ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/18/90)
Date: 02-15-90 (08:38) Number: 2907 (Echo)
To: ALL Refer#: NONE
From: DAVID ALBERT Read: (N/A)
Subj: FORTH MODULARITY Status: PUBLIC MESSAGE
Hi everyone,
I am a Modula-2 programmer who recently got into TIL languages. The
reason for this was that I had to build a custom 80188 computer for a
handicapped person and I needed a quick way to get it up and running
interactively, so I read up on TILs and then implemented one. My TIL
has been refined several times and now works rather well.
I do however have several questions: First, I have seen that several
implementations of Forth use a small "inner interpreter loop" using
DS:SI for example as the instructioni pointer. I chose just to use CALL
and RET as the entry and exit to my words. Therfore, CS:IP is my
instruction pointer and word pointer. Here's the question: Why do
people use the separate inner interpreter loop? It seems that the call
and return are much more flexible and that I can more easily manipulate
return addresses since they are just on the stack. I use BP for my
parameter stack pointer.
Question number 2: Since I have come from a Modula-2 background,
reusable, compiled libraries are very important (to me anyway) in the
course of developing good applications. Due to the interpretive nature
of TILs, I seem to be having some trouble implementing libraries. I
would like to allow several libraries (vocabularies) which can be
manipulated independently and linked into my final application. I have
read some of the solutions on late binding, but none I've seen have been
satisfactory. Does anyone have any ideas, hints, etc. Do you know how
anyone else does it? Any advice would be appreciated.
Well, thanks for taking the time to read all of this and I look
forward to your replies! See ya :-)
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/19/90)
Category 2, Topic 8
Message 33 Sun Feb 18, 1990
L.ZETTEL at 17:31 EST
To David Albert (GEnie msg 32).
Baldly put, the advice of many Forthers on the subject of implementing
compiled libraries would be: don't bother. Allow me to explain.
A good Forth compiler is as fast or faster than the linker in many
conventional langauages. Good Forth code is inherently modular. If you
can rapidly create the final product direct from source, why
bother with compiled libraries? If you deal with blocks (as I prefer)
or many short files (as Mitch Bradley prefers), then using load screens
or some equivalent lets you mix, match, and select to your heart's
content.
There are also some collateral advantages. Always going from source
solves one aspect of the "version problem": there is no doubt that the
stuff executing matches the source. You are also spared the temptation
to fatten up the modules by throwing in something that you don't need
now but might be handy some day - when some day arrives (usually much
less often than thought) the additional stuff can be added.
My personal preferred Forth programming style is to start with the
minimum kernal and compile everything else up to the spot I am working
on when I take up the problem again. After that, at the first hint that
something has been corrupted (because I maybe miscalculated an address,
for instance) I trash the executable and rebuild it. Debugging is much
more effective because I always know where I'm at.
This style also encourages review and revision of the lower level
stuff as you gain more insight into what the application should really
look like, so the final product tends to be really tight and well
integrated. Try it for a while!
-LenZ-
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/20/90)
Category 2, Topic 8
Message 34 Sun Feb 18, 1990
F.SERGEANT [Frank] at 20:39 CST
To: David Albert Re: inner interpreter (next) vs CALL/RET on 8088/8086
The reasons some people use LODS, AX JMP, rather than CALL/RET are
1. the former probably runs faster
2. the former probably takes less space.
I say 'probably' in #1 because I've given up thinking I know how long an
instruction takes due to variables such as the mix of instructions affecting
whether the queue stays filled and the actual processor (eg V20 vs 8088), and
in #1 & #2 because of a number of other variabls. My Intel book for the 8088
says
LODSW 1 byte 16 cycles memptr16 CALL 3 bytes 29 cycles
AX JMP 2 bytes 11 cycles RET 1 byte 20 cycles
so, at first glance it seems that letting CS:IP thread thru the address list
takes 49 cycles to DS:SI's 27 cycles. There are a number of complicating
factors if you use CALL/RET.
