Subscribe
Hello Edwin Eyan Moragas haa...@gmail.com :-) You are now subscribed -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: solicting user stories of picolisp
On Mon, Jul 19, 2010 at 10:39 PM, Mateusz Jan Przybylski dexen.devr...@gmail.com wrote: However, a (quick'n'dirty) HTML HTTP application in PicoLisp got me a v= ery good grade for `Programming languages paradigms' course at Uni. The lecturer never heard of Lisp before; after listening to my explanatio= ns he wrapped it up with: =A0``So this Lisp is a newfangled language, quite like Ruby, right?'' Geez... i really hope you were kidding. -- Mateusz Jan Przybylski ``One can't proceed from the informal to the formal by formal means.'' -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Unsubscribe
Good bye Edwin Eyan Moragas haa...@gmail.com :-( You are now unsubscribed -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Subscribe
Hello Edwin Eyan Moragas e...@yndy.org :-) You are now subscribed -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
warning while compiling picolisp
Hi all, compiling picoLisp-3.0.3 on openbsd 4.6 gives some warnings. the others i am familiar with but this one is beyond my C programming foo at the moment: flow.c: In function `doCatch': flow.c:1351: warning: argument `x' might be clobbered by `longjmp' or `vfork' i'm willing to test patches if anybody has a solution to this. otherwise, i'll just have to dig deeper. thank you. /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: warning while compiling picolisp
On Mon, Jul 26, 2010 at 12:59 AM, dexen deVries dexen.devr...@gmail.com wrote: On Sunday 25 of July 2010 17:41:11 you wrote: compiling picoLisp-3.0.3 on openbsd 4.6 gives some warnings. the others i am familiar with but this one is beyond my C programming foo at the moment: flow.c: In function `doCatch': flow.c:1351: warning: argument `x' might be clobbered by `longjmp' or `vfork' Not a patch, but this is what i've found: http://www.technovelty.org/code/c/cdecl.html thank you. learn something... old.. every day. :) -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: PL vs. BSD License
Hi Alex, all, On Mon, Aug 2, 2010 at 2:29 PM, Alexander Burger a...@software-lab.de wrote: Hi all, as this discussion popped up recently and in the past, and will surely pop up in the future: What do you think if PicoLisp were released under the BSD license instead of GPL? i go for BSD. My reason for asking this is simple: On several occasions I experienced that the acceptance of PicoLisp is limited by the GPL. Companies are reluctant to use it because they are afraid of being forced to publish their little secrets. Their fear is sometimes rational and sometimes not, but the effects are the same. For me it is critical, as my economic survival depends on it. the way i understand it (from reading all of the pros and cons posts of GPL vs BSD over the net for quite some time) is that the GPL forces you to publish changes when you want to distribute the code outside of the company. the most clear statement i have read somewhere tells me the GPL is a distribution license. i think that captures the spirit of the GPL. or most of it. Ideologically, I prefer the GPL. It guarantees that freedom propagates. But it does this by cutting down on freedom, so it is schizophrenic. I used to compare the situation with freedom in a society: A society should not give an individual member so much freedom that he can make himself a dictator and thus destroy freedom. But this comparison is wrong. Using free software in a non-free project doesn't decrease its freedom; it just doesn't increase it the way the GPL tries to enforce. So is this just much ado about nothing? this is a complex issue and i am not an expert. my additional arguments would be: 1) the BSD license does not force one to publish changes. good for companies. the GPL does not enforce this too *unless* you distribute your changes to pre-existing GPL code. 2) the BSD license still ensures that the code remains free. the GPL does not guarantee that changes in the pre-existing code will be made publicly available (see point 1). 3) the GPL is complex. so many ifs. i'm not keen on using complex things. best regards, eyan PS: hat tip to Alex for PicoLisp and bringing this issue up in the list. -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: PL vs. BSD License
Hi Dan, all, On Mon, Aug 2, 2010 at 7:24 PM, Daniel Elliott danelliotts...@gmail.com w= rote: Hello. Does the GPL affect code that I write that runs on PicoLisp, or just changes to the PicoLisp interpreter? picoLisp is made up of 1) the interpreter and 2) libraries and support files that come with the distribution, ie the *.l files. If only changes to the interpreter are affected, than GPL is the way to go as far as I am concerned. if the interpreter were GPL and the support files were BSD, then your code would not be affected. with this i am sure. however, if you use any of the .l files (which is GPL), is your code also forced to be GPL? i don't know. However, I should add that many large companies (at least in the US) are irrationally terrified of anything GPL. =A0It is very annoying. - dan On Mon, Aug 2, 2010 at 3:50 AM, Mansur Mamkin mmam...@mail.ru wrote: Hi all, Hi all, as this discussion popped up recently and in the past, and will surely pop up in the future: What do you think if PicoLisp were released unde= r the BSD license instead of GPL? Why not? It would be great to see PicoLisp under BDS license Oops, sorry, I mean BSD -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
documentation license?
Hi Alex, what's the license of the documentation that comes with picoLisp? best, /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
first swing at running local wiki
Hi all, built picoLisp, fetched the wiki, untarred it, then followed the instructions here: http://www.mail-archive.com/picolisp@software-lab.de/msg01556.html in summary it went like: $ cd ~/picoLisp $ mkdir -p db/wiki $ ./dbg wiki/main.l lib/too.l -main wiki/init.l -go opened the browser to: http://host:5000 and i got this output on the terminal: e...@obsddev $ ./dbg wiki/main.l lib/too.l -main wiki/init.l -go /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'Nil' /bin/picolisp:lib/ht: undefined symbol 'prin' /bin/picolisp:lib/ht: undefined symbol 'prin' /bin/picolisp:lib/ht: undefined symbol 'bufSize' /bin/picolisp:lib/ht: undefined symbol 'bufSize' /bin/picolisp:lib/ht: undefined symbol 'bufString' /bin/picolisp:lib/ht: undefined symbol 'bufString' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'Env' /bin/picolisp:lib/ht: undefined symbol 'outString' /bin/picolisp:lib/ht: undefined symbol 'outString' /bin/picolisp:lib/ht: undefined symbol 'evList' /bin/picolisp:lib/ht: undefined symbol 'evList' /bin/picolisp:lib/ht: undefined symbol 'evList' /bin/picolisp:lib/ht: undefined symbol 'evList' /bin/picolisp:lib/ht: undefined symbol 'evList' /bin/picolisp:lib/ht: undefined symbol 'evList' /bin/picolisp:lib/ht: undefined symbol 'evList' /bin/picolisp:lib/ht: undefined symbol 'evList' /bin/picolisp:lib/ht: undefined symbol 'evList' /bin/picolisp:lib/ht: undefined symbol 'name' /bin/picolisp:lib/ht: undefined symbol 'Intern' /bin/picolisp:lib/ht: undefined symbol 'ihash' /bin/picolisp:lib/ht: undefined symbol 'hashed' /bin/picolisp:lib/ht: undefined symbol 'begString' /bin/picolisp:lib/ht: undefined symbol 'begString' /bin/picolisp:lib/ht: undefined symbol 'endString' /bin/picolisp:lib/ht: undefined symbol 'endString' /bin/picolisp:lib/ht: undefined symbol 'firstByte' /bin/picolisp:lib/ht: undefined symbol 'firstByte' /bin/picolisp:lib/ht: undefined symbol 'firstByte' /bin/picolisp:lib/ht: undefined symbol 'firstByte' /bin/picolisp:lib/ht: undefined symbol 'firstByte' /bin/picolisp:lib/ht: undefined symbol 'firstByte' /bin/picolisp:lib/ht: undefined symbol 'outName' /bin/picolisp:lib/ht: undefined symbol 'outName' /bin/picolisp:lib/ht: undefined symbol 'mkChar' /bin/picolisp:lib/ht: undefined symbol 'mkChar' /bin/picolisp:lib/ht: undefined symbol 'mkChar' /bin/picolisp:lib/ht: undefined symbol 'evCnt' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'Chr' /bin/picolisp:lib/ht: undefined symbol 'getChar' /bin/picolisp:lib/ht: undefined symbol 'getChar' /bin/picolisp:lib/ht: undefined symbol 'cons'
Re: documentation license?
