June 2017 - Oberon - mailman3.inf.ethz.ch

Oberon Wikibook.
by peter＠easthope.ca 22 Jun '17

22 Jun '17

Thanks to the anonymous contributor who made several edits in the Wikibook last Sunday, 2017-06-18. I've incorporated those contributions and the Origin section into a new introduction. If there is no objection, will remove most the old introduction soon. By the way, the Wikibook is public. Anyone is welcome to edit it and anyone can revert an edit. Anonymous edits are welcome but edits from an account are better in some ways. Regards, ... Lyall E. -- 123456789 123456789 123456789 123456789 123456789 123456789 123456789 Tel: +1 360 639 0202 Pender Is.: +1 250 629 3757 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

3 2

Re: Acronyms and Versions
by peter＠easthope.ca 20 Jun '17

20 Jun '17

>From jwr at robrts.net Thu Jan 28 14:57:15 2016 > ... summary of acronyms and version names ... John, thanks for the extensive list. I've added some to the glossary. https://en.wikipedia.org/wiki/Oberon_%28operating_system%29#Glossary Can extend further when time is available. Also more hyperlinks. > Is ETH-Linux-Oberon the same as LEO or LNO? (Probably it is LEO.) > Is Linux-ETH-Oberon the same as LEO? Same as ETH-Linux-Oberon? > See https://lists.inf.ethz.ch/pipermail/oberon/2015/007996.html > and https://lists.inf.ethz.ch/pipermail/oberon/2008/005410.html I don't know. As usual, depending upon Chris, Joerg and other knowledgeable people to identify errors and suggest corrections. Regards, ... Lyall E. -- 123456789 123456789 123456789 123456789 123456789 123456789 123456789 Tel: +1 360 639 0202 Pender Is.: +1 250 629 3757 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

2 1

Re (2): Re (2): LONGINT, RETURN & etc.
by peter＠easthope.ca 18 Jun '17

18 Jun '17

From: Chris Burrows <chris(a)cfbsoftware.com> Date: Sat, 17 Jun 2017 14:52:57 +0930 > The CONST keyword protects the parameter from being changed within the > body of the procedure and any procedures that it calls i.e. the value > is read-only. Makes perfect sense, thanks. The meaning should be obvious but I missed it. From: Chris Burrows <chris(a)cfbsoftware.com> Date: Sun, 18 Jun 2017 08:51:11 +0930 > NIL is a reserved word representing a null POINTER, " " is a blank CHAR. > They do not denote that 'expression' is optional. > > RETURN NIL END is valid > > RETURN " " END is valid > > RETURN END is not valid Oops. Thanks, ... Lyall E. -- 123456789 123456789 123456789 123456789 123456789 123456789 123456789 Tel: +1 360 639 0202 Pender Is.: +1 250 629 3757 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

1 0

Re (2): LONGINT, RETURN & etc.
by peter＠easthope.ca 17 Jun '17

17 Jun '17

From: Chris Burrows <chris(a)cfbsoftware.com Date: Sat, 17 Jun 2017 13:40:15 +0930 > ... the Oberon 'LOOP' statement with multiple 'EXIT's was abolished > at the same time. An even simpler possibility is to eliminate FOR, REPEAT and WHILE and retain LOOP with the restriction of exactly one exit, at the beginning or at the end. Analogous to the RETURN revision. Some time in the last year or two (I can't find a citation at present) Joerg commented, > LOOP R.I.P. With this hypothetical revision, the cases of the content of the LOOP being executed zero times or once remain immediately clear. Dropping FOR, REPEAT and WHILE in favour of LOOP would impose more work to bring old sources to new syntax. If there is a stronger argument against it, I haven't caught it. LOOP or not, Oberon is a superb language. Regards, ... Lyall E. -- 123456789 123456789 123456789 123456789 123456789 123456789 123456789 Tel: +1 360 639 0202 Pender Is.: +1 250 629 3757 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

2 1

LONGINT, RETURN & etc.
by peter＠easthope.ca 17 Jun '17

17 Jun '17

In pages 5 and 6 of _Project Oberon_, Revised Edition 2013, Prof. Wirth states, "1. The data types LONGINT, ... have been discarded, ... ... 3. The RETURN statement has been discarded ..." While in Peter De Wachter's emulator, "MODULE Edit; (*JG 2.11.90 / NW 28.11.2015*) ... PROCEDURE Max(i, j: LONGINT): LONGINT; VAR m: LONGINT; BEGIN IF i >= j THEN m := i ELSE m := j END ; RETURN m END Max;" Wirth's revisions of syntax not yet implemented? Thanks, ... Lyall E. -- 123456789 123456789 123456789 123456789 123456789 123456789 123456789 Tel: +1 360 639 0202 Pender Is.: +1 250 629 3757 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

5 9

Re: [Oberon] Help: Display Init Strings for Native Oberon
by peter＠easthope.ca 17 Jun '17

