2014-09-26T00:00:10Z boogie joined #lisp
2014-09-26T00:01:55Z urandom__ quit (Quit: Konversation terminated!)
2014-09-26T00:02:38Z lambdAnon joined #lisp
2014-09-26T00:02:44Z lambdAnon: hi everyone
2014-09-26T00:02:49Z hugoduncan joined #lisp
2014-09-26T00:02:51Z Xach: hello
2014-09-26T00:03:21Z boogie quit (Client Quit)
2014-09-26T00:03:43Z hugoduncan is now known as aduadu
2014-09-26T00:05:13Z klltkr quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2014-09-26T00:07:12Z bbyler_tho quit (Remote host closed the connection)
2014-09-26T00:09:44Z hitecnologys quit (Quit: hitecnologys)
2014-09-26T00:12:01Z Bicyclidine joined #lisp
2014-09-26T00:14:08Z innertracks joined #lisp
2014-09-26T00:15:34Z svetlyak40wt joined #lisp
2014-09-26T00:19:21Z attila_lendvai quit (Quit: Leaving.)
2014-09-26T00:19:52Z svetlyak40wt quit (Ping timeout: 250 seconds)
2014-09-26T00:19:55Z drmeister_: Hello
2014-09-26T00:20:15Z drmeister_: I should have spent a month testing my build system - I feel sick.
2014-09-26T00:26:21Z Zhivago: You discovered everything is a lie? :)
2014-09-26T00:26:44Z drmeister_: No, just that getting things to build on other peoples computers is hard.
2014-09-26T00:27:10Z Zhivago: I'm looking forward to when people start distributing virtual computers instead.
2014-09-26T00:27:53Z zz_karupa is now known as karupa
2014-09-26T00:35:35Z resttime: maybe releasing a set of binaries as well?
2014-09-26T00:37:01Z drmeister_: restime: How do you do that?  Do I just tar the build directory?
2014-09-26T00:37:25Z drmeister_: The built directory that contains the executables and libraries?   I have everything in one hierarchy.
2014-09-26T00:38:32Z drmeister_: That would be a relief.  I was working with two people simultaneously, with two different problems.  One working on OS X and the other on linux - I almost lost it.  And it's all my fault.
2014-09-26T00:38:51Z drmeister_: Ok, ok, calm down.   Phew.
2014-09-26T00:39:01Z drmeister_ breathes deeply.
2014-09-26T00:39:07Z drmeister_ into a paper bag.
2014-09-26T00:39:09Z joshe: build systems are hard
2014-09-26T00:39:26Z joshe: well, ones that work well anyway
2014-09-26T00:40:10Z drmeister_: I knew that but I didn't _know_ that.
2014-09-26T00:40:19Z oGMo: this is why as much as everyone whines about autotools sucking, it's still the only one that works solidly and without a lot of effort :p
2014-09-26T00:40:46Z oGMo: (and it does suck, but everything else *is* *far* *worse*)
2014-09-26T00:42:10Z drmeister_: I really appreciate it when a system has ./configure; make and it works.
2014-09-26T00:42:17Z harish_ quit (Ping timeout: 245 seconds)
2014-09-26T00:42:29Z drmeister_: I didn't use autotools because I use boost build.   There is too much overlap of functionality
2014-09-26T00:43:06Z oGMo: yeah unfortunately trying to use two together is usually exponential headache
2014-09-26T00:43:20Z drmeister_: Yes.
2014-09-26T00:44:20Z drmeister_: Ok, the problems are all fixable and I'm fixing them now.
2014-09-26T00:45:54Z drmeister_: There's a reason it took me so long to get this out.  I was so afraid of this.  Ok, I'm over-sharing.   Time to get a grip and fix some code.
2014-09-26T00:46:28Z drmeister_: It's all just ones and zeros.    And two's don't exist.
2014-09-26T00:46:36Z oGMo: drmeister_: you should hardly sweat those kinds of things though heh, they're usually easy to fix and you don't encounter them til others test
2014-09-26T00:46:54Z nyef: drmeister_: This is also why people often do a "stealth" release, and wait until the second or third version to start hitting up the publicity.
2014-09-26T00:47:08Z oGMo: eh
2014-09-26T00:47:30Z drmeister_: nyef: Point taken.
2014-09-26T00:48:08Z nyef: Even SBCL occasionally gets fixes to its build process, sometimes fairly hefty fixes.
2014-09-26T00:48:40Z nyef: And I never want to try building CMUCL.
