January 2003 - Oberon - mailman3.inf.ethz.ch

Re: [Oberon] Programming Editor
by mcintosh＠vima.austin.tx.us 23 Jan '03

23 Jan '03

> John M. Drake wrote > Version 2.3 doesn't automatically put them in, but > you can insert them by selecting the errors and > clicking Builder.MarkErrors from the "Compiler" > menu in the Programming.Tool. Thanks. This lead me to a good environment. Under the BlackBox May 22, 2002 release: Programming.Tool is not present The Desktops.OpenDoc ^ viewer doesn't appear to edit its popup menus. The Edit.Open window does not appear, well, to have extensive functionality. This appears to be the window opened with the MM+MR command from System.Log. (I use System.Log extensively as a scratch pad.) Builder.Tool is present. Working in the Desktops.OpenDoc ^ viewer and executing Builder.Compile * from the System.Log or the Builder.Tool , the overall "look and feel" is familiar, ... including many active error elements :-( . Aubrey McIntosh <mcintosh(a)vima.austin.tx.us> Assistant Professor of Chemistry University of Minnesota, Morris 600 E. 4th St. Morris, MN 56267-2134

3 2

Defining code for MM + MR
by mcintosh＠vima.austin.tx.us 23 Jan '03

23 Jan '03

In V4, the "Viewers" each have "Handlers" and it is the procedure Handle where the MM+MR interclick is interpreted. I looked at the S3 (BlueBottle 5/22/02) procedure TextFrames.Handle, which handles the type Oberon.InputMsg with the procedure TextFrames.Edit. In that procedure, the keys variable handles the various mouse clicks. The MM+MR command, for example, is located deep under IF X < F.X + Min(F.left, barW) THEN ... ELSE ... ELSIF 1 IN Keys THEN ... IF (pos >= 0) & ~(2 IN keysum) THEN CallCmd(OpenCmd, F, pos, FALSE) END One thing that this means is that each type of Viewer, and indeed each frame in a viewer, has a custom action for the MM+MR command. In principle, I could have one text open with two viewers and have two different actions for the MM+MR key sequence. (Hopefully, they would both be conceptually some type of "Open" action. Aubrey McIntosh <mcintosh(a)vima.austin.tx.us> Assistant Professor of Chemistry University of Minnesota, Morris 600 E. 4th St. Morris, MN 56267-2134

1 0

Trap
by mcintosh＠vima.austin.tx.us 23 Jan '03

23 Jan '03

My direct email ( mcintosa(a)mrs.umn.edu ) While debugging version 2 of the SMTP receiver, codenamed "MailBottle" I received a trap in the log viewer. 1. I don't know how to copy text between Bluebottle viewers. 2. The highlights appear to be: AosFS.Volume.AllocBlock pc=584 State AosFS: AosDiskFS.AlloceSector pc=14648 AosDiskFS.WriteBuf pc=17623 AosDiskFS.File.Writebytes pc=13687 AosFS.Rider0.Send pc=133 Aos.Write.Char pc=382 Aos.Write.Ln pc=1363 SMTPP.SMTPAgent.ToFile pc=2815 SMTPP.SMTPAgent.DATA pc=3152 SMTPP.SMTPAgent.MAIL pc=3633 SMTPP.SMTPAgent.reset pc=3951 SMTPP.SMTPAgent.$$ pc=174 3. The input sent via telnet to port 27 was helo butterfly.net mail from: gandalf@tower rcpt to: frodo@shire rcpt to: bilbo@shire data subject: Grew and Yhite The situation is grim. <Trap happened here. Boy, Life imitates art. Following the trap, the mouse and window manager was active. I could pan to the second screen, raise windows by clicking on the border. Anything that would access the disk appeared to fail. After this is sent, I will place the source on the net at http://cda.mrs.umn.edu/~mcintosa/Bluebottle (this may have capitalization spelling errors ... I'm going to send this now in case I'm just ahead of a hardware crash.) Aubrey McIntosh <mcintosh(a)vima.austin.tx.us> Assistant Professor of Chemistry University of Minnesota, Morris 600 E. 4th St. Morris, MN 56267-2134

