freenode/lisp - IRC Chatlog

3:12:22 loke pierpa: The spec does not place any restrictions or guarantees on either of those functions.

3:13:19 Bike that's what they led with

3:27:54 pierpa yes, but the WITH-HASH-TABLE-ITERATOR version would be unreasonable if it was not O(1). The HASH-TABLE-COUNT would not be completely crazy if in some circumstances it would not be O(1)

3:28:41 pierpa for example, in Racket (the equivalent of) HASH-TABLE-COUNT is O(1) only for non-weak tables.

3:29:53 pierpa so, in Racket the (equivalent of) second function would be preferable.

3:37:44 loke pierpa: I could imagine the case where the second version is also not O(1), although I agree with you that it's less likely.

5:52:06 Harag1 ** NICK Harag

9:28:05 John[Lisbeth] Is there a common lisp implementantion of latex so I can point out people should use that instead of latex?

9:29:53 Bike what on earth would be the point of rewriting latex

9:30:10 John[Lisbeth] So I don't have to learn Latex and I only have to learn common lisp

9:31:22 Bike you would still have to learn latex to use an implementation of it

9:31:41 John[Lisbeth] I couldn't use some domain specific langauge that lets me make pretty pdfs in common lisp?

9:31:51 Bike you could, but that would not be latex

9:32:01 John[Lisbeth] well I mean a common lisp version of it

9:32:04 John[Lisbeth] such as latex or tex

9:32:15 Bike you mean a common lisp typesetting system, not a common lisp latex

9:32:19 Bike latex is a pretty specific thing

9:32:25 John[Lisbeth] I don't know the word for it sorry

9:32:35 John[Lisbeth] yes a lisp typesetting system so I can format pretty characters in pdfs in common lisp

9:33:31 Bike i know fare uses this thing https://docs.racket-lang.org/scribble/

9:33:37 Bike dunno about something in lisp

9:33:49 John[Lisbeth] probably not highly sought after

9:34:01 Bike well, latex is really good

9:34:01 schjetne John[Lisbeth]: try CommonDoc

9:34:06 schjetne http://commondoc.github.io

9:34:56 John[Lisbeth] It says it uses vertex

9:35:07 John[Lisbeth] is this common lisp? https://github.com/CommonDoc/vertex

9:35:09 schjetne You don't have to use it if you don't want to

9:35:15 Bike and then that using tex in a vertex document sucks. heh.

9:35:16 John[Lisbeth] I am just curious about it

9:36:12 John[Lisbeth] oh well thanx

10:00:17 kdas__ ** NICK kushal

10:45:23 ``Erik_ ** NICK ``Erik

10:56:29 VitoVan Hey guys, good day.

10:57:00 VitoVan I recently is reading <On Lisp>, I am stuck in chapter 12.4, http://ep.yimg.com/ty/cdn/paulgraham/onlisp.pdf

10:57:05 VitoVan page 184

10:58:02 VitoVan Which the author is trying to construct a macro _f to make things like this: (_f + x y) the same as (incf x y)

10:58:52 VitoVan and then have given a wrong example: (defmacro _f (op place &rest args) '(setf ,place (,op ,place ,@args)))

10:59:28 VitoVan which I can't find anything wrong with this example, and I can't understand why "Unfortunately, we can’t define a correct f with define-modify-macro, because the operator to be applied to the generalized variable is given as an argument."

10:59:41 VitoVan Hope anyone can help, thank you very much!

11:02:24 VitoVan Forgive my poor English, I hope that won't bothering you.

11:03:15 malice` VitoVan: setf is a quasi-magical macro that replaces "setf X Y" with proper form

11:03:32 Grue`` fwiw I tried (let ((x (list 1))) (_f + (car x) 10) x) and it works

11:03:45 Grue`` there might be some complex case where it fails though

11:04:46 Grue`` something about ,place being evaluated twice

11:04:47 VitoVan malice`: then what's the problem?

11:05:38 Grue`` for example Graham offers the example (incf (aref a (incf i))). with the naive implementation i would be incremented twice

11:05:47 Grue`` or something

11:06:05 VitoVan Grue``: Yes, I aware the eval twice thing, but it should be *not good* instead of **wrong**

11:06:33 VitoVan Thank you Grue``

11:06:37 jdz in C that would be "not good", in CL it is "wrong"

11:07:51 VitoVan Right, the double evals of incf is truly wrong.

11:14:22 VitoVan Thank you very much Grue``, and malice and jdz, problem solved, what a magical place~

11:19:20 pipping there's no place like (user-homedir-pathname)

11:36:18 vito_ ** NICK Guest44279

11:36:43 Guest44279 ** NICK VitoVan

11:39:02 kdas__ ** NICK kushal

12:02:44 flip214 does lparallel have some sort function as well?

12:05:54 scymtym flip214: there is LPARALLEL:PSORT which can sort vectors in parallel (according to the documentation string)

12:10:08 flip214 scymtym: thanks a lot, didn't see that in the documentation at https://lparallel.org/api/

12:10:37 flip214 ah, hiding in "Cognates"

12:32:52 ovenpasta1 ** NICK ovenpasta

12:37:16 beach Good afternoon everyone!

