based machine? Was, Oberon Text and new systems for science.
by peter＠easthope.ca 12 Jun '21

12 Jun '21

Chris Burrows cfbsoftware at gmail.com Thu Jun 10 09:05:51 CEST 2021 > I'm left wondering: your V4 software will still run today on Windows > 10 running Linz V4 will it not? If not, why not? If it does run OK, > what are you hoping to gain by repeating the exercise on the FPGA? Near 49:50 in lecture "Reviving a computer system of 25 years ago - Wirth, 2014", https://www.youtube.com/watch?v=EXY78gPMvl0 , a similar question was asked, with Prof. WIrth answering "... no side doors and no back doors." As anticipated, cyberattacks have been in the news recently. https://www.power-technology.com/features/the-five-worst-cyberattacks-again… Criminal interference is as much a concern in science and engineering as in banking and commerce. Sooner or later, the Skulski machines might attract serious interest. Regards, ... P.L. -- tel: +1 604 670 0140 Bcc: peter at easthope. ca

1 0

Oberon Text and new systems; was Re (n): Question about Oakwood guidelines and design choices
by peter＠easthope.ca 10 Jun '21

10 Jun '21

From: "Skulski, Wojciech" <skulski(a)pas.rochester.edu> Date: Mon, 7 Jun 2021 17:21:05 +0000 > Since you said Text, please be aware that there are multiple Text > flavors out there. The section at https://en.wikibooks.org/wiki/Oberon#Oberon_Text is intended to be understandable to a novice. Under the subheading "Additional Details" is the link to https://en.wikibooks.org/wiki/Oberon/Text where the variant systems are addressed in more detail. Sorry to say, V4 is not yet included. =8~( When encountering Text, visualization of the record types was difficult for me. The tables you see in the second page cited above are my way of mitigating difficulty. > 3) I believe that System 3 did about the same as #2 above, but in a > more complex way. In ETH Oberon and in A2, Texts.TextDesc is an extension of Objects.ObjDesc. Consequently, understanding the Text type requires understanding records defined in two modules--Objects and Texts. An additional difficulty for a newbie. In the tables, the extension is represented by background color shading. The display of types concisely in one place should help a novice. > 6) I suspect that porting #2 [V4] to the FPGA Oberon would entail a > complete replacement of the 2013 Text and text frames with the ones > from Linz V4. This will not be easy not because of the code, but > because of the language: "they" changed the language so V4 will no > longer compile. Didn't Andreas discuss or mention that months ago. Three steps to put an "old" system, S, on a new machine. (1) In an extant S, translate the sources into Oberon-07. In the process, the sources can be recompiled by the compiler in S. This is just work; time consuming but not lethal. =8~) (2) In the new system, compile all the sources of S. (3) Install and run the revised S on the new machine. S can be V4 or ETH Oberon or another. > I am afraid that (4+Gadgets) is unlikely to fit into a small memory > footprint. For basic Text functionality, the complete Gadgets machinery is not necessary. Build a system with the Objects module and the Texts module? I haven't evaluated the memory requirement. If someone drafts a V4 section and adds it to https://en.wikibooks.org/wiki/Oberon/Text , the contribution will be welcome. Accept that merciless editing can happen. =8~| Should a universal Oberon Text type be derived from V4, ETH Oberon or V5? Also acknowledge that HTML5 is far more ubiquitous than any Oberon Text type. Hypothetically an Oberon system can be built with plain ASCII text and HTML5 and no Oberon Text. Discussion by the knowledgeale and experienced people can help. Regards, ... P.L. -- tel: +1 604 670 0140 Bcc: peter at easthope. ca

6 7

Correction: Re: Oberon Text and new systems;
by peter＠easthope.ca 08 Jun '21

08 Jun '21

From: peter(a)easthope.ca Date: Tue, 08 Jun 2021 09:24:51 -0700 > ... HTML5 is far more ubiquitous ... Poor word usage. Better to say, HTML5 is ubiquitous. Regards, ... P.L. -- tel: +1 604 670 0140 Bcc: peter at easthope. ca

