Re: Printed representation of (char 0) ?

[Thorsten Jolitz]

Hi Alex and beneroth,
there are system interfaces that don't handle binary data and send them as
hex strings to middleware for conversion ...
All I wanted was a quick way to cross check that conversion with PicoLisp,
and that was really easy, except that NUL problem, solved now.
But very interesting discussion, I'm playing around with rd and pr right
now.
Thanks for the input
Cheers
Thorsten

Am Mi., 14. Feb. 2024 um 08:04 Uhr schrieb Alexander Burger <
picolisp@software-lab.de>:

> Hi Thorsten,
>
> > I wonder if there actually is a way to directly print ^@ in PicoLisp for
> a
> > "non-printable" hex "00", instead of NIL?
>
> As we see from the previous discussion, this is not an issue of
> printability.
> Other control characters may also be non-printable. It is an issue of
> binary data vs. symbol names.
>
> But you can of course print a caret and an at-mark instead of NIL
>
>(prin (or (something) "\^@"))
>
>
> > Wrt the application, I just have to deal with fixed length hex strings
> (!)
> > where the values at certain offsets carry semantics, conversions are
> done,
> > and it's crucial that values stay in that position, the NUL values
> matter.
>
> Yes, but why do you need to convert it to a string? I would process these
> data
> all exclusively as a list of numbers, and do the final printing explicitly
> (if
> needed at all). This printing may print '0' as "\^@", and also take care of
> other control characters and non-printable stuff.
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Printed representation of (char 0) ?

[Thorsten Jolitz]

Hi Alex, Thomas,
thanks for your input, this is actually what I was looking for :
[image: image.png]
or even better:
[image: image.png]
I wonder if there actually is a way to directly print ^@ in PicoLisp for a
"non-printable" hex "00", instead of NIL?
Wrt the application, I just have to deal with fixed length hex strings (!)
where the values at certain offsets carry semantics, conversions are done,
and it's crucial that values stay in that position, the NUL values matter.
And I don't want to write a PicoLisp application for this, I just wanted an
easy way to produce the expected conversion result in PicoLisp as a
reference for comparison, and the above solution is fine for that.

Am Di., 13. Feb. 2024 um 09:09 Uhr schrieb Alexander Burger <
picolisp@software-lab.de>:

> Hi Thorsten,
>
> > But shouldn't hex 23232424 print to something like ##^N^N$$ instead
> of
> > ##$$ ?
>
> The problem is that you try to handle binary data as symbols. This is not
> a good
> idea. Binary data are numbers.
>
> First of all, do you really have a hex message? Where does
> it come from? Normally I would expect a list of numbers
> as obtained with e.g.
>
>(make (do 96 (link (rd 1
>
> If it is really a hexadecimal string, you can obtain the list
> of numbers with
>
>: (make (for (L (chop "23232424") (cut 2 'L)) (link (hex @
>-> (35 35 0 0 36 36)
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Printed representation of (char 0) ?

[Thorsten Jolitz]

Hi Alex,
But shouldn't hex 23232424 print to something like ##^N^N$$ instead of
##$$ ?
So the printed ASCII string (as char) carries all the information from the
hex string, and can be converted back to the exact same hex string?
At least in some special cases when it's needed?

Alexander Burger  schrieb am Di., 13. Feb. 2024,
07:56:

> Hi Thorsten,
>
> > it's been some time .. ;-)
>
> Welcome back! :)
>
>
> > I'm playing around a bit with hex<->ascii conversion in PicoLisp, and I
> > have the problem that (char 0) = NIL
> >
> >  (hex "00")
> > -> 0
> > : (char (hex "00"))
> > -> NIL
>
> This is correct.
>
> 'char' converts a number to a (transient) symbol here.
>
> A symbol's name is a string, a null-terminated sequence of UTF-8
> characters. In
> case of 'char', this string has a single character and a terminating null
> byte.
> This is the same as in other languages like C.
>
> So the number 65 gives a symbol "A":
>
>: (char 65)
>-> "A"
>
> But what happens with 0?
>
> It gives an empty string, i.e. a null-byte
>
>: (char 0)
>-> NIL
>
> and an empty string in PicoLisp is NIL.
>
>: ""
>-> NIL
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Printed representation of (char 0) ?

[Thorsten Jolitz]

Ok I understand that.
But when I have a hexadecimal message string with fixed length, and the
positions inside the string carry semantics? A certain value in a certain
position has a meaning?

Say fixed length is 96 like in the example above, and 2323 ist two start
chars, and then 01 or 02 is a handshake.

And this message has to be translated to ASCII to be understood by a third
party, that sends ASCII answers back, that has to be translated to hex.

This does not work when a round-trip conversion of 232301 results in 23231.

Tomas Hlavaty  schrieb am Di., 13. Feb. 2024,
07:49:

> On Tue 13 Feb 2024 at 00:25, Thorsten Jolitz 
> wrote:
> > I would like to achieve a roundtrip like this:
>
> why?
>
> NUL is often a string sentinel value
> so trying to use it as a character value
> will lead to issues
> do not do that
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Printed representation of (char 0) ?

[Thorsten Jolitz]

I would like to achieve a roundtrip like this:
: (hex (char (char (hex "23"]
-> "23"

but not like this:
: (hex (char (char (hex "01"]
-> "1"
: (hex (char (char (hex "00"]
-> "0"

It should be possible to do this, so that W = X in the end:
: (setq X
"23230200010001000F00323032342D30322D303931302D35342D303520202424")
: (length X)
-> 96
: (setq Y (chop X))
: (length Y)
-> 96
: (setq W (pack (mapcar '((A) (hex (char A))) (make (while Y (link (char
(hex (pack (cut 2 'Y)]
->
"232320001010F000323032342D30322D303931302D35342D303520202424"
: (length W)
-> 72

Am Mo., 12. Feb. 2024 um 23:56 Uhr schrieb Tomas Hlavaty <
picolisp@software-lab.de>:

> On Mon 12 Feb 2024 at 23:25, Thorsten Jolitz 
> wrote:
> > Shouldn't the (char 0) representation print to something
> > like ^N or so too, like (char 1), (char 2) etc?
>
> ^@
>
> what are you trying to achieve?
>
> why not use base64, for example?
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Printed representation of (char 0) ?

[Thorsten Jolitz]

Ah sorry, the moment I sent the question, I figured out the answer :
this is just a side effect of the final (pack ..).

When using (str ...) I get what I want:
:  (str (make (while Y (link (char (hex (pack (cut 2 'Y)]
-> "\"#\" \"#\" \"\^B\" NIL NIL NIL \"\^A\" NIL \"\^A\" NIL \"\^O\" NIL NIL
NIL NIL NIL NIL NIL \"2\" \"0\" \"2\" \"4\" \"-\" \"0\" \"2\" \"-\" \"0\"
\"9\" \"1\" \"0\" \"-\" \"5\" \"4\" \"-\" \"0\" \"5\" \" \" \" \" NIL NIL
NIL NIL NIL NIL NIL NIL \"$\" \"$\""

although somehow I still think it would be fine to have 'pack that prints
the NIL too, in this special case:
: (pack (make (while Y (link (char (hex (pack (cut 2 'Y)]
-> "##^B^A^A^O2024-02-0910-54-05  $$"

Am Mo., 12. Feb. 2024 um 23:25 Uhr schrieb Thorsten Jolitz <
tjol...@gmail.com>:

> Hello List,
> it's been some time .. ;-)
>
> I'm playing around a bit with hex<->ascii conversion in PicoLisp, and I
> have the problem that (char 0) = NIL
>
>  (hex "00")
> -> 0
> : (char (hex "00"))
> -> NIL
>
>  and disappears from the resulting ascii string.
>
> : (setq X
> "23230200010001000F00323032342D30322D303931302D35342D303520202424")
> : (setq Y (chop X))
> : (pack (make (while Y (link (char (hex (pack (cut 2 'Y)]
> -> "##^B^A^A^O2024-02-0910-54-05  $$"
>
> So how would it be possible to convert the ascii string back to the
> original hex string, when all the "00" are gone in the ascii
> representation? Shouldn't the (char 0) representation print to something
> like ^N or so too, like (char 1), (char 2) etc?
>
> When I try this online converter, the "00" are maintained as blanks:
> Hex to ASCII Text String Converter (rapidtables.com)
> <https://www.rapidtables.com/convert/number/hex-to-ascii.html>
> [image: image.png]
>
> and here is another representation that seems to enable the hex -> ascii
> -> hex conversion, getting exactly the same hex string in the end as in the
> beginning (not exactly the same conversion as above! just an example how it
> might look) :
> [image: image.png]
> Cheers
> Thorsten
>

Printed representation of (char 0) ?

[Thorsten Jolitz]

Hello List,
it's been some time .. ;-)

I'm playing around a bit with hex<->ascii conversion in PicoLisp, and I
have the problem that (char 0) = NIL

 (hex "00")
-> 0
: (char (hex "00"))
-> NIL

 and disappears from the resulting ascii string.

: (setq X
"23230200010001000F00323032342D30322D303931302D35342D303520202424")
: (setq Y (chop X))
: (pack (make (while Y (link (char (hex (pack (cut 2 'Y)]
-> "##^B^A^A^O2024-02-0910-54-05  $$"

So how would it be possible to convert the ascii string back to the
original hex string, when all the "00" are gone in the ascii
representation? Shouldn't the (char 0) representation print to something
like ^N or so too, like (char 1), (char 2) etc?

When I try this online converter, the "00" are maintained as blanks:
Hex to ASCII Text String Converter (rapidtables.com)

[image: image.png]

and here is another representation that seems to enable the hex -> ascii ->
hex conversion, getting exactly the same hex string in the end as in the
beginning (not exactly the same conversion as above! just an example how it
might look) :
[image: image.png]
Cheers
Thorsten

Re: org-mode with picolisp

[Thorsten Jolitz]

Hi List,
I did some tests now with PIL21 in Org-mode on Emacs doom.

I did not touch picolisp-mode.el or inferior-picolisp.el from my repo yet
tj64/picolisp-mode: GNU Emacs mode for PicoLisp programming (github.com)
<https://github.com/tj64/picolisp-mode>

However, and to my surprise, I had to do some changes to the ob-picolisp.el
from my repo, they are committed to the repo file:
tj64/ob-picolisp: Adding PicoLisp to Org Babel, the multi-language
programming environment included in Emacs Org Mode. (github.com)
<https://github.com/tj64/ob-picolisp>

My version of ob-picolisp.el had some additional functionality, while the
version I found in the former org-contrib repo had some cleanup in the
comments and some improvements in the code:
emacsPackages/ob-picolisp.el at master · NotBrianZach/emacsPackages
(github.com)
<https://github.com/NotBrianZach/emacsPackages/blob/master/org-plus-contrib-20181210/ob-picolisp.el>
I kept the additional functionality in my version, but copied all the
changes from the other version.

Then I tested in Emacs Doom with PIL21 installed, and it worked (in
general), did not test the whole functionality, see PS.
Cheers
Thorsten

PS
#+title: Ob Picolisp Test1

#+begin_src emacs-lisp
(+ 3 4)
#+end_src

#+RESULTS:
: 7


#+begin_src picolisp :results value
 (+ 2 5)
#+end_src

#+RESULTS:
: 7

#+begin_src picolisp :results value :session test1
 (setq A 12)
 (setq B 24)
#+end_src

#+RESULTS:
: No

#+begin_src picolisp :results output :session test1
 (if (> A B)
(print A)
(print B) )
#+end_src

#+RESULTS:
:   24




Am Mi., 30. Nov. 2022 um 23:14 Uhr schrieb Thorsten Jolitz <
tjol...@gmail.com>:

> Ah no, there was another file with 2 versions, led.l and eled.l for line
> editing vim or emacs style.
> Like Alex said, edit.l and eedit.l were used for symbol editing 
>
> Am Mi., 30. Nov. 2022 um 23:06 Uhr schrieb Thorsten Jolitz <
> tjol...@gmail.com>:
>
>> Yes, and I added eedit.l to picolisp iirc, so one could chose between vim
>> and emacs keybindings in the repl line editor.
>> Inferior-picolisp.el is responsable for the picolisp repl in Emacs, so
>> when starting it the user could choose between the 2 editing styles.
>> You don't necessary see that repl in your org-mode file with a picolisp
>> src-block, but it's there in the background, therefore the error.
>>
>> I think they are not needed anymore. I cannot test it right now, but when
>> those lines are outcommented in the .el files, it might work (then using
>> readline library for line editing in the repl in emacs, just like in
>> standalone picolisp).
>>
>> Am Mi., 30. Nov. 2022 um 22:48 Uhr schrieb Alexander Burger <
>> a...@software-lab.de>:
>>
>>> On Wed, Nov 30, 2022 at 03:25:31PM -0600, Galaxy Being wrote:
>>> > The old picolisp file structure had ~/.../lib/edit.l and .../lib/el
>>> which
>>> > is no longer the case. That functionality went somewhere else? And yet
>>> both
>>>
>>> lib/edit.l was an in-memory editor for symbols in pil64, which called Vim
>>> internally. It is now obsolete, as this functionality is handled by Vip
>>> (the
>>> built-in Vi-style editor) in pil21.
>>>
>>> It is strange that lib/edit.l is loaded at all. Org-Mode should not care
>>> about
>>> it, and lib/edit.l is loaded at start-up by pil64, but never by pil21.
>>>
>>> lib/el/ is no longer part of the distro in pil21 to keep things small.
>>> It should
>>> probably better be made a separate tarball or repo.
>>>
>>> ☺/ A!ex
>>>
>>> --
>>> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>>>
>>

Re: org-mode with picolisp

[Thorsten Jolitz]

Yes, and I added eedit.l to picolisp iirc, so one could chose between vim
and emacs keybindings in the repl line editor.
Inferior-picolisp.el is responsable for the picolisp repl in Emacs, so when
starting it the user could choose between the 2 editing styles.
You don't necessary see that repl in your org-mode file with a picolisp
src-block, but it's there in the background, therefore the error.

I think they are not needed anymore. I cannot test it right now, but when
those lines are outcommented in the .el files, it might work (then using
readline library for line editing in the repl in emacs, just like in
standalone picolisp).

Am Mi., 30. Nov. 2022 um 22:48 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> On Wed, Nov 30, 2022 at 03:25:31PM -0600, Galaxy Being wrote:
> > The old picolisp file structure had ~/.../lib/edit.l and .../lib/el which
> > is no longer the case. That functionality went somewhere else? And yet
> both
>
> lib/edit.l was an in-memory editor for symbols in pil64, which called Vim
> internally. It is now obsolete, as this functionality is handled by Vip
> (the
> built-in Vi-style editor) in pil21.
>
> It is strange that lib/edit.l is loaded at all. Org-Mode should not care
> about
> it, and lib/edit.l is loaded at start-up by pil64, but never by pil21.
>
> lib/el/ is no longer part of the distro in pil21 to keep things small. It
> should
> probably better be made a separate tarball or repo.
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: org-mode with picolisp

[Thorsten Jolitz]

Hi,
I'm more or less responsable for PicoLisp in org-mode.

Before diving any deeper, note the readme of GitHub -
flexibeast/plisp-mode: PicoLisp support for Emacs
 :
"The plisp-mode in this package has been built from scratch, and is not
based on, nor connected with, the PicoLisp support for Emacs provided in the
PicoLisp distribution , or the more
recently updated version of that support
. At this stage, the main advantages
provided by this package are [...]"

So you are actually in unsupported territory there, the original
picolisp-mode which was used for ob-picolisp is this:
GitHub - tj64/picolisp-mode: GNU Emacs mode for PicoLisp programming


I don't think that ob-picolisp cares about pil64 or pil21, most likely the
problems you see are related to plisp-mode itself.
Maybe try the original mode standalone, and if that works, try it with
org-mode.

plisp-mode was developed by someone else, and I don't know if it was ever
used "in production".
Cheers
Thorsten



Am Mi., 30. Nov. 2022 um 16:47 Uhr schrieb Galaxy Being :

> The problem might be with plisp-mode. It gives the same error (on "send
> definition and go" i.e., start REPL and hand code off to it), and I think
> ob-picolisp.el relies on plisp. In any case, you've changed your files
> around and there is no edit.l anymore. Here's my .emacs section for
> picolisp. ;; is commented out
>
> ;; (add-to-list 'load-path "~/opt/picoLisp/lib/el")
> ;;(load "/home/galaxybeing/.emacs.d/modes/tsm.el") ;; Picolisp
> TransientSymbolsMarkup (*Tsm)
> (autoload 'run-picolisp "inferior-picolisp")
> (autoload 'plisp-mode "picolisp" "Major mode for editing Picolisp." t)
>
> (require 'plisp-mode)
>
> (setq picolisp-program-name "/home/galaxybeing/opt/pil21/pil")
> (add-to-list 'auto-mode-alist '("\\.l$" . picolisp-mode))
>
> ;; (add-hook 'picolisp-mode-hook
> ;;(lambda ()
> ;;   (paredit-mode +1) ;; Loads paredit mode automatically
> ;;   (tsm-mode) ;; Enables TSM
> ;;   (define-key picolisp-mode-map (kbd "RET") 'newline-and-indent)
> ;;   (define-key picolisp-mode-map (kbd "C-h")
> 'paredit-backward-delete) ) )
>
> On Wed, Nov 30, 2022 at 1:09 AM Alexander Burger 
> wrote:
>
>> Hi Lawrence,
>>
>> > Having trouble with getting picolisp to work with org-mode Babel code
>> > blocks. I'm guessing it's because I have an out of date ob-picolisp.el
>> > which I had to scrounge on the Internet. (Got one latest 2021.)
>>
>> Sorry, yes. Org-Mode is not ported to Pil21 (yet).
>>
>> Let us investigate a little :)
>>
>> ☺/ A!ex
>>
>> --
>> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>>
>
>
> --
> ⨽
> Lawrence Bottorff
> Grand Marais, MN, USA
> borg...@gmail.com
>

Re: PicoLisp Wiki: Embed Content?

[Thorsten Jolitz]

Hi Alex,
thanks!
And in case it's a personal private wiki one could even outcomment the new
syntax element and eliminate the check for an  schrieb am Do., 13. Okt. 2022, 08:24:

> Hi Thorsten,
>
> > I thought about security too. I even would like to have 2 embed options,
> > one only for iframes, the other for any html (produced by an external
> tool)
>
> This is indeed a potential security risk. A public Wiki can be edited by
> everyone and malicious HTML code might sneak in.
>
> As discussed yesterday, I commented the change in the release and on the
> public
> site. It is useful in environments where edit access is controlled, so it
> can be
> un-commented in such cases.
>
> Released.
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: PicoLisp Wiki: Embed Content?

[Thorsten Jolitz]

Hi Alex,
simplification looks good, will test later.

I thought about security too. I even would like to have 2 embed options,
one only for iframes, the other for any html (produced by an external tool)

Re: PicoLisp Wiki: Embed Content?

[Thorsten Jolitz]

PS
or better a check like this, then if the §{content} does not start with
:

> Hi Alex,
> it's actually extremely easy to embed content in a picolisp wiki file,
> when adding the syntax element from the PS 1 to wiki/lib.l :
> §{ }
> I added a little check for  syntax user cannot insert any kind of code.
>
> With this element its easy to embed youtube videos in wiki pages, they all
> have an embed link (see PS 2)
>
> Would you consider to add something like this in wiki/lib.l ?
> Cheers
> Thorsten
>
>
>
> PS 1
> 
>
> 1 file changed, 5 insertions(+), 1 deletion(-)
>
>
> lib.l | 6
>
> +-
>
>
> modified   lib.l
>
>
> @@ -54,7 +54,7 @@
>
>
> (recur (Nest)
>
>
> (use C
>
>
> (loop
>
>
>-   (ht:Prin (till "^J123456&/!_*+-%~|<@>=\^:$\\#}" T))
>
>
> +   (ht:Prin (till "^J123456&/!_*+-%~|<@>=\^:$\\#§}" T))
>
>
> (NIL (setq C (char)))
>
>
> (T (and Nest (= C "}")))
>
>
> (unless (= C "\\")
>
>
> @@ -159,6 +159,10 @@
>
>
> (renderBlock ht:Prin)) )
>
>
> ("\\" (prin "{"))  # Escaped brace
>
>
> ("#" (renderBlock prog))  # Comment
>
>
> +   ("§"  # iframe
>
>
> +  (let Lnk (till "}" T)
>
>
> +   (when (= " (chop Lnk) " ")
>
> + (prin Lnk) ) )
>
>
> (T (prin C "{")) ) )
>
>
> (T (ht:Prin C)) ) ) ) ) ) ) )
> 
>
> PS 2
> https://www.youtube.com/embed/k7CHDscLREk; title="YouTube video player"
> frameborder="0" allow="accelerometer; autoplay; clipboard-write;
> encrypted-media; gyroscope; picture-in-picture" allowfullscreen>
>
>
> Am So., 17. Juli 2022 um 15:46 Uhr schrieb Alexander Burger <
> a...@software-lab.de>:
>
>> Hi Thorsten,
>>
>> > is there a way to embed content in a PicoLisp Wiki file?
>> > ...
>> > But when sharing e.g. a youtube video, there is the "embed" option, that
>> > gives this iframe:
>> > https://www.youtube.com/embed/xEKHU4zCRpY; title="YouTube video player"
>> > ...
>> > I could not figure out how to include this in a page with the PicoLisp
>> Wiki
>> > Syntax - probably it's not possible?
>>
>> I'm afraid so. I think there is no syntax for that in the markup
>> language. Maybe
>> Erik has an idea?
>>
>> ☺/ A!ex
>>
>> --
>> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>>
>

Re: PicoLisp Wiki: Embed Content?

