Built FLOAD for Toad the retro Forth. I didn't
install it as a block file but left it as a
sequential file and pasted it into Toad's input.
So this begs the question, if sequential files
can just be pasted in then is FLOAD really needed?
Whatever.
FLOAD currently just uses one of the block buffers
for its input and SCAN and SPLIT finds a line and
puts its address in TIB and calls INTERPRET.
FLOAD is using what's available without any change
on their part. After all the file is processed,
TIB is reset back to S0 and IN is restored to
what is was so command line picks up where it left
off.
--
me

On 9/08/2023 7:51 am, S Jack wrote:
> Built FLOAD for Toad the retro Forth. I didn't
> install it as a block file but left it as a
> sequential file and pasted it into Toad's input.
> So this begs the question, if sequential files
> can just be pasted in then is FLOAD really needed?

Do you want to nest source to several levels?

On Wednesday, August 9, 2023 at 8:59:40 PM UTC-5, dxforth wrote:
> On 9/08/2023 7:51 am, S Jack wrote:
>
> Do you want to nest source to several levels?

No, I don't want to but the question should be do I need
to? It's somewhat primitive but it's workable. I was and
still am in the mode of not implementing features until I
feel the pain. Frog, Toad's predecessor and more advanced
has yet to implement includes (meaning can have up to some number, eight,
of nestings). With Frog I got some nesting
or maybe better to say concurrent loading by using block
loads along with FLOAD. All sequential files that needed
to be loaded for a given task were loaded with FLOADs
in a block file using LOAD. A file being loaded by FLOAD
could also LOAD codes from block files but of course
if block file also did a FLOAD it would terminate the
initial FLOAD (because only one file buffer). But I could
get around this somewhat with FLOAD restarts. A file
would start with FLOAD then found it needed another
sequential file it would execute a marker, load the
needed file, set the marker and restart the original file.
Would want to have all the NEEDS and REQUIRES up front.
In most cases never had more then 3 needed files.
Granted, it's not as smooth as having includes but it's
workable.
Toad by the way now allocates its own buffer so it
can work along with the block files.
--
me

In article <19a653ac-e3f6-4336-af1f-b0d2f4fdc8f2n@googlegroups.com>,
S Jack <sdwjack69@gmail.com> wrote:
>On Wednesday, August 9, 2023 at 8:59:40 PM UTC-5, dxforth wrote:
>> On 9/08/2023 7:51 am, S Jack wrote:
>>
>> Do you want to nest source to several levels?
>
>No, I don't want to but the question should be do I need
>to? It's somewhat primitive but it's workable. I was and
>still am in the mode of not implementing features until I
>feel the pain. Frog, Toad's predecessor and more advanced
>has yet to implement includes (meaning can have up to some number, eight,
>of nestings). With Frog I got some nesting
>or maybe better to say concurrent loading by using block
>loads along with FLOAD. All sequential files that needed
>to be loaded for a given task were loaded with FLOADs
>in a block file using LOAD. A file being loaded by FLOAD
>could also LOAD codes from block files but of course
>if block file also did a FLOAD it would terminate the
>initial FLOAD (because only one file buffer). But I could
>get around this somewhat with FLOAD restarts. A file
>would start with FLOAD then found it needed another
>sequential file it would execute a marker, load the
>needed file, set the marker and restart the original file.
>Would want to have all the NEEDS and REQUIRES up front.
>In most cases never had more then 3 needed files.
>Granted, it's not as smooth as having includes but it's
>workable.
>Toad by the way now allocates its own buffer so it
>can work along with the block files.

ciforth's take on this:

Blocks are used to have small facilities (3 blocks)
Block loading can be nested, but the supervisory block is locked.
This is contrary to the implementation suggested by the standard.
The complicated BLK/BLOCK/SRC manipulation (ISO94) is circumvented
without making the programs itself unstandard.

Files are slurped in normal programs. I have added an INCLUDE-FILE
that reads files buffered. The block system consists of 1 k buffers
and they are use for this. The incremental code is minute, assuming
the code to implement the READ-LINE standard function is present.

>--
>me
--
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 Friday, August 11, 2023 at 6:45:21 AM UTC-5, none albert wrote:

"Peering into the Abyss" words which came to mind when I opened the
man page for GCC. First time I tried to save Toad image it failed.
Looking into the image found a huge empty gap between the image header
and the start of code (Yes Albert, had my save searched for the magic
it would have worked, but then I wouldn't have known of the gap).
The little bit of GCC parameters looked ok; fixed it when I gave the text
section a good ending. I guess GCC lets you know it doesn't like something
by leaving holes in you image. Think some HTML tools do that.