1 0

IRC oberon community is moving elsewhere
by Diego Sardina 08 Jun '21

08 Jun '21

Due to recent changes concerning operational control of freenode, we have decided to move elsewhere. You can read more about it in: https://en.wikipedia.org/wiki/Freenode#Ownership_change_and_conflict If you want to go deeper (and you have time to do that): https://gist.github.com/joepie91/df80d8d36cd9d1bde46ba018af497409 Many opensource projects and communities moved to the new network Libera.Chat or the freenode's alternative OFTC (Open and Free Technology Community). While some of us already moved to Libera.Chat, some others asked to move to OFTC. In my opinion, since Libera.Chat and OFTC are very crowded, it makes sense to be present in both networks. In recent years, in particular with Project Oberon 2013, lot of people coming from other channels joined #oberon to ask information both about the system and the language. The channel also increased from a number of users less than 10 to an average of 20 with a peak of 31 during last years. For these reasons being in both networks is the best way to widespride Oberon and let new users to join the (small) community. So people that already moved to #oberon on Libera.Chat are also invited to join and populate #oberon on OFTC. The channel on Freenode will be shut down on 2021-06-21. -- Diego Sardina

1 0

Question about Oakwood guidelines and design choices
by Robert Doiel 08 Jun '21

08 Jun '21

Good Morning, I was looking at the Oakwood Guidelines this morning trying to sort out why an implementation of `Out.Real()` might only accept a single parameter versus two (like `Out.Int()` ). The Oakwood guidelines I am reading in PDF form are from Oct. 20, 1995 (URL https://www.math.bas.bg/bantchev/place/oberon/oakwood-guidelines.pdf downloaded on 2021-06-06) and indicate that two parameters are expected. Next I looked at how the Oxford compiler implemented Out and saw that it was a very thin wrapper on Files using STDOUT instead of a user supplied file handle. That makes complete sense to me, at least from a POSIX point of view. But Files WriteReal() in Oakwood on;y has one parameter. Does anyone know the history behind why Oakwood as to why Out.Real() and Files.WriteReal() would be different? I noticed in Native Oberon 2.3.6 that Files.WriteReal() does have two parameters. Just curious to know the story behind the evolution from 1995 to Native Oberon 2.3.6. All the best, Robert Doiel

6 8

Re (2): ETH Oberon, source NetSystem.SetUser*.
by peter＠easthope.ca 06 Jun '21

06 Jun '21

From: Michael Schierl <schierlm(a)gmx.de> Date: Sun, 6 Jun 2021 21:01:19 +0200 > In your except you missed a WHILE loop. There is a comment above the > SetUser procedure declaration that describes why there may be multiple > URLs in one call to SetUser. entered[0] is only 0X for the first of > them. Thanks Michael. That allows for a command such as this. The first three authentications share one password. NetSystem.Setuser ssh:peter:sshpassword@localhost ssh:peter@otherhost ssh:pete@anotherhost pop:peter:mailpassword@localhost pop:peter:otherpassword@mailserver.isp.net ~ Another interesting point: the SetUser in ETH Oberon has two instances of RETURN; neither according to Oberon-07 syntax. In UnixAos, SetUser is revised to remove both RETURNs. Years ago, when user ids in the pattern <user>@<smarthost> came into use, misinterpretation of "@" by percent encoding was considered. NetSystem.Setuser pop:<user>@<maildomain>@<smarthost> ~ would be represented NetSystem.Setuser pop:<user>%40<maildomain>@<smarthost> ~ . Bad idea. Unnecessary complication. Better to recognize that the host name begins at the last "@". Thanks, ... P. -- tel: +1 604 670 0140 Bcc: peter at easthope. ca

1 0

ETH Oberon, source NetSystem.SetUser*.
by peter＠easthope.ca 06 Jun '21

06 Jun '21

== Reference == https://en.wikibooks.org/wiki/Oberon/ETH_Oberon/NetSystem.Mod == Observation == Excerpt from the body of SetUser. BEGIN Texts.OpenReader(R, Oberon.Par.text, Oberon.Par.pos); ok := TRUE; entered[0] := 0X; verbose := FALSE; ... IF host[0] # 0X THEN IF (usr[0] = 0X) OR ((pwd[0] = 0X) & (entered[0] = 0X)) THEN ... IF pwd[0] = 0X THEN (* no pwd specified *) IF entered[0] = 0X THEN (* prompt first time *) InputStr("Enter password: ", FALSE, entered); == Question == Isn't (entered[0] = 0X) inevitable when the IF statements are executed? The two ests, entered[0] = 0X, are redundant? Thanks, ... P. -- tel: +1 604 670 0140 Bcc: peter at easthope. ca

3 2