2014-09-26T00:49:14Z drmeister_: nyef: Thanks - I really appreciate it.
2014-09-26T00:50:10Z Bicyclidine: i'm sure you'll be fine publicity wise. i think most people who see "C++ CL interop" and get excited about it know that there are gonna be weird difficulties.
2014-09-26T00:50:53Z drmeister_: Yeah and I figure we'll expose some cool libraries and post some more announcements.  That's when the rubber really hits the road.
2014-09-26T00:51:24Z drmeister_: Oh and when we incorporate Cleavir.  I'm really looking forward to that.
2014-09-26T00:54:27Z jusss joined #lisp
2014-09-26T00:56:55Z drmeister_:  It's ok - in one case we got to a minimal repl - there was just a problem resolving pathnames in logical-pathname-translations (I think).
2014-09-26T00:57:16Z varjag_ quit (Quit: Connection closed for inactivity)
2014-09-26T00:58:37Z zlrth quit (Ping timeout: 260 seconds)
2014-09-26T00:58:50Z zRecursive joined #lisp
2014-09-26T01:01:09Z drmeister_: The rest of the problems were all hardwired paths in makefiles that work on my system but not others.
2014-09-26T01:02:13Z nyef: Here's another SBCL build system quirk: Some SBCL versions behave differently if they are built in a directory that has "src" somewhere in the parent chain.
2014-09-26T01:02:32Z nyef: (As opposed to not having the "src" directory component.)
2014-09-26T01:03:58Z drmeister_ thinks he might still need a hug.  Trundles off to find his cat.
2014-09-26T01:05:47Z Longlius joined #lisp
2014-09-26T01:08:27Z kushal joined #lisp
2014-09-26T01:09:40Z clop2 joined #lisp
2014-09-26T01:11:06Z drmeister_: I'm taking a cold eyed look at what's need and what's not needed for the release.     How do you use git to just get the most recent commit?   Is it: git clone --depth 1 URL   ?
2014-09-26T01:12:22Z zRecursive: drmeister_: git show 'HEAD@{1}..HEAD'
2014-09-26T01:12:37Z zRecursive: woow, no
2014-09-26T01:13:10Z drmeister_: Wait - what I want to do is suggest to users that they can use:  git clone --depth 1 URL    to just get the most recent commit without the history.
2014-09-26T01:13:16Z nyef: I'm not aware of it being possible to create a repository with only the HEAD commit.
2014-09-26T01:13:42Z nyef: That said, why not tarball the working directory, excluding the .git directory?
2014-09-26T01:13:50Z lambdAno` joined #lisp
2014-09-26T01:15:41Z jleija joined #lisp
2014-09-26T01:15:46Z drmeister_: But if I had ECL in the repo and then I:   git rm -r ecl    users who clone the repo will still get ECL when they clone it - correct?
2014-09-26T01:16:01Z nyef: Yes.
2014-09-26T01:16:05Z svetlyak40wt joined #lisp
2014-09-26T01:16:19Z drmeister_: Well, there's one other thing I miss about svn.
2014-09-26T01:16:38Z drmeister_: That and small integer revision numbers.
2014-09-26T01:17:02Z drmeister_: Is there any way to prune that old stuff out?
2014-09-26T01:17:32Z nyef: Hrm. Actually, a quick google search suggests that git clone --depth 1 is a reasonable thing to try.
2014-09-26T01:17:49Z lambdAnon quit (Ping timeout: 260 seconds)
2014-09-26T01:18:48Z nyef: The thing is, a git revision is the full state of the tree at that point, plus a commit message. Unlike a darcs commit, which is the delta to the previous commit plus a commit message.
2014-09-26T01:18:50Z drmeister_: Yeah - I looked that up - I was asking more because I was fishing for peoples experience with it.
2014-09-26T01:19:31Z nyef: This might be one of those case where you just need to take the ten minutes or so to try it and see what happens.
2014-09-26T01:20:17Z drmeister_: You are correct.
2014-09-26T01:21:09Z svetlyak40wt quit (Ping timeout: 272 seconds)
2014-09-26T01:23:20Z lambdAno` quit (Read error: Connection reset by peer)
2014-09-26T01:25:49Z P4Titan joined #lisp
2014-09-26T01:29:33Z ltbarcly quit (Quit: Computer has gone to sleep.)
