July 2016 - Oberon - mailman3.inf.ethz.ch

ReadNum and WriteNum in C
by Srinivas Nayak 12 Jul '16

12 Jul '16

Dear All, Here I tried to implement ReadNum and WriteNum in C. I couldn't do it simply using / and % imitating how NW has written them using DIV and MOD and Shifts. I would like to see a better way to do it... #include <stdio.h> #define ELSIF }else if( #define ELSE }else{ #define IF if( #define THEN ){ #define END } #define WHILE while( #define DO ){ #define REPEAT do{ #define UNTIL }while(!( #define RETURN return unsigned int ROR(unsigned int x, int n) { return (x >> (n%32))|(x << ((32-n)%32)); } int ASR(int x, int n) { return (x >> n%32); } void ReadNum(FILE* R, int* x) { int y, n; unsigned char b; n = 32; y = 0; b = fgetc(R); //printf("%hhu ", b); WHILE b >= 0x80 DO y = ROR(y|(b-0x80), 7); n = n-7; b = fgetc(R); //printf("%hhu ", b); END //printf("\n"); IF n <= 4 THEN *x = ROR(y|(b&0xF), 4); ELSE *x = ASR(ROR(y|b, 7), n-7); END } void WriteNum(FILE* R, int x) { unsigned char b; WHILE (x < -0x40) || (x >= 0x40) DO b = (x & 0x7F)|0x80; fputc(b, R); //printf("%hhu ", b); x = ASR(x, 7); END b = (x & 0x7F); fputc(b, R); //printf("%hhu ", b); //printf("\n"); } int main() { FILE *f; int i = 0; int m = 0, n = 0; for (i = -8192; i <= 8192; i++) { m = i; //printf("\nnum write = %d\n", m); f = fopen("test.bin", "wb"); WriteNum(f, m); fclose(f); f = fopen("test.bin", "rb"); ReadNum(f, &n); fclose(f); //printf("num read = %d\n", n); if(m != n) { printf("%d FAILED\n", m); } } return 0; } With thanks and best regards, Yours sincerely, Srinivas Nayak Home: http://www.mathmeth.com/sn/ Blog: http://srinivas-nayak.blogspot.in/

3 3

Use of DIV and MOD for bit manipulation
by Srinivas Nayak 12 Jul '16

12 Jul '16

Dear All, While reading compiler code(OBG.Mod.txt), I see that in many places we have used DIV, MOD, multiplication etc. No doubt, one can argue that, we want to do a mathematical operation there, but to be honest, the interest there is not to do a mathematical operation; rather to do some bit manipulation. And, since we know that DIV and MOD can help us there, we employ them. How shall we justify it (to obscure the natural meaning)? Isn't the code clarity being compromised there? I would like to know, how Oberoners think about it... With thanks and best regards, Yours sincerely, Srinivas Nayak Home: http://www.mathmeth.com/sn/ Blog: http://srinivas-nayak.blogspot.in/

3 3

On Close in ORG.Mod.txt
by Srinivas Nayak 11 Jul '16

11 Jul '16

Dear All, In ORG.Mod.txt we see PROCEDURE Close*(VAR modid: ORS.Ident; key, nofent: LONGINT); ... i := (i+4) DIV 4 * 4; INC(comsize, i+4) ... Is this correct? Should it be i := (i+3) DIV 4 * 4; INC(comsize, i+4) ? With thanks and best regards, Yours sincerely, Srinivas Nayak Home: http://www.mathmeth.com/sn/ Blog: http://srinivas-nayak.blogspot.in/

2 1

Version control system for NW's oberon/compiler code
by Srinivas Nayak 10 Jul '16

10 Jul '16

Dear All, I see, time to time FPGA Oberon code and compiler code is being updated. And the latest code is available at Professor Wirth's site and www.projectoberon.com Do we keep these in any version control system? Will it be helpful? With thanks and best regards, Yours sincerely, Srinivas Nayak Home: http://www.mathmeth.com/sn/ Blog: http://srinivas-nayak.blogspot.in/

1 0

Treatment of parameter-less procedure type in ORP.Mod.txt
by Srinivas Nayak 10 Jul '16

10 Jul '16

Dear All, I think ORP.Mod.txt missed to make ptype.nofpar := 0; in PROCEDURE ProcedureType() for parameter-less procedure types... Such as: TYPE Command* = PROCEDURE; BEGIN body (*initialize module*) END Not sure how Oberon compiler manages to get nofpar = 0! Or, I am missing it somewhere? Or, is there a chance of any memory chunk pre-initialized to 0 in Oberon? With thanks and best regards, Yours sincerely, Srinivas Nayak Home: http://www.mathmeth.com/sn/ Blog: http://srinivas-nayak.blogspot.in/