[Thorsten Jolitz]

Hi Alex,
it's actually extremely easy to embed content in a picolisp wiki file, when
adding the syntax element from the PS 1 to wiki/lib.l :
§{ }
I added a little check for =\^:$\\#}" T))


+   (ht:Prin (till "^J123456&/!_*+-%~|<@>=\^:$\\#§}" T))


(NIL (setq C (char)))


(T (and Nest (= C "}")))


(unless (= C "\\")


@@ -159,6 +159,10 @@


(renderBlock ht:Prin)) )


("\\" (prin "{"))  # Escaped brace


("#" (renderBlock prog))  # Comment


+   ("§"  # iframe


+  (let Lnk (till "}" T)


+   (when (= "https://www.youtube.com/embed/k7CHDscLREk; title="YouTube video player"
frameborder="0" allow="accelerometer; autoplay; clipboard-write;
encrypted-media; gyroscope; picture-in-picture" allowfullscreen>


Am So., 17. Juli 2022 um 15:46 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> Hi Thorsten,
>
> > is there a way to embed content in a PicoLisp Wiki file?
> > ...
> > But when sharing e.g. a youtube video, there is the "embed" option, that
> > gives this iframe:
> > https://www.youtube.com/embed/xEKHU4zCRpY; title="YouTube video player"
> > ...
> > I could not figure out how to include this in a page with the PicoLisp
> Wiki
> > Syntax - probably it's not possible?
>
> I'm afraid so. I think there is no syntax for that in the markup language

PicoLisp Wiki: Embed Content?

[Thorsten Jolitz]

Hi List,
is there a way to embed content in a PicoLisp Wiki file?
I can download the video, add it to the page, and then reference it by
name @{}.

But when sharing e.g. a youtube video, there is the "embed" option, that
gives this iframe:
https://www.youtube.com/embed/xEKHU4zCRpY; title="YouTube video player"
frameborder="0" allow="accelerometer; autoplay; clipboard-write;
encrypted-media; gyroscope; picture-in-picture" allowfullscreen>

I could not figure out how to include this in a page with the PicoLisp Wiki
Syntax - probably it's not possible?
Cheers
Thorsten

Small error in httpGate doc

[Thorsten Jolitz]

Hi Alex,
just tried to build httpGate on WSL2 and got an error:

$ make gate
make: *** No rule to make target 'gate'.  Stop.

First I thought it's some WSL2 peculiarity as usual, but I think it's just
outdated documentation:

The 'httpGate' Proxy Server (software-lab.de)
:
"Building httpGate
Next, go to the 'src' directory in the distribution and run make gate. When
this is done, there should be an httpGate executable in the 'bin'
directory."

Looking into the makefile, it seems now there is a single "all" target, and
httpGate is built together with pil21?
Just to let you know
Cheers
Thorsten

Re: Internal WIKI Links do not work on WSL2 Debian

[Thorsten Jolitz]

Hi Olaf,
*"Did you try 127.0.0.1:5000 <http://127.0.0.1:5000/> instead of
localhost:5000  ?"*
just as a documentation, my experience is that:

1) WSL2 / w3m : all of these work
http://localhost:5000
http://127.0.0.1:5000
http://0.0.0.0:5000

2) WIN10 / edge : only this works
http://localhost:5000

Cheers
Thorsten


Am Mi., 19. Jan. 2022 um 14:43 Uhr schrieb O.Hamann :

> Hi Thorsten, thanks for your solution!
>
> Just to be curious:
> Did you try 127.0.0.1:5000 instead of localhost:5000  ?
> Same pn-replacing effect?
>
> Regards,  Olaf
>
>
> On 19.01.22 00:20, Thorsten Jolitz wrote:
> > Hi List,
> > today I figured out what seems to be the problem here:
> > Somehow the URLs are built with the computername instead of the port:
> > http://0.0.0.0:asuspn/?home
> >
> > Replacing asuspn with 5000 all of these work, even from (say) Edge in
> > Win10, when pil wiki server was started on WSL2:
> >
> > http://localhost:5000/?home
> > http://localhost:5000/?help
> > http://localhost:5000/?*Menu=+0&*Tab=+1&*ID=&*ID=$login
> >
> > with the last one, I can actually login, get a session id, and can
> > navigate in the wiki and in my account.
> > Strange, the port looks ok in the server call, and on Archlinux the URLs
> > are ok too, but on WSL2 the servername instead of port is used to build
> > the URL.
> >
> > $ pil wiki/main.l -main -go +
> > (server (or (format (sys "PORT")) *WikiPort) "!wiki")
> > ! *WikiPort
> > -> 5000
> >
> > Cheers
> > Thorsten
> >
> > PS
> > I start the wiki the normal way
> > $ pil wiki/main.l -main -go +
> > *Socket
> > 1877 = 40895 20654634285828729~
> > : *Socket
> > -> NIL
> > : 1877 * 2022-01-19 00:00:37 admin
> > 1883 = 57451 41629218673747370~
> > :
> >
> >
> >
> > Am Di., 28. Dez. 2021 um 23:42 Uhr schrieb Thorsten Jolitz
> > mailto:tjol...@gmail.com>>:
> >
> > Hi Alex,
> > yes, *ID is NIL when I debug it (see PS1).
> > I'm not sure how to debug (app), but all global vars except *PID
> > seem to be NIL ...
> >
> > I think it's a WSL2 localhost problem, not a bug in PicoLisp.
> > Not too long ago I tried accessing the pil wiki server running on
> > WSL2 from the Win10 Browser (say Edge), and that did not work at all
> > Now it seems they have fixed that ,what is very nice, but still not
> > the same like pure Linux apparently.
> >
> > Checking localhost everything seems normal (see PS 2) , but e.g this
> > article describes a lot of problems:
> > Fixing WSL2 localhost access issue - abdus.dev
> > <https://abdus.dev/posts/fixing-wsl2-localhost-access-issue/>
> >
> > Cheers
> > Thorsten
> >
> > PS 1
> > $  pil wiki/main.l -main -go +
> > (and (== 'login *ID) (app))
> > ! *ID
> > -> NIL
> > ! (setq *ID 123456789)   # naive try
> > -> 123456789
> >
> > # naiv try did not work
> > Can't load -2.html?*Menu=+0&*Tab=+1&*ID=+123456789&*ID=$login
> >
> > PS 2
> > $ ping -a localhost
> > PING localhost (127.0.0.1) 56(84) bytes of data.
> >
> > # type c (peek current url) in W3M : wiki start page
> > http://localhost:8080/
> >
> > # $ sudo cat /etc/hosts
> > 127.0.0.1   localhost
> > 127.0.1.1   xyz.localdomain  xyz
> >
> > Am Di., 28. Dez 2021 um 20:44 Uhr schrieb Alexander Burger
> > mailto:a...@software-lab.de>>:
> >
> > Hi Thorsten, Olaf,
> >
> > > it's a session problem.
> > > Port 8080 does not help, but I cross checked on Archlinux,
> > when I first
> > > click on the Login link, a session prefix is added to the
> > internal links,
> > > and after login, that session prefix is everywhere in the html
> > source.
> > > But on WSL2 that does not work somehow
> >
> > This is indeed strange. This logic should not depend on the
> system.
> >
> > I think the session is started after clicking on "Log in" in
> >
> >(dm (html> . +Doc) ()
> >   (and (== 'login *ID) (app))
> >
> > So somehow this check fails? The *ID value is passed on the URL,
> > perhaps it is
> > destroyed somehow? The browser?
> >
> > ☺/ A!ex
> >
> > --
> > UNSUBSCRIBE: mailto:picolisp@software-lab.de
> > <mailto:picolisp@software-lab.de>?subject=Unsubscribe
> >
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subjectUnsubscribe
>

Re: Internal WIKI Links do not work on WSL2 Debian

[Thorsten Jolitz]

Hi Alex,
yes, $NAME was the computername ("asuspn"), and by setting name to
   $ echo $NAME
   5000
everything works, I even get the wiki page with CSS then like in the real
wiki.
So, problem identified and solved, thanks!
Cheers
Thorsten


Am Mi., 19. Jan. 2022 um 17:56 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> Hi Thorsten,
>
> > I looks actually as designed:
> > in function baseHRef in http.l is this line
> >
> > (or Port (if *SesId *Port *Port1))
> >
> > and *Port1 is the Computername.
>
> Correct, though not the Computername.
>
> IFF the PicoLisp server is started behind a httqGate, the "port" is the
> application name. For example, on picolisp.com, the httpGate config file
> has a
> line
>
>wiki 5000 app /home/app log/wiki pil21/pil wiki/main.l @lib/app.l -main
> patch.l -go -wait +
>
> so that httpGate translates "wiki" in requests to 5000.
>
> The question is how this happens in your setup:
>
> *Port1 is assigned in the 'server' function from the "NAME" environment
> variable. So I suspect $NAME is set in your env to the machine's name, and
> not
> overridden by httpGate. That would explain it!
>
> You could unset NAME in your shell before starting PicoLisp, or (even
> better)
> use httpGate.
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Internal WIKI Links do not work on WSL2 Debian

[Thorsten Jolitz]

Hi Alex, Hi Olaf,
I would not say I found a "solution", just trying to figure out the cause.

I looks actually as designed:
in function baseHRef in http.l is this line

(or Port (if *SesId *Port *Port1))

and *Port1 is the Computername.


So apparently the landing page is build with these URLs with name instead
of port, and only once a session id is acquired, a numeric port is used.

These links are dead on the landing page, but when I enter the URLs  in the
browser, they work, e.g. http://localhost:5000/?help.

Once I enter this in the browser:
http://localhost:5000/?*Menu=+0&*Tab=+1&*ID=&*ID=$login , and login, all
the links work again on the page, since a *'SesId exists and they are build
with the port number (e.g. http://localhost:49651/62212780669042004~?Help)


I wonder how these links

http://localhost:asuspn/?help <http://localhost:5000/?help>

http://localhost:asuspn/?h <http://localhost:5000/?help>ome

are supposed to work before a session id is acquired, but somehow they do
work on normal Linux.

Since (session P H) is called with the Port P (5000), why not use it for
URL construction (instead of *Port1)?


Cheers

Thorsten


PS

(out 2 (prinl *Pid " = " *Port " " *SesId))

1549 = 49651 62212780669042004~

(or "Port" (if *SesId *Port *Port1))

(or "Port" (if *SesId *Port *Port1))

(or "Port" (if *SesId *Port *Port1))

(or "Port" (if *SesId *Port *Port1))

(or *SesId (bye))

: 1549 * 2022-01-19 17:07:16 admin

Am Mi., 19. Jan. 2022 um 14:43 Uhr schrieb O.Hamann :

> Hi Thorsten, thanks for your solution!
>
> Just to be curious:
> Did you try 127.0.0.1:5000 instead of localhost:5000  ?
> Same pn-replacing effect?
>
> Regards,  Olaf
>
>
> On 19.01.22 00:20, Thorsten Jolitz wrote:
> > Hi List,
> > today I figured out what seems to be the problem here:
> > Somehow the URLs are built with the computername instead of the port:
> > http://0.0.0.0:asuspn/?home
> >
> > Replacing asuspn with 5000 all of these work, even from (say) Edge in
> > Win10, when pil wiki server was started on WSL2:
> >
> > http://localhost:5000/?home
> > http://localhost:5000/?help
> > http://localhost:5000/?*Menu=+0&*Tab=+1&*ID=&*ID=$login
> >
> > with the last one, I can actually login, get a session id, and can
> > navigate in the wiki and in my account.
> > Strange, the port looks ok in the server call, and on Archlinux the URLs
> > are ok too, but on WSL2 the servername instead of port is used to build
> > the URL.
> >
> > $ pil wiki/main.l -main -go +
> > (server (or (format (sys "PORT")) *WikiPort) "!wiki")
> > ! *WikiPort
> > -> 5000
> >
> > Cheers
> > Thorsten
> >
> > PS
> > I start the wiki the normal way
> > $ pil wiki/main.l -main -go +
> > *Socket
> > 1877 = 40895 20654634285828729~
> > : *Socket
> > -> NIL
> > : 1877 * 2022-01-19 00:00:37 admin
> > 1883 = 57451 41629218673747370~
> > :
> >
> >
> >
> > Am Di., 28. Dez. 2021 um 23:42 Uhr schrieb Thorsten Jolitz
> > mailto:tjol...@gmail.com>>:
> >
> > Hi Alex,
> > yes, *ID is NIL when I debug it (see PS1).
> > I'm not sure how to debug (app), but all global vars except *PID
> > seem to be NIL ...
> >
> > I think it's a WSL2 localhost problem, not a bug in PicoLisp.
> > Not too long ago I tried accessing the pil wiki server running on
> > WSL2 from the Win10 Browser (say Edge), and that did not work at all
> > Now it seems they have fixed that ,what is very nice, but still not
> > the same like pure Linux apparently.
> >
> > Checking localhost everything seems normal (see PS 2) , but e.g this
> > article describes a lot of problems:
> > Fixing WSL2 localhost access issue - abdus.dev
> > <https://abdus.dev/posts/fixing-wsl2-localhost-access-issue/>
> >
> > Cheers
> > Thorsten
> >
> > PS 1
> > $  pil wiki/main.l -main -go +
> > (and (== 'login *ID) (app))
> > ! *ID
> > -> NIL
> > ! (setq *ID 123456789)   # naive try
> > -> 123456789
> >
> > # naiv try did not work
> > Can't load -2.html?*Menu=+0&*Tab=+1&*ID=+123456789&*ID=$login
> >
> > PS 2
> > $ ping -a localhost
> > PING localhost (127.0.0.1) 56(84) bytes of data.
> >
> > # type c (peek current url) in W3M : wiki start page
> > http://localhost:8080/
> >
> > # $ sudo cat /etc/hosts
> > 127.0.0.1   localhost
> > 1

Re: Internal WIKI Links do not work on WSL2 Debian

[Thorsten Jolitz]

Hi List,
today I figured out what seems to be the problem here:
Somehow the URLs are built with the computername instead of the port:
http://0.0.0.0:asuspn/?home

Replacing asuspn with 5000 all of these work, even from (say) Edge in
Win10, when pil wiki server was started on WSL2:

http://localhost:5000/?home
http://localhost:5000/?help
http://localhost:5000/?*Menu=+0&*Tab=+1&*ID=&*ID=$login

with the last one, I can actually login, get a session id, and can navigate
in the wiki and in my account.
Strange, the port looks ok in the server call, and on Archlinux the URLs
are ok too, but on WSL2 the servername instead of port is used to build the
URL.

$ pil wiki/main.l -main -go +
(server (or (format (sys "PORT")) *WikiPort) "!wiki")
! *WikiPort
-> 5000

Cheers
Thorsten

PS
I start the wiki the normal way
$ pil wiki/main.l -main -go +
*Socket
1877 = 40895 20654634285828729~
: *Socket
-> NIL
: 1877 * 2022-01-19 00:00:37 admin
1883 = 57451 41629218673747370~
:



Am Di., 28. Dez. 2021 um 23:42 Uhr schrieb Thorsten Jolitz <
tjol...@gmail.com>:

> Hi Alex,
> yes, *ID is NIL when I debug it (see PS1).
> I'm not sure how to debug (app), but all global vars except *PID seem to
> be NIL ...
>
> I think it's a WSL2 localhost problem, not a bug in PicoLisp.
> Not too long ago I tried accessing the pil wiki server running on WSL2
> from the Win10 Browser (say Edge), and that did not work at all.
> Now it seems they have fixed that ,what is very nice, but still not the
> same like pure Linux apparently.
>
> Checking localhost everything seems normal (see PS 2) , but e.g this
> article describes a lot of problems:
> Fixing WSL2 localhost access issue - abdus.dev
> <https://abdus.dev/posts/fixing-wsl2-localhost-access-issue/>
>
> Cheers
> Thorsten
>
> PS 1
> $  pil wiki/main.l -main -go +
> (and (== 'login *ID) (app))
> ! *ID
> -> NIL
> ! (setq *ID 123456789)   # naive try
> -> 123456789
>
> # naiv try did not work
> Can't load -2.html?*Menu=+0&*Tab=+1&*ID=+123456789&*ID=$login
>
> PS 2
> $ ping -a localhost
> PING localhost (127.0.0.1) 56(84) bytes of data.
>
> # type c (peek current url) in W3M : wiki start page
> http://localhost:8080/
>
> # $ sudo cat /etc/hosts
> 127.0.0.1   localhost
> 127.0.1.1   xyz.localdomain  xyz
>
> Am Di., 28. Dez. 2021 um 20:44 Uhr schrieb Alexander Burger <
> a...@software-lab.de>:
>
>> Hi Thorsten, Olaf,
>>
>> > it's a session problem.
>> > Port 8080 does not help, but I cross checked on Archlinux, when I first
>> > click on the Login link, a session prefix is added to the internal
>> links,
>> > and after login, that session prefix is everywhere in the html source.
>> > But on WSL2 that does not work somehow
>>
>> This is indeed strange. This logic should not depend on the system.
>>
>> I think the session is started after clicking on "Log in" in
>>
>>(dm (html> . +Doc) ()
>>   (and (== 'login *ID) (app))
>>
>> So somehow this check fails? The *ID value is passed on the URL, perhaps
>> it is
>> destroyed somehow? The browser?
>>
>> ☺/ A!ex
>>
>> --
>> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>>
>

Re: Internal WIKI Links do not work on WSL2 Debian

[Thorsten Jolitz]

Hi Alex,
yes, *ID is NIL when I debug it (see PS1).
I'm not sure how to debug (app), but all global vars except *PID seem to be
NIL ...

I think it's a WSL2 localhost problem, not a bug in PicoLisp.
Not too long ago I tried accessing the pil wiki server running on WSL2 from
the Win10 Browser (say Edge), and that did not work at all.
Now it seems they have fixed that ,what is very nice, but still not the
same like pure Linux apparently.

Checking localhost everything seems normal (see PS 2) , but e.g this
article describes a lot of problems:
Fixing WSL2 localhost access issue - abdus.dev


Cheers
Thorsten

PS 1
$  pil wiki/main.l -main -go +
(and (== 'login *ID) (app))
! *ID
-> NIL
! (setq *ID 123456789)   # naive try
-> 123456789

# naiv try did not work
Can't load -2.html?*Menu=+0&*Tab=+1&*ID=+123456789&*ID=$login

PS 2
$ ping -a localhost
PING localhost (127.0.0.1) 56(84) bytes of data.

# type c (peek current url) in W3M : wiki start page
http://localhost:8080/

# $ sudo cat /etc/hosts
127.0.0.1   localhost
127.0.1.1   xyz.localdomain  xyz

Am Di., 28. Dez. 2021 um 20:44 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> Hi Thorsten, Olaf,
>
> > it's a session problem.
> > Port 8080 does not help, but I cross checked on Archlinux, when I first
> > click on the Login link, a session prefix is added to the internal links,
> > and after login, that session prefix is everywhere in the html source.
> > But on WSL2 that does not work somehow
>
> This is indeed strange. This logic should not depend on the system.
>
> I think the session is started after clicking on "Log in" in
>
>(dm (html> . +Doc) ()
>   (and (== 'login *ID) (app))
>
> So somehow this check fails? The *ID value is passed on the URL, perhaps
> it is
> destroyed somehow? The browser?
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Internal WIKI Links do not work on WSL2 Debian

[Thorsten Jolitz]

Hallo Olaf,
it's a session problem.
Port 8080 does not help, but I cross checked on Archlinux, when I first
click on the Login link, a session prefix is added to the internal links,
and after login, that session prefix is everywhere in the html source.
But on WSL2 that does not work somehow, I cleaned up the Wiki DB and
started again, but same problem
Regards, Thorsten

PS
Before that, the links look the same:
WSL2/Debian
   Log in
Archlinux:
   Log in
=> click on login
   Log
in

Am Di., 28. Dez. 2021 um 16:27 Uhr schrieb O.Hamann :

> And another idea:
>
> Did you try other more common ports, sth. like 8080 or so?
> (in case there is some firewall mechanism blocking, or so?)
>
> Regards, Olaf
>
>
> On 24.12.21 14:06, Thorsten Jolitz wrote:
> > Hi List,
> > I have the newest pil21 and the current picolisp wiki running on
> > - Win10 / WSL2 (Debian unstable)
> > -  Archlinux
> > and while in both cases I do see the start page of the wiki on
> > http://localhost:5000, only the external Menu links work on WSL2, not
> > the wiki internal links (nor the login).
> >
> > What is quite nice now, on Win10 I can access the running wiki server on
> > localhost from WSL2/Debian (via W3M) and from the Windows browser
> > (chrome or edge), I think that didn't work before.
> >
> > But these kind of links do not work in either case,
> >  > href="-2.html?*Menu=+0&*Tab=+1&*ID=&*ID=$login">Log in
> >
> > W3M says: "Can't load -2.html?*Menu=+0&*Tab=+1&*ID=&*ID=$login", while
> > Edge shows "about:blank#blocked". I started the server in debug mode
> > (pil wiki/main.l -main -go +), but no messages show up in the server
> > window.
> >
> > Did anyone experience the same issue?
> > Cheers
> > Thorsten
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subjectUnsubscribe
>

Internal WIKI Links do not work on WSL2 Debian

[Thorsten Jolitz]

Hi List,
I have the newest pil21 and the current picolisp wiki running on
- Win10 / WSL2 (Debian unstable)
-  Archlinux
and while in both cases I do see the start page of the wiki on
http://localhost:5000, only the external Menu links work on WSL2, not the
wiki internal links (nor the login).