17 Jun '17

From: Treutwein Bernhard <Bernhard.Treutwein(a)Verwaltung.Uni-Muenchen.DE> Date: Fri, 16 Jun 2017 13:42:02 +0000 > Does anybody remember how to create display Init Strings for Native > Oberon? https://en.wikibooks.org/wiki/Oberon/EO#FAQ If you can add to that--good. Regards, ... Lyall E. -- 123456789 123456789 123456789 123456789 123456789 123456789 123456789 Tel: +1 360 639 0202 Pender Is.: +1 250 629 3757 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

1 0

CONST parameters
by Chris Burrows 17 Jun '17

17 Jun '17

> -----Original Message----- > From: Oberon [mailto:oberon-bounces@lists.inf.ethz.ch] On Behalf Of > peter(a)easthope.ca > Sent: Wednesday, 14 June 2017 2:25 PM > To: oberon(a)lists.inf.ethz.ch > Subject: Re: [Oberon] Text buffer analogue of Strings.Prefix. > > From: Peter Matthias <PeterMatthias(a)web.de> > Date: Tue, 13 Jun 2017 21:55:53 +0200 > > Which version of PC Native O. has such CONST declaration? > > Will check the alpha release again tomorrow and might need to correct > my statement. > > LinuxAos (rev.6661) certainly has it. Still puzzled that a parameter > is constant. > Maybe you are thinking that CONST parameters are like CONST declarations. i.e. that the parameter always has the same value? That is not so. Constant parameters allow any different value to be passed to the procedure via the parameter whenever the procedure is called. The CONST keyword protects the parameter from being changed within the body of the procedure and any procedures that it calls i.e. the value is read-only. This feature helps programmers to improve the security, reliability, testability and maintainability of their programs. In other languages (e.g. Ada, Component Pascal etc.) they are declared as 'IN' parameters (which might be less confusing). A long-standing problem in Pascal, Modula-2, Oberon and Oberon-2 is that if you wanted to protect an ARRAY parameter from being accidentally modified by a procedure you would omit the VAR keyword when declaring the parameter. The system would then make a copy of the array every time you called the procedure. If the array is large the time taken to make the copy could be a significant overhead. Hence, programmers have got into the bad habit of declaring these arrays as VAR just to avoid performance problems. The tradeoff was the increased probability of accidental changes to the array occurring, requiring more extensive testing to be done. In Oberon-07 ARRAYs that not declared as VAR parameters are enforced as read-only parameters at compile time. No copy of the array is required. This is the default behaviour - no CONST or IN keyword is required. This means that Oberon programmers no longer have to take the risk of declaring large ARRAYs as VAR parameters just to avoid performance problems. Regards, Chris Burrows CFB Software http://www.astrobe.com Oberon for ARM Cortex-M3, M4 and M7 microcontrollers

1 0

Re: Wikibook _versus_ svn or git.
by peter＠easthope.ca 16 Jun '17

16 Jun '17

>From joerg.straube at iaeth.ch Wed Apr 20 18:50:55 2016 > Perhaps even with some approvers to review the contributions to > guarantee a minimal level of quality... As described here, anyone can quickly see a list of all changes to a book within a recent interval. https://en.wikibooks.org/wiki/Using_Wikibooks/How_To_Edit_A_Wikibook#Watchi… Recent changes to the Oberon book here. https://en.wikibooks.org/wiki/Special:RecentChangesLinked/Oberon The world now has over 7 x 10^9 people. With any luck, a few have knowledge, interest and motivation to use the facility and review as you mention. > In my point of view, WikiBook doesn't seem to be the best choice > for this task. A wiki tailored to source code is hypothetically possible. I don't know of one reliable enough to use. We work with what we have. If you can suggest something better than this wikibook, I'm interested. Regards, ... Lyall E. -- 123456789 123456789 123456789 123456789 123456789 123456789 123456789 Tel: +1 360 639 0202 Pender Is.: +1 250 629 3757 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

1 0

Oberon0 - Compiler Construction
by Tomas Kral 15 Jun '17

15 Jun '17

Hi, There is a new compiler update by prof. Wirth. I need to read the books and comments, meanwhile can anyone explain? Many thanks -- Tomas Kral <thomas.kral(a)email.cz>

2 1

Re (2): Text buffer analogue of Strings.Prefix.
by peter＠easthope.ca 15 Jun '17

15 Jun '17

From: Peter Matthias <PeterMatthias(a)web.de> Date: Tue, 13 Jun 2017 21:55:53 +0200 > Which version of PC Native O. has such CONST declaration? Correction: no version of PC N.O. that I have. I should have cited Oberon in LinuxAos (rev.6661). Thanks for catching the error, ... Lyall E. -- 123456789 123456789 123456789 123456789 123456789 123456789 123456789 Tel: +1 360 639 0202 Pender Is.: +1 250 629 3757 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

1 0