I actually don't understand, why no Forth creator inserts
into his design WORDS, that allow to examine what's actually
„inside”. Perhaps the worst example is FPC Forth — 1500 words
making just „zooooooooooom!” throught the screen, like a thunder,
when one types WORDS.

It doesn't seem to be that difficult to create „paged” WORDS; here's
my simple take (it's fig-Forth):

: WORDS ( -- )
0 OUT C! 22 OUT 1+ C! CR
CONTEXT @ @
BEGIN
DUP ID. SPACE PFA LFA @ DUP 0=
OUT C@ C/L >
IF CR 0 OUT C! OUT 1+ DUP C@ 1- SWAP C! THEN
OUT 1+ C@ 0= DUP
IF DROP 22 OUT 1+ C! ." >>> ENTER to continue -- any other key to end <<<"
CR KEY 13 = 0= 0 OUT C! THEN ( OUT-lo "reset" because of the message )
OR
UNTIL DROP ;

(yes, a few „magic numbers”, but no big deal; 22 is the lines' count)
--
Z.

On Saturday, December 23, 2023 at 6:26:04 PM UTC, LIT wrote:
> I actually don't understand, why no Forth creator inserts
> into his design WORDS, that allow to examine what's actually
> „inside”. Perhaps the worst example is FPC Forth — 1500 words
> making just „zooooooooooom!” throught the screen, like a thunder,
> when one types WORDS.
>
> It doesn't seem to be that difficult to create „paged” WORDS; here's
> my simple take (it's fig-Forth):
>
> : WORDS ( -- )
> 0 OUT C! 22 OUT 1+ C! CR
> CONTEXT @ @
> BEGIN
> DUP ID. SPACE PFA LFA @ DUP 0=
> OUT C@ C/L >
> IF CR 0 OUT C! OUT 1+ DUP C@ 1- SWAP C! THEN
> OUT 1+ C@ 0= DUP
> IF DROP 22 OUT 1+ C! ." >>> ENTER to continue -- any other key to end <<<"
> CR KEY 13 = 0= 0 OUT C! THEN ( OUT-lo "reset" because of the message )
> OR
> UNTIL DROP ;
>
> (yes, a few „magic numbers”, but no big deal; 22 is the lines' count)
> --
> Z.

possibly because a better way to do it would be to decouple the source of
the character stream (ie WORDS), and the paging mechanism, so that the
latter could be used for any input stream. In a similar way to how in Unix/Linux
you can 'pipe' a stream to 'more' to get paged output.

(I would hope that at least some Forths work like this, but I confess I have
no idea if this is true)

> possibly because a better way to do it would be to decouple the source of
> the character stream (ie WORDS), and the paging mechanism, so that the
> latter could be used for any input stream. In a similar way to how in Unix/Linux
> you can 'pipe' a stream to 'more' to get paged output.

I'm not opposed; if not the proposed way, then „by decoupling”,
just it should be done — one way, or another
--
Z.

On 24/12/2023 5:25 am, LIT wrote:
> I actually don't understand, why no Forth creator inserts
> into his design WORDS, that allow to examine what's actually
> „inside”. Perhaps the worst example is FPC Forth — 1500 words
> making just „zooooooooooom!” throught the screen, like a thunder,
> when one types WORDS.
>
> It doesn't seem to be that difficult to create „paged” WORDS; here's
> my simple take (it's fig-Forth):

Paged mode with msg can be less convenient if one wants to do whole
dumps e.g. for documentation.

p.s. F-PC had no delay/calibration for stuff like WORDS ?

> Paged mode with msg can be less convenient if one wants to do whole
> dumps e.g. for documentation.

But I mean strictly WORDS, which (I believe) is hardly
ever used for that, and its purpose is rather to allow
„quick view”, interactive way, into „what do we offer”.

> p.s. F-PC had no delay/calibration for stuff like WORDS ?

I didn't actually use FPC that extensively — maybe indeed
I missed something? Even, if it has something like that to
adjust, still its defaults (regarding WORDS) could be more
sane
--
Z.