What is quite nice now, on Win10 I can access the running wiki server on
localhost from WSL2/Debian (via W3M) and from the Windows browser (chrome
or edge), I think that didn't work before.

But these kind of links do not work in either case,
Log
in

W3M says: "Can't load -2.html?*Menu=+0&*Tab=+1&*ID=&*ID=$login", while Edge
shows "about:blank#blocked". I started the server in debug mode (pil
wiki/main.l -main -go +), but no messages show up in the server window.

Did anyone experience the same issue?
Cheers
Thorsten

Re: Manually install (newest) PIL21 on Debian/WSL2

[Thorsten Jolitz]

Hi Mike, Hi Karl,
yes shame on me, I somehow forgot temporarily that it's PIL21 now, ;-(
Cheers
Thorsten

PS
with this line from INSTALL:
$ sudo apt install make clang llvm libreadline-dev libffi-dev libssl-dev
pkg-config

nbv3@nbox:~/bin/pil21$ (cd src/; make)
nbv3@nbox:~/bin/pil21$ ./pil +
: (version)
21.8.20
-> (21 8 20)

Am So., 22. Aug. 2021 um 23:18 Uhr schrieb Karl-Heinz Kreis <
karl-heinz.kr...@gmx.de>:

>
> Hi,
>
> opt is part of llvm assembler, version 11 .
> The Toolchain needs other fairly actual versions, perl3 ...
>
> Greetings Karl
>

Manually install (newest) PIL21 on Debian/WSL2

[Thorsten Jolitz]

Hi List,
on Win10 with WSL2 and Debian unstable, I can install Pil21 by "sudo
apt-get picolisp", and I get :

nbv3@nbox:~$ pil +
: (version )
21.6.30


When I download the "rolling release" version, and try to install it
locally, I'm asked to first install MAKE and CLANG. Then I get the error in
PS.
(There is an /OPT directory in this debian install, if that's what the
error refers to)

Did anyone try (and maybe solve) this too?
Cheers
Thorsten

PS
*Error*
nbv3@nbox:~/bin/pil21$ (cd src/; make)
make: opt: No such file or directory
make: *** [Makefile:40: base.bc] Error 127

*src/Makefile: *

# 19aug21 Software Lab. Alexander Burger

SILENT:

CC = clang
PIL = ../pil  # pil
* ASM = opt -O3  # llvm-as  *
[...]
base.bc: base.ll

  *   $(ASM) -o base.bc base.ll*

Re: Jitsi screen sharing issues

[Thorsten Jolitz]

Hi Alex,
since I cannot attend every PilCon, I would not want to provoke a change
that maybe causes problems for others that attend more frecuently.
Otherwise it would of course be ok to try SSH again. For me, the jitsi
problem was just that the shared screen comes and goes, and was mostly gone

Re: Jitsi screen sharing issues

[Thorsten Jolitz]

Hi Mansur,
I had exactly the same issue the last 2 PilCons I attended, on Win10, with
both Edge and Chrome.
It almost makes no sense for me to attend that way, because I do not see
what Alex presents.

I read somewhere if someone with Android connects, the others get problems,
but it seems everybody else was just fine except you and me.

Cheers
Thorsten

Mansur Mamkin  schrieb am Mi., 21. Juli 2021, 17:19:

> Hi all,
>
> During last PilCon I repeatedly lost shared screen after several seconds
> of connection (or little more). Helped reconnection only. The same
> behavior on Linux+Chrome and MacOS+(Chrome or Safari). Maybe anyone
> knows how to solve such issue or some workaround?
>
> Best regards,
>
> Mansur Mamkin
>
>
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: PilCon tomorrow

[Thorsten Jolitz]

Hi Alex,
I won't be able to participate during the whole meeting, maybe just half an
hour, and I already was in the first meeting about the topic.

But I have questions about distributed databases.

Would it be possible to use Picolisp as "super fat client" by installing
the server app on the client side too and just let the databases
synchronize?

So you can actually allow the user to insert and execute code , because
it's done on his client, and not on the server, but the client & server
databases synchronize and the generated html is available on the server?

I think about a Notebook type application.  This is not possible with e.g.
Emacs Org-Mode, for obvious security reasons. Any Org-Mode server that does
website generation by exporting Org-Mode files to html has to deactivate
the execution of code blocks in that file.

Now imagine we want to add the syntax "code block" to the Picolisp wiki
syntax, so that the user can create dynamic documents, e.g. with a code
block that updates weather info every 30min. Obviously, Alex wouldn't want
to blindly execute user code on his server.

But what if the user could clone the whole server app on his client
computer, as a "super fat client", does the code execution on his client,
the DB synchronizes, and the server shows an updated html page every half
hour, without any code execution on the server? A static site generator,
but with dynamic sources, created by and executed at the risk of the user?

Not sure if that is a valid idea ;-)
Would probably only possible with Picolisp, due to the small size and easy
installation.

Cheers
Thorsten

Alexander Burger  schrieb am Mo., 19. Juli 2021, 10:47:

> Hi all,
>
> next PilCon will be tomorrow, July 20th, at 16:00 UTC on
>
>https://meeting.itship.ch/PilCon
>
> The plan was to recapitulate the three database layers in PicoLisp,
> starting
> from the basics. Any questions or other input welcome!
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: PilCon tomorrow (Anatomy of Vip)

[Thorsten Jolitz]

Hi Alex,
thanks for your effort here, I can follow your reasoning, emulating Emacs
produces a lot of commands , which is not exactly in the spirit of
simplicity, and one can just adapt it's workflow instead of adding more
commands.
But maybe I can reuse your code anyway!
Cheers Thorsten

Alexander Burger  schrieb am Mo., 19. Apr. 2021, 07:14:

> On Sun, Apr 18, 2021 at 07:13:50PM +0200, Alexander Burger wrote:
> > OK, thinking about it, I simplified it a little and will indeed use it
> > occasionally perhaps :)
> >
> >(de *F9  # Eval lines till mark "e"
> >   (evCmd
> >  (run (str (getText (jmpMark "e" ) )
> > ...
> > So now I do "me" somewhere to set ark "e" and then "F9" from somewhere
> before
> > that position to execute the code.
>
> No, thinking about it again, I will remove it frmm my ~/.pil/viprc :)
>
> It is useless, because
>
> 1. using the existing Ctrl-E on a few expressions, one after the other, is
> more
>interactive and intuitive than pressing many keys to select a region
> and then
>evaluate it with other keys.
>
> 2. it is easier to type ":l" to reload the whole file. In general I
> write
>sources to be 'load'able as a whole any time.
>
> 3. it wastes a precious function key.
>
> 4. it requires more rules to remember. Keep it simple!
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: PilCon tomorrow (Anatomy of Vip)

[Thorsten Jolitz]

Hi Alex,
that is cool, a bit like in Emacs, where you work in a major mode buffer
and eval expressions there, maybe sending the results to a repl buffer.

What do you think about some more enhancements in that spirit:

[X] eval atom
[X] eval expression
[ ] eval definition/form (?)
[ ] eval region
[ ] eval buffer/window

(In Emacs all these commands often have "eval-x-and-go" siblings, that eval
the expression and then jump to the repl)

In Emacs one uses markers for regions, in VIP I'm not sure ...

The difference between expression and definition is, that in Emacs the
cursor might be inside the expression/definition, and first jumps to the
opening paren, then evaluates.

(de foo (X) (+ X| 2))

With the cursor behind X would then either return X+2 or foo. Saves some
cursor movements before evaluation, but needs different keybindings, not
only one.

Cheers
Thorsten



Alexander Burger  schrieb am So., 18. Apr. 2021, 16:39:

> Hi all,
>
> when a question or request came up at PilCon about evaluating expressions
> in
> Vip's edit buffer, I explained how this works in a line-oriented way in the
> bottom command window.
>
> Now (version 21.4.18) I added a similiar feature to normal edit buffers:
> Hitting
> Ctrl-E (edit) on either an open parenthesis or an atom causes the
> evaluation of
> that expression, with output directed to the command window.
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Native Calls with SEXP (a variant type, with subtypes for all R’s data structures).

[Thorsten Jolitz]

Picolisp has "native" calls as an FFI to call C shared libraries from
Picolisp, and build Picolisp wrapper functions on top of C functions
defined in header files.

A lot of programming languages allow that, there are e.g. many Python
wrappers for all kinds of data science libraries. But with Picolisp you
need only very little code to do it. It's still not easy when the C
functions are non-trivial (and you have only limited C knowledge).

There are quite a few Picolisp wrapper libs from AW on GitHub with good
explanations

Cesar Rabak  schrieb am Do., 8. Apr. 2021, 21:54:

> So, if I understand correctly we could use the introspective power of
> picolisp and write a {meta?}program to produce the "library" for Pil, and
> then use this to call R from picolisp?
>
> --
> Cesar Rabak
>
> On Thu, Apr 8, 2021 at 4:14 PM Thorsten Jolitz  wrote:
>
>> Hi Alex,
>> after digging deeper in R Internals I found out that there are actually C
>> access functions for the various subtypes of SEXP that can be wrapped with
>> native too:
>>
>> rinc: (setupRinC)
>> -> NIL
>> rinc: (evalQuietlyInR "V <- c(1.5, 3.4, 4.2)")
>> -> NIL
>> rinc: (evalInR "V")
>> -> 65007640
>> rinc: (REAL (evalInR "V"))
>> -> (2 3 4)
>> rinc: (teardownRinC)
>> -> NIL
>>
>> I think this is already quite nice.
>> Setup an R session, do quiet evaluations in R, do evaluations with return
>> value, convert that return value (SEXP subtype) to its appropriate Pil data
>> type, tear down R session.
>>
>> So there are only a few challenges left:
>> 1. write wrappers for all type conversion functions (for all SEXP
>> subtypes)
>> 2. figure out the SEXP subtype from the pointer returned
>> 3. dynamically get the length for all SEXP subtypes with a length, to be
>> able to dynamically define return types like '(1.0 . 3) in native or struct
>> (I think there are C functions for this in R internals too)
>>
>> Question on topic 2:
>> given the C data type described below for an R "node", would it be
>> somehow possible to extract just the SEXPTYPE from a SEXP return value, i.e.
>> the first field of the sexpinfo_struct, that is the first struct in the
>> SEXPREC_HEADER, that is the first field in the SEXPREC struct to which the
>> SEXP return value points?
>> Without specifying that complicated C structure in native or struct,
>> because we have the C access functions for that. Ignoring all the rest of
>> the pointer content, just extract that one field SEXPTYPE?
>>
>> Knowing which SEXPTYPE the return value represents, one could then read
>> the returned pointer with the appropriate data type into Pil (like REAL in
>> the example above).
>> if the return value is atomic, one is done. If its a vector, list, or so
>> one would need the length of it too, not only the type (or is it a valid
>> strategy to just use some very big value for the specification like '(1.0 .
>> ) or so)?
>>
>> Cheers
>> Thorsten
>>
>>
>> *Composition of an R node: *
>> the SEXP subtypes are defined like this:
>> typedef unsigned int SEXPTYPE;
>>   52   #define NILSXP›0›   /* nil = NULL */
>>   53   #define SYMSXP›1›   /* symbols */
>>   54   #define LISTSXP›   2›   /* lists of dotted pairs */
>>   55   #define CLOSXP›3›   /* closures */
>>   56   #define ENVSXP›4›   /* environments */
>>   57   #define PROMSXP›   5›   /* promises: [un]evaluated closure
>> arguments */
>>   58   #define LANGSXP›   6›   /* language constructs (special lists)
>> */
>>   59   #define SPECIALSXP   7› /* special forms */
>>   60   #define BUILTINSXP   8› /* builtin non-special forms */
>>   61   #define CHARSXP›   9›   /* "scalar" string type (internal
>> only)*/
>>   64  #define INTSXP›   13›   /* integer vectors */
>>   65   #define REALSXP›  14›   /* real variables */
>>   66   #define CPLXSXP›  15›   /* complex variables */
>>   67   #define STRSXP›   16›   /* string vectors */
>>   68   #define DOTSXP›   17›   /* dot-dot-dot object */
>>   69   #define ANYSXP›   18›   /* make "any" args work.
>>   72   #define VECSXP›   19›   /* generic vectors */
>>   73   #define EXPRSXP›  20›   /* expressions vectors */
>>   74   #define BCODESXP21/* byte code */
>>   62   etc
>>
>> this type is part of the sxpinfo structure:
>> struct sxpinfo_struct {
>>  142   SEXPTYPE type  :  TYPE_BITS;
>&g

Re: Native Calls with SEXP (a variant type, with subtypes for all R’s data structures).

[Thorsten Jolitz]

Hi Alex,
after digging deeper in R Internals I found out that there are actually C
access functions for the various subtypes of SEXP that can be wrapped with
native too:

rinc: (setupRinC)
-> NIL
rinc: (evalQuietlyInR "V <- c(1.5, 3.4, 4.2)")
-> NIL
rinc: (evalInR "V")
-> 65007640
rinc: (REAL (evalInR "V"))
-> (2 3 4)
rinc: (teardownRinC)
-> NIL

I think this is already quite nice.
Setup an R session, do quiet evaluations in R, do evaluations with return
value, convert that return value (SEXP subtype) to its appropriate Pil data
type, tear down R session.

So there are only a few challenges left:
1. write wrappers for all type conversion functions (for all SEXP subtypes)
2. figure out the SEXP subtype from the pointer returned
3. dynamically get the length for all SEXP subtypes with a length, to be
able to dynamically define return types like '(1.0 . 3) in native or struct
(I think there are C functions for this in R internals too)

Question on topic 2:
given the C data type described below for an R "node", would it be somehow
possible to extract just the SEXPTYPE from a SEXP return value, i.e.
the first field of the sexpinfo_struct, that is the first struct in the
SEXPREC_HEADER, that is the first field in the SEXPREC struct to which the
SEXP return value points?
Without specifying that complicated C structure in native or struct,
because we have the C access functions for that. Ignoring all the rest of
the pointer content, just extract that one field SEXPTYPE?

Knowing which SEXPTYPE the return value represents, one could then read the
returned pointer with the appropriate data type into Pil (like REAL in the
example above).
if the return value is atomic, one is done. If its a vector, list, or so
one would need the length of it too, not only the type (or is it a valid
strategy to just use some very big value for the specification like '(1.0 .
) or so)?

Cheers
Thorsten


*Composition of an R node: *
the SEXP subtypes are defined like this:
typedef unsigned int SEXPTYPE;
  52   #define NILSXP›0›   /* nil = NULL */
  53   #define SYMSXP›1›   /* symbols */
  54   #define LISTSXP›   2›   /* lists of dotted pairs */
  55   #define CLOSXP›3›   /* closures */
  56   #define ENVSXP›4›   /* environments */
  57   #define PROMSXP›   5›   /* promises: [un]evaluated closure
arguments */
  58   #define LANGSXP›   6›   /* language constructs (special lists) */
  59   #define SPECIALSXP   7› /* special forms */
  60   #define BUILTINSXP   8› /* builtin non-special forms */
  61   #define CHARSXP›   9›   /* "scalar" string type (internal only)*/
  64  #define INTSXP›   13›   /* integer vectors */
  65   #define REALSXP›  14›   /* real variables */
  66   #define CPLXSXP›  15›   /* complex variables */
  67   #define STRSXP›   16›   /* string vectors */
  68   #define DOTSXP›   17›   /* dot-dot-dot object */
  69   #define ANYSXP›   18›   /* make "any" args work.
  72   #define VECSXP›   19›   /* generic vectors */
  73   #define EXPRSXP›  20›   /* expressions vectors */
  74   #define BCODESXP21/* byte code */
  62   etc

this type is part of the sxpinfo structure:
struct sxpinfo_struct {
 142   SEXPTYPE type  :  TYPE_BITS;
 143   /* ==> (FUNSXP == 99) %% 2^5 == 3 ==
CLOSXP
 144   › ›   ›,* -> warning: `type' is narrower than values
 145   › ›   ›,*  of its type
 146   › ›   ›,* when SEXPTYPE was an enum */
 147   unsigned int scalar:  1;
 148   unsigned int obj   :  1;
 149etc

which then becomes part of an header:
#define SEXPREC_HEADER \
 209   struct sxpinfo_struct sxpinfo; \
 210   struct SEXPREC *attrib; \
 211   struct SEXPREC *gengc_next_node, *gengc_prev_node
 212

which finally becomes part of the SEXPREC structure. A SEXP is a pointer to
a SEXPREC structure.
 213   /* The standard node structure consists of a header followed by the
 214  node data. */
 215   typedef struct SEXPREC {
 216   SEXPREC_HEADER;
 217   union {
 218   › struct primsxp_struct primsxp;
 219   › struct symsxp_struct symsxp;
 220   › etc




Am Mi., 7. Apr. 2021 um 15:55 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> On Wed, Apr 07, 2021 at 07:46:20AM +0200, Alexander Burger wrote:
> > OK, so now we have a pointer to a structure filled by evalInR().
> > ...
> >(struct (evalInR "6*4") ...)
>
> In fact this depends on what exactly evalInR() returns.
>
> If it returns a pointer to a dynamically allocated structure, and the
> caller is
> responsible to de-allocate it, you would do
>
>(let P (evalInR "6*4")
>   (prog1
>  (struct P ...)
>  (free P) ) )
>
> If, however, the structure is statically allocated, you can also give the
> structure's structure directly as the return value specification:
>
>(native `*RinC "evalInR" '(...) "Cmd")
>
> For example, as a (somewhat conceived) test 

Re: Native Calls with SEXP (a variant type, with subtypes for all R’s data structures).

[Thorsten Jolitz]

Hi Cesar,
it works both ways, and there is an old R C API  and more modern C++
implementations (with C headers too) for the use case here:
- call R from C(++).
or better:
- call R from PicoLisp via (native) C (calls)
See my last answer to Alex to see that the call to R actually works.
Cheers
Thorsten.


Am Di., 6. Apr. 2021 um 22:20 Uhr schrieb Cesar Rabak :

> Hi Thorsten,
>
> The quote you copied here (coming from the ref. of yours R's C Interface),
> describes IIUC a structure, more accurately the model of, for calling
> "foreing" C functions in R and not the converse.
>
> The way to call R functions in picolisp would be to call the functions
> made available through the API, in case of R a process similar to FFI using
> as reference the include files, wouldn't?
>
> HTH
> --
> Cesar Rabak
>
>
> On Mon, Apr 5, 2021 at 4:30 PM Thorsten Jolitz  wrote:
>
>> Hello List,
>> I wonder how to deal with the R SEXP Data structure in native calls.
>>
>> *"Technically, [a SEXP] is a pointer to a structure with typedef SEXPREC

Re: Native Calls with SEXP (a variant type, with subtypes for all R’s data structures).

[Thorsten Jolitz]

Hi Alex,
thanks for the hints, I tried both, T and 'P as result values , the first
gives a segment fault, but the second actually works:

 (de evalInR ("Cmd")
  (native `*RinC "evalInR" 'P "Cmd"))


##  SEXP evalInR(char * cmd);

rinc: (evalInR "print(6*4)")
[1] 24
-> 65814160

If I use "print" in the R command, I actually see the R output.
And with the 'P, I actually get a return value.
But what can I do in PicoLisp with that pointer? If I skip the "print" and
just call R for the return value:

rinc: (evalInR "6*4")
-> 65814160

how can I extract the result from that Pointer in PicoLisp (lack of C
skills ... ;-)?
Cheers
Thorsten

Am Di., 6. Apr. 2021 um 10:23 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> Hi Thorsten,
>
> > I wonder how to deal with the R SEXP Data structure in native calls.
> >
> > *"Technically, [a SEXP] is a pointer to a structure with typedef SEXPREC.
> > A SEXP is a variant type, with subtypes for all R’s data structures"*
> >
> > E.g.
> >
> >- INTSXP: integer vector
> >- LGLSXP: logical vector
> >- STRSXP: character vector
> >- ... (and a dozen more)
>
> I do not really understand the implications in the context of R, but Pil21
> has a
> new 'T' result specification for raw Lisp data.
>
> It allows to pass a pointer to any Lisp data item to a native function,
> and/or
> to return such data.
>
>
> > When I have an imaginary generic C function like this:
> >
> > SEXP fun(SEXP x, char * cmd)
>
> As an example, we might call the standard 'prog1' function in the Pil21
> executable (which has the internal label "_prog1", try (vi 'prog1)). It
> takes a
> list for the body, and returns the resulting value:
>
>: (%@ "_prog1" T '(T prog1 7 (println 1 2 3)))
>1 2 3
>-> 7
>
> This is equivalent to
>
>: (prog1 7 (println 1 2 3))
>1 2 3
>-> 7
>
>
> > how can I specify Return Value/Arguments in a 'native' call, if I cannot
> > know in advance to which R subtype the  SEXP Points?
>
> If the above is not what you intended, the 'P' result specification can be
> used
> as an unspecific pointer (void*, an unsigned 64 bit number).
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Native Calls with SEXP (a variant type, with subtypes for all R’s data structures).