Toad is a done deal; has a nice assortment of baseline functions.

:) toady VERSION "LIB 0 0 (LINE) CR TYPE CR .." "CR BYE"

Toad-Forth 1.0.B
XYZZY TOADLIB.SCR TOAD-FORTH LIBRARY
CURRENT and CONTEXT are FORTH Base: 10 Latest: SYSSAVE
:)

It can execute command line arguments, load sequential files,
evaluate strings, allocate buffers, make system calls, and save
its image. It an assortment of extensions for unsigned math,
doubles, stack operators, search and compare, shifts and other
odds and ends. All extensions are high level words.

It has USRINIT in COLD to initialize any user code on startup.
Currently it allocates wanted buffers e.g. string input ring buffer,
fload buffer etc. The only pre-allocated buffers are the block buffers
that's part of the Fig model.

Have ton of code that could be imported but no plan to do much;
Toad is to serve for base level evaluation.

Frog was Perseverance, Toad is Artemis, on the drawing board is
MU-Forth (mutated toad) for the long voyage.
Where Toad is pure Fig with extensions MU will be an evolved FIG
to the point of having no Fig appearance and little in Fig
behavior.

Toad is so much fun that MU may never get off the paper.
--
me

On Monday, August 14, 2023 at 6:00:39 PM UTC-5, S Jack wrote:
Toad gets recognition.
:) toady

Toad-Forth 1.0.B
FLOAD f/recog.f OK
FLOAD f/sml.f OK
"rcdg 'CODE APPEND' f/misc.f > /tmp/foo" /O
OK
FLOAD /tmp/foo OK
LATEST ID. APPEND OK
"rcdg APPEND f/misc.f" /O
[r] APPEND ( s1 s2 a -- s3 )
-- Append string s1 to sting s2 placing result at address a.
-- Return result string s3 (which is address a)
--.
OK
&WORLD& %HELLO % PAD APPEND CR TELL
HELLO WORLD OK
BYE :)
--
me

( Some Toad UI )

13 -LIST
SCR# 13
0 ( TOADY+ )
1 FLOAD f/sml.f ( SIMPLE MARKUP LANGUAGE )
2 FLOAD f/recog.f ( INTERPRET WITH RECOGNIZER )
3 FLOAD f/mpad.f ( TEXT OUTPUT BUFFER )
4 FLOAD f/env.f ( ENVIRONMENT )
5 FLOAD f/misc.f ( ASSORMENT OF EXTENSIONS )
6 FLOAD f/display.f ( TERMINAL SCREEN DISPLAY )
7 FLOAD f/rtc.f ( BackForth )
8 FLOAD f/bdisplay.f ( BACKFORTH DISPLAY )
9 ;S OK
"rcdi f/bdisplay.f" CR /O

[#] bdisplay.f -- Back Display
[s] Word listing
[r] const FAUXHDR
[r] CODE FAUXHDR
[r] BNAMES ( lfa ->nfa -- )
[r] CODE BNAMES
[r] //SEEN ( nfa -- nfa | recoil )
[r] CODE //SEEN
[r] NAMEZ ( nfa -- nfa u )
[r] CODE NAMEZ
[r] .NAME ( nfa u -- )
[r] CODE .NAME
[r] LSW
[r] CODE LSW
[r] BVOCNAME ( nfa 1 -- ) ( cfa 2 > cfa -- )
[r] CODE BVOCNAME
[r] BVOCWALK ( >nfa 1 -- ) ( >cfa 2 --)
[r] CODE BVOCWALK
[r] LSWA
[r] CODE LSWA
[#] //
OK
"rcdg 'LSW' f/bdisplay.f" CR /O

[r] LSW
-- Print names of words in context vocabulary
[r] CODE LSW
: LSW CR PWV MAXCOLUMNS @ OUT !
CONTEXT @ BNAMES //SEEN NAMEZ .NAME ;
--.
[r] LSWA
-- Print names of words in all vocabularies
[r] CODE LSWA
: LSWA MAXCOLUMNS @ OUT ! BVOCWALK BVOCNAME BNAMES
//SEEN NAMEZ .NAME ;
--.
OK
LSW
CURRENT and CONTEXT are WRK
LSWA BVOCWALK BVOCNAME LSW .NAME NAMEZ //SEEN BNAMES FAUXHDR OK
--
me