2014-09-26T01:29:40Z chitofan joined #lisp
2014-09-26T01:33:01Z theethicalegoist joined #lisp
2014-09-26T01:33:47Z lisper29 joined #lisp
2014-09-26T01:34:06Z innertracks quit (Ping timeout: 246 seconds)
2014-09-26T01:37:27Z ltbarcly joined #lisp
2014-09-26T01:37:57Z zeebrah quit (Ping timeout: 244 seconds)
2014-09-26T01:41:29Z harish_ joined #lisp
2014-09-26T01:44:43Z MutSbeta quit (Quit: Leaving.)
2014-09-26T01:44:55Z c107 joined #lisp
2014-09-26T01:50:04Z _ku quit (Quit: WeeChat 0.4.3)
2014-09-26T01:52:05Z jusss quit (Read error: No route to host)
2014-09-26T01:52:10Z Bicyclidine quit (Ping timeout: 250 seconds)
2014-09-26T01:57:40Z kushal quit (Quit: Leaving)
2014-09-26T01:59:34Z huza joined #lisp
2014-09-26T02:03:54Z jusss joined #lisp
2014-09-26T02:08:16Z yacks joined #lisp
2014-09-26T02:16:31Z svetlyak40wt joined #lisp
2014-09-26T02:20:06Z lowfyr joined #lisp
2014-09-26T02:21:13Z svetlyak40wt quit (Ping timeout: 260 seconds)
2014-09-26T02:21:43Z lowfyr: hello #lisp
2014-09-26T02:23:33Z pillton quit (Quit: ERC Version 5.3 (IRC client for Emacs))
2014-09-26T02:25:59Z P4Titan: hola
2014-09-26T02:26:13Z z_ joined #lisp
2014-09-26T02:26:19Z baotiao joined #lisp
2014-09-26T02:26:25Z z_ is now known as Guest55664
2014-09-26T02:26:28Z ltbarcly quit (Quit: Computer has gone to sleep.)
2014-09-26T02:28:49Z zophy joined #lisp
2014-09-26T02:28:57Z zeitue quit (Ping timeout: 245 seconds)
2014-09-26T02:30:43Z viaken: hi
2014-09-26T02:31:14Z madmalik quit (Quit: Connection closed for inactivity)
2014-09-26T02:33:14Z boogie joined #lisp
2014-09-26T02:35:03Z marsbot is now known as marsam
2014-09-26T02:35:08Z boogie quit (Remote host closed the connection)
2014-09-26T02:35:21Z ndrei joined #lisp
2014-09-26T02:35:35Z boogie joined #lisp
2014-09-26T02:36:13Z zeebrah joined #lisp
2014-09-26T02:38:52Z chitofan quit (Ping timeout: 246 seconds)
2014-09-26T02:39:50Z ndrei quit (Ping timeout: 250 seconds)
2014-09-26T02:46:00Z theethicalegoist quit (Quit: Leaving)
2014-09-26T02:49:01Z jlongster joined #lisp
2014-09-26T02:52:54Z lisper29 left #lisp
2014-09-26T02:57:23Z bgs100 quit (Quit: bgs100)
2014-09-26T02:58:03Z nyef quit (Quit: G'night all.)
2014-09-26T03:06:16Z eMBee quit (Ping timeout: 250 seconds)
2014-09-26T03:08:10Z gluegadget quit (Read error: Connection reset by peer)
2014-09-26T03:10:12Z gluegadget_ joined #lisp
2014-09-26T03:13:46Z lyanchih_ joined #lisp
2014-09-26T03:16:56Z svetlyak40wt joined #lisp
2014-09-26T03:17:10Z blakbunnie27 quit (Quit: EliteBNC free bnc service - http://elitebnc.org - be a part of the Elite!)
2014-09-26T03:20:09Z marsam is now known as marsbot
2014-09-26T03:21:24Z svetlyak40wt quit (Ping timeout: 260 seconds)
2014-09-26T03:21:33Z hiyosi_ quit (Ping timeout: 246 seconds)
2014-09-26T03:22:08Z zacharias_ joined #lisp
2014-09-26T03:25:13Z lyanchih_ quit (Quit: lyanchih_)
2014-09-26T03:25:23Z drmeister_: How do people communicate on github?   Is there any sort of wiki or chatroom associated with github?
2014-09-26T03:25:35Z drmeister_: Or what do people use?
2014-09-26T03:26:00Z zacharias quit (Ping timeout: 258 seconds)
2014-09-26T03:27:13Z phao quit (Quit: Leaving)
2014-09-26T03:27:52Z White_Flame: Is there a library which counts the number of "1" bits in a bit vector using specialized CPU instructions?