[Thorsten Jolitz]

Hello List,
I wonder how to deal with the R SEXP Data structure in native calls.

*"Technically, [a SEXP] is a pointer to a structure with typedef SEXPREC.
A SEXP is a variant type, with subtypes for all R’s data structures"*

E.g.

   - INTSXP: integer vector
   - LGLSXP: logical vector
   - STRSXP: character vector
   - ... (and a dozen more)

When I have an imaginary generic C function like this:

SEXP fun(SEXP x, char * cmd)

how can I specify Return Value/Arguments in a 'native' call, if I cannot
know in advance to which R subtype the  SEXP Points?

On the one hand, a SEXP is like a C Datatype for a Lisp list, so one could
think that's ideal for calling C from Pil with list args and list return
values. On the other hand, how can one give the primitive specifications in
the specification lists for 'native' args/return vals, without knowing the
concrete subtype (R data structure), i.e. when CMD can be any R command
that may take/return any of the subtypes (char, logical, int, vector, list
)?

Is there maybe a generic solution on the 'native' side too, without the
primitive specifications?
Thanks in advance and Cheers
Thorsten

PS
For those interested here two links:
Rinternals.h source code:
r-source/Rinternals.h at a1425adea54bcc98eef86081522b5dbb3e149cdc ·
wch/r-source (github.com)


R's C Interface (blog post explaining the code/data structures)
R's C interface · Advanced R. (had.co.nz)
.

Re: Picolisp Outlook

[Thorsten Jolitz]

I'm just saying that it would be nice to have a Picolisp app that has a
real demand and a little ecosystem around it ...

Like e.g. Moodle or some CMS system, that probably have an ecosystem of PHP
programmers around them, and many organisations using them (with need for
some support). For them it's irrelevant if it's PHP or anything else.

C K Kashyap  schrieb am Di., 23. Feb. 2021, 15:56:

> I also believe that the simplicity of PicoLisp by itself is a "killer
> app". I've explored several languages in the past but PicoLisp is the only
> one that allowed me to understand the implementation of the language all
> the way (I still have to work on the external symbol bit :) . I was so used
> to seeing documents of concepts of other languages that don't directly map
> to the source code that it took me a while to realize the accuracy of
> PicoLisp documentation.
>
> Regards,
> Kashyap
>
> On Tue, Feb 23, 2021 at 2:26 AM Manuel Cano  wrote:
>
>> Hi,
>>
>> That's surely because it isn't made in Picolisp! :D
>>
>> Kind regards,
>> Manu
>>
>>
>> El mar, 23 feb 2021 a las 10:32, Alexander Burger ()
>> escribió:
>>
>>> Hi all,
>>>
>>> sorry for the multiple mails! Seems I have a DNS problem on
>>> the server ...
>>>
>>> On Tue, Feb 23, 2021 at 09:18:57AM +0100, Alexander Burger wrote:
>>> > Hi Thorsten,
>>> > ...
>>>
>>> --
>>> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>>>
>>>

Re: Picolisp Outlook

[Thorsten Jolitz]

Hi Alex,


Alexander Burger  schrieb am Di., 23. Feb. 2021, 09:33:

> Hi Thorsten,
>
> > Maybe we should sponsor Alex (3 month work?) to build that "killer app"
> > with the clear goals
>
> Hmm, no need to specially sponsor me, but thanks for the proposal!
>
> The problem is that such an app would probably not be accepted by those
> millions
> of (data) scientists. As always, Lisp looks too suspicious to them ;)
>

That's too bad, but probably the harsh truth.

Then there is only one option left ;-)
Build a low-code/no-code system where Picolisp simplicity shines under the
hood, but no average user sees any lisp.
These are getting quite popular nowadays in the enterprise world, but then
again, which enterprise doesn't go with the market leader for almost any
application?

Cheers
Thorsten


> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Picolisp Outlook

[Thorsten Jolitz]

I thought interfacing with important data science libraries would be a pro,
but of course it has to be done first, it's quite a lot of work and not
always that easy, even if it's C, not to mention C++.

I guess you are right about the popularity of Lisp ...

pd  schrieb am Di., 23. Feb. 2021, 08:57:

> In my opinion picolisp has two disadvantages to be a scientific
> programming language:
>
> 1 lack of flota support
> 2 lack of libraries
>
> Both together make far easier to develop a science app in Python rather
> than picolisp
>
> For sure It has got adventages too like db and pil but disadvantages are a
> heavy stone
>
> Also lisp is not a fancy language, at least most people thinks it's weird
> compared to Python and similars and this sure doesn't helps. This is the
> reason for "science languages" are so similar in syntax (python,R...)
>
> Greets
>
>
> El lun., 22 feb. 2021 22:40, Thorsten Jolitz  escribió:
>
>> Maybe we should sponsor Alex (3 month work?) to build that "killer app"
>> with the clear goals
>>
>> - for all those millions of (data) scientists that work with R etc, it
>> should be the easiest (because fully integrated) way to build applications
>> on top of their data
>> - for those who like Picolisp it should be the Pil App to potentially
>> make money with (without being superstar programmers).
>>
>> If Alex charges moderate rates (and doesn't think the idea is a waste ;-)
>> I would be willing to take over one month of development ... ;-)
>>
>>
>>
>>
>>  schrieb am Mo., 22. Feb. 2021, 21:51:
>>
>>> Yeah I had kinda similiar ideas, Thorsten.
>>>
>>> PicolispDB is certainly a killer feature - multi-paradigm database
>>> (Key-Value, Object, Document, Graph, Relational.. really everything
>>> covered), ACID (transactions), many indexing capabilities (including
>>> text and spatial indexing), performant, extremely flexible and nicely
>>> well maintainable.
>>>
>>> In the past, a lack of (digitalized) data was often an obstacle to get
>>> useful (business) insights with software.
>>> This changed, we are drowning in data.
>>>
>>> Now, and for the foreseeable future, the problem is to make sense of the
>>> data, to be able to filter, map and connect various formats and data
>>> sources together - while requirements change all the time.
>>>
>>> I believe Picolisp Database is a tool outstandingly suited for this.
>>>
>>> On 22.02.21 17:04, Thorsten Jolitz wrote:
>>> > hallo list,
>>> > I always thought a "killer app" would be nice, to make those "killer
>>> > features" popular, and I always thought that could be a "data science
>>> > application builder" with 3 features:
>>> >
>>> > - easy data import into a Picolisp DB
>>> > - ffi/java wrappers for many data science libs (Rmath,  Weka, ...)
>>> > - easy web app development
>>> >
>>> > because for data scientists it seems often quite difficult to build
>>> > applications on top of their data, and with Picolisp it would be all
>>> > integrated into one single tool.
>>> >
>>> > But then I should be the one who implements that, and I made some
>>> > attempts, but never had the time/energy/stamina/skills to bring it on

Re: Picolisp Outlook

[Thorsten Jolitz]

Maybe we should sponsor Alex (3 month work?) to build that "killer app"
with the clear goals

- for all those millions of (data) scientists that work with R etc, it
should be the easiest (because fully integrated) way to build applications
on top of their data
- for those who like Picolisp it should be the Pil App to potentially make
money with (without being superstar programmers).

If Alex charges moderate rates (and doesn't think the idea is a waste ;-) I
would be willing to take over one month of development ... ;-)




 schrieb am Mo., 22. Feb. 2021, 21:51:

> Yeah I had kinda similiar ideas, Thorsten.
>
> PicolispDB is certainly a killer feature - multi-paradigm database
> (Key-Value, Object, Document, Graph, Relational.. really everything
> covered), ACID (transactions), many indexing capabilities (including
> text and spatial indexing), performant, extremely flexible and nicely
> well maintainable.
>
> In the past, a lack of (digitalized) data was often an obstacle to get
> useful (business) insights with software.
> This changed, we are drowning in data.
>
> Now, and for the foreseeable future, the problem is to make sense of the
> data, to be able to filter, map and connect various formats and data
> sources together - while requirements change all the time.
>
> I believe Picolisp Database is a tool outstandingly suited for this.
>
> On 22.02.21 17:04, Thorsten Jolitz wrote:
> > hallo list,
> > I always thought a "killer app" would be nice, to make those "killer
> > features" popular, and I always thought that could be a "data science
> > application builder" with 3 features:
> >
> > - easy data import into a Picolisp DB
> > - ffi/java wrappers for many data science libs (Rmath,  Weka, ...)
> > - easy web app development
> >
> > because for data scientists it seems often quite difficult to build
> > applications on top of their data, and with Picolisp it would be all
> > integrated into one single tool.
> >
> > But then I should be the one who implements that, and I made some
> > attempts, but never had the time/energy/stamina/skills to bring it on.
> > So this is just an idea I had for quite some time ...
> >
> > Cheers
> > Thorsten
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>
>

Re: Picolisp Outlook

[Thorsten Jolitz]

hallo list,
I always thought a "killer app" would be nice, to make those "killer
features" popular, and I always thought that could be a "data science
application builder" with 3 features:

- easy data import into a Picolisp DB
- ffi/java wrappers for many data science libs (Rmath,  Weka, ...)
- easy web app development

because for data scientists it seems often quite difficult to build
applications on top of their data, and with Picolisp it would be all
integrated into one single tool.

But then I should be the one who implements that, and I made some attempts,
but never had the time/energy/stamina/skills to bring it on. So this is
just an idea I had for quite some time ...

Cheers
Thorsten




pd  schrieb am Mo., 22. Feb. 2021, 12:26:

>
>
> El lun., 22 feb. 2021 9:31, Alexander Burger 
> escribió:
>
>> ... and immune to temporary hypes
>>
>
> What a nice desire being rejected by history of humankind again and again
> ;)
>
> Specially in computer science
>
>>

Re: Call native wrapper function with double argument

[Thorsten Jolitz]

Hi Alex,
thanks for your tips ... and patience!
My lack of experience with C, Pointers and low level programming clearly
shows in my questions, and most real world libraries of 'native' wrappers I
find don't really implement complex C signatures, so I have to experiment
myself.

I'll try to digest your hints and do some more experimentation over the
weekend.
Luckily, If I manage to wrap one such C function, I can reuse that
understanding for many others 
Cheers
Thorsten

Am Di., 24. Nov. 2020 um 07:50 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> On Mon, Nov 23, 2020 at 06:17:46PM +0100, Alexander Burger wrote:
> > Something like '("Dx" (24 . 1.0) '(1.0 1.0 2.0 3.0)) could make sense,
> if the C
> > argument is "double dx[3];".
>
> oops
>
> (24 . 1.0) is of course not returning "double dx[3];", but a single double.
>
> Try (24 1.0 1.0 1.0) or better (24 (1.0 . 3)).
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Call native wrapper function with double argument

[Thorsten Jolitz]

Hi Alex,
one more 'native' question, this time a bit more complicated:

libblas.so is on my machine, the C function "idamax_" is found,which is
actually a C wrapper for Fortran Code.
In the Fortran docs, the parameters are described exactly, see below.
Since the Parameters of the C function are input/output Parameters, I try
to define the input values in the CDDR, but obviously not correctly.

How do I define an input array correctly (especially DX and INCX)?
If I have in/out Parameters, how do I get the int return value too when the
pil functions returns?

hope these questions are not too messy ...
cheers
Thorsten

PS
when I understand how this works, I will use the input values for native as
parameters for the pil wrapper, but for now I just write them inside of
'native'.


*> \param[in] N
33 *> \verbatim
34 *> N is INTEGER
35 *> number of elements in input vector(s)
36 *> \endverbatim
37 *>
38 *> \param[in] DX
39 *> \verbatim
40 *> DX is DOUBLE PRECISION array, dimension ( 1 + ( N - 1 )*abs( INCX ) )
41 *> \endverbatim
42 *>
43 *> \param[in] INCX
44 *> \verbatim
45 *> INCX is INTEGER
46 *> storage spacing between elements of DX
47 *> \endverbatim
48 *



 (symbols 'blas 'pico)
 (default *LibBlas "libblas.so")


 (de idamax ()
(use "N" "Dx" "Incx"
   (native `*LibBlas "idamax_" 'I '("N" (4 . 'I) 3) '("Dx" (8 . 1.0)
'(1.0 1 2 3)) '("Incx" (4 . 'I) 0))(list "N" "Dx" "Incx") )
)
  ## BLAS_extern int/* IDAMAX - return the index of the element with
max abs value */   ## F77_NAME(idamax)(const int *n,
const double *dx, const int *incx)


# when INCX = 0
: (blas~idamax)
-> ('NIL 0 'NIL)
# when INCX = 1 segfault


Am Di., 17. Nov. 2020 um 23:45 Uhr schrieb Thorsten Jolitz <
tjol...@gmail.com>:

> Hi Alex,
> the wrappers with primitive arguments do work now when defined like this
> with transient symbols:
>
> (de pgamma ("X" "Y" "Z" "I" "J")
>(native `*LibRmath "pgamma" 1.0 (cons "X" 1.0) (cons "Y" 1.0) (cons
> "Z" 1.0) "I" "J" ) )
>## double›  pgamma(double, double, double, int, int);
>
> With your tips I could define wrappers for functions with pointer args too
>
> (de pnorm_both ("X" "I" "J")
>   (use "Y" "Z"
> (native `*LibRmath "pnorm_both" NIL (cons "X" 1.0) '("Y" (8 .
> 1.0)) '("Z" (8 . 1.0)) "I" "J" )
> (list "Y" "Z") ) )
>## void›pnorm_both(double, double *, double *, int, int);/* both
> tails */
>
> does work (scl 3):
> : (rmath~pnorm_both 1.2 2 3)
> -> (-122 -2162)
>
> now I only have to understand the meaning of some of these functions.
> What I find irritating is that the pointer args are not used for input but
> only as a kind of return values.
>
> Thanks for your help!
> Cheers
> Thorsten
>
> Am So., 15. Nov. 2020 um 22:55 Uhr schrieb Alexander Burger <
> a...@software-lab.de>:
>
>> Hi Thorsten,
>>
>> hmm, this is not correct:
>>
>> > (de pnorm_both ("X" "Y" "Z" "I" "J")
>> > (! native `*LibRmath "pnorm_both" 1.0 (cons "X" 1.0) '("Y" (1.0 .
>> 4))
>> > '("Z" (1.0 . 4)) "I" "J" ) )
>>
>> "Z" is an argument to the function, so it is bound to some evaluated
>> value.
>>
>> But this value is ignored, because in
>>
>>'("Z" (1.0 . 4))
>>
>> "Z" is a *return* value. Also, (1.0 . 4) makes no sense here. A structure
>> argument is
>>
>>'(Var ( . >
>> but 1.0 is no size (it is too big, something like 100) and 4 is no
>> return
>> spec.
>>
>>
>> If you want to pass a buffer to receivea double (the double* in the C
>> signature), you would do:
>>
>>(use MyDouble
>>   (native `*LibRmath "pnorm_both" 1.0 ...
>>  '(MyDouble (8 . 1.0)) ... )
>>   ... do something with MyDouble ...)
>>
>> (8 . 1.0) allocates 8 bytes on the stack, passes the pointer to the C
>> function,
>> receives a double in this place, and stores it in the symbol MyDouble.
>>
>> ☺/ A!ex
>>
>> --
>> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>>
>

Re: Call native wrapper function with double argument

[Thorsten Jolitz]

Hi Alex,
the wrappers with primitive arguments do work now when defined like this
with transient symbols:

(de pgamma ("X" "Y" "Z" "I" "J")
   (native `*LibRmath "pgamma" 1.0 (cons "X" 1.0) (cons "Y" 1.0) (cons
"Z" 1.0) "I" "J" ) )
   ## double›  pgamma(double, double, double, int, int);

With your tips I could define wrappers for functions with pointer args too

(de pnorm_both ("X" "I" "J")
  (use "Y" "Z"
(native `*LibRmath "pnorm_both" NIL (cons "X" 1.0) '("Y" (8 .
1.0)) '("Z" (8 . 1.0)) "I" "J" )
(list "Y" "Z") ) )
   ## void›pnorm_both(double, double *, double *, int, int);/* both
tails */

does work (scl 3):
: (rmath~pnorm_both 1.2 2 3)
-> (-122 -2162)

now I only have to understand the meaning of some of these functions.
What I find irritating is that the pointer args are not used for input but
only as a kind of return values.

Thanks for your help!
Cheers
Thorsten

Am So., 15. Nov. 2020 um 22:55 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> Hi Thorsten,
>
> hmm, this is not correct:
>
> > (de pnorm_both ("X" "Y" "Z" "I" "J")
> > (! native `*LibRmath "pnorm_both" 1.0 (cons "X" 1.0) '("Y" (1.0 . 4))
> > '("Z" (1.0 . 4)) "I" "J" ) )
>
> "Z" is an argument to the function, so it is bound to some evaluated value.
>
> But this value is ignored, because in
>
>'("Z" (1.0 . 4))
>
> "Z" is a *return* value. Also, (1.0 . 4) makes no sense here. A structure
> argument is
>
>'(Var ( . 
> but 1.0 is no size (it is too big, something like 100) and 4 is no
> return
> spec.
>
>
> If you want to pass a buffer to receivea double (the double* in the C
> signature), you would do:
>
>(use MyDouble
>   (native `*LibRmath "pnorm_both" 1.0 ...
>  '(MyDouble (8 . 1.0)) ... )
>   ... do something with MyDouble ...)
>
> (8 . 1.0) allocates 8 bytes on the stack, passes the pointer to the C
> function,
> receives a double in this place, and stores it in the symbol MyDouble.
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Call native wrapper function with double argument

[Thorsten Jolitz]

Hi Alex,
I actually do use a namespace, and there is not much going on wrt to
assignments:
>From line 8 on there are only some 100+ mostly trivial wrapper functions,
see PS. I first thought it might have something to do with Picolisp
Linebreaks "delta J", but I have seen the same issue now with an Y
parameter name, not just J.

With a definition like in PS 1, I would be on the safe side?
When I use (bt) on the original signature specification, it looks like
this:
(native NIL "pwilcox" 1000 (cons X 1000) (cons Y 1000) (cons Z 1000) I "J")
 ! (bt)
("pwilcox" 2700 6200 2100 3 ..)
   X 2700
   Y 6200
   Z 2100
   I 3
   "J" 2

Cheers
Thorsten

PS 1
(de pnorm_both ("X" "Y" "Z" "I" "J")
(! native `*LibRmath "pnorm_both" 1.0 (cons "X" 1.0) '("Y" (1.0 . 4))
'("Z" (1.0 . 4)) "I" "J" ) )
  ## void›pnorm_both(double, double *, double *, int, int);/* both
tails */

: (rmath~pnorm_both 2.1 3.2 4.1 3 2)
(native "libRmath.so" "pnorm_both" NIL (cons "X" 1000) '("Y" (1000 . 4))
'("Z" (1000 . 4)) "I" "J")
! (bt)
("pnorm_both" 2100 3200 4100 3 ..)
   "X" 2100
   "Y" 3200
   "Z" 4100
   "I" 3
   "J" 2

PS 2

   1 # libRmath.l - (native) PicoLisp bindings for Rmath
   2 # copyright (C) 2020 Thorsten Jolitz
   3
   4 (symbols 'rmath 'pico)
   5
   6 (default *LibRmath "libRmath.so")
   7
   8 (de Rlog1p (X)
   9 (native `*LibRmath "Rlog1p" 1.0 (cons X  1.0) ) )
  10 ## double  Rlog1p(double);

Am So., 15. Nov. 2020 um 18:50 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> Hi Thorsten,
>
> > But with my real wrapper functions, where ever I use J, as first or
> second
> > arg, its interpreted as a transient symbol
>
> So somewhere 'J' is bound to "J".
>
>
> > But when I rename J to H in my real wrapper function, the problem is
> gone:
>
> Right, so it is a binding issue. Think hard what values are exactly bound
> to
> which variables at runtime. Such things happen usually only in FEXPRs or
> when
> evaluable expressions are passed around through binding environments.
>
>
> > So I think I know whats the problem here: its reading the function
> > definitions from a library file vs defining the definition directly in
> the
> > repl.
> > (load "libRmath.l") seems to have a problem with the char J used for a
>
> Is "libRmath.l" written by you?
>
>
> > The easy solution is therefore to avoid J as parameter name ;-)
> > But maybe I found a bug?
>
> I don't think it is a bug. But you must take care of the implications of
> dynamic
> binding, exwlained e.g. in:
>
>https://software-lab.de/doc/faq.html#problems
>
> Such things are best fixed with transient or private symbols, or even a
> separate
> namespace.
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>
>

Re: Call native wrapper function with double argument

[Thorsten Jolitz]

Hi Alex,
this is strange indeed.
I checked on Bash/Archlinux as well, same problem as Win Term / WSL2, so
this is not about WSL or so.