3 4

BYTE to/from INTEGER conversion
by Srinivas Nayak 06 Jul '16

06 Jul '16

In ORB.Mod.txt we see PROCEDURE Read(VAR R: Files.Rider; VAR x: INTEGER); VAR b: BYTE; BEGIN Files.ReadByte(R, b); IF b < 80H THEN x := b ELSE x := b - 100H END END Read; and PROCEDURE Write(VAR R: Files.Rider; x: INTEGER); BEGIN Files.WriteByte(R, x) END Write; Here, Read gets a BYTE from Files.ReadByte(R, b) and converts that BYTE to INTEGER using IF .. END But, Write passes the INTEGER directly to Files.WriteByte(R, x) where prototype of WriteByte is PROCEDURE WriteByte*(VAR r: Rider; x: BYTE); That means, compiler is able to convert a 4 byte INTEGER implicitly to a 1 byte BYTE, But the reverse is not possible... How can I understand this? Please shed some light on this. With thanks and best regards, Yours sincerely, Srinivas Nayak Home: http://www.mathmeth.com/sn/ Blog: http://srinivas-nayak.blogspot.in/

2 3

frames, displays and screens;
by Andreas Pirklbauer 04 Jul '16

04 Jul '16

That's in effect what virtual screens are. Question of course is how to initiate the jumping between virtual screens (other than via the command System.SetDisplay). I didn't want the scroll bar to extend above or below the visible part of the current virtual display or to clutter the display by introducing additional visual items (such as a small arrow somewhere at the top and the bottom of the current display that allows one to jump to the previous or next virtual display). One could though.. it would be a trivial addition. -------------- peter at easthope.ca peter at easthope.ca Mon Jul 4 17:14:37 CEST 2016 - Previous message: [Oberon] Oberon in Atlantic City - Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] From: Andreas Pirklbauer <andreas_pirklbauer at yahoo.com> Date: Sat, 2 Jul 2016 14:17:51 +0200 > In essence one can create sev= > eral (separate!) logical display areas that all look like the standard Obero= > n display (works like virtual screens in macOS, i.e. one can just switch fro= > m one to another). OK, thanks. Rather than scroll a big display up-down and left-right on a smaller screen, the user will jump from one logical screen to another. Both scrolling and jumping could be in one system. Eg. arrange your virtual screens vertically (or horizontally) and allow scrolling and jumping up/down (or left/right). > It's just a small experiment to prepare Oberon for a tablet version. Seems you are on the right track. Thanks! ... Lyall E.

1 0

frames, displays and screens; was Re (2): I have been defeated.
by Andreas Pirklbauer 04 Jul '16

04 Jul '16

This refers to "virtual logical display areas" in Experimental Oberon (see the demo video on its github page to see how it fits). It's just a small experiment to prepare Oberon for a tablet version. In essence one can create several (separate!) logical display areas that all look like the standard Oberon display (works like virtual screens in macOS, i.e. one can just switch from one to another). peter at easthope.ca peter at easthope.ca Fri Jul 1 17:13:10 CEST 2016 Previous message: [Oberon] On praise of repeat-until construct Next message: [Oberon] Oberon in Atlantic City Messages sorted by: [ date ] [ thread ] [ subject ] [ author ] From: Andreas Pirklbauer <andreas_pirklbauer at yahoo.com> Date: Sat, 25 Jun 2016 22:23:07 +0000 > ... the copy is created in a separate virtual logical display area. Please explain why the "separate virtual logical display" will be necessary or helpful. Where or how will it fit in the screen with the primary display? Thanks, ... Lyall E. -- http://easthope.ca/Peter.html Bcc: peter at easthope. ca

2 1

Oberon in Atlantic City
by Chris Burrows 02 Jul '16

02 Jul '16

Astrobe for Cortex-M3 has been used to develop Oberon software to program an Opus-Two LPC1768 Microcontroller-based system used in the largest organ in the world in Atlantic City. More information and links to the news articles about this newly restored organ are on the Astrobe forum: http://www.astrobe.com/forum/viewtopic.php?f=3&t=505 Regards, Chris Burrows CFB Software http://www.astrobe.com

2 1

frames, displays and screens; was Re (2): I have been defeated.
by peter＠easthope.ca 01 Jul '16

01 Jul '16

From: Andreas Pirklbauer <andreas_pirklbauer(a)yahoo.com> Date: Sat, 25 Jun 2016 22:23:07 +0000 > ... the copy is created in a separate virtual logical display area. Please explain why the "separate virtual logical display" will be necessary or helpful. Where or how will it fit in the screen with the primary display? Thanks, ... Lyall E. -- 123456789 123456789 123456789 123456789 123456789 123456789 123456789 12 Tel +1 360 639 0202 http://easthope.ca/Peter.html Bcc: peter at easthope. ca

1 0