2014-09-26T03:28:53Z White_Flame: drmeister_: I believe each project gets a wiki, though I'm not sure if that's meant for general communication purposes
2014-09-26T03:30:23Z P4Titan left #lisp
2014-09-26T03:35:06Z eMBee joined #lisp
2014-09-26T03:35:21Z marsbot is now known as marsam
2014-09-26T03:36:57Z pillton joined #lisp
2014-09-26T03:45:00Z replcated_ quit (Ping timeout: 246 seconds)
2014-09-26T03:45:38Z replcated joined #lisp
2014-09-26T03:47:04Z boogie quit (Remote host closed the connection)
2014-09-26T03:48:08Z hiyosi_ joined #lisp
2014-09-26T03:48:51Z marsam is now known as marsbot
2014-09-26T03:52:48Z hiyosi_ quit (Ping timeout: 244 seconds)
2014-09-26T03:53:09Z KCL_ joined #lisp
2014-09-26T03:53:48Z blakbunnie27 joined #lisp
2014-09-26T03:55:19Z gz quit (Connection reset by peer)
2014-09-26T03:55:42Z KCL quit (Ping timeout: 245 seconds)
2014-09-26T04:00:48Z KCL joined #lisp
2014-09-26T04:01:39Z KCL_ quit (Ping timeout: 258 seconds)
2014-09-26T04:07:42Z jusss quit (Ping timeout: 245 seconds)
2014-09-26T04:09:03Z gingerale joined #lisp
2014-09-26T04:12:08Z Kuran joined #lisp
2014-09-26T04:12:31Z Kuran is now known as Guest92288
2014-09-26T04:12:57Z Neet quit (Read error: Connection reset by peer)
2014-09-26T04:13:23Z Neet joined #lisp
2014-09-26T04:16:17Z wchun joined #lisp
2014-09-26T04:17:24Z svetlyak40wt joined #lisp
2014-09-26T04:18:19Z madrik joined #lisp
2014-09-26T04:18:19Z mac_ified joined #lisp
2014-09-26T04:20:23Z node joined #lisp
2014-09-26T04:21:44Z svetlyak40wt quit (Ping timeout: 244 seconds)
2014-09-26T04:22:34Z boogie joined #lisp
2014-09-26T04:23:02Z mac_ified quit (Remote host closed the connection)
2014-09-26T04:23:53Z svetlyak40wt joined #lisp
2014-09-26T04:23:57Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T04:24:48Z kami joined #lisp
2014-09-26T04:25:16Z node quit (Ping timeout: 246 seconds)
2014-09-26T04:25:20Z node_ joined #lisp
2014-09-26T04:25:51Z kpreid quit (Quit: Quitting)
2014-09-26T04:25:58Z jleija quit (Quit: leaving)
2014-09-26T04:26:14Z kpreid joined #lisp
2014-09-26T04:26:23Z node_ left #lisp
2014-09-26T04:32:42Z araujo quit (Quit: Leaving)
2014-09-26T04:32:47Z kristof joined #lisp
2014-09-26T04:33:13Z robot-beethoven joined #lisp
2014-09-26T04:38:29Z vinleod joined #lisp
2014-09-26T04:41:04Z Adlai: drmeister_: discussions around issues and pull requests
2014-09-26T04:45:03Z drmeister_: Adlai: I don't understand.
2014-09-26T04:45:44Z drmeister_: Is there a way with git to checkout a particular commit?    I have commit 3025b00b7f127c3bcd04c31e8b9cab4639ad6510
2014-09-26T04:46:00Z Adlai: drmeister_: an example: https://github.com/bitcoin/bitcoin/issues/273
2014-09-26T04:46:06Z drmeister_: I tried git checkout 3025b00b7f127c3bcd04c31e8b9cab4639ad6510   but all I get is    fatal: reference is not a tree: 3025b00b7f127c3bcd04c31e8b9cab4639ad6510
2014-09-26T04:46:24Z Adlai: drmeister_: yes, git checkout -b newbranchname commitID
2014-09-26T04:46:46Z Adlai: hm, if 'git checkout ID' says it's not valid, then it's not valid
2014-09-26T04:46:56Z Adlai is rusty on git
2014-09-26T04:46:58Z drmeister_: So git checkout -b newbranchname  3025b00b7f127c3bcd04c31e8b9cab4639ad6510
2014-09-26T04:49:00Z hiyosi_ joined #lisp
2014-09-26T04:49:46Z gingerale quit (Ping timeout: 272 seconds)
2014-09-26T04:50:42Z baotiao quit (Ping timeout: 245 seconds)
2014-09-26T04:51:06Z drmeister_: Damn - this was exactly what I was afraid of.   I grabbed a TOT llvm3.6 build and I was told I could checkout this specific commit.  The problem is that users then have to download the entire LLVM history to get that one commit.