I can like you define a dummy wrapper, and debug it, and it looks fine:

: (de foo (I J) (! native NIL NIL 1.0 I J))
-> foo
: (foo 2 3)
(native NIL NIL 1 I J)
! J
-> 3

So here I can use J as argument name, and it works.

But with my real wrapper functions, where ever I use J, as first or second
arg, its interpreted as a transient symbol and is NIL, even if I create a
fake wrapper without any real lib reference:

: (rmath~exp_rand_fake2 3 2)
(native NIL NIL 1 "J" B)
! J
-> NIL

But when I rename J to H in my real wrapper function, the problem is gone:
: (rmath~pwilcox 1.2 2.1 3.2 3 2)
(native "libRmath.so" "pwilcox" 1 (cons X 1) (cons Y 1) (cons Z 1) I H)
! H
-> 2

So I think I know whats the problem here: its reading the function
definitions from a library file vs defining the definition directly in the
repl.
(load "libRmath.l") seems to have a problem with the char J used for a
paramter, and somehow converts it into a transient symbol.

The easy solution is therefore to avoid J as parameter name ;-)
But maybe I found a bug?

Cheers
Thorsten

Am So., 15. Nov. 2020 um 09:01 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> Hi Thorsten,
>
> > (de pwilcox (X Y Z I J)
> > ...
> > : (rmath~pwilcox 2.7 6.20 5.4 1 3)
> > (native "libRmath.so" "pwilcox" 1 (cons X 1) (cons Y 1) (cons Z 1) I "J")
> > ...
> > ! J
> > -> NIL
> > Why is that second Integer argument interpreted as transient symbol, and
> > then NIL although the actual arg = 3?
>
> This is strange. I tried the same here, and get
>
>(native NIL "pwilcox" 100 (cons X 100) (cons Y 100) (cons Z
> 100) I J)
>
> And I get 3 as expected for 'J'.
>
> Note the scaled integers! It seems your *Scl is zero. But this does not
> explain
> what you see. Are you sure you pasted exactly what you tested?
>
>
> > In the docs I only find:
> > "The number of fixpoint arguments is limited to six."
> > but that looks irrelevant here.
>
> Yes, for pil64. For pil21 this no longer holds. There is no more limit on
> the
> number of float or double arguments.
>
> ☺/ A!ex
>
> --
> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>

Re: Call native wrapper function with double argument

[Thorsten Jolitz]

Hi Alex,
one more question:

(de pwilcox (X Y Z I J)
(! native `*LibRmath "pwilcox" 1.0 (cons X 1.0) (cons Y 1.0) (cons Z
1.0) I J ) )
## double pwilcox(double, double, double, int, int);


: (rmath~pwilcox 2.7 6.20 5.4 1 3)
(native "libRmath.so" "pwilcox" 1 (cons X 1) (cons Y 1) (cons Z 1) I "J")
! J
-> NIL
! I
-> 1
!
-> -4

Why is that second Integer argument interpreted as transient symbol, and
then NIL although the actual arg = 3?
In the docs I only find:
"The number of fixpoint arguments is limited to six."
but that looks irrelevant here.

So how do I specify a
 ## double pwilcox(double, double, double, int, int);
signature correctly?

Cheers
Thorsten

Am Fr., 13. Nov. 2020 um 17:15 Uhr schrieb Thorsten Jolitz <
tjol...@gmail.com>:

> Hi Alex,
> yes that works with (cons X 1.0), I knew it was a trivial problem.
> Thanks!
> Cheers
> Thorsten
>
> Am Fr., 13. Nov. 2020 um 08:00 Uhr schrieb Alexander Burger <
> a...@software-lab.de>:
>
>> Hi Thorsten,
>>
>> welcome back! :)
>>
>> > I'm playing around with the native function again (after a long long
>> time
>> > ;-) and somehow I don't manage to call a native wrapper with double arg.
>>
>> >
>> > Using rmath from R, random value from poisson distribution:^
>> >  ## double›  rpois(double);
>> > ...
>> > This works
>> > : (native "libRmath.so" "rpois" 1.0 (2.567 . 1.0))
>> > ...
>> > but this dumps
>> > : (de rpois (X) (native "libRmath.so" "rpois" 1.0  (X . 1.0) ) )
>> > ...
>> > When I debug it, X=3 when the function is called.
>>
>> The problem is (X . 1.0), it calls 'X' as a function.
>>
>> So this would work:
>>
>>: (de rpois (X)
>>   (native "libRmath.so" "rpois" 1.0 (cons X 1.0) ) )
>>
>> ☺/ A!ex
>>
>> --
>> UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>>
>>

Re: Call native wrapper function with double argument

[Thorsten Jolitz]

Hi Alex,
yes that works with (cons X 1.0), I knew it was a trivial problem.
Thanks!
Cheers
Thorsten

Am Fr., 13. Nov. 2020 um 08:00 Uhr schrieb Alexander Burger <
a...@software-lab.de>:

> Hi Thorsten,
>
> welcome back! :)
>
> > I'm playing around with the native function again (after a long long time
> > ;-) and somehow I don't manage to call a native wrapper with double arg

Call native wrapper function with double argument

[Thorsten Jolitz]

Hello List, Hi Alex,
I'm playing around with the native function again (after a long long time
;-) and somehow I don't manage to call a native wrapper with double arg.

Using rmath from R, random value from poisson distribution:^
 ## double›  rpois(double);

This works
: (native "libRmath.so" "rpois" 1.0 (2.567 . 1.0))
-> 5
: (native "libRmath.so" "rpois" 1.0 (2.567 . 1.0))
  -> 2

This works too:
  ## int› imax2(int, int);

: (de imax2 (X) (native "libRmath.so" "imax2" 'I 3 6))
-> imax2
: (imax2 3 6)
-> 6

but this dumps
: (de rpois (X) (native "libRmath.so" "rpois" 1.0  (X . 1.0) ) )
-> rpois
: (rpois 2.567)
Segmentation fault

When I debug it, X=3 when the function is called.
So probably my function call is wrong?

Cheers
Thorsten

subscribe

[Thorsten Jolitz]

Re: Vip without [N]curses

[Thorsten Jolitz]

Alexander Burger  writes:

Hi Alex,

>> (for termux I always have to adapt the first line in /bin/vip,
>> i.e. replace /usr with termux $PREFIX, /data/data/com.termux/files/usr)
>
> This is no longer necessary if you install in Termux
>
>$ apt install termux-exec
>
> With that, hashbangs like #!/usr/bin/pil automagically work! :)
> Most scripts now work on Linux and Termux without change.

I do have that installed, but the for the library on that first
line it does not work, I can leave the first (/usr/bin ...) as is, but have to 
change
the second entry (/usr/lib ...), otherwise I get a library-not-found error.

#!/usr/bin/picolisp /usr/lib/picolisp/lib.l

>> I looked into vip.l and some vt100 docs, but I find it really hard to
>> understand how e.g. "split window (qs)" is implemented with the vt100
>> escape sequences?
>> Could you elaborate on this a bit?
>
> The windows are maintained by Vip directly and explicitly. Simply positioning
> the cursor and writing stuff. Windows are just areas on the screen like
> everything else.
>
> As you see in @lib/vip.l, it uses only one single escape sequence for all such
> positionings:
>
>(de cup (Y X)
>   (prin "^[[" Y ";" X "H") )
>
> The other escapes, like character attributes, line clearing, show/hide cursor
> and switching between first and second screen are more or less just cosmetics.

Ok, thanks, maybe I understand that somehow now. 

If you do a (horizontal) split, then 2 +Window objects refer
to the same +Buffer object, so changes are synchronized.
The split is realized by dividing the current window hight by 2,
adjust the existing +Window object, and create a new one. 
Vip itself than draws the separating status line for each window
(except the "mini buffer", in Emacs terminology).

So its really just PicoLisp drawing on one terminal screen, not much
vt100 "intelligence" involved. Thats cool ;-)

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Vip without [N]curses

[Thorsten Jolitz]

Alexander Burger  writes:

Hi Alex,

> It should be a problem only on the first bootstrap. After you
> downloaded the *.s
> files once and got a running system, it will self-build from that time on.

I could fix this now, maybe a rather Archlinux specific problem.
Fist I figured out, that java verson shows 7, but in the ArchLinux repos
there are Open JDK/JRE 12 already. When I installed the newest version,
the verbose output gave the decisive hint, problem was the default Java
ENV.

This works now, did not have to download anything. 
Thanks 
Thorsten

PS
(3/4) Installiere jre-openjdk (12)
Default Java environment is already set to 'java-7-openjdk'
See 'archlinux-java help' to change it

=>

[tj@arch ~]$ archlinux-java help
archlinux-java 

COMMAND:
status  List installed Java environments and enabled one
get Return the short name of the Java environment
set as default
set   Force  as default
unset   Unset current default Java environment
fix Fix an invalid/broken default Java environment
configuration

[tj@arch ~]$ sudo archlinux-java get
java-7-openjdk
[tj@arch ~]$ sudo archlinux-java set java-12-openjdk

=>

[tj@arch picoLisp]$ (cd src64;make) [...]

[tj@arch picoLisp]$ pil +
: (version)
19.8.9
-> (19 8 9)
:


-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Vip without [N]curses

[Thorsten Jolitz]

George Orais  writes:

Hi Geo,

> I'm not sure but looking at those errors are coming from Java, so maybe
> you can try  the other two ways to build the 64bit:
>
> * Download one of the pre-generated "*.s" file packages
> * Build a 32-bit version first, and use the resulting bin/picolisp to
>  generate the "*.s" files:

I tried the same thing with the current stable version (19 6), same
error, so it's not really a problem with the dev version. But I never
ever had a problem with installing a new pil on this machine. 

If nobody else has this problem, seems to be a local (java) thing.
Since new PicoLisp versions a coming quite frequently, I would prefer
not to fall back to alternative build methods but rather fix my system
so that the normal build works again. But thanks for your hints anyway. 

I googled this now, this seems to be a well known Java problem with
versions of JRE and JDK which do not work together (the java class file
was compiled with a newer Java version than the installed runtime Java).

So not really a PicoLisp problem, sorry for the noise ...
Cheers 
Thorsten

 
PS
[tj@arch bin]$ ls
lisp  picoLisp  picoLisp-19.6.tgz  PilBox  pil-dev  pil-old  tmp
waffles
[tj@arch bin]$ cd picoLisp
[tj@arch picoLisp]$ (cd src64/;make)
/mkAsm x86-64 ".linux" .s Linux base "" ../lib/map  version.l glob.l
main.l gc.l apply.l flow.l sym.l subr.l big.l io.l db.l net.l err.l
sys/x86-64.linux.code.l
Exception in thread "main" java.lang.UnsupportedClassVersionError:
PicoLisp : Unsupported major.minor version 52.0
at java.lang.ClassLoader.defineClass1(Native Method) [...]
make: *** [Makefile:179: x86-64.linux.base.s] Fehler 1

[tj@arch picoLisp]$ java -version
java version "1.7.0_171"
OpenJDK Runtime Environment (IcedTea 2.6.13) (Arch Linux build
7.u171_2.6.13-1-x86_64)
OpenJDK 64-Bit Server VM (build 24.171-b02, mixed mode)




>  
>
> BR,
> Geo
> On Sunday, 11 August 2019, 09:54:04 pm GMT+9, Thorsten Jolitz
>  wrote: 
>
> Alexander Burger  writes:
>
> Hi Alex,
>
>> more and more I got frustrated with all the quirks of Ncurses (as
>> discussed here
>> and in IRC), so I decided to abandon them, and implement Vip directly
>> with ANSI
>> escape sequences (VT-100).
>>
>> To my surprise this turned out quite easy, and the result is both
>> smaller and
>> simpler!
>>
>> I tested on Termux, Tmux, XTerm and Linux Console. If anybody is
>> interested, it
>> is in the rolling picoLisp.tgz release :)
>
> thats very interesting, ncurses seemed to be a real curse so to say ...
> I have 
> - version (19 7 31) on Android/termux 
> - version (19 8 9) on Win10/wsl 
> now, both seem to have that change, and both could be installed and
> work (for termux I always have to adapt the first line in /bin/vip,
> i.e. replace /usr with termux $PREFIX, /data/data/com.termux/files/usr)
>
> But on my desktop ArchLinux machine with the newest PicoLisp dev
> download I get a make error, see PS1.
>
> Cheers 
> Thorsten
>
> PS 0
> I looked into vip.l and some vt100 docs, but I find it really hard to
> understand how e.g. "split window (qs)" is implemented with the vt100
> escape sequences?
> Could you elaborate on this a bit?
>
> PS 1
> On an up-to-date ArchLinux today, with the freshly downloaded dev
> version of PicoLisp:
>
> [tj@arch picoLisp]$ (cd src64/;make)
> /mkAsm x86-64 ".linux" .s Linux base "" ../lib/map  version.l glob.l
> main.l gc.l apply.l flow.l sym.l subr.l big.l io.l db.l net.l err.l
> sys/x86-64.linux.code.l
> Exception in thread "main" java.lang.UnsupportedClassVersionError:
> PicoLisp : Unsupported major.minor version 52.0
> at java.lang.ClassLoader.defineClass1(Native Method)
> at java.lang.ClassLoader.defineClass(ClassLoader.java:803)
> at
> java.security.SecureClassLoader.defineClass
> (SecureClassLoader.java:142)
> at java.net.URLClassLoader.defineClass(URLClassLoader.java:442)
> at java.net.URLClassLoader.access$100(URLClassLoader.java:64)
> at java.net.URLClassLoader$1.run(URLClassLoader.java:354)
> at java.net.URLClassLoader$1.run(URLClassLoader.java:348)
> at java.security.AccessController.doPrivileged(Native Method)
> at java.net.URLClassLoader.findClass(URLClassLoader.java:347)
> at java.lang.ClassLoader.loadClass(ClassLoader.java:425)
> at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:312)
> at java.lang.ClassLoader.loadClass(ClassLoader.java:358)
> at
> sun.launcher.LauncherHelper.checkAndLoadMain
> (LauncherHelper.java:482)
> make: *** [Makefile:179: x86-64.linux.base.s] Fehler 1
>
> PS 2
> [tj@arch picoLisp]$ lscpu
> Architektur:x86_64
> CPU Operationsmodus:32-bit, 64-bit
>
> -- 
> cheers,
> Thorsten

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Vip without [N]curses

[Thorsten Jolitz]

Alexander Burger  writes:

Hi Alex,

> more and more I got frustrated with all the quirks of Ncurses (as
> discussed here
> and in IRC), so I decided to abandon them, and implement Vip directly
> with ANSI
> escape sequences (VT-100).
>
> To my surprise this turned out quite easy, and the result is both
> smaller and
> simpler!
>
> I tested on Termux, Tmux, XTerm and Linux Console. If anybody is
> interested, it
> is in the rolling picoLisp.tgz release :)

thats very interesting, ncurses seemed to be a real curse so to say ...
I have 
- version (19 7 31) on Android/termux 
- version (19 8 9) on Win10/wsl 
now, both seem to have that change, and both could be installed and
work (for termux I always have to adapt the first line in /bin/vip,
i.e. replace /usr with termux $PREFIX, /data/data/com.termux/files/usr)

But on my desktop ArchLinux machine with the newest PicoLisp dev
download I get a make error, see PS1.

Cheers 
Thorsten

PS 0
I looked into vip.l and some vt100 docs, but I find it really hard to
understand how e.g. "split window (qs)" is implemented with the vt100
escape sequences?
Could you elaborate on this a bit?

PS 1
On an up-to-date ArchLinux today, with the freshly downloaded dev
version of PicoLisp:

[tj@arch picoLisp]$ (cd src64/;make)
/mkAsm x86-64 ".linux" .s Linux base "" ../lib/map  version.l glob.l
main.l gc.l apply.l flow.l sym.l subr.l big.l io.l db.l net.l err.l
sys/x86-64.linux.code.l
Exception in thread "main" java.lang.UnsupportedClassVersionError:
PicoLisp : Unsupported major.minor version 52.0
at java.lang.ClassLoader.defineClass1(Native Method)
at java.lang.ClassLoader.defineClass(ClassLoader.java:803)
at
java.security.SecureClassLoader.defineClass(SecureClassLoader.java:142)
at java.net.URLClassLoader.defineClass(URLClassLoader.java:442)
at java.net.URLClassLoader.access$100(URLClassLoader.java:64)
at java.net.URLClassLoader$1.run(URLClassLoader.java:354)
at java.net.URLClassLoader$1.run(URLClassLoader.java:348)
at java.security.AccessController.doPrivileged(Native Method)
at java.net.URLClassLoader.findClass(URLClassLoader.java:347)
at java.lang.ClassLoader.loadClass(ClassLoader.java:425)
at sun.misc.Launcher$AppClassLoader.loadClass(Launcher.java:312)
at java.lang.ClassLoader.loadClass(ClassLoader.java:358)
at
sun.launcher.LauncherHelper.checkAndLoadMain(LauncherHelper.java:482)
make: *** [Makefile:179: x86-64.linux.base.s] Fehler 1

PS 2
[tj@arch picoLisp]$ lscpu
Architektur: x86_64
CPU Operationsmodus: 32-bit, 64-bit

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Yet Another Emacs PicoLisp Mode - hybrid

[Thorsten Jolitz]

Grant Shangreaux 
writes:

Hello Grant,

> I would be happy to make
> a pull-request into the tj64 version if it is welcome, and Alexis I
> would be interested in helping to improve your version as well
> (especially since it is the one listed on MELPA).
>
> Would be glad to hear from Emacs users on the list, I know these threads
> pop up time to time, i've been trying to catch up :) !

go ahead an send me a pull request, from what I read in the thread I
conclude that its just merging tested code from another version into the
one I maintain on github, and this merge is already tested by you (and
others) so I can accept the PR without extra testing?

BTW
It would be nice to have even one more PicoLisp mode for Emacs!
There is this new concept of LSP out there now (Language Server
Protocol), with the basic idea that every language implements the server
side once, and every editor implements the client side once (in a
generic way). Then, a new editor mode for a language is just an adoption
of the LSP client of that editior for that specific language. And every editor
with an LSP client implementaton can easily offer a mode for a language that
has its LSP server implementation.

>From a PicoLisp point of view there are two pretty cool options here, I
think:

1. implement a Picolisp LSP Server implementation, enabling all those
editors out there to have their LSP PicoLisp modes (Emacs e.g. already
has two competing LSP client implementations and several new or
rewritten LSP major modes afaik)

2. implement a Vip LSP Client, thereby enabling language modes in Vip
for all that (growing numbers) of languages with LSP server
implemenations.

This is from Microsoft, but not evil at all, rather seems like the
future of editing modes. Maybe somebody finds this interesting too ... ;-)  

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Setup for Emacs org-mode?

[Thorsten Jolitz]

r...@tamos.net writes:

Hi Rick,

> Hi Lawrence,
>
> Welcome back to the list!
>
>> Anyone have the definitive setup they could share for picolisp on
>> Emacs org-mode?
>
> AFAIK, there is no "definitive setup" in Emacs.
>
>> There seems to be two picolisp-modes
>
> Three is the count I have.
>
> 1. picolisp-mode bundled in the picolisp distro.
>- N.B. hasn't been updated in years.
>
> 2. Thorsten/tj64 version
>- https://github.com/tj64/picolisp-mode
>- first commit dated Apr 2012

[...]

Thanks for summing this up, I'm a bit rusty here and I did not change
anything in the picolisp-mode repo for quite some time, but it worked
when I left it, and sometime ago somebody complained loudly about
ob-picolisp not working in org-mode, but I checked and answered (with a
pretty extensive example) on the org-mode and picolisp mailing lists,
and everything was still working fine, the problem was more a general
misunderstanding about the working of org-babel.

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: PicoLisp on Windows WSL first tryout fails ...

[Thorsten Jolitz]

Joe Bogner  writes:

Hey Joe,
>
> For WSL, you need to build picoLisp on a linux machine and then transfer
> it down. You can follow the download/install instructions, but here is
> generally what I did

maybe I don't really understand what you mean by transfer down, but in
my case (WSL with Suse) I just opened the Suse "App" and then installed
picolisp the same way I do on a standalone Linux System, and it works
just fine.

BTW isn't WSL the best thing (produced by microsoft) ever since sliced bread? 
;-)
Now I have Linux with Tmux, Emacs and PicoLisp on my average Win10
Notebook (only Tmux needs a little hack to work) - as an official
Windows App.

I really like it ...