Hi Alex, On Tue, Aug 10, 2010 at 2:09 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, what's the license of the documentation that comes with picoLisp? I assumed that it is also under the GPL until now, as the GPL is the only license included in the distribution. As we are currently discussing the license issue, what would be your proposal for the documentation? Is a separate license preferable? no proposal from here. :) just asking. it's more of a mental preparation for future projects. I think the documentation is tightly coupled and in sync with the code, so I felt that it should be under the same license as the code. thank you. Cheers, - Alex -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: first swing at running local wiki
Hi Alex, On Tue, Aug 10, 2010 at 2:23 PM, Alexander Burger a...@software-lab.de wrote: I didn't test under OpenBSD since last year. Something's wrong with the link flags (PICOLISP-FLAGS or DYNAMIC-LIB-FLAGS) in src/Makefile? will check. thank you /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: first swing at running local wiki
Hi Alex, On Tue, Aug 10, 2010 at 2:39 PM, Edwin Eyan Moragas e...@yndy.org wrote: I didn't test under OpenBSD since last year. Something's wrong with the link flags (PICOLISP-FLAGS or DYNAMIC-LIB-FLAGS) in src/Makefile? will check. thank you no problems with the link flags as far as i can tell. it compiles cleanly. diverging a bit. what's picolisp's protocol of loading shared libs? i'm asking to just save effort. i can trace it in the code but... you know the drill. :) the errors i posted earlier comes *after* i stop the browser from waiting for the request. in another train of thought, i'm thinking of creating a static binary instead of dynamic libs, so the my question is still in line with this effort. patience and peace, /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: first swing at running local wiki
On Tue, Aug 10, 2010 at 10:19 PM, Alexander Burger a...@software-lab.de wr= ote: Hi Edwin, no problems with the link flags as far as i can tell. it compiles cleanl= y. OK diverging a bit. what's picolisp's protocol of loading shared libs? A simple dlopen() followed by dlsym(). This happens e.g. in src/main.c in the sharedLib() function. got it. thank you. i was sprinkling some printf()s in the sharedLib() function. again, happens after browser is stopped. i also looked at tab.c. saw the functions being registered there. Hmm, probably upon the first access to a function in the shared object file. I'm not sure at the moment exactly _when_ the first of those functions is called. You could try manually som functions from lib/ht, e.g.: =A0 $ ./dbg =A0 : (ht:Prin abcd) =A0 alt;bamp;cgt;d- abcd =A0 : (ht:Fmt 'abc def 123 (4 5 6)) =A0 - $abcdef+123_+4_+5_+6 If this works, we are done :) didn't. dbg spewed out similar errors. digging deeper... Cheers, - Alex -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: first swing at running local wiki
e...@obsddev $ ./dbg wiki/main.l lib/too.l -main wiki/init.l -go !? (ht:Pack @U) ht:Pack -- Undefined ? (bye) File not found the error is from dlerror(). i'm also posting to m...@openbsd for some clues. haven't dealt with dlopen() and friends before. :) i did another run. the buffer was too short so i hardcoded the absolute path. sorry. dlerror() now returns Cannot load specified object. /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: first swing at running local wiki
Hi Alex, On Thu, Aug 12, 2010 at 2:28 AM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, made ext.c and ht.c as part of the picolisp binary by editing the makefile. also tweaked tab.c (and io.c, flow.c since there were clashing symbols there) and pico.h to include the missing protoypes. Hmm, this should not be the way ;-) indeed. i just wanted to have my fix. :) (snip) Are you sure that the build of lib/ht and the other DLLs was successful? Are there any strange file permissions? Could you unpack a yes. at least as much as i can tell. gmake didn't give out any errors. picoLisp.tgz file in some temporary space, and do the steps described in INSTALL? i did. still the same errors. (snip instructions for bootstrapping wiki) thank your for the instructions. the bootstrap was successful. i have a few minor errors but i was able to remove them (used the non-dynamic-lib build, my bad). Sorry for the rather complicated procedure, but I didn't care to provide for a clean bootstrap procedure, because I created the DB manually during development of the first version of the wiki ;-) no worries. i should be the one apologizing for bugging the list. moving forward, i'm playing with some flags to diagnose the problem. the obsd guys game some hints that i can work with. /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
looks like solved (was Re: first swing at running local wiki)
On Thu, Aug 12, 2010 at 2:31 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, thank your for the instructions. the bootstrap was successful. i have Great! moving forward, i'm playing with some flags to diagnose the problem. the obsd guys game some hints that i can work with. Thanks. I'd really be interested to know what goes wrong. i got the thing behaving correctly by tweaking the Makefile: ifeq ($(shell uname), OpenBSD) OS = OpenBSD PICOLISP-FLAGS = -m32 -rdynamic -Wl,-E LIB-FLAGS = -lc -lm DYNAMIC-LIB-FLAGS = -Wl,-E -Wl,-shared STRIP = strip else Cheers, - Alex -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: looks like solved (was Re: first swing at running local wiki)
Hi Alex, possible additions to the Makefile (INSTALL instructions sensitive) : # Clean up clean: rm -f *.o + rm -f ../lib/ht + rm -f ../lib/ext On Thu, Aug 12, 2010 at 3:21 PM, Edwin Eyan Moragas e...@yndy.org wrote: On Thu, Aug 12, 2010 at 2:31 PM, Alexander Burger a...@software-lab.de w= rote: Hi Edwin, thank your for the instructions. the bootstrap was successful. i have Great! moving forward, i'm playing with some flags to diagnose the problem. the obsd guys game some hints that i can work with. Thanks. I'd really be interested to know what goes wrong. i got the thing behaving correctly by tweaking the Makefile: ifeq ($(shell uname), OpenBSD) =A0 =A0 =A0 =A0OS =3D OpenBSD =A0 =A0 =A0 =A0PICOLISP-FLAGS =3D -m32 -rdynamic -Wl,-E =A0 =A0 =A0 =A0LIB-FLAGS =3D -lc -lm =A0 =A0 =A0 =A0DYNAMIC-LIB-FLAGS =3D -Wl,-E -Wl,-shared =A0 =A0 =A0 =A0STRIP =3D strip else Cheers, - Alex -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: looks like solved (was Re: first swing at running local wiki)
On Thu, Aug 12, 2010 at 3:56 PM, Alexander Burger a...@software-lab.de wro= te: Hi Edwin, possible additions to the Makefile (INSTALL instructions sensitive) : =A0# Clean up =A0clean: =A0 =A0 =A0 =A0 rm -f *.o + =A0 =A0 =A0 rm -f ../lib/ht + =A0 =A0 =A0 rm -f ../lib/ext That's right. It is not absolutely necessary, though, as just removing '*.o' will trigger a complete rebuild anyway. If we include the libraries, then we should also include all other targets, which would make things harder to maintain and keep in sync. copy that. Cheers, - Alex -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
difference between a cons pain and plain list
Hi all, yes, a lisp newcomer here. looking at the example in the app dev doc: : (h1 '(id . bar) Title) two questions 1) if the cons pair was written as code(id bar)/code instead of as with the example, would h1 treat it different? how? 2) what's the fundamental difference between a cons pair and a simple list? as far as simple symbols go, it would look something like this: (a . b ) == [a, b] (a b) == [a, b, NIL] making it three: 3) when is a cons pair appropriate to use? cheers, /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: difference between a cons pain and plain list
Hi Alex, On Tue, Aug 17, 2010 at 8:34 PM, Alexander Burger a...@software-lab.de wro= te: (a . b ) =3D=3D [a, b] (a b) =3D=3D [a, b, NIL] Well, it depends how your '[' and '[' notation is interpreted. I would draw it as cell boxes. (a . b) is a single cell (two pointers), with 'a' in the CAR and 'b' in the CDR: =A0 +-+-+ =A0 | =A0a =A0| =A0b =A0| =A0 +-+-+ while (a b) consists of two cells, where the CDR of the second cell is NIL: =A0 +-+-+ =A0 =A0 +-+-+ =A0 | =A0a =A0| =A0---+| =A0b =A0| NIL | =A0 +-+-+ =A0 =A0 +-+-+ thank you for this. clearer now. this is how it is implemented in picolisp underneath, right? (from pico.h) so if i had a different base data structure, it wouldn't be like this? say in other lisp implementations? reason i'm asking is that if i had arrays as base structures in C instead of linked cells, a dotted pair would look something like: +-+-+ | s | t | +-+-+ and a list as: +-+-+-+ | s | t | NIL | +-+-+-+ is this a possible way to think about this or am i just chasing windmills? 3) when is a cons pair appropriate to use? When you have data structures that typically appear in pairs. A good example are two-dimensional coordinates, x and y. You can put a single coordinate into a single cell (x . y) instead of a two-element list (x y), as you know there will never be a third element. In that way you save half of the space. A three-dimensional coordinate, btw, could be put into a two-cell structure (x y . z), needing only two cells instead of three. like so? +-+-+ +-+-+ | x | ---+| y | z | +-+-+ +-+-+ Besides saving memory, cons pairs also make access easier sometimes. To access the 'z' in (x y . z) you can use the function 'cddr', doing two pointer dereferences. To access it in (x y z), you need three pointer operations (i.e. 'caddr'). Basically, you are free to decide your data structure depending on the application. Sometimes a list has advantages over a pair, for example when you need uniform access (e.g. 'mapping' over the list). thank you for taking time to answer my silly questions. /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: script for generating tags for picolisp
Hi Alex, On Wed, Aug 18, 2010 at 3:41 PM, Alexander Burger a...@software-lab.de wro= te: Hi Doug, On Tue, Aug 17, 2010 at 09:19:23PM -0700, Doug Snead wrote: ... # make a tags file for pico lisp source files. ... As I understand it, this is not exactly what Edwin intended. I think he wanted a tags file for the main interpreter functions written in C Doug sent something that i actually intended. there's ctags for C functions=
Re: script for generating tags for picolisp
Hi Doug,
thanks for this. this would come in handy.