1st, addressing the data stack is more difficult as you need to do at least
one pair of SP BP XCHG instructions in order to be able to use the PUSH & POP
instructions to address the data stack.
2nd, addressing the return stack becomes easier since you no longer need to
do the SP BP XCHG for it.
3rd, you can eliminate docol (nest) saving both time & space.
4th, you have a smaller, faster exit (unnest), using a 1 byte RET rather
than a 2 byte address for a jump to a central routine.
5th, each entry in the address list takes 3 bytes rather than 2 bytes.
6th, colon definitions can (often) end w/ a 3 byte JMP to the final word in
the list rather than using a 3 byte CALL followed by a 1 byte RET.
Where is our breakeven point? I certainly don't know. I would welcome a
detailed analysis from someone(s). Let's look just at space in a colon
definition: Using CALL/RET we save a 3 byte jump to nest and one byte for
unnest. Each word in the colon definition would take 3 bytes instead of two.
The space breakeven point seems to be 4 words. Unfortunately perhaps, the
code I write averages more than 4 words per colon definition, so space-wise I
seem to be better off using direct threading rather than CALL/RET.
And, there are even more factors to consider such as in-line code and
optimizations vs the ease of decompiling (SEE).
I hope to do some experimenting when I get the time.
Obviously the best solution is to cease using Intel processors.
-- Frank
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/23/90)
Date: 02-21-90 (15:06) Number: 2939 (Echo)
To: ALL Refer#: NONE
From: IAN GREEN Read: HAS REPLIES
Subj: FORTH Status: PUBLIC MESSAGE
Can someone document some 'Forth' instructions for me. I want to know
the words to STORE the top of the stack to a specified memory location
and a word to PUSH on the stack, the data contained in a specified
memory location. What I want is the equavalent to load and store
instructions I suppose.
Ian Green
NET/Mail : British Columbia Forth Board - Burnaby BC - (604)434-5886
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/23/90)
Date: 02-21-90 (15:53) Number: 2940 (Echo)
To: IAN GREEN Refer#: 2601
From: MICHAEL HOBSON Read: NO
Subj: FORTH Status: PUBLIC MESSAGE
re: Beginner's Forth
There are some exellant Forth Tutorials on this board (check the file
areas). For your immediate questions:
For sixteen bit numbers:
@ - 'fetch' (addr - number) replace address on top of
stack with the value at that address.
! - 'store' (number addr - ) stores number below top of
stack into address at top of stack. Both values are
consumed (POPed).
For eight bit numbers:
C@ - 'c-fetch' (addr - char) rreplace address at top of stack
character value at that address. For iapx86 machines, the
value is zero-extended into an unsigned integer.
C! - 'c-store' (char addr - ) store character value below top
of stack into address at top of stack (byte store). Both
items are consumed (POPed)
Good luck with RTX2001A kit, Ian.
"The Elf" [^]-[^]
\---/
Elf - A wise and helpful variety of magical being.
NET/Mail : British Columbia Forth Board - Burnaby BC - (604)434-5886
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'koopman@a.gp.cs.cmu.edu (Philip Koopman) (02/23/90)
In article <514.UUL1.3#5129@willett.UUCP>, ForthNet@willett.UUCP (ForthNet articles from GEnie) writes: > F.SERGEANT [Frank] at 20:39 CST > To: David Albert Re: inner interpreter (next) vs CALL/RET on 8088/8086 > in #1 & #2 because of a number of other variabls. My Intel book for the 8088 > says I have recently seen some work that suggests that the Intel numbers for 80x86 series processors are extremely optimistic on real programs. The book claims that you should add about 5% for prefetch delays, etc. Apparently some folks took measurements that suggest the real number is 30%-40% on an 80C286. If you are thinking of tuning your Forth for clock cycles, get an 80286 reference manual and use it as well since there are an awful lot of ATs out there. You will find out that the clock cycle counts are considerably different for almost all instructions. Harris makes an 80C286 and will probably ship you an "80C286 Hardware Reference Manual" if you call the 800 sales line. PUSH reg = 3 clocks POP reg = 5 clocks LODS = 9 clocks CALL = about 9 clocks (depends on prefetch fill) RET = about 13 clocks JMP reg = about 9 clocks Phil Koopman koopman@greyhound.ece.cmu.edu Arpanet 2525A Wexford Run Rd. Wexford, PA 15090 Senior Scientist at Harris Semiconductor, adjunct professor at CMU. I don't speak for them, and they don't speak for me.
ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/24/90)
Date: 02-22-90 (08:50) Number: 2941 (Echo) To: IAN GREEN Refer#: 2939 From: STEVE PALINCSAR Read: NO Subj: FORTH Status: PUBLIC MESSAGE Ian, the words are ! (store) and @ (fetch). Before discussing their usage, I must explain that the default behavior of a forth variable is to place its address on the stack. Now let's define a variable: VARIABLE SNIP <cr> When we invoke SNIP what we do is place its address on the stack. To store 5 at the memory location we've established with SNIP we simply put 5 on the stack, put SNIP's address on the stack by invoking its name, and then use the ! operator: 5 SNIP ! The stack diagram for ! is ( n addr -- ) which indicates that it wants an integer value with an address on top of it as its stack arguments; after execution, no arguments are returned. SNIP @ will put the value stored at SNIP on the stack. @'s stack diagram is ( addr -- n ). ----- This message came from GEnie via willett through a semi-automated process. Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'
ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/24/90)
Date: 02-22-90 (08:41) Number: 2947 (Echo)
To: MICHAEL HOBSON Refer#: 2940
From: IAN GREEN Read: 02-22-90 (10:10)
Subj: FORTH Status: PUBLIC MESSAGE
Thanks, for the clarification. I assume that the address can be a
symbol like 'HERE' or 'THERE' or 'CONTAINER' etc. As an extremely crude
example, lets try the following.
CONSTANT UARTBASE 3F8 ;
: CLR8250 6 FOR UARTBASE + 0 ! NEXT;
Is this correct, the logic I am attempting to do is to write 0's to each
register in the UART located at $3F8. If not could you fix up the word
so I can understand the control structure. If I am going to use the RTX
I am going to need to be able to use Forth. Thanks again.
Ian Green
NET/Mail : British Columbia Forth Board - Burnaby BC - (604)434-5886
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/24/90)
Date: 02-22-90 (10:15) Number: 2948 (Echo) To: IAN GREEN Refer#: 2605 From: MICHAEL HOBSON Read: 02-22-90 (20:20) Subj: FORTH Status: PUBLIC MESSAGE > Thanks, for the clarification. I assume that the address can be a >symbol like 'HERE' or 'THERE' or 'CONTAINER' etc. As an extremely crude >example, lets try the following. > >CONSTANT UARTBASE 3F8 ; > >: CLR8250 6 FOR UARTBASE + 0 ! NEXT; > >Is this correct, the logic I am attempting to do is to write 0's to eac This example should be more like the following: HEX ( Set Base to Hexadecimal ) 3F8 CONSTANT UARTBASE ( CONSTANT expects a value on the stack. No semicolon is needed here, because the text scan for the constant's name terminates with at the first whitespace character ) : CLR8250 6 ( # of registers to clear ) FOR 0 ( value to store must be pushed first ) UARTBASE I + ( add loop index to base address ) ! ( write zero to port address ) NEXT ; ( semi-colon is a WORD and MUST be separated by whitespace, the open-paren for defining these comments is also a WORD and MUST be separated by whitespace. ) Of course, the 8250 base address will be whatever you decide to decode from the RTX 2001A address lines, so substitute as appropriate. I hope this helps somewhat. "The Elf" [^]-[^] \---/ Elf - A wise and helpful variety of magical being. NET/Mail : British Columbia Forth Board - Burnaby BC - (604)434-5886 ----- This message came from GEnie via willett through a semi-automated process. Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'
ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/24/90)
Date: 02-22-90 (20:23) Number: 2949 (Echo)
To: MICHAEL HOBSON Refer#: 2606
From: IAN GREEN Read: NO
Subj: FORTH Status: PUBLIC MESSAGE
Thanks for clarifying my attempt to understand an elemantary control
structure. As a novice to the language, I appreciate the assistance
greatly.