> ON LINUX
> 1. wget https://software-lab.de/picoLisp.tgz
> 2. tar -zxvf picoLisp.tgz
> 3. cd picoLisp/src
> 4. make
> 5 cd ../src64
> 6. make
>
> ON WINDOWS BASH
> 1. wget https://software-lab.de/picoLisp.tgz 
> 2. tar -zvxf picoLisp.tgz
> 3. cd picoLisp
> 4. scp user@domain:/path/to/bin/picoLisp bin/picoLisp
>
> You should be able to then run ./pil
>
> The key here is to build on linux and then transfer down to your windows
> bash install. I used SCP to do the transfer
>
> Hope this helps. If you do not have access to a linux machine, you may
> want to try out vagrant on windows. I can help with that if you'd like
>
> Also, several of us are active on #picoLisp -- if you are unfamiliar
> with irc you can try here: https://webchat.freenode.net/
>
> NOTE: WSL has an issue with file locking with the picoLisp DB. I will
> look into that next
>
> Joe
>
> On Tue, Apr 17, 2018 at 7:25 AM, Alexander Burger
>  wrote:
>
>  Hi Philipp, Arie,
>
>  > pil is just a wrapper around picolisp, it loads a few libraries
>  etc as
>
>  Yes, but
>
>  > standard, but it relies on the intepreter being at
>  /usr/bin/picolisp,
>
>  This is not completely correct.
>
>  Note that there are two 'pil's in the distribution: One in bin/
>
>  #!/usr/bin/picolisp /usr/lib/picolisp/lib.l
>  (load "@lib/misc.l" "@lib/btree.l" "@lib/db.l" "@lib/pilog.l")
>
>  which indeed calls #!/usr/bin/picolisp, but this is not meant to be
>  called here.
>  It is intended to be copied to - or linked from - /usr/bin.
>
>  The other 'pil' looks different:
>
>  exec ${0%/*}/bin/picolisp ${0%/*}/lib.l @ext.l "$@"
>
>  and it is the main workhorse. It can be called locally
>
>  $ ./pil +
>
>  or with a relative or absolute path from anywhere
>
>  $ /foo/bar/pil +
>
>  and will always load everything from its local environment.
>
>  ♪♫ Alex
>
>  -- 
>  UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>
>

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Emacs mode can't run pil

[Thorsten Jolitz]

Lawrence Bottorff 
writes:

Hi Lawrence,
[just for the protocoll, I send you the identical answer on the
org-mode mailing list already].

> I'm looking at picolisp -- and wondering how it works, or better, why it
> doesn't really work work with babel. First problem, I couldn't get any
> form of picolisp to work in Emacs -- until I stopped starting Emacs with
> [...]
> I don't mean to complain or sound negative, but picolisp as is can't
> really be included as a babel language, can it? Maybe it worked once,
> but doesn't now?

sometimes the bug actually sits in front of the computer, as we all know
;-)

Apparently you are not aware of the 'session' concept of org source
blocks (please refer to the org manual). 
This is nothing specific to ob-picolisp, but holds for all ob languages
that do allow for sessions:

,
| * Picolisp scr-block test
| 
| #+BEGIN_SRC picolisp  :session pil1
|  (setq X1 (+ 3 4))
| #+END_SRC
| 
| #+results:
| : 7
| 
| 
| #+BEGIN_SRC picolisp  :session pil1
| (setq X2 (+ X1 1))
| #+END_SRC
| 
| #+results:
| : 8
| 
| 
| #+BEGIN_SRC picolisp  :session pil1
| (de foo1 (X) (+ X 2))
| #+END_SRC
| 
| #+results:
| : foo1
| 
| #+BEGIN_SRC picolisp  :session pil1 :results raw
| (setq X3 (foo1 8))
| #+END_SRC
| 
| #+results:
| 10
`

And, with a session, you do have a related interactive repl buffer in
Emacs called "pil1", thats reflects all evaluations of the code blocks
in the org file, and allows for user input just like the PicoLisp repl:

,
| (setq X1 (+ 3 4))
| 'org-babel-picolisp-eoe
| : -> 7
| : -> org-babel-picolisp-eoe
| : (setq X2 (+ X1 1))
| 'org-babel-picolisp-eoe
| -> 8
| : -> org-babel-picolisp-eoe
| : (de foo1 (X) (+ X 2))
| 'org-babel-picolisp-eoe
| -> foo1
| : -> org-babel-picolisp-eoe
| : (setq X3 (foo1 8))
| 'org-babel-picolisp-eoe
| -> 10
| : -> org-babel-picolisp-eoe
| : 
`

Hope that helps

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Chinese input causes mess in the REPL

[Thorsten Jolitz]

Danilo Kordic 
writes:

>   GNU Emacs can be used as a line editor.  Execute elisp expression
> ``(term "/absolute/path/to/pil")'' then activate `term-line-mode' with
> ``C-C C-j''.
>
>   Is this of any help?  How much does it count :) ?

Cool! I did not know this ...

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Which emacs mode to use?

[Thorsten Jolitz]

Tim Johnson  writes:

Hi Tim,

> I note that my emacs (25.1.1 on ubuntu) will install and provide 
> 'picolisp-mode as an elpa package.

I'm not the author of this mode, but I somehow took over maintainership
(more or less) and did some contributions to the mode (in my github repo
tj64: https://github.com/tj64/picolisp-mode).

But I have no idea who added this as an elpa package (elpa or melpa?)

> In addition, my installation of picolisp has at :
> /usr/share/emacs/site-lisp/picolisp
> tsm.el, inferior-picolisp.el, and picolisp.el (providing mode
> 'picolisp)
>
> I'd welcome opinions as to which I should use.

I think only diffing the various versions can help to find out, which
one is the newest. IIRC Alex simply added new versions of the mode to
the picolisp distribution, when I sent them to him once in a while. But
for a long time, I did not touch these libraries anymore.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Method undefined - why?

[Thorsten Jolitz]

Thorsten Jolitz <tjol...@gmail.com>
writes:

> so somehow my understanding of read macros is false here:
>
> (class ~(any (pack '+ ClsNm)))

Grrr ... of course when reading (de foo (ClsNm) ...) ClsNm is still NIL,
so the outcome of the above is (class +) which can't work. Sorry for the
noise.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Method undefined - why?

[Thorsten Jolitz]

Thorsten Jolitz <tjol...@gmail.com>
writes:

> Alexander Burger <a...@software-lab.de> writes:

Ok, this works ;-)

: (de foo2 (ClsNm MethNm) (class +Bar) (dm plus> (X) (+ 1 X)))
-> foo2
: (foo2)
-> plus>
: (setq B (new '(+Bar)))
-> $177641167640474
: (plus> B 3)
-> 4
'))

so somehow my understanding of read macros is false here:

(class ~(any (pack '+ ClsNm)))

> Hi Alex,
>
>>> : (de +Test
>>>(T (Hi) (=: hi Hi))
>>>(hi> (Nm) (or (text (: hi) Nm) "Dear Sir or Madam,")) )
>>> -> +Test
>>
>> .. while this is half of the lunch ...
>>
>>> : (hi> Foo "Alex")  
>>> !? (hi> Foo "Alex")
>>> hi> -- Undefined  # => WHY?
>>
>> .. it is better (as Joe Bogner suggested) to use 'dm'.
>>
>> The reason is that 'dm' does a little more: It also defines the symbol
>> 'hi>' to
>> behave as a sender of messages to objects, equivalent to
>>
>>: (setq hi> meth)
>>-> 22951574276
>>
>> With that, the following works
>>
>>: (hi> Foo "Thorsten")
>
> Ok, I see.
>
> But, in a source file it's obvious how to use (class) and (dm)
>
> (class 1 ...)
> (dm ...)
> (dm ...)
>
> (class 2 ...)
> (dm ...)
> (dm ...)
>
> The reader seems to remember which class was defined last, and
> associates the following methods to that class.
>
> How to use (class) and (dm) in a program is not that obvious for me:
>
> :(de foo (ClsNm MethNm) (class ~(any (pack '+ ClsNm))) (dm ~(any (pack
> MethNm ">") (X) (+ 1 X
> -> foo
> : (foo "Bar" "plus")
> !? (val (setq *Class (car Lst)))
> 270136 -- Variable expected
> '
>
> Would you rather write the classical definitions to a file and then load
> that?

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Method undefined - why?

[Thorsten Jolitz]

Alexander Burger  writes:

Hi Alex,

>> : (de +Test
>>(T (Hi) (=: hi Hi))
>>(hi> (Nm) (or (text (: hi) Nm) "Dear Sir or Madam,")) )
>> -> +Test
>
> .. while this is half of the lunch ...
>
>> : (hi> Foo "Alex")  
>> !? (hi> Foo "Alex")
>> hi> -- Undefined  # => WHY?
>
> .. it is better (as Joe Bogner suggested) to use 'dm'.
>
> The reason is that 'dm' does a little more: It also defines the symbol
> 'hi>' to
> behave as a sender of messages to objects, equivalent to
>
>: (setq hi> meth)
>-> 22951574276
>
> With that, the following works
>
>: (hi> Foo "Thorsten")

Ok, I see.

But, in a source file it's obvious how to use (class) and (dm)

(class 1 ...)
(dm ...)
(dm ...)

(class 2 ...)
(dm ...)
(dm ...)

The reader seems to remember which class was defined last, and
associates the following methods to that class.

How to use (class) and (dm) in a program is not that obvious for me:

:(de foo (ClsNm MethNm) (class ~(any (pack '+ ClsNm))) (dm ~(any (pack
MethNm ">") (X) (+ 1 X
-> foo
: (foo "Bar" "plus")
!? (val (setq *Class (car Lst)))
270136 -- Variable expected
'

Would you rather write the classical definitions to a file and then load
that?

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Method undefined - why?

[Thorsten Jolitz]

Joe Bogner <joebog...@gmail.com> writes:

HiJoe,

> Hi Thorsen, what about using (class) and (dm) instead?
>
> joebo@joebo:~/dev/picoLisp$ pil + 
> : (class +Test) 
> -> +Test 
> : (dm T (Hi) (=: hi Hi)) 
> -> T 
> : (dm hi> (Nm) (or (text (: hi) Nm) "Dear Sir or Madam,")) 
> -> hi> 
> : (setq Foo (new '(+Test) "Hi @1")) 
> -> $177463554467256 
> : (hi> Foo "Alex") 
> -> "Hi Alex" 

yes, thats the more classical way, but I want to dynamically build the
class definition, so using (de) seems easier.

And I find it a bit strange that everything else works and looks good,
but not the method call in the code below.

This works

>  : (methods Foo)
>  -> ((T . +Test) (hi> . +Test))

but this not

>  : (hi> Foo "Alex")
>  !? (hi> Foo "Alex")
>  hi> -- Undefined

?

Should it work?
Does it work for others, but not for me?

This would be my questions ... ;-)

> On Thu, Mar 23, 2017 at 3:17 PM, Thorsten Jolitz
> <tjol...@gmail.com> wrote:
>
>  Hi List,
>
>  playing around a bit wih Pil classes/object, here is something I
>  don't
>  understand:
>
>  #+BEGIN_SRC picolisp
>
>  : (de +Test
>  (T (Hi) (=: hi Hi))
>  (hi> (Nm) (or (text (: hi) Nm) "Dear Sir or Madam,")) )
>  -> +Test
>
>  : (setq Foo (new '(+Test) "Hi @1"))
>  -> $176777024346263
>
>  : (hi> Foo "Alex")
>  !? (hi> Foo "Alex")
>  hi> -- Undefined # => WHY?
>  ?
>
>  : (getl Foo)
>  -> (("Hi @1" . hi))
>
>  : (can 'hi>)
>  -> ((hi> . +Test))
>
>  : (type Foo)
>  -> (+Test)
>
>  : (show Foo)
>  $176777024346263 (+Test)
>  hi "Hi @1"
>  -> $176777024346263
>
>  : (methods Foo)
>  -> ((T . +Test) (hi> . +Test))
>
>  #+END_SRC
>
>  Is that strange, or is the problem sitting in front of the computer?
>
>  --
>  cheers,
>  Thorsten
>
>  --
>  UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
>
>

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Method undefined - why?

[Thorsten Jolitz]

Hi List, 

playing around a bit wih Pil classes/object, here is something I don't
understand:


#+BEGIN_SRC picolisp  

: (de +Test
   (T (Hi) (=: hi Hi))
   (hi> (Nm) (or (text (: hi) Nm) "Dear Sir or Madam,")) )
-> +Test

: (setq Foo (new '(+Test) "Hi @1"))
-> $176777024346263

: (hi> Foo "Alex")  
!? (hi> Foo "Alex")
hi> -- Undefined  # => WHY?
?

: (getl Foo)
-> (("Hi @1" . hi))

: (can 'hi>)
-> ((hi> . +Test))

: (type Foo)
-> (+Test)

: (show Foo)
$176777024346263 (+Test)
   hi "Hi @1"
-> $176777024346263

: (methods Foo)
-> ((T . +Test) (hi> . +Test))

#+END_SRC


Is that strange, or is the problem sitting in front of the computer?

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: JavaCode with static class methods

[Thorsten Jolitz]

Alexander Burger  writes:

Hi Joe, Hi Alex,

>> I'm experimenting a bit with the 'java' function, and the StringBuilder
>> ...
>> (java "java.lang.Math" 'sqrt 4)
>
> It needs a different call:

>: (java "java.lang.Math" 'sqrt  (-6 . 64.0))

thanks for the hints, so actually I was pretty close, just ignorant
about the peculiarities of those numeric arguments.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

JavaCode with static class methods

[Thorsten Jolitz]

Hi List, 

I'm experimenting a bit with the 'java' function, and the StringBuilder
and GregorianCalendar example from the JavaCode Wiki entry work fine,
but I wonder how to use a constructorless class like e.g. java.lang.Math
with its static class methods?

#+NAME: JavaCode - none of these work
#+BEGIN_SRC picolisp
(java "java.lang.Math" 'sqrt 4)
(java java.lang.Math 'sqrt 4)
(java 'java.lang.Math 'sqrt 4)

(setq M (java "java.lang.Math" T))  // returns null, no constructor
#+END_SRC

Thanks for any hints.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: PicoLisp as Femto extension language

[Thorsten Jolitz]

Thorsten Jolitz <tjol...@gmail.com>
writes:

Hi List, 

[funny enough, in GNUS Ricks posts to this thread don't show up for me,
while I can see them in the www.mail-archive.com, so I follow up to my
own post instead of his.]

My original idea was to compile Femto as shared library, create FFI
wrapper libraries for PicoLisp, and have buffers instead of only a
single line to edit and evaluate PicoLisp code. A bit naive probably.

Replacing Femto-lisp with PicoLisp as extension language seems to be the
way to go, but is over my head I'm afraid. I thought, with the core
editor functions already defined in C (using the gap buffer concept), it
would be just about wrapping these basic functions (and data structures)
in PicoLisp and one could start to write cool interactive editor
features based on this Core libraries, just like in Emacs (Lisp).

But solving the fundamental communication problem between the two
systems is something else ...

Thanks for the anwsers, Rick.


> Hi List,
> recently I found out about Femto, a minimal Emacs implementation in less
> than 2k lines of C (including ncurses, though):
>
> ,
> | git clone https://github.com/hughbarney/femto.git
> `
>
> ,
> | Femto is an extended version of Atto Emacs.
> | 
> | In Defining Atto as the lowest functional Emacs I have had to
> | consider the essential feature set that makes Emacs, 'Emacs'. [...]
> `
>
> ,
> |  Goals of Femto Emacs
> | 
> |   * To extend Atto Emacs with filename completion, dired, buffer
> | menu and the ability to run a shell command.
> |   * Provide a rich level of functionality in the smallest amount of
> | code
> |   * Femto Emacs will eventually have it own lisp extension
> | languauge.
> |   * Be easy to understand without extensive study (to encourage
> | further experimentation).
> `
>
> There is already a project underway using FemTo-Lisp as extension
> language
>
> ,
> | git clone https://github.com/FemtoEmacs/Femto-Emacs.git
> `
>
> but I thought this would be a perfect use case for an 'AW-style' FFI
> library making use of 'native'. Then, PicoLisp could have its own
> (Window/Buffer)-Editor, not just the LineEditor, with all the core
> Editor-Functions just wrappers around Femto's C-functions.
>
> But lacking C skills have me stuck in the very beginning: how to turn
> this C application into a shared library for 'native', does it even
> make sense?
>
> I guess I should:
>
>  1. create a shared library for all .c files in the repo except main.c
>  2. create the application (main.o) linking the shared library? 
>
> But then, the shared library by itself does not make much sense, it
> would need the running C Application to do some useful work (?) 
>
> So the queston seems rather how to embed PicoLisp in this C Application,
> and make Femto extensible in PicoLisp (like Emacs is extensible in Emacs
> Lisp) by having FFI wrappers for all core editor functions?
>
> The 
>
> ,
> | /home/tj/gitclone/Femto-Emacs/femtolisp
> `
>
> subdirectory makes the general "Lisp integration task" look rather
> complicated, but maybe there is (like so often) a really simple solution
> with PicoLisp?

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

PicoLisp as Femto extension language

[Thorsten Jolitz]

Hi List,
recently I found out about Femto, a minimal Emacs implementation in less
than 2k lines of C (including ncurses, though):

,
| git clone https://github.com/hughbarney/femto.git
`

,
| Femto is an extended version of Atto Emacs.
| 
| In Defining Atto as the lowest functional Emacs I have had to
| consider the essential feature set that makes Emacs, 'Emacs'. [...]
`

,
|  Goals of Femto Emacs
| 
|   * To extend Atto Emacs with filename completion, dired, buffer
| menu and the ability to run a shell command.
|   * Provide a rich level of functionality in the smallest amount of
| code
|   * Femto Emacs will eventually have it own lisp extension
| languauge.
|   * Be easy to understand without extensive study (to encourage
| further experimentation).
`

There is already a project underway using FemTo-Lisp as extension
language

,
| git clone https://github.com/FemtoEmacs/Femto-Emacs.git
`

but I thought this would be a perfect use case for an 'AW-style' FFI
library making use of 'native'. Then, PicoLisp could have its own
(Window/Buffer)-Editor, not just the LineEditor, with all the core
Editor-Functions just wrappers around Femto's C-functions.

But lacking C skills have me stuck in the very beginning: how to turn
this C application into a shared library for 'native', does it even
make sense?

I guess I should:

 1. create a shared library for all .c files in the repo except main.c
 2. create the application (main.o) linking the shared library? 

But then, the shared library by itself does not make much sense, it
would need the running C Application to do some useful work (?) 

So the queston seems rather how to embed PicoLisp in this C Application,
and make Femto extensible in PicoLisp (like Emacs is extensible in Emacs
Lisp) by having FFI wrappers for all core editor functions?

The 

,
| /home/tj/gitclone/Femto-Emacs/femtolisp
`

subdirectory makes the general "Lisp integration task" look rather
complicated, but maybe there is (like so often) a really simple solution
with PicoLisp?

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Extending the wiki markup syntax

[Thorsten Jolitz]

Alexander Burger  writes:

Hi Alex, Hi List,

thanks for the cool tractatus and the as always very interesting views
about wiki markup. 

I don't know if I should mention it, because it's in a raw state and I
don't have much time to polish it now, but as you know there does exist
already an alternative wiki syntax: (Emacs) Org-mode.  Note that
Org-mode is not just dumb wiki syntax like Markdown or so, under the
syntax surface is another very powerful and featurerich Lisp application
that can do amazing things, see

,
| orgmode.org
`

The gitrepo of my iOrg project can be found here:

,
| https://github.com/tj64/iorg
`

Its the PicoLisp Wiki, but calls the Org-mode exporter to render the
(Org-mode) markup, instead of using the PicoLisp render function from
Alex. This works!

There is even support for emacs-w3m mode for editing the wiki pages,
i.e. instead of opening a small text buffer for editing an textarea (the usual
behaviour), emacs-w3m opens an Org-mode buffer for editing iOrg wiki
pages. The changes from the Org-mode buffer are then committed to the
web application (the wiki) just like any other html form that is edited
with emacs-w3m.

I haven't touched it for a long time, but this part of the project
worked very well IIRC. Thank to Alex for all his help back then!

The project was more ambitious, there is a pretty far developed web
interface for org-mode trees/nodes, that almost worked, but then I had
to move to other shores and could not finish it. But I'm still
interested in this project, because Emacs Org-mode is the most powerful
static website/document generator anyone can think of, and using Emacs
Org-mode as syntax for the PicoLisp wiki would combine my two favorite
pieces of software.

Under the hood, an Org-mode document is a nested (Emacs Lisp) list,
which is of course very convenient from a PicoLisp point of
view. Unfortunately, the Org-mode parser (org-element.el) makes use of
some unique Emacs Lisp features (circular lists, text properties ...)
that complicated my life tremendously when trying to make the two apps
talk to each other.

So if extending the PicoLisp markup is on the agenda right now - why not
try a revival of the iOrg project? I would be happy about feedback and
contributions.

> True, but again it holds what I said about too much abstraction. I've
> shot myself into the food this way many times.

That made me really curios - was it vegetables, meat, or rather seafood?
;-)

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Emacs REPL not "full-featured"?

[Thorsten Jolitz]

Lawrence Bottorff 
writes:

> I've got a REPL started in Emacs (called from ELPA version of picolisp
> major mode with run-picolisp), but I'm noticing it doesn't have such
> features as more, edit, what, etc. Starting picolisp from the command
> line with pil is however "full-featured". What am I missing here?