On Wed, Aug 18, 2010 at 12:19 PM, Doug Snead semaphore_2...@yahoo.com wro=
te:
Here's one shell script I used to use for that (I called it lisptags) ...=
hopefully it still works :-)
#!/bin/sh
# make a tags file for pico lisp source files.
# use:
# =A0 =A0 =A0 lisptags foo.l bar.l baz.l ... bof.l
# generate the file 'tags'
# [based on lisptags csh script by John Foderaro, c.1982]
awk '
/^[ \t]*\([bd][em][ \t]/ {
n=3D$2;
gsub(/^\(/,,n);
gsub(/\)$/,,n);
print n \t FILENAME \t?^ $0 $?;
}
/^[ \t]*\(class[ \t]/ {
n=3D$2;
gsub(/^\(/,,n);
gsub(/\)$/,,n);
print n \t FILENAME \t?^ $0 $?;
if (n ~ /^\+/) {
=A0gsub(/^\+/,,n);
=A0print n \t FILENAME \t?^ $0 $?;
}
}
' $* | sort tags
=3D =3D =3D
--- On Tue, 8/17/10, Edwin Eyan Moragas e...@yndy.org wrote:
From: Edwin Eyan Moragas e...@yndy.org
Subject: script for generating tags for picolisp
To: picolisp@software-lab.de
Date: Tuesday, August 17, 2010, 12:05 AM
Hi all,
anybody who made an attempt to create such kind of script?
or maybe a
resource somewhere in the interwebs for this?
cheers,
/e
--
UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe
--
UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe
--
UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
(extend cls)
Hi all, i'm going thru the wiki code. encountered `(extend)` in it but couldn't find an example in the tutorial but found the definition in the docs. after reading thru, seems to me that *Class is used to change class definitions (from wiki/er.l and doc/family.l). examining wiki/gui.l, seems to me that class definitions (say +Doc) are kept in top-level. execution flow now would look like: a) define classes using `(extend)` or `(class)` b) open the db c) accept requests d) do whatever e) render the page f) goto #c questions: 1) am i on the right track so far? 2) as far as i can tell, only 1 db can be opened at any time. how does one deal with multiple databases? thanks for your patience. :) /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
how to run tests
Hi all, i can't find anything on how to run tests that come with the source distribution. anybody has an URL or a clue how to do this? thank you. /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: how to run tests
Hi Alex, On Mon, Aug 23, 2010 at 1:56 PM, Alexander Burger a...@software-lab.de wro= te: Hi Edwin, i can't find anything on how to run tests that come with the source distribution. anybody has an URL or a clue how to do this? Do you mean the unit tests residing in the test/ directory? yes. They are started via lib/test.l, with a somewhat special command (written as a comment in the beginning of lib/test.l): =A0 $(/bin/pwd)/p lib/test.l -bye This special path syntax for 'p' is necessary so that the functions checking pathes and the current working directory return the expected results. When all goes well, OK is printed, otherwise an error handler is entered: =A0 $ $(/bin/pwd)/p lib/test.l -bye =A0 OK great. thank you. :) more stuff to do for me. i want this to pass on openbsd. in due time. :) Cheers, - Alex -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: non forking picoLisp appserver
On Mon, Aug 23, 2010 at 7:10 PM, Alexander Burger a...@software-lab.de wro= te: Hi Edwin, can i make the picoLisp app server not fork? that is, when i run There was a discussion about that here in 2008. You can easily make a non-forking server by not calling the standard 'server' function which automatically forks upon a connect, but use a 'listen'ing loop instead: #!bin/picolisp lib.l (load ext.l lib/http.l lib/xhtml.l) (allowed () @start) (de start () =A0 (html 0 Hello NIL NIL =A0 =A0 =A0Hello World! ) ) (let (P (port 8080) =A0H @start) =A0 (setq *Home (cons H (chop H))) =A0 (loop =A0 =A0 =A0(when (listen P) =A0 =A0 =A0 =A0 (http @) =A0 =A0 =A0 =A0 (close @) ) ) ) This loops infinitely, listening on port 8080. When a request arrives, it is handled, and the connection closed. This works, but has disadvantages like denying further requests while a transaction is running, and not being able to keep state information in a session the way the child processes do after a fork. For certain special cases it might well make sense, though. thank you Alex. cheers, /eyan -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
port handling and httpgate
Hi all, i start httpgate with: $ bin/httpGate 3000 8080 8081 8082 then run the sample app. it shows me the login screen and then the location in the browser points at: http://192.168.1.253:3000/45313/713963214703762...@start?*menu=+3*Tab=+1*ID= which shows a blank page. console shows: e...@saidin $ ./dbg app/main.l -main -go : 22487 * 2010-08-25 09:01:10 admin which i guess is alright. running the sample app without httpgate gives me no problems. where's the best place to look to start tracking down this problem. i am running this on an openbsd box by the way. haven't tested it on linux (which is a well known deployment environment anyway). cheers /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: port handling and httpgate
Hi Alex, Do you need this kind of setup for some special reason? After all, the not really. i was just using it for tests. i didn't even give port 80 a shot yet. :). main purpose of 'httpGate' is to proxy all requests to the default port to app-specific ports like 8080. If you pass :3000 to the browser, you could as well directly pass :8080. If such a setup were indeed necessary (e.g. because you have no permission to start 'httpGate' at port 443, but you need it for HTTPS), then the protocol between 'httpGate' and the application server would need to be extended a little, passing the alternate port (3000). No big deal, but it seems nobody needed that until now ;-) it's not necessary. yet. i was on the 'completeness' frame of mind. if this works on port 80, then it will work with other ports as well. will try the usual port and if i don't succeed, then it will be a concern. cheers /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: port handling and httpgate
Hi Alex, will try the usual port and if i don't succeed, then it will be a concern. it is a concern. $ ./bin/httpGate 80 8080 gives the same result. cheers, /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: where are the values of CAR and CDR stored?