I assume that inside the loop structure, I can substitute any
sequence of instructions I desire. My next question is how would I
implement a nested FOR loop. In pseudo code:
for control1 = 1 to limit1
for control2 = 1 to limit2
for control3 = 1 to limit3
dosomething
next
next
next
If it is not a problem, perhaps we could start a thread that others can
use to try to move from a 'conventional' language to Forth. With this
contest as the carrot, I feel that learning at least some degree of
Forth would be a valuable addition to my experience in programming. It
would also make life a lot easier when I attempt to 'read' Forth
programs outlining various algorithms.
Thank for answering my stupid questions.
Ian Green
NET/Mail : British Columbia Forth Board - Burnaby BC - (604)434-5886
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/25/90)
Date: 02-23-90 (13:18) Number: 2951 (Echo) To: IAN GREEN Refer#: 2949 From: MICHAEL HOBSON Read: NO Subj: FORTH Status: PUBLIC MESSAGE >If it is not a problem, perhaps we could start a thread that others can >use to try to move from a 'conventional' language to Forth. With this >contest as the carrot, I feel that learning at least some degree of >Forth would be a valuable addition to my experience in programming. It >would also make life a lot easier when I attempt to 'read' Forth >programs outlining various algorithms. > >Thank for answering my stupid questions. Ian, you have probably already heard this before, but no question is stupid! The desire to learn more is something that I understand, and I have always appreciated the patience of those who answered my 'stupid' questions. As for a new thread on this subject, why not. Let's ask the sysop about it. Maybe we'll call it 'GOTO Forth' :-) "The Elf" [^]-[^] \---/ Elf - A wise and helpful variety of magical being. ----- This message came from GEnie via willett through a semi-automated process. Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'
ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/26/90)
Category 2, Topic 8 Message 42 Sat Feb 24, 1990 F.SERGEANT [Frank] at 20:30 CST To Phil Koopman >I have recently seen some work that suggests that the Intel numbers >for 80x86 series processors are extremely optimistic on real programs. Thanks for the suggestions. I'll check into getting the '286 manual. Some days I want to tune for speed and other days I'm happy when the code runs at all. As you say, there are a lot of ATs out there. They don't have to be good if they are cheap enough. I did enjoy the old days of predictable instruction cycle timings - a selling point for the RTX 2000/1, I believe. I wonder how our world would be different today if Motorola had gotten the IBM PC contract instead of Intel. - Frank ----- This message came from GEnie via willett through a semi-automated process. Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'
ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/26/90)
Date: 02-24-90 (07:55) Number: 2959 (Echo)
To: ALL Refer#: NONE
From: IAN GREEN Read: (N/A)
Subj: FORTH Status: PUBLIC MESSAGE
Can someone write a demo program to outline a nested FOR loop.
Something like the following pseudo code:
for i = 1 to n
for j = 1 to m
dosomething
next
next
-------
How about declaring data structures like an array, or a queue. A
stack is automatic but I am unsure how to implement these other
elementary data structures.
