March 2018 - Oberon - mailman3.inf.ethz.ch

Re (2): Documentation.
by peter＠easthope.ca 08 Mar '18

08 Mar '18

> André Fischer's tutorial is linked both, in Wikipedia ... and > in the second paragraph of "Getting Started" of the Wikibook ... OK, good! There is a peculiarity that André's "Contents" are * Objective * Welcome to the Oberon Tutorial System! * Notation used for describing the syntax of commands * Bibliography * The tutorial package * What's next? whereas the actual contents are listed in http://www.ethoberon.ethz.ch/ethoberon/tutorial/index.html under the heading Welcome to the Oberon Tutorial System! * Compiler and Builder User's Guide * Browser and Watson User's Guide * Columbus User's Guide ... * Using Gadgets ... Consequently a page such as "Using Gadgets" is easily missed. =8~( I wonder whether he might rectify that. Would a contents summary in wikibook ... #Getting_Started be acceptable? > And even more information can be found in the Oberon Companion. Also good. Appears that André created his pages after the Companion existed. Therefore he would have incorported some improvements. An obvious bigger concern is that at some point André will be fully retired and no longer revising his work. Then, if the pages cease to be available from ETH, access will depend upon Wayback and similar archives. At least until copyright expires. For Switzerland 70 pma. See https://en.wikipedia.org/wiki/Wikipedia:Non-U.S._copyrights#Dates_of_restor… Subsequent publication will depend upon a copy surviving the copyright period. A dicey situation. We should hope the mood of Lady Luck is good during those decades. Not a good way to manage critical information. 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

Documentation; was "Viewer marking."
by peter＠easthope.ca 07 Mar '18

07 Mar '18

From: SvZyZw== <joerg.straube(a)iaeth.ch> Date: Tue, 6 Mar 2018 18:51:28 +0100 > Did you read this intro into the Gadget system: > http://www.ethoberon.ethz.ch/ethoberon/tutorial/GadgetsIntro.html Finding this tutorial isn't a certainty. The left margin of http://www.ethoberon.ethz.ch contains no direct link. There is the link "Documentation" anchored at http://www.ethoberon.ethz.ch/compiler/index.html#report . That page addresses the language and doesn't lead to http://www.ethoberon.ethz.ch/ethoberon/tutorial/ . If there is a link elsewhere, I'm afraid most readers must depend on Luck to find it. No need to say, Luck is a poor basis for engineering. =8~) http://www.ethoberon.ethz.ch/ should have a link directly to http://www.ethoberon.ethz.ch/ethoberon/tutorial/ . Further confusion. The 3rd paragraph in http://www.ethoberon.ethz.ch/ethoberon/tutorial begins "The document you are viewing right now, is a hypertext document which you may have opened under the control of the Oberon System itself (with Desktops.OpenDoc index.html) ..." Where is index.html? Linux A2 Generic, Jan 31 2018 has no such file. If available in S3, it's in an optional archive I didn't install. Wouldn't this be better? Desktops.OpenDoc http://www.ethoberon.ethz.ch/ethoberon/tutorial/ No offense to Andre Fischer. His documentation is a wonderful help. Unfortunately a snag in access can hobble it slightly. In a wiki, a link could be added easily. 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

ORD and scalar types
by August Karlstrom 07 Mar '18

07 Mar '18

According to the Oberon-07 language report, the procedure ORD accepts parameters of type CHAR, BOOLEAN and SET. In Project Oberon 2013, however, ORD accepts parameters of any scalar type; even NIL. The procedure Texts.WriteReal, for instance, uses ORD(x) for the real number x. Is this to be considered a language extension? -- August

2 1

Android
by jwr＠robrts.net 07 Mar '18

07 Mar '18

Peter Matthias wrote on 28 Jan: Android uses Linux Kernel. OLR/ARM Linux runs, but is not compatible with Android runtime system. E.g. Android updates the framebuffer frequently so that OLR output on the framebuffer is overwritten. I don't know how Android works with a framebuffer, but this reminds me of requirements I once learned for writing software for the Macintosh. After the MacOS overwrote portions of an application's window with a dialog box or menu, it sent an event to the app with the window handle and coordinates of the rectangle the app needed to redraw. It was the app's responsibility to redraw what used to be there. Does Android send a message to the running application(s) notifying that a portion of the screen needs to be redrawn? Does OLR on Android need to write to a (persistent) off-screen bitmap, and then (re)copy it to the framebuffer when notified? It seems odd to me that use of the framebuffer on Android should differ from use of the framebuffer on "normal" Linux; why was this not a problem before? Lyall Easthope wrote on 26 Jan: Oberon or A2 on Android should be possible. Appealing, but is it worth the trouble? Pessimism not intended. On the one hand, I don't see Oberon being particularly functional on a touchscreen smartphone or tablet, unless you also use an external mouse and keyboard. And in that case, why not just use a netbook or small laptop instead? I don't see a particular useful purpose (use case/application) for Oberon on a phone or tablet, that is materially different than on a small laptop. But just because my imagination is limited doesn't mean Oberon on a phone or tablet isn't particularly useful to some people for some purpose. On the other hand, it's easy enough to carry along a mouse and tiny keyboard, and many people carry their phone and/or tablet everywhere. So I think it IS useful to figure out how to run OLR on Android effectively, as an enabler for anyone who does have something useful to do with Oberon on such a device. I specifically say OLR, because especially for smartphones, I don't think it makes sense to run Oberon or A2 as the host OS: you would want to preserve phone and email and other native capabilities of the smartphone, and run Oberon essentially as a guest application which you can switch to (and then off of to other apps) as desired. -- John Roberts