Hi Santiago,
thank you for this.
i should have been clearer in my question. how is, say, a symbol,
stored using cells?.
On Wed, Aug 25, 2010 at 11:06 AM, santiago ottonello sano...@gmail.com wr=
ote:
I guess the actual data is in car
i.e.
hi
|h| *|-|i|*|-NIL
But just a guess :-)
you've hit it right in the head. assuming your answer is correct, the
i guess i get it.
:-)
2010/8/24 Edwin Eyan Moragas e...@yndy.org
Hi list,
looking at pico.h i see the definition of the cell:
typedef struct cell { =A0 =A0 =A0 =A0 =A0 =A0// PicoLisp primary data ty=
pe
=A0 struct cell *car;
=A0 struct cell *cdr;
} cell, *any;
these are pointers to... where? this would answer questions like:
where is the actual symbol (or internal string representation of it)
stored?
i can dig deeper in the source but where to start?
cheers,
/e
--
UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe
--
UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: where are the values of CAR and CDR stored?
Hi Mansur, On Wed, Aug 25, 2010 at 12:27 PM, Mansur Mamkin mmam...@mail.ru wrote: Hi all! Look at http://www.software-lab.de/doc/ref.html#vm yes. i have read thru it and several times over. thank you. what i need is confirmation of my understanding. so i guess this mean that symbols are stored like picolisp ints? CAR holding a digit (in the case of symbols, a single character) and the CDR pointing to the next cell, which holds a digit in its CAR, and so on? Hi Santiago, thank you for this. i should have been clearer in my question. how is, say, a symbol, stored using cells?. -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: where are the values of CAR and CDR stored?
Hi Alex, As you observed, a cell is defined as a structure of two pointers. This is just to keep the C compiler happy. In truth, the CAR and the CDR contain either a pointer to another cell, or a plain binary value. this is what i was after. thank you. cheers, /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
httpGate redux
Hi list, using the dev version, i was able to install and run the wiki with httpGate successfully. i can log in and do things fine. *but* i am redirected to a different port (http://192.168.1.253:34205/523083596357234...@start?*menu=+3*Tab=+1*ID=*Evt=+1*Got=_+2_+1) after logging in. how can i just make it stick with port 80? thanks much /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: httpGate redux
Hi Alex, Sticking with port 80 is usually handled by 'httpGate'. To the outside world, there will always only port 80 (or 443) be visible. there's the rub. i'm running it behind httpGate. thank you, /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: httpGate redux
Hi Tomas, On Thu, Aug 26, 2010 at 2:52 PM, Tomas Hlavaty t...@logand.com wrote: I use nginx as a web server and proxy on port 80 and it allows you more freedom of configuration than when using httpGate. =A0You can read a bit about my setup at http://logand.com/blog/picolisp-behind-nginx-proxy.html Hope it helps. i found your resource some time ago and plan on using it as a base for configuring apache to proxy. thank you for this. back to my problem at hand, i'll spell out what i did. i'm running things on: e...@saidar $ uname -a OpenBSD myhost 4.6 GENERIC#58 i386 launched httpGate using: e...@saidar $ sudo ./bin/httpGate 80 8080 8081 8082 after initializing the wiki prereqs and setting *Wikiport to 8080, i launch it using: e...@saidar $ ./dbg wiki/main.l lib/too.l -main -go pointed my browser to: http://192.168.1.253/ and this displays the login page. logged in using admin/admin and it redirects me to: http://192.168.1.253:13678/228846712525427...@start?*menu=3d+3*Tab=3D+1*I= D=3D*Evt=3D+1*Got=3D_+2_+1 and since OpenBSD is not officially supported, i'll just have to do things on my own. however, if there are any clues as to where to start, it would indeed be most helpful. after a cursory look at httpGate.c at around line 220, i think that's where URL rewriting happens. i'll look into that some more. Cheers, Tomas -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=3dunsubscribe -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
where is the tail?
Hi list,
reading thru the reference:
A property is a key-value-pair, represented as a cell in the symbol's tail.
looking at the definition of a cell (from pico.h):
typedef struct cell {
struct cell *car;
struct cell *cdr;
} cell, *any;
where is the 'tail' in light of what's given? is it in the 'car' or
'cdr'? the ASCII art in http://www.software-lab.de/doc/ref.html#cell
is a bit vague.
best,
/e
--
UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
symbols, values and properties
Hi list, reading up on the docs again, it mentions symbols, values and properties. the way i understand it, symbols evaluate to a value. symbols also have properties. several questions: 1) how are symbol properties implemented on the C level? 2) how are symbol properties stored in the db? what i'd like to accomplish is a decent understanding of the implementation before i delve deeper into the code. thank you. /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: symbols, values and properties
Hi Alex,
On Sun, Sep 26, 2010 at 3:11 PM, Alexander Burger a...@software-lab.de wro=
te:
Note that - as opposed to association lists - the key is in the CDR of
the cell. This has advantages when manipulating the value, because the
CAR of a cell is a variable ('var' called in the documentation).
Searching for a property by a key goes through the list, and looks for a
CDR that is '=3D=3D' (pointer equal) to the key. When a property is found=
,
it is moved to the front of the list, as a side effect to accelerate
further accesses (last recently used scheme).
thank you for this.
2) how are symbol properties stored in the db?
The whole symbol (value and properties) is simply stored in the PLIO
format, a binary serialization format. The name of an external symbol
determines its location, at which offset into which file it resides.
The PLIO format is also used by other functions like 'pr', 'rd', 'tell',
'hear' etc. You can experiment with it also outside the database context
by writing it to a plain file:
=A0 : (put 'A 'a 1) =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 # Create some=
properties for 'A'
=A0 - 1
=A0 : (put 'A 'b 2)
=A0 - 2
=A0 : (put 'A 'c 3)
=A0 - 3
=A0 : (getl 'A) =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 # See the=
property list
=A0 - ((3 . c) (2 . b) (1 . a))
=A0 : (out 'a (pr (getl 'A))) =A0 =A0 =A0 =A0 =A0 # Print property list t=
o file a
=A0 - ((3 . c) (2 . b) (1 . a))
=A0 : (hd 'a) =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 # Hexdu=
mp of the PLIO representation
=A0 =A001 01 04 06 02 05 63 01 04 04 02 05 62 01 04 02 =A0..=
c.b...
=A0 0010 =A002 05 61 03 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =
=A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0..a.
=A0 - NIL
Does this help?
helped a lot.
it became clearer too after reading the answer to the second question.
that and reading the source in sym.c of doGetl.
i'm still on the path of trying to see all the little details.
thank you much.
/e
--
UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
opt/pilog.l?
hi all, being a newbie and all, what is the significance of pilog.l in opt/ in the source distribution? best, /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: opt/pilog.l?
