2014-10-10T00:16:13Z zRecursive joined #lisp 2014-10-10T00:16:45Z whmark joined #lisp 2014-10-10T00:16:51Z zRecursive quit (Remote host closed the connection) 2014-10-10T00:17:05Z zRecursive joined #lisp 2014-10-10T00:17:15Z knosys quit (Quit: WeeChat 0.4.2) 2014-10-10T00:17:28Z innertracks joined #lisp 2014-10-10T00:19:33Z whmark quit (Remote host closed the connection) 2014-10-10T00:21:49Z Stunt123 joined #lisp 2014-10-10T00:22:06Z banjara quit (Quit: Leaving.) 2014-10-10T00:29:05Z zz_karupa is now known as karupa 2014-10-10T00:31:51Z innertracks quit (Ping timeout: 258 seconds) 2014-10-10T00:34:20Z gabriel_laddel quit (Ping timeout: 260 seconds) 2014-10-10T00:35:59Z varjag_ quit (Quit: Connection closed for inactivity) 2014-10-10T00:36:12Z pjb: nowhere_man: 1- what REPL? 2- you can check what command is bound to a key chord with C-h k For example, here, in slime repl, C-h k M-p gives slime-repl-previous-input and C-h k M-n gives slime-repl-next-input. 2014-10-10T00:36:30Z pjb: nowhere_man: once you've identified the command that's bound to the key, you can do something. 2014-10-10T00:36:39Z pjb: depending on that command. 2014-10-10T00:38:43Z milosn quit (Ping timeout: 272 seconds) 2014-10-10T00:41:24Z ltbarcly quit (Quit: Computer has gone to sleep.) 2014-10-10T00:43:07Z simulacrum quit (Ping timeout: 245 seconds) 2014-10-10T00:44:37Z nell quit (Quit: WeeChat 1.1-dev) 2014-10-10T00:44:55Z simulacrum joined #lisp 2014-10-10T00:47:36Z Bicyclidine joined #lisp 2014-10-10T00:51:53Z nowhere_man: pjb: M-p is bound to (slime-repl-previous-input) 2014-10-10T00:54:30Z zyaku quit (Ping timeout: 250 seconds) 2014-10-10T00:54:32Z knosys joined #lisp 2014-10-10T00:57:10Z devon: Is there a reader macro 2014-10-10T00:57:31Z devon: for \n and so on in strings? 2014-10-10T00:58:20Z rme quit (Quit: rme) 2014-10-10T00:59:42Z oGMo: there's cl-interpolation or something like that 2014-10-10T00:59:59Z oGMo: cl-interpol 2014-10-10T01:01:20Z Vivitron`: devon: http://weitz.de/cl-interpol/ 2014-10-10T01:05:15Z _xinix_ joined #lisp 2014-10-10T01:05:42Z leo2007 quit (Read error: Connection reset by peer) 2014-10-10T01:08:39Z erikc joined #lisp 2014-10-10T01:09:02Z _xinix_: Hi all 2014-10-10T01:10:11Z zyaku joined #lisp 2014-10-10T01:10:39Z erikc quit (Client Quit) 2014-10-10T01:10:51Z zolk3ri quit (Quit: Lost terminal) 2014-10-10T01:10:56Z emma quit (Read error: Connection reset by peer) 2014-10-10T01:12:03Z xyjprc joined #lisp 2014-10-10T01:12:09Z rme joined #lisp 2014-10-10T01:13:47Z nell joined #lisp 2014-10-10T01:15:00Z macdice`` joined #lisp 2014-10-10T01:17:47Z macdice` quit (Ping timeout: 245 seconds) 2014-10-10T01:19:20Z urandom__ quit (Quit: Konversation terminated!) 2014-10-10T01:20:01Z c107 joined #lisp 2014-10-10T01:21:40Z znode joined #lisp 2014-10-10T01:21:58Z jusss joined #lisp 2014-10-10T01:22:18Z ivan\ joined #lisp 2014-10-10T01:25:59Z milosn joined #lisp 2014-10-10T01:28:30Z _xinix_: hiyosi, I am a guy from scheme to common lisp, i realized that in clisp functions used in a strange way compared to scheme, sometimes I must add "#'" before functions, and sometimes not... It is hard for me to distinguish the occasions prefixing "#'" or not. Could you tell me how? 2014-10-10T01:28:49Z oGMo: _xinix_: #' means "the function called..." 2014-10-10T01:28:51Z cneira quit (Ping timeout: 246 seconds) 2014-10-10T01:29:11Z oGMo: it's like 'foo but means (function foo) instead of (quote foo) 2014-10-10T01:29:48Z oGMo: you usually don't need it except when you really want to say "the function..." 2014-10-10T01:31:18Z oGMo: (actually saying it's like 'foo is misleading, because it's not quoting per se .. it evaluates to the function you named) 2014-10-10T01:31:39Z huza joined #lisp 2014-10-10T01:32:52Z leo2007 joined #lisp 2014-10-10T01:33:31Z innertracks joined #lisp 2014-10-10T01:34:17Z _xinix_: do you mean that foo is only a symbol-like thing, and #'foo can be evaluated to a FUNCTION OBJECT such as '0xSOMEADDR'? 2014-10-10T01:35:26Z pjb: nowhere_man: sounds good. Why doesn't it "work"? What happens when you type (+ 1 2) RET and then M-n ? 2014-10-10T01:35:34Z oGMo: _xinix_: in CL you should think of symbols as physical objects that name things (because they are) .. foo isn't a symbol-like thing, it _is_ a symbol 2014-10-10T01:36:53Z ynniv quit (Quit: ynniv) 2014-10-10T01:36:54Z oGMo: _xinix_: when you evaluate foo, if it's a variable (i.e. defvar, not lexical), you get the value of the variable named foo; when you evaluate 'foo you get the symbol foo; when you evaluate #'foo you get the function named foo 2014-10-10T01:37:04Z jusss quit (Remote host closed the connection) 2014-10-10T01:38:16Z nowhere_man: pjb: I get "End of history" in the mini-buffer 2014-10-10T01:38:30Z _xinix_: oGMo, what about foo? 2014-10-10T01:38:41Z oGMo: _xinix_: that's the first thing i said 2014-10-10T01:39:38Z pjb: devon: http://paste.lisp.org/display/137262 2014-10-10T01:39:38Z Bicyclidine: (eval 'foo) => value of foo, (eval ''foo) => the symbol FOO, (eval '(function foo)) => the function called foo 2014-10-10T01:40:02Z pjb: nowhere_man: end-of-history is because you don't have anymore entries in that direction. 2014-10-10T01:40:15Z pjb: Try to type more expressions in the repl, and navigate after. 2014-10-10T01:41:04Z _xinix_: oGMo, thank you very much, i think i have caught something... thank you very much 2014-10-10T01:41:38Z oGMo: _xinix_: also e.g. (defun foo ...) (funcall #'foo ...) 2014-10-10T01:41:57Z nowhere_man: pjb: if I do M-x slime-repl-previous-input to get back, I can get forward with M-n 2014-10-10T01:42:15Z oGMo: but you can also just say 'foo and it'll find it 2014-10-10T01:42:26Z pjb: nowhere_man: sounds logical. 2014-10-10T01:42:29Z resttime joined #lisp 2014-10-10T01:42:51Z nowhere_man: seems to be my WM interfering someway 2014-10-10T01:42:58Z _xinix_: enn, i need more exercises.... 2014-10-10T01:43:03Z pjb: nowhere_man: history is a doubly-linked-list. Not a circular list. You can go previous, until you reach the first one. And you can go next, until you reach the last one. 2014-10-10T01:43:25Z pjb: nowhere_man: of course, you should use a sane window manager. Try stumpwm or ratpoison. 2014-10-10T01:43:37Z nowhere_man: pjb: :-) 2014-10-10T01:44:59Z innertracks quit (Quit: innertracks) 2014-10-10T01:46:31Z znode quit (Quit: (null)) 2014-10-10T01:49:04Z grumio2 joined #lisp 2014-10-10T01:50:39Z cneira joined #lisp 2014-10-10T01:55:05Z ivan\ quit (Quit: ERC Version 5.3 (IRC client for Emacs)) 2014-10-10T01:56:09Z huza quit (Ping timeout: 246 seconds) 2014-10-10T01:57:58Z zRecursive: How to get the contents of emacs mini-buffer ? I want to copy the error messages of Mew 2014-10-10T01:58:25Z ivan4th quit (Ping timeout: 260 seconds) 2014-10-10T01:58:32Z huza joined #lisp 2014-10-10T02:01:50Z MutSbeta joined #lisp 2014-10-10T02:03:27Z Stunt124 joined #lisp 2014-10-10T02:04:52Z Stunt123 quit (Ping timeout: 260 seconds) 2014-10-10T02:05:29Z cpc26 joined #lisp 2014-10-10T02:07:17Z Bicyclidine quit (Ping timeout: 260 seconds) 2014-10-10T02:07:33Z Stunt123 joined #lisp 2014-10-10T02:07:42Z Stunt124 quit (Ping timeout: 246 seconds) 2014-10-10T02:07:45Z Stunt123 quit (Client Quit) 2014-10-10T02:12:14Z huza quit (Remote host closed the connection) 2014-10-10T02:13:11Z xyjprc quit (Remote host closed the connection) 2014-10-10T02:13:21Z huza joined #lisp 2014-10-10T02:14:27Z ekinmur joined #lisp 2014-10-10T02:17:22Z ynniv joined #lisp 2014-10-10T02:20:19Z ynniv quit (Client Quit) 2014-10-10T02:20:57Z huza quit (Ping timeout: 272 seconds) 2014-10-10T02:21:20Z Bicyclidine joined #lisp 2014-10-10T02:24:52Z beach joined #lisp 2014-10-10T02:25:00Z beach: Good morning everyone! 2014-10-10T02:25:00Z minion: beach, memo from pjb: for conforming code, it should not matter what implementation of environment is used to walk it. The semantics should be conserved, whatever that environment implementation. 2014-10-10T02:25:06Z ynniv joined #lisp 2014-10-10T02:26:39Z beach: pjb: Yes, sure. I have already taken care of that part. 2014-10-10T02:28:47Z ivan\ joined #lisp 2014-10-10T02:30:17Z beach: So I am thinking in the code that generates an AST from a form, when it sees (EVAL-WHEN (:COMPILE-TOPLEVEL) ) AND the compiler is the file compiler, AND the form is a top-level form, it should translate to an AST and then call the AST interpreter on the result. Unless I am not thinking clearly, it should work both in the native compiler and the cross compiler. 2014-10-10T02:31:05Z Bicyclidine: and then return nil or the ast equivalent, yes? 2014-10-10T02:31:19Z beach: Right. 2014-10-10T02:31:36Z Bicyclidine: nice and easy for :compile-toplevel :load-toplevel too 2014-10-10T02:31:45Z beach: Yes. 2014-10-10T02:32:23Z pjb: zRecursive: messages are written into the *Messages* buffer. 2014-10-10T02:32:25Z beach: Before, I thought I would have to create a generic function EVAL to be specialized for each implementation, but this solution seems to avoid that. 2014-10-10T02:32:53Z zRecursive: pjb: it is , thanks 2014-10-10T02:33:09Z pjb: C-x b *Messages* RET locate and copy-and-paste. 2014-10-10T02:33:23Z pjb: But C-h e seems to do too. 2014-10-10T02:33:42Z zRecursive: yeah, just got it from #emacs 2014-10-10T02:33:53Z slyrus joined #lisp 2014-10-10T02:34:59Z ndrei joined #lisp 2014-10-10T02:35:58Z zarkone joined #lisp 2014-10-10T02:42:24Z wglb: nowhere_man: I misspoke earlier. M-p and M-n work fine for me. Hope I did not add too much to your confusion. 2014-10-10T02:42:37Z KarlDscc joined #lisp 2014-10-10T02:44:25Z beach: pjb: I created a bunch of generic functions that are similar in spirit to the CLtL2 environment functions, but they create standard objects instead of returning multiple values, and they are generic, so each implementation can (has to) provide methods on those functions for their particular environments. 2014-10-10T02:45:14Z huza joined #lisp 2014-10-10T02:48:05Z ynniv quit (Quit: ynniv) 2014-10-10T02:49:06Z pjb: It's in the sicp repo right? 2014-10-10T02:49:11Z cneira quit (Ping timeout: 272 seconds) 2014-10-10T02:49:30Z MrWoohoo quit (Quit: ["Textual IRC Client: www.textualapp.com"]) 2014-10-10T02:49:41Z beach: pjb: Are you talking to me? Do you mean SICL? If so, yes. 2014-10-10T02:49:55Z pjb: Yes, yes, thanks. :-) 2014-10-10T02:49:58Z beach: pjb: In ../Code/Cleavir/Environment/ 2014-10-10T02:51:04Z Dasyatid1 joined #lisp 2014-10-10T02:53:55Z pjb: In hostile.lisp, (not (eq symbol (macroexpand-1 symbol))) shouldn't it use the environment passed to variable-info to macroexpand the symbol? (What's the degree of "hostility" of this example?) 2014-10-10T02:54:22Z beach: The hostility means that it can not do things like that. 2014-10-10T02:54:45Z emma joined #lisp 2014-10-10T02:54:57Z ekinmur quit 2014-10-10T02:55:12Z pjb: Yes. Similarly, the test boundp should not take precedence on looking the symbol in the environment, since it could be a lexical variable. A strange example, which seems to do all that shouldn't be done :-) 2014-10-10T02:55:28Z beach: It means the implementation did not supply the necessary methods to create local environment augmentations, so we have to provide our own, but we can not expect our augmentations to work with the native macro expander. 2014-10-10T02:56:15Z beach: pjb: I don't remember that particular code. 2014-10-10T02:56:53Z pjb: I guess it was some experiment. It's the only example, and others will probably have to be written. 2014-10-10T02:57:03Z pjb: some real examples, I mean. 2014-10-10T02:57:13Z beach: Yes, the hostile environment is not representative. 2014-10-10T02:57:47Z beach: The minimal compiler (in the sense of "minimal compilation") and the AST generator are better examples. 2014-10-10T02:59:25Z KarlDscc quit (Remote host closed the connection) 2014-10-10T03:00:26Z beach: A cute thing to do would be to provide methods on the augmentation functions for SBCL. 2014-10-10T03:00:47Z pjb: I can't justify to myself why name and types are not attributes of the abstract class variable-info. And why constant-variable doesn't have type? 2014-10-10T03:01:56Z beach: pjb: You may be right, but let me see if I can explain my thinking... 2014-10-10T03:02:08Z pjb: Obsolete comment: ;;; VARIABLE-INFO. / (defgeneric function-info (environment function-name)) 2014-10-10T03:02:08Z pjb: 2014-10-10T03:02:35Z pjb: Well, it's more a question of OO style. More fundamental is why constant-variable doesn't have type? 2014-10-10T03:02:36Z beach: pjb: Constant variable doesn't have a type, because the type is the TYPE-OF the value. 2014-10-10T03:03:28Z beach: pjb: I am not quite awake yet, but variable-info ought to contain a type. 2014-10-10T03:03:35Z beach: Oh, no. 2014-10-10T03:03:49Z beach: Because of constant variable not having a type. 2014-10-10T03:04:18Z beach: But I suppose it could be in the abstract class and initialized to TYPE-OF of the value. 2014-10-10T03:04:24Z beach: Is that what you are suggesting? 2014-10-10T03:04:36Z pjb: "variable name" is not in the glossary, but there is: "non-constant variable n. a variable that is not a constant variable." So if declare type was specified to take a non-constant variable, they would have said so. 2014-10-10T03:05:00Z beach: Hmm. OK. 2014-10-10T03:05:03Z pjb: I'm suggesting that (defconstant +e+ 42) (declaim (type integer +e+)) is valid (in any order). 2014-10-10T03:05:06Z beach: Like I said, you may be right. 2014-10-10T03:05:19Z beach: OK. Taking notes! Thanks. 2014-10-10T03:05:21Z pjb: and that (declaim (type character +e+)) (defconstant +e+ 42) should signal a type error. 2014-10-10T03:05:31Z drmeister: General question: Are enabling floating point exceptions essential to the proper operation of numerical functions? 2014-10-10T03:05:44Z pjb: Not in CL, AFAIK. 2014-10-10T03:06:01Z pjb: drmeister: however, you might still want to do that because of NANs if you haev a IEEE implementation. 2014-10-10T03:06:35Z pjb: drmeister: also, for your kind of applications, you probably will want to be able to handle IEEE fp exceptions from lisp. Some CL imlpementations allow to do it. 2014-10-10T03:06:49Z drmeister: Ok. 2014-10-10T03:07:28Z beach: pjb: Fixed the incorrect comment. Thanks! 2014-10-10T03:07:37Z drmeister: I'm filling in some missing CL functionality from some remaining ECL Common Lisp source files that I haven't incorporated yet. 2014-10-10T03:08:04Z pjb: beach: how do we ignore functions? 2014-10-10T03:08:20Z pjb: Oh, there's (declare (ignore (function f))). Didn't know that. 2014-10-10T03:09:20Z drmeister: ECL does some messing about with floating point exceptions during bootstrapping that I can't quite figure out. 2014-10-10T03:09:48Z nydel: does "ecl" refer to the ancient ecl, emacs common lisp, or embedded common lisp? 2014-10-10T03:09:52Z drmeister: "When in doubt - comment it out"! 2014-10-10T03:10:01Z pjb: nydel: emacs-cl is the former. 2014-10-10T03:10:04Z drmeister: Embedded Common Lisp 2014-10-10T03:10:15Z nydel: tyty 2014-10-10T03:10:21Z beach: What is "Emacs Common Lisp"? 2014-10-10T03:10:33Z nell quit (Ping timeout: 258 seconds) 2014-10-10T03:11:01Z zRecursive: seems a CL impl developed using EL :) 2014-10-10T03:11:01Z beach: nydel: tyty? 2014-10-10T03:11:26Z nydel: beach: oh i suppose emacs lisp is never called that. my mistake. i was just hoping someone was talking about that old e lisp that i could never find much information on but was fascinated by. 2014-10-10T03:11:46Z nydel: beach: thank you, twice, abreviated. is what i meant by tyty. 2014-10-10T03:11:50Z beach: nydel: M-x define-mode-abbrevtytyThank you, thank you! 2014-10-10T03:12:15Z pjb: beach: perhaps we could add an expander-like attribute to special-operators, for code walking. That is, the code walker could use that slot to perform walking specific stuff. For an interpreter, this function would interpret the special operator form. For a compiler, it would generate the corresponding code, etc. 2014-10-10T03:12:48Z beach: pjb: Yeah, sounds like a good idea. I would have to work out the details, though. 2014-10-10T03:12:50Z pjb: it could be done in a subclass specific to the minimal compiler/code walker. 2014-10-10T03:13:01Z beach: Sure, that sounds good. 2014-10-10T03:13:33Z nydel: beach: i meant tyty. it means something different from what you suggest. i think you mean "speak properly and with intention" and while i'm quite very sure i'm doing the later, i'll be sure to work on the former when you're around, if that would please? 2014-10-10T03:15:27Z beach: pjb: I would have to think about the advantages of that solution compared to providing a method on the appropriate generic function, specialized to the name of the special operator (EQL). 2014-10-10T03:15:49Z beach: pjb: But I am pretty sure you are a faster thinker than I am, so maybe you can figure it out. 2014-10-10T03:15:55Z pjb: Yes, it seems a non warranted extension. 2014-10-10T03:17:40Z beach: pjb: I am already pleased that you haven't found any fundamental design flaw in the Cleavir environments, even after having looked at the code for several minutes. :) 2014-10-10T03:17:59Z nydel: does anyone recall what was named/called the elisp that existed before emacs lisp? having trouble looking it up. 2014-10-10T03:18:28Z Harag joined #lisp 2014-10-10T03:18:32Z beach: nydel: Emacs Lisp was created specifically for GNU Emacs. 2014-10-10T03:18:37Z Bicyclidine: never heard of it. did it have distinguishing characteristics? 2014-10-10T03:18:48Z pjb: beach: I'm a little puzzled. What's the relationship of the classes lexical-variable and lexical-variable-info (in augmentation-functions and in query)? 2014-10-10T03:18:51Z beach: nydel: Multics Emacs was written in Multics MacLisp. 2014-10-10T03:19:26Z nydel: beach: That's right bud, but there was a dialect of Lisp at some point called ELisp that existed before GNU. 2014-10-10T03:19:27Z beach: pjb: LEXICAL-VARIABLE is an environment used to augment existing environment. 2014-10-10T03:19:31Z martinjungblut joined #lisp 2014-10-10T03:20:13Z beach: pjb: The -INFO classes are used for query; they summarize everything knows by combining several augmentation classes and the global environment. 2014-10-10T03:20:30Z pjb: nydel: cf. https://www.gnu.org/gnu/rms-lisp.html 2014-10-10T03:20:31Z beach: nydel: Oh, OK. Never heard of it. 2014-10-10T03:21:21Z pjb: beach: you mean that those classes are instanciated as response to query functions? All those classes are only API classes, and the implementation of environments is opaque? 2014-10-10T03:21:43Z beach: Yes. 2014-10-10T03:21:54Z beach: [If I understand what you are saying] 2014-10-10T03:23:10Z Harag quit (Client Quit) 2014-10-10T03:23:23Z beach: pjb: The augmentation classes are Cleavir specific, and are only used for implementations that do not provide their own, or that do not need to provide their own. 2014-10-10T03:24:06Z beach: pjb: So for instance, SICL will use the Cleavir-specific augmentation classes, because it is a new implementation, so it does not have to fit the Cleavir functions into an existing framework. 2014-10-10T03:25:17Z Harag joined #lisp 2014-10-10T03:26:12Z beach: pjb: An existing implementation such as SBCL or Clasp where the maintainer does not want to ditch the existing code will provide augmentation classes of its own, methods on ADD-LEXICAL-VARIABLE etc. that instantiate those classes, and methods on VARIABLE-INFO that take those classes as arguments. 2014-10-10T03:27:08Z pjb: Right. But I have a problem here. Augmentation functions don't return a new environment. 2014-10-10T03:27:20Z simulacrum quit (Remote host closed the connection) 2014-10-10T03:27:21Z beach: They do. Or they should. 2014-10-10T03:27:39Z beach: Pretty sure they do, because I use that fact in both the minimal compiler and in the AST generator. 2014-10-10T03:28:25Z pjb: They return new instances of the classes defined in default-augmentation-functions.lisp. Those classes only inherit from entry which doesn't inherit from anything. 2014-10-10T03:28:55Z pjb: Perhaps the methods default-augmentation-functions.lisp don't return the right result? 2014-10-10T03:29:06Z beach: pjb: That's just the default. ENTRY should really be called LOCALLY-AUGMENTED-ENVIRONMENT. 2014-10-10T03:30:17Z pjb: I see, but it is not reflected in the class structure. I think it should or the default-augmentation-functions is not code for machine consumption. 2014-10-10T03:31:02Z beach: I guess I am not awake yet, because I don't understand. 2014-10-10T03:32:14Z pjb: No, it's me. I was puzzled, because no environment class is defined. So entry is as good an environment. 2014-10-10T03:32:28Z vinleod quit (Quit: ["Textual IRC Client: www.textualapp.com"]) 2014-10-10T03:32:29Z beach: Whew! 2014-10-10T03:32:31Z theos makes a cup of strong cappuccino for beach 2014-10-10T03:33:05Z pjb: This makes me wonder why you defined classes and default methods at all? Perhaps it would be better to just define the generic functions, even for the accessors, instead of using defclass. 2014-10-10T03:33:30Z huza quit (Ping timeout: 244 seconds) 2014-10-10T03:33:40Z beach: The default augmentation classes you mean? 2014-10-10T03:33:44Z pjb: Yes. 2014-10-10T03:34:18Z beach: Because I want to provide assistance to creators of new implementations who don't have any opinion about the representation of locally-augmented environments. 2014-10-10T03:34:27Z beach: They might as well use the one I provide. 2014-10-10T03:34:35Z pjb: And the classes in query.lisp too. Just defining the reader generic functions should do. 2014-10-10T03:34:41Z beach: I am one such creator (of SICL) and I intend to use the default provided by Cleavir. 2014-10-10T03:34:49Z pjb: ok. 2014-10-10T03:35:52Z beach: pjb: It helps to think of Cleavir as a toolbox for creators of Common Lisp systems. I try to supply as many tools as possible. 2014-10-10T03:36:15Z beach: ... both for maintainers of existing implementations and for creators of new ones. 2014-10-10T03:36:16Z sheilong quit (Quit: Konversation terminated!) 2014-10-10T03:36:23Z pjb: So in conclusion, I approve :-) Re: pjb: I created a bunch of generic functions that are similar in spirit to the CLtL2 environment functions, but they create standard objects instead of returning multiple values, and they are generic, so each implementation can (has to) provide methods on those functions for their particular environments. 2014-10-10T03:36:23Z pjb: 2014-10-10T03:37:10Z beach: pjb: Thanks. Your opinion is very important to me, because you have been giving these things some thought already. 2014-10-10T03:37:35Z pjb: beach: however, this should be documented, whether subclasses of these classes are expected, or whether it is only required to provide objects for which those generic functions are defined (which is called duck-typing). 2014-10-10T03:37:47Z beach: Definitely! 2014-10-10T03:38:11Z pjb: You seem to be doing duck-typing, and this is not necessarily expected in CLOS. 2014-10-10T03:38:17Z _xinix_ quit (Ping timeout: 260 seconds) 2014-10-10T03:38:28Z beach: Of environments? Yes, sure. 2014-10-10T03:39:00Z kanru` joined #lisp 2014-10-10T03:39:03Z beach: pjb: I only came up with this solution recently, so I haven't had time to document it fully. I think there is some (perhaps wrong) material in the Cleavir documentation already, but it is definitely not complete. 2014-10-10T03:39:24Z pjb: And for the default methods returning subclasses of entry. I guess you define your own classes in the rest of SICL. 2014-10-10T03:39:48Z banjara joined #lisp 2014-10-10T03:39:53Z beach: Maybe so, yes. I may need some more info in there. 2014-10-10T03:40:18Z beach: I should add more to the default classes though, such as SOURCE-LOCATION perhaps. 2014-10-10T03:41:05Z beach: Normally, for SICL I think I can just create subclasses of the existing classes. 2014-10-10T03:41:14Z beach: ... or that was my thinking. 2014-10-10T03:42:48Z beach: So, again, an interesting experiment would be to provide classes and methods that will work for SBCL. 2014-10-10T03:43:20Z nug700 joined #lisp 2014-10-10T03:43:24Z beach: But I won't take the time right now, because it looked complicated, and I have more urgent things to do. 2014-10-10T03:43:31Z knosys quit (Quit: WeeChat 0.4.2) 2014-10-10T03:46:25Z erikc joined #lisp 2014-10-10T03:51:49Z _xinix_ joined #lisp 2014-10-10T03:53:23Z erikc quit (Quit: erikc) 2014-10-10T03:57:48Z resttime quit (Quit: resttime) 2014-10-10T03:58:11Z rme quit (Quit: rme) 2014-10-10T03:58:11Z rme quit (Quit: rme) 2014-10-10T03:58:51Z zacts quit (Ping timeout: 272 seconds) 2014-10-10T03:59:19Z zacts joined #lisp 2014-10-10T04:03:39Z gingerale joined #lisp 2014-10-10T04:03:57Z zacts quit (Ping timeout: 260 seconds) 2014-10-10T04:05:04Z zacts joined #lisp 2014-10-10T04:08:28Z kcj joined #lisp 2014-10-10T04:09:30Z zacts quit (Ping timeout: 246 seconds) 2014-10-10T04:09:38Z zeitue joined #lisp 2014-10-10T04:10:47Z zacts joined #lisp 2014-10-10T04:12:33Z erikc joined #lisp 2014-10-10T04:13:46Z kjeldahl quit (Read error: Connection reset by peer) 2014-10-10T04:14:12Z kristof joined #lisp 2014-10-10T04:14:22Z kristof is now known as Clarice 2014-10-10T04:14:43Z kjeldahl joined #lisp 2014-10-10T04:15:15Z erikc quit (Client Quit) 2014-10-10T04:17:01Z zacts quit (Ping timeout: 260 seconds) 2014-10-10T04:17:44Z zacts joined #lisp 2014-10-10T04:22:04Z zacts quit (Ping timeout: 244 seconds) 2014-10-10T04:22:22Z svetlyak40wt joined #lisp 2014-10-10T04:22:39Z MrWoohoo joined #lisp 2014-10-10T04:23:30Z zacts joined #lisp 2014-10-10T04:28:15Z zacts quit (Ping timeout: 272 seconds) 2014-10-10T04:29:00Z zacts joined #lisp 2014-10-10T04:29:04Z zacts quit (Changing host) 2014-10-10T04:29:05Z zacts joined #lisp 2014-10-10T04:29:47Z erikc joined #lisp 2014-10-10T04:29:47Z erikc quit (Client Quit) 2014-10-10T04:31:09Z MoALTz_ joined #lisp 2014-10-10T04:32:39Z MoALTz__ joined #lisp 2014-10-10T04:33:05Z svetlyak40wt quit (Remote host closed the connection) 2014-10-10T04:33:49Z zacts quit (Ping timeout: 260 seconds) 2014-10-10T04:33:52Z edgar-rft joined #lisp 2014-10-10T04:34:05Z boogie joined #lisp 2014-10-10T04:34:12Z MoALTz quit (Ping timeout: 260 seconds) 2014-10-10T04:34:42Z zacts joined #lisp 2014-10-10T04:35:09Z MoALTz joined #lisp 2014-10-10T04:35:45Z MoALTz_ quit (Ping timeout: 246 seconds) 2014-10-10T04:37:14Z MoALTz__ quit (Ping timeout: 250 seconds) 2014-10-10T04:39:22Z zacts quit (Ping timeout: 245 seconds) 2014-10-10T04:40:08Z beach: pjb: I started documenting the environments by writing comments. Later, I will put more information in the Cleavir documentation. 2014-10-10T04:40:30Z zacts joined #lisp 2014-10-10T04:41:49Z zarkone: hello. could you please advice me where to find sources of `nth` in sbcl. Im very new to CLisp.. 2014-10-10T04:42:17Z nug700 quit (Read error: Connection reset by peer) 2014-10-10T04:42:40Z pjb: zarkone: if you use slime, just typing (nth M-. in a lisp buffer should jump to the source of nth. 2014-10-10T04:42:56Z zarkone: thanks! 2014-10-10T04:43:02Z pjb: zacts: sbcl is not clisp. clisp is not sbcl. Both clisp and sbcl are CL implementations. 2014-10-10T04:44:09Z Bicyclidine quit (Quit: leaving) 2014-10-10T04:44:25Z zarkone: yes, i know, if this was for me =) 2014-10-10T04:45:02Z pjb: Yes. Completion interupted. :-) 2014-10-10T04:45:21Z zacts quit (Ping timeout: 272 seconds) 2014-10-10T04:45:38Z zarkone: when i said "CLisp" i meant "Common Lisp" 2014-10-10T04:46:15Z zacts joined #lisp 2014-10-10T04:46:19Z pjb: Since there's an implementation named clisp, we restrict ourselves to CL or Common Lisp. 2014-10-10T04:46:34Z nug700 joined #lisp 2014-10-10T04:47:22Z Clarice quit (Ping timeout: 245 seconds) 2014-10-10T04:47:55Z zarkone: ok, i will remember that 2014-10-10T04:48:56Z gingerale quit (Ping timeout: 244 seconds) 2014-10-10T04:49:26Z ltbarcly joined #lisp 2014-10-10T04:50:48Z zacts quit (Ping timeout: 246 seconds) 2014-10-10T04:51:26Z nug700 quit (Read error: Connection reset by peer) 2014-10-10T04:51:51Z beach: zarkone: It is probably similar to this: (defun nth (n list) (loop for rest = list then (cdr rest) repeat n finally (return (car rest)))) 2014-10-10T04:52:02Z ltbarcly quit (Client Quit) 2014-10-10T04:52:06Z zacts joined #lisp 2014-10-10T04:52:37Z Bike: it's actually (car (nthcdr ...)), how simple! 2014-10-10T04:52:54Z malglim quit (Ping timeout: 258 seconds) 2014-10-10T04:53:03Z beach: That's "similar" :) 2014-10-10T04:54:30Z nug700 joined #lisp 2014-10-10T04:54:57Z malglim joined #lisp 2014-10-10T04:56:25Z ltbarcly joined #lisp 2014-10-10T04:56:44Z zacts quit (Ping timeout: 250 seconds) 2014-10-10T04:57:58Z zacts joined #lisp 2014-10-10T04:58:11Z ltbarcly quit (Client Quit) 2014-10-10T04:59:00Z protist: I think nth also works on vectors 2014-10-10T04:59:00Z minion: protist, memo from pjb: you can make commercial products with GPL code. Only you must keep distributing the code of your commercial products! 2014-10-10T04:59:09Z protist: therefore it must be more complicated than that 2014-10-10T04:59:10Z pjb: protist: nope. elt does. 2014-10-10T04:59:20Z pjb: nope. this is the actual source. 2014-10-10T04:59:22Z protist: pjb: just saw your memo :p 2014-10-10T04:59:29Z protist: pjb: ah cool 2014-10-10T04:59:43Z pjb: :-) 2014-10-10T04:59:46Z ltbarcly joined #lisp 2014-10-10T05:00:12Z defaultxr quit (Ping timeout: 245 seconds) 2014-10-10T05:02:22Z zacts quit (Ping timeout: 250 seconds) 2014-10-10T05:03:42Z zacts joined #lisp 2014-10-10T05:03:46Z dddd joined #lisp 2014-10-10T05:04:22Z oleo__ quit (Quit: Verlassend) 2014-10-10T05:06:31Z defaultxr joined #lisp 2014-10-10T05:07:19Z zarkone: in nthcdr definition there are calls of fast-nthcdr 2014-10-10T05:07:31Z grumio2 quit (Ping timeout: 272 seconds) 2014-10-10T05:08:16Z Bike: probably an assembly routine. 2014-10-10T05:08:48Z zarkone: seems like it's more optimized than just (loop .. 2014-10-10T05:09:17Z Bike: wait, fast-nthcdr is just an flet, duh 2014-10-10T05:09:34Z Bike: those do's are pretty much like loop 2014-10-10T05:09:45Z tajjada joined #lisp 2014-10-10T05:10:02Z innertracks joined #lisp 2014-10-10T05:10:34Z jasom: It never ceases to amaze me how do manages to make iteration more ugly than mapfoo and a lambda. 2014-10-10T05:11:08Z Bike: sbcl defines loop pretty late and also the code is old, is i think the explanation 2014-10-10T05:11:35Z zarkone: oh, flet -- it's like let but about functions 2014-10-10T05:12:22Z karswell` joined #lisp 2014-10-10T05:12:32Z pjb: zarkone: the optimization only comes from the (declare (type index …)), which makes sbcl generate faster code for the loop in fast-nthcdr. 2014-10-10T05:13:14Z karswell quit (Read error: Connection reset by peer) 2014-10-10T05:13:15Z pjb: The loop in the alternate branch of typedef works on bignums, while in fast-nthcdr, it's on fixnums. 2014-10-10T05:13:32Z zarkone: pjb: yes, i see now.. 2014-10-10T05:14:19Z Bike: oh, the slow version checks for circular lists. i wonder who wanted that. 2014-10-10T05:14:24Z innertracks quit (Client Quit) 2014-10-10T05:15:34Z beach: Bike: LOOP presumably exists in the host implementation used for bootstrapping, so that's not a valid reason not to use it. 2014-10-10T05:15:58Z Bike: but sbcl uses that implementation to compile itself, and then that sbcl to compile the product sbcl 2014-10-10T05:16:30Z zarkone: What this is all about. I implemented A* and try to make it faster. I've created 2 versions, single- and multithread. Now experiment with datatypes. Hash tables was very fast, but i crash the heap when i try to load > 1M of elements.. 2014-10-10T05:17:06Z pjb: Bike: that's a nice optimization, checking for circular lists. 2014-10-10T05:17:13Z zarkone: now i want to try arrays 2014-10-10T05:17:26Z Bike: pjb: it's neat, i just wonder why somebody bothered 2014-10-10T05:17:26Z beach: Bike: I think I managed to convince myself that no such dumbing down is necessary. 2014-10-10T05:18:17Z boogie quit (Remote host closed the connection) 2014-10-10T05:19:38Z pjb: Bike: a reason is needed, why sbcl images are so big :-) 2014-10-10T05:19:51Z ananna quit (Ping timeout: 246 seconds) 2014-10-10T05:19:51Z banjara quit (Quit: Leaving.) 2014-10-10T05:19:53Z zarkone: by the way, may be any advices about hash-tables? I shouldn't use it for big data or smth wrong in my code..? 2014-10-10T05:20:21Z pjb: zarkone: you shouldn't use it for small data, but for big data, it's ok in general. 2014-10-10T05:20:41Z pjb: With less than 5-35 entries, it's faster to use a-lists or p-lists. 2014-10-10T05:20:50Z zarkone: for small i can use :key props form lists 2014-10-10T05:21:02Z pjb: yes. 2014-10-10T05:21:16Z zarkone: so in this case it's something wrong with the code.. 2014-10-10T05:21:45Z pjb: But unless you have a lot of small dictionaries, using hashtables for them won't hurt either. 2014-10-10T05:21:52Z pjb: Do you have any real problem? 2014-10-10T05:22:45Z nalssee joined #lisp 2014-10-10T05:23:59Z dddd quit (Ping timeout: 272 seconds) 2014-10-10T05:25:03Z dddd joined #lisp 2014-10-10T05:28:16Z zarkone: yes -- i load the field representation into hash-table. The hash is like "el-i-j", and the item itself is an list with :keys. While I read file by symbols, I create corresponding items, create hash and setf item to it. So, when i try to store ~ 1 000 000 vertexex in hash, SBCL downs with "cannot allocate" 2014-10-10T05:29:16Z zarkone: i replace hashtable with list and the problem disappered but the process became very slow 2014-10-10T05:29:56Z zarkone: obviously because of access-time difference between this structures 2014-10-10T05:33:21Z pranavrc joined #lisp 2014-10-10T05:35:59Z harish joined #lisp 2014-10-10T05:36:56Z zarkone: I have ~ 6,5G free memory, 2014-10-10T05:37:24Z ltbarcly quit (Quit: Computer has gone to sleep.) 2014-10-10T05:37:59Z Bike: might need to let sbcl get at more of it with --dynamic-space-size 2014-10-10T05:39:49Z tajjada quit (Ping timeout: 272 seconds) 2014-10-10T05:40:14Z zarkone: thanks, i'll try it 2014-10-10T05:40:25Z zRecursive: After `git pull`, i used to run `git show 'HEAD@{1}..HEAD'` to see the newest update on git server. Now what is the corresponding svn command to `git show` ? 2014-10-10T05:45:55Z dddd quit (Read error: Connection reset by peer) 2014-10-10T05:54:16Z beach: pjb: Thinking about it some more, I am now convinced that an existing implementation could simply modify its implementation of MACROEXPAND-1 so that it can handle the Cleavir default augmentation classes as well. That might be less work if this solution is acceptable for that implementation. 2014-10-10T05:57:15Z hiroakip joined #lisp 2014-10-10T05:58:19Z Bike: well, more than likely that just means genericizing macro-function, which might not be a bad idea 2014-10-10T05:58:22Z defaultxr quit (Quit: gnight) 2014-10-10T05:59:02Z beach: Good idea! 2014-10-10T05:59:43Z Bike: i don't think i've seen a macroexpand-1 implementation that wouldn't work with other environments just from that 2014-10-10T05:59:59Z beach: I understand. 2014-10-10T06:00:04Z Bike: well, symbol macros would be weird. 2014-10-10T06:01:16Z beach: Hmm, so the environment parameter is optional in MACRO-FUNCTION. One would have to trampoline to GENERIC-MACRO-FUNCTION where the environment is a required parameter. 2014-10-10T06:01:32Z beach: But I can supply GENERIC-MACRO-FUNCTION in Cleavir no problem. 2014-10-10T06:01:42Z Shinmera joined #lisp 2014-10-10T06:01:51Z Bike: that doesn't seem to be asking too much 2014-10-10T06:01:59Z beach: You are right. 2014-10-10T06:02:01Z Bike: i guess you'd need, like, symbol-macro-expansion or something, though 2014-10-10T06:02:15Z beach: Maybe so. Not too much to ask either. 2014-10-10T06:03:11Z Bike: it'd be less trivial, e.g. look at how macroexpand-1 on symbols works in sbcl 2014-10-10T06:03:45Z beach: Hmm. OK, I'll give it some more thought. 2014-10-10T06:03:53Z Bike: well, that entire block looks just like the macro-function implementation actually. 2014-10-10T06:04:07Z beach: Their problem, then! :) 2014-10-10T06:04:24Z Bike: i should probably install ccl so i can do weird source comparisons on multiple implementations 2014-10-10T06:04:31Z harish quit (Ping timeout: 272 seconds) 2014-10-10T06:04:45Z beach: That would be great. 2014-10-10T06:04:49Z beach: I have a very good feeling about the appropriateness of this environment design. 2014-10-10T06:05:12Z zarkone quit (Ping timeout: 260 seconds) 2014-10-10T06:05:20Z Bike: how do you handle environment augmentation, again? 2014-10-10T06:05:58Z beach: Each augmentation is a small class instance with a NEXT slot that refers to the next less specific environment. 2014-10-10T06:06:15Z beach: The last link in the chain is an object that represents the global environment. 2014-10-10T06:06:29Z svetlyak40wt joined #lisp 2014-10-10T06:06:31Z Bike: i mean the functional interface, the thing that takes an environment and returns an augmented one. 2014-10-10T06:06:36Z beach: Implementations that don't have first-class global environments supply a dummy object to dispatch on. 2014-10-10T06:06:48Z mishoo joined #lisp 2014-10-10T06:07:11Z beach: A bunch of functions: add-lexical-variable, add-variable-type, add-optimize, etc. 2014-10-10T06:07:18Z Bike: ah. 2014-10-10T06:07:27Z Bike: so not like cltl2's keyword orgy. 2014-10-10T06:07:37Z Bike: add-optimize? not add-declaration? 2014-10-10T06:07:40Z beach: Right, because I need to do dispatch on them. 2014-10-10T06:07:55Z beach: Yes, one function for each type of declarations. 2014-10-10T06:08:01Z beach: Excluding SPECIAL of course :) 2014-10-10T06:08:18Z Bike: what about user decls? 2014-10-10T06:08:25Z beach: I haven't thought about those yet. 2014-10-10T06:08:39Z beach: I suppose implementations will have to supply their own augmentation classes for those. 2014-10-10T06:09:25Z beach: I am not smart enough to think about everything right away! :) 2014-10-10T06:09:38Z Bike: well, i mean, you can do (declaim (declaration biketastic)) (locally (declare (biketastic 39 hut hut hut)) ...), not that nobody ever does 2014-10-10T06:09:51Z beach: Yeah I know. 2014-10-10T06:09:57Z beach: I just haven't thought about it yet. 2014-10-10T06:10:00Z Bike: so it's not just the implementation i mean. 2014-10-10T06:10:11Z beach: Oh, I see what you mean. 2014-10-10T06:10:12Z beach: Sure. 2014-10-10T06:10:24Z Bike: hm, you can't locally declare declaration, how boring 2014-10-10T06:10:30Z beach: I guess I have to add an ADD-DECLARATION function. 2014-10-10T06:10:44Z beach: Oh that simplifies things a bit. 2014-10-10T06:10:54Z nipra quit (Quit: Leaving.) 2014-10-10T06:10:57Z nalssee quit (Quit: This computer has gone to sleep) 2014-10-10T06:13:18Z pjb: beach: it may be interesting for a specific compiler to use add-{specific-declaration} methods, but for a generic code walker, or to keep note of user declarations (declaim (declaration thingy)), it might be needed to have an add-declaration operation. 2014-10-10T06:13:46Z beach: Yeah. 2014-10-10T06:13:48Z pjb: So, ok. 2014-10-10T06:13:54Z pjb: Time is strange. 2014-10-10T06:14:09Z Harag quit (Ping timeout: 260 seconds) 2014-10-10T06:15:45Z c107 quit (Remote host closed the connection) 2014-10-10T06:17:24Z beach: pjb: Strange? How so? 2014-10-10T06:19:31Z rtra` joined #lisp 2014-10-10T06:22:00Z hiroakip quit (Ping timeout: 260 seconds) 2014-10-10T06:22:56Z rtra quit (Ping timeout: 260 seconds) 2014-10-10T06:23:09Z Dasyatid1 left #lisp 2014-10-10T06:23:16Z rtra` is now known as rtra 2014-10-10T06:25:36Z prxq joined #lisp 2014-10-10T06:26:23Z beach: OK, time to get some work done. I'll read the logs if anyone has any more remarks. 2014-10-10T06:26:28Z beach left #lisp 2014-10-10T06:28:47Z bgs100 quit (Quit: bgs100) 2014-10-10T06:30:36Z Ven joined #lisp 2014-10-10T06:30:37Z cneira joined #lisp 2014-10-10T06:30:52Z phao quit (Ping timeout: 240 seconds) 2014-10-10T06:32:05Z pt1 joined #lisp 2014-10-10T06:34:40Z moore33 joined #lisp 2014-10-10T06:37:19Z Shinmera quit (Quit: しつれいしなければならないんです。) 2014-10-10T06:38:43Z _xinix_ quit (Ping timeout: 272 seconds) 2014-10-10T06:39:22Z moore33 quit (Client Quit) 2014-10-10T06:41:24Z vaporatorius joined #lisp 2014-10-10T06:42:22Z jusss joined #lisp 2014-10-10T06:43:01Z stassats joined #lisp 2014-10-10T06:44:41Z MoALTz quit (Quit: Leaving) 2014-10-10T06:46:15Z angavrilov joined #lisp 2014-10-10T06:47:05Z freaksken joined #lisp 2014-10-10T06:47:48Z pavelpenev joined #lisp 2014-10-10T06:49:26Z nalssee joined #lisp 2014-10-10T06:50:29Z mvilleneuve joined #lisp 2014-10-10T06:51:06Z banjara joined #lisp 2014-10-10T06:52:03Z vaporatorius quit (Quit: Leaving) 2014-10-10T06:53:31Z ananna joined #lisp 2014-10-10T06:53:40Z nalssee quit (Client Quit) 2014-10-10T06:56:42Z zarkone joined #lisp 2014-10-10T06:57:02Z nalssee joined #lisp 2014-10-10T07:01:13Z nalssee quit (Client Quit) 2014-10-10T07:01:51Z Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) 2014-10-10T07:02:45Z zRecursive left #lisp 2014-10-10T07:02:47Z nalssee joined #lisp 2014-10-10T07:07:06Z loke quit (Remote host closed the connection) 2014-10-10T07:07:07Z DGASAU quit (Read error: Connection reset by peer) 2014-10-10T07:08:04Z DGASAU joined #lisp 2014-10-10T07:10:50Z oD_C joined #lisp 2014-10-10T07:11:17Z marvimias is now known as perdent 2014-10-10T07:11:31Z perdent is now known as marvimias 2014-10-10T07:13:35Z munksgaard joined #lisp 2014-10-10T07:16:20Z oudeis joined #lisp 2014-10-10T07:17:19Z nalssee quit (Quit: This computer has gone to sleep) 2014-10-10T07:18:17Z oD_C: hello there all 2014-10-10T07:18:37Z psy_ quit (Ping timeout: 272 seconds) 2014-10-10T07:18:57Z oD_C: i have zero knowledge of any programming language. (i am 29 years old). Do i think right when i think that i am going to start learning from zero with lisp? 2014-10-10T07:18:58Z theZacAttack joined #lisp 2014-10-10T07:19:11Z oD_C: i mean do i need a math background in order to achieve this goal 2014-10-10T07:19:29Z stassats: you don't 2014-10-10T07:19:36Z stassats: minion: tell oD_C about gentle 2014-10-10T07:19:36Z minion: oD_C: look at gentle: "Common Lisp: A Gentle Introduction to Symbolic Computation" is a smoother introduction to lisp programming. http://www.cs.cmu.edu/~dst/LispBook/ 2014-10-10T07:20:23Z oD_C: thank you amte i will give it a look 2014-10-10T07:20:26Z theZacAttack quit (Remote host closed the connection) 2014-10-10T07:21:16Z stassats: minion: land of lisp? 2014-10-10T07:21:16Z minion: i like lisp... i'm written in it 2014-10-10T07:21:17Z dim: I think you need a project with a easy goal to start, and lots of time and patience, because there's so many details to learn about along the way, and you need to be prepared to change your mind about how to implement your project often 2014-10-10T07:21:22Z stassats: minion: land-of-lisp? 2014-10-10T07:21:23Z minion: Sorry, I couldn't find anything in the database for ``land-of-lisp''. 2014-10-10T07:21:27Z stassats: minion: LoL? 2014-10-10T07:21:27Z minion: LoL: Land of Lisp - http://landoflisp.com 2014-10-10T07:21:39Z stassats: oD_C: if you like pictures 2014-10-10T07:21:56Z stassats: but that one isn't available for free 2014-10-10T07:22:23Z vinleod joined #lisp 2014-10-10T07:23:24Z Harag joined #lisp 2014-10-10T07:24:11Z oD_C: ok thank you for the information. i will print both books (if i find the second) and i will give it a try. 2014-10-10T07:24:58Z nalssee joined #lisp 2014-10-10T07:25:03Z theZacAttack joined #lisp 2014-10-10T07:28:08Z loke joined #lisp 2014-10-10T07:29:39Z nalssee quit (Client Quit) 2014-10-10T07:31:57Z drdanmaku quit (Quit: Connection closed for inactivity) 2014-10-10T07:32:40Z ehu joined #lisp 2014-10-10T07:33:07Z przl joined #lisp 2014-10-10T07:34:06Z theZacAttack quit (Quit: theZacAttack) 2014-10-10T07:35:32Z nalssee joined #lisp 2014-10-10T07:36:12Z cneira quit (Ping timeout: 258 seconds) 2014-10-10T07:36:46Z robot-beethoven joined #lisp 2014-10-10T07:38:48Z przl quit (Ping timeout: 250 seconds) 2014-10-10T07:39:57Z nalssee quit (Client Quit) 2014-10-10T07:41:23Z stepnem joined #lisp 2014-10-10T07:42:11Z nalssee joined #lisp 2014-10-10T07:44:01Z psy joined #lisp 2014-10-10T07:44:15Z kanru` quit (Remote host closed the connection) 2014-10-10T07:44:24Z fridim_ joined #lisp 2014-10-10T07:46:11Z zarkone: why not to learn lisp as a firest language? just curious, my first was C, and it was very hard to begin with lisp. May be if lisp would first it would be easier to learn another..? never seen the person who starts with lisp or functional language.. 2014-10-10T07:46:31Z zarkone: /s/firest/first 2014-10-10T07:47:51Z xificurC joined #lisp 2014-10-10T07:49:03Z xificurC quit (Client Quit) 2014-10-10T07:49:22Z xificurC joined #lisp 2014-10-10T07:49:44Z pranavrc quit (Ping timeout: 260 seconds) 2014-10-10T07:50:24Z pranavrc joined #lisp 2014-10-10T07:50:24Z pranavrc quit (Changing host) 2014-10-10T07:50:24Z pranavrc joined #lisp 2014-10-10T07:53:04Z Ven joined #lisp 2014-10-10T07:53:07Z martinjungblut quit (Quit: Leaving.) 2014-10-10T07:53:14Z jusss quit (Quit: ERC Version 5.2 (IRC client for Emacs)) 2014-10-10T07:53:21Z martinjungblut joined #lisp 2014-10-10T07:53:24Z Zhivago: Probably the lack of resources aimed at complete novices. 2014-10-10T07:56:58Z robot-beethoven quit (Quit: ERC Version 5.3 (IRC client for Emacs)) 2014-10-10T07:57:14Z banjara quit (Quit: Leaving.) 2014-10-10T07:59:25Z pjb: zarkone: definitely, lisp should be taught as a first language, to children. 2014-10-10T07:59:31Z cneira joined #lisp 2014-10-10T08:00:06Z pjb: We need somebody to set up a Raspberry Pi flash for children with a nice lisp setup. 2014-10-10T08:00:39Z dim: something like a lisp machine setup where you can hack your way around all in lisp without even quitting the lisp env? 2014-10-10T08:00:55Z pjb: Eventually. 2014-10-10T08:01:19Z dim: and with Smalltalk like ideas where you play with letters, numbers, images and sounds maybe 2014-10-10T08:02:32Z Cymew joined #lisp 2014-10-10T08:03:29Z pjb: But perhaps not as sophisticated as alice. Just having an API to a bitmap window would be enough. Programming to the terminal is a nice way to learn. LightSpeed Pascal was like that, you could open a graphic window and draw stuff, but otherwise you had a stdin/stdout interface. 2014-10-10T08:04:17Z cneira quit (Ping timeout: 260 seconds) 2014-10-10T08:04:33Z dim: a termio thingy might be ok (ncurses maybe) yes, I had great fun moving that strange square sprite around the 80x24 screen back in the day 2014-10-10T08:04:52Z dim: we could have a selection of the classic bsdgames in lisp then 2014-10-10T08:04:57Z dim: and offer kids to hack them away 2014-10-10T08:05:07Z dim: snake, worm and hangman were a hit here 2014-10-10T08:05:20Z dim: https://github.com/dimitri/kids that's what I began with 2014-10-10T08:05:58Z pt1_ joined #lisp 2014-10-10T08:09:15Z pt1 quit (Ping timeout: 246 seconds) 2014-10-10T08:10:00Z fzappa joined #lisp 2014-10-10T08:11:13Z nha joined #lisp 2014-10-10T08:11:56Z pavelpenev quit (Ping timeout: 256 seconds) 2014-10-10T08:17:34Z kushal joined #lisp 2014-10-10T08:19:45Z munksgaard quit (Ping timeout: 246 seconds) 2014-10-10T08:22:09Z prxq quit (Remote host closed the connection) 2014-10-10T08:26:23Z zacharias joined #lisp 2014-10-10T08:31:55Z przl joined #lisp 2014-10-10T08:31:55Z moore33 joined #lisp 2014-10-10T08:33:32Z wasamasa: dim: didn't know you can use keywords instead of symbols in LOOP 2014-10-10T08:35:22Z banjara joined #lisp 2014-10-10T08:35:31Z simulacrum joined #lisp 2014-10-10T08:36:08Z dim: I like it for the syntax highlighting mainly 2014-10-10T08:36:36Z H4ns: awful 2014-10-10T08:37:34Z aldo joined #lisp 2014-10-10T08:37:37Z H4ns: you're really making syntax choices because your editor is stupid? shameful. :) 2014-10-10T08:39:50Z dim: it seems I can't feel shame on that behavior, so I still do it 2014-10-10T08:40:08Z banjara quit (Ping timeout: 260 seconds) 2014-10-10T08:40:28Z wasamasa: and elisp doesn't support that variant 2014-10-10T08:40:30Z wasamasa: what a shame 2014-10-10T08:40:47Z nalssee quit (Quit: This computer has gone to sleep) 2014-10-10T08:40:58Z dim: elisp lacks packages or any other way to have namespaces 2014-10-10T08:41:04Z Posterdati quit (Ping timeout: 260 seconds) 2014-10-10T08:41:04Z pt1_ quit (Remote host closed the connection) 2014-10-10T08:42:06Z nalssee joined #lisp 2014-10-10T08:42:28Z jkaye joined #lisp 2014-10-10T08:44:04Z wasamasa: I hope this will change if it should evolve towards CL 2014-10-10T08:44:06Z oudeis quit (Quit: This computer has gone to sleep) 2014-10-10T08:44:33Z |3b|: wasamasa: keywords are symbols, and CL:LOOP only looks at symbols' name not package 2014-10-10T08:46:50Z jkaye quit (Ping timeout: 250 seconds) 2014-10-10T08:47:15Z stassats: CL is a strange destination for evolution 2014-10-10T08:47:30Z |3b|: so it is more that CL:LOOP doesn't support packages than that there is an extra syntax it does support 2014-10-10T08:47:35Z moore33: Could be worse. 2014-10-10T08:47:36Z minion: moore33, memo from pjb: subclasses are useful mostly if you have to define distinct methods for them. This is rarely done at run-time. Therefore if you have new name/value pairs at run-time, but not new methods, put them in a plist. 2014-10-10T08:47:45Z pjb: wasamasa: it's better to use keywords, because some of those symbols are not exported from CL, therefore you might actually use them before using a package that exports symbols with the same name, which would give an error! 2014-10-10T08:48:22Z stassats: that's not a valid reason 2014-10-10T08:48:32Z loke: pjb: That's only a problem if you USE packages 2014-10-10T08:48:41Z |3b|: or just don't :use packages, particularly not after you have started using the destination package 2014-10-10T08:49:06Z stassats: do you name your variables in the keyword package? 2014-10-10T08:49:19Z pjb: loke: this is something you may do often in the REPL. 2014-10-10T08:49:42Z pjb: stassats: variables are not named WHILE or FOR usually. 2014-10-10T08:49:42Z |3b|: pjb: it isn't something i do often in the repl 2014-10-10T08:49:43Z loke: pjb: Some people, pehaps. Not me though. 2014-10-10T08:49:54Z nalssee quit (Quit: This computer has gone to sleep) 2014-10-10T08:49:58Z stassats: pjb: what makes WHILE and FOR so special? 2014-10-10T08:50:06Z pjb: That's why I said "better" and "might". Fuck! Read the words! 2014-10-10T08:50:20Z pjb: stassats: they are commonly used as macro names. 2014-10-10T08:50:43Z |3b|: pjb: then you will get conflicts from the next package you USE anyway that also defines them :p 2014-10-10T08:51:20Z stassats: there is no problem even with uncommon names, so this whole thing is made up 2014-10-10T08:51:34Z |3b|: then you will accidentally redefine one of the functions you don't use from one of those 2 packages, breaking some 3rd package that wants to use that function 2014-10-10T08:51:44Z pjb: Telling newbies to use keywords instead of symbols for loop keywords will save them half a day. 2014-10-10T08:51:53Z stassats: it won't 2014-10-10T08:52:04Z moore33: No, it will take years off their lives :) 2014-10-10T08:52:07Z pjb: It would have saved me half a day, if I had been told. 2014-10-10T08:52:20Z stassats: you're an exception, we already know that 2014-10-10T08:52:24Z moore33: And corrupt their mortal souls. 2014-10-10T08:52:29Z loke: Better to tell newbies not to use :USE instead. That's save them more time. 2014-10-10T08:52:42Z litian_ joined #lisp 2014-10-10T08:52:54Z |3b|: yeah, i've seen things silently break from USEing packages... conflict on USE is nice and obvious 2014-10-10T08:53:13Z litian_ quit (Client Quit) 2014-10-10T08:53:18Z Grue`: when I use LOOP in macros I use keywords, and in normal code I use normal symbols. very idiosyncratic, I know 2014-10-10T08:53:22Z stassats: even with :use, who runs loop inside a new package before calling use-package? don't do that and expect good results 2014-10-10T08:53:40Z stassats: or any code, for that matter 2014-10-10T08:54:05Z stassats: Grue`: i really hope you are not expanding your macros into LOOP 2014-10-10T08:54:18Z stassats: if it's just in the loop body, then it is very peculiar indeed 2014-10-10T08:54:19Z loke: stassats: True. I never even thought about that. This so-called problem is even less of a problem :-) 2014-10-10T08:55:17Z Grue`: I probably do expand some macros into LOOP, you don't expect me to use DO, do you? ;) 2014-10-10T08:55:24Z stassats: DO is worse 2014-10-10T08:55:41Z loke: stassats: The most bizarre argument for using keywords that I saw was someone on c.l.l explaining that it could be disasterous if someone changes the name of a symbol that you have :USE'd... 2014-10-10T08:55:45Z stassats: Grue`: the problem with LOOP is LOOP-FINISH, and the problem with DO is BLOCK NIL 2014-10-10T08:56:05Z loke: like foo:while being actually changed so that it has a different name, and you :USE foo 2014-10-10T08:56:15Z stassats: loke: i don't have any knives at home either, or i may cut myself 2014-10-10T08:56:24Z loke: stassats: :-) 2014-10-10T08:56:31Z Grue`: block nil is useful when it's documented 2014-10-10T08:56:53Z stassats: if it's a do-like macro, sure 2014-10-10T08:56:57Z moore33: stassats: What's the problem with LOOP-FINISH? 2014-10-10T08:57:22Z pt1 joined #lisp 2014-10-10T08:57:31Z stassats: moore33: (loop ... (macro-expanding-into-loop ... (when x (loop-finish)))) what do you expect to happen? 2014-10-10T08:57:46Z Grue`: most of my macros are for top-level use, so there's no problem with scope 2014-10-10T08:57:49Z Posterdati joined #lisp 2014-10-10T08:57:54Z stassats: loke: but, how can you change the name of a symbol? 2014-10-10T08:58:08Z moore33 has to look up LOOP-FINISH :) 2014-10-10T08:58:15Z munksgaard joined #lisp 2014-10-10T08:58:31Z stassats: if you want to make iterating macros, you need to use TAGBODY 2014-10-10T08:58:53Z stassats: loke: i know _i_ can, but that isn't really defined 2014-10-10T08:59:06Z loke: stassats: I know. But that was the argument :-) 2014-10-10T09:00:09Z moore33: stassats: Heh. Got it. 2014-10-10T09:00:53Z stassats: loop also has block nil, but you can (loop named (gensym)), but that still leaves loop-finish, so, there's no point in doing that 2014-10-10T09:01:27Z nha quit (Ping timeout: 245 seconds) 2014-10-10T09:02:52Z splittist: we just need to extend loop with loop-named-finish (: 2014-10-10T09:06:14Z splittist: but, yeah. 2014-10-10T09:06:54Z Grue`: (symbol-macrolet ((and '(incf i))) (loop for i from 1 to 10 if (< i 5) do (print i) and collect i)) I'm disappointed this doesn't work 2014-10-10T09:08:20Z pjb: dim: Here is what I have (dating from 2006); it's for clisp 2.49: git clone http://git.informatimago.com/public/ipcode 2014-10-10T09:08:53Z pjb: ./lisp.sh RET (load "ipl-ex-graphic.lisp") RET (IN-PACKAGE "IPL-USER") RET (ROTATING-SQUARE) RET etc. 2014-10-10T09:11:51Z dim: cool, will have a look someday ;-) 2014-10-10T09:12:01Z dim: (did clone and skim through, no time for more now) 2014-10-10T09:12:11Z pjb: I agree, that asdf and defpackage and file based system construction is important, for reproductibility and distribution, but we still have a REPL, and one has to know how to use it. Therefore use-package, import, export, etc should not be forgotten or forbidden. At the REPL. 2014-10-10T09:12:22Z pjb: dim: sure :-) 2014-10-10T09:12:23Z EvW joined #lisp 2014-10-10T09:13:03Z stassats: loke: and that's actually only a problem with macros, since they store expanded loop code, otherwise when you recompile existing it reads the right symbol, but we already established that expanding into loop is a bad idea. another case is EVAL and COMPILE 2014-10-10T09:13:12Z stassats: so, that reason is pretty contrived 2014-10-10T09:13:16Z dim: it's just hard to keep track of multiple packages all at the REPL 2014-10-10T09:13:31Z pjb: Grue`: this is not surprising, loop parsing occurs before macroexpanding. 2014-10-10T09:13:42Z pjb: dim: not if you have the right commands. 2014-10-10T09:13:51Z pjb: 1- set up your repl to display the current package in the prompt. 2014-10-10T09:14:02Z pjb: 2- use pushp and popp. 2014-10-10T09:15:17Z pjb: (ql:quickload :com.informatimago.common-lisp.interactive) (use-package :com.informatimago.common-lisp.interactive.interactive) 2014-10-10T09:16:11Z pjb: 3- also mkupack 2014-10-10T09:17:18Z pjb: Grue`: in any case, in ccl it works perfectly: it prints from 1 to 4 and returns (1 2 3 4). 2014-10-10T09:17:47Z svetlyak40wt quit (Remote host closed the connection) 2014-10-10T09:19:45Z svetlyak40wt joined #lisp 2014-10-10T09:21:06Z pt1 quit (Remote host closed the connection) 2014-10-10T09:21:33Z pt1 joined #lisp 2014-10-10T09:23:32Z Harag quit (Ping timeout: 260 seconds) 2014-10-10T09:23:45Z chu joined #lisp 2014-10-10T09:24:20Z nug700 quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/) 2014-10-10T09:25:41Z pavelpenev joined #lisp 2014-10-10T09:28:35Z vinleod quit (Quit: ["Textual IRC Client: www.textualapp.com"]) 2014-10-10T09:30:41Z nipra joined #lisp 2014-10-10T09:30:42Z Harag joined #lisp 2014-10-10T09:31:07Z hitecnologys joined #lisp 2014-10-10T09:32:02Z gabriel_laddel joined #lisp 2014-10-10T09:32:28Z svetlyak40wt quit (Remote host closed the connection) 2014-10-10T09:33:05Z nalssee joined #lisp 2014-10-10T09:33:26Z sz0 joined #lisp 2014-10-10T09:33:44Z chu quit (Quit: ERC Version 5.3 (IRC client for Emacs)) 2014-10-10T09:36:53Z nalssee quit (Client Quit) 2014-10-10T09:36:56Z banjara joined #lisp 2014-10-10T09:37:20Z aldo quit (Ping timeout: 258 seconds) 2014-10-10T09:41:18Z heurist joined #lisp 2014-10-10T09:41:45Z banjara quit (Ping timeout: 272 seconds) 2014-10-10T09:42:27Z dim: I prefer using emacs and slime and not fiddling too much at the repl when it's about code organisation 2014-10-10T09:45:18Z zarkone quit (Ping timeout: 272 seconds) 2014-10-10T09:52:08Z zickzackv joined #lisp 2014-10-10T09:54:55Z KarlDscc joined #lisp 2014-10-10T09:58:08Z hardenedapple joined #lisp 2014-10-10T10:01:21Z kuzy000_ joined #lisp 2014-10-10T10:03:21Z xificurC quit (Quit: WeeChat 1.0) 2014-10-10T10:08:42Z sz0 quit (Quit: My computer has gone to sleep. ZZZzzz…) 2014-10-10T10:09:24Z thawes joined #lisp 2014-10-10T10:10:48Z kuzy000_ quit (Read error: Connection reset by peer) 2014-10-10T10:11:59Z KarlDscc quit (Remote host closed the connection) 2014-10-10T10:12:33Z kuzy000_ joined #lisp 2014-10-10T10:14:22Z svetlyak40wt joined #lisp 2014-10-10T10:14:35Z dim: latest QL release had me fix bugs in my tooling, but debian upload are in progress now 2014-10-10T10:15:10Z banjara joined #lisp 2014-10-10T10:16:39Z kuzy000_ quit (Read error: Connection reset by peer) 2014-10-10T10:17:43Z kuzy000_ joined #lisp 2014-10-10T10:19:12Z karupa is now known as zz_karupa 2014-10-10T10:19:27Z macdice`` quit (Ping timeout: 245 seconds) 2014-10-10T10:19:37Z banjara quit (Ping timeout: 260 seconds) 2014-10-10T10:22:57Z theos quit (Ping timeout: 258 seconds) 2014-10-10T10:26:17Z MutSbeta quit (Remote host closed the connection) 2014-10-10T10:32:33Z Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) 2014-10-10T10:32:35Z nalssee joined #lisp 2014-10-10T10:35:00Z svetlyak40wt quit (Read error: Connection timed out) 2014-10-10T10:35:29Z sunwukong joined #lisp 2014-10-10T10:35:40Z theos joined #lisp 2014-10-10T10:35:52Z svetlyak40wt joined #lisp 2014-10-10T10:36:08Z gabriel_laddel quit (Ping timeout: 244 seconds) 2014-10-10T10:39:17Z oudeis joined #lisp 2014-10-10T10:40:22Z urandom__ joined #lisp 2014-10-10T10:40:27Z ndrei quit (Ping timeout: 246 seconds) 2014-10-10T10:41:54Z sunwukong quit (Quit: Bye) 2014-10-10T10:44:52Z fmu quit (Ping timeout: 240 seconds) 2014-10-10T10:46:12Z fmu joined #lisp 2014-10-10T10:49:44Z ndrei joined #lisp 2014-10-10T10:53:09Z aldo joined #lisp 2014-10-10T10:56:21Z varjag_ joined #lisp 2014-10-10T11:00:52Z aldo quit (Ping timeout: 240 seconds) 2014-10-10T11:01:06Z aldo joined #lisp 2014-10-10T11:02:39Z kcj quit (Read error: Connection reset by peer) 2014-10-10T11:05:47Z nipra quit (Quit: Leaving.) 2014-10-10T11:08:52Z farhaven quit (Ping timeout: 240 seconds) 2014-10-10T11:13:09Z Shinmera joined #lisp 2014-10-10T11:13:22Z gabriel_laddel joined #lisp 2014-10-10T11:15:41Z knosys joined #lisp 2014-10-10T11:15:53Z banjara joined #lisp 2014-10-10T11:16:00Z przl quit (Ping timeout: 260 seconds) 2014-10-10T11:20:27Z banjara quit (Ping timeout: 258 seconds) 2014-10-10T11:21:40Z ivan4th joined #lisp 2014-10-10T11:22:32Z Harag quit (Ping timeout: 260 seconds) 2014-10-10T11:25:58Z nalssee quit (Quit: Leaving) 2014-10-10T11:28:54Z ehu quit 2014-10-10T11:31:54Z Ven joined #lisp 2014-10-10T11:32:40Z Svetlana quit (Killed (sinisalo.freenode.net (Nickname regained by services))) 2014-10-10T11:32:43Z ggole joined #lisp 2014-10-10T11:32:58Z Sveta joined #lisp 2014-10-10T11:34:10Z nipra joined #lisp 2014-10-10T11:34:42Z simulacrum quit (Quit: No Ping reply in 180 seconds.) 2014-10-10T11:36:08Z simulacrum joined #lisp 2014-10-10T11:39:46Z pavelpenev quit (Remote host closed the connection) 2014-10-10T11:42:40Z svetlyak40wt quit (Remote host closed the connection) 2014-10-10T11:45:01Z moore33: Gotta love reddit: http://vintage-digital.com/hefner/misc/lisp-programmers.jpg 2014-10-10T11:45:20Z oudeis quit (Quit: This computer has gone to sleep) 2014-10-10T11:45:54Z mingvs quit (Ping timeout: 246 seconds) 2014-10-10T11:46:33Z fe[nl]ix: moore33: what do you mean reddit ? 2014-10-10T11:47:39Z stassats: it's old and it's from ahefner 2014-10-10T11:47:47Z mingvs joined #lisp 2014-10-10T11:48:13Z pavelpenev joined #lisp 2014-10-10T11:48:38Z moore33: fe[nl]ix: I mean that's where I saw it. 2014-10-10T11:48:40Z martinjungblut: you know, a bunch of those are actually accurate, and I'm not even a lisp programmer 2014-10-10T11:50:31Z moore33: I like the Forth-using Unibomber. 2014-10-10T11:53:38Z stassats` joined #lisp 2014-10-10T11:55:40Z stassats quit (Ping timeout: 260 seconds) 2014-10-10T11:55:54Z oudeis joined #lisp 2014-10-10T11:57:59Z splittist: the factor image is ambiguous. Probably fittingly so. 2014-10-10T11:59:33Z KarlDscc joined #lisp 2014-10-10T11:59:40Z gabriel_laddel quit (Ping timeout: 250 seconds) 2014-10-10T12:01:26Z pavelpenev quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.) 2014-10-10T12:02:16Z Beetny quit (Ping timeout: 250 seconds) 2014-10-10T12:02:46Z svetlyak40wt joined #lisp 2014-10-10T12:03:19Z knobo_ joined #lisp 2014-10-10T12:03:57Z knobo_: I remember there was a shell with lisp syntax. What was the name? 2014-10-10T12:04:37Z knobo_: scsh 2014-10-10T12:04:40Z knobo_: :) 2014-10-10T12:06:04Z knobo_: maybe.. 2014-10-10T12:06:08Z Hache_ joined #lisp 2014-10-10T12:07:23Z knobo_: is scsh guile? 2014-10-10T12:07:31Z eudoxia joined #lisp 2014-10-10T12:09:10Z Harag joined #lisp 2014-10-10T12:11:05Z yrk quit (Remote host closed the connection) 2014-10-10T12:12:50Z gravicappa joined #lisp 2014-10-10T12:16:12Z Harag quit (Ping timeout: 260 seconds) 2014-10-10T12:16:39Z banjara joined #lisp 2014-10-10T12:18:57Z ehu joined #lisp 2014-10-10T12:21:25Z banjara quit (Ping timeout: 260 seconds) 2014-10-10T12:21:48Z stepnem quit (Ping timeout: 260 seconds) 2014-10-10T12:22:40Z prxq joined #lisp 2014-10-10T12:22:54Z Poenikatu joined #lisp 2014-10-10T12:22:54Z Poenikatu quit (Changing host) 2014-10-10T12:22:54Z Poenikatu joined #lisp 2014-10-10T12:23:24Z davazp joined #lisp 2014-10-10T12:23:56Z stepnem joined #lisp 2014-10-10T12:26:02Z oudeis quit (Quit: This computer has gone to sleep) 2014-10-10T12:27:48Z dim: if you don't insist on CL, M-x eshell might be a good fit 2014-10-10T12:29:25Z chu joined #lisp 2014-10-10T12:30:00Z Sveta is now known as Svetlana 2014-10-10T12:30:03Z paul0` joined #lisp 2014-10-10T12:30:59Z przl joined #lisp 2014-10-10T12:32:46Z Harag joined #lisp 2014-10-10T12:33:09Z paul0 quit (Ping timeout: 246 seconds) 2014-10-10T12:34:11Z reb`` quit (Read error: Connection reset by peer) 2014-10-10T12:34:28Z reb`` joined #lisp 2014-10-10T12:39:58Z fantazo joined #lisp 2014-10-10T12:41:03Z kcj joined #lisp 2014-10-10T12:42:57Z knob joined #lisp 2014-10-10T12:44:12Z yrk joined #lisp 2014-10-10T12:44:38Z yrk quit (Changing host) 2014-10-10T12:44:38Z yrk joined #lisp 2014-10-10T12:46:33Z chu quit (Quit: ERC Version 5.3 (IRC client for Emacs)) 2014-10-10T12:48:20Z pavelpenev joined #lisp 2014-10-10T12:50:07Z LiamH joined #lisp 2014-10-10T12:50:50Z oleo joined #lisp 2014-10-10T12:52:53Z BitPuffin joined #lisp 2014-10-10T12:53:32Z madrik joined #lisp 2014-10-10T12:54:13Z wbooze joined #lisp 2014-10-10T12:55:26Z pgomes joined #lisp 2014-10-10T12:58:27Z eudoxia quit (Quit: Lost terminal) 2014-10-10T12:58:57Z kcj quit (Read error: Connection reset by peer) 2014-10-10T13:07:20Z munksgaard quit (Ping timeout: 272 seconds) 2014-10-10T13:08:28Z TDog quit (Ping timeout: 260 seconds) 2014-10-10T13:10:07Z drdanmaku joined #lisp 2014-10-10T13:11:54Z munksgaard joined #lisp 2014-10-10T13:12:17Z sz0 joined #lisp 2014-10-10T13:14:07Z przl quit (Quit: leaving) 2014-10-10T13:17:40Z blackwolf joined #lisp 2014-10-10T13:18:14Z banjara joined #lisp 2014-10-10T13:18:15Z paul0` is now known as paul0 2014-10-10T13:19:36Z blackwolf: anyone have a persistence solution suggestion? elephant appears inactive, and I'd rather not take the time to roll my own for a prototype. 2014-10-10T13:19:54Z fe[nl]ix: blackwolf: postgres 2014-10-10T13:19:54Z stassats`: postmodern 2014-10-10T13:20:46Z thawes quit (Quit: Konversation terminated!) 2014-10-10T13:20:47Z DGASAU quit (Read error: Connection reset by peer) 2014-10-10T13:21:12Z blackwolf: yes, have hacked with postmodern a bit. any other alternatives? 2014-10-10T13:21:22Z stassats`: write your own 2014-10-10T13:21:23Z pjb: knobo_: scsh is indeed a scheme. You may also use clash, which is clisp with a convenient reader macro to run unix commands. 2014-10-10T13:21:35Z DGASAU joined #lisp 2014-10-10T13:21:50Z thawes joined #lisp 2014-10-10T13:21:57Z pjb: knobo_: http://www.clisp.org/clash.html 2014-10-10T13:22:24Z pjb: blackwolf: http://www.cliki.net/site/search?query=database has 11 pages of results… 2014-10-10T13:22:40Z blackwolf: stassats`: prefer not to for a quick-and-dirty prototype 2014-10-10T13:22:47Z banjara quit (Ping timeout: 245 seconds) 2014-10-10T13:23:03Z pjb: blackwolf: http://www.cliki.net/Database lists tens of RDBMS, ORMs and OODBMS. 2014-10-10T13:23:22Z blackwolf: pjb: ya, was hoping to winnow that list down a bit :) 2014-10-10T13:23:42Z pjb: blackwolf: and said that, realize that you are the only lisper who has to use a database this month, therefore ALL those projects are dead and unmaintained right now! 2014-10-10T13:24:47Z pjb: For example, currently beach is writting a compiler, I'm working on patchwork, xach is working for ccl customers and on quicklisp, dim is working on a SQL dumper, etc. Nobody currently needs or works on a persistent solution. 2014-10-10T13:25:17Z p_l: perec worked fine for postgres quite recently 2014-10-10T13:25:21Z pjb: That's the beauty of working with a rare programming language: you have the opportunity to be a significant member of the "community" :-) 2014-10-10T13:25:22Z moore33: no code written for a month != unmaintained 2014-10-10T13:25:26Z p_l: postmodern too 2014-10-10T13:25:35Z p_l: moore33: jokes ;) 2014-10-10T13:25:38Z moore33: pjb:Ha! 2014-10-10T13:25:50Z pjb: I wouldn't be surprised if postmodern hadn't have any commit for the last two years at least. 2014-10-10T13:25:58Z pjb: for example. 2014-10-10T13:26:20Z p_l: i sometimes run CL code that is pushing two decades without updates 2014-10-10T13:26:27Z pjb: indeed. 2014-10-10T13:26:51Z moore33: pjb: Easily checked... last commit 9 days ago. 2014-10-10T13:26:52Z pjb: for example, according to /topic, the only hot things are sbcl 1.2.4 (as always), yason and drakma. 2014-10-10T13:27:06Z pjb: glad to be wrong :-) 2014-10-10T13:27:21Z blackwolf: 2 years is a bit better that 5 ... will see what the level of effort we'll be to get something hacked up with postmodern. thx .... 2014-10-10T13:27:26Z pjb: somebody should update the /topic :-) 2014-10-10T13:27:44Z p_l: the stuff that would normally warrant update to postmodern I used to handle by few classes abd macros in perec (new sql types) 2014-10-10T13:27:59Z pjb: Now, anyways my point is that it is a good thing to have so many stable libraries. 2014-10-10T13:28:15Z zeitue quit (Quit: Leaving) 2014-10-10T13:28:16Z moore33: Not that I'm a user of postmodern, but I've always liked PostgreSQL. 2014-10-10T13:28:21Z stassats`: except that the lock of activity doesn't indicate stability or fitness 2014-10-10T13:28:24Z pjb: It's really reposing and it let you make some progress, when you don't have to follow moving targets all the time. 2014-10-10T13:28:28Z stassats`: s/lock/lack/ 2014-10-10T13:28:29Z wooden quit (Ping timeout: 272 seconds) 2014-10-10T13:28:47Z dim: I found that Postmodern works beautifully 2014-10-10T13:28:51Z pjb: Neither does constant maintainance. 2014-10-10T13:29:07Z stassats`: but pjb just keeps spinning the same story, oh well 2014-10-10T13:29:49Z pjb: Yes, because people keep coming with bullshit like "elephant appears inactive". 2014-10-10T13:29:52Z Cymew quit (Ping timeout: 245 seconds) 2014-10-10T13:30:12Z dim: take Lyons 2014-10-10T13:30:21Z farhaven joined #lisp 2014-10-10T13:30:22Z Mon_Ouie joined #lisp 2014-10-10T13:30:22Z Mon_Ouie quit (Changing host) 2014-10-10T13:30:22Z Mon_Ouie joined #lisp 2014-10-10T13:30:22Z stassats`: pjb: so, you just want to pile some more on? 2014-10-10T13:30:30Z dim: apparently they sleep 22h a day *average* when free, more in captivilty 2014-10-10T13:30:37Z dim: captivity even 2014-10-10T13:30:40Z pjb: :-) 2014-10-10T13:31:04Z Zhivago: Do cities sleep? 2014-10-10T13:31:12Z pjb: Well, in captivity they're fed, so they've got even less hunting to do. 2014-10-10T13:31:47Z Zhivago: They can probably get in a good 26 hours. 2014-10-10T13:32:09Z phao joined #lisp 2014-10-10T13:32:10Z pjb: Zhivago: more or less. Eg. in Paris, the metro is closed between 2:00 and 6:00, so there is not much activity then. 2014-10-10T13:32:12Z dim: Zhivago: that's "paresseux", that I don't know the english name of yet 2014-10-10T13:32:29Z Zhivago: However, the statistics seem to be the reverse. 2014-10-10T13:32:33Z dim: http://en.wikipedia.org/wiki/Sloth apparently 2014-10-10T13:32:43Z Zhivago: In captivity they only get about 10-15 hours, presumably because of noisy children. 2014-10-10T13:32:45Z fe[nl]ix: minion: we're not talking about one month here 2014-10-10T13:32:45Z minion: you speak nonsense 2014-10-10T13:32:46Z dim: they are 23h30 a day average IIRC 2014-10-10T13:32:53Z fe[nl]ix: moore33: we're not talking about one month here 2014-10-10T13:33:39Z stassats`: we're not talking any concrete facts either, i would prefer somebody who actually uses the thing in question to chime in 2014-10-10T13:34:03Z stassats`: instead of some philosophical discussion on what it means to be unmaintained 2014-10-10T13:34:38Z moore33: fe[nl]ix: Talk to minion instead ;) 2014-10-10T13:34:43Z fe[nl]ix: a concrete fact is that is hasn't been developed in 5 years 2014-10-10T13:35:12Z pjb: A more concrete fact would be whether it's featureful and documented and usable. 2014-10-10T13:35:17Z fe[nl]ix: and when I used to maintain lisp packages for Gentoo I couldn't even compile it 2014-10-10T13:35:42Z pjb: A good datapoint indeed would be if there is an ASDF file for it. 2014-10-10T13:35:45Z fe[nl]ix: every other sbcl release tended to break it 2014-10-10T13:35:51Z oudeis joined #lisp 2014-10-10T13:36:00Z stassats`: probably has to do with structure dumping 2014-10-10T13:36:03Z simulacrum quit (Remote host closed the connection) 2014-10-10T13:36:11Z pjb: (not that its absence would be a big impediment, if the project is cleanly structured, writing an asdf file is easy). 2014-10-10T13:36:33Z pjb: So a good question is how conforming its code would be. 2014-10-10T13:37:49Z pjb: is<->would be 2014-10-10T13:38:56Z pjb: My previous parenthesis means a good question would be whether it's _maintainable_, rather than maintained. 2014-10-10T13:39:17Z rme joined #lisp 2014-10-10T13:39:37Z Svetlana quit (Quit: sleep) 2014-10-10T13:39:57Z pjb: (that said, I'm yet to find a lisp project that's not maintainable, at least compared to what I've seen in C++ or java…) 2014-10-10T13:40:57Z Xach: elephant is very difficult to build for quicklisp and nobody is improving the situation. the primary author has moved on. 2014-10-10T13:41:29Z stassats`: maybe he just forgot 2014-10-10T13:41:33Z loke_erc joined #lisp 2014-10-10T13:41:34Z pavelpenev quit (Quit: No Ping reply in 180 seconds.) 2014-10-10T13:41:51Z dim: hey Xach! do you think https://github.com/dimitri/pgloader/issues/119 could be a Quicklisp bug, could it be QL naming files that way? 2014-10-10T13:43:12Z pavelpenev joined #lisp 2014-10-10T13:43:26Z Xach: dim: looks like it's a ql bug, yes. it should not try to construct a file like that. 2014-10-10T13:43:56Z madrik quit (Ping timeout: 260 seconds) 2014-10-10T13:44:08Z dim: ok, so I won't open an issue on CCL I think, because it's unclear that their behavior should be different reall 2014-10-10T13:44:10Z dim: +y 2014-10-10T13:44:16Z Xach: I have to figure out a good way to deal with slashed system names in general. 2014-10-10T13:44:36Z Xach: dim: right. 2014-10-10T13:44:49Z dim: Xach: if you need more information or tests, feel free to chime in right in the pgloader issue, if you prefer, I can open a new issue on Quicklisp referencing that one too 2014-10-10T13:45:29Z dim: on azerty keyboards we often used to use § as a separator 2014-10-10T13:45:44Z skbierm joined #lisp 2014-10-10T13:45:46Z dim: nowadays I'm on qwerty but we have unicode, so § still work ;-) 2014-10-10T13:46:06Z dim: but then you need to know about the filesystem names encoding maybe 2014-10-10T13:46:14Z pjb: I must have used § once, 30 years ago, on Macintosh :-) 2014-10-10T13:47:10Z dim: it's a single byte in iso-8859-1 IIRC, I don't know in unicode 2014-10-10T13:47:25Z dim: buffer code: #xC2 #xA7 2014-10-10T13:47:25Z pjb: unicode is a superset of iso-8859-1. 2014-10-10T13:47:27Z Zhivago: Well, as unicode isn't an encoding ... 2014-10-10T13:47:37Z dim: well, I meant utf-8 here sorry 2014-10-10T13:47:41Z pjb: so it's the same code point. 2014-10-10T13:47:45Z thawes_ joined #lisp 2014-10-10T13:48:00Z zz_karupa quit (Ping timeout: 250 seconds) 2014-10-10T13:48:30Z thawes quit (Ping timeout: 272 seconds) 2014-10-10T13:49:10Z pranavrc quit 2014-10-10T13:51:01Z Xach: dim: a backtrace would save me some research time 2014-10-10T13:51:21Z dim: sure, pgloader --debug will provide that, asking on the issue 2014-10-10T13:51:49Z blackwolf quit (Quit: Bye) 2014-10-10T13:51:57Z dim: mmm. 2014-10-10T13:52:26Z dim: backtrace of creating the systems or of something else? 2014-10-10T13:52:39Z dim: to repro, git clone pgloader, make CL=ccl (with ccl 1.10) 2014-10-10T13:52:56Z dim: I'm not sure what to hack in the process to get the backtrace you need 2014-10-10T13:52:59Z Xach: Actually, I have it narrowed down now, thanks. 2014-10-10T13:54:15Z dim: oh, just though /tmp/build.log should do it 2014-10-10T13:54:23Z dim: awesome, thank you Xach 2014-10-10T13:56:05Z sz0 quit (Ping timeout: 244 seconds) 2014-10-10T13:56:12Z loke_erc quit (Remote host closed the connection) 2014-10-10T13:57:12Z zz_karupa joined #lisp 2014-10-10T13:57:48Z zz_karupa is now known as karupa 2014-10-10T13:58:38Z billstclair quit (Ping timeout: 272 seconds) 2014-10-10T13:59:45Z billstclair joined #lisp 2014-10-10T14:01:32Z pgomes quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client) 2014-10-10T14:02:16Z Xach: https://github.com/quicklisp/quicklisp-client/issues/107 2014-10-10T14:02:27Z Xach: 107 is my lucky number, so maybe I'll be able to fix it and release a new client soon! 2014-10-10T14:04:10Z lambda` joined #lisp 2014-10-10T14:05:41Z gravicappa quit (Remote host closed the connection) 2014-10-10T14:06:13Z Xach scratches head 2014-10-10T14:06:52Z BitPuffin quit (Ping timeout: 240 seconds) 2014-10-10T14:09:39Z Xach: dim: a backtrace would help, after all. instead of using :GO in the debugger. 2014-10-10T14:10:15Z bobbysmith0071 quit (Quit: Leaving.) 2014-10-10T14:12:11Z Xach: Ah, no, found it. 2014-10-10T14:13:20Z thawes_ quit (Ping timeout: 260 seconds) 2014-10-10T14:14:39Z ozzloy quit (Ping timeout: 246 seconds) 2014-10-10T14:14:39Z tessier quit (Ping timeout: 246 seconds) 2014-10-10T14:16:08Z tessier joined #lisp 2014-10-10T14:16:37Z ozzloy joined #lisp 2014-10-10T14:17:26Z yeticry quit (Read error: Connection reset by peer) 2014-10-10T14:17:51Z simulacrum joined #lisp 2014-10-10T14:19:03Z banjara joined #lisp 2014-10-10T14:19:04Z banjara quit (Changing host) 2014-10-10T14:19:05Z banjara joined #lisp 2014-10-10T14:19:12Z justinmcp quit (Ping timeout: 246 seconds) 2014-10-10T14:19:15Z justinmcp_ joined #lisp 2014-10-10T14:19:47Z finnrobi quit (Ping timeout: 272 seconds) 2014-10-10T14:20:46Z FracV quit (Changing host) 2014-10-10T14:20:46Z FracV joined #lisp 2014-10-10T14:22:40Z lambda` quit (Ping timeout: 260 seconds) 2014-10-10T14:23:14Z rick-monster: ha - I was just trying to fix a bug in xml-emitter when I found someone beat me to it here https://github.com/billitch/xml-emitter/ 2014-10-10T14:23:22Z banjara quit (Ping timeout: 240 seconds) 2014-10-10T14:23:28Z yeticry joined #lisp 2014-10-10T14:23:35Z davazp quit (Remote host closed the connection) 2014-10-10T14:23:39Z Xach: |3b|: someone asked me to add classimp to quicklisip. how do you feel about that? 2014-10-10T14:25:42Z rick-monster: what's the protocol to try and get that fix pushed upstream? contact original maintainer I guess? 2014-10-10T14:25:43Z |3b|: Xach: no opinion 2014-10-10T14:25:55Z Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) 2014-10-10T14:26:00Z Xach: rick-monster: i think that's a good approach 2014-10-10T14:27:22Z didi joined #lisp 2014-10-10T14:28:19Z |3b|: or alternately, adding it is good, but then i'd feel bad about still not having updated it for new cffi :p 2014-10-10T14:28:40Z didi: Can I use a keyword as `eql-specializer-form' in a method declaration? e.g. (defmethod foo ((bar (eql :baz))) ...) 2014-10-10T14:28:52Z Bike: sure 2014-10-10T14:28:55Z didi: Cool. 2014-10-10T14:29:04Z stassats`: why would you assume that you can't? 2014-10-10T14:29:13Z didi: *shrug* 2014-10-10T14:29:16Z didi: They look different. 2014-10-10T14:29:50Z didi: I still have problems identifying the differences between `eq', `eql', `equal' and `equalp'. 2014-10-10T14:30:20Z stassats`: you should write more code 2014-10-10T14:30:26Z didi nods 2014-10-10T14:31:30Z svetlyak40wt quit (Read error: Connection timed out) 2014-10-10T14:31:49Z moore33: didi: Don't forget =, string=, and string-equal... 2014-10-10T14:32:02Z Xach: dwim= 2014-10-10T14:32:06Z Zhivago: EQL is identity, EQ is optimized for a subset of EQL's domain. EQUAL and EQUALP are vague and waffly equivalences. 2014-10-10T14:32:07Z stassats`: dwim 2014-10-10T14:32:16Z svetlyak40wt joined #lisp 2014-10-10T14:32:27Z fzappa quit (Remote host closed the connection) 2014-10-10T14:33:28Z dlowe: also char= and char-equal 2014-10-10T14:33:28Z hiroakip joined #lisp 2014-10-10T14:33:43Z stassats`: and char/= 2014-10-10T14:33:53Z stassats`: who knows more comparison operators! 2014-10-10T14:34:07Z dlowe: the subject is equality 2014-10-10T14:34:32Z stassats`: what's next, brotherhood? 2014-10-10T14:36:18Z protist quit (Quit: Konversation terminated!) 2014-10-10T14:36:29Z Zhivago: Just way until you get to EGAL. 2014-10-10T14:36:37Z Zhivago: er, wait. 2014-10-10T14:42:58Z innertracks joined #lisp 2014-10-10T14:44:01Z svetlyak40wt quit (Remote host closed the connection) 2014-10-10T14:46:37Z bgs100 joined #lisp 2014-10-10T14:46:50Z finnrobi joined #lisp 2014-10-10T14:46:56Z zickzackv quit (Ping timeout: 250 seconds) 2014-10-10T14:49:11Z c107 joined #lisp 2014-10-10T14:49:23Z pavelpenev quit (Quit: No Ping reply in 180 seconds.) 2014-10-10T14:50:55Z pavelpenev joined #lisp 2014-10-10T14:52:38Z hrs joined #lisp 2014-10-10T14:55:53Z wglb quit (Ping timeout: 260 seconds) 2014-10-10T14:56:31Z Ven joined #lisp 2014-10-10T14:58:25Z xyjprc joined #lisp 2014-10-10T14:58:59Z gingerale joined #lisp 2014-10-10T15:01:01Z xyjprc quit (Remote host closed the connection) 2014-10-10T15:04:14Z thawes_ joined #lisp 2014-10-10T15:04:32Z _xinix_ joined #lisp 2014-10-10T15:07:55Z innertracks quit (Ping timeout: 272 seconds) 2014-10-10T15:08:24Z munksgaard quit (Ping timeout: 260 seconds) 2014-10-10T15:10:24Z banjara joined #lisp 2014-10-10T15:10:39Z didi left #lisp 2014-10-10T15:12:02Z banjara quit (Read error: Connection reset by peer) 2014-10-10T15:12:11Z banjara joined #lisp 2014-10-10T15:14:24Z _xinix_ quit (Quit: Leaving) 2014-10-10T15:14:56Z fantazo quit (Ping timeout: 260 seconds) 2014-10-10T15:15:23Z cheater joined #lisp 2014-10-10T15:16:32Z banjara quit (Ping timeout: 245 seconds) 2014-10-10T15:18:52Z thawes_ quit (Ping timeout: 240 seconds) 2014-10-10T15:20:49Z harish_ joined #lisp 2014-10-10T15:25:49Z |3b| supposes i shouldn't fully unroll a 512 point FFT... first it broke sbcl, then it broke my lisp->glsl compiler, finally optimized it enough that they would compile it and now gl drivers won't take it :p 2014-10-10T15:26:04Z moore33: Ouch. 2014-10-10T15:26:26Z |3b|: only 17922 instructions according to GL :/ 2014-10-10T15:28:51Z ndrei quit (Ping timeout: 258 seconds) 2014-10-10T15:32:04Z |3b| wonders how i should break it up instead 2014-10-10T15:34:25Z harish_ quit (Quit: Leaving) 2014-10-10T15:34:54Z harish_ joined #lisp 2014-10-10T15:35:30Z ndrei joined #lisp 2014-10-10T15:36:20Z thawes_ joined #lisp 2014-10-10T15:40:02Z rvirding_: A while back I read an article/blog on lisp which amongst other things stated that a problem with lisp was a lack of common libraries, and that one reason for this is that it is so easy to roll your own so this is what people 2014-10-10T15:40:08Z Krystof: do 2014-10-10T15:40:21Z rvirding_: do, which results in common libraries not being written. Or something like that 2014-10-10T15:41:04Z rvirding_: problem is I cannot remember either name or author of the article :-( 2014-10-10T15:41:12Z munksgaard joined #lisp 2014-10-10T15:41:19Z rvirding_: does it ring a bell in someone? 2014-10-10T15:41:42Z rvirding_: do you feel that the premise is true? 2014-10-10T15:41:53Z RenRenJuan joined #lisp 2014-10-10T15:42:05Z Krystof: no 2014-10-10T15:42:24Z Krystof: it's not that easy to roll your own, and common libraries do get written 2014-10-10T15:42:27Z rvirding_: is it too easy to write your own libraries so that people do that instead of building/using common libraries? 2014-10-10T15:42:28Z Krystof: so I think it's basically completely bogus 2014-10-10T15:42:40Z ggole: Is that the "bipolar lisp programmer" article by any chance? 2014-10-10T15:43:30Z Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…) 2014-10-10T15:43:48Z fe[nl]ix: Krystof: it's not entirely bogus, but also not specific to CL 2014-10-10T15:43:49Z Krystof: jsnell's response to the "bipolar lisp programmer" article was the finest. 2014-10-10T15:44:05Z rvirding_: ggole: that could be it 2014-10-10T15:44:06Z Krystof: (I don't know if it's the article rvirding_ was thinking of) 2014-10-10T15:44:49Z rvirding_: where is that response? 2014-10-10T15:45:10Z Krystof: https://groups.google.com/forum/#!original/comp.lang.lisp/eicqvm3GXiE/3qvWZ0ngjBoJ 2014-10-10T15:45:13Z fe[nl]ix: Krystof: I have to maintain a node.js server at the moment, and I find the node ecosystem much worse 2014-10-10T15:46:46Z mvilleneuve quit (Quit: This computer has gone to sleep) 2014-10-10T15:48:23Z rvirding_: reason I am asking is that I see myself doing just that. Currently not in lisp but in erlang which makes it easy to do concurrent things. 2014-10-10T15:49:03Z rvirding_: So I roll my own small "library" functions, like a parallel map, easier to do it myself rather than find and read the docs 2014-10-10T15:49:23Z pt1 quit (Remote host closed the connection) 2014-10-10T15:49:33Z hrs quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…) 2014-10-10T15:50:32Z rvirding_: one thing which supports this in erlang is the handling of errors in concurrent systems, there are so many ways you could reasonably handle this and it is usually very application specific that it is easier to do it yourself and get it as you wish 2014-10-10T15:51:05Z rvirding_: so I remembered the article and wondered if ppl feel it is true or not 2014-10-10T15:51:26Z rvirding_: of course writing BIG packages is another thing 2014-10-10T15:53:19Z jsnell: I don't think there's any particular truth to "common libraries not being written" 2014-10-10T15:54:23Z Krystof: comp.lang.lisp had its moments 2014-10-10T15:54:25Z jsnell: there might be to them not being used though. a lot of people hate adding dependencies, and those kinds of utility libraries often have divisive style issues 2014-10-10T15:54:35Z dlowe: it was also pre-quicklisp 2014-10-10T15:54:36Z psy quit (Ping timeout: 260 seconds) 2014-10-10T15:54:48Z dlowe: which profoundly and unexpectedly changed the entire landscape there 2014-10-10T15:55:06Z hugod quit (Remote host closed the connection) 2014-10-10T15:56:26Z zacharias quit (Ping timeout: 272 seconds) 2014-10-10T15:56:51Z redline6561: dlowe: +1 2014-10-10T15:57:11Z redline6561: Quicklisp was and is a big deal. 2014-10-10T15:57:33Z fe[nl]ix: dlowe: but it's a real cargo cult, this one 2014-10-10T15:57:42Z rvirding_: sorry for going on there 2014-10-10T15:57:43Z Xach: rvirding_: I think it's pretty easy to roll your own, and harder to make something useful to a small internal group (like your company) and harder to polish something up to be suitable for a wide audience 2014-10-10T15:57:46Z psy joined #lisp 2014-10-10T15:57:59Z fe[nl]ix: years after QL came out, I still hear people furiously avoiding dependencies 2014-10-10T15:58:20Z fe[nl]ix: 'cause "deps are bad" 2014-10-10T15:58:38Z Shinmera: I am scarred by The Java Experience. 2014-10-10T15:58:40Z Xach: rvirding_: i've heard it estimated that there's a 3x difficulty factor at each level 2014-10-10T15:58:42Z dlowe: I avoid dependencies for some things, not others. local-time could use regexes, but it probably shouldn't pull in cl-pcre 2014-10-10T15:59:57Z rvirding_: Xach: I would say that one problem with making it suitable for a wide audience is how you are going to handle all the different ways users want to use it 2014-10-10T16:00:32Z pjb: |3b|: beach wrote a very efficient FFT in CL. 2014-10-10T16:00:42Z Xach: rvirding_: I am not thinking about that as much. I mean more like "write documentation, make sure the interfaces are clean and exported", and things like that 2014-10-10T16:00:48Z rvirding_: do you a big handle-everything case with lots of options to control it? Or do you write a number of simpler ones for different case? 2014-10-10T16:01:07Z Shinmera: rvirding_: Imo the problem I have most of the time with libs is that I don't really understand how I'm supposed to use them. Often there's a lack of a good "introduction" and listing of what it is capable of. 2014-10-10T16:01:43Z rvirding_: I remember in the 80's there was this discussion in the unix world, "cat -v considered harmful" 2014-10-10T16:01:59Z rvirding_: documentation is always difficult 2014-10-10T16:02:04Z |3b|: pjb: and pkhuong wrote a more efficient one from what i understand, neither runs on GPU though (also, not as educational as writing my own, though probably more efficient) 2014-10-10T16:02:16Z Shinmera: rvirding_: http://harmful.cat-v.org/ 2014-10-10T16:04:25Z rvirding_: hehehe the programmer general declares cat -v is harmful 2014-10-10T16:04:52Z munksgaard quit (Ping timeout: 260 seconds) 2014-10-10T16:10:39Z fe[nl]ix: https://i.imgur.com/r2G4LZO.jpg 2014-10-10T16:11:46Z pavelpenev quit (Read error: Connection reset by peer) 2014-10-10T16:11:46Z pavelpenev_ joined #lisp 2014-10-10T16:12:57Z banjara joined #lisp 2014-10-10T16:14:22Z ndrei quit (Ping timeout: 245 seconds) 2014-10-10T16:14:57Z Krystof: "because if it did compilation would be much harder"? 2014-10-10T16:16:02Z fe[nl]ix: damned pragmatism 2014-10-10T16:16:37Z moore33: What I love from that whole bipolar programmer thing is how eager people are to share the life story... yuck. 2014-10-10T16:16:43Z moore33: their life story 2014-10-10T16:17:28Z banjara quit (Ping timeout: 260 seconds) 2014-10-10T16:18:13Z moore33: Also, the reference to http://www.google.com/url?q=http%3A%2F%2Fwww.idlewords.com%2F2005%2F04%2Fdabblers_and_blowhards.htm&sa=D&sntz=1&usg=AFQjCNFeA-Wx3V_FQHNRhaASKC30Zy9FWw, completely awesome. 2014-10-10T16:19:00Z drewc: mmmm ... c.l.l in '06, erlang, Java , node.js .... thank g-d we are discussing past 'problems' within CL that other langs still have, otherwise I would be wondering :) 2014-10-10T16:19:21Z eudoxia joined #lisp 2014-10-10T16:19:35Z eudoxia: why do people *still* insist CL doesn't have libraries 2014-10-10T16:19:47Z Ven joined #lisp 2014-10-10T16:19:55Z eudoxia: i mean, i don't remember i time i thought "boy, there's a complete lack of any usable libraries for X" 2014-10-10T16:20:07Z eudoxia: sure, it's missing some things, but let's not exaggerate 2014-10-10T16:20:38Z drewc: eudoxia: I would say that whomever insists on such a thing is not a common lisper. I have never heard that at all personally. 2014-10-10T16:20:41Z Ven_ joined #lisp 2014-10-10T16:20:50Z Shinmera: I haven't found a CL program that'll make my windows dance funny on my desktop yet, and you claim there's no lack of libraries? 2014-10-10T16:21:02Z eudoxia: drewc: i agree 2014-10-10T16:21:28Z eudoxia: i'm mostly referring to the ocassional HN commenter who seems to believe there actually is no such thing as lisp libraries 2014-10-10T16:21:35Z moore33: I also enjoy the dating advice from Mark Tarver... wtf. 2014-10-10T16:21:40Z boogie joined #lisp 2014-10-10T16:21:44Z drewc: I mean, Portable Common Loops is a library, non? And we standardised on CLOS, which implementations used what library to implement? 2014-10-10T16:22:23Z eudoxia: i'm not really talking about that kind of foundational libraries 2014-10-10T16:22:41Z eudoxia: more like the sort of high-level stuff you find here http://github.com/fukamachi/ 2014-10-10T16:23:04Z drewc: eudoxia: Ahh ... those of us who think that PG is clueless also do not read that capitalist^Hhacker news. :P 2014-10-10T16:23:30Z eudoxia: drewc: i mostly use it to get stars on GitHub for my projects, and so far it's worked 2014-10-10T16:23:51Z drewc: heh ... look at the very first and reminds me of my first high level lisp library. 2014-10-10T16:24:14Z eudoxia: it helps when you build a pretty, but otherwise useless, project website, because people these days only use projects whose CSS they can trust 2014-10-10T16:24:21Z drewc: As it turns out, most people climb different mountains at different times... 2014-10-10T16:24:31Z moore33: drewc: It's a bit of a stretch to call Portable Common Loops a library, IMHO. It was widely ported, but required a great deal of magic integration with the guts of the host implementation. 2014-10-10T16:24:33Z Ven quit (Ping timeout: 272 seconds) 2014-10-10T16:25:10Z drewc: so 'high level' is not neccessarily the level on the mountain I want to climb. 2014-10-10T16:25:53Z drewc: moore33: ok, so bordeaux-threads is not a library? hunchentoot is not? usocket is not? 2014-10-10T16:26:11Z eudoxia: shrug. people have already done enough work on the foundations. now it's time for the skyscrapers. 2014-10-10T16:26:40Z moore33: drewc: Sure, those are libraries, as far as I know. 2014-10-10T16:26:48Z Harag quit (Ping timeout: 260 seconds) 2014-10-10T16:27:06Z drewc: eudoxia: on those of us that live on the 3/4 of the world where buildings are impossible and there is no foundation? what of us? 2014-10-10T16:27:21Z eudoxia: drewc: you mean embedded systems? 2014-10-10T16:27:50Z drewc: eudoxia: no, I mean the ocean, I live on a sailboat on the Pacific. 2014-10-10T16:27:55Z Shinmera: 3/4 sounds more like ARM 2014-10-10T16:28:20Z eudoxia: pretty sure clack will run on a tiny computer on a sailboat :P 2014-10-10T16:28:21Z pjb: eudoxia: clearly, we still dont' have a general AI and the software for a robot to clean up the bathrooms in lisp, so it's very lacking of "libraries", indeed. 2014-10-10T16:28:37Z Shinmera: At least lisp can clean our floors 2014-10-10T16:28:47Z oleo: lol 2014-10-10T16:28:47Z drewc agrees with pjb :) 2014-10-10T16:29:21Z bugtroll joined #lisp 2014-10-10T16:29:21Z bugtroll quit (Client Quit) 2014-10-10T16:30:32Z drewc: eudoxia: why would I want to use clack? Just because there is a high level library means we should use it because, after all, solid foundations correct? I thought that that lack of libraries means that ... wait ... 2014-10-10T16:30:37Z beach joined #lisp 2014-10-10T16:30:51Z beach: Good evening everyone! 2014-10-10T16:30:54Z drewc will now say "Good Morning" 2014-10-10T16:31:07Z ndrei joined #lisp 2014-10-10T16:31:19Z eudoxia: drewc: i wouldn't use clack directly, i'd rather use a web framework built on *top* of clack 2014-10-10T16:31:58Z drewc: eudoxia: so, you like to build lightning rods at the top of skyscrapers? URL please! 2014-10-10T16:32:00Z bugtroll joined #lisp 2014-10-10T16:32:02Z eudoxia: for the same reason i wouldn't use usocket for build a web server. abstractions are nice 2014-10-10T16:32:09Z MoALTz joined #lisp 2014-10-10T16:32:19Z Shinmera: Abstractions are nice when they work. 2014-10-10T16:32:24Z drewc: usocket is a high level skyscraper? 2014-10-10T16:32:32Z hiroakip quit (Ping timeout: 272 seconds) 2014-10-10T16:32:54Z eudoxia: no, no, usocket is a foundational library 2014-10-10T16:33:03Z jlongster joined #lisp 2014-10-10T16:33:08Z drewc is now very confused ... wants to be a part of this new community for a new language full of skyscrapers! 2014-10-10T16:33:10Z eudoxia: clack is not really a skyscraper but it's certainy higher-level 2014-10-10T16:34:18Z drewc: eudoxia: so how does clack differ from, say, UCW, LOL, FTW, and other web 'frameworks' that I have stopped using/developing? 2014-10-10T16:34:27Z tajjada joined #lisp 2014-10-10T16:34:33Z moore33 quit 2014-10-10T16:34:45Z eudoxia: drewc: it's maintained and has like 300 stars on github, which is comforting for potential users 2014-10-10T16:34:48Z eudoxia: social proof and all 2014-10-10T16:34:50Z drewc want to know why we are talking about clack in specific .. are there no libraries? 2014-10-10T16:35:02Z drewc: Oh .... ok. 2014-10-10T16:35:16Z ehu quit (Ping timeout: 250 seconds) 2014-10-10T16:35:36Z normanrichards joined #lisp 2014-10-10T16:36:44Z beach: Bike: This is what I came up with for generic EVAL and MACRO-FUNCTION: http://paste.lisp.org/+333S 2014-10-10T16:37:08Z Fare joined #lisp 2014-10-10T16:37:11Z beach: pjb: I am sorry, I didn't see your PM before. 2014-10-10T16:37:25Z drewc makes an X on his checklist, as 300 stars of github means nothing but I like common lisp, yes not CL + some docstring readmacro and a whole bunch of other misuses of #\@. Thanks eudoxia :) 2014-10-10T16:39:06Z eudoxia: it does, however, mean a lot to potential users, because frequent maintenance, open issues, and 300 stars generally means that others are interested in it or are actually using it 2014-10-10T16:39:31Z beach: martinjungblut: Same with you. I am sorry that I didn't see your PM. 2014-10-10T16:39:34Z eudoxia: Shinmera: is there an HTTP file downloader somewhere in the vastness of Radiance? 2014-10-10T16:39:37Z Xach: eudoxia: there are a lot of people in CL who are content to do things in a way only they personally can understand and enjoy 2014-10-10T16:39:49Z fantazo joined #lisp 2014-10-10T16:40:07Z Xach: eudoxia: the warm nest of their own creation is the only home they can enjoy 2014-10-10T16:40:19Z drewc: eudoxia: I bet it is big on hacker news! ;) 2014-10-10T16:41:09Z Beluki joined #lisp 2014-10-10T16:41:45Z zarkone joined #lisp 2014-10-10T16:42:25Z eudoxia: Xach: well at least we have fukamachi and Shinmera to counterbalance that 2014-10-10T16:42:50Z drewc: all said and done, it reminds me of the UCW that segv wrote in about '05, which I used (was forced to use/needed to use) 'till last year... 2014-10-10T16:43:06Z hitecnologys quit (Quit: hitecnologys) 2014-10-10T16:43:40Z pavelpenev_ quit (Quit: No Ping reply in 180 seconds.) 2014-10-10T16:43:51Z drewc: so why is fukamachi's code superior? I do not know, but somebody might. 2014-10-10T16:44:24Z Xach: eudoxia: indeed 2014-10-10T16:44:30Z hardenedapple quit (Quit: WeeChat 1.0.1) 2014-10-10T16:44:30Z eudoxia: drewc: he writes (and maintains!) lots of good libraries with good documentation 2014-10-10T16:44:52Z pavelpenev joined #lisp 2014-10-10T16:45:11Z eudoxia: the code itself is also very clear, one package per file, that sort of stuff 2014-10-10T16:45:16Z BitPuffin joined #lisp 2014-10-10T16:45:36Z eudoxia: consider https://github.com/fukamachi/clack/blob/master/src/core/app/urlmap.lisp 2014-10-10T16:46:41Z eudoxia: I also try to follow Ariel Network's CL style guide, but only used annotations/decorators in Crane http://labs.ariel-networks.com/cl-style-guide.html 2014-10-10T16:46:55Z Shinmera is not sure why eudoxia mentioned him in reply to Xach since he makes libs mostly because he either doesn't like others or just doesn't care to understand them, so Xach was pretty much spot on. 2014-10-10T16:47:24Z Shinmera: eudoxia: Why would there be an HTTP downloader when there's drakma. 2014-10-10T16:47:26Z eudoxia: Shinmera: what do you mean? there wasn't anything like clack or cl-dbi 2014-10-10T16:47:35Z drewc: eudoxia: if you consider @export ,@doc and other such stuff 'clear', I would enjoy having a beer with you and looking at the 'clear' bottle. :) 2014-10-10T16:47:47Z Xach: Shinmera: I think that is different from "only you personally can use and understand" 2014-10-10T16:48:07Z eudoxia: Shinmera: i was wondering if you had one, because trivial-download is kinda slow and maybe i was doing something extremely wrong 2014-10-10T16:48:10Z Shinmera: eudoxia: the 'he' in my /me was referring to myself 2014-10-10T16:48:13Z Xach: drewc: a bottle of monadic combinator beer? 2014-10-10T16:48:22Z Shinmera: Xach: I suppose so 2014-10-10T16:48:52Z drewc: Xach: indeed ... In common lisp even! 2014-10-10T16:48:58Z eudoxia: i also agree with Xach, making a library that does the same as an existing one is not wrong if your new library has better docs and that kind of stuff 2014-10-10T16:49:08Z Shinmera shrugs 2014-10-10T16:49:13Z Bike: i don't think i understand the three argument thing, but you have a good reason i'm sure 2014-10-10T16:51:10Z eudoxia: drewc: i personally don't use the @ stuff, but i don't think it confuses things 2014-10-10T16:51:26Z oudeis quit (Quit: This computer has gone to sleep) 2014-10-10T16:52:01Z drewc looks at clack and sees UCW, looks at cl-dbi and sees CLSQL ... and now shuts up because if such things did not exist, then he is out to lunch. 2014-10-10T16:52:11Z Ven_ quit (Quit: My MacBook has gone to sleep. ZZZzzz…) 2014-10-10T16:53:01Z eudoxia: the thing is, cl-dbi, sxql, integral, crane are all separate projects, while CLSQL is a "great whale" projects. it includes the equivalent of cl-dbi, sxql, and the ORM, all rolled into one 2014-10-10T16:53:18Z eudoxia: but you can use cl-dbi independently of an ORM, sxql independently as well, etc. 2014-10-10T16:53:48Z eudoxia: i think having lots of small libraries, and a couple that merge them together into a larger framework, is better than trying to have huge frameworks like CLSQL 2014-10-10T16:53:59Z banjara joined #lisp 2014-10-10T16:54:00Z drewc: and you cannot use CLSQL without the ORM? having never used the ORM, I guess it is breakfast time... 2014-10-10T16:54:04Z banjara quit (Changing host) 2014-10-10T16:54:04Z banjara joined #lisp 2014-10-10T16:54:28Z drewc: But please write a library or documentation that explains to me why I am so wrong :) 2014-10-10T16:54:40Z eudoxia: drewc: you can, i just don't think CLSQL should implement its own database bindings, SQL generator, etc. 2014-10-10T16:55:35Z psy quit (Disconnected by services) 2014-10-10T16:55:41Z drewc: eudoxia: so, rather then patch it, much better to create all new libraries that do most of what is needed! It will be in quicklisp after all :D 2014-10-10T16:55:43Z eudoxia: i mean, looking at clsql.asd, it doesn't look like it depends on a single external library except cffi 2014-10-10T16:55:58Z eudoxia: that's a massive repetition of effort 2014-10-10T16:56:18Z psy_ joined #lisp 2014-10-10T16:57:28Z drewc: eudoxia: can you show me what libraries it should have been based on? If there was a massive repetition of effort, I would like to know... after all, the lack of libraries meant that every lib counts! 2014-10-10T16:58:45Z eudoxia: drewc: there's cl-dbi, which is a backend-independent database lib. sxql, which is a DSL for generating SQL. then there's integral and crane, which implement an ORM on top of those two libraries. cl-dbi, in turn, depends on other libraries, pg for Postgres, cl-sqlite3 for sqlite3, some mysql lib for mysql 2014-10-10T16:58:51Z drewc stopped using CLSQL around '06 or so, so anything after that does not really count for my education 2014-10-10T16:58:57Z eudoxia: small, modular, independently-tested components 2014-10-10T16:59:09Z drewc: eudoxia: ok, so why is CLSQL not based on those! 2014-10-10T16:59:22Z Jesin quit (Quit: Leaving) 2014-10-10T16:59:25Z eudoxia: drewc: because they were written after CLSQL was written 2014-10-10T16:59:48Z eudoxia: but if i had been writing CLSQL, that's the approach i would have taken. write the database bindings as separate libraries, then write cl-dbi, then sxql, then crane 2014-10-10T17:00:20Z drewc: so why should clsql have been based on? and why is the new massive repetition of effort better then a patch would have been? 2014-10-10T17:00:36Z hlavaty left #lisp 2014-10-10T17:00:57Z Xach: clsql was for a long time barely maintained 2014-10-10T17:01:05Z Xach: it is more maintained now, which is good. 2014-10-10T17:01:14Z drewc: eudoxia: Ok, you are a victor writing history? 2014-10-10T17:01:17Z ndrei quit (Ping timeout: 272 seconds) 2014-10-10T17:01:56Z Shinmera: What's with the pissing contest in here, stop it. It's embarrassing to read. 2014-10-10T17:02:13Z drewc: Xach: I stopped using it a while ago, when postmodern pre-release release started working ... So I have no idea tbh. 2014-10-10T17:02:18Z drewc: Shinmera: you win! 2014-10-10T17:02:29Z boogie quit (Quit: Leaving...) 2014-10-10T17:02:35Z Xach: drewc: That much is very clear. 2014-10-10T17:03:01Z drewc shuts up now .. sorry for trying to learn. 2014-10-10T17:03:04Z reb`` quit (Remote host closed the connection) 2014-10-10T17:03:28Z eudoxia quit (Quit: leaving) 2014-10-10T17:04:34Z Lefeni joined #lisp 2014-10-10T17:04:47Z thawes_ quit (Ping timeout: 245 seconds) 2014-10-10T17:06:13Z thawes_ joined #lisp 2014-10-10T17:07:42Z BitPuffin quit (Ping timeout: 245 seconds) 2014-10-10T17:12:47Z drewc: #+name: tangle_code-block-noweb-body 2014-10-10T17:12:57Z drewc: yikes, sorry, mis-paste. 2014-10-10T17:15:06Z reb` joined #lisp 2014-10-10T17:16:57Z Poenikatu left #lisp 2014-10-10T17:17:46Z ltbarcly joined #lisp 2014-10-10T17:20:08Z InvalidCo: hmm 2014-10-10T17:20:33Z InvalidCo: what would be the best way to make a lambda that has it's code generated at compile-time? 2014-10-10T17:20:54Z dlowe: (lambda () .. code to be generated at compile time ..) 2014-10-10T17:21:05Z InvalidCo: err, no 2014-10-10T17:21:25Z svetlyak40wt joined #lisp 2014-10-10T17:21:34Z InvalidCo: the problem is the code generator function needs to be executed after a few certain symbols are bound 2014-10-10T17:21:45Z InvalidCo: by some other code that is macrogenerated 2014-10-10T17:21:53Z dlowe: that's precisely what macros are 2014-10-10T17:23:04Z dlowe: (lambda () (my-macro-that-generates-code)) 2014-10-10T17:23:23Z InvalidCo: I don't think you understand 2014-10-10T17:23:38Z InvalidCo: let me explain with a simplified version of the code 2014-10-10T17:24:15Z InvalidCo: (macro-that-generates-code-which-binds-symbol) (make-instance 'this-and-that :selector #'(lambda () (macro-which-needs-symbol-bound))) 2014-10-10T17:24:27Z InvalidCo: now, if you type these at the repl they work fine 2014-10-10T17:24:40Z InvalidCo: because the first code-generating macro also gets to execute the code in question 2014-10-10T17:25:03Z InvalidCo: but when they are in the same file and being compiled, sbcl seems to evaluate macros first 2014-10-10T17:25:18Z InvalidCo: and when it gets to macro-which-needs-symbol-bound the symbol is not bound yet 2014-10-10T17:25:21Z dlowe: you're binding a special variable there/ 2014-10-10T17:25:22Z dlowe: ? 2014-10-10T17:25:32Z dlowe: in the first one 2014-10-10T17:25:40Z InvalidCo: not me 2014-10-10T17:25:45Z InvalidCo: it's generating a prototype class 2014-10-10T17:25:53Z dlowe: ah, ok 2014-10-10T17:26:01Z InvalidCo: I'm using sheeple but I'd imagine it would be the same situation with clos 2014-10-10T17:26:06Z kushal quit (Excess Flood) 2014-10-10T17:26:40Z dlowe: no, clos would resolve the symbol at run time, when the lambda was executed 2014-10-10T17:26:48Z InvalidCo: as both generate the (add-class (make-class) *class-list*) or whatever 2014-10-10T17:27:05Z InvalidCo: you still don't understand 2014-10-10T17:27:17Z InvalidCo: the _macro_ needs the symbol bound 2014-10-10T17:27:18Z kushal joined #lisp 2014-10-10T17:27:34Z dlowe: then you need to evaluate the add-class at compile time 2014-10-10T17:27:39Z InvalidCo: it uses it to calculate stylesheet specificity and it needs to know the kind of class it is operating on 2014-10-10T17:27:57Z dlowe: which you can do by the first macro expanding to eval-when 2014-10-10T17:28:05Z InvalidCo: I guess I'll have to fork sheeple, then 2014-10-10T17:28:50Z InvalidCo: although the relevant line is in the root of a progn 2014-10-10T17:29:04Z dlowe: (eval-when (:compile-toplevel :load-toplevel :execute) (macro-that-generates-code-which-binds-symbol)) 2014-10-10T17:31:27Z InvalidCo: yup, now it works 2014-10-10T17:32:20Z hugod joined #lisp 2014-10-10T17:32:20Z InvalidCo: but I'd still love to be able to delay the lambda-expansion...maybe even by adding an eval there 2014-10-10T17:32:50Z InvalidCo: *lambda's expansion 2014-10-10T17:33:07Z hugod quit (Remote host closed the connection) 2014-10-10T17:33:41Z InvalidCo: otherwise I'll have to add a lot of eval-whens => invent a new macro 2014-10-10T17:35:48Z wbooze quit (Quit: Client Quit) 2014-10-10T17:36:42Z InvalidCo: but, thanks! 2014-10-10T17:37:11Z ARM9 joined #lisp 2014-10-10T17:37:27Z InvalidCo: I didn't realize that I could/should put the class definition itself in an eval-when 2014-10-10T17:37:37Z ARM9 left #lisp 2014-10-10T17:38:29Z Beluki quit (Quit: Beluki) 2014-10-10T17:39:57Z normanrichards quit (Ping timeout: 258 seconds) 2014-10-10T17:40:33Z zarkone quit (Ping timeout: 272 seconds) 2014-10-10T17:41:17Z kjeldahl quit (Ping timeout: 265 seconds) 2014-10-10T17:44:06Z kjeldahl joined #lisp 2014-10-10T17:52:43Z hrs joined #lisp 2014-10-10T17:53:34Z cheater__ joined #lisp 2014-10-10T17:55:52Z bugtroll left #lisp 2014-10-10T17:56:57Z cheater quit (Ping timeout: 245 seconds) 2014-10-10T17:56:57Z cheater__ is now known as cheater 2014-10-10T17:57:01Z backupthrick quit (Ping timeout: 272 seconds) 2014-10-10T17:57:03Z svetlyak40wt quit (Remote host closed the connection) 2014-10-10T17:57:43Z kjeldahl quit (Ping timeout: 265 seconds) 2014-10-10T17:58:02Z prxq quit (Ping timeout: 272 seconds) 2014-10-10T17:58:06Z backupthrick joined #lisp 2014-10-10T17:58:57Z hugod joined #lisp 2014-10-10T18:00:31Z _leb joined #lisp 2014-10-10T18:00:49Z knobo_ quit (Ping timeout: 272 seconds) 2014-10-10T18:02:47Z kjeldahl joined #lisp 2014-10-10T18:02:48Z thawes_ quit (Ping timeout: 250 seconds) 2014-10-10T18:03:03Z thawes_ joined #lisp 2014-10-10T18:03:19Z atgreen joined #lisp 2014-10-10T18:07:55Z TDog joined #lisp 2014-10-10T18:08:00Z cyc0 joined #lisp 2014-10-10T18:08:40Z kjeldahl quit (Ping timeout: 255 seconds) 2014-10-10T18:09:22Z Fare quit (Ping timeout: 245 seconds) 2014-10-10T18:10:09Z moore33 joined #lisp 2014-10-10T18:12:06Z Xach: I am somewhat gratified by the amount of space the Apple iOS docs spend on explaining how to draw lines that don't look fuzzy because they straddle pixels. Maybe I should link to it in the vecto docs! 2014-10-10T18:12:19Z Hache_ quit (Remote host closed the connection) 2014-10-10T18:13:30Z wasamasa: vecto? 2014-10-10T18:13:52Z ivan4th quit (Ping timeout: 272 seconds) 2014-10-10T18:13:58Z drewc: Xach: I would enjoy reading it, and part of the vecto docs makes it solve at least 2 problems! 2014-10-10T18:14:33Z drewc: (1: drew has not read it. 2: it is not in the vecto docs) 2014-10-10T18:15:44Z Adlai: InvalidCo: seeing as Sheeple isn't actively developed anymore, you could just ask the original developers (aka me and sykopomp) to pick up development 2014-10-10T18:15:58Z Adlai: ie, "fork" is an overstatement 2014-10-10T18:16:43Z ehaliewicz joined #lisp 2014-10-10T18:17:04Z nha joined #lisp 2014-10-10T18:18:10Z Adlai would still be curious to hear how sheeple is working out 2014-10-10T18:19:22Z erikc joined #lisp 2014-10-10T18:19:33Z kjeldahl joined #lisp 2014-10-10T18:19:33Z beach left #lisp 2014-10-10T18:22:34Z moore33 quit 2014-10-10T18:24:34Z kjeldahl quit (Ping timeout: 276 seconds) 2014-10-10T18:25:59Z drewc: Adlai: I started using it recently ... so far, it is 'better' than the ALISTs I used before for a similar task, and way better than ContextL that I also used for a similar task. 2014-10-10T18:26:27Z TDog quit (Ping timeout: 245 seconds) 2014-10-10T18:26:36Z oleo is now known as Guest29532 2014-10-10T18:27:32Z drewc: Adlai: thanks to https://github.com/mmontone/descriptions , sheeple is now a part of my 'I use this commercially!' library set... so at least in that regard, yay! 2014-10-10T18:28:11Z Adlai: yay! 2014-10-10T18:28:20Z oleo__ joined #lisp 2014-10-10T18:28:21Z Adlai uses chanl commercially, which is also nice 2014-10-10T18:29:40Z Guest29532 quit (Ping timeout: 250 seconds) 2014-10-10T18:30:17Z hrs quit (Quit: Textual IRC Client: www.textualapp.com) 2014-10-10T18:30:41Z Krystof: you people and your commercial use of lisp! Crazy talk 2014-10-10T18:30:57Z felideon: drewc: will there be a lisp-on-lines remake on sheeple then? :) 2014-10-10T18:31:05Z drewc: Adlai: I have not used (or even looked into) chanl, but will do so on my lunch break :) 2014-10-10T18:31:25Z Krystof: next you'll be telling me that people can make a living by programming 2014-10-10T18:32:03Z Adlai: for me, at least, the exciting bit is commercial use of software that i forgot even existed for a few years 2014-10-10T18:32:45Z drewc: felideon: Well, LoL is essentially descriptions + UCW ... Now I use my own FTW for the web call/cc type stuff, so I might as well port LoL over to it and descriptions ... so yeah, I guess so :) 2014-10-10T18:32:46Z felideon: drewc: oh, I guess that's what mmontone -is- 2014-10-10T18:33:01Z Adlai is not entirely satisfied with chanl, but is focusing on other things at the moment 2014-10-10T18:33:11Z drewc: felideon: yeah, almost exactly really. 2014-10-10T18:34:03Z Adlai: my general sense being that i just haven't done enough work yet with chanl to know exactly what to change (or what to look for in a replacement) 2014-10-10T18:34:05Z drewc knows that it is based on his work for LoL , and even has a patch in it :) 2014-10-10T18:34:18Z vaporatorius joined #lisp 2014-10-10T18:34:58Z felideon: ok, I now realize mmontone is the developer, not the name of the library. 'descriptions' as a name threw me off. 2014-10-10T18:35:44Z slyrus quit (Ping timeout: 250 seconds) 2014-10-10T18:36:12Z aleamb joined #lisp 2014-10-10T18:36:39Z yrk quit (Quit: ERC Version 5.3 (IRC client for Emacs)) 2014-10-10T18:37:36Z aleamb left #lisp 2014-10-10T18:38:05Z madmalik joined #lisp 2014-10-10T18:39:07Z drmeister: Here's what's missing from Clasp to make it a fully ANSI compliant Common Lisp (bugs notwithstanding). 2014-10-10T18:39:10Z drmeister: https://www.irccloud.com/pastebin/0xaaJEWA 2014-10-10T18:39:23Z drmeister: 65 symbols left to implement. 2014-10-10T18:39:52Z drewc: felideon: Did not throw me off at all ... 2005 IIRC : https://github.com/drewc/lisp-on-lines/blob/master/src/description.lisp :) 2014-10-10T18:41:09Z slyrus joined #lisp 2014-10-10T18:41:23Z gravicappa joined #lisp 2014-10-10T18:41:31Z drewc: drmeister: no CL:PI ... that is easy to implement and impossible to implement 'properly' :) Beyond that, looks like you're almost there! 2014-10-10T18:41:34Z Grue`: some of these look easy 2014-10-10T18:41:57Z drdanmaku quit (Quit: Connection closed for inactivity) 2014-10-10T18:41:58Z drmeister: Well, I was thinking of rounding it down to 3.14. 2014-10-10T18:42:40Z drewc: in my head it is usually 3.14159. 2014-10-10T18:42:40Z MutSbeta joined #lisp 2014-10-10T18:42:55Z drmeister: I am the "implementor" after all. 2014-10-10T18:43:07Z drewc: true that! :D 2014-10-10T18:43:36Z Bike: i already threw in a few definitions for the weird list crap, but i have no idea where they go in the source tree 2014-10-10T18:44:01Z Krystof: drmeister: have you tried the all-important "build SBCL" CL implementation stress-test yet? 2014-10-10T18:44:03Z ndrei joined #lisp 2014-10-10T18:45:57Z aleamb joined #lisp 2014-10-10T18:47:25Z drmeister: Bike, where did you put those definitions? 2014-10-10T18:47:46Z Bike: a paste in #clasp yesterday, and a comment on the integer-length issue. 2014-10-10T18:48:18Z Bike: don't think i've ever seen pairlis actually used, but whatever, it's something i guess 2014-10-10T18:48:47Z slyrus quit (Ping timeout: 255 seconds) 2014-10-10T18:48:57Z ndrei quit (Ping timeout: 272 seconds) 2014-10-10T18:49:13Z Krystof: sbcl uses pairlis in its compiler 2014-10-10T18:49:42Z Krystof: I'm slightly surprised by that -- I thought it would use vector-pop, but no, it uses pairlis 2014-10-10T18:50:13Z Krystof: in other words, you can build sbcl with a lisp that doesn't implement vector-pop, but not with one that doesn't implement pairlis 2014-10-10T18:50:16Z slyrus joined #lisp 2014-10-10T18:50:34Z Krystof: (building sbcl is the only thing CL implementations are for, right? [including sbcl itself...]) 2014-10-10T18:50:46Z Bike: well, clasp doesn't have vector-pop either, so it's all good 2014-10-10T18:51:28Z EvW quit (Ping timeout: 260 seconds) 2014-10-10T18:51:33Z normanrichards joined #lisp 2014-10-10T18:51:49Z ehu joined #lisp 2014-10-10T18:52:03Z drewc: Krystof: I use ALISTs all the time, yet have never used PAIRLIS and, though I remember reading about it a long time ago, did not really know it.... 2014-10-10T18:52:23Z drmeister: These? http://sprunge.us/hWTU 2014-10-10T18:52:32Z Bike: those 2014-10-10T18:52:45Z tajjada quit (Ping timeout: 272 seconds) 2014-10-10T18:52:49Z aleamb quit (Changing host) 2014-10-10T18:52:49Z aleamb joined #lisp 2014-10-10T18:52:56Z drmeister: Where did these come from? 2014-10-10T18:53:01Z drewc: Krystof: Now, well, I know at least one place where I can replace my code with PAIRLIS ... so thank you! 2014-10-10T18:53:08Z drmeister has some paperwork for Bike to sign. 2014-10-10T18:53:11Z drmeister: [I'm kidding] 2014-10-10T18:53:26Z Bike: some are modified copies from sbcl (public domain files), the rest are like two lines a piece 2014-10-10T18:53:38Z Bike: (append (mapcar #'cons keys data) alist) is original, i'm gonna patent that 2014-10-10T18:53:54Z drmeister: Great! Thank you. 2014-10-10T18:54:05Z Bike: oh, and you should see if numlib implements cis already, probably 2014-10-10T18:54:41Z MutSbeta quit (Quit: Leaving.) 2014-10-10T18:54:43Z drmeister: numlib does implement cis. 2014-10-10T18:54:53Z Bike: coerce-fdesignator is sbcl's %coerce-callable-to-fun, which you might already have somewhere 2014-10-10T18:55:00Z swflint joined #lisp 2014-10-10T18:57:09Z Bike: i'd also never seen get-properties, it's a surprisingly complicated function. returns three values! 2014-10-10T18:57:11Z whmark joined #lisp 2014-10-10T18:57:22Z oleo__ quit (Quit: Verlassend) 2014-10-10T18:57:24Z drmeister: Bike: Do you want me to put your full name in the file comment or just Bike from #lisp? 2014-10-10T18:57:35Z Bike: bike from lisp is fine 2014-10-10T18:57:54Z Bike: are you just going to dump that file in there? i put everything in one file just because i wasn't sure where they'd go in the source tree. 2014-10-10T18:58:44Z whmark quit (Read error: Connection reset by peer) 2014-10-10T18:58:53Z kjeldahl joined #lisp 2014-10-10T18:58:55Z drmeister: I put them in clasp/src/lisp/kernel/lsp/assorted.lsp 2014-10-10T18:59:02Z drmeister: It's building them into the source now. 2014-10-10T18:59:11Z Longlius joined #lisp 2014-10-10T18:59:13Z kushal quit (Quit: Leaving) 2014-10-10T18:59:34Z Bike: ...well, i guess that works. 2014-10-10T18:59:43Z stassats` quit (Ping timeout: 272 seconds) 2014-10-10T18:59:48Z Bike: you should probably figure out the CL interface to the reader, i imagine actual code won't work if you can't set macro chars 2014-10-10T18:59:56Z drmeister: The primary concern about where in the source tree is - are the macros that the file depends on in place. 2014-10-10T19:01:15Z drmeister: I don't want to add the functions to other source files because I like to keep the ECL Common Lisp source code as close to the original as possible. That way I can "ediff" the ECL source code against mine and extract updates and bug fixes. 2014-10-10T19:01:31Z oleo__ joined #lisp 2014-10-10T19:01:36Z oleo__ quit (Read error: Connection reset by peer) 2014-10-10T19:01:36Z Bike: yeah, i used loop for a few things sbcl did with do, but i figured you could just put this at the end of the build because they're pretty obscure 2014-10-10T19:01:40Z wooden joined #lisp 2014-10-10T19:01:59Z oleo joined #lisp 2014-10-10T19:02:20Z drmeister: There you go - now that you say that I have to move it later in the build process, after loop. I didn't catch that. 2014-10-10T19:04:13Z drmeister: Here's the build order - although unless you are very familiar with ECL Common Lisp source it won't tell you everything. 2014-10-10T19:04:16Z drmeister: https://www.irccloud.com/pastebin/BODFAdLb 2014-10-10T19:05:06Z drmeister: It's on line 62, right after "lsp/loop2" and before "lsp/defpackage" and "lsp/format" 2014-10-10T19:06:26Z Jesin joined #lisp 2014-10-10T19:07:58Z ndrei joined #lisp 2014-10-10T19:10:19Z Jesin quit (Client Quit) 2014-10-10T19:11:32Z kjeldahl quit (Ping timeout: 260 seconds) 2014-10-10T19:12:22Z ndrei quit (Ping timeout: 245 seconds) 2014-10-10T19:12:31Z Jesin joined #lisp 2014-10-10T19:13:38Z drmeister: Once I add those functions that you provided Bike - do you think it would be worth trying the ANSI tests again? Or will they fail for missing symbols? 