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

Wanted: Oberon Syntax in TextMate, Sublime Text or Atom Grammar
by Treutwein Bernhard 01 Jan '22

01 Jan '22

Does somebody have grammars for Oberon / Oberon-2 / Component Pascal / Active Oberon as grammar for TextMate / Sublime Text / Atom Background: Github wrongly classifies Oberon / Oberon-2 / Component Pascal / Active Oberon as Modula-2. I opened quite some tiome ago an issue in the github/linguist project (https://github.com/github/linguist/issues/3888) and they don't accept EBNF grammars. -- Bernhard Treutwein

6 9

PO2013 - Real time measurement
by Tomas Kral 23 Apr '19

23 Apr '19

Hi, I wish to do some real time input from basic sensors. This discipline is well established with so called home automation, developed through Arduino project, see MySensors lib, etc. [a] Arduino can do A/D conversions for Oberon while it can talk over WiFi, the same WiFi, PO2013 was designed with, more sensors can be linked in a HUB. [b] PIC processors can also do A/D conversion. PIC can possibly interconnect through GPIO, while I cannot see an easy way to add more devices to the hub (* multiplexing GPIO input???*). It also has some leverage in NW's PICL project. Both options require some more studying and understanding on my part. Generally what is your experience, recommendations to pursue [a], [b] for PO2013? Many thanks -- Tomas Kral <thomas.kral(a)email.cz>

9 52

FPGA - DIY Dialogs
by Tomas Kral 06 Feb '19

06 Feb '19

Hi, Trying to draw a selection rectangle, aka `marching ants'. I do this, it makes selection east-rightwards only. I am surprised I need to call `DrawBox' in three places, otherwise it trails visual marks. I wish to keep the code minimal. How to make it in all directions? PROCEDURE DrawBox(x, y, w, h: INTEGER); (*selection rectangle*) BEGIN Display.ReplPattern(8, Display1.grey2, x, y, w, 1, Display.invert); Display.ReplPattern(8, Display1.grey2, x, y+h-1, w, 1, Display.invert); Display.ReplPattern(8, Display1.grey2, x, y, 1, h, Display.invert); Display.ReplPattern(8, Display1.grey2, x+w-1, y, 1, h, Display.invert) END DrawBox; PROCEDURE TrackSelection(M: Oberon.InputMsg); (*eastwards*) VAR x, y: INTEGER; keys: SET; BEGIN x := M.X; y := M.Y; REPEAT DrawBox(M.X, M.Y, ABS(M.X-x), ABS(M.Y-y)); Input.Mouse(keys, x, y); Oberon.DrawMouse(Crosshair, x, y); DrawBox(M.X, M.Y, ABS(M.X-x), ABS(M.Y-y)) UNTIL keys = {}; DrawBox(M.X, M.Y, ABS(M.X-x), ABS(M.Y-y)) END TrackSelection; -- Tomas Kral <thomas.kral(a)email.cz>

5 23

FPGA - Simple Graph Fill
by Tomas Kral 14 Jan '19

14 Jan '19

> In short, from a given (x, y), evaluate F(x+1, y-1/2) which dictates > wether to go East or South-East for the next pixel. Hi, I have this simple example, drawing a circle, and attempting to fill it, starting off the centre. Fill goes well east and north, and fills exactly one quadrant. Uncommenting west south then the recursive does not stop. GraphTest.DoCircle 11 300 200 4 ~ colour x y radius GraphTest.DoFill 0 8 300 200 ~ target replacement x y col x y 0 300 200 east start 0 301 200 0 302 200 0 303 200 11 304 200 east stop col # target col 0 303 201 north start 11 304 201 east resume 11 303 202 north resume 0 302 201 0 303 201 11 304 201 11 303 202 .... never terminates CONST ppw = 8; dpt = 4; bpl = dpt*128; (*pixels per word, depth, bytes per line*) PROCEDURE Fill*(tgc, rpc, x, y: INTEGER); VAR c, a: INTEGER; (*tgc, rpc - target replacement colour*) BEGIN a := D.Base + (x DIV ppw)*4 + y*bpl; S.GET(a, c); c := ASR(c, -(x MOD ppw * 4)) MOD 10H; O.Int(c, 4); O.Int(x, 4); O.Int(y, 4); O.Ln; IF c = tgc THEN D.Dot(rpc, x, y, D.replace); Fill(tgc, rpc, x+1, y); (*east*) Fill(tgc, rpc, x, y+1); (*north*) (*Fill(tgc, rpc, x-1, y); (*west*) Fill(tgc, rpc, x, y-1); (*south*)*) END END Fill; -- Tomas Kral <thomas.kral(a)email.cz>

6 11

SystemV- Heap + Module Space
by Andreas Pirklbauer 01 Jan '19

01 Jan '19

> Can you please just give me a small hint (*comment*) in the code for [a],[b], I will try then on my own. It depends on what heuristics you want to use to find the gap. For example, if you just want to find the very first block "at the lowest possible end that fits", simply replace the lines (in Modules.Load) mod := root; WHILE (mod # NIL) & ~((mod.name[0] = 0X) & (mod.size >= size)) DO mod := mod.next END ; with: m := root; mod := NIL; WHILE m # NIL DO IF (m.name[0] = 0X) & (m.size >= size) THEN mod := m END ; m := m.next END ; If you really feel like it, you can also strengthen the condition in the IF statement to find a gap that meets additional criteria, i.e. the “best possible fit”. You may also think about splitting the gap into a “now re-used again” one and a “still not used” portion, and about merging adjacent gaps (to keep fragmentation small). A lot of ink has been spilled over questions like that --> you can make any number of student exercises out of it... But I question the usefulness of any of this in a system like Oberon.