IIRC even when all libraries are loaded, there remain a few things that
don't work (correctly), e.g.  (doc 'pp)

1. Emacs REPL (prints unreadable results):

,
| : (doc 'pp)
| (doc 'pp)
| 
1,03kbloadedinstead,itshouldbearangeofnumberswhicharetriedinturn.Whenvarisgiven,itisboundtotheportnumber.:(port0'A)#Allocatefreeport->4:A->1034#Got1034:(port(4000.4008)'A)#Tryoneoftheseports->5:A->4002(pp
| 'sym) -> sym (pp 'sym 'cls) -> sym (pp '(sym . cls)) ->
| 
symPretty-printsthefunctionormethoddefinitionofsym.Theoutputformatwouldregeneratethatsamedefinitionwhenreadandexecuted.Seealsopretty,debugandvi.:(pp'tab)(detab(Lst.@)(forNLst(letV(next)≪
| ↑ ↓ Viewing libgpm: zero screen dimension, assuming 80x25.
`

2. pil REPL (opens W3M on the console):

,
| (pp 'sym) -> sym
| (pp 'sym 'cls) -> sym
| (pp '(sym . cls)) -> sym
| Pretty-prints the function or method definition of sym. The
| output format would regenerate that same definition when read
| and executed. See also pretty, debug and vi.
| 
| : (pp 'tab)
| (de tab (Lst . @)
|(for N Lst
|   (let V (next)
|  (and (gt0 N) (space (- N (length V
|  (prin V)
|  (and
| (lt0 N)
| (space (- 0 N (length V))) ) ) )
|(prinl) )
| -> tab
`

But I guess one could configure Emacs to show the 'doc' results in
emacs-w3m or eww instead of the *picolisp* buffer itself.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: org mode picolisp?

[Thorsten Jolitz]

"O.Hamann"  writes:

Hi All,

as the author of of ob-picolisp I should make some smart comment now,
but I'm afraid I cannot add much to the discussion, the former answers
explained it all very well. As the answers show, the library still seems
to work fine, so the original problem looks not really ob-picolisp
specific but rather like a problem with using Org code blocks correctly.

Here is a link to the relevant manual section:
,
| http://orgmode.org/manual/Working-With-Source-Code.html
`

and here one to the ob-picolisp doc:

,
| http://orgmode.org/worg/org-contrib/babel/languages/ob-doc-picolisp.html
`

> I'm not really sure about namespaces, so I can't compare.
> Manual explains :session this way:
> ==
> The :session header argument starts a (possibly named) session for an
> interpreted language where the interpreter’s state is preserved. All
> code blocks sharing the same name are exectuted by the same
> interpreter process. By default, a session is not started. 
> ==
> (fetched from: http://orgmode.org/manual/session.html)

This is not related to PicoLisp namespaces but rather to the
communication between Emacs and PicoLisp. 

 - without session, Emacs/Org-mode starts a new PicoLisp instance for
   every query, so e.g. setting X to 3 in one code block and then adding 2
   to X in the next code block won't work, since the newly started
   instance for the second call knows nothing about X.
 - with session, Emacs starts one PicoLisp instance and keeps it
   alive, so that state is preserved. The call from the second code
   block goes to the same PicoLisp instance where symbol X was set to
   3 before, so the state of X is preserved and can be used for (+
   X 2) e.g.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Single File Function Reference

[Thorsten Jolitz]

 writes:

Hi List,

> There are two books about picolisp available, one containing also the
> reference [...] Thought I'm not absolutely sure that the reference in
> there is on the newest current state, maybe someone else can
> confirm/deny this.

The function reference reflects the current state at the time of
publishing of the books. I guess that whats in there is still 99% valid,
and only a few tiny changes and corrections have be done. But some new
functions may be missing in the ref.

One day there should be a 2nd edition of these books, since so many
exciting things have been and are happening in the PicoLisp universe
that should and could be added to the book. E.g.:

 - PicoLisp on Hardware
 - PicoLisp and Java
 - Emacs and PicoLisp
 - PicoLisp Ports
 - etc, etc ...
 
Don't have the time for it though. Maybe in the future ... or somebody
else takes over the task?

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Let and Glue

[Thorsten Jolitz]

Rick Hanson cryptor...@gmail.com writes:

Hi Rick,

 Here is a problem (again!) with using a pil backquote expression,
 where the user, like you or me, is stuck on CL-unquote thinking.

 Let's call g again, but now we switch the places of the dates and
 times.  In this case, we might expect the answer now to be No;
 however, the answer will remain Yes.

 : (g Is $S1$ before $S2$?
The answer is
`(if (and
(= $D2$ $D1$)
(= $T2$ $T1$) )
Yes
No ) )
 - Is \1999-09-09 20:12:05\ before \2000-03-04 18:03:03\? The
 answer is \Yes\

 Why?

 Because when the backquote expression gets resolved by the reader, the
 symbols $D1$, $D2$, $T1$, and $T2$ are not bound to any values
 (i.e. will evaluate to NIL).  And since

 (= NIL NIL)

 evals to T, g will always see Yes as its last argument.

 Sorry.  That means you have to come up with another way to eval code
 in your DSL (i.e the stuff under g).

 I hope this make sense.  This is what Alex has been trying to tell us
 all along.  (I only now just realized this; so you can see how far
 I've come along: not much.)

thanks for your improvements of my rather crude function and for the
final hint that using (date) in the examples was a rather bad idea since
(date) works with NIL args too, it just gives different results.

This backquote/comma mechanism in Emacs Lisp seemed so natural to me I
never really thought about the way it works. So I should rather think
about function calls then read syntax. But maybe not in public, I don't
want to drive more people off the mailing list ;-)

However, I will use your factorized code as starting point for new
investigations, thanks again! 

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Let and Glue

[Thorsten Jolitz]

Alexander Burger a...@software-lab.de writes:

Hi Alex,

sorry for being thick as a brick ...

 Well, 'bind' is the evaluating version of 'let'. It could be used to
 implement the outher functions.

I'll check that out.

 What you probably mean is something different. It is a different way of
 interpreting the arguments to a function. It has nothing to do with how
 the symbols are bound, and nothing with the reader:

   (de g Args
  (glue  
(mapcar
   '((X)
  (if (pair X) (val (car X)) X) )
   Args ) ) )


   (setq Temp 33)

   (g Current \temperature\ in Berlin is (Temp) °Celsius \(its really
   hot!\) )


- Current \temperature\ in Berlin is 33 °Celsius (its really hot!)

This comes pretty close to what I was looking for, thanks. The only
drawback is that normal parens (and double quotes) are very common in
text so a lot of escaping would be necessary. Something like this:

   (g Current temperature in Berlin is {Temp} \{°Celsius\} (its
   really hot!) )

resulting in:

- Current \temperature\ in Berlin is 33 {°Celsius} (its really hot!)

would be much more comfortable as syntax, but harder to implement.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Let and Glue

[Thorsten Jolitz]

Thorsten Jolitz tjol...@gmail.com
writes:

Hi Alex,

 When I quote the reference:

 ,
 | A single backquote character ` will cause the reader to evaluate
 | the following expression, and return the result.
 | 
 | : '(a `(+ 1 2 3) z)
 | - (a 6 z)
 `

 it looks to me as if the difference between PicoLisp and others (like
 Emacs Lisp) must be rather in the 'let than in the read macros, since
 the combination quote/backquote in PicoLisp is equivalent to the
 combination backquote/comma in other Lisps, and works the same, except
 inside let bindings:

 PicoLisp:

 ,
 | $ pil +
 | : (let X (+ 2 3) '(3 4 `X))
 | - (3 4 NIL)
 | : '(3 4 `(+ 2 3))
 | - (3 4 5)
 `

 vs Emacs Lisp:

 ,
 | (let ((X (+ 2 3))) `(3 4 ,X))
 | - (3 4 5)
 | 
 | `(3 4 ,(+ 2 3))
 | - (3 4 5)
 `

 How would the above behavior inside let bindings be achieved in
 PicoLisp? 

Thinking about it it seems that Read Macros are equivalent, but the
readers work differently: 

 - PicoLisp :: read without eval, except when encountering a read
  macro
 - Emacs Lisp :: read without eval, except in special situations (local
  assignments, read macros, ???)

So assignments inside 'let are treated special too, like implicit read
macros, and evaluated during reading.

Not only that I have a concrete use case for that behavior right now
(wrt to let and job), I find that very useful in general. 

The only advantage of the current PicoLisp behavior I can see is speed
of reading (because of less assignments in the process), but there
might be others of course. Would it be conveivable to have 'evaluating
versions' of some of the binding environments in PicoLisp too?

,
| let, let?, bind, recur, with, for, job, use ...
`

Here a possible use case:

Imagine these two functions are given:

,
| (de g  Txt
|(glue   Txt) ) 
| 
| (de getCurrTemp (Url) # should retrieve temperature from url
|   33 )
`

than even a non-programmer could specify a dynamic report in PicoLisp
by writing:

,
| (job
|   CurrTemp (getCurrTemp /path/to/current/temp/in/berlin)
| 
|   (g Current temperature in Berlin is `CurrTemp °Celsius) )
`
  
i.e. by using PicoLisp as a kind of smart/dynamic markup language
instead of defining and calling functions like a programmer. Of course
this could easily be done by 'normal' programming, but I'm looking for
something like the above ...   

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Let and Glue

[Thorsten Jolitz]

Alexander Burger a...@software-lab.de writes:

Hi Alex,

uhh ... bad timing from my side, writing more confused question while
you are actually answering them at the same time...

 Either
: (let X (+ 2 3) (list 3 4 X))  # I would prefer a simple 'list'
- (3 4 5)

 or
: (let X (+ 2 3) (fill (3 4 X) 'X))  # Rick tried 'fill' too
- (3 4 5)

 or

: (let @X (+ 2 3) (fill (3 4 @X)))
- (3 4 5)

 or whatever function you like to build the list.


 Note that also in CL and Emacs the backquote function builds a list at
 *runtime*, just like 'list' or 'fill' do.

I have to try fill, and think more about the whole issue, but the last
version with @X looks pretty close to what I'm looking for.

Again, thanks for the explanations (and patience)!

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Let and Glue

[Thorsten Jolitz]

Thorsten Jolitz tjol...@gmail.com
writes:

Hi List,

 This comes pretty close to what I was looking for, thanks. The only
 drawback is that normal parens (and double quotes) are very common in
 text so a lot of escaping would be necessary. Something like this:

(g Current temperature in Berlin is {Temp} \{°Celsius\} (its
really hot!) )

 resulting in:

 - Current \temperature\ in Berlin is 33 {°Celsius} (its really hot!)

 would be much more comfortable as syntax, but harder to implement.

I implemented something close to my goal, with $xyz$ syntax for
arguments to be evaluated:

,
| (de f Args
|   (mapcar
|  '((Arg)
|(if (atom Arg)
|   (let (ChopArgs (chop Arg)
| ArgsTail (tail 2 ChopArgs)
| PunctChars (chop .,;:?!\'_-{[]} )
| ArgLst NIL
| Punct NIL )
|  (ifn (and
| (= (car ChopArgs) $)
| (or
|(= (last ChopArgs) $)
|(and
|   (= (car ArgsTail) $)
|   (member (last ArgsTail) PunctChars)
|   (setq Punct (last ArgsTail))
|   (setq ChopArgs (head -1 ChopArgs)
|  ) ) ) )
| Arg
| (pack
|(val
|   (intern
|  (pack
| (tail -1 (head -1 ChopArgs)) ) ) )
|Punct ) ) )
|   (f Arg) ) ) # use recur/recurse instead?
|   Args ) )
`

using it returns an almost perfect list result:

,
| (setq
|D1 (date 1999 09 09)
|D2 (date 2000 03 04) 
|T1 (time 20 12 05)
|T2 (time 18 03 03)
|S1 (stamp D1 T1)
|S2 (stamp D2 T2) )
| 
| (f Is $S1$ before $S2$?
|The answer is
|`(if (and
|(= $D1$ $D2$)
|(= $T1$ $T2$) )
|Yes
|No ) )
| 
| 
| - (Is 1999-09-09 20:12:05 before 2000-03-04 18:03:03? The answer is 
Yes)
`

only that

  2000-03-04 18:03:03?

should rather be

 2000-03-04 18:03:03?

I did not yet make it to write a function g in the spirit of:

,
| (de g Args
|   (glue   (f Args)) ) # does not work
`

that then returns a single string:


,
| Is \1999-09-09 20:12:05\ before \2000-03-04 18:03:03\? The answer is 
\Yes\
`


-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Let and Glue

[Thorsten Jolitz]

Alexander Burger a...@software-lab.de writes:

Hi Alex,

 I wonder how I can use local variable X inside of the 'glue'
 argument list:
 
 ,
 | : (let X (+ 3 4) X)
 | - 7
 | 
 | : (let X (+ 3 4) (glue   '(Number  X `(- 4 9
 | - Number X -5
 | 
 | : (let X (+ 3 4) (glue   '(Number  `X `(- 4 9
 | - Number  -5
 `
 
 X does not seem defined in the last expression although its surrounded
 by (let X (+ 3 4) ...). 

 Hmm, read-macros seem indeed a lot misunderstood. NEVER use a read-macro
 to insert values which are defined at *run*time! As the name says, they
 are evaulated at *read* time!


 You could do this:

: (let X (+ 3 4) (glue   (list Number  X `(- 4 9

 (the second read-macro is OK, as the (- 4 9) may be evaluated at
 read-time)

Ah, I see, thanks.  

I think the (working) second read-macro fooled me a bit, and I did not
have separation of read- and runtime very clear. Well, nice that, as
always with PicoLisp, there is an easy way to make it work ;-)

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Let and Glue

[Thorsten Jolitz]

Rick Hanson cryptor...@gmail.com writes:

 Yeah, sorry.  I had this on the mind -- a different animal altogether.

   $ sbcl
   * (let ((X (+ 3 4))) `(hello ,X ,(- X 9)))
   (HELLO 7 -2)

Maybe I was confused by Emacs Lisp a bit too:

,
| (let ((x (+ 3 4))
|   (y (+ 5 6)))
|   `(+ 5 x ,y))
| 
| - (+ 5 x 11)
`

 Thanks, Alex, for taking the time and writing a very nice explanation.
 I believe I understand it, but I will re-read and ponder it more.

1+
Many thanks for that in detail explanation, its actually a bit confusing
especially when one is used to other Lisps too.  

When I quote the reference:

,
| A single backquote character ` will cause the reader to evaluate
| the following expression, and return the result.
| 
| : '(a `(+ 1 2 3) z)
| - (a 6 z)
| 
| A tilde character ~ inside a list will cause the reader to evaluate
| the following expression, and (destructively) splice the result into
| the list.
| 
| : '(a b c ~(list 'd 'e 'f) g h i)
| - (a b c d e f g h i)
`

it looks to me as if the difference between PicoLisp and others (like
Emacs Lisp) must be rather in the 'let than in the read macros, since
the combination quote/backquote in PicoLisp is equivalent to the
combination backquote/comma in other Lisps, and works the same, except
inside let bindings:

PicoLisp:

,
| $ pil +
| : (let X (+ 2 3) '(3 4 `X))
| - (3 4 NIL)
| : '(3 4 `(+ 2 3))
| - (3 4 5)
`

vs Emacs Lisp:

,
| (let ((X (+ 2 3))) `(3 4 ,X))
| - (3 4 5)
| 
| `(3 4 ,(+ 2 3))
| - (3 4 5)
`

How would the above behavior inside let bindings be achieved in
PicoLisp? 


-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Let and Glue

[Thorsten Jolitz]

Rick Hanson cryptor...@gmail.com writes:

Hi Rick,

 Reading your other post I was wondering how it worked for you and
 thought it must be because of debug mode, and well ... correctly
 guessed ;-)

 Hi Thorsten!  Yes. :) And btw thanks for picolisp-mode.  Please count
 me as a happy user.

Oh, you have to thanks Guillermo R. Palavecino in the first place, since
he wrote it:

,
| ;; picolisp-mode: Major mode to edit picoLisp.
| ;; Version: 1.3
| 
| ;;; Copyright (c) 2009, Guillermo R. Palavecino
| ;;; Copyright (c) 2011, 2012 Thorsten Jolitz
`

I just became the (rather inactive) maintainer a few years ago and did
some enhancements later on, expecially to inferior-picolisp.el.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Open Sound Control

[Thorsten Jolitz]

Robert Herman rpjher...@gmail.com
writes:

Hi Rob,

 I have come the long way round to PicoLisp, and I have been tinkering
 with livecoding (audio/video, not just programming) for fun.

 I started with fluxus: www.pawfal.org/fluxus/
 It is a great environment where you code and 3D objects show behind
 your code, and you can drive their parameters from an audio feed or
 file. It was written in a scheme which is now Racket.

 I am not a fan of clojure, so I only tried overtone, which is a
 Clojure wrapper for the Supercollider sound server. They also copied
 Shadertoy with their 'Shadertone' which allows for the graphics part
 of the livecoding of music and graphics.

What I see in the livecoding scene is the combination of rather complex
programming with rather simplistic music - relentless techno beats ;-)
I would like it the other way around.

 I personally like Extempore, but I couldn't get it built on my Windows
 machine, the OS X install had some issues with Jack and timing, and my
 Linux distro had a few issues too. All in all, it is very complete and
 complex, but too much fuss for my skills.
 http://extempore.moso.com.au/

If it takes days to make it run it looses attraction ...

 I have been sticking with learning PicoLisp, and I would like to
 somehow get it to work with Grace (a single cross-platform executable,
 that you program music pieces in a Scheme or simplified Scheme called
 Sal). http://commonmusic.sourceforge.net/
 Grace or CM lacks a video creation component or library. I was hoping
 to hook into the CM libraries with PicoLisp, and then use Alex's z3d.l
 library to do graphics in PicoLisp. I am not near enough of a
 programmer to do so, only aware that it can be done (I think?).

 Livecoding video and audio in a Lisp! Pure heaven...maybe CEPL in
 PicoLisp??? https://www.youtube.com/watch?v=I0kWZP9L9Kc

What would be the minimalistic setup? A midi cmdline tool or a C shared
library that can be called from PicoLisp?
Or would Supercollider be the easiest thing to work with, now that the
OSC Protocol is implemented in PicoLIsp? 

 Have fun!

 Rob

 On Thu, Jul 30, 2015 at 1:26 PM, Thorsten Jolitz tjol...@gmail.com
 wrote:

 Erik Gustafson
 erik.d.gustaf...@gmail.com writes:
 
 Hi Erik,
 
  https://github.com/erdg/picolisp-osc
 
  If interested, more info about OSC can be found here:
 
  opensoundcontrol.org/introduction-osc
  opensoundcontrol.org/spec-1_0
 
 I find the combination of sound  picolisp very interesting, are
 you
 aware of SoundCollider and the Clojure Libraries Overtone and
 Leipzig
 (both on Github)?
 
 There are interesting videos on Youtube about making music with
 emacs/vim and clojure:
 
 ,
 | 1.
 | Functional Composition - Chris Ford - YouTube
 |
 | www.youtube.com/watch?v=Mfsnlbd-4xQ8. Jan. 2013 - 39 Min.
 | ► - Hochgeladen von ClojureTV Music theory is one of the
 | 39:21 most naturally elegant and functional domains. It's a
 | perfect fit for ...
 |
 | 2.
 | Creating music with Clojure and Overtone - YouTube
 |
 | www.youtube.com/watch?v=RYZeQ6t_5SA23. Juli 2014 - 71 Min.
 | ► - Hochgeladen von Manchester Geek Nights Chris Ford shows
 | 70:50 how to make music with Clojure, starting with the basic
 | building block of ...
 `
 
 And I noticed that you have another music related picolisp lib on
 github:
 
 ,
 | 1. erdg/picolisp-aubio · GitHub
 |
 | https://github.com/erdg/picolisp-aubio
 `
 
 I'm not so much interested in the technical (syntheziser) stuff
 but
 rather in the musical side of it, and I have a few questions:
 
 1. How much would it take not to rewrite Overtone in PicoLisp but
 rather
 to define a handfull of musical instruments that can easily be
 used in a
 music creating PicoLisp program? I'm thinking of a basic rhythm
 section
 with a few rhythm instruments (maybe just a snare drum for
 creating
 swing and a Cajon and maybe Handclaps for creating Flamenco/World
 Music
 beats) and, most important, a (acoustic contra) bass.
 