13:04:37 Xach Hello beach

13:54:30 Posterdati hi

13:56:06 Posterdati I've got problem to quickload antik on sbcl 1.3.8-26 for x86_64: caught error: illegal function call

13:56:44 beach What code is causing the problem?

13:58:46 Posterdati cffi-libffi

13:59:11 beach Can you be more specific? What form is causing the problem?

14:00:31 Posterdati Illegal function call

14:00:51 jackdaniel Posterdati: beach asks for a particular code snippet which contains that illegal function call

14:00:55 Posterdati While evaluating the form starting at line 30, column 0

14:01:03 jackdaniel you may find it with a backtrace for instance

14:01:29 beach Posterdati: Can you give us the form starting at line 30, column 0?

14:02:52 Posterdati (eval-when (:compile-toplevel :execute) (asdf:oos 'asdf:load-op :cffi-grovel) (asdf:oos 'asdf:load-op :trivial-features))

14:04:33 beach I don't see anything wrong with that code.

14:05:10 Posterdati *features* reports :asdf3.1

14:05:23 beach There is no illegal function call in that form.

14:18:02 scymtym antik seems to depend on gsll. over the last weeks people have reported problems with gsll caused by changes in the gsl C library that are not (yet) reflected in the gsll ffi bindings. the only suggestion i have seen is downgrading the gsl C library

14:18:42 sampiero hi

14:18:56 beach Hello sampiero.

14:19:16 LiamH antik-base can be loaded without GSLL

14:20:21 sampiero http://paste.lisp.org/display/322684

14:20:49 sampiero this is the problem trying to load :cffi-libffi

14:21:02 sampiero and thus :antik

14:21:58 beach It looks like there is a conditional form that is not recognized as such.

14:22:44 beach Can you paste the contents of the file asdf-compat.lisp?

14:22:55 sampiero LiamH: I need gsll too

14:23:43 LiamH sampiero: OK, well just keep in mind it will not load with GSL 2.x. However, I have in the repo a version that may load.

14:23:59 sampiero :(

14:25:48 sampiero http://paste.lisp.org/display/322684#1

14:26:45 beach I don't think the code in that paste corresponds to the code giving the error.

14:27:29 scymtym beach: i think it does. it may be OS-COND again

14:27:48 sampiero I pasted the file you told me

14:28:51 sampiero and?

14:29:02 scymtym assuming OS-COND to be a function would put things like ((UIOP/OS:OS-WINDOWS-P) "dll") mentioned in the error message in an evaluated position

14:29:05 beach I OK, I see it.

14:29:24 beach Yes, os-cond is not recognized as a macro.

14:29:39 beach That is what I meant by a conditional form that is not recognized as such.

14:29:48 sampiero ok

14:29:53 sampiero is there a cure?

14:29:59 scymtym sampiero: not sure, but your asdf may be too old (see http://ircbrowse.net/browse/lisp?q=os-cond).

14:30:14 scymtym what is your asdf version?

14:30:30 sampiero :ASDF-PACKAGE-SYSTEM :ASDF3.1 :ASDF3 :ASDF2 :ASDF :OS-UNIX

14:31:11 scymtym please try (asdf:component-version (asdf:find-system :asdf))

14:31:43 sampiero "3.1.5"

14:34:29 sampiero ?

14:35:35 scymtym sorry, that macro seems to be provided by the uiop system. what does (asdf:component-version (asdf:find-system :uiop)) report?

14:37:17 sampiero "3.1.0.4"

14:38:13 sampiero from the latest quicklisp installation

14:38:25 Xach quicklisp doesn't provide asdf 3. it comes with your implementation.

14:38:30 Xach or manually

14:38:42 sampiero uiop too?

14:38:45 jackdaniel yes

14:39:12 Xach quicklisp *does* provide uiop.

14:40:02 sampiero is "3.1.0.4" the latest uiop?

14:40:26 Xach No, 3.1.6 is available. Not sure if that's the newest, but it is newer.

14:42:27 jackdaniel oh, sorry

14:43:28 devon g'day all

14:43:37 sampiero not to all

14:43:44 beach Hello devon.

14:44:48 devon Know of a file output stream that automatically takes care of preventing partially written files from appearing under the real file name?

14:46:43 devon E.g., as Emacs makes foo a placeholder link to nowhere, writes to #foo# and then renames it to the final real foo.

14:49:17 cromachina http://clhs.lisp.se/Body/f_finish.htm these functions help with flushing streams

14:51:52 scymtym cffi-toolchain.asd seems to assume that the asdf version is identical to the uiop version

14:52:31 sampiero scymtym: I'm trying with ecl 16.1.2

14:55:05 devon Here I go reinventing this old wheel once again. I should set it to music, _the Lisper's Lament_

14:55:24 sampiero http://paste.lisp.org/display/322684#2

14:55:31 sampiero scymtym: http://paste.lisp.org/display/322684#2