Hi Alex, On Tue, Oct 5, 2010 at 3:10 PM, Alexander Burger a...@software-lab.de wrote: The first case were some Pilog extensions from cle, and as there is already a lib/pilog.l and a misc/pilog.l, we agreed to put such things into a new directory opt/. thank you. cheers, /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
warning in io.c
Hi all, compiling picolisp gives me this warning: io.c: In function =91doEcho=92: io.c:2161: warning: =91op=92 may be used uninitialized in this function my problem is that i can't find where `op is defined. any help would be appreciated. best, /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: picolisp in java
Hi Tomas, On Thu, Oct 14, 2010 at 1:43 AM, Tomas Hlavaty t...@logand.com wrote: In that email, I just wanted to say that I implemented some functions directly in Java and that speeds up the fibo benchmark significantly. For getting better picture about interpretation overhead, there are now three versions: one (the original and slow) completely interpreted, one with native functions and one completely native. thank explains it. by the way, thank you for releasing this in the wild. your java implementation gave me a better understanding how picolisp works under the hood. reading Alex's code is really a great exercise to become a solid C head. this is a Good Thing (tm). cheers, /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: wiki password
On Wed, Oct 27, 2010 at 2:54 PM, Alexander Burger a...@software-lab.de wro= te: Hi Tomas, This is a big flaw. =A0You should not be able ever to find out his password. Why not? Nobody could stop me anyway. I could trace the program during execution, for example, to get the passwords. oh yes! :D /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
just a big THANK YOU
Hi list (esp Alex), was browsing thru the rosetta code examples. not to mention all the docs and written materials made available freely via the web. to everyone who has made picolisp what it is today, THANK YOU so much! have a merry christmas y'all! cheers! /e -- UNSUBSCRIBE: mailto:picol...@software-lab.de?subject=unsubscribe
Re: composite primary keys
Hi Henrik, On Wed, Mar 16, 2011 at 3:34 PM, Henrik Sarvell hsarv...@gmail.com wrote: Might the +Aux relation be what you're looking for? i'm approaching this from with a SQL background so forgive me for my naivete. checking the docs out, +Aux is a Prefix class maintaining auxiliary keys for +relations, in addition to +Ref or +Idx indexes. Expects a list of auxiliary attributes of the same object, and combines all keys in that order into a single index key. which means that there must exist a primary key for the +relation in question. however, i'm looking for a +Key with multiple columns. i'm not sure if this is applicable at all, please note. so if +Aux is indeed the right direction, i'll give it a swing. thank you. /e On Wed, Mar 16, 2011 at 2:19 PM, Edwin Eyan Moragas e...@yndy.org wrote: Hi List, anyone encountered creating a composite primary key for Pilog? or it is applicable in the Pilog context? how would one go about using composite primary keys? thank you. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: composite primary keys
Hi Alex, On Wed, Mar 16, 2011 at 3:54 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, anyone encountered creating a composite primary key for Pilog? You mean for database indexes? not just for indexes but for a composite primary key. i'm not sure tho if my head is right on this. If so, there is the '+Aux' prefix class which does this, and there is also a function 'aux' for direct access. The Pilog rules know about '+Aux' and handle it accordingly. thank you. hope you've read my response to Henrik's. whack me with a clue stick anytime. :) Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: composite primary keys
Hi Alex, On Wed, Mar 16, 2011 at 4:44 PM, Alexander Burger a...@software-lab.de wrote: In my mind, I distinguish between objects and indexes in the DB. The objects are the payload data, and also have direct links between each other (the +Link and +Joint relations). uhm. this helps a lot. thank you. Henrik, thanks to you too. those articles helped. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
setting a cookie from a script
Hi List, i got as far as figuring out where a script is executed in a (listen)-ing http server. i think it's somewhere here in http.l: ((= '@ (car @U)) #URL starts with @ #@U is @start by default (if (disallowed) (prog (msg *Url not allowed) (http404)) (and *SesId (timeout *Timeout)) # put this here to find out what the (apply (val... kungfu below is doing (out NIL (println ((val (intern (ht:Pack (cdr @U L))) (apply (val (intern (ht:Pack (cdr @U L) ) ) all well and good. now how do i set a cookie? what i'm trying to do is create a non forking server, store sessions in an idx variable (maybe) and have a way to connect the session with the cookie. i'm hoping this is not too hard. *crossing fingers, hands and all*. thank you. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: setting a cookie from a script
Hi Alex, On Thu, Mar 17, 2011 at 12:46 AM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, i got as far as figuring out where a script is executed in a (listen)-ing http server. i think it's somewhere here in http.l: ((= '@ (car @U)) #URL starts with @ #@U is @start by default (if (disallowed) (prog (msg *Url not allowed) (http404)) (and *SesId (timeout *Timeout)) # put this here to find out what the (apply (val... kungfu below is doing (out NIL (println ((val (intern (ht:Pack (cdr @U L))) (apply (val (intern (ht:Pack (cdr @U L) ) ) Actually, this is not the place where a script is executed, but a direct Lisp function call. For that, the funtion's name is written in the URL as @foo. ah. yes. my mistake. A lisp script (source file) is executed a little more down in the source, at ((tail '(. l) @U) (and *SesId (timeout *Timeout)) (apply script L *Url) ) i.e. when the name ends with .l. all well and good. now how do i set a cookie? You can create cookies by calling the 'cookie' function in your function or script: (cookie Name This is my name) or (cookie Address My City My Street) The first argument is the cookie's name, and all following arguments are stored as a list as the cookie's value. They will be transmitted in the HTTP response. Upon the next request, all cookies will be returned in an assoc list in the global variable '*Cookies'. it works! thank you. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
understanding (task)
Hi list, in http.l, i see a lot of (task (close S)) usage. from the docs, (task) says it's a front end to the *Run global and created a (job) environment. isn't (close) simple enough to just let it rip? why should it be enclosed in a (task). /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: understanding (task)
Hi Alex, On Fri, Mar 25, 2011 at 4:00 PM, Alexander Burger a...@software-lab.de wrote: Instead of enclosing it in (task (close S)), two separate calls (close S) and (task S) would also be all right (just a little bigger). But both are necessary: (close S) to free the socket file descriptor, and (task S) to remove the corresponding entry from the list in '*Run'. is this in the context of multiple picolisp processes? figured S is just a socket and calling (close) in a non forking server would be ok? thank you. /e Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: understanding (task)
Hi Alex On Fri, Mar 25, 2011 at 4:46 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, But both are necessary: (close S) to free the socket file descriptor, and (task S) to remove the corresponding entry from the list in '*Run'. is this in the context of multiple picolisp processes? figured S is just a socket and calling (close) in a non forking server would be ok? It doesn't matter whether there are multiple picolisp processes: Both the open file descriptor, and the list of tasks in '*Run', are private to each process. i see. in this light, i might have been asking the wrong question... So if you just close the file descriptor, but don't remove it from the task list, you'll get a Key conflict error when the next time a new socket with that file descriptor number is assigned a 'task', or when the event handler processing '*Run' passes the closed file descriptor to select(2). what's the reason for running it via (task) in the first place? /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: understanding (task)
On Fri, Mar 25, 2011 at 9:22 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, So if you just close the file descriptor, but don't remove it from the task list, you'll get a Key conflict error when the next time a new socket with that file descriptor number is assigned a 'task', or when the event handler processing '*Run' passes the closed file descriptor to select(2). what's the reason for running it via (task) in the first place? A task is a piece of code to be run whenever a certain event happens. An event is either the expiry of a timer, or the arrival of data on a file descriptor. There may be several timers and many file descriptors where the process is expecting something to happen, so they are all put into a list in '*Run', and handled with I/O Multiplexing (the POSIX select(2) system call). there you go. thank you. Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
paging Tomas Hlavaty for java.wl license