On 24/12/2023 6:51 pm, LIT wrote:
>> Paged mode with msg can be less convenient if one wants to do whole
>> dumps e.g. for documentation.
>
> But I mean strictly WORDS, which (I believe) is hardly
> ever used for that, and its purpose is rather to allow
> „quick view”, interactive way, into „what do we offer”.

That too. Every new forth I tried I would dump its words to keep
as a sorted quick reference.

>
>> p.s. F-PC had no delay/calibration for stuff like WORDS ?
>
> I didn't actually use FPC that extensively — maybe indeed
> I missed something? Even, if it has something like that to
> adjust, still its defaults (regarding WORDS) could be more
> sane

Had a quick look and could only find FUDGE which calibrated MS.
Has no effect on WORDS AFAICS. Strange to have MS and not use
it in WORDS.

> That too. Every new forth I tried I would dump its words to keep
> as a sorted quick reference.

But actually paged WORDS will make it easier — one page,
screen-dump, the second page, screen-dump etc.

I mean DOS' environment, of course
--
Z.

On 24/12/2023 7:42 pm, LIT wrote:
>> That too.  Every new forth I tried I would dump its words to keep
>> as a sorted quick reference.
>
> But actually paged WORDS will make it easier — one page,
> screen-dump, the second page, screen-dump etc.
>
> I mean DOS' environment, of course

Don't know about easier. If one has 'MS' it's cheaper to use that
than to implement paging.

LIT wrote:

>> That too. Every new forth I tried I would dump its words to keep
>> as a sorted quick reference.

The limited usefulness of WORDS has always irritated me, too.
In the old days, the teletype or VDU's baudrate slowed it down to
readable speed.

iForth uses a console window to interact with the user, and this console
can be configured to have any size scrollback buffer. Mark, search, and
moused copy-and-paste (to the line editor) all work on modern consoles.

Then there is WORDS-DELAY :
FORTH> help words-delay
WORDS-DELAY IFORTH
( -- a-addr )
Contains the number of milliseconds to wait per line of output in
WORDS WORDS: and WORDS?
See also: WORDS WORDS: and WORDS?

'WORDS: abc' shows only those words containing substring abc, WORDS?
allows a grep-like pattern like *ape or a?pe.

Old CLF discussions resulted in doWORDS

FORTH> help doWORDS
doWORDS IFORTH
( xt -- count-of-matched-entries )
The wordlist traversing factor of WORDS WORDS: and WORDS?
Supply the xt of a word that can filter ( addr -- bool ) where
addr is the address of a counted string (name of the next word
in the list). For example, to realize a clone of WORDS that only
prints short words you can do:
: SWORDS-XT ( addr -- bool ) C@ 3 < ; ' SWORDS-XT doWORDS DROP .

Further manipulation is possible with HELP <word>, LOCATE <word>, e.g.
FORTH> locate +
File: d:dfwforth/x64/inc/kernel.inc
1834: T: NEGATE NEGATE T; ANS <w> --- <-w>
1835: T: BSWAP BSWAP T; <w1> --- <w2>
1836: T: - - T; ANS <u> <u> --- <u>
1837>> T: + + T; ANS <u> <u> --- <u>
1838: T: CELL+ CELL+ T; ANS <addr> --- <addr+>
1839: T: CELL- CELL- T; <addr-> --- <addr->
1840: T: CELLS CELLS T; ANS <n> --- <n*> ok

-marcel

In article <980a047687a8dfd0e2df160ec6e93d93@news.novabbs.com>,
LIT <zbigniew2011@gmail.com> wrote:
>I actually don't understand, why no Forth creator inserts
>into his design WORDS, that allow to examine what's actually
>„inside”. Perhaps the worst example is FPC Forth — 1500 words
>making just „zooooooooooom!” throught the screen, like a thunder,
>when one types WORDS.