Thanks,
Ian Green
NET/Mail : British Columbia Forth Board - Burnaby BC - (604)434-5886
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'rob@idacom.uucp (Rob Chapman) (02/27/90)
> From: IAN GREEN Read: (N/A) > Can someone write a demo program to outline a nested FOR loop. > Something like the following pseudo code: > for i = 1 to n > for j = 1 to m > dosomething > next > next This translates quite easily with a sprinkling of postfix to: n FOR m FOR dosomething NEXT NEXT In botForth, if n or m are 0 the 'dosomething' will not be executed. If they are nonzero, then 'dosomething' will be executed n m * times as you wanted. Others please correct me if I err. In ANS Forth I believe the same code will execute n 1 + m 1 + * times and if n or m are 0, the 'dosomething' will be executed at least once. To handle the 0 case ?DUP IF...THEN construncts must be used. I'm not sure, but I think RTXForth follows the ANS Forth definition. (aside to the ANS people: isn't it tougher to explain and harder to use this way?) > How about declaring data structures like an array, or a queue. A > stack is automatic but I am unsure how to implement these other > elementary data structures. I have found queues to be a fundamental piece for interconnecting processes. The queue definition and it's operators extracted from botForth are: ( ==== Queues ===== ) QUEUE ( #words -- ) create a queue with #words entries >Q ( n \ queue -- ) append n to end of queue Q> ( queue -- n ) return first value in queue Q ( queue -- n ) get a copy of first value in queue 0Q ( queue -- ) remove all items from a queue Q? ( queue -- flag ) return number of items in a queue : QUEUE ( #words -- ) ( Queues: | >insert | >remove | >end | queue... | ) <BUILDS HERE 6 + , HERE 4 + , 1 + 2* DUP HERE + , ALLOT ; : Q? ( queue -- n ) @+ @+ >R SWAP - DUP 0< IF R @ R> - + 2/ EXIT ENDIF R> DROP 2/ ; : >Q ( n \ queue -- ) DUP>R @ !- DUP 4 - R - IF R> ! EXIT ENDIF @ R> ! ; : Q> ( queue -- n ) 2 + DUP>R @ @- DUP 2 - R - IF R> ! EXIT ENDIF @ R> ! ; : Q ( queue -- n ) 2 + @ @ ; : 0Q ( queue -- ) @+ ! ; ( example ) 10 QUEUE dataq ( queue to hold 10 data items ) dataq 0Q 1 dataq >Q 2 dataq >Q 3 dataq >Q dataq Q? . ( emit a 3 ) dataq Q> dataq Q> dataq Q> . . . ( emit 3 2 1 since the stack ) ( reverses the order ) dataq Q? . ( emit a 0 since it's now empty ) The above definitions have been written for speed on the RTX20xx. Again I'm guessing about the Forth that you are using so I've included definitions and explanations for some non-standard but useful words: ( ==== Incrementing/Decrementing memory access ==== ) @+ ( a -- n \ a+ ) fetch a word from address a and increment a by 2 @- ( a -- n \ a- ) fetch a word from address a and decrement a by 2 !- ( n \ a -- a- ) store a word at address a and decrement a by 2 : @+ ( a -- n \ a+ ) DUP @ SWAP 2 + ; : @- ( a -- n \ a- ) DUP @ SWAP 2 - ; : !- ( n \ a -- a- ) TUCK ! 2 - ; Well this is quite a bit already so I'll leave arrays to someone else. rob@idacom " It's so simple, it has to work! "
ForthNet@willett.UUCP (ForthNet articles from GEnie) (02/27/90)
Category 2, Topic 8
Message 44 Sun Feb 25, 1990
M.HAWLEY at 20:07 EST
This will be close:
: NESTED-LOOPS
N 1 DO
M 1 DO
DOSOMETHING
LOOP
LOOP
;
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'ForthNet@willett.UUCP (ForthNet articles from GEnie) (03/01/90)