1 0

(Re: [Oberon] Programming Editor) & etc.
by eas-lab＠absamail.co.za 23 Jan '03

23 Jan '03

Aubrey McIntosh wrote: > Can some kind soul provide a Tool (or name some existing ones) that > have a few relevant commands for the { Edit Compile } Store cycle? I got no reply to the same questions. Eth 'insiders' are obviously hiding the good stuff from us. I recently revisited 1980's DOS Turbo-Pascal Edit/Compile/Store facilities - much more productive than what I have to use with S3. That's why I don't do oberon programming. S3 is superb for general text handling; why such weak programming tools? I can't believe that eth 'insiders' use what we are offered ? peter_easthope(a)gulfnet.sd64.bc.ca wrote: > In PC Native 24.08.2002 my Oberon.Text:System: > InitCommands contains "ET.ReplaceSystemEditor" > yet MM+MR on a file name opens Edit. Do I have > something configured incorrectly; is this a > repaired bug or a pending repair? For me MM+MR opens the editor type: Edit/ET/Sctript/Desktops; according to the type which 'contains' the file name. OK ! But not for ET. Seems to be an inconsistency. I'm using ver. beta?alpha 2001. I'd be interested to know the mechanism whereby MM+MR DOES opens an editor of the same 'type' as where the MM+MR was executed. Where is the system wide mouse handling done ? An early tutorial showed how MM <Module>.<Proc> caused the M.P to be executed. I guess the MM+MR handling is related to this ? > Friday, 2002-07-19 Chris Glur suggested, > cg> You shouldn't try writing video drivers > before doing simpler experiments. Use > successive refinement. After you've done > RS232 and UBS 'experiments' you will better > cope with a video driver. > > A reasonable suggestion, although I have no > USB hardware yet. Nor I. Nor do I know what UBS is; except that as a 'few-pin' interface replacing the serial port, it must be much simpler than a video driver. The correct route to start from zero experience for writing hardware drivers is: * get hands-on-feedback as soon as possible, by exercising the simplest case. The PC-speaker is ideal. Unfortunately our http://www.oberon.ethz.ch/speaker.html is a dud, because it introduces much irrelevant complications. * the next hardware that can be played with, without crashing the machine is the parallel port. Again, results must be observable. A speaker (isolated via a capacitor) can give observable output. Parallel port can also exercise inputs. After this it's time to be able to see voltage changes caused on outputs: voltmeter, oscilloscope ....etc. * Even if you use n-o's : "s := s - {0, 1}" syntax, you have to think in terms of bits, bytes, registers - not sets. I.e use the terminology & paradigm of the hardware device/card suppliers. > I spent a good part of the Christmas break > studying RFCs and Edgar's PPP modules. It > was a good learning exercise and this system > can now connect to the GulfNet Vicom router > once again. Is it difficult to condense your newly aquired knowledge and explain what the problem was and what are the essential steps and material to cover, to understand the ppp process ? Eg. a sequence of: "Doc: xyz explains xxx, which is essential." Even knowing exactly what to read is a big part of the knowledge of having researched something. -- Chris Glur.

2 1

(Re: [Oberon] Programming Editor) & etc.
by peter_easthope＠gulfnet.sd64.bc.ca 23 Jan '03

23 Jan '03

Sorry if this mixture of topics is incongruous. Wednesday, 2003-01-22 Aubrey McIntosh enquired, ami> In the S3 environment, the Edit viewer ... Aubrey, what about ET ... and is Pet still in use? In PC Native 24.08.2002 my Oberon.Text:System: InitCommands contains "ET.ReplaceSystemEditor" yet MM+MR on a file name opens Edit. Do I have something configured incorrectly; is this a repaired bug or a pending repair? Friday, 2002-07-19 Chris Glur suggested, cg> You shouldn't try writing video drivers before doing simpler experiments. Use successive refinement. After you've done RS232 and UBS 'experiments' you will better cope with a video driver. A reasonable suggestion, although I have no USB hardware yet. I spent a good part of the Christmas break studying RFCs and Edgar's PPP modules. It was a good learning exercise and this system can now connect to the GulfNet Vicom router once again. Thursday, 1999-05-06 Chris Glur remarked, cg> The real limitation is getting documentation for the ever increasing mass of software. Thursday, 2002-09-05 Pat commented, p> For a programmer the source code is the documentation. There is truth in both of these statements. I appreciate Edgar's nicely written sources and wouldn't want to be without the RFCs as well. ... Peter E.