Hi Tomas, was wondering if you'd like to attach an explicit license to wl (hopefully MIT too, or new BSD :) since i cannot find one (or maybe i'm just a moron not to find it). i'm hoping i could use java.wl for some open source stuff (currently vaporware) i'm doing. btw, wl.java is a great start for wannabe language implementors like me. thank you for that (and the very helpful picolisp community). /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
autogenerated keys for pilog
Hi List, what's the best way to go about autogenerated keys for Pilog? i've been digging around a bit and can't find anything related. hints and pointers would be most appreciated. thank you. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: autogenerated keys for pilog
H Alex, On Thu, Mar 31, 2011 at 2:42 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, what's the best way to go about autogenerated keys for Pilog? i've been digging around a bit and can't find anything related. Hmm, I am not sure what that is. You mean something along autogenerated keys in JDBC? I find JDBC's auto-generated keys feature provides a way to retrieve values from columns that are part of an index or have a default value assigned In the PicoLisp DB there are no keys used to join tables. Instead, objects are directly linked, i.e. they refer to each other (in properties) by containing other objets, or lists of objects. check. another insight. thank you. The relation maintenance classes for that are '+Link' and '+Joint'. In fact, there is a function 'genKey' which generates keys, but this is different from how I understand the above autogenerated keys. 'genKey' simply returns a new number not yet existent in an index. i think this hits the nail squarely. Can you give an example what you'd like to do? suppose there is a song entitled evergreen. one was sung by barbra streisand, the other by will young. different lyrics, but same song title. i'd like to use SongTitle as a key, but it won;t cut it, as in the example above. so i have to revert to something like an autoincrement key. this is it! thank you. Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
understanding (id) (was Re: autogenerated keys for pilog)
Hi Henrik, On Thu, Mar 31, 2011 at 2:48 PM, Henrik Sarvell hsarv...@gmail.com wrote: You get the same behavior as auto increment in for instance MySQL with this one: http://software-lab.de/doc/refG.html#genKey I use it a lot in the VizReader code. great. thank you. I'm not so sure I would've used it much though if I had know about the (id) function from the start: http://software-lab.de/doc/refI.html#id i'd like to understand this more. can you please give an illustration on how to do this? In my case it's all about displaying a nice id that is easy to relate for humans and JavaScript in my non-traditional gui and as you can see both approaches accomplish that but that latter is imo more elegant. /Henrik Sarvell On Thu, Mar 31, 2011 at 1:26 PM, Jakob Eriksson ja...@vmlinux.org wrote: On Thu, Mar 31, 2011 at 09:57:06AM +0800, Edwin Eyan Moragas wrote: Hi List, what's the best way to go about autogenerated keys for Pilog? i've been digging around a bit and can't find anything related. hints and pointers would be most appreciated. thank you. Hi! I have copied your (very good) question to: http://stackoverflow.com/questions/5496015/autogenerated-keys-in-pilog As I want to promote the Picolisp language on that site, but I have a hard time thinking up good questions and answers on my own. best regards, Jakob -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: understanding transient symbols, (====) and dynamic binding
On Wed, Apr 13, 2011 at 2:15 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, if i understand correctly, () when called outside a method works only when a file is (load)ed? I'm not sure what you mean here. Isn't that the case for _any_ function? check. i realized this after sending my first email. To my understanding '' is a normal function, which clears the transient symbol table as a side effect. right. again, more insight. only the transient table is cleared, but the transient symbols themselves are left alone. additionally, the reference says: That means, a transient symbol cannot be accessed then by its name, and there may be several transient symbols in the system having the same name. i had a hard time understanding this until i checked the source, which just wipes out the transient hash table but leaves the heap alone. replying to myself, wrote a test... (de test1 () (if (not (num? ctr)) (setq ctr 1) (inc 'ctr) ) (prinl ctr) () ) (test1) (test1) (prinl ctr) did a face palm and tells myself that's how a static behaves in C!. did that explain that right? Yes, exactly. 'test1' will always refer to ctr, even if it is gone out of visibility (transient scope) for the outside world. after writing the test (damn, i should be writing more tests rather than asking...), reading the C code and reading and re-reading the reference i now have a clearer picture of how transient symbols are used. this is just part of unlearning i think. elegance! thank you Alex for sharing all of this. (wishful thinking: alternative syntax for representing transient symbols without embedded spaces. maybe something like \xxx) /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: understanding transient symbols, (====) and dynamic binding
On Wed, Apr 13, 2011 at 3:22 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, (wishful thinking: alternative syntax for representing transient symbols without embedded spaces. maybe something like \xxx) Yeah, that's something I'm missing too. Transient symbols which represent not just strings simply look ugly. In fact, there exists something along that line, though not really a syntax in the traditional sense: Transient symbol markup. this is way too complex. perhaps i didn't express myself clearly. i was looking at something far simpler. at the moment, we have transient symbols as my transient symbols. this syntax for this is just fine. what if we can express MyParam, a transient symbol as \MyParam (or something similar) (a personal quirk, i dislike double quotes too much). maybe implement '\' as a read macro (?) to convert \MyParam to MyParam during read time. anyway, i'm not pushing this too much. was just thinking out loud. This is something tc.rucho and I were experimenting with for quite a while. You can see it going in and out over the years if you search for '*Tsm' in CHANGES. In the end (since 3.0.6) we disabled it by default, as it caused too much confusion. But the mechanics for *Tsm are still in the system (doc/refT.html#*Tsm). You can switch in on by loading lib/tsm.l: $ ./p lib/tsm.l + or $ ./dbg lib/tsm.l Transient symbol markup causes transient symbols to be displayed in an underlined font on the console, without the double quotes. Once you get used to it, it is a nice thing. The drawback is that it is difficult to be supported in most editors (tc.rucho implemented some support for it in emacs, and partially in vim). Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: understanding transient symbols, (====) and dynamic binding
On Wed, Apr 13, 2011 at 7:45 PM, Alexander Burger a...@software-lab.de wrote: On Wed, Apr 13, 2011 at 10:57:17AM +0200, Alexander Burger wrote: I was thinking of re-introducing a syntax like :MyParam, or perhaps MyParam, but this saves only a single character over MyParam, and doesn't look so very much better. The main reason of not doing this was that it adds yet another syntax rule, increasing the confusion about transient symbols. If both :MyParam and My Param are transient symbols, people will forget about the fact that they are the same internally, and then a statement where indeed a I forgot to mention an even bigger problem if two alternative syntaxes are used for transient symbols: Which one should be used for output (print, pp, edit etc.)? more problems. :) will leave this for now and maybe someday when i understand the internals more i can present some solution. Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
named, interned symbols
Hi list, several things that need help on. i encounter named symbol and interned symbol a lot in the docs. i just want to make sure that i understand things correctly. 1) named symbols are symbols that have a name in their property cell 2) interned symbols are symbols that are bound to symbol name, ie (setq) and having an entry in the intern hash 3) anonymous symbols are symbols that aren't bound to a symbol name, but can be accessed as a member of a list 4) transient symbols are named and bound (ie having an entry in the transient hash). did i get those right? /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: Changed '@' to '!' for functions in URLs
On Sun, May 8, 2011 at 6:31 PM, Jakob Eriksson ja...@vmlinux.org wrote: In which Linux distributions? This is great news! debian for one. not sure about the others yet. // Jakob -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
more questions on transient symbols
Hi Alex, the effect of () is that it clears the transient index. how then does the interpreter find the correct one to achieve the as identifiers with a limited access scope (like, for example, static identifiers in the C language family) feature as pointed out in the reference after calling ()? this is more of a curiosity since i can't figure it out from the code. thank you. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: more questions on transient symbols