Date: 02-27-90 (10:02) Number: 1609 (Echo) To: IAN GREEN Refer#: 1608 From: STEVE PALINCSAR Read: NO Subj: ADVANCED BEGINNERS Status: PUBLIC MESSAGE You can't use an undefined word. Since you're talking about variables here, and since they are forth words, they need to be defined. If you tried to use an undefined word, the outer interpreter would simply give you an error message. OTOH, you don't "allocate space on the stack." You put numbers on the stack, and the stack itself takes care of management of the memory. Actually, the code example given (with N 1 DO ) wouldn't work with a variable (since you must use @ to get the value of the variable) but would instead set up a loop that looped from 1 to 1 less than the numeric value of the address of the variable N, which isn't what you wanted to do at all! OTOH some forths have a data structure called a VAR which instead of returning its address on the stack when it is invoked instead returns the value stored at that address. Forth constants also return the value instead of the address. It's also useful to know that you aren't dealing with variables in a loop. The actual values for the starting and ending points of the loop _as numbers_ must be on the parameter stack, since DO wants two arguments; and DO puts those two values on the return stack. Similarly, I (which gets at the current index of the inner loop) and J (which gets at the current index of the outer loop) are not variables either, but instead are words which copy the current index value from the Rstack. ----- This message came from GEnie via willett through a semi-automated process. Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'
ForthNet@willett.UUCP (ForthNet articles from GEnie) (03/01/90)
Date: 02-27-90 (23:55) Number: 1610 (Echo) To: IAN GREEN Refer#: 1608 From: NICK JANOW Read: NO Subj: ADVANCED BEGINNERS Status: PUBLIC MESSAGE No, 'N' and 'M' are the maximum counts you desire; the '1's are the starting value. For example: 20 10 DO LOOP Counts from 10 19. Starting with 1 gives you the familiar 1 to n count. The count index is 'I'. Within the loop, 'I' puts the present count value on the top of the stack. For nested loops, as in your question, 'J' returns the count value (index) of the next outer loop. You don't have to declare 'I' or 'J'. NET/Mail : British Columbia Forth Board - Burnaby BC - (604)434-5886 ----- This message came from GEnie via willett through a semi-automated process. Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'
ForthNet@willett.UUCP (ForthNet articles from GEnie) (03/02/90)
Date: 02-28-90 (07:39) Number: 1612 (Echo)
To: NICK JANOW Refer#: 1610
From: IAN GREEN Read: NO
Subj: ADVANCED BEGINNERS Status: PUBLIC MESSAGE
OK, thanks. In the RTX docs I have, it mentions that I have a
FOR-NEXT loop construct, so I will use that as it seems to be vaguely
usable from my limited understanding of Forth. A code fragment that I
was given examplifies what I am aspiring to:
HEX
CONSTANT UART-BASE 2F8 ; ( com2 )
: CLEAR8250 ( a word to reset a UART )
6 FOR ( 6 registers )
0 ( the goose egg )
UART-BASE ( push the base address )
I ( fetch up the loop index )
+ ( compute address )
! ( barf because of I/O mapping )
NEXT
;
Now from what I understand the following is also kosher, but highly
wasteful of CPU cycles:
: CLEAR8250-2 ( mark 2 )
3 FOR ( outer loop )
2 FOR ( inner loop )
0 ( the goose egg again )
UART-BASE ( the invisible UART )
I ( fetch up the outer loop index )
J ( don't forget the inner loop )
+ ( compute )
+ ( rediculus isn't it )
! ( finally )
NEXT
NEXT
;
From what I understand of the syntax of this basic control structure, I
can implement 'nested' loops to handle rasters and arrays etc. One idea
I have is to try to translate my fractal program into Forth as an
exercise. It does require floating point but as I recall, the F83
program has floats internally so that I should be OK.
As for data containers I assume I can simply declare a variable as
follows:
VARIABLE CONTAINER ;
or is my syntax bogus?