No it isn't. It is supposed to run on a PC that is so slow that
you have plenty of time to hit the space button before the first
screen ;-)

>
>It doesn't seem to be that difficult to create „paged” WORDS; here's
>my simple take (it's fig-Forth):
>
>: WORDS ( -- )
> 0 OUT C! 22 OUT 1+ C! CR
> CONTEXT @ @
> BEGIN
> DUP ID. SPACE PFA LFA @ DUP 0=
> OUT C@ C/L >
> IF CR 0 OUT C! OUT 1+ DUP C@ 1- SWAP C! THEN
> OUT 1+ C@ 0= DUP
> IF DROP 22 OUT 1+ C! ." >>> ENTER to continue -- any other key to end <<<"
> CR KEY 13 = 0= 0 OUT C! THEN ( OUT-lo "reset" because of the message )
> OR
> UNTIL DROP ;
>
>(yes, a few „magic numbers”, but no big deal; 22 is the lines' count)

Nevertheless valuable. I just tried out fpc and was pleasantly surprised
of its interactive quality and its build in documentation.

WORDS should display the words in recent last order such as in ciforth.
You are interested in the words you defined yourself.

It is trivial, leave a sentinel 0 on the stack
: WORDS
0 'NOOP CONTEXT @ FOR-WORDS \ Leave the addresses of all the words
BEGIN DUP WHILE ID. REPEAT DROP \ Now print the id giving its address.
;

I can't believe that nobody has copied this.
Note that it is less code than your solution, and in a system you
can subtract the original code that WORDS occupied for a total
expense of half a dozen WOC and no extra headers.

>--
>Z.

Groetjes Albert
--
Don't praise the day before the evening. One swallow doesn't make spring.
You must not say "hey" before you have crossed the bridge. Don't sell the
hide of the bear until you shot it. Better one bird in the hand than ten in
the air. First gain is a cat spinning. - the Wise from Antrim -

> WORDS should display the words in recent last order such as in ciforth.
> You are interested in the words you defined yourself.

This has a rationale behind it exactly when WORDS isn't paged. When
it's paged you'll find your just-defined words on the very beginning,
and not running away at light speed.

> It is trivial, leave a sentinel 0 on the stack
> : WORDS
> 0 'NOOP CONTEXT @ FOR-WORDS Leave the addresses of all the words
> BEGIN DUP WHILE ID. REPEAT DROP Now print the id giving its address.
> ;

> I can't believe that nobody has copied this.
> Note that it is less code than your solution, and in a system you
> can subtract the original code that WORDS occupied for a total
> expense of half a dozen WOC and no extra headers.

It's still not paged, unfortunately. What if I want to browse
the „initial” words? In Linux' environment I can use scrollback
buffer (but not always! Kudos to Mark Manning for implementing
his WORDS as „paged”) fortunately — but in MS-DOS (or similar)
there's a problem. The simple solution is: just to split WORDS'
output into separate pages. In such situation the „old” usual
list order is even better: you'll find your words immediately,
without a need to get to the last screen
--
Z.

In article <2232c750331a17c5160fa56c6bc99f2a@news.novabbs.com>,
<SNIP>
>It's still not paged, unfortunately. What if I want to browse
>the „initial” words? In Linux' environment I can use scrollback
>buffer (but not always! Kudos to Mark Manning for implementing
>his WORDS as „paged”) fortunately — but in MS-DOS (or similar)
>there's a problem. The simple solution is: just to split WORDS'
>output into separate pages. In such situation the „old” usual
>list order is even better: you'll find your words immediately,
>without a need to get to the last screen

Of corse you have a separate window with the Forth words ordered
by initial letter, as well as ordered by wordset in html where
you can click on ALSO's if something is not clear.

>--
>Z.

Groetjes Albert
--
Don't praise the day before the evening. One swallow doesn't make spring.
You must not say "hey" before you have crossed the bridge. Don't sell the
hide of the bear until you shot it. Better one bird in the hand than ten in
the air. First gain is a cat spinning. - the Wise from Antrim -

On 12/23/23 2:51 PM, JKN wrote:
> [...]
>
> possibly because a better way to do it would be to decouple the source
> of the character stream (ie WORDS), and the paging mechanism, so that
> the latter could be used for any input stream. In a similar way to how
> in Unix/Linux you can 'pipe' a stream to 'more' to get paged output.
>
> (I would hope that at least some Forths work like this, but I confess
> I have no idea if this is true)