On Mon, May 23, 2011 at 12:19 AM, Edwin Eyan Moragas e...@yndy.org wrote: Hi Alex, the effect of () is that it clears the transient index. how then does the interpreter find the correct one to achieve the as identifiers with a limited access scope (like, for example, static identifiers in the C language family) feature as pointed out in the reference after calling ()? this is more of a curiosity since i can't figure it out from the code. thank you. /e here's a code sample of what i've been playing with: eyan@faya $ dbg : (setq a 234) - 234 : (de x (a b) () (pack a b)) - x : (setq y (x 3 4)) - 34 : (prinl y) 34 - 34 : (prinl a) a - a : inside (x), all's well even with the transient index cleared. the thing i'm afraid of is clearing the transient index and getting the wrong transient symbol in the process. any more clues or pointers would be most appreciated. thank you. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: more questions on transient symbols
Hi Alex, On Mon, May 23, 2011 at 2:00 PM, Alexander Burger a...@software-lab.de wrote: The interpreter _first_ 'read's the list (de x (a b) () (pack a b)) and _then_ executes it, i.e. assignes the list ((a b) () (pack a b)) to the symbol 'x'. The function '' is not executed yet, and the transient index is not changed. aha! thank you. hmmm. i need to understand the difference between run-time and read-time more. cheers! /e Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
sorting db queries
Hi list, i'm currently playing with pilog and the database. i'm trying to look for an example of sorting the output of database queries by a given attribute. as a trivial example: (class +MyClass +Entity) (rel date (+Date)) and if i run a query, i'd like to sort the output, say, by increasing date. sorry if i missed this somewhere. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: sorting db queries
Hi Alex, thank you. just what i need. /e On Tue, May 24, 2011 at 2:09 AM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, i'm trying to look for an example of sorting the output of database queries by a given attribute. ... (class +MyClass +Entity) (rel date (+Date)) ... and if i run a query, i'd like to sort the output, say, by increasing date. There are two possibilities: 1. If this sorting is often necessary, you maintain an index for that attribute: (rel date (+Ref +Date)) Then all queries on that attribute will directly return sorted results: (collect 'date '+MyClass) (collect 'date '+MyClass fromDate tillDate) (pilog '((db date +MyClass @M) ...) ...) Note, however, if 'select' is used with multiple search attributes, the resulting order is not defined. 2. If you need to do that only once or seldom, sort the results after fetching them: (by '((This) (: date)) sort (collect 'someOtherAttribute '+MyClass) ) This has the disadvantage that first all results must be fetched, of course, before they can be sorted. Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
limiting results of select/3
Hi list, after checking the 'select' predicate doc, i can't find any way to limit the number of results of the query. i'm thinking along the lines of the LIMIT clause of SQL. thank you. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
running test suite in mac osx error: IP bind error: Address already
Hi guys, any chance ideas on how to fix this on osx? /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: running test suite in mac osx error: IP bind error: Address
Hi Alex, On Fri, Jun 10, 2011 at 1:16 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, any chance ideas on how to fix this on osx? In which situation does this error appear? not sure which specific one. but the ouput's here: eyan@faya $ $(/bin/pwd)/p lib/test.l -bye + [test/lib.l:5] !? (port T ) IP bind error: Address already in use ? # lsof -i :port was using netstat and the lsof invocation above. the port's ok. this also happens in obsd by the way. btw, if it is significant, i'm sticking with 3.0.5. i tried this with the current version too and it's the same story. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: running test suite in mac osx error: IP bind error: Address
Hi Alex, On Fri, Jun 10, 2011 at 4:05 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, eyan@faya $ $(/bin/pwd)/p lib/test.l -bye + [test/lib.l:5] !? (port T ) IP bind error: Address already in use Oh, I see. This are the unit tests, and it is the place where it tries to bind to a UDP port. So it seems the UDP handling of PicoLisp doesn't work under Mac OS (and OpenBSD, where Mac OS is derived from, iirc). check this. In the past, we have tested it a lot under FreeBSD, and couldn't see any problems with that. Is there anybody who can debug it under Mac OS or OpenBSD? i'll be on it. not sure about time frame tho since i'm diddling with an app at the moment. looks like i'm the only one running it on obsd so i guess it's not an urgent issue. :) btw, if it is significant, i'm sticking with 3.0.5. i tried this with the current version too and it's the same story. No problem, I think. The network stuff didn't change recently. What comes to my mind, though: PicoLisp is not IPv6 ready yet (see previous discussions here). Could this be a reason? probably not. IPv6 not enabled in my obsd setup. thanks Alex. was silly i should have mentioned running the test suite in my first email. /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: running test suite in mac osx error: IP bind error: Address
On Fri, Jun 10, 2011 at 5:43 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, i'll be on it. not sure about time frame tho since i'm diddling with Great! an app at the moment. looks like i'm the only one running it on obsd so i guess it's not an urgent issue. :) We are now discussing it also in IRC. ckeen checked on his BSD system, and the UDP binding worked. We ran into other problems, though, the signal() calls don't seem to work :( ctrl-c when repl is running is not working on obsd. not sure if this is related to the signal() call. it's in my todo list. Too bad ... Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: running test suite in mac osx error: IP bind error: Address
whopee! thanks Alex and ckeen! On Fri, Jun 10, 2011 at 8:24 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, We are at it ... :) We've got it!! Thanks to ckeen! :) There were actually two problems with the OpenBSD version: 1. Signal handling was wrong. There was the possibility of an array out-of-bound access because of a non-portable max signal number. 2. UDP handling was indeed wrong. The port was used twice during the tests, and not closed in between. It is just that Linux doesn't give an error in that case. Both should be fixed now. If you like to try, please download the latest version, e.g. wget software-lab.de/picoLisp.tgz Many thanks for your bug report! Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: running test suite in mac osx error: IP bind error: Address
downloaded and ran the unit tests now runs fine on osx snow leopard and obsd 4.8 32 bit. thanks a bunch again guys! On Fri, Jun 10, 2011 at 9:01 PM, Edwin Eyan Moragas e...@yndy.org wrote: whopee! thanks Alex and ckeen! On Fri, Jun 10, 2011 at 8:24 PM, Alexander Burger a...@software-lab.de wrote: Hi Edwin, We are at it ... :) We've got it!! Thanks to ckeen! :) There were actually two problems with the OpenBSD version: 1. Signal handling was wrong. There was the possibility of an array out-of-bound access because of a non-portable max signal number. 2. UDP handling was indeed wrong. The port was used twice during the tests, and not closed in between. It is just that Linux doesn't give an error in that case. Both should be fixed now. If you like to try, please download the latest version, e.g. wget software-lab.de/picoLisp.tgz Many thanks for your bug report! Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: android picolisp using ndk/jni
Hi Doug, you've been pretty busy. :) cheering everybody on... /e On Thu, Jul 7, 2011 at 4:43 PM, Doug Snead semaphore_2...@yahoo.com wrote: http://developer.android.com/sdk/ndk/overview.html : Typical good candidates for the NDK are self-contained, CPU-intensive operations that don't allocate much memory, such as signal processing, physics simulation, and so on. Ok, here's a plan for :-) a real Android picolisp ... start by taking a version of picolisp, and begin cutting it back by dropping everything ndk c/c++ headers do not support. So, the idea would be you'd get much of picolisp, hopefully more than minipicolisp, at least as much as ersatz, (I would think), and with no java interpretation penalty for the lisp interpreter. re: ... that don't allocate much memory. Fine; allocate what you need per application on init (from java), pass it to android picolisp to use. -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: The PicoLisp Ticker
On Fri, Jul 15, 2011 at 4:13 PM, Doug Snead semaphore_2...@yahoo.com wrote: I think you created a bot-trap :-) aha! that figures. i'm subscribed to google alerts with 'picolisp' and 'pico lisp' as keyword. i have gotten 4 so far from that wacky ticker of yours. but don't remove it, please! it *is* fun. --- On Fri, 7/15/11, Alexander Burger a...@software-lab.de wrote: From: Alexander Burger a...@software-lab.de Subject: The PicoLisp Ticker To: picolisp@software-lab.de Date: Friday, July 15, 2011, 12:58 AM Hi all, let me share my funny experiences about the PicoLisp Ticker, and how it is handled by Mr. Google: http://picolisp.com/5000/-2-1i.html Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subjectUnsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