2 1

Bug in multiplier?
by Hellwig Geisse 07 Mar '18

07 Mar '18

Hi all, I'm trying to understand the RISC5 implementation and stumbled over a corner case in unsigned multiplication. What exactly is "Multiplier.v" supposed to compute in case u=1 in the instruction (i.e., u=0 in Multiplier.v, because of ".u(~u)" in the instantiation of the module)? In NW's "The Design of a RISC Architecture and its implementation with an FPGA" (revision 1.9.2015) I read "..., and the MUL instruction with u set to 1 considers the operands as unsigned numbers, yielding a 64-bit unsigned product". So (1 << 31)*(1 << 31) should give a result of (1 << 62). But when I feed the multiplier's x and y inputs both with 0x80000000, the result is 0xC000000000000000 instead of 0x4000000000000000, at least in my simulation of the module. Am I missing something? Hellwig

2 1

Project Oberon / MIPS Linux
by Peter Matthias 06 Mar '18

06 Mar '18

is now available at http://oberon.wikidot.com . It has a native MIPS32 compiler. Regards, Peter

1 0

Re: [Oberon] Viewer marking.
by peter＠easthope.ca 06 Mar '18

06 Mar '18

From: Andreas Pirklbauer <andreas_pirklbauer(a)yahoo.com> Date: Tue, 6 Mar 2018 20:59:24 +0100 > Perhaps the reason why Sort traps is that Oberon.MarkedViewer > returns the filler viewer, whose fields Viewers.FillerViewer.dsc > and Viewers.FillerViewer.next are both always set to NIL. Thanks. Will think about that possibility and try to make the test module behave well before adding more to Sort. ... 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

Viewer marking.
by Andreas Pirklbauer 06 Mar '18

06 Mar '18

> Sort should terminate silently but traps or sorts an unmarked viewer. Perhaps the reason why Sort traps is that Oberon.MarkedViewer returns the filler viewer, whose fields Viewers.FillerViewer.dsc and Viewers.FillerViewer.next are both always set to NIL.

1 0

Re (2): Viewer marking.
by peter＠easthope.ca 06 Mar '18

06 Mar '18

From: Andreas Pirklbauer <andreas_pirklbauer(a)yahoo.com> Date: Tue, 6 Mar 2018 18:21:27 +0100 > If you can modify module Oberon, just add the following procedure > ... From: SvZyZw== <joerg.straube(a)iaeth.ch> Date: Tue, 6 Mar 2018 18:51:28 +0100 > Did you read this intro into the Gadget system: > http://www.ethoberon.ethz.ch/ethoberon/tutorial/GadgetsIntro.html Thanks Andreas and Joerg. Will read the documentation before fumbling more. For interest, when nothing is marked to sort, the behaviour of the extant Sort is wrong. It should terminate silently but traps or sorts an unmarked viewer. Thanks again, ... 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

Viewer marking.
by peter＠easthope.ca 06 Mar '18

06 Mar '18

Suppose this test is compiled in A2. If the context is removed it can compile in S3 as well. MODULE MarkingTest IN Oberon; IMPORT Documents, Oberon, Texts; VAR W: Texts.Writer; PROCEDURE T*; (* No parameters. *) BEGIN IF Oberon.MarkedViewer() # NIL THEN Texts.WriteString(W, "A Text viewer is marked."); Texts.WriteLn(W); Texts.Append(Oberon.Log, W.buf) ELSIF Documents.MarkedDoc() # NIL THEN Texts.WriteString(W, "A Document viewer is marked."); Texts.WriteLn(W); Texts.Append(Oberon.Log, W.buf) ELSE Texts.WriteString(W, "No viewer is marked."); Texts.WriteLn(W); Texts.Append(Oberon.Log, W.buf) END END T; BEGIN Texts.OpenWriter(W) END MarkingTest. MarkingTest.T always reports "A Text viewer is marked." Pressing F2 to remove the marker has no effect. How is viewer marking detected correctly? 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

2 1