\ Portable Forth Environment 0.33.73w (Aug 21 2021 09:34:46)
Copyright (C) Dirk Uwe Zoller 1993 - 1995.
Copyright (C) Tektronix, Inc. 1998 - 2003.
Copyright (C) Guido U. Draheim 2005 - 2008.
ANS/ffa ITC Forth - Please enter LICENSE and WARRANTY.
Running on x86_64 darwin20.6.0 - to quit say BYE. ok
words
C K-SHIFT-MASK p EKEY>FKEY C K-SHIFT-F10
C K-SHIFT-F9 C K-SHIFT-F8 C K-SHIFT-F7
C K-SHIFT-F6 C K-SHIFT-F5 C K-SHIFT-F4
C K-SHIFT-F3 C K-SHIFT-F2 C K-SHIFT-F1
....
p EXITCODE p ARGV p ARGC
p APPLICATION p EDIT-BLOCKFILE p HELP
p STANDARD-I/O p (TYPE) p (KEY)
p (EXPECT) p (EMIT) p *TYPE*
p *KEY* p *EXPECT* p *EMIT*
more?

Pressing return gets you another line, and the space bar gets you
another page. I don't remember the significance of the C and p
prefixes, which is also not explained by

help words
/usr/local/include/pfe/tools-ext.h:
/** WORDS ( -- )
* uses CONTEXT and lists the words defined in that vocabulary.
* usually the vocabulary to list is named directly in before.
example:
FORTH WORDS or LOADED WORDS
*/
ok

At least the second line hints at which source file to look in to find
that out, namely, tools-ext.c.

-- David W.

"David N. Williams" <williams@umich.edu> writes:
>\ Portable Forth Environment 0.33.73w (Aug 21 2021 09:34:46)
>Copyright (C) Dirk Uwe Zoller 1993 - 1995.
>Copyright (C) Tektronix, Inc. 1998 - 2003.
>Copyright (C) Guido U. Draheim 2005 - 2008.
>ANS/ffa ITC Forth - Please enter LICENSE and WARRANTY.
>Running on x86_64 darwin20.6.0 - to quit say BYE. ok
>words
>C K-SHIFT-MASK p EKEY>FKEY C K-SHIFT-F10
>C K-SHIFT-F9 C K-SHIFT-F8 C K-SHIFT-F7
>C K-SHIFT-F6 C K-SHIFT-F5 C K-SHIFT-F4
>C K-SHIFT-F3 C K-SHIFT-F2 C K-SHIFT-F1
>...
>p EXITCODE p ARGV p ARGC
>p APPLICATION p EDIT-BLOCKFILE p HELP
>p STANDARD-I/O p (TYPE) p (KEY)
>p (EXPECT) p (EMIT) p *TYPE*
>p *KEY* p *EXPECT* p *EMIT*
....
>I don't remember the significance of the C and p
>prefixes

Looking at these words, my guess is:

C: constant
p: primitive

- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: https://forth-standard.org/
EuroForth 2023: https://euro.theforth.net/2023

I decoupled the paging with switches -PAGE to enable paging and +PAGE to
disable paging anything being writing to the display. Just a matter of having
a CR that counts.
Linux provides the scrollback. If on DOS, would implement plan B.
Cross that bridge when come to it but my first choice would not be to put
paging into words.

zbigniew2011@gmail.com (LIT) writes:
>It doesn't seem to be that difficult to create "paged" WORDS

In Gforth, if you want the output of words to be paged, you

require more.fs

I find it cool that this can be added in a Forth system without having
to change every word that outputs multiple lines.

One other "better WORDS" is Martin Bitter's MWORDS. This lists all
words that contain a certain string. You can find it on
<http://theforth.net/package/mwords>.

- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: https://forth-standard.org/
EuroForth 2023: https://euro.theforth.net/2023

On 25/12/2023 2:25 am, Anton Ertl wrote:
> zbigniew2011@gmail.com (LIT) writes:
>> It doesn't seem to be that difficult to create "paged" WORDS
>
> In Gforth, if you want the output of words to be paged, you
>
> require more.fs
>
> I find it cool that this can be added in a Forth system without having
> to change every word that outputs multiple lines.
> ...