Ian Green
NET/Mail : British Columbia Forth Board - Burnaby BC - (604)434-5886
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'ForthNet@willett.UUCP (ForthNet articles from GEnie) (03/02/90)
Category 2, Topic 8 Message 49 Thu Mar 01, 1990 M.HAWLEY at 22:30 EST FOR-NEXT looks like a regression to BASIC to me, but then FORTH is nothing if not flexible. Regarding fractals you might be interested in my MANDELZEN file ( I think that is what I called it. ) Entirely integer F83. ----- This message came from GEnie via willett through a semi-automated process. Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'
ForthNet@willett.UUCP (ForthNet articles from GEnie) (03/03/90)
Category 2, Topic 8
Message 50 Fri Mar 02, 1990
D.RUFFER [Dennis] at 01:03 EST
You've got two of them up there. Here they are:
[(Silly Question:) Does anyone want me to port these to the Internet? -dwp]
No. File Name Type Address YYMMDD Bytes Accesses Lib
1471 2-MANDEL.ARC X M.HAWLEY 890121 7560 20 6
Desc: F-PC MANDELBROT DOUBLE PRECISION
1247 MANDLZEN.ARC X M.HAWLEY 880918 6300 28 3
Desc: One screen. Draws Mandelbrot Set.
Ian, your syntax for variable is close, but drop the ; at the end.
i.e. VARIABLE CONTAINER
DaR
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'louis@marco.UUCP (Juergen Fluk) (03/07/90)
RRRRRR TTTTTTT FFFFFFF M M could you please stop these boring R R T F MM MM questions ? there are very good books R R T F M M M M (e.g. Leo Brodie's "Starting Forth") RRRRRR T FFFFF M M M which can answer all your questions. R R T F M M you could even find someone to answer R R T F M M your questions via e-mail (not me :-), R R T F M M but don't drown this newsgroup. louis What about "How can I add two numbers, say 3 and 4 ..."
ForthNet@willett.UUCP (ForthNet articles from GEnie) (03/22/90)
Date: 03-20-90 (08:02) Number: 1624 (Echo) To: ALL Refer#: NONE From: MARK ANDREWS Read: (N/A) Subj: D85 FORTH AND EMACS Status: PUBLIC MESSAGE I have copies of D85 and the DOS version of Emacs. I read in the D85 docs that it was possible to have D85 call Emacs to use as its editor. Unfortunately, the D85 info assumes you know Forth well enough to see how another Forther added the Qedit editor, and substitute Emacs in its place. I can't even see how D85 can call Qedit, much less Emacs; I am out of my depth. Any hints from anyone? ----- This message came from GEnie via willett through a semi-automated process. Report problems to: 'uunet!willett!dwp' or 'willett!dwp@gateway.sei.cmu.edu'
ForthNet@willett.UUCP (ForthNet articles from GEnie) (07/02/90)
Category 18, Topic 57
Message 7 Sat Jun 30, 1990
D.RUFFER [Dennis] at 23:18 EDT
Re: seg@PacBell.COM (S. E. Grove)
> I would like a decompilation of the FORTH word U.R.
: U.R ( u n) SWAP 0 <# #S #> ROT OVER - SPACES TYPE ;
This work in polyFORTH and should be pretty standard. The main difference
will probably be on what size of number #S converts. In polyFORTH it takes a
double number, thus the 0 is put on before conversion. This word also does
not do any checking that the converted number is larger than the field. I
contend that the programmer should know this fact before writing the
application, but others may disagree with me and insist on the check before
the SPACES.
Anyway, that is a start. DaR
-----
This message came from GEnie via willett through a semi-automated process.