 With some musical instruments available, one could take some
 inspiration
 from Overtone and Leipzig and maybe a python program like
 
 ,
 | 1. MMA Home Page - Mellowood
 |
 | www.mellowood.ca/mma/
 | ‎
 | + Im Cache
 | + Ä hnliche Seiten
 | 13 Jun 2015 ... MMA-Musical MIDI Accompaniment is an
 | accompaniment generator. ... MMA's templating track system
 | puts you in control of your music.
 `
 
 and create background tracks for practising in PicoLisp. I think
 that
 would be fun ;-)
 
 2. How to use (picolisp-)aubio to get a score of what I play?
 
 Reading about Aubio, it seems that I could plugin my guitar into
 my
 computer, record some stuff

Re: Open Sound Control

[Thorsten Jolitz]

Erik Gustafson
erik.d.gustaf...@gmail.com writes:

Hi Erik,

 https://github.com/erdg/picolisp-osc

 If interested, more info about OSC can be found here:

 opensoundcontrol.org/introduction-osc
 opensoundcontrol.org/spec-1_0

I find the combination of sound  picolisp very interesting, are you
aware of SoundCollider and the Clojure Libraries Overtone and Leipzig
(both on Github)?

There are interesting videos on Youtube about making music with
emacs/vim and clojure:

,
|  1.   
   
| Functional Composition - Chris Ford - YouTube 
   
|   
   
|   www.youtube.com/watch?v=Mfsnlbd-4xQ8. Jan. 2013 - 39 Min.   
   
| ► - Hochgeladen von ClojureTV Music theory is one of the  
   
| 39:21 most naturally elegant and functional domains. It's a   
   
|   perfect fit for ... 
   
|   
   
|  2.   
   
| Creating music with Clojure and Overtone - YouTube
   
|   
   
|   www.youtube.com/watch?v=RYZeQ6t_5SA23. Juli 2014 - 71 Min.  
   
| ► - Hochgeladen von Manchester Geek Nights Chris Ford shows   
   
| 70:50 how to make music with Clojure, starting with the basic 
   
|   building block of ...   
   
`

And I noticed that you have another music related picolisp lib on
github:

,
|  1. erdg/picolisp-aubio · GitHub
| 
| https://github.com/erdg/picolisp-aubio  
`

I'm not so much interested in the technical (syntheziser) stuff but
rather in the musical side of it, and I have a few questions:

1. How much would it take not to rewrite Overtone in PicoLisp but rather
to define a handfull of musical instruments that can easily be used in a
music creating PicoLisp program? I'm thinking of a basic rhythm section
with a few rhythm instruments (maybe just a snare drum for creating
swing and a Cajon and maybe Handclaps for creating Flamenco/World Music
beats) and, most important, a (acoustic contra) bass.

With some musical instruments available, one could take some inspiration
from Overtone and Leipzig and maybe a python program like

,
|  1. MMA Home Page - Mellowood   
| 
| www.mellowood.ca/mma/   
| ‎   
|   + Im Cache
|   + Ä hnliche Seiten
| 13 Jun 2015 ... MMA-Musical MIDI Accompaniment is an  
| accompaniment generator. ... MMA's templating track system  
| puts you in control of your music.  
`
 
and create background tracks for practising in PicoLisp. I think that
would be fun ;-)

2. How to use (picolisp-)aubio to get a score of what I play?

Reading about Aubio, it seems that I could plugin my guitar into my
computer, record some stuff, and the use Aubio to extract a midi score
of what I played (and then use other programs to convert that midi score
to conventional musical notation). 

,
|  1. aubio, a library for audio labelling
| 
| aubio.org/  
| ‎   
|   + Im Cache
|   + Ä hnliche Seiten
| aubio, a collection of algorithms and tools to extract  
| musical meaning from audio signals, such as tempo, pitch, and   
| onset.  
`

A fascinating perspective, but how to do that in practice? I tried to
use aubio on mp3 and ogg files as input

,
| $ aubionotes --help
| usage: aubionotes [ options ]
|-i  --inputinput file
|-r  --samplerate   select samplerate
|-B  --bufsize  set buffer size
|-H  --hopsize  set hopsize
|-O  --onsetselect onset detection algorithm
|-t  --onset-threshold  set onset detection threshold
|-p  --pitchselect pitch 

Re: Property lists and keywords in PicoLisp

[Thorsten Jolitz]

Alexander Burger a...@software-lab.de writes:

Hi Alex,

  In PicoLisp, a property list cannot be seen separated from a symbol.
  After all, these are properties OF a symbol. It is just an
  implementation detail that they are a list internally.
 
 not in PicoLisp, but e.g. in Emacs Lisp there are standalone plists like
 the above, and functions like plist-{get|put|member} to work with them

 OK, but then I feel that they use the wrong term in Emacs Lisp. A
 property is something an item *has*, i.e. a symbol in Lisp.

 A stand-alone list of key-value pairs is traditionally called an
 association list.

yes, the naming is confusing

 Yes, named parameters is my use-case, e.g. wrapping an R function like
 plot() with *many* named parameters, most of them with decent default
 values and thus omitted in function calls, in PicoLisp glue code.
 
  What exactly would be your goal?
 
 there are actually two goals:
 
  1. create functions with named parameters

 Yes. So the approaches shown in the rosetta code tasks help?

they did, after some experimenting

  2. use flat lists like (:a b :c d) like alists ((:a . b) (:b . c))

 You can use a flat list with 'memq' (instead of 'assoc' or 'asoq'):

: (cdr (asoq 'b '((a . 1) (b . 2) (c . 3
- 2

: (cadr (memq 'b '(a 1 b 2 c 3)))
- 2

ok, I see, so no need for more syntax. 

Thanks!

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Property lists and keywords in PicoLisp

[Thorsten Jolitz]

Alexander Burger a...@software-lab.de writes:

Hi Alex,

 using flat property lists with keywords like
 
 ,
 | (:a b :c d)
 `
 
 as data (not as a symbol's property list) is pretty common in e.g. Emacs
 Lisp and elsewhere. In Clojure the keywords even work like functions
 AFAIK

 I feel that I don't completely understand the question.


 In PicoLisp, a property list cannot be seen separated from a symbol.
 After all, these are properties OF a symbol. It is just an
 implementation detail that they are a list internally.

not in PicoLisp, but e.g. in Emacs Lisp there are standalone plists like
the above, and functions like plist-{get|put|member} to work with them
(additional to 'get' and 'put' for symbol's plists).

 A stand-alone list of key-value pairs is an association list, as used
 with 'assoc' or 'asoq'.


 These 'stand-alone' property lists are pretty useful for function
 keyword arguments too, e.g.
 
 ,
 | (defun foo (keys a b c (d . 5) e (f . g)) ...)
 `

 There were some rosetta tasks in that like. I don't remember exactly if
 they address the same problem, but you may look at the Named
 parameters and Optional parameters tasks. Is this similar?

Yes, named parameters is my use-case, e.g. wrapping an R function like
plot() with *many* named parameters, most of them with decent default
values and thus omitted in function calls, in PicoLisp glue code.

 What exactly would be your goal?

there are actually two goals:

 1. create functions with named parameters

 2. use flat lists like (:a b :c d) like alists ((:a . b) (:b . c))

Both could be solved when (let L ...) in the following code would work:

#+BEGIN_SRC picolisp  
: (let (:a 'b :c 'd :e 'f) :c)
- d
: (de foo L (let L :c))
- foo
: (foo :a 'b :c 'd :e 'f)
- NIL
#+END_SRC

I tried (let `L ...), (let (L) ...) and quite a lot of other
combinations, but always get NIL or an error. 

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Property lists and keywords in PicoLisp

[Thorsten Jolitz]

Hi List, 

using flat property lists with keywords like

,
| (:a b :c d)
`

as data (not as a symbol's property list) is pretty common in e.g. Emacs
Lisp and elsewhere. In Clojure the keywords even work like functions
AFAIK

,
| (:a Lst) - b
`

These 'stand-alone' property lists are pretty useful for function
keyword arguments too, e.g.

,
| (defun foo (keys a b c (d . 5) e (f . g)) ...)
`

called like

,
| (foo :a 3 :d 4)
`

I wonder if this would make sense in PicoLisp? Or is it easily emulated
with existing functionality? Selecting values by key works of cause for
nested symbol property-lists, but requires some manual list processing
for flat lists with key/val pairs - or am I missing something?

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Strange PicoLisp site

[Thorsten Jolitz]

Hi List, 

have you seen this website:

,
| http://www.picolisp.org/
` 

Its a nice looking website, but I cannot see no connection to PicoLisp
whatsover. Or does 'picolisp' have other no technical meanings to native
English speakers?

Even the camel case they use is the same as in PicoLisp, maybe Alex
should sue them for copyrights violations ... ;-)

-- 
cheers,
Thorsten


-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Strange PicoLisp site

[Thorsten Jolitz]

Jakob Eriksson ja...@aurorasystems.eu writes:

 someone registered that domain for PicoLisp, then did not renew the fee.

 When it was expired, the domain was snatched (like most expired
 domains are)
 by one of these companies which do nothing else but buy expired domains and
 resell them.

 The boilerplate text is just random junk to keep the page in google's
 index,
 to keep the value of the domain up a tiny little bit.

I see ... but I must say that the 'random junk' looks pretty good, with
a different title like 'PoetsClub' or so it would make a nice webpage.

 On October 27, 2014 at 9:49 AM Thorsten Jolitz
 tjol...@gmail.com wrote:

 Hi List,

 have you seen this website:

 ,
 | http://www.picolisp.org/
 `

 Its a nice looking website, but I cannot see no connection to PicoLisp
 whatsover. Or does 'picolisp' have other no technical meanings to native
 English speakers?

 Even the camel case they use is the same as in PicoLisp, maybe Alex
 should sue them for copyrights violations ... ;-)

 --
 cheers,
 Thorsten


 --
 UNSUBSCRIBE:
 mailto:picolisp@software-lab.de?subject=Unsubscribe

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Strange PicoLisp site

[Thorsten Jolitz]

Alexander Burger a...@software-lab.de writes:

Hi Alex,

 I think the above site is evil.

so the message to take home is that evilness might come along with a
pretty attractive appearance ...

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: The many uses of PilMCU

[Thorsten Jolitz]

Christophe Gragnic
christophegrag...@gmail.com writes:

Hi Christophe,

 This thread is meant to collect ideas about PilMCU.
 At least mine (because I need to clean them up a bit)
 and ideas of other PicoLispers (out of curiosity).

 Maybe some items will look more like questions like «is it even possible?»!

that last sentence describes the situation very well for people who have
not been in the 'embedded world' so far (like me) - they can let their
imagination (about PilMCU) flow, but most likely some seasoned embedded
programmer will bring them back-to-earth soon. 

For me the fascination about the PilMCU chip lies in the fact that, even
if it might be slower or more expensive than available mass products, I
can program it in PicoLisp. For others this might be a huge advantage
over doing embedded programming in other environments (which they
already know), for me it would allow me to try out embedded programming
without having to learn all the usual related stuff, I would simply skip
the painfull part and directly go to the fun part, so to say.

The problem is that I don't know exactly how that fun part would look
like in the end? Whats the state-of-the-art in robotics e.g.? How
difficult would it be to use the PilMCU chip as the brain of an
otherwise fully functional roboter(-tool-kit) - if at all possible? 

What about (connecting) smart home devices etc? All my ideas are
application oriented, i.e. I imagine an interesting use of a chip that
is programmed in PicoLisp, but these ideas can be turned down easily by
saying this would be cheaper  faster when simply running 64bit
PicoLisp on a regular mass-product chip.

So I think this thread is a good idea, and hope it helps me to make up
my mind about realistic possibilities of PilMUC. 

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Programming environment for PilMCU

[Thorsten Jolitz]

George Orais gpor...@yahoo.com writes:

Hi George,

 The pilMCU is the PicoLisp interpreter itself :) To interact with it,
 as of the moment we will use UART and use a PC as terminal. But later
 we plan to add PS/2 and VGA as the means to inter act with the
 hardware interpreter, i hope this answer your inquiry? Thanks!

Yes, thank you!
This stuff is exciting, but a bit new for me too ...

 On Monday, September 22, 2014 5:29 PM, Thorsten Jolitz
 tjol...@gmail.com wrote:

 Alexander Burger a...@software-lab.de writes:

 Hi Alex,

 Assuming PilMCU hardware exists and someone wants to use or program
 it, how
 would that look like? Where would one type the commands to manage
 the
 file system (whats the PilCMU terminal/console?), how would one
 interact
 with the PicoLisp REPL?

 You saw the copy/pasted session in my first post? That's exactly how
 you
 interact with it.

 yes, but I wondered what would be the device the PicoLisp REPL runs on
 in this case. 

 We have two I/O ports defined as TTY in- and output. On the real
 hardware you connect a terminal(program).

 ok

 The SSD images contain a database, with a simple file system
 implemented
 in external symbols. These images are generated with a normal
 PicoLisp
 running on a standard PC, and then transferred t o the SSDs.

 ok

 I made the images so far by copying normal *.l files from standard
 PicoLisp to an init/ directory, and edited the files so that they
 were
 the way we need them for PilMCU.

 thanks for the info!

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: PicoLisp roots

[Thorsten Jolitz]

jerome moliere jer...@javaxpert.com writes:

 Thanks for your quick reply...
 Is there a guide explaining major differences between Common Lisp 
 PicoLisp ? I guess, reading your answer , that there 's no just a few
 syntactic differences between the 2 dialects ... I can read between
 the lines some philosophical major differences , right?
 What are the most difficult tasks , hot topics to be aware of when
 porting a software from CL to PicoLisp ?

You might want to have a look at rosettacode.org, there are hundreds of
tasks with solutions for both, PicoLisp and CL, so you have a perfect
comparison. 

OTOH, although people told me that they have only the parenthesis in
common, I always felt that writing Emacs Lisp and writing PicoLisp has
the same enjoyable (dynamic) feeling, and I think Emacs Lisp was mostly
influenced by MacLisp too, not by CL.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Announce: PicoLisp in Hardware (PilMCU)

[Thorsten Jolitz]

Alexander Burger a...@software-lab.de writes:

Hi Alex (and George),

 we are proud to announce PilMCU, the Lisp Machine on a Chip! :)

though not really a hardware/low-level guy, I think this sounds pretty
exiting!

 How shall we proceed? We need investors (or crowdfunding) to polish,
 manufacture and distribute the real thing.

I suggest to proceed in 2 steps:

 1. make me a team member

 2. repeat the {Microsoft|Apple}-Story

;-)

 We imagine something in the line of an Embedded Lisp Machine or a
 Lisp Machine Kit. Perhaps for home brewing, educational institutions
 and/or robotics research?

 Is anybody interested -- or knows people who are?

I think I have VC-Companies and Robotics-Research-Faculties in my
neighborhood, so once you have a business-idea based on PilMCU's USPs, I
could try to make first contacts if that helps. 

Not sure what would be a realistic business idea, but maybe start by
figuring out where the real money is nowadays (energy sector,
automotive sector, mobile-phones etc) and then think about a possible niche to
fill. 

Makes things much easier when potential clients drown in profits ;)

If you find out e.g. how an Embedded Lisp Machine can be really useful
for the car industry, we will all have PicoLisp jobs pretty soon!

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Announce: PicoLisp in Hardware (PilMCU)

[Thorsten Jolitz]

Loyall, David david.loy...@nebraska.gov
writes:

 The Internet would like to run this locally.  Would you post the
 verilog source and build files?  Or a link to a repository?

I think this has the potential to make a very nice and successfull
kickstarter project, so why not try to build a business idea around it
instead of just giving away the verilog source and build files?

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Announce: PicoLisp in Hardware (PilMCU)

[Thorsten Jolitz]

Loyall, David david.loy...@nebraska.gov
writes:

 From: Thorsten Jolitz
 Loyall, David writes:
  The Internet would like to run this locally.  Would you post the
  verilog source and build files?  Or a link to a repository?
 
 I think this has the potential to make a very nice and successfull
 kickstarter
 project, so why not try to build a business idea around it instead
 of just giving
 away the verilog source and build files?

 It's a Lisp machine.  It probably shouldn't be born crippled (with
 closed design). :)

I'm sure its technical design is not crippled at all.

 It still needs additional development, right Geo and Alex?  Many hands
 make light work.

With 'many hands' involved we would not have most amazing PicoLisp but
rather a kind of 'small common lisp', thats for sure ...

 Have you seen https://news.ycombinator.com/item?id=8340283 ?  Folks
 are looking for the source already.

Of course they are, but what will folks give in return?

 You can still make money on open source hardware.  In fact, that's a
 new trend.  https://www.google.com/search?q=open+hardware

Open-source software has been pretty much a one-way business for Alex so
far - he creates the software and gives a lot of assistence to (new)
users, but how many of them go on and press the 'donate' button on the
website?

It would be only fair (and very beneficial for the PicoLisp project) if
there would be donations in both directions, and PilMCU looks pretty
interesting wrt this goal.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Announce: PicoLisp in Hardware (PilMCU)

[Thorsten Jolitz]

Christophe Gragnic
christophegrag...@gmail.com writes:

 On Fri, Sep 19, 2014 at 10:53 PM, Loyall, David
 david.loy...@nebraska.gov wrote:
 If you sell a FPGA configured to be an open source Lisp CPU, I'll
 buy a few

 Someone on Hacker News: «where's the kickstarter page? I want a few of
 those.»
 I'd buy a few too.

Thats the idea, I would say: buy the chips and support the project. Give
them time to prepare a nice kickstarter project. Enjoy the opportunity
to support a wonderful free software project to become not only a
technical but an economic success too.

You ask them to give away their most important 'capital' to the public
before even starting the business. Not a good advice, really ...

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: (index 'any 'lst) implementation

[Thorsten Jolitz]

Alexander Burger a...@software-lab.de writes:

Hi Alex,

 I see that the index function evaluates the entire lst before
 searching for a match. Wouldn’t it be more efficient to evaluate the
 elements of lst one by one, testing the result for a match, and only
 proceed with evaluation of the next element if a match was not found? Is
 there a specific reason why you didn’t implement it that way?

 What you describe here is call lazy evaluation. While some languages
 support this, PicoLisp (like most other Lisps) doesn't.

 Think about what this would mean: The function 'list' must know somehow
 that it should stop, because some function up in the call history
 doesn't need the whole list.

 And even if it knew that, can it know that some side effects (like your
 'prinl' above) may be omitted?

 The fundamental rule of Lisp function calls (a function recursively
 evaluates its arguments before it starts to run) is broken here.

Funny how views differ -  in the Clojure world lazy evaluation seems
to be a big feature ...

OTOH, Paul Graham writes 150pages about the beauty and power of macros
in 'On Lisp', what made me think that macros are the ultimate thing in
programming, only to read later that 'macros are cludge' in the PicoLisp
docs ;)

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: try picolisp

[Thorsten Jolitz]

Christophe Gragnic
christophegrag...@gmail.com writes:

 On Mon, Sep 1, 2014 at 10:02 PM, Thorsten Jolitz
 tjol...@gmail.com wrote:

 could you elaborate on this a bit?

 Sorry, I cannot give you better info that what is in the ref:
 http://www.tcl.tk/man/tcl8.6/TclCmd/interp.htm#M4

thx for the link!

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: try picolisp

[Thorsten Jolitz]

Christophe Gragnic
christophegrag...@gmail.com writes:

Hi Christophe,

 On Mon, Sep 1, 2014 at 5:52 AM, Tomas Hlavaty

 Or simply call something that crashes PicoLisp, e.g.
 […]

 Here comes the power of embedded interpreters, which PicoLisp
 does not have.

could you elaborate on this a bit?

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: try picolisp

[Thorsten Jolitz]

Alexander Burger a...@software-lab.de writes:

Hi Alex,

 Is it hard to implement minimalist version of minipicolisp in browser ?
 Like http://tryclj.com/ ?

 In addition to the proposed solutions involving JavaScript versions
 of PicoLisp:


 Perhaps not many people are aware that standard PicoLisp comes with a
 REPL in the browser GUI since many years. [...]

just tried it out, its neat!

Imagine the interactive PicoLisp web tutorials that could be written if
there would actually be a safe PicoLisp interpreter as proposed -
thats a nice idea. 

Or would it suffice to define the tutorial functions and start a
server that only allows these functions and nothing else?

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe

Re: Documentation again

[Thorsten Jolitz]

Alexander Burger a...@software-lab.de writes:

 I think you'll know the function

http://software-lab.de/doc/refD.html#doc


 I mostly use it directly from a shell window, with a script 'doc' in my
 executable path

#!/usr/bin/pil @lib/debug.l
(raw T)
(doc (opt) (opt))
(bye)

 It is called as

$ doc mapcar

 and directly opens the right browser window. I have 'w3m' as my default
 browser, so this is blindingly fast, but any other browser works just as
 well. If e.g. Firefox is already open, it connects to it and just opens
 a new tab.

In case you are an Emacs user, you can reuse your muscle memory, since
in eled.l (the emacs version of the pil line editor) I reused some
well-known C-h bindings from Emacs:

,
| Help/Info/Debugging
| 
| | action| keys  |
| |---+---|
| | debug | C-h d |
| | unbug | C-u C-h d |
| | file info | C-h i |
| | symbol doc| C-h f |
| | show symbol   | C-h s |
| | pretty print (pp) | C-h p p   |
| | pretty print (pretty) | C-h p r   |
`

(see http://picolisp.com/wiki/?emacsStyleLed for the complete keybinding
reference)

Thus calling 'doc' on function/symbol FOO is just 'C-h f FOO RET', and
thats already hardcoded in every Emacs user's brain I would guess.

-- 
cheers,
Thorsten

-- 
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe