2016-08-18T00:00:42Z Petit_Dejeuner quit (Ping timeout: 276 seconds)
2016-08-18T00:01:10Z shka quit (Ping timeout: 250 seconds)
2016-08-18T00:01:41Z cagomez joined #lisp
2016-08-18T00:06:14Z phadthai joined #lisp
2016-08-18T00:08:55Z Polyphony joined #lisp
2016-08-18T00:09:53Z mastokley quit (Ping timeout: 265 seconds)
2016-08-18T00:11:11Z fUD joined #lisp
2016-08-18T00:15:27Z phoe quit (Ping timeout: 264 seconds)
2016-08-18T00:16:54Z cagomez quit
2016-08-18T00:17:51Z lexicall joined #lisp
2016-08-18T00:18:37Z lexicall quit (Client Quit)
2016-08-18T00:21:25Z jdtest3 joined #lisp
2016-08-18T00:21:27Z jdtest quit (Read error: Connection reset by peer)
2016-08-18T00:25:20Z BusFactor1 quit (Quit: My Mac has gone to sleep. ZZZzzz…)
2016-08-18T00:40:44Z phoe joined #lisp
2016-08-18T00:42:37Z oleo_ joined #lisp
2016-08-18T00:45:35Z emma is now known as em
2016-08-18T00:46:42Z oleo quit (Ping timeout: 265 seconds)
2016-08-18T00:51:05Z IPmonger joined #lisp
2016-08-18T00:52:56Z discardedes quit (Remote host closed the connection)
2016-08-18T00:55:57Z IPmonger quit (Ping timeout: 276 seconds)
2016-08-18T00:56:41Z shdeng joined #lisp
2016-08-18T01:08:42Z fiddlerwoaroof_: phoe: have you looked at xelf at all?
2016-08-18T01:08:47Z wildlander quit (Quit: Saliendo)
2016-08-18T01:09:07Z phoe: fiddlerwoaroof_: not yet.
2016-08-18T01:09:14Z fiddlerwoaroof_: I played with it a bit once, and it seemed to have a nice API for game stuff
2016-08-18T01:09:19Z phoe: oh, wait!
2016-08-18T01:09:21Z fiddlerwoaroof_ is now known as fiddlerwoaroof
2016-08-18T01:09:22Z phoe: xelf! I did.,
2016-08-18T01:09:25Z fiddlerwoaroof: http://xelf.me/
2016-08-18T01:09:44Z phoe: I'll poke it again tomorrow.
2016-08-18T01:09:45Z phoe: Night!
2016-08-18T01:09:49Z fiddlerwoaroof: night
2016-08-18T01:14:29Z Petit_Dejeuner joined #lisp
2016-08-18T01:15:40Z vaitel quit (Quit: Leaving)
2016-08-18T01:16:05Z vaitel joined #lisp
2016-08-18T01:18:56Z zacts joined #lisp
2016-08-18T01:20:45Z Arathnim joined #lisp
2016-08-18T01:21:20Z malabaris joined #lisp
2016-08-18T01:21:56Z malabaris quit (K-Lined)
2016-08-18T01:27:32Z ovenpasta quit (Ping timeout: 240 seconds)
2016-08-18T01:32:25Z SeanP left #lisp
2016-08-18T01:33:35Z SeanP joined #lisp
2016-08-18T01:38:20Z EvW quit (Ping timeout: 265 seconds)
2016-08-18T01:38:44Z eSVG joined #lisp
2016-08-18T01:40:30Z defaultxr joined #lisp
2016-08-18T01:40:56Z dyelar quit (Quit: Leaving.)
2016-08-18T01:42:34Z aphprentice joined #lisp
2016-08-18T01:42:45Z mastokley joined #lisp
2016-08-18T01:44:43Z kn-928 quit (Quit: bye)
2016-08-18T01:47:43Z kn-928 joined #lisp
2016-08-18T01:51:16Z yrk quit (Read error: Connection reset by peer)
2016-08-18T01:53:05Z arescorpio joined #lisp
2016-08-18T01:56:07Z nikki93 quit (Remote host closed the connection)
2016-08-18T01:56:36Z pseudan joined #lisp
2016-08-18T01:56:45Z nikki93 joined #lisp
2016-08-18T01:59:17Z tmtwd_ joined #lisp
2016-08-18T02:00:58Z nikki93 quit (Ping timeout: 252 seconds)
2016-08-18T02:01:36Z slyrus quit (Ping timeout: 276 seconds)
2016-08-18T02:02:31Z shka joined #lisp
2016-08-18T02:13:28Z rpg: anyone know if there's a way to get a backtrace out of SLDB? I'd like one so that I can file a bug report....
2016-08-18T02:14:11Z rpg: I mean, I have the buffer, but I'd rather dump the backtrace to a file.
2016-08-18T02:14:28Z arescorpio quit (Read error: Connection reset by peer)
2016-08-18T02:15:36Z Bike: well, i know how to get the backtrace from the implementation, but i don't know if sldb has an export
2016-08-18T02:17:04Z rpg: Bike: Tried to call (asdf:raw-print-backtrace...) and boom! new error and my debugger context is gone :-(
2016-08-18T02:17:30Z rpg sometimes really wishes he could disable sldb.
2016-08-18T02:17:43Z Bike: you uh, can, can't you?
2016-08-18T02:17:53Z rpg: Bike: You can, but then SLIME breaks horribly.
2016-08-18T02:18:01Z Bike: huh. weird.
2016-08-18T02:18:34Z rpg: AFAICT using SLIME with SLDB disabled just doesn't work. I suspect that not all people assumed that sldb could be not there, so there are dependencies on it, even if it's disabled.
2016-08-18T02:18:50Z Bike: well, i was able to call uiop/image:raw-print-backtrace fine. from the debugger frame, of course.
2016-08-18T02:19:40Z rpg: Bike: I typoed somehow, trying to dump it to a file, got a file error, and that somehow blew away the SLDB context. So it's very useful to know it worked for you.
2016-08-18T02:19:57Z Bike: ah. sorry.
2016-08-18T02:20:27Z rpg: No need! It's super-helpful to know that *I* messed up, and that it's actually possible. So I know to keep trying.
2016-08-18T02:20:36Z Bike: oh. i thought you were being sarcastic.
2016-08-18T02:20:39Z Bike: you could add an emacs function to do the call for you?
2016-08-18T02:20:46Z rpg: No! On the contrary!
2016-08-18T02:21:04Z Jesin quit (Quit: Leaving)
2016-08-18T02:21:45Z prole quit (Remote host closed the connection)
2016-08-18T02:21:59Z rpg: wow! exciting! bactrace.txt is 547.9 Mb!
2016-08-18T02:22:28Z Bike: impressive
2016-08-18T02:22:43Z algae quit (Quit: leaving)
2016-08-18T02:22:58Z rpg: And it's growing!
2016-08-18T02:23:03Z rpg: Now it's 7.1Gb
2016-08-18T02:23:06Z Bike: that seems like it would blow most control stacks i've ever used
2016-08-18T02:23:08Z Bike: er.
2016-08-18T02:23:23Z rpg: Indeed, it blew up the lisp
2016-08-18T02:23:34Z rpg: I'll do this in the shell with rlwrap!
2016-08-18T02:24:10Z kn-928 quit (Quit: Textual IRC Client: www.textualapp.com)
2016-08-18T02:24:33Z kn-928 joined #lisp
2016-08-18T02:24:57Z Jesin joined #lisp
2016-08-18T02:25:26Z Bike: also, if you're unlucky enough to get recursive errors, you should be able to drop to earlier levels... there's a restart like [ABORT] Return to SLDB level #
2016-08-18T02:30:24Z rpg is busy being CLX troublemaker.
2016-08-18T02:32:07Z rpg: I will try to see what happens if I turn off sldb again (I tried a year or so ago, and it turned out badly, but I didn't write up the problems well enough
2016-08-18T02:42:33Z pseudan quit (Ping timeout: 240 seconds)
2016-08-18T02:43:09Z karswell joined #lisp
2016-08-18T02:45:10Z tmtwd_ quit (Ping timeout: 244 seconds)
2016-08-18T02:46:24Z kn-928 quit (Quit: Textual IRC Client: www.textualapp.com)
2016-08-18T02:48:46Z Petit_Dejeuner quit (Ping timeout: 244 seconds)
2016-08-18T02:49:36Z vlatkoB joined #lisp
2016-08-18T02:52:07Z zacharias_ joined #lisp
2016-08-18T02:54:59Z pseudan joined #lisp
2016-08-18T02:55:16Z zacharias quit (Ping timeout: 265 seconds)
2016-08-18T02:55:40Z rjnw joined #lisp
2016-08-18T03:04:54Z atgreen joined #lisp
2016-08-18T03:04:58Z jleija quit (Quit: leaving)
2016-08-18T03:07:31Z jokleinn quit (Quit: WeeChat 1.5)
2016-08-18T03:07:33Z mikaelj quit (Remote host closed the connection)
2016-08-18T03:07:41Z FreeBirdLjj joined #lisp
2016-08-18T03:08:43Z pseudan quit (Quit: Leaving)
2016-08-18T03:16:46Z joekarma joined #lisp
2016-08-18T03:20:01Z joekarma quit (Client Quit)
2016-08-18T03:20:44Z d4ryus quit (Ping timeout: 260 seconds)
2016-08-18T03:23:19Z d4ryus joined #lisp
2016-08-18T03:24:07Z karswell quit (Remote host closed the connection)
2016-08-18T03:24:24Z tmtwd_ joined #lisp
2016-08-18T03:24:42Z karswell joined #lisp
2016-08-18T03:31:31Z quazimodo joined #lisp
2016-08-18T03:31:34Z quazimod1 joined #lisp
2016-08-18T03:35:06Z IPmonger joined #lisp
2016-08-18T03:37:21Z quazimod1 quit (Ping timeout: 244 seconds)
2016-08-18T03:37:33Z rpg is now known as rpg[Away]
2016-08-18T03:37:48Z quazimodo quit (Ping timeout: 265 seconds)
2016-08-18T03:38:09Z rpg[Away] quit (Quit: My Mac has gone to sleep. ZZZzzz…)
2016-08-18T03:42:29Z kobain quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
2016-08-18T03:49:23Z bugrum joined #lisp
2016-08-18T03:54:13Z arescorpio joined #lisp
2016-08-18T03:57:01Z IPmonger quit (Ping timeout: 252 seconds)
2016-08-18T04:11:00Z IPmonger joined #lisp
2016-08-18T04:15:36Z IPmonger quit (Ping timeout: 250 seconds)
2016-08-18T04:15:57Z slyrus joined #lisp
2016-08-18T04:17:15Z IPmonger joined #lisp
2016-08-18T04:21:54Z pillton: What verb is customary for macros which are syntactic sugar for make-X functions?
2016-08-18T04:23:08Z asc232 quit (Quit: Saliendo)
2016-08-18T04:23:31Z loke__: pillton: Huh?
2016-08-18T04:23:45Z loke__: I'm trying to parse that sentence, and I _may_ have figured out what you mean.
2016-08-18T04:24:02Z loke__: But it'd be helpful if you could provide an example of verbs that you have dismissed.
2016-08-18T04:26:23Z pillton: define-X typically modifies the global environment. make-X is typically a function.
2016-08-18T04:26:46Z loke__: pillton: Right.
2016-08-18T04:26:47Z pillton: I'm staring at the theasurus now.
2016-08-18T04:27:07Z loke__: So why is define-X not acceptable to you?
2016-08-18T04:27:24Z loke__: I mean, make-X
2016-08-18T04:28:00Z pillton: Because it is typically a function. I'm writing a macro.
2016-08-18T04:28:41Z loke__: pillton: Does it matter?
2016-08-18T04:28:49Z FreeBirdLjj quit (Remote host closed the connection)
2016-08-18T04:29:06Z loke__: And also: Why is ti a macro?
2016-08-18T04:29:23Z loke__: Can you be more specific?
2016-08-18T04:35:02Z beach: Good morning everyone!
2016-08-18T04:36:56Z Bike: good morning.
2016-08-18T04:37:17Z beach: Bike: I saw you wanted to discuss type inference.
2016-08-18T04:37:22Z BlueRavenGT joined #lisp
2016-08-18T04:37:48Z Bike: ah. yeah. what's done so far? the file in hir transformations is pretty empty.
2016-08-18T04:38:01Z beach: Let me look...
2016-08-18T04:40:27Z beach: There is some stuff in the subdirectory Type-inference.
2016-08-18T04:41:43Z Bike: hm, i saw that but assumed it was too old. guess not, now that i look at the files
2016-08-18T04:43:13Z beach: In order to avoid complicated dataflow analysis, type inference has to be done after the transformation that removes lexical variables that are shared between nested functions.
2016-08-18T04:43:47Z Bike: oh, there is such a transformation? that's convenient, that was problem number two in my mind
2016-08-18T04:43:50Z tmtwd_ quit (Ping timeout: 244 seconds)
2016-08-18T04:44:08Z lexicall joined #lisp
2016-08-18T04:44:26Z beach: That transformation converts such variables to "cells".
2016-08-18T04:44:49Z beach: Initially, for type inference, it is possible to assume that cells can contain anything.
2016-08-18T04:45:01Z lexicall quit (Client Quit)
2016-08-18T04:46:37Z beach: The main difficulty with type inference is to define a lattice that is not only finite, but that also does not take too much time to compute.  So for instance, one must avoid having every 64-bit integer as member of the lattice.
2016-08-18T04:46:47Z beach: ... even though that would be finite.
2016-08-18T04:46:56Z pillton: loke__: http://paste.lisp.org/display/323522
2016-08-18T04:47:29Z arescorpio quit (Quit: Leaving.)
2016-08-18T04:48:29Z vibs29 quit (Ping timeout: 244 seconds)
2016-08-18T04:49:27Z defaultxr quit (Quit: gnight)
2016-08-18T04:51:54Z loke__: pillton: How about BUILD-TABLE?
2016-08-18T04:51:55Z Bike: i figured for straight line code with conditionals it doesn't really matter how complex the type system is, but with loops and recursive functions you start having to compute fixed points and that gets impossible quickly
2016-08-18T04:52:13Z beach: Right.
2016-08-18T04:54:31Z vibs29 joined #lisp
2016-08-18T04:56:38Z tmtwd_ joined #lisp
2016-08-18T04:58:09Z vaitel quit (Ping timeout: 260 seconds)
2016-08-18T04:59:15Z pillton: loke__: That would work. I was wondering if there was a word which shouts "macro" like define-X and with-Y do.
2016-08-18T05:01:01Z eSVG quit (Ping timeout: 250 seconds)
2016-08-18T05:01:09Z Bike: well, i see this is mostly the same idea that i was prototyping. only so many ways to skin a cat slash traverse a graph
2016-08-18T05:04:04Z FreeBirdLjj joined #lisp
2016-08-18T05:04:10Z beach: Yes, the algorithm itself is pretty straightforward.
2016-08-18T05:07:05Z IPmonger quit (Ping timeout: 249 seconds)
2016-08-18T05:10:25Z Bike: i'm not sure how it's supposed to go afterwards, though? like, once you have a type for each arc, that opens up many optimizations... it would be nice to keep information around.
2016-08-18T05:11:29Z beach: It can be kept in a hash table.
2016-08-18T05:11:34Z nullman quit (Ping timeout: 265 seconds)
2016-08-18T05:11:37Z beach: Not necessarily in the graph itself.
2016-08-18T05:13:13Z beach: Arcs where at least one variable has the type NIL can never be taken.
2016-08-18T05:13:52Z beach: So a branching instruction with one such successor can be replaced by a NOP or some other appropriate instruction.
2016-08-18T05:14:40Z mathi_aihtam joined #lisp
2016-08-18T05:14:57Z Harag1 joined #lisp
2016-08-18T05:16:07Z Harag quit (Read error: Connection reset by peer)
2016-08-18T05:16:07Z Harag1 is now known as Harag
2016-08-18T05:16:11Z mvilleneuve joined #lisp
2016-08-18T05:16:16Z oleo_ quit (Quit: Leaving)
2016-08-18T05:18:57Z angavrilov joined #lisp
2016-08-18T05:20:41Z rjnw quit (Quit: Connection closed for inactivity)
2016-08-18T05:23:26Z bocaneri joined #lisp
2016-08-18T05:23:32Z mathi_aihtam quit (Quit: mathi_aihtam)
2016-08-18T05:24:06Z mathi_aihtam joined #lisp
2016-08-18T05:24:16Z nikki93 joined #lisp
2016-08-18T05:26:03Z BlueRavenGT quit (Ping timeout: 240 seconds)
2016-08-18T05:29:39Z Grue`` joined #lisp
2016-08-18T05:34:50Z eSVG joined #lisp
2016-08-18T05:35:40Z puchacz joined #lisp
2016-08-18T05:37:29Z Harag quit (Quit: Harag)
2016-08-18T05:43:32Z mishoo joined #lisp
2016-08-18T05:46:11Z mvilleneuve quit (Quit: This computer has gone to sleep)
2016-08-18T05:46:27Z Largeaux joined #lisp
2016-08-18T05:46:47Z nikki93 quit (Remote host closed the connection)
2016-08-18T05:47:16Z nikki93 joined #lisp
2016-08-18T05:51:40Z Bike: beach: for typeq, i get why the first successor has meet(type, input-type), but why does the second have the join? shouldn't it be more like meet(complement(type), input-type)?
2016-08-18T05:51:44Z milanj quit (Quit: This computer has gone to sleep)
2016-08-18T05:52:21Z smokeink quit (Ping timeout: 276 seconds)
2016-08-18T05:52:43Z mathi_aihtam quit (Quit: mathi_aihtam)
2016-08-18T05:53:39Z mathi_aihtam joined #lisp
2016-08-18T05:54:37Z tmtwd_ quit (Ping timeout: 244 seconds)
2016-08-18T05:54:52Z beach: Bike: That sounds right.  I don't remember the details.
2016-08-18T05:54:57Z beach: Bike: Sorry! :(
2016-08-18T05:55:42Z nullman joined #lisp
2016-08-18T05:55:46Z Bike: it's fine. pretty much intending to figure out how it works by breaking it, here.
2016-08-18T05:56:10Z mathi_aihtam quit (Client Quit)
2016-08-18T05:56:39Z Bike: adding complement kind of destroys the descriptor thing, so that's a mite unfortunate
2016-08-18T06:01:08Z beach: In what way does it destroy it?
2016-08-18T06:01:41Z Kaisyu quit (Ping timeout: 250 seconds)
2016-08-18T06:02:38Z Bike: well... i guess it doesn't. it makes the alternate branches of typeq and fixnump and such useless, but i guess that can be lived with.
2016-08-18T06:03:27Z tmtwd_ joined #lisp
2016-08-18T06:03:29Z angavrilov quit (Ping timeout: 260 seconds)
2016-08-18T06:04:15Z stepnem joined #lisp
2016-08-18T06:04:37Z Bike: anyway, i think i'm going to try adding function and values types. i think you need the latter for the and typeq to be totally correct, anyway
2016-08-18T06:04:49Z FreeBirdLjj quit (Remote host closed the connection)
2016-08-18T06:04:57Z beach: OK.
2016-08-18T06:05:34Z beach: HIR has special data types for multiple values.
2016-08-18T06:06:33Z Bike: values-location, right. i think the inferencer currently restricts itself to lexical locations
2016-08-18T06:06:48Z beach: I think that's right, yes.
2016-08-18T06:07:42Z FreeBirdLjj joined #lisp
2016-08-18T06:07:57Z pok quit (Quit: leaving)
2016-08-18T06:08:17Z pok joined #lisp
2016-08-18T06:08:22Z beach: Bike: I'll try to look at that code later today, and we can talk about it again later.
2016-08-18T06:08:30Z Bike: ok, sure.
2016-08-18T06:08:58Z beach: But now I need to go do some other work.
2016-08-18T06:09:10Z Bike: the fairly arbitrary goal i decided on is letting multiple-value-bind not enclose, so i need to mess with a lot of stuff anyway
2016-08-18T06:09:18Z Bike: seeya later
2016-08-18T06:09:26Z beach: Good luck!
2016-08-18T06:09:29Z beach left #lisp
2016-08-18T06:11:27Z eivarv joined #lisp
2016-08-18T06:14:46Z angavrilov joined #lisp
2016-08-18T06:22:13Z lnostdal_ joined #lisp
2016-08-18T06:22:15Z shka quit (Ping timeout: 276 seconds)
2016-08-18T06:24:49Z pareidolia joined #lisp
2016-08-18T06:25:19Z puchacz quit (Quit: Konversation terminated!)
2016-08-18T06:28:50Z mvilleneuve joined #lisp
2016-08-18T06:32:17Z eivarv quit (Quit: Sleep)
2016-08-18T06:38:17Z FreeBirdLjj quit (Remote host closed the connection)
2016-08-18T06:40:30Z davsebamse quit (Ping timeout: 265 seconds)
2016-08-18T06:40:45Z flamebeard joined #lisp
2016-08-18T06:42:16Z davsebamse joined #lisp
2016-08-18T06:47:30Z nikki93__ joined #lisp
2016-08-18T06:48:15Z nikki93 quit (Read error: Connection reset by peer)
2016-08-18T06:51:08Z bocaneri quit (Ping timeout: 265 seconds)
2016-08-18T06:53:08Z FreeBirdLjj joined #lisp
2016-08-18T06:57:00Z jstewart777 joined #lisp
2016-08-18T06:58:14Z mbuf joined #lisp
2016-08-18T06:59:08Z u03bb joined #lisp
2016-08-18T07:01:19Z eivarv joined #lisp
2016-08-18T07:05:10Z nikki93__ quit (Remote host closed the connection)
2016-08-18T07:05:39Z nikki93 joined #lisp
2016-08-18T07:05:40Z Petit_Dejeuner joined #lisp
2016-08-18T07:06:37Z FreeBirdLjj quit (Remote host closed the connection)
2016-08-18T07:07:10Z FreeBirdLjj joined #lisp
2016-08-18T07:07:25Z gravicappa joined #lisp
2016-08-18T07:08:24Z voidlily quit (Ping timeout: 276 seconds)
2016-08-18T07:09:24Z yeticry quit (Read error: Connection reset by peer)
2016-08-18T07:10:04Z nikki93 quit (Ping timeout: 252 seconds)
2016-08-18T07:10:41Z shka joined #lisp
2016-08-18T07:10:50Z FreeBird_ joined #lisp
2016-08-18T07:10:53Z FreeBirdLjj quit (Read error: Connection reset by peer)
2016-08-18T07:11:09Z Grue`` quit (Ping timeout: 260 seconds)
2016-08-18T07:12:18Z tmtwd_ quit (Ping timeout: 258 seconds)
2016-08-18T07:12:38Z arduo quit (Read error: Connection reset by peer)
2016-08-18T07:13:12Z eivarv quit (Quit: Sleep)
2016-08-18T07:14:11Z zacharias_ quit (Ping timeout: 244 seconds)
2016-08-18T07:14:49Z FreeBirdLjj joined #lisp
2016-08-18T07:14:52Z FreeBird_ quit (Ping timeout: 240 seconds)
2016-08-18T07:15:43Z yeticry joined #lisp
2016-08-18T07:18:09Z davsebamse quit (Ping timeout: 276 seconds)
2016-08-18T07:18:32Z nikki93 joined #lisp
2016-08-18T07:19:17Z davsebamse joined #lisp
2016-08-18T07:20:08Z nikki93 quit (Remote host closed the connection)
2016-08-18T07:20:28Z bugrum quit (Remote host closed the connection)
2016-08-18T07:20:45Z nikki93 joined #lisp
2016-08-18T07:21:11Z edgar-rft joined #lisp
2016-08-18T07:24:41Z Atarian: Anyone familiar with cl-who?
2016-08-18T07:25:36Z FreeBird_ joined #lisp
2016-08-18T07:25:55Z Atarian: I want to return one div if a value is true, and another if the value is false
2016-08-18T07:26:26Z Atarian: But it doesn't recognise the :div tag inside the cond
2016-08-18T07:27:54Z Bike: i think you need htm? http://weitz.de/cl-who/#syntax
2016-08-18T07:28:01Z Largeaux quit (Ping timeout: 258 seconds)
2016-08-18T07:28:30Z lexicall joined #lisp
2016-08-18T07:29:16Z FreeBirdLjj quit (Ping timeout: 264 seconds)
2016-08-18T07:30:56Z Atarian: Bike: that looks like it, thanks
2016-08-18T07:31:28Z lexicall quit (Client Quit)
2016-08-18T07:34:42Z Atarian: I'm still getting htm is not a symbol or lambda expression...
2016-08-18T07:37:31Z Bike: paste your code?
2016-08-18T07:39:25Z Munksgaard joined #lisp
2016-08-18T07:40:00Z mathi_aihtam joined #lisp
2016-08-18T07:40:52Z mathi_aihtam quit (Client Quit)
2016-08-18T07:44:56Z FreeBird_ quit (Read error: Connection reset by peer)
2016-08-18T07:45:47Z lnostdal_ quit (Read error: Connection reset by peer)
2016-08-18T07:48:41Z FreeBirdLjj joined #lisp
2016-08-18T07:53:33Z bdr3552 joined #lisp
2016-08-18T07:53:41Z FreeBirdLjj quit (Read error: Connection reset by peer)
2016-08-18T07:54:44Z FreeBirdLjj joined #lisp
2016-08-18T07:56:10Z FreeBird_ joined #lisp
2016-08-18T07:57:57Z impaktor: I've installed GSLL (am a CL noob). Now I want to read in matrices for multiplying, inverting, and doing some function fitting to data. Big question is: can I just use CL vectors for this? It doesn't seem like gsl_matrix_alloc is available in GSLL (gsll:gsl-lookup "gsl_matrix_alloc") --> NIL, so I'm having trouble finding what is meant by a "matrix" in GSLL.
2016-08-18T07:59:14Z ante joined #lisp
2016-08-18T07:59:39Z knicklux joined #lisp
2016-08-18T07:59:46Z FreeBirdLjj quit (Ping timeout: 265 seconds)
2016-08-18T08:01:42Z davsebamse quit (Ping timeout: 265 seconds)
2016-08-18T08:01:43Z Bike: docs say a few things. and that there are examples if you call #'examples.
2016-08-18T08:02:20Z ante is now known as Cymew
2016-08-18T08:03:24Z davsebamse joined #lisp
2016-08-18T08:03:29Z hhdave joined #lisp
2016-08-18T08:03:47Z Bike: https://gitlab.common-lisp.net/antik/gsll/blob/master/series-acceleration.lisp#L91-118 maybe this grid:make-foreign-array bit.
2016-08-18T08:03:55Z impaktor: Bike: Yeah, I've looked at some that define some kind of unit matrix. I guess that's the way to go, and then all my data would be in some GSLL format, rather than CL vector.
2016-08-18T08:04:18Z Bike: well, gsll just calls gsl, it makes sense that they'd lean on foreign arrays
2016-08-18T08:05:48Z impaktor: Yeah, but that makes it stranger to me that there are no method to read them in, that I can tell.
2016-08-18T08:07:20Z hhdave_ joined #lisp
2016-08-18T08:08:12Z hhdave quit (Ping timeout: 276 seconds)
2016-08-18T08:08:12Z hhdave_ is now known as hhdave
2016-08-18T08:08:16Z varjag joined #lisp
2016-08-18T08:12:07Z Bike quit (Quit: leaving)
2016-08-18T08:13:31Z pierpa joined #lisp
2016-08-18T08:18:44Z nikki93 quit (Read error: Connection reset by peer)
2016-08-18T08:18:54Z nikki93__ joined #lisp
2016-08-18T08:18:59Z zacharias joined #lisp
2016-08-18T08:21:28Z attila_lendvai joined #lisp
2016-08-18T08:21:28Z attila_lendvai quit (Changing host)
2016-08-18T08:21:28Z attila_lendvai joined #lisp
2016-08-18T08:23:26Z warweasle joined #lisp
2016-08-18T08:24:44Z warweasle quit (Remote host closed the connection)
2016-08-18T08:29:10Z impaktor: Hmm, so GSLL uses this Antik thingy. Antik claims to have a "detailed manual", link is sadly 404. But it looks like one should use (GRID:MAKE-FOREIGN-ARRAY :initial-contents '(...)). (I assume "GRID" is some package in GSLL or Antik?)
2016-08-18T08:29:37Z mastokley quit (Ping timeout: 244 seconds)
2016-08-18T08:31:06Z cibs quit (Ping timeout: 240 seconds)
2016-08-18T08:33:14Z cibs joined #lisp
2016-08-18T08:36:37Z mbuf quit (Quit: Ex-Chat)
2016-08-18T08:38:28Z warweasle joined #lisp
2016-08-18T08:39:14Z attila_lendvai quit (Ping timeout: 260 seconds)
2016-08-18T08:44:09Z eivarv joined #lisp
2016-08-18T08:45:37Z FreeBird_ quit (Remote host closed the connection)
2016-08-18T08:46:30Z SumoSudo joined #lisp
2016-08-18T08:51:03Z Karl_Dscc joined #lisp
2016-08-18T08:54:32Z DeadTrickster joined #lisp
2016-08-18T08:55:37Z edgar-rft: impaktor: after (ql:quickload "antik") the manual is in ~/quicklisp/dists/quicklisp/software/antik-20131003-git/documentation
2016-08-18T08:56:12Z FreeBirdLjj joined #lisp
2016-08-18T08:59:20Z Lexicall joined #lisp
2016-08-18T08:59:21Z eSVG quit (Ping timeout: 250 seconds)
2016-08-18T09:01:56Z Sigyn quit (Quit: I'll Be Back)
2016-08-18T09:02:26Z Sigyn joined #lisp
2016-08-18T09:02:48Z ASau quit (Ping timeout: 276 seconds)
2016-08-18T09:04:33Z ramky joined #lisp
2016-08-18T09:06:31Z moei quit (Quit: Leaving...)
2016-08-18T09:06:56Z eivarv quit (Quit: Sleep)
2016-08-18T09:08:09Z mbrock joined #lisp
2016-08-18T09:08:12Z nostoi joined #lisp
2016-08-18T09:09:52Z davsebamse quit (Ping timeout: 240 seconds)
2016-08-18T09:11:20Z impaktor: edgar-rft: thanks.
2016-08-18T09:12:00Z davsebamse joined #lisp
2016-08-18T09:12:40Z milanj joined #lisp
2016-08-18T09:15:20Z przl joined #lisp
2016-08-18T09:16:44Z nikki93__ quit (Remote host closed the connection)
2016-08-18T09:16:56Z nikki93 joined #lisp
2016-08-18T09:21:20Z Karl_Dscc quit (Remote host closed the connection)
2016-08-18T09:22:03Z ASau joined #lisp
2016-08-18T09:22:35Z FreeBirdLjj quit (Remote host closed the connection)
2016-08-18T09:35:00Z m00natic joined #lisp
2016-08-18T09:37:15Z CEnnis91 quit (Quit: Connection closed for inactivity)
2016-08-18T09:38:25Z lnostdal_ joined #lisp
2016-08-18T09:39:18Z przl quit (Ping timeout: 250 seconds)
2016-08-18T09:43:30Z Karl_Dscc joined #lisp
2016-08-18T09:43:42Z test1600 joined #lisp
2016-08-18T09:47:25Z przl joined #lisp
2016-08-18T09:50:49Z FreeBirdLjj joined #lisp
2016-08-18T09:51:47Z SumoSudo quit (Ping timeout: 250 seconds)
2016-08-18T09:55:33Z FreeBirdLjj quit (Ping timeout: 244 seconds)
2016-08-18T09:56:34Z lnostdal_ quit (Quit: lnostdal_)
2016-08-18T09:57:30Z ASau quit (Ping timeout: 250 seconds)
2016-08-18T09:58:31Z gargaml quit (Quit: WeeChat 1.5)
2016-08-18T09:58:32Z eivarv joined #lisp
2016-08-18T09:59:18Z mathi_aihtam joined #lisp
2016-08-18T10:12:28Z HeyFlash joined #lisp
2016-08-18T10:13:12Z FreeBirdLjj joined #lisp
2016-08-18T10:13:53Z warweasle quit (Remote host closed the connection)
2016-08-18T10:15:23Z nikki93__ joined #lisp
2016-08-18T10:16:09Z nikki93 quit (Read error: Connection reset by peer)
2016-08-18T10:19:07Z FreeBirdLjj quit (Remote host closed the connection)
2016-08-18T10:34:03Z nostoi quit (Quit: Verlassend.)
2016-08-18T10:37:46Z eivarv quit (Quit: Sleep)
2016-08-18T10:41:56Z eivarv joined #lisp
2016-08-18T10:53:50Z eivarv quit (Quit: Sleep)
2016-08-18T10:59:05Z Grue`` joined #lisp
2016-08-18T10:59:10Z Grue`` is now known as Grue`
2016-08-18T11:01:09Z przl quit (Ping timeout: 244 seconds)
2016-08-18T11:03:27Z Atarian quit (Remote host closed the connection)
2016-08-18T11:04:49Z `lain quit (Quit: bye)
2016-08-18T11:08:06Z Atarian joined #lisp
2016-08-18T11:08:22Z scymtym joined #lisp
2016-08-18T11:09:50Z sjl joined #lisp
2016-08-18T11:10:06Z FreeBirdLjj joined #lisp
2016-08-18T11:11:38Z shdeng quit (Quit: Leaving)
2016-08-18T11:12:52Z nikki93__ quit (Remote host closed the connection)
2016-08-18T11:13:00Z smokeink joined #lisp
2016-08-18T11:13:20Z nikki93 joined #lisp
2016-08-18T11:17:33Z nikki93 quit (Ping timeout: 240 seconds)
2016-08-18T11:17:53Z FreeBirdLjj quit (Remote host closed the connection)
2016-08-18T11:26:39Z przl joined #lisp
2016-08-18T11:32:18Z przl quit (Ping timeout: 276 seconds)
2016-08-18T11:32:21Z sulky quit (Remote host closed the connection)
2016-08-18T11:33:40Z mvilleneuve quit (Quit: This computer has gone to sleep)
2016-08-18T11:36:37Z ggole joined #lisp
2016-08-18T11:39:17Z EvW joined #lisp
2016-08-18T11:42:25Z dmitigr joined #lisp
2016-08-18T11:47:44Z mathi_aihtam quit (Quit: mathi_aihtam)
2016-08-18T11:49:24Z SumoSudo joined #lisp
2016-08-18T11:49:27Z sweater joined #lisp
2016-08-18T11:51:25Z ovenpasta joined #lisp
2016-08-18T11:52:44Z mvilleneuve joined #lisp
2016-08-18T11:57:28Z ovenpasta quit (Ping timeout: 244 seconds)
2016-08-18T11:59:33Z FreeBirdLjj joined #lisp
2016-08-18T12:00:05Z rpg joined #lisp
2016-08-18T12:00:21Z ksool joined #lisp
2016-08-18T12:00:23Z ksool quit (Max SendQ exceeded)
2016-08-18T12:00:35Z ksool joined #lisp
2016-08-18T12:01:15Z gilez joined #lisp
2016-08-18T12:01:42Z dyelar joined #lisp
2016-08-18T12:03:17Z gravicappa quit (Ping timeout: 265 seconds)
2016-08-18T12:04:12Z DGASAU quit (Read error: Connection reset by peer)
2016-08-18T12:04:51Z dyelar quit (Client Quit)
2016-08-18T12:06:08Z DGASAU joined #lisp
2016-08-18T12:07:57Z DGASAU quit (Read error: Connection reset by peer)
2016-08-18T12:09:33Z DGASAU joined #lisp
2016-08-18T12:13:03Z nullman quit (Ping timeout: 264 seconds)
2016-08-18T12:13:54Z nikki93 joined #lisp
2016-08-18T12:14:39Z nullman joined #lisp
2016-08-18T12:16:03Z DGASAU quit (Ping timeout: 264 seconds)
2016-08-18T12:16:36Z DeadTrickster quit (Ping timeout: 244 seconds)
2016-08-18T12:16:58Z Lexicall quit (Ping timeout: 252 seconds)
2016-08-18T12:17:39Z DeadTrickster joined #lisp
2016-08-18T12:18:30Z nikki93 quit (Ping timeout: 244 seconds)
2016-08-18T12:19:26Z rpg quit (Quit: My Mac has gone to sleep. ZZZzzz…)
2016-08-18T12:21:35Z asc232 joined #lisp
2016-08-18T12:23:33Z warweasle joined #lisp
2016-08-18T12:26:46Z EvW quit (Ping timeout: 244 seconds)
2016-08-18T12:27:10Z iqbalansari joined #lisp
2016-08-18T12:27:17Z FreeBirdLjj quit (Remote host closed the connection)
2016-08-18T12:27:53Z Lexicall joined #lisp
2016-08-18T12:28:10Z Atarian quit (Remote host closed the connection)
2016-08-18T12:28:42Z asc232 quit (Remote host closed the connection)
2016-08-18T12:30:34Z Atarian joined #lisp
2016-08-18T12:35:51Z DGASAU joined #lisp
2016-08-18T12:39:50Z kolko quit (Read error: Connection reset by peer)
2016-08-18T12:40:57Z dmitigr quit (Remote host closed the connection)
2016-08-18T12:41:22Z smokeink quit (Read error: Connection reset by peer)
2016-08-18T12:41:26Z kolko joined #lisp
2016-08-18T12:43:53Z nalik joined #lisp
2016-08-18T12:46:14Z smokeink joined #lisp
2016-08-18T12:52:52Z elimik31 joined #lisp
2016-08-18T12:55:46Z eivarv joined #lisp
2016-08-18T12:56:19Z jokleinn joined #lisp
2016-08-18T12:58:51Z DGASAU quit (Read error: Connection reset by peer)
2016-08-18T13:00:10Z DGASAU joined #lisp
2016-08-18T13:01:32Z `lain joined #lisp
2016-08-18T13:05:31Z przl joined #lisp
2016-08-18T13:10:45Z sweater quit (Read error: Connection reset by peer)
2016-08-18T13:11:02Z sweater joined #lisp
2016-08-18T13:13:49Z CEnnis91 joined #lisp
2016-08-18T13:16:28Z impaktor quit (Ping timeout: 258 seconds)
2016-08-18T13:17:10Z ovenpasta joined #lisp
2016-08-18T13:18:32Z sjl: is the best way to (fill ...) a multidimensional array to make a 1d array displaced to it and fill that?
2016-08-18T13:19:39Z davsebamse quit (Ping timeout: 265 seconds)
2016-08-18T13:21:00Z impaktor joined #lisp
2016-08-18T13:21:00Z eivarv quit (Quit: Sleep)
2016-08-18T13:21:18Z davsebamse joined #lisp
2016-08-18T13:23:29Z varjag: probably.. or use :initial-element if it's for initialization
2016-08-18T13:23:49Z pierpa: sjl: or maybe a LOOP using ROW-MAJOR-AREF
2016-08-18T13:24:06Z sjl: varjag: not for init unfortunately
2016-08-18T13:24:15Z aphprentice quit (Quit: Connection closed for inactivity)
2016-08-18T13:24:30Z sjl: pierpa: loop would work, but I think a compiler might be able to optimize it better if I use fill
2016-08-18T13:24:40Z varjag: row-major-aref is also a displacement essentially
2016-08-18T13:24:50Z pierpa: no
2016-08-18T13:24:52Z pierpa: (loop for i below (array-total-size a) do (setf (row-major-aref a i) init))
2016-08-18T13:25:20Z varjag: clhs row-major-aref
2016-08-18T13:25:20Z specbot: http://www.lispworks.com/reference/HyperSpec/Body/f_row_ma.htm
2016-08-18T13:25:30Z varjag: see notes
2016-08-18T13:25:32Z gargaml joined #lisp
2016-08-18T13:26:23Z pierpa: that's an explanation of what it does
2016-08-18T13:26:38Z pierpa: of course there's no displacement involved
2016-08-18T13:26:57Z pierpa: there's direct access to the array storage
2016-08-18T13:27:24Z atgreen quit (Ping timeout: 260 seconds)
2016-08-18T13:27:42Z IPmonger joined #lisp
2016-08-18T13:28:23Z LiamH joined #lisp
2016-08-18T13:28:42Z eSVG joined #lisp
2016-08-18T13:29:23Z oleo joined #lisp
2016-08-18T13:29:24Z oleo quit (Changing host)
2016-08-18T13:29:24Z oleo joined #lisp
2016-08-18T13:31:03Z happy-dude joined #lisp
2016-08-18T13:31:31Z eivarv joined #lisp
2016-08-18T13:31:32Z cmack joined #lisp
2016-08-18T13:31:35Z jerme joined #lisp
2016-08-18T13:32:33Z IPmonger quit (Ping timeout: 276 seconds)
2016-08-18T13:34:25Z przl quit (Read error: Connection reset by peer)
2016-08-18T13:34:31Z eSVG quit (Ping timeout: 250 seconds)
2016-08-18T13:34:39Z lambda-smith joined #lisp
2016-08-18T13:37:26Z IPmonger joined #lisp
2016-08-18T13:39:30Z kobain joined #lisp
2016-08-18T13:41:12Z przl joined #lisp
2016-08-18T13:42:39Z sellout- joined #lisp
2016-08-18T13:42:43Z sweater quit (Read error: Connection reset by peer)
2016-08-18T13:43:04Z lambda-smith quit (Quit: Konversation terminated!)
2016-08-18T13:43:19Z elimik31 quit (Ping timeout: 252 seconds)
2016-08-18T13:43:36Z al-damiri joined #lisp
2016-08-18T13:44:10Z prole joined #lisp
2016-08-18T13:45:02Z sjl: okay so yeah, fill is much faster http://paste.lisp.org/display/323574
2016-08-18T13:45:08Z sjl: at least on sbcl
2016-08-18T13:47:54Z sjl: pierpa: varjag: ^^
2016-08-18T13:49:13Z sjl: unfortunately I don't think sbcl can stack-allocate that intermediate displaced array, but even with the consing it's way faster than the loop
2016-08-18T13:49:26Z dyelar joined #lisp
2016-08-18T13:51:24Z przl quit (Ping timeout: 276 seconds)
2016-08-18T13:52:39Z mason- joined #lisp
2016-08-18T13:53:28Z eivarv quit (Quit: Sleep)
2016-08-18T13:53:28Z cmack quit (Remote host closed the connection)
2016-08-18T13:53:45Z cmack joined #lisp
2016-08-18T13:54:06Z vaitel joined #lisp
2016-08-18T13:55:30Z eivarv joined #lisp
2016-08-18T13:55:34Z IPmonger quit (Ping timeout: 258 seconds)
2016-08-18T14:01:15Z przl joined #lisp
2016-08-18T14:02:21Z Karl_Dscc quit (Remote host closed the connection)
2016-08-18T14:03:14Z pierpa: sjl: of course is slower. You didn't declare any type in the loop version!
2016-08-18T14:03:23Z iqbalansari quit (Remote host closed the connection)
2016-08-18T14:03:51Z cromachina quit (Read error: Connection reset by peer)
2016-08-18T14:04:04Z pierpa: well, ok, at least try declare i fixnum, and compile it with safety & debug 0
2016-08-18T14:04:05Z milanj quit (Quit: This computer has gone to sleep)
2016-08-18T14:04:46Z pierpa: FILL is certainly compiled with these declarations in effect, so for fairness the loop should do too
2016-08-18T14:05:40Z mason-: Hi all. Is there a preferred library for numerical linear algebra?
2016-08-18T14:06:07Z mason-: lisp-matrix looks the most up-to-date on github, but still no commits in over a year.
2016-08-18T14:07:40Z eivarv quit (Quit: Sleep)
2016-08-18T14:08:49Z fourier joined #lisp
2016-08-18T14:08:54Z fourier quit (Changing host)
2016-08-18T14:08:55Z fourier joined #lisp
2016-08-18T14:09:34Z rpg joined #lisp
2016-08-18T14:09:45Z sjl: pierpa: fill is still far faster: http://paste.lisp.org/display/323574#1
2016-08-18T14:09:48Z test1600 quit (Quit: Leaving)
2016-08-18T14:10:11Z sjl: even with type declaring the loop index properly
2016-08-18T14:11:06Z sjl: doesn't actually make much of a different, I wonder if SBCL inferred the type of (array-total-size)
2016-08-18T14:12:50Z przl quit (Ping timeout: 244 seconds)
2016-08-18T14:12:52Z sjl: hmm, let me run this in a fresh image
2016-08-18T14:13:51Z pierpa: sjl: ok, now we know which is faster :)
2016-08-18T14:13:53Z elimik31 joined #lisp
2016-08-18T14:14:18Z pierpa: you could look at the source of fill to learn what trick they use to make it so fast, and then reports to us :)
2016-08-18T14:14:27Z pierpa: -s
2016-08-18T14:15:36Z nikki93 joined #lisp
2016-08-18T14:17:29Z jdz: sjl: it looks like you're not telling the compiler what type the array in fill-loop is, so it must use a generic version of ROW-MAJOR-AREF.
2016-08-18T14:19:55Z nikki93 quit (Ping timeout: 244 seconds)
2016-08-18T14:20:37Z pierpa: TBH, it doesn't tell this to FILL  either
2016-08-18T14:21:33Z jdz: Yes, but I'd guess FILL would figure that out once, and then use whatever specific filling routine it has.  In fill-loop case this has to be done on every iteration.
2016-08-18T14:21:39Z sjl: pierpa: yeah but fill can do it once at the start
2016-08-18T14:21:54Z pierpa: yes
2016-08-18T14:22:00Z sjl: okay in a fresh image, compiled with those optimizations, and type hinting the ftype to say that it's an array
2016-08-18T14:22:08Z sjl: loop is 13s, fill is 8s
2016-08-18T14:22:13Z sjl: +/- 1
2016-08-18T14:23:07Z jdz: sjl: try adding (check-type array (simple-array t (* * ))) at the start of fill-loop.
2016-08-18T14:23:26Z sjl: jdz: it should work for n-dimensional arrays
2016-08-18T14:23:28Z fourier quit (Ping timeout: 250 seconds)
2016-08-18T14:23:37Z sjl: so (simple-array t *)?
2016-08-18T14:24:01Z jdz: It's a two-dimensional array, right?  So it's (simple-array t (* *))
2016-08-18T14:24:06Z Karl_Dscc joined #lisp
2016-08-18T14:24:12Z sjl: I mean, in my particular case it is
2016-08-18T14:24:26Z sjl: but in general one should be able to fill n-dimensional arrays
2016-08-18T14:24:34Z atgreen joined #lisp
2016-08-18T14:24:48Z pierpa: so, the FILL version is both faster and more general
2016-08-18T14:24:49Z mason- quit (Quit: Using Circe, the loveliest of all IRC clients)
2016-08-18T14:24:54Z jdz: not for me
2016-08-18T14:25:04Z KaliLinuxGR quit (Changing host)
2016-08-18T14:25:04Z KaliLinuxGR joined #lisp
2016-08-18T14:25:19Z sjl: pierpa: not more general, the loop version will work with n-dimensional arrays too
2016-08-18T14:25:31Z pierpa: yes, sorry. they are equally general
2016-08-18T14:25:53Z jdz: for 1000 iterations test-fill is 1.063 seconds and test-loop is 0.472 seconds (0 bytes consed)
2016-08-18T14:26:07Z cagomez joined #lisp
2016-08-18T14:26:15Z pierpa: hmmm
2016-08-18T14:26:28Z sjl: jdz: how big of an array?
2016-08-18T14:26:45Z sjl: the overhead of creating the displaced array probably makes the fill version slower for small arrays
2016-08-18T14:26:46Z jdz: The same array, I just reduced the iteration count in test functions.
2016-08-18T14:26:50Z mbrock quit (Quit: Connection closed for inactivity)
2016-08-18T14:27:57Z sjl: fill is still faster for me with 1k iterations
2016-08-18T14:28:02Z sjl: what lisp are you in?
2016-08-18T14:28:19Z jdz: OK, with original 10k iterations it's like this for me: test-fill -- 11.257, test-loop: 4.717.
2016-08-18T14:29:17Z jdz: With safety restricted to 1, and debug to 2.
2016-08-18T14:29:23Z gargaml quit (Quit: WeeChat 1.5)
2016-08-18T14:29:52Z jdz: Anyway, the point is that for me the loop version is faster than fill version, and no consing.
2016-08-18T14:30:34Z gargaml joined #lisp
2016-08-18T14:30:43Z pierpa: btw, in ccl they are about the same speed
2016-08-18T14:30:59Z sjl: and here it's 2.2 for loop and .8 for fill, on speed 3 debug 2 safety 1
2016-08-18T14:31:01Z sjl: sbcl 1.3.6
2016-08-18T14:33:42Z gargaml quit (Client Quit)
2016-08-18T14:35:14Z Munksgaard quit (Quit: Leaving.)
2016-08-18T14:35:40Z adolf_stalin joined #lisp
2016-08-18T14:38:05Z lnostdal joined #lisp
2016-08-18T14:38:53Z ovenpasta quit (Ping timeout: 258 seconds)
2016-08-18T14:39:51Z scymtym: sjl: if you don't care about portable code, try (fill (sb-ext:array-storage-vector (the (array t) array)) item)
2016-08-18T14:41:29Z IPmonger joined #lisp
2016-08-18T14:41:36Z yrk joined #lisp
2016-08-18T14:41:43Z KaliLinuxGR quit (Quit: Keep Hacking)
2016-08-18T14:42:09Z yrk quit (Changing host)
2016-08-18T14:42:10Z yrk joined #lisp
2016-08-18T14:42:13Z rpg quit (Ping timeout: 250 seconds)
2016-08-18T14:42:23Z sjl: scymtym: not sure I want to go quite THAT far, but I'll try it for fun :)
2016-08-18T14:42:28Z sjl: here's the other results btw http://paste.lisp.org/display/323574#2
2016-08-18T14:42:32Z KaliLinuxGR joined #lisp
2016-08-18T14:42:50Z sjl: and yeah if I say (simple-array t (* *)) the loop gets down nearly as fast as the fill
2016-08-18T14:43:01Z sjl: but then I can't use this on n-dimensional arrays
2016-08-18T14:44:04Z jdz: sjl: why do you want to fill an existing array instead of creating a new array?
2016-08-18T14:44:34Z sjl: jdz: recomputing dijkstra maps on every tick
2016-08-18T14:44:59Z Petit_Dejeuner quit (Ping timeout: 260 seconds)
2016-08-18T14:45:19Z sjl: I could just throw it away and let the gc deal with it
2016-08-18T14:45:24Z jdz: Yes!
2016-08-18T14:45:36Z gargaml joined #lisp
2016-08-18T14:45:37Z sjl: or I could not generate garbage for no reason
2016-08-18T14:45:50Z IPmonger quit (Ping timeout: 258 seconds)
2016-08-18T14:46:52Z jdz: Yes, so fill-displaced is out of the question, right?
2016-08-18T14:47:09Z scymtym: in SBCL, displaced arrays are the worst
2016-08-18T14:48:28Z jdz: I'd bet allocating arrays with :initial-element as required is definitely better than filling existing arrays.
2016-08-18T14:48:46Z jdz: For a few values of "better".
2016-08-18T14:49:12Z sjl: would be another thing to benchmark
2016-08-18T14:49:24Z jdz: http://blog.jonis.lv/old/3-bb-cl-1/brians-brain-on-cl-take-2.html
2016-08-18T14:51:00Z sellout- quit (Quit: Leaving.)
2016-08-18T14:51:43Z arduo joined #lisp
2016-08-18T14:52:29Z sjl: scymtym: the sbcl-specific version seems a little bit faster, not significantly so though
2016-08-18T14:54:09Z sjl: nevermind, I copypasted wrong.  it's definitely faster
2016-08-18T14:54:35Z eivarv joined #lisp
2016-08-18T14:55:08Z varjag quit (Quit: ERC Version 5.3 (IRC client for Emacs))
2016-08-18T14:55:17Z asc232 joined #lisp
2016-08-18T14:57:02Z zacts quit (Quit: WeeChat 1.5)
2016-08-18T14:57:21Z scymtym: sjl: the problem with the displaced array version is that it accumulates weak pointers to displaced arrays when the arrays are created and only removes them when the arrays are gced. that list is traversed each time a displaced array is created. so it can get much worse with more repetitions and depends on gc behavior
2016-08-18T14:59:47Z elimik31 quit (Ping timeout: 265 seconds)
2016-08-18T14:59:49Z IPmonger joined #lisp
2016-08-18T15:01:16Z sjl: scymtym: ugh, that's an sbcl-specific thing then?
2016-08-18T15:01:25Z sellout- joined #lisp
2016-08-18T15:02:37Z fourier joined #lisp
2016-08-18T15:02:50Z sjl: jdz: alloc version is better than loop, worse than fill and the sbcl-specific on http://paste.lisp.org/display/323574#3
2016-08-18T15:03:35Z sjl: *one
2016-08-18T15:04:30Z scymtym: sjl: probably. i don't know how other implementations handle the problem. the reason for this weak pointer mechanism is that resizing the displaced-to array can propagate to the displaced array which has to be a pretty rare case
2016-08-18T15:05:20Z arduo quit (Quit: ERC (IRC client for Emacs 25.1.1))
2016-08-18T15:05:36Z arduo joined #lisp
2016-08-18T15:06:23Z sjl: ugh, yeah that is annoying
2016-08-18T15:06:39Z fiddlerwoaroof: Hmm, does fill-displaced actually need to construct a new array every time it's called?  Why not just use two arrays?
2016-08-18T15:06:49Z fiddlerwoaroof: (new displaced array that is)
2016-08-18T15:07:11Z BusFactor1 joined #lisp
2016-08-18T15:07:13Z jdz: sjl: also, you have fixed the typo in ftype declaration before timing, right?
2016-08-18T15:07:15Z sjl: fiddlerwoaroof: just create a second array alongside the main one?
2016-08-18T15:07:42Z fiddlerwoaroof: Yeah, have *arr* be the n-dimensional array and *arr-1d* be a displaced array of the appropriate size
2016-08-18T15:08:01Z fiddlerwoaroof: then just (fill *arr-1d* 1) when you need to clear the array
2016-08-18T15:08:21Z sjl: jdz: the one for the sbcl version?  it doesn't change anything
2016-08-18T15:08:58Z jdz: Yes, it has its own declaration using THE.
2016-08-18T15:09:23Z sjl: fiddlerwoaroof: yeah that would work too, though now I have to keep track of two separate arrays.  would be nice to just have a single replacement for `fill`
2016-08-18T15:09:29Z ardoc joined #lisp
2016-08-18T15:09:39Z sjl: but that would be one way to avoid the garbage
2016-08-18T15:09:58Z fiddlerwoaroof: sjl: just bundle them in a struct / class and have your replacement for fill take the struct as its argument
2016-08-18T15:10:18Z ardoc quit (Remote host closed the connection)
2016-08-18T15:10:41Z arduo quit (Remote host closed the connection)
2016-08-18T15:11:24Z sjl: fiddlerwoaroof: yeah I know, that would work.  at this point I'm just satisfying curiosity
2016-08-18T15:11:39Z sjl: though I might stick a `fill-multidimensional` in my little utility lib at the end of this
2016-08-18T15:11:47Z sjl: just on principle
2016-08-18T15:12:25Z lambda-smith joined #lisp
2016-08-18T15:13:29Z fiddlerwoaroof: Yeah, the standards support for multidimensional arrays has always seemed slightly lacking to me
2016-08-18T15:13:34Z IPmonger quit (Ping timeout: 260 seconds)
2016-08-18T15:13:58Z sjl: yeah they do seem to be slightly less supported everywhere
2016-08-18T15:14:23Z sjl: e.g. I had to write an iterate driver to iterate over them nicely https://github.com/sjl/cl-losh/blob/master/losh.lisp#L652
2016-08-18T15:14:28Z lambda-smith quit (Client Quit)
2016-08-18T15:14:38Z eivarv quit (Quit: Sleep)
2016-08-18T15:14:48Z sjl: but at least lisp HAS them, unlike some languages
2016-08-18T15:16:29Z oGMo: but you never really _need_ them :P
2016-08-18T15:16:39Z nikki93 joined #lisp
2016-08-18T15:16:45Z sjl: heh
2016-08-18T15:16:59Z sjl: what's a few extra pointer derefs between friends
2016-08-18T15:17:18Z oGMo: it's trivial to map an N-dimensional array to a 1-dimensional array .. the main thing CL does nicely is let you resize them properly heh
2016-08-18T15:17:54Z sjl: yeah you can wrap it yourself if necessary
2016-08-18T15:18:01Z sjl: but it's nice to have built in
2016-08-18T15:18:13Z oGMo: the other nice thing CL does is let you map annoying multidimensional arrays to a 1-dimensional array ;)
2016-08-18T15:18:47Z oGMo: sjl: it can be handy if you really need them
2016-08-18T15:19:04Z sjl: yeah the displaced arrays are nice.  I don't need them often but it's been really nice to have
2016-08-18T15:19:07Z sjl: when I did
2016-08-18T15:19:44Z oGMo: yeah displaced arrays are probably my favorite CL array feature
2016-08-18T15:20:10Z cagomez quit (Remote host closed the connection)
2016-08-18T15:21:46Z nikki93 quit (Ping timeout: 252 seconds)
2016-08-18T15:21:56Z ramky quit (Ping timeout: 265 seconds)
2016-08-18T15:22:17Z rpg joined #lisp
2016-08-18T15:22:20Z pierpa: anyway, the cost of allocating a displaced array should matter only in the most performance critical of applications, if ever.
2016-08-18T15:23:38Z sjl: pierpa: well from what scymtym said, if creating a displaced array creates a weak pointer that's never gc'ed until the array is
2016-08-18T15:23:44Z sjl: and the array sticks around for a long time
2016-08-18T15:23:56Z sjl: and we're clearing the array every tick in a 60fps game...
2016-08-18T15:24:19Z eivarv joined #lisp
2016-08-18T15:24:23Z arduo joined #lisp
2016-08-18T15:24:32Z pierpa: 60 fps games is a most performance critical application :)
2016-08-18T15:24:40Z sjl: haha
2016-08-18T15:24:44Z sjl: good point
2016-08-18T15:25:20Z knicklux quit (Quit: Leaving)
2016-08-18T15:25:44Z FreeBirdLjj joined #lisp
2016-08-18T15:27:54Z jdz: It might make sense to separate game logic processing from drawing.  I bet it's common to run the game logic at 50Hz, or 30Hz.
2016-08-18T15:28:11Z jdz: Which is orthogonal to the issue at hand :)
2016-08-18T15:29:15Z arduo quit (Remote host closed the connection)
2016-08-18T15:29:49Z flamebeard quit (Quit: Leaving)
2016-08-18T15:30:57Z zeissoctopus joined #lisp
2016-08-18T15:34:03Z prole quit (Remote host closed the connection)
2016-08-18T15:35:11Z eivarv quit (Quit: Sleep)
2016-08-18T15:36:40Z zeissoctopus quit (Quit: Leaving)
2016-08-18T15:36:46Z FreeBirdLjj quit (Remote host closed the connection)
2016-08-18T15:37:53Z rszeno joined #lisp
2016-08-18T15:39:55Z fourier quit (Ping timeout: 252 seconds)
2016-08-18T15:42:33Z FreeBirdLjj joined #lisp
2016-08-18T15:49:11Z Lexicall quit (Remote host closed the connection)
2016-08-18T15:51:32Z mathi_aihtam joined #lisp
2016-08-18T15:52:04Z arbv quit (Read error: Connection reset by peer)
2016-08-18T15:54:38Z araujo joined #lisp
2016-08-18T15:55:09Z tristam__ is now known as Tristam
2016-08-18T15:55:30Z Tristam quit (Changing host)
2016-08-18T15:55:30Z Tristam joined #lisp
2016-08-18T15:57:03Z Grue` quit (Ping timeout: 244 seconds)
2016-08-18T15:57:18Z sellout- quit (Quit: Leaving.)
2016-08-18T15:57:41Z IPmonger joined #lisp
2016-08-18T15:59:33Z papachan joined #lisp
2016-08-18T16:02:16Z IPmonger quit (Ping timeout: 250 seconds)
2016-08-18T16:05:05Z shka_ joined #lisp
2016-08-18T16:06:30Z zacts joined #lisp
2016-08-18T16:08:30Z davsebamse quit (Read error: Connection reset by peer)
2016-08-18T16:08:39Z kilfer joined #lisp
2016-08-18T16:09:09Z JuanDaugherty quit (Quit: Hibernate, reboot, exeunt, etc.)
2016-08-18T16:09:55Z BlueRavenGT joined #lisp
2016-08-18T16:11:06Z gargaml quit (Quit: WeeChat 1.5)
2016-08-18T16:12:29Z davsebamse joined #lisp
2016-08-18T16:14:23Z IPmonger joined #lisp
2016-08-18T16:15:09Z Grue` joined #lisp
2016-08-18T16:15:23Z Rikeo joined #lisp
2016-08-18T16:15:26Z mvilleneuve quit (Quit: This computer has gone to sleep)
2016-08-18T16:18:02Z rpg quit (Quit: My Mac has gone to sleep. ZZZzzz…)
2016-08-18T16:18:39Z zaon joined #lisp
2016-08-18T16:21:26Z Rikeo quit (Ping timeout: 264 seconds)
2016-08-18T16:23:04Z beach joined #lisp
2016-08-18T16:23:13Z Rikeo_ joined #lisp
2016-08-18T16:23:14Z wccoder joined #lisp
2016-08-18T16:23:37Z Rikeo_: hello guys, anyone develop good in squeak or clisp?
2016-08-18T16:24:15Z beach: Rikeo_: Do you really mean "clisp".  CLISP is an implementation of Common Lisp.
2016-08-18T16:24:31Z beach: Good evening everyone, by the way.
2016-08-18T16:25:18Z warweasle is now known as warweasle_lab
2016-08-18T16:25:21Z beach: Rikeo_: This channel is dedicated to Common Lisp, so many people here develop in Common Lisp.
2016-08-18T16:27:28Z rpg joined #lisp
2016-08-18T16:29:24Z Bike joined #lisp
2016-08-18T16:35:20Z kokonaisluku joined #lisp
2016-08-18T16:35:28Z beach: Bike: I looked at the code.
2016-08-18T16:36:00Z araujo_ joined #lisp
2016-08-18T16:36:02Z Bike: morning.
2016-08-18T16:36:07Z beach: Hello!
2016-08-18T16:36:08Z beach: Do you want me to update it before you start working, or do you want me to tell you what to do?
2016-08-18T16:36:33Z Bike: if you have something you want to do to it i can do it, sure. though i already defined a few transfer methods
2016-08-18T16:36:43Z beach: Great!
2016-08-18T16:36:56Z beach: First, the creation of the dictionary needs to return RESULT.
2016-08-18T16:37:15Z beach: Second, we can remove the methods on consp-instruction and fixnump-instruction.
2016-08-18T16:37:32Z beach: Those instructions are deprecated.  TYPEQ can do it all.
2016-08-18T16:37:33Z araujo_ quit (Max SendQ exceeded)
2016-08-18T16:37:33Z Bike: Have those been replaced by typeq?
2016-08-18T16:37:35Z Bike: right.
2016-08-18T16:37:43Z beach: Well, nobody is using them.
2016-08-18T16:37:57Z beach: Maybe drmeister is using fixnump, but I don't think so.
2016-08-18T16:38:14Z beach: It would be good if the code remains intact even after we generalize the lattice.
2016-08-18T16:38:17Z Bike: pretty easy to fix if so
2016-08-18T16:38:22Z beach: yes.
2016-08-18T16:38:28Z araujo_ joined #lisp
2016-08-18T16:38:34Z Bike: generalize the lattice, meaning expanding the type descriptors?
2016-08-18T16:38:38Z beach: For that, we seem to need an operation meet-complement or complement-meet.
2016-08-18T16:38:42Z beach: Yes.
2016-08-18T16:39:03Z araujo quit (Ping timeout: 240 seconds)
2016-08-18T16:39:07Z Bike: i think it'd just be "difference".
2016-08-18T16:39:16Z beach: sure.
2016-08-18T16:39:33Z beach: But then, perhaps we should rename meet->intersection and join->union?
2016-08-18T16:39:57Z beach: Anyway, not important right now.
2016-08-18T16:39:58Z Bike: that did confuse me at first, but I guess meet and join are the terms you use for lattices?
2016-08-18T16:40:03Z beach: Yes.
2016-08-18T16:40:15Z Bike: for the first thing, do you mean compute-initial-dictionary? because that already returns result
2016-08-18T16:40:27Z beach: Oh?
2016-08-18T16:40:41Z beach: No, wait, computing the worklist.
2016-08-18T16:40:49Z beach: It fails to return the result I think.
2016-08-18T16:40:52Z Bike: ah. oh, yes, that is an issue.
2016-08-18T16:41:19Z beach: So you will do some of that?
2016-08-18T16:41:29Z Bike: sure.
2016-08-18T16:41:50Z zacharias quit (Quit: WeeChat 1.5)
2016-08-18T16:41:59Z beach: Great!
2016-08-18T16:42:25Z beach: I need to go cook dinner, and then spend time with my (admittedly small) family.  I'll be back tomorrow morning (UTC+2).
2016-08-18T16:42:39Z Bike: have a good meal.
2016-08-18T16:42:39Z beach: I may check in here briefly after dinner in case you have any questions.
2016-08-18T16:42:43Z beach: Thanks!
2016-08-18T16:43:33Z hhdave quit (Ping timeout: 244 seconds)
2016-08-18T16:43:34Z mathi_aihtam quit (Read error: Connection reset by peer)
2016-08-18T16:44:35Z stepnem quit (Ping timeout: 265 seconds)
2016-08-18T16:45:58Z IPmonger quit (Quit: ZNC 1.6.3 - http://znc.in)
2016-08-18T16:46:33Z pierpa quit (Ping timeout: 240 seconds)
2016-08-18T16:46:38Z fourier joined #lisp
2016-08-18T16:47:49Z DGASAU quit (Read error: Connection reset by peer)
2016-08-18T16:48:51Z Polyphony quit (Ping timeout: 276 seconds)
2016-08-18T16:49:20Z DGASAU joined #lisp
2016-08-18T16:49:28Z gravicappa joined #lisp
2016-08-18T16:53:32Z IPmonger joined #lisp
2016-08-18T16:54:15Z mathi_aihtam joined #lisp
2016-08-18T16:56:24Z HeyFlash quit (Remote host closed the connection)
2016-08-18T16:57:02Z kokonaisluku quit (Quit: ChatZilla 0.9.92 [Firefox 45.3.0/20160803111628])
2016-08-18T16:57:28Z papachan quit (Ping timeout: 252 seconds)
2016-08-18T16:57:50Z smokeink quit (Ping timeout: 250 seconds)
2016-08-18T16:58:10Z discardedes joined #lisp
2016-08-18T16:59:07Z algae joined #lisp
2016-08-18T17:04:52Z warweasle_lab is now known as warweasle
2016-08-18T17:04:52Z u03bb quit (Quit: bye)
2016-08-18T17:06:48Z anunnaki quit (Ping timeout: 244 seconds)
2016-08-18T17:07:59Z mastokley joined #lisp
2016-08-18T17:08:02Z varjag joined #lisp
2016-08-18T17:09:09Z Bike: ok, so currently type inference doesn't actually happen, both because of the work-list bug, and because process-instruction needs predecessors, which the enter-instructions it starts on don't have. nice.
2016-08-18T17:09:11Z bdr3553 joined #lisp
2016-08-18T17:10:44Z IPmonger quit (Ping timeout: 250 seconds)
2016-08-18T17:11:02Z papachan joined #lisp
2016-08-18T17:11:24Z bdr3552 quit (Ping timeout: 260 seconds)
2016-08-18T17:12:28Z arrsim quit (Ping timeout: 250 seconds)
2016-08-18T17:13:06Z arrsim joined #lisp
2016-08-18T17:13:18Z moei joined #lisp
2016-08-18T17:13:53Z IPmonger joined #lisp
2016-08-18T17:17:49Z zacts quit (Ping timeout: 252 seconds)
2016-08-18T17:18:14Z IPmonger quit (Client Quit)
2016-08-18T17:21:22Z MoALTz joined #lisp
2016-08-18T17:22:19Z warweasle quit (Quit: meeting...)
2016-08-18T17:22:28Z Rikeo_ quit (Quit: Page closed)
2016-08-18T17:23:48Z IPmonger joined #lisp
2016-08-18T17:24:11Z arduo joined #lisp
2016-08-18T17:24:32Z knobo: Can I find out which asdf system a file is used in?
2016-08-18T17:25:02Z cagomez joined #lisp
2016-08-18T17:25:48Z knobo: I was thinking about making an emacs function that moves the defpackage form to (merge-pathnames "packages.lisp" (asdf:system-source-directory :system-name))
2016-08-18T17:26:47Z knobo: or, in the :module directory. if the file exists.
2016-08-18T17:28:04Z IPmonger quit (Ping timeout: 250 seconds)
2016-08-18T17:29:08Z m00natic quit (Remote host closed the connection)
2016-08-18T17:31:01Z IPmonger joined #lisp
2016-08-18T17:31:39Z fourier quit (Ping timeout: 250 seconds)
2016-08-18T17:31:45Z mastokley quit (Ping timeout: 276 seconds)
2016-08-18T17:34:34Z rpg: so you have a filename, and you want to reverse search to find the system?
2016-08-18T17:41:15Z Cymew quit (Ping timeout: 264 seconds)
2016-08-18T17:43:37Z FreeBirdLjj quit (Remote host closed the connection)
2016-08-18T17:44:39Z Grue`: seems like you just need to find a packages.lisp in one of the parent directories of current file
2016-08-18T17:45:05Z gingerale joined #lisp
2016-08-18T17:45:06Z Grue`: because there's probably only one
2016-08-18T17:45:28Z __acher__ joined #lisp
2016-08-18T17:47:54Z slyrus quit (Ping timeout: 265 seconds)
2016-08-18T17:48:08Z BlueRavenGT quit (Ping timeout: 244 seconds)
2016-08-18T17:49:52Z puchacz joined #lisp
2016-08-18T17:50:17Z KaliLinuxGR quit (Remote host closed the connection)
2016-08-18T17:50:36Z IPmonger quit (Ping timeout: 250 seconds)
2016-08-18T17:52:39Z fiddlerwoaroof: Is there a way to have slime translate remote pathnames?
2016-08-18T17:52:58Z ggole quit (Remote host closed the connection)
2016-08-18T17:53:02Z fiddlerwoaroof: I'm connecting to a swank that's running in a vm under a different username from my username
2016-08-18T17:53:09Z fiddlerwoaroof: Which breaks M-.
2016-08-18T17:53:22Z ggole joined #lisp
2016-08-18T17:53:31Z Bike: https://common-lisp.net/project/slime/doc/html/Setting-up-pathname-translations.html#Setting-up-pathname-translations this?
2016-08-18T17:53:55Z fiddlerwoaroof: I think so
2016-08-18T17:53:57Z fiddlerwoaroof: Thanks
2016-08-18T17:56:42Z knobo: rpg, Grue`: right..
2016-08-18T17:57:44Z rpg: knobo: If you really want to do that, I *think* you have to find the loaded systems (there's a function for that), then find all their children, and try to match component-pathnames.
2016-08-18T17:58:56Z vlatkoB quit (Remote host closed the connection)
2016-08-18T17:58:59Z rpg: Potentially computationally expensive, but it shouldn't be that difficult to program.
2016-08-18T18:00:33Z vlatkoB joined #lisp
2016-08-18T18:03:29Z wildlander joined #lisp
2016-08-18T18:04:44Z cagomez quit
2016-08-18T18:04:51Z Davidbrcz joined #lisp
2016-08-18T18:05:06Z ovenpasta joined #lisp
2016-08-18T18:07:07Z pierpa joined #lisp
2016-08-18T18:07:42Z atgreen quit (Quit: Leaving)
2016-08-18T18:08:37Z milanj joined #lisp
2016-08-18T18:11:18Z milanj quit (Client Quit)
2016-08-18T18:12:52Z milanj joined #lisp
2016-08-18T18:13:20Z Denommus joined #lisp
2016-08-18T18:16:34Z milanj quit (Client Quit)
2016-08-18T18:17:54Z Davidbrcz quit (Ping timeout: 260 seconds)
2016-08-18T18:19:21Z zacts joined #lisp
2016-08-18T18:19:29Z zacharias joined #lisp
2016-08-18T18:19:41Z DGASAU quit (Read error: Connection reset by peer)
2016-08-18T18:20:24Z DGASAU joined #lisp
2016-08-18T18:21:16Z BusFactor1 quit (Quit: My Mac has gone to sleep. ZZZzzz…)
2016-08-18T18:23:13Z BusFactor1 joined #lisp
2016-08-18T18:26:34Z ovenpasta quit (Ping timeout: 250 seconds)
2016-08-18T18:28:07Z nikki93 joined #lisp
2016-08-18T18:28:08Z nikki93 quit (Remote host closed the connection)
2016-08-18T18:28:37Z nikki93 joined #lisp
2016-08-18T18:31:48Z mastokley joined #lisp
2016-08-18T18:32:33Z Davidbrcz joined #lisp
2016-08-18T18:32:49Z BusFactor1 quit (Quit: My Mac has gone to sleep. ZZZzzz…)
2016-08-18T18:34:16Z BusFactor1 joined #lisp
2016-08-18T18:34:58Z NeverDie quit (Quit: http://radiux.io/)
2016-08-18T18:36:53Z rszeno quit (Quit: Leaving.)
2016-08-18T18:44:16Z adolf_stalin quit (Quit: Leaving...)
2016-08-18T18:48:36Z warweasle joined #lisp
2016-08-18T18:50:09Z knicklux joined #lisp
2016-08-18T18:51:58Z zygentoma joined #lisp
2016-08-18T18:52:41Z EvW joined #lisp
2016-08-18T18:53:16Z asc232 quit (Remote host closed the connection)
2016-08-18T18:56:41Z pareidolia: A question for the group
2016-08-18T18:57:06Z davsebamse quit (Ping timeout: 265 seconds)
2016-08-18T18:57:25Z pareidolia: When editing a multi-file project in Emacs, do you use a fancy solution to switch between these files, or (like me) do you just switch between buffers and keep a dired open?
2016-08-18T18:58:39Z davsebamse joined #lisp
2016-08-18T19:00:54Z raydeejay: pareidolia: I sometimes pop up speedbar
2016-08-18T19:01:11Z dlowe: I don't even keep a dired open.
2016-08-18T19:01:28Z dlowe: but I do have keys bound to compile and run tests
2016-08-18T19:02:06Z pareidolia: raydeejay: I'll look up speedbar
2016-08-18T19:02:11Z pareidolia: Thanks to both :)
2016-08-18T19:03:58Z jean377 quit (Ping timeout: 252 seconds)
2016-08-18T19:08:22Z xantoz: pareidolia: I used to do that but recently I tend to use find-file-in-project which works for git repos
2016-08-18T19:08:54Z xantoz: I bound it to C-x C-M-f
2016-08-18T19:09:05Z PuercoPop: Y normally just use C-x C-f or C-x b, when exploring a project I use direct (through C-d after C-x C-f)
2016-08-18T19:11:33Z rumbler31 joined #lisp
2016-08-18T19:12:51Z pareidolia: I just remembered I have Projectile
2016-08-18T19:13:22Z EvW quit (Ping timeout: 250 seconds)
2016-08-18T19:14:45Z ovenpasta joined #lisp
2016-08-18T19:17:39Z zacts quit (Quit: WeeChat 1.5)
2016-08-18T19:18:28Z zacts joined #lisp
2016-08-18T19:19:29Z IPmonger joined #lisp
2016-08-18T19:20:18Z PuercoPop: dlowe: how do you run tests in lisp? Do you do something like (swank:eval `(asdf:test-system ,current-system)) and set the current-system with dir-locals.el? (I normally just use (asdf:test-system :foo) in the REPL)
2016-08-18T19:20:57Z EvW joined #lisp
2016-08-18T19:21:15Z zacts quit (Client Quit)
2016-08-18T19:23:34Z pareidolia: Grrrrr. Package.el seems to ignore the protocol in the address. I need http:// on marmalade because GnuTLS gives me errors
2016-08-18T19:24:24Z IPmonger quit (Ping timeout: 260 seconds)
2016-08-18T19:25:10Z phoe quit (Quit: Lost terminal)
2016-08-18T19:25:29Z vlatkoB quit (Remote host closed the connection)
2016-08-18T19:27:31Z mathi_aihtam quit (Read error: Connection reset by peer)
2016-08-18T19:27:46Z npatrick04 joined #lisp
2016-08-18T19:28:24Z nikki93__ joined #lisp
2016-08-18T19:29:16Z Grue`: pareidolia: I almost always use C-x 5 f, each file in its own wind^H^H^H^H frame
2016-08-18T19:29:55Z DGASAU quit (Ping timeout: 244 seconds)
2016-08-18T19:30:18Z slyrus joined #lisp
2016-08-18T19:30:42Z nikki93 quit (Remote host closed the connection)
2016-08-18T19:32:25Z dyelar quit (Remote host closed the connection)
2016-08-18T19:32:47Z pareidolia quit (Remote host closed the connection)
2016-08-18T19:33:18Z pareidolia joined #lisp
2016-08-18T19:33:28Z dyelar joined #lisp
2016-08-18T19:33:48Z mathi_aihtam joined #lisp
2016-08-18T19:34:56Z pareidolia quit (Remote host closed the connection)
2016-08-18T19:38:48Z warweasle quit (Quit: bye)
2016-08-18T19:41:18Z rumbler31: pareidolia: what os
2016-08-18T19:42:33Z DGASAU joined #lisp
2016-08-18T19:42:56Z davsebamse quit (Ping timeout: 250 seconds)
2016-08-18T19:44:02Z ggole quit (Ping timeout: 244 seconds)
2016-08-18T19:44:12Z __acher__ quit (Ping timeout: 240 seconds)
2016-08-18T19:44:47Z davsebamse joined #lisp
2016-08-18T19:45:18Z NeverDie joined #lisp
2016-08-18T19:46:37Z __acher__ joined #lisp
2016-08-18T19:47:08Z lnostdal quit (Ping timeout: 244 seconds)
2016-08-18T19:52:09Z npatrick04 quit (Ping timeout: 276 seconds)
2016-08-18T19:53:27Z angavrilov quit (Remote host closed the connection)
2016-08-18T19:53:29Z asc232 joined #lisp
2016-08-18T19:54:02Z gravicappa quit (Remote host closed the connection)
2016-08-18T19:56:04Z Karl_Dscc quit (Remote host closed the connection)
2016-08-18T19:56:34Z npatrick04 joined #lisp
2016-08-18T19:57:23Z moore33 joined #lisp
2016-08-18T20:00:10Z holly2 quit (Ping timeout: 250 seconds)
2016-08-18T20:02:01Z holly2 joined #lisp
2016-08-18T20:02:25Z fourier joined #lisp
2016-08-18T20:04:01Z IPmonger joined #lisp
2016-08-18T20:04:36Z jerme quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
2016-08-18T20:07:27Z nalik quit (Quit: Leaving)
2016-08-18T20:07:35Z gilez quit (Ping timeout: 265 seconds)
2016-08-18T20:08:08Z attila_lendvai joined #lisp
2016-08-18T20:08:08Z attila_lendvai quit (Changing host)
2016-08-18T20:08:08Z attila_lendvai joined #lisp
2016-08-18T20:08:30Z IPmonger quit (Ping timeout: 250 seconds)
2016-08-18T20:10:21Z npatrick04 quit (Ping timeout: 276 seconds)
2016-08-18T20:11:09Z asc232 quit (Remote host closed the connection)
2016-08-18T20:11:39Z holly2 quit (Ping timeout: 276 seconds)
2016-08-18T20:14:18Z fourier quit (Ping timeout: 258 seconds)
2016-08-18T20:17:30Z IPmonger joined #lisp
2016-08-18T20:21:50Z IPmonger quit (Ping timeout: 250 seconds)
2016-08-18T20:24:41Z IPmonger joined #lisp
2016-08-18T20:25:36Z BusFactor1 quit (Quit: My Mac has gone to sleep. ZZZzzz…)
2016-08-18T20:26:56Z holly2 joined #lisp
2016-08-18T20:27:17Z BusFactor1 joined #lisp
2016-08-18T20:28:41Z asc232 joined #lisp
2016-08-18T20:29:12Z IPmonger quit (Ping timeout: 250 seconds)
2016-08-18T20:32:27Z Grue` quit (Ping timeout: 276 seconds)
2016-08-18T20:33:09Z gargaml joined #lisp
2016-08-18T20:33:35Z knicklux quit (Remote host closed the connection)
2016-08-18T20:36:20Z zacts joined #lisp
2016-08-18T20:36:51Z Ven_ joined #lisp
2016-08-18T20:38:43Z bdr3553 left #lisp
2016-08-18T20:39:08Z prole joined #lisp
2016-08-18T20:41:19Z Blukunfando quit (Ping timeout: 252 seconds)
2016-08-18T20:41:42Z puchacz quit (Quit: Konversation terminated!)
2016-08-18T20:44:02Z pareidolia joined #lisp
2016-08-18T20:44:58Z pareidolia: programisto: Saluton!
2016-08-18T20:45:17Z mathi_aihtam quit (Ping timeout: 265 seconds)
2016-08-18T20:45:50Z pareidolia: rumbler31: Linux
2016-08-18T20:45:57Z drmeister: beach: If you are up late I'm here to chat if you like.
2016-08-18T20:47:09Z Bike: hey, drmeister. i decided to see if i could do that multiple-value-bind stuff you wanted in february and probably don't remember.
2016-08-18T20:48:31Z drmeister: Hi bike - I kind of recall - help is always greatly appreciated!
2016-08-18T20:48:57Z drmeister: Would you like a capsule summary of what I've been up to clasp-wise?
2016-08-18T20:50:31Z Bike: yeah, sure.
2016-08-18T20:53:35Z rpg quit (Quit: Textual IRC Client: www.textualapp.com)
2016-08-18T20:54:15Z wccoder quit (Remote host closed the connection)
2016-08-18T20:56:23Z drmeister: I've incorporated a new llvm feature "Link Time Optimization" - basically everything, C++ and Common Lisp is compiled to LLVM bitcode and the linker links bitcode together.
2016-08-18T20:56:32Z papachan quit (Ping timeout: 240 seconds)
2016-08-18T20:56:54Z drmeister: The linker does inlining and other whole-program optimization with the bitcode before lowering it to native code.
2016-08-18T20:57:47Z Bike: i remember this, i think.
2016-08-18T20:57:49Z drmeister: This means that there is a lot more inlining and function calls that can't be inlined and do not export symbols are converted to the "fastcc" calling convention where registers are matched across the caller and the callee.
2016-08-18T20:58:37Z drmeister: To incorporate LLVM LTO I rewrote the entire build system.  I ditched boost build and switched to the 'waf' build system.
2016-08-18T20:58:51Z drmeister: It makes it a lot easier to implement LTO.
2016-08-18T20:58:52Z Bike: oh, geez.
2016-08-18T20:59:28Z drmeister: I gotta do what I gotta do.
2016-08-18T20:59:45Z Bike: sounds like the pain is worth it at least.
2016-08-18T20:59:58Z drmeister: waf is also better in that I can incorporate tests into the build process that fail early if libraries are not available.
2016-08-18T21:00:07Z faheem joined #lisp
2016-08-18T21:00:19Z drmeister: Yeah, even after working with boost build for more than 5 years I still didn't understand it.
2016-08-18T21:00:19Z mathi_aihtam joined #lisp
2016-08-18T21:00:34Z Bike: do you understand waf?
2016-08-18T21:00:42Z drmeister: waf is Python with a build system built into it.   It's much more understandable.   When I run into problems I just write more code.
2016-08-18T21:01:17Z Bike: cool.
2016-08-18T21:01:28Z IPmonger joined #lisp
2016-08-18T21:02:04Z drmeister: Sort of, but not completely, but I got it to work.
2016-08-18T21:03:07Z drmeister: waf uses a lot of higher order function capabilities of Python.   It's a bit bewildering but there are lots of examples and fairly comprehensive documentation.
2016-08-18T21:03:17Z drmeister: There is also the #waf IRC channel.
2016-08-18T21:03:29Z Bike: boost doesn't have an irc channel?
2016-08-18T21:03:43Z drmeister: It does, but good luck getting an answer.
2016-08-18T21:03:54Z Bike: eheh.
2016-08-18T21:04:03Z drmeister: Boost build implements a language but it's awful.
2016-08-18T21:04:12Z Bike: boost is just so massive
2016-08-18T21:04:43Z drmeister: Yeah.  There's boost and within it is "boost build", the build system that builds boost.
2016-08-18T21:05:22Z drmeister: "So much information - and yet - I feel like I knew it all along."
2016-08-18T21:09:23Z Bike: well, i've been working on cleavir. i added inlining for a couple types of normal calls, but you need type inference for multiple-value-call, so i'm doing that now.
2016-08-18T21:09:42Z drmeister: That's awesome!
2016-08-18T21:09:50Z drmeister: I saw beach's comment about TYPEQ vs FIXNUMP.
2016-08-18T21:09:52Z wccoder joined #lisp
2016-08-18T21:10:02Z Bike: oh, yeah. are you still using fixnump?
2016-08-18T21:10:03Z drmeister: I don't have a problem switching to TYPEQ - it should be simple
2016-08-18T21:10:21Z Bike: it's just as well, as of right now there is no type inference anyway
2016-08-18T21:10:54Z drmeister: Yes, I use fixnump but I should be able to convert (typeq x 'fixnum) to (fixnump x)  - that's the issue right?
2016-08-18T21:11:15Z Bike: the other way around
2016-08-18T21:11:32Z Bike: also i think you don't quote the type for typeq.
2016-08-18T21:12:35Z drmeister: The other way around?   My thinking was Cleavir will generate (typeq x fixnum) as HIR and I would lower that to (fixnump x) in MIR.  Or am I confused?
2016-08-18T21:13:46Z drmeister: Where HIR is "His (beach's) Intermediate Representation" and MIR is "My Intermediate Representation".  I'm being funny but HIR->MIR
2016-08-18T21:13:50Z Bike: er, well the problem is the fixnump HIR instruction, which is unnecessary.
2016-08-18T21:14:12Z __acher__ quit (Ping timeout: 240 seconds)
2016-08-18T21:14:45Z drmeister: Yes, I agree with that, and that is because its simpler generally to just use TYPEQ and specifically (TYPEQ x FIXNUM) to test for fixnums.
2016-08-18T21:15:22Z papachan joined #lisp
2016-08-18T21:15:37Z Bike: right. that's what we were talking about. if you have... oh, sorry, i see what you meant. you mean that right now you are using HIR fixnump, but you can easily adjust away from that by adding a transform from HIR typeq to MIR fixnump.
2016-08-18T21:16:20Z drmeister: Yes
2016-08-18T21:16:42Z Bike: which, yeah, that's what you need to do. i wasn't thinking about it since i'm up in cloudcuckooland, here.
2016-08-18T21:17:10Z drmeister: Ok
2016-08-18T21:17:19Z drmeister: It sounds exciting
2016-08-18T21:18:58Z grimsley joined #lisp
2016-08-18T21:19:04Z Bike: it's fun. the code i added to inline functions with &optional and &rest looks nasty, though
2016-08-18T21:22:33Z IPmonger quit (Ping timeout: 244 seconds)
2016-08-18T21:25:42Z IPmonger joined #lisp
2016-08-18T21:26:10Z arduo quit (Remote host closed the connection)
2016-08-18T21:26:25Z arduo joined #lisp
2016-08-18T21:27:23Z npatrick04 joined #lisp
2016-08-18T21:29:25Z EvW quit (Ping timeout: 244 seconds)
2016-08-18T21:30:12Z IPmonger quit (Ping timeout: 258 seconds)
2016-08-18T21:31:12Z drmeister: Is it commited to cleavir?
2016-08-18T21:31:45Z drmeister: How are you debugging it?
2016-08-18T21:31:48Z Bike: yup. beach is quick https://github.com/robert-strandh/SICL/pull/39
2016-08-18T21:31:54Z Bike: Debugging what, my code?
2016-08-18T21:32:16Z drmeister: Yes.   Is it tested?  Are you looking at HIR graphs or actually running it somehow.
2016-08-18T21:32:45Z drmeister: You can smoke test it in Clasp.
2016-08-18T21:33:35Z Bike: for inlining i only needed AST level, so i used AST-to-source and AST-interpreter
2016-08-18T21:34:19Z drmeister: I see - and is it all working?
2016-08-18T21:34:31Z Bike: yes, but the tests i put in certainly don't cover all actual use
2016-08-18T21:35:23Z drmeister: If I pulled the update and compiled clasp - that should test it pretty hard.
2016-08-18T21:35:40Z Bike: indeed. when you run into a problem i will fix it
2016-08-18T21:35:53Z Bike: i still haven't built clasp myself. i think.
2016-08-18T21:37:04Z Bike: 19 MB of deltas. mhm, i have not tried this in a while
2016-08-18T21:38:06Z arduo quit (Remote host closed the connection)
2016-08-18T21:38:15Z shka_ quit (Ping timeout: 258 seconds)
2016-08-18T21:38:31Z Bike: http://imgur.com/a/KF0Dw mmhm, definitely time to nuke everything and start over.
2016-08-18T21:39:11Z Bike: "you may want to set your merge.renamelimit variable to at least 26269 and retry the command"
2016-08-18T21:41:40Z mathi_aihtam quit (Read error: Connection reset by peer)
2016-08-18T21:41:46Z drmeister: Is that from updating clasp?   Yes,  nuke it from orbit and clone fresh.
2016-08-18T21:41:50Z drmeister: Everything has changed.
2016-08-18T21:41:55Z Bike: yeah. done.
2016-08-18T21:42:06Z Bike: i'm not foolish enough to wade through twenty thousand merge conflicts.
2016-08-18T21:44:49Z cmack quit (Ping timeout: 258 seconds)
2016-08-18T21:45:11Z gingerale quit (Remote host closed the connection)
2016-08-18T21:45:12Z Bike: Why is it using an old LLVM?
2016-08-18T21:45:36Z cmack joined #lisp
2016-08-18T21:46:24Z gargaml quit (Quit: WeeChat 1.5)
2016-08-18T21:48:33Z mathi_aihtam joined #lisp
2016-08-18T21:51:47Z npatrick` joined #lisp
2016-08-18T21:53:10Z IPmonger joined #lisp
2016-08-18T21:53:33Z mastokley quit (Ping timeout: 240 seconds)
2016-08-18T21:53:39Z Ven_ quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2016-08-18T21:53:58Z npatrick04 quit (Ping timeout: 258 seconds)
2016-08-18T21:56:17Z pareidolia quit (Ping timeout: 244 seconds)
2016-08-18T21:56:29Z mathi_aihtam quit (Quit: mathi_aihtam)
2016-08-18T21:58:15Z LiamH quit (Quit: Leaving.)
2016-08-18T21:58:23Z cagomez joined #lisp
2016-08-18T22:00:52Z SumoSudo quit (Ping timeout: 258 seconds)
2016-08-18T22:01:19Z attila_lendvai quit (Ping timeout: 260 seconds)
2016-08-18T22:02:13Z npatrick` quit (Ping timeout: 265 seconds)
2016-08-18T22:03:49Z zacts quit (Quit: WeeChat 1.5)
2016-08-18T22:04:38Z algae quit (Ping timeout: 265 seconds)
2016-08-18T22:09:29Z IPmonger quit (Ping timeout: 260 seconds)
2016-08-18T22:09:33Z drmeister: Bike - whoops - I need to catch you up there as well.
2016-08-18T22:09:34Z eivarv joined #lisp
2016-08-18T22:09:55Z drmeister: Hang on...
2016-08-18T22:10:00Z Bike: i do appreciate the directions for arch linux, i just like to avoid downgrading things as much as i can
2016-08-18T22:12:51Z MoALTz quit (Quit: Leaving)
2016-08-18T22:13:38Z drmeister: It uses a VERY recent version of llvm3.9    you can build it using externals-clasp.
2016-08-18T22:13:40Z drmeister: https://github.com/drmeister/externals-clasp/blob/master/README.md
2016-08-18T22:14:03Z drmeister: In a few weeks I hope people can start getting llvm3.9 with package managers.
2016-08-18T22:14:10Z Bike: ...oh, so the clasp readme is out of date, i see.
2016-08-18T22:14:25Z drmeister: You need to use the 'testing' branch of clasp.
2016-08-18T22:14:32Z drmeister: I'll update the README now.
2016-08-18T22:14:56Z drmeister: I've been off programming and working on the build system and not documenting.
2016-08-18T22:15:25Z Bike: understandable. so, i need to download and make externals-clasp, and then i can build the 'testing' branch of clasp?
2016-08-18T22:15:57Z Denommus quit (Quit: Bye)
2016-08-18T22:16:29Z __acher__ joined #lisp
2016-08-18T22:17:03Z drmeister: Yeesh - that is out of date.  I have to run and catch a train.  I'll edit it on the train.
2016-08-18T22:17:36Z drmeister: Yes - (1) externals-clasp  (2) edit the clasp/local.config to point it at externals-clasp (3) make clasp
2016-08-18T22:17:45Z drmeister: There are dependencies that are listed here.
2016-08-18T22:17:47Z Bike: got it.
2016-08-18T22:17:52Z larsen quit (Ping timeout: 240 seconds)
2016-08-18T22:18:24Z drmeister: https://github.com/drmeister/clasp/wiki/Building-Clasp-0.4-on-Ubuntu
2016-08-18T22:18:43Z drmeister: I think that's correct other than the "boehm" gc library has been added as a dependency.
2016-08-18T22:18:53Z eivarv quit (Quit: Sleep)
2016-08-18T22:19:19Z Bike: i think i have that anyway.
2016-08-18T22:20:27Z zygentoma quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
2016-08-18T22:20:37Z zygentoma joined #lisp
2016-08-18T22:21:32Z cmack quit (Ping timeout: 250 seconds)
2016-08-18T22:23:38Z cagomez quit
2016-08-18T22:26:03Z Bike quit (Quit: Lost terminal)
2016-08-18T22:26:06Z Davidbrcz quit (Ping timeout: 244 seconds)
2016-08-18T22:30:14Z zygentoma quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
2016-08-18T22:38:12Z pierpa quit (Ping timeout: 240 seconds)
2016-08-18T22:38:22Z npatrick` joined #lisp
2016-08-18T22:42:44Z klltkr joined #lisp
2016-08-18T22:44:58Z jleija joined #lisp
2016-08-18T22:45:18Z edgar-rft quit (Quit: edgar-rft)
2016-08-18T22:46:02Z zaon quit (Quit: Konversation terminated!)
2016-08-18T22:49:32Z rumbler31 quit (Remote host closed the connection)
2016-08-18T22:50:32Z mishoo quit (Ping timeout: 240 seconds)
2016-08-18T22:51:12Z __acher__ quit (Ping timeout: 240 seconds)
2016-08-18T22:51:33Z npatrick` quit (Ping timeout: 276 seconds)
2016-08-18T22:54:09Z klltkr quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2016-08-18T22:59:21Z BusFactor1 quit (Ping timeout: 276 seconds)
2016-08-18T23:00:17Z mishoo joined #lisp
2016-08-18T23:07:22Z EvW joined #lisp
2016-08-18T23:15:11Z wccoder quit (Remote host closed the connection)
2016-08-18T23:15:25Z wccoder joined #lisp
2016-08-18T23:17:03Z mishoo quit (Ping timeout: 265 seconds)
2016-08-18T23:19:32Z algae joined #lisp
2016-08-18T23:22:51Z scymtym quit (Ping timeout: 265 seconds)
2016-08-18T23:24:09Z DeadTrickster quit (Ping timeout: 260 seconds)
2016-08-18T23:29:28Z ovenpasta quit (Ping timeout: 258 seconds)
2016-08-18T23:30:58Z al-damiri quit (Quit: Connection closed for inactivity)
2016-08-18T23:31:21Z varjag quit (Ping timeout: 244 seconds)
2016-08-18T23:31:30Z zelcon5 joined #lisp
2016-08-18T23:32:02Z sellout- joined #lisp
2016-08-18T23:32:34Z zelcon5: why doesn't this work (xor swap)?
2016-08-18T23:32:35Z zelcon5: http://paste.lisp.org/display/323602/raw
2016-08-18T23:33:34Z papachan quit (Ping timeout: 265 seconds)
2016-08-18T23:36:36Z sjl: zelcon5: it does, but you're only returning the final value of a
2016-08-18T23:37:02Z sjl: if you're expecting it to mutate whatever you're passing in, that's not how it works
2016-08-18T23:39:23Z sjl: zelcon5: http://paste.stevelosh.com/57b6471f222cab000757bb27
2016-08-18T23:39:35Z klltkr joined #lisp
2016-08-18T23:40:32Z sjl: but you might just want (psetf a b b a) unless this is just for bit twiddling practice
2016-08-18T23:41:12Z wildlander quit (Quit: Saliendo)
2016-08-18T23:42:55Z cromachina joined #lisp
2016-08-18T23:45:15Z milanj joined #lisp
2016-08-18T23:48:13Z IPmonger joined #lisp
2016-08-18T23:52:57Z zelcon5: thanks
2016-08-18T23:53:34Z IPmonger quit (Ping timeout: 258 seconds)
2016-08-18T23:53:46Z zelcon5: I thought SETF modified the place in memory like "*place = value"
2016-08-18T23:54:55Z kn-928 joined #lisp
2016-08-18T23:55:50Z algae quit (Quit: leaving)
2016-08-18T23:58:02Z z3r0_ joined #lisp