Report problems to: uunet!willett!dwp or willett!dwp@hobbes.cert.sei.cmu.eduForthNet@willett.UUCP (ForthNet articles from GEnie) (07/11/90)
Date: 07-07-90 (23:43) Number: 1811 (Echo) To: ALL Refer#: NONE From: BOB CAVANAUGH Read: (N/A) Subj: RESTART APPLICATION Status: PUBLIC MESSAGE Can anybody out there give me a way to jump back to a word after an error condition is set? Specifically, I wrote a program to read either S-records or Intel hex records and burns EPROM on a circuit card. When I encounter a bad checksum, I want to restart the program from my main menu routine, MAIN. I tried ' main execute and it went into never-never land. I am using V3.5 Thanks -- Bob Cavanaugh ----- This message came from GEnie via willett through a semi-automated process. Report problems to: uunet!willett!dwp or willett!dwp@hobbes.cert.sei.cmu.edu
ForthNet@willett.UUCP (ForthNet articles from GEnie) (07/11/90)
Category 2, Topic 8 Message 53 Mon Jul 09, 1990 D.RUFFER [Dennis] at 20:32 EDT Re: BOB CAVANAUGH > Can anybody out there give me a way to jump back to a word > after an error condition is set? ... I tried > ' main execute and it went into never-never land. Bob, I'd suspect that you are overflowing your return stack when you try to do that. If you are doing this fairly deeply into you program and your main routine calls a bunch of things before getting to the menu, then this is pretty likely your problem. You need to figure out how to reset the return stack before executing main. Even if this is not what is hurting you yet, you will need to deal with the return stack problem eventually. Another possiblity is that your main was not defined when you define the ' main execute and you are executing some other definition of main that is doing something nasty. Check your load order to find out if that is the case. If so, you will need to use a DEFERed word to resolve the forward reference for you. > I am using V3.5 I assume this is F-PC, but I don't use it much. Does it have some kind if contol over ABORT? Realistically, your main word is nothing more than another version of QUIT (where ABORT usually goes to). Many systems give you a way to change where ABORT goes to so that you can define alternative QUIT's. Check out how ABORT or ABORT" works and you will probably find a way that you can use. DaR ----- This message came from GEnie via willett through a semi-automated process. Report problems to: uunet!willett!dwp or willett!dwp@hobbes.cert.sei.cmu.edu
ForthNet@willett.UUCP (ForthNet articles from GEnie) (07/16/90)
Category 2, Topic 8 Message 54 Sun Jul 15, 1990 NMORGENSTERN at 14:53 EDT Re: BOB CAVANAUGH > Can anybody out there give me a way to jump back to a word > after an error condition is set? In many cases a simple loop is enough. For example, if the word DO-IT leaves TRUE on the stack if ok and FALSE if not, you can repeat until satisfactory by this: BEGIN DO-IT UNTIL. If you want to do a fixup, you can use this: BEGIN DO-IT NOT WHILE FIXUP REPEAT These do not permit jumping out of a nested word. Several methods have been proposed that do just that. For example: CATCH THROW, which I believe is in the library. I proposed a method at ACM SIGFORTH, Austin, 1989, which uses 5 words. I have tested it in F83, and it should work in F-PC, but I haven't tried it. The idea is to set a "safety-net" that will catch you if you fall. You can have as many nets as you need, and can change them while the program is running. The method permits a Jump-Back-To a previous point, but you cannot jump forward into. : NET CREATE 0 , 0 , 0 , ; : 3! ( n1 n2 n3 a -- ) 2DUP ! 2+ NIP 2! ; : 3@ ( a -- n1 n2 n3) DUP 2+ 2@ ROT @ ; : SET-NET ( net -- ) R@ SP@ ROT RP@ SWAP 3! ; : FALL ( net -- ) 2+ RP! SWAP >R 4 + SP! ; For example: To declare a net: NET SAFETY1 To set the net ... SAFETY1 SET-NET ( point A ) FOO FAH .... To go back to point A from any word nested within FOO or FAH simply incorporate the phrase SAFETY1 FALL The data stack and return stack will be restored to their original depth and the program will take off from point A. It is possible to incorporate a net in ABORT". The details are given in the Austin paper. I have recently modified this scheme in several ways, mainly to permit recursion, but have not completed all the details. ----- This message came from GEnie via willett through a semi-automated process. Report problems to: uunet!willett!dwp or willett!dwp@hobbes.cert.sei.cmu.edu