2014-10-10T19:14:24Z ndrei joined #lisp 2014-10-10T19:15:18Z Bike: well, you'll still have missing symbols, obviously 2014-10-10T19:15:48Z Patzy quit (Remote host closed the connection) 2014-10-10T19:15:55Z Nizumzen joined #lisp 2014-10-10T19:15:56Z Patzy joined #lisp 2014-10-10T19:16:11Z cheater quit (Ping timeout: 272 seconds) 2014-10-10T19:17:01Z drmeister: Isn't GCL missing things like LOAD-TIME-VALUES? I read that somewhere and it boggles my mind. 2014-10-10T19:18:01Z Bike: gcl sucks, near as i can tell. 2014-10-10T19:18:07Z Bike: though load-time-value is weird. 2014-10-10T19:18:08Z drmeister wonders if he is incorrect then he could unboggle his mind. 2014-10-10T19:18:47Z swflint is now known as swflint_away 2014-10-10T19:19:30Z drewc: is it missing that with -enable-ansi at compile time? Regardless, I agree with Bike on the suction. 2014-10-10T19:19:31Z drmeister: Ok. 2014-10-10T19:19:34Z _leb quit (Quit: Computer has gone to sleep.) 2014-10-10T19:20:14Z swflint_away left #lisp 2014-10-10T19:21:32Z Jesin quit (Ping timeout: 245 seconds) 2014-10-10T19:24:14Z drmeister: On another note: My lab is moving into a new building in a few weeks. The Science Education and Research Center at Temple University. We had the ribbon cutting today. http://cst.temple.edu/research/science-education-and-research-center/SERC-floor-by-floor. 2014-10-10T19:25:24Z drmeister: The floor above me is all computational biology people and they have a high performance computing cluster up there. 2014-10-10T19:25:46Z drmeister: It's going to be fun trying to get the manager let me run Common Lisp code on it. 2014-10-10T19:26:12Z drmeister: Not a lot of PI's come in with code that was compiled on compilers they wrote themselves. 2014-10-10T19:27:49Z drmeister: There's an attitude that FORTRAN is the epitome of software evolution. 2014-10-10T19:28:24Z kjeldahl joined #lisp 2014-10-10T19:29:20Z Jubb quit (Ping timeout: 244 seconds) 2014-10-10T19:29:53Z TDog joined #lisp 2014-10-10T19:30:11Z ThomasH joined #lisp 2014-10-10T19:30:15Z ThomasH: Greetings lispers! 2014-10-10T19:30:48Z aleamb left #lisp 2014-10-10T19:31:45Z Sgeo_ quit (Ping timeout: 246 seconds) 2014-10-10T19:31:55Z aldo quit (Ping timeout: 244 seconds) 2014-10-10T19:32:33Z nug700 joined #lisp 2014-10-10T19:33:07Z aleamb joined #lisp 2014-10-10T19:33:57Z drewc: ThomasH: '(|Greetings, good afternoon!|) 2014-10-10T19:34:10Z cyc0 quit (Ping timeout: 255 seconds) 2014-10-10T19:34:11Z ThomasH: Hello drewc! 2014-10-10T19:34:30Z edgar-rft quit (Quit: activity destroyed by paranoid bleeding) 2014-10-10T19:37:23Z ltbarcly quit (Ping timeout: 255 seconds) 2014-10-10T19:37:48Z cheater__ joined #lisp 2014-10-10T19:37:51Z cheater__ is now known as cheater 2014-10-10T19:38:32Z ltbarcly joined #lisp 2014-10-10T19:40:35Z RenRenJuan quit (Quit: This computer has gone to sleep) 2014-10-10T19:45:10Z Tau joined #lisp 2014-10-10T19:45:16Z Tau left #lisp 2014-10-10T19:45:59Z cyc0 joined #lisp 2014-10-10T19:47:26Z Adlai: drmeister: why not use whatever value of pi the available system's math libraries provide? 2014-10-10T19:47:52Z knob quit (Quit: Leaving) 2014-10-10T19:53:31Z TDog quit (Ping timeout: 255 seconds) 2014-10-10T19:53:43Z Bicyclidine joined #lisp 2014-10-10T19:53:51Z Nizumzen quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/) 2014-10-10T19:54:02Z nipra quit (Quit: Leaving.) 2014-10-10T19:54:32Z malbertife joined #lisp 2014-10-10T19:54:58Z KarlDscc quit (Remote host closed the connection) 2014-10-10T19:55:24Z Aiwass joined #lisp 2014-10-10T20:02:05Z rvirding_: drmeister: where's the fun implementing a language if its compiler is not written in itself? 2014-10-10T20:02:33Z Bicyclidine: you've clearly never had to compile perl :p 2014-10-10T20:03:46Z Jubb joined #lisp 2014-10-10T20:04:42Z drmeister: Adlai: I just haven't gotten around to PI. Get it? 2014-10-10T20:05:13Z rvirding_: Writing a perl compiler in perl would be "interesting", even non-trivial 2014-10-10T20:05:16Z drmeister: Damn, it took a long time to get someone to set that up for me. 2014-10-10T20:05:22Z Adlai: drmeister: here you go! https://leejagers.files.wordpress.com/2006/07/procrastination.jpg 2014-10-10T20:06:27Z Aiwass quit (Quit: ERC (IRC client for Emacs 25.0.50.1)) 2014-10-10T20:08:22Z drmeister: And with that I'm adding PI to Clasp. 2014-10-10T20:08:33Z drmeister: Thankyew, thankyew, thankyew. 2014-10-10T20:09:22Z rme: (defconstant pi ...) ;whew! 2014-10-10T20:09:35Z thawes_ quit (Quit: Konversation terminated!) 2014-10-10T20:10:36Z Adlai: (defconstant pi (float 3)) 2014-10-10T20:13:19Z Patzy quit (Ping timeout: 255 seconds) 2014-10-10T20:13:26Z tajjada joined #lisp 2014-10-10T20:13:55Z Patzy joined #lisp 2014-10-10T20:14:48Z fantazo quit (Ping timeout: 244 seconds) 2014-10-10T20:18:30Z fridim_ quit (Quit: Leaving) 2014-10-10T20:20:52Z ndrei quit (Ping timeout: 240 seconds) 2014-10-10T20:24:06Z karswell` quit (Remote host closed the connection) 2014-10-10T20:24:13Z dim: (deconstant π (/ 355.0d0 113)) 2014-10-10T20:24:52Z karswell` joined #lisp 2014-10-10T20:28:45Z mindCrime_ quit (Quit: Konversation terminated!) 2014-10-10T20:30:46Z dim: or even (/ 2646693125139304345.0d0 842468587426513207) 2014-10-10T20:30:46Z Adlai: aha! (defconstant pi (imagpart (log -1))) 100% ANSI CL 2014-10-10T20:31:18Z dim: pretty cool ;-) 2014-10-10T20:31:36Z dim: the big numbers above give one more decimal place apparently 2014-10-10T20:31:53Z dim: I mean when compared to (defconstant pi (imagpart (log -1.0d0))) 2014-10-10T20:32:27Z Adlai: you could probably find big numbers that give you arbitrarily more precise estimates... 2014-10-10T20:32:49Z dim: you'd be surprised, from what I found 2014-10-10T20:33:03Z dim: http://qin.laya.com/tech_projects_approxpi.html 2014-10-10T20:34:02Z dim: of course there's also a whole wikipedia page about approximating pi 2014-10-10T20:35:40Z martinjungblut: reminds me of fabrice bellard's programme for calculating the then largest known approximation of pi 2014-10-10T20:35:45Z cheater quit (Ping timeout: 272 seconds) 2014-10-10T20:35:55Z martinjungblut: until some asians beat him 2014-10-10T20:36:00Z mishoo quit (Read error: No route to host) 2014-10-10T20:36:04Z dim: Bellard is a very impressive hacker 2014-10-10T20:36:20Z martinjungblut: true 2014-10-10T20:36:22Z dim: he did beat the worldwide competition on his desktop machine… 2014-10-10T20:36:33Z mishoo joined #lisp 2014-10-10T20:36:38Z martinjungblut: if I'm not mistaken, the machine he ran the programme on was not that powerful 2014-10-10T20:36:57Z Bicyclidine: bellard also has x86 with linux in js, there is no hope 2014-10-10T20:37:04Z dim: http://bellard.org/pi/pi2700e9/ 2014-10-10T20:37:25Z dim: Bellard wrote tinycc, then tccboot, then qemu, then jslinux, yeah 2014-10-10T20:37:29Z martinjungblut: the asian guys' machine was much, much more powerful, but I'm not if that is due to Bellard's brilliance, or it is just that the more you approximate pi, the more hardware you need 2014-10-10T20:37:38Z martinjungblut: he's incredible 2014-10-10T20:37:47Z Bicyclidine: is there a particular reason you don't just say 'kondo' 2014-10-10T20:37:50Z cheater__ joined #lisp 2014-10-10T20:37:55Z cheater__ is now known as cheater 2014-10-10T20:38:12Z martinjungblut: s/I'm not if/I'm not sure if/ 2014-10-10T20:38:40Z martinjungblut: what is a kondo? 2014-10-10T20:38:49Z Shinmera: The "asian guy" 2014-10-10T20:38:50Z Bicyclidine: shigeru kondo is apparently "the asian guy" 2014-10-10T20:38:58Z ndrei joined #lisp 2014-10-10T20:39:16Z ltbarcly quit (Ping timeout: 260 seconds) 2014-10-10T20:39:20Z martinjungblut: I didn't know his name until now 2014-10-10T20:39:29Z martinjungblut: lol 2014-10-10T20:39:45Z erikc quit (Quit: erikc) 2014-10-10T20:40:21Z dim: My computation used a single Core i7 Quad Core CPU at 2.93 GHz giving a peak processing power of 46.9 Gflops. So the supercomputer is about 2000 times faster than my computer. However, my computation lasted 116 days, which is 96 times slower than the supercomputer for about the same number of digits. So my computation is roughly 20 times more efficient. 2014-10-10T20:40:46Z dim: that's a quote from Bellard's FAQ about his Pi digits computation 2014-10-10T20:40:47Z martinjungblut: oh, wow, 5 trillion digits vs. bellard's 2.700 billion 2014-10-10T20:41:31Z martinjungblut: ah, there we go, that's the question I was asking myself 2014-10-10T20:41:33Z martinjungblut: thanks dim! 2014-10-10T20:42:28Z ltbarcly joined #lisp 2014-10-10T20:43:24Z mishoo_ joined #lisp 2014-10-10T20:44:15Z mishoo quit (Ping timeout: 244 seconds) 2014-10-10T20:44:43Z cyc0 quit (Ping timeout: 258 seconds) 2014-10-10T20:46:54Z jkaye joined #lisp 2014-10-10T20:47:01Z ltbarcly quit (Ping timeout: 258 seconds) 2014-10-10T20:47:44Z ggole quit 2014-10-10T20:48:04Z ltbarcly joined #lisp 2014-10-10T20:48:15Z jegaxd26 joined #lisp 2014-10-10T20:48:22Z nipra joined #lisp 2014-10-10T20:50:58Z Shinmera quit (Quit: しつれいしなければならないんです。) 2014-10-10T20:51:14Z jkaye quit (Ping timeout: 258 seconds) 2014-10-10T20:51:36Z Ven joined #lisp 2014-10-10T20:52:00Z gravicappa quit (Ping timeout: 244 seconds) 2014-10-10T20:53:51Z dim: today I read some article about early Common Lisp GUI innovations, that were compared to modern notebook projects, and now I want to have a SQL notebook! 2014-10-10T20:57:49Z svetlyak40wt joined #lisp 2014-10-10T20:59:47Z resttime joined #lisp 2014-10-10T21:01:10Z nipra quit (Quit: Leaving.) 2014-10-10T21:02:22Z Bicyclidine quit (Ping timeout: 255 seconds) 2014-10-10T21:02:35Z svetlyak40wt quit (Ping timeout: 272 seconds) 2014-10-10T21:03:05Z MoALTz quit (Ping timeout: 260 seconds) 2014-10-10T21:07:27Z dim: and https://github.com/Inaimathi/cl-notebook might even be a good start 2014-10-10T21:08:42Z nipra joined #lisp 2014-10-10T21:14:30Z martinjungblut1 joined #lisp 2014-10-10T21:17:41Z martinjungblut quit (Ping timeout: 258 seconds) 2014-10-10T21:17:52Z cheater quit (Ping timeout: 240 seconds) 2014-10-10T21:19:54Z cheater__ joined #lisp 2014-10-10T21:19:58Z cheater__ is now known as cheater 2014-10-10T21:20:39Z Ven quit (Read error: Connection reset by peer) 2014-10-10T21:23:28Z moore33 joined #lisp 2014-10-10T21:23:57Z mrSpec joined #lisp 2014-10-10T21:24:04Z mrSpec quit (Changing host) 2014-10-10T21:24:04Z mrSpec joined #lisp 2014-10-10T21:25:28Z ehu_ joined #lisp 2014-10-10T21:25:33Z malbertife quit (Quit: Leaving) 2014-10-10T21:25:46Z ehu quit (Ping timeout: 255 seconds) 2014-10-10T21:26:40Z cheater quit (Ping timeout: 255 seconds) 2014-10-10T21:28:31Z drdanmaku joined #lisp 2014-10-10T21:29:02Z angavrilov quit (Remote host closed the connection) 2014-10-10T21:30:50Z martinjungblut1 quit (Quit: Leaving.) 2014-10-10T21:35:56Z innertracks joined #lisp 2014-10-10T21:36:39Z Beetny joined #lisp 2014-10-10T21:42:13Z mishoo_ quit (Ping timeout: 258 seconds) 2014-10-10T21:42:56Z mrSpec quit (Ping timeout: 250 seconds) 2014-10-10T21:51:13Z innertracks quit (Ping timeout: 260 seconds) 2014-10-10T21:52:11Z LiamH quit (Quit: Leaving.) 2014-10-10T21:54:22Z Lycurgus joined #lisp 2014-10-10T21:54:23Z zickzackv joined #lisp 2014-10-10T21:59:22Z gingerale quit (Ping timeout: 245 seconds) 2014-10-10T22:02:39Z Bicyclidine joined #lisp 2014-10-10T22:04:45Z EvW joined #lisp 2014-10-10T22:05:27Z madmalik quit (Quit: Connection closed for inactivity) 2014-10-10T22:08:52Z ehu_ quit (Ping timeout: 260 seconds) 2014-10-10T22:11:49Z nha quit (Quit: "If an MMO game is serving more people than it's bandwidth can handle, then you get lag. Lag leads to anger, anger leads to hate and hate leads to people not coming back again..." -- Dean Harding) 2014-10-10T22:13:13Z urandom__ quit (Quit: Konversation terminated!) 2014-10-10T22:15:36Z zeitue joined #lisp 2014-10-10T22:18:40Z devon: (ql:quickload "bordeaux-fft") -- CCL::INVALID-MEMORY-OPERATION -- do you get this? 2014-10-10T22:20:25Z phao quit (Remote host closed the connection) 2014-10-10T22:21:25Z rme: It loads for me on current trunk ccl, darwin/x8664. Quicklisp fetched bordeaux-fft-20101006-http.tgz. 2014-10-10T22:25:22Z freaksken quit (Ping timeout: 240 seconds) 2014-10-10T22:27:42Z chu joined #lisp 2014-10-10T22:27:42Z tajjada quit (Read error: Connection reset by peer) 2014-10-10T22:32:45Z jlongster quit (Ping timeout: 244 seconds) 2014-10-10T22:34:47Z normanrichards quit (Ping timeout: 245 seconds) 2014-10-10T22:37:55Z Jesin joined #lisp 2014-10-10T22:37:58Z zickzackv quit (Ping timeout: 250 seconds) 2014-10-10T22:39:10Z kcj joined #lisp 2014-10-10T22:43:37Z ndrei quit (Ping timeout: 255 seconds) 2014-10-10T22:47:37Z jkaye joined #lisp 2014-10-10T22:51:57Z jkaye quit (Ping timeout: 245 seconds) 2014-10-10T22:53:31Z banjara quit (Ping timeout: 258 seconds) 2014-10-10T22:56:45Z banjara joined #lisp 2014-10-10T22:58:27Z paul0 quit (Read error: No route to host) 2014-10-10T22:59:02Z phao joined #lisp 2014-10-10T23:01:04Z vaporatorius quit (Remote host closed the connection) 2014-10-10T23:03:03Z jlongster joined #lisp 2014-10-10T23:03:33Z Bicyclidine quit (Ping timeout: 272 seconds) 2014-10-10T23:08:24Z devon: rme: still here? 2014-10-10T23:08:38Z ynniv joined #lisp 2014-10-10T23:09:15Z devon: (list *FEATURES* *LOAD-PATHNAME* (LISP-IMPLEMENTATION-TYPE) (LISP-IMPLEMENTATION-VERSION) (MACHINE-VERSION) (SOFTWARE-VERSION)) 2014-10-10T23:09:15Z devon: ==> ((:QUICKLISP :ASDF-PACKAGE-SYSTEM :ASDF3.1 :ASDF3 :ASDF2 :ASDF :OS-MACOSX :OS-UNIX :ASDF-UNICODE :PRIMARY-CLASSES :COMMON-LISP :OPENMCL :CCL :CCL-1.2 :CCL-1.3 :CCL-1.4 :CCL-1.5 :CCL-1.6 :CCL-1.7 :CCL-1.8 :CCL-1.9 :CCL-1.10 :CLOZURE :CLOZURE-COMMON-LISP :ANSI-CL :UNIX :OPENMCL-UNICODE-STRINGS :OPENMCL-NATIVE-THREADS :OPENMCL-PARTIAL-MOP :MCL-COMMON-MOP-SUBSET :OPENMCL-MOP-2 :OPENMCL-PRIVATE-HASH-TABLES :X86-64 :X8 2014-10-10T23:09:15Z devon: 6_64 :X86-TARGET :X86-HOST :X8664-TARGET :X8664-HOST :DARWIN-HOST :DARWIN-TARGET :DARWINX86-TARGET :DARWINX8664-TARGET :DARWINX8664-HOST :64-BIT-TARGET :64-BIT-HOST :DARWIN :LITTLE-ENDIAN-TARGET :LITTLE-ENDIAN-HOST) NIL "Clozure Common Lisp" "Version 1.10-r16196 (DarwinX8664)" "MacBookAir6,2" "13.4.0") 2014-10-10T23:10:40Z ynniv: Can I undeclare a constant? I’m working interactively in slime and I declared a constant. Now every time I reload the file it complains that I’m redefining a constant. But I can’t make it a parameter now that it’s a constant… Can I remove the constant declaration from the namespace somehow? 2014-10-10T23:11:28Z ynniv: Obviously I can start a new session, so this is mostly trivia. 2014-10-10T23:11:47Z rme: ynniv: unintern it? 2014-10-10T23:12:32Z ynniv: Yup, that’s the magic I was looking for. 2014-10-10T23:12:38Z nand1 joined #lisp 2014-10-10T23:14:39Z wilfredh joined #lisp 2014-10-10T23:16:04Z rme: devon: I think it's a CCL bug, perhaps http://trac.clozure.com/ccl/ticket/1237. If you update ccl (with svn up) and rebuild with (rebuild-ccl :full t), it should work. 2014-10-10T23:16:13Z kuzy000_ quit (Ping timeout: 272 seconds) 2014-10-10T23:19:39Z nand1 quit (Remote host closed the connection) 2014-10-10T23:19:56Z nand1 joined #lisp 2014-10-10T23:21:33Z MutSbeta joined #lisp 2014-10-10T23:23:37Z c107 quit (Remote host closed the connection) 2014-10-10T23:24:39Z boogie joined #lisp 2014-10-10T23:24:46Z nand1 quit (Remote host closed the connection) 2014-10-10T23:25:02Z nand1 joined #lisp 2014-10-10T23:25:04Z devon: rme: fatal error: ../platform-darwinx8664.h:24:10: fatal error: 'sys/signal.h' file not found 2014-10-10T23:25:45Z rme: devon: You have to run "xcode-select --install" from a terminal in order to install the command-line compiler tools. 2014-10-10T23:26:00Z j_king_: is it possible to determine if a given vector is :adjustable? or specialize a method on an adjustable array vs. non-adjustable array? 2014-10-10T23:26:57Z j_king_: or is the appropriate approach to just try vector-push-extend and provide a restart to resize the array manually? 2014-10-10T23:29:36Z Sgeo joined #lisp 2014-10-10T23:32:06Z Xach: j_king_: the return value of adjust-array will be eq to the original if it's actually adjustable 2014-10-10T23:33:02Z j_king_: Xach: thanks 2014-10-10T23:33:23Z rme: Does adjustable-array-p work, or is it possible for that to return t always? 2014-10-10T23:33:38Z Xach: rme: that looks pretty good 2014-10-10T23:34:14Z Xach: rme: i think it's possible for every array to be adjustable, not that the test returns t always 2014-10-10T23:44:13Z EvW quit (Quit: EvW) 2014-10-10T23:51:25Z drmeister: The BOOLE function operates on integers with different integer-length's - correct? 2014-10-10T23:53:18Z cneira joined #lisp 2014-10-10T23:53:53Z Sgeo_ joined #lisp 2014-10-10T23:55:38Z ltbarcly quit (Quit: Computer has gone to sleep.) 2014-10-10T23:57:56Z Sgeo quit (Ping timeout: 255 seconds) 2014-10-10T23:58:33Z jgoss joined #lisp