3 4

Re (2): Re (2): Location and name for percent encoding.
by peter＠easthope.ca 31 Dec '18

31 Dec '18

From: Dieter <d.gloetzel(a)web.de> Date: Mon, 31 Dec 2018 10:43:42 +0100 > Could somebody be so kind and explain to me, what percent encoding/decoding means? https://de.wikipedia.org/wiki/URL-Encoding https://en.wikipedia.org/wiki/Percent-encoding The articles contain links to RFC 3986 and 3987. Regards, ... Lyall E. -- Message composed and transmitted by software designed to avoid the complication, overhead and vulnerability of antivirus software. 123456789 123456789 123456789 123456789 123456789 123456789 123456789 Tel: +1 360 639 0202 +1 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

1 0

Re (2): Location and name for percent encoding.
by peter＠easthope.ca 31 Dec '18

31 Dec '18

From: Colby Russell <colby(a)colbyrussell.com> Date: Sun, 30 Dec 2018 12:36:30 -0600 > I'm not a fan of bikeshedding, ... One of the early books recommended meaningful names. "Bike shed" strikes me as reasonably meaningful. Better than "precipitation deflecting velocipede depot" for example. > If you do ultimately value accuracy over ergonomics, HTTPForm.Encode, > WWWForm.Encode, or WebForm.Encode would be the natural choices. OK, thanks. I wonder how many first year students would get the meaning of HTTPForm. From: =?utf-8?Q?J=C3=B6rg_Straube?= <joerg.straube(a)iaeth.ch> Date: Sun, 30 Dec 2018 19:40:23 +0100 > MODULE RFC3986; > or > MODULE URI; > > with the procedures Encode() and Decode(). I hadn't thought of RFC3986. It explains motivation along with syntax. Seems a good choice. Thanks, ... Lyall E. -- Message composed and transmitted by software designed to avoid the complication, overhead and vulnerability of antivirus software. 123456789 123456789 123456789 123456789 123456789 123456789 123456789 Tel: +1 360 639 0202 +1 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

2 1

SystemV- Heap + Module Space
by Andreas Pirklbauer 30 Dec '18

30 Dec '18

> Also, based on similar idea, the trivia of Apple Macintosh memory > management was built, around time, when 640K was a limit under DOS. BTW, the OS built for the original 1984 Macintosh is a marvel of engineering. It is written in Pascal and absolutely worth studying.. they only had 128K to play with and therefore optimal memory management was critical.

1 0

SystemV- Heap + Module Space
by Andreas Pirklbauer 30 Dec '18

30 Dec '18

> Should it not start at lower end with first unused gap? (the other way round) The module loader (procedure Modules.Load) could of course do that, but starting with mod = Modules.root is just the simplest possible code. Doing it the other way around would also be only a single line of code, but one would need to traverse the *entire* module list find the best fit at the "lower end". You can try to refine the module allocation/deallocation strategy. However, this is one area where it was thought that this would be an optimization in the wrong place. On "production systems” module unloading is rare, whereas on development systems, one generally tends to unload “the last loaded” module anyway. And for that, the few lines of code I provided are entirely adequate. A much more significant contribution would be to make module unloading “safe”, i.e. prevent unloading of there are still client, type or procedure variable references.

2 1

SystemV- Heap + Module Space
by Andreas Pirklbauer 30 Dec '18

30 Dec '18

> How can I compact unused module space after clean up, or is it really needed? One small change you could introduce to your system is to change Modules.Mod ... From… PROCEDURE Free*(name: ARRAY OF CHAR); VAR mod, imp: Module; p, q: INTEGER; BEGIN mod := root; res := 0; WHILE (mod # NIL) & (mod.name # name) DO mod := mod.next END ; IF mod # NIL THEN IF mod.refcnt = 0 THEN mod.name[0] := 0X; p := mod.imp; q := mod.cmd; WHILE p < q DO SYSTEM.GET(p, imp); DEC(imp.refcnt); INC(p, 4) END ; ELSE res := 1 END END END Free; To… PROCEDURE Free*(name: ARRAY OF CHAR); VAR mod, imp: Module; p, q: INTEGER; BEGIN mod := root; res := 0; WHILE (mod # NIL) & (mod.name # name) DO mod := mod.next END ; IF mod # NIL THEN IF mod.refcnt = 0 THEN mod.name[0] := 0X; p := mod.imp; q := mod.cmd; WHILE p < q DO SYSTEM.GET(p, imp); DEC(imp.refcnt); INC(p, 4) END ; IF mod = root THEN (*increase size of available module space*) p := mod.size; mod := mod.next; WHILE (mod # NIL) & (mod.name[0] = 0X) DO INC(p, mod.size); mod := mod.next END ; AllocPtr := AllocPtr - p; root := mod END ELSE res := 1 END END END Free; and rebuild the system … ORP.Compile Modules.Mod ~ Boot.Link Modules ~ Boot.Load Modules.bin ~ This small change will automatically clean up unused module blocks at the “end* of the module list (i.e. those shown first by System.ShowModules) when such modules are unloaded, but not those “in the middle”. You could also try to compact those in the middle and make one big unused module block out of them whenever possible, but this would be overkill (probably).

2 1