1 0

Programming Editor
by mcintosh＠vima.austin.tx.us 22 Jan '03

22 Jan '03

In V4 Oberon, the Edit viewer, ErrorElems, and the Folds compiler interact to place active error descriptors in the text in the event of a compiler time error. In the S3 environment, the Edit viewer does not seem to have a replace command, and I haven't found the equivalent of ErrorElems. Can some kind soul provide a Tool (or name some existing ones) that have a few relevant commands for the { Edit Compile } Store cycle? Aubrey McIntosh <mcintosh(a)vima.austin.tx.us> Assistant Professor of Chemistry University of Minnesota, Morris 600 E. 4th St. Morris, MN 56267-2134

2 1

Bluebottle / Native Oberon
by Capaul Giachen F (KADA 12) 22 Jan '03

22 Jan '03

Hi all, what is the difference between Bluebottle & Native Oberon? Would you recommend having a play with both, or is one of them targeted for specific needs? Regards, Flurin

1 0

Hardware driver construction 'capacity building' [3D Acc.]
by eas-lab＠absamail.co.za 22 Jan '03

22 Jan '03

> > > As for the specs of graphics cards not > > > being available, that's only partially true. > > > Thanks to Linux the source code for many > > > advanced graphics cards such as the NVidia > > > chips are freely available. Now the problem > > > is you have to wade through thousands of lines > > > of ugly C code just to figure out what could > > > have likely been documented in a few pages of > > > plane text. I've downloaded the NVidia code > > > myself, but I haven't taken the time > > > yet to seriously look at it. Is anyone else > > > interested? If BlueBottle had drivers for > > > NVidia and ATI chips it would be well > > > positioned in the 3D graphics arena. > > > > Thomas Frey wrote: > > Yes. Currently there are almost only these two > > cards around. But writing the driver based > > on "really ugly" C code can be > > very arduous. I have no time for this in > > the foreseeable future. > > John M. Drake wrote: > I realize that it would be a difficult task and > I wasn't suggesting you take it on by yourself > or even at all. What I meant when I said > "Is anyone else interested" is "is anyone else > interesting in joining in the effort as an > open source project." It could be hosted at > SourceForge. I would be willing to spearhead > such a project if there's enough interest. Well yes, it's uncomfortable translating C code; but I'm still waiting to hear if it's only me who has pain in translating between: Clear Bits 0,1 of s s -> Port(61H) and this monster: s := s - {0, 1}; SYSTEM.PORTOUT (61H, SYSTEM.VAL(CHAR, s)) This is terribly unintuitive ! Also, SYSTEM.PORTIN (61H, SYSTEM.VAL(CHAR, s)); seems absurd, since by conventional (and also n-o) syntax, SYSTEM.VAL(CHAR, s) must execute before: SYSTEM.PORTIN (61H, <byte>). I.e. the 'value' must be converted to 'CHAR type', *before* it is obtained/available ?! Is there a gentle tutorial to introduce this strange syntax. A colaborative project to 'write hardware drivers in general' rather than just 3D Acc. is needed. Chris Glur.

2 1

(no subject)
by Michael Grzeschik 21 Jan '03

21 Jan '03

Hy, i have ret on the site http://www.oberon.ethz.ch/projects.html - that a guy named christian plattner has writen an driver for the midi-3 interface of steinberg. So i have got a question, whas this driver for linux. And if yes where can i get this driver?? Greetz Michael G. -- +++ GMX - Mail, Messaging & more http://www.gmx.net +++ NEU: Mit GMX ins Internet. Rund um die Uhr für 1 ct/ Min. surfen!

2 1

Hardware 3D Acc.
by Brandon Elam Barker 21 Jan '03

21 Jan '03

Which (if any) cards/chipsets are supported for hardware 3D acceleration in BlueBottle? What 3D APIs are implemented? Also, how portable should the BlueBottle codebase be to other nonintel platforms? Thanks, Brandon Barker

3 4