That would get very annoying, very quickly - no? Paging poses a similar
problem in that the user is required to enter a key to progress.

dxf <dxforth@gmail.com> writes:
>On 25/12/2023 2:25 am, Anton Ertl wrote:
>> zbigniew2011@gmail.com (LIT) writes:
>>> It doesn't seem to be that difficult to create "paged" WORDS
>>
>> In Gforth, if you want the output of words to be paged, you
>>
>> require more.fs
>>
>> I find it cool that this can be added in a Forth system without having
>> to change every word that outputs multiple lines.
>> ...
>
>That would get very annoying, very quickly - no? Paging poses a similar
>problem in that the user is required to enter a key to progress.

That's why more.fs is not loaded by default. Some people work in an
environment where it is useful, most have scroll-back support and
don't need it.

- anton
--
M. Anton Ertl http://www.complang.tuwien.ac.at/anton/home.html
comp.lang.forth FAQs: http://www.complang.tuwien.ac.at/forth/faq/toc.html
New standard: https://forth-standard.org/
EuroForth 2023: https://euro.theforth.net/2023

>>> I find it cool that this can be added in a Forth system without having
>>> to change every word that outputs multiple lines.

I wrote just about WORDS.

Yes, such universal solution is of advantage.

> That's why more.fs is not loaded by default. Some people work in an
> environment where it is useful, most have scroll-back support and
> don't need it.

The problem with 'more' in Gforth is, that it's asking
a question 'more?' — but ignores my answer, be it 'N', 'Q',
'X', 'Esc' or any other asnwer of the type 'no, thanks'.
It gives me no choice — while the solution I pasted allows
to give up after the first screen is shown (or after any next part)
--
Z.

> That would get very annoying, very quickly - no? Paging poses a similar
> problem in that the user is required to enter a key to progress.

Option like 'C' (for 'continuous output' / no paging
at the moment) would be helpful in case the user wants
to suspend paging switched on by default
--
Z.

On 12/24/23 10:23 AM, Anton Ertl wrote:
> "David N. Williams" <williams@umich.edu> writes:
>> \ Portable Forth Environment 0.33.73w (Aug 21 2021 09:34:46)
>> Copyright (C) Dirk Uwe Zoller 1993 - 1995.
>> Copyright (C) Tektronix, Inc. 1998 - 2003.
>> Copyright (C) Guido U. Draheim 2005 - 2008.
>> ANS/ffa ITC Forth - Please enter LICENSE and WARRANTY.
>> Running on x86_64 darwin20.6.0 - to quit say BYE. ok
>> words
>> C K-SHIFT-MASK p EKEY>FKEY C K-SHIFT-F10
>> C K-SHIFT-F9 C K-SHIFT-F8 C K-SHIFT-F7
>> C K-SHIFT-F6 C K-SHIFT-F5 C K-SHIFT-F4
>> C K-SHIFT-F3 C K-SHIFT-F2 C K-SHIFT-F1
>> ...
>> p EXITCODE p ARGV p ARGC
>> p APPLICATION p EDIT-BLOCKFILE p HELP
>> p STANDARD-I/O p (TYPE) p (KEY)
>> p (EXPECT) p (EMIT) p *TYPE*
>> p *KEY* p *EXPECT* p *EMIT*
> ...
>> I don't remember the significance of the C and p
>> prefixes
>
> Looking at these words, my guess is:
>
> C: constant
> p: primitive

That sounds right. Thanks!

-- David W.

On 25/12/2023 3:46 am, Anton Ertl wrote:
> dxf <dxforth@gmail.com> writes:
>> On 25/12/2023 2:25 am, Anton Ertl wrote:
>>> zbigniew2011@gmail.com (LIT) writes:
>>>> It doesn't seem to be that difficult to create "paged" WORDS
>>>
>>> In Gforth, if you want the output of words to be paged, you
>>>
>>> require more.fs
>>>
>>> I find it cool that this can be added in a Forth system without having
>>> to change every word that outputs multiple lines.
>>> ...
>>
>> That would get very annoying, very quickly - no? Paging poses a similar
>> problem in that the user is required to enter a key to progress.
>
> That's why more.fs is not loaded by default. Some people work in an
> environment where it is useful, most have scroll-back support and
> don't need it.

WRT MORE one should be able to type:

more words

and have it limited to the target. A New Year's challenge?