doProve (was Re: Ninety-Nine Lisp Problems: pilog)
(BTW - Many people might not realise the heart of pilog, the prove function, is written in C and seems very fast. For some prolog applications, it may be fast or faster than compiled prologs - would like to see some benchmarks to investigate that feeling. But for years, I was under the false impression that pilog was completely written in lisp and so pilog programs were doubly interpreted, but I was wrong: the heart of the pilog interpreter is C.) now this got me interested. does Alex, or anybody else, have a picolisp version of the doProve() and doUnify() c functions? /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: doProve (was Re: Ninety-Nine Lisp Problems: pilog)
Hi Alex, On Sun, Jul 17, 2011 at 3:55 PM, Alexander Burger a...@software-lab.de wrote: Hmm, after some searching, it seems I have only a fragment of some initial version, when I experimented in Lisp before I rewrote the time critical parts in C. There is 'prove', but 'unify' seems missing. Anway: thanks for this! i'll find this useful for my personal research. :) cheers /e (de prove (Q) (when Q (use (N NL Alt Tp1 Tp2 Env E) (back) (while (or Tp1 Tp2) (cond (Alt (ifn (setq E (unify (car NL) (cdar Tp1) N (caar Alt) Env) ) (or (setq Alt (cdr Alt)) (back)) (when (cdr Alt) (push Q (cons N NL (cdr Alt) Tp1 Tp2 Env) ) ) (push 'NL N) (inc 'N) (setq Tp2 (cons (cdr Tp1) Tp2) Tp1 (cdar Alt) Env E ) (off Alt) ) ) ((not Tp1) (setq Tp1 (pop 'Tp2) NL (cdr NL)) ) ((=T (car Tp1)) (while (= (caaar Q) (car NL)) (pop Q) ) (pop 'Tp1) ) ((num? (caar Tp1)) (setq E (eval (cdar Tp1))) (push 'NL (get NL (caar Tp1))) (setq Tp2 (cons (cdr Tp1) Tp2) Tp1 E) ) ((pat? (caar Tp1)) (if (and (eval (cdar Tp1)) (unify (car NL) (caar Tp1) (car NL) @ Env) ) (setq Tp1 (cdr Tp1) Env @) (back) ) ) ((not (setq Alt (get (caar Tp1) T))) (back) ) ) ) (let R NIL (for (L Env (cdr L) (cdr L)) (when (=0 (caaar L)) (push 'R (cons (cdaar L) (lookup 0 (cdaar L) Env)) ) ) ) (or R (bool Env)) ) ) ) ) (de back () (setq Env (pop Q) N (pop 'Env) NL (pop 'Env) Alt (pop 'Env) Tp1 (pop 'Env) Tp2 (pop 'Env) ) ) (de - X (lookup (get NL (or (cadr X) 1)) (car X) Env ) ) (de unifyEnv (Lst) (setq Env (unify (cadr NL) Lst (car NL) Lst Env) ) ) Cheers, - Alex -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
difference between (let) and (bind)
Hi list, looking at the docs, looks like (bind) is just a fancy way of doing (let). is this correct? i tried looking at the source but i'm still trying to decipher it. also checked out wl.java. both (let) and (bind) are quite similar. is this so? /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: difference between (let) and (bind)
Hi Tomas, On Thu, Jul 28, 2011 at 1:27 AM, Tomas Hlavaty t...@logand.com wrote: Hi Edwin, roughly speaking, bind is a function while let is a macro. They differ in argument evaluation. bind is programmable while let is more of a syntax thingy. thank you for this. thinking about it a bit, i can just do away with one and live with the other? say, can i just live with (bind)? Cheers, Tomas -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: difference between (let) and (bind)
On Thu, Jul 28, 2011 at 4:48 AM, Tomas Hlavaty t...@logand.com wrote: Hi Edwin, thinking about it a bit, i can just do away with one and live with the other? say, can i just live with (bind)? yes, but it is more convenient to write let in your programs instead of the bind with constructing the binding list manually. However, in certain situations, one needs an evaluating version, e.g. when you construct code programmatically. I hit this situation in the wl reading the docs again after this, (let) evaluates while (bind) does not. i hope i got that right. (yes sir, still learning. i'm actually in the process of skinning minipicolisp to retain the wl functions. what i'm trying to do is understand the minipicolisp interpreter) interpreter (the one written in Java) where I wanted to implement as much as possible in pure picolisp (as opposed to Java). There I needed evaluating versions of some functions. I don't remember off the top of my head which cases were those functions but I usually named them with a leading dot character to make it obvious they were internal. When you have the evaluating version it becomes easy to implement the macro version on top of them. i see. but i still have to experience them firsthand. looking at the java.wl, you have (up.) and (if.). i'll let this stew some. please don't hesitate to shed further enlightenment on this tho. thank you. /e Cheers, Tomas -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
picolisp on openbsd amd64
Hi list, tried compiling picolisp on an openbsd amd64 install and it seems out of wack. i tried diddling the gcc flags to no avail. as of now, i don't have any solution to this. to make things compile: - added '-fPIC' to CFLAGS in the Makefile. - changed all instances of -m32 to -m64 giving dbg a spin: eyan@imango $ ./dbg n/li open: No such file or directory ? i'm not using and64 full time at the moment, and won't be running picolisp code in it. someone out there might just have the solution. best /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: picolisp on openbsd amd64
Hi Tomas On Tue, Aug 9, 2011 at 2:50 AM, Tomas Hlavaty t...@logand.com wrote: Hi Edwin, tried compiling picolisp on an openbsd amd64 install and it seems out - changed all instances of -m32 to -m64 the -m32 is there for a reason, you can't just replace it and expect it to work:-D i figured as much. haha! was just shooting blindly. I suppose you are trying to compile the C version, which is 32 bit only. ah no. just trying to see if the 32 bit version builds correctly on obsd-amd64. Alex implemented the 64 bit version in assembly so it is a different program (in src64 directory). Try to build this one if you are after 64 bit picolisp. yeah. :) Cheers, Tomas -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
evaluating versions (was Re: when to use (prog) or (run))
Hi Tomas, On Tue, Aug 9, 2011 at 2:46 AM, Tomas Hlavaty t...@logand.com wrote: Hi Edwin, having difficulty knowing when to use (prog) or (run). bottomline is, when is it best to use one over the other? 'run' is more general and evaluating version, i.e. it's a function suitable for manipulating programs (aka eval) with other bell and whistles like environment control. 'prog' is a macro, i.e. a convenience shortcut for writing text of a program by hand. You can build the macro version on to of the evaluating version. thanks much for the replies. every time i think i got something, i run into more questions. i've run across 'evaluating' and 'non evaluating' versions of functions. not just in this list but also in reading materials. i haven't experienced the difference between the two yet. in any case, i guess i'll just have to grow into it. think and program in the language more. even tho it would be slow. :) best /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Re: evList2() in minipicolisp
Hi Alex,
thanks for a thorough explanation as always!
cheers
On Tue, Aug 16, 2011 at 5:01 AM, Alexander Burger a...@software-lab.de wrote:
Hi Edwin,
need some help. when is evList2() called in minipicolisp?
i can't decipher when it is called.
comments are mine.
if (isNum(foo = car(ex)))
return ex; /* return numbers as is */
Right. This is the case where the evaluated list starts with a number,
e.g. (1 2 3 4), or also (123 a b c).
if (isCell(foo)) {
'foo' is the CAR of the list, and it is a list here. We might have a
call like ((get 'x 'y) 1 2 3), so 'foo' is (get 'x 'y) which - when
evaluated - is supposed to return a function.
if (isNum(foo = evList(foo)))
return evSubr(foo, ex); /* if a list evaluates to a number,
call a C subroutine */
Yes. Evaluation of (get 'x 'y) resulted in a number, so a C function is
called.
return evList2(foo, ex); /* then... ? */
'foo' is not a number, so these are the remaining possibilities: It must
be either a symbol or a list. It might be a symbol if we have e.g. (x 1
2 3) where the value of 'x' is 'y', which in turn is defined as a
function (or something else suitable to a recursive invocation of the
whole 'evList' machinery). Otherwise, it is a list which is taken as a
function (i.e. (params . body).
'evList2' is put into a separate function for a single reason: The value
must be pushed on the stack to be safe from garbage collection (the
'Push' marcro at the start of 'evList2').
This problem we have only in the C version, because the compiler
allocates space in the stack frame for 'c1' on every call to 'evList',
which is a waste of stack space in most cases, as normal function
calls don't pass through here.
Cheers,
- Alex
--
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
--
UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
help in deciphering isSymb() macro
Hi List, came across these macros in pico.h: #define isSym(x)(num(x)WORD) #define isSymb(x) ((num(x)(WORD+2))==WORD) what is the intent of isSymb()? as compared to isSym()? yes, still on that long path of understanding the C implementation. :). this is still minipicolisp by the way. cheers /e -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Unsubscribe
Good bye Edwin Eyan Moragas e...@yndy.org :-( You are now unsubscribed -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
Subscribe
Hello Edwin Eyan Moragas e...@yndy.org :-) You are now subscribed -- UNSUBSCRIBE: mailto:picolisp@software-lab.de?subject=Unsubscribe