2014-09-26T04:52:07Z drmeister_: If the download git clone --depth 1 XXXX then they only get the last commit and that specific commit from a couple of days ago wont be there.
2014-09-26T04:52:37Z drmeister_: If they get the last commit LLVM keeps changing.  There are already changes that break my build.  Crap.
2014-09-26T04:53:32Z hiyosi_ quit (Ping timeout: 245 seconds)
2014-09-26T04:53:55Z drmeister_: Adlai - sorry.  You were saying discussions around issues and pull requests are used as a way to communicate to and between users.
2014-09-26T04:54:31Z gryAway is now known as Svetlana
2014-09-26T04:54:52Z Adlai: drmeister_: if you add it as a submodule referencing that specific commit, your users just need to run `git submodule update', and git will download the history up to the relevant commit, and no further
2014-09-26T04:54:57Z drmeister_: Thank you - that is helpful.  I didn't realize that would be useful for discussions.
2014-09-26T04:55:16Z drmeister_: Brrr, submodules.
2014-09-26T04:55:29Z Adlai: indeed, it seems as though the actual development discussion for that project has shifted out of the forums in favor of IRC and github
2014-09-26T04:57:01Z Adlai: wrt checking out a commit - if you don't use -b to create a new branch, you'll get a dangling head. in other words, the thin abstractions over git's raw internals will melt away and you'll probably learn much more than you ever wanted to know about git
2014-09-26T04:57:01Z mr-foobar quit (Quit: Leaving...)
2014-09-26T04:57:25Z vinleod quit (Quit: Computer has gone to sleep.)
2014-09-26T04:58:01Z drmeister_: Actually, maybe submodules aren't such a bad idea here.
2014-09-26T04:58:05Z oleo__ quit (Quit: Verlassend)
2014-09-26T04:58:18Z baotiao joined #lisp
2014-09-26T04:58:22Z drmeister_: I can put git submodule update in the makefile.
2014-09-26T05:01:01Z vanila joined #lisp
2014-09-26T05:02:34Z c107 quit (Remote host closed the connection)
2014-09-26T05:04:37Z buzzy_ quit (Remote host closed the connection)
2014-09-26T05:04:57Z baotiao_ joined #lisp
2014-09-26T05:05:40Z huza quit (Ping timeout: 258 seconds)
2014-09-26T05:07:35Z baotiao quit (Ping timeout: 258 seconds)
2014-09-26T05:07:36Z baotiao_ is now known as baotiao
2014-09-26T05:09:40Z kristof quit (Ping timeout: 260 seconds)
2014-09-26T05:10:25Z drmeister_: git submodule is a nightmare.
2014-09-26T05:11:20Z drmeister_: I'm going to clone the latest commit of llvm, delete out all the git stuff and add it to the externals-clasp repo - I don't see any other way to quickly lock down a specific revision of llvm/clang that will work with Clasp.
2014-09-26T05:12:24Z drmeister_: There's probably some tricky git wizardry that will make submodule or subtree or gitslave (very poor choice of name) work but I can only tolerate so much pain in my life.
2014-09-26T05:13:26Z rme quit (Quit: rme)
2014-09-26T05:14:21Z vanila: hey drmeister_ good luck with the release
2014-09-26T05:14:32Z vanila: i came to say congrats
2014-09-26T05:15:20Z drmeister_: Thank you.
2014-09-26T05:16:53Z bjorkintosh quit (Ping timeout: 240 seconds)
2014-09-26T05:17:31Z Adlai: drmeister_: what is bad with submodule for your use-case?
2014-09-26T05:17:58Z drmeister_: http://codingkilledthecat.wordpress.com/2012/04/28/why-your-company-shouldnt-use-git-submodules/
2014-09-26T05:18:01Z Adlai: it's far from ideal as a general purpose tool, but it does do exactly what it's supposed to
2014-09-26T05:18:48Z Adlai: i'd replace "They even support pointing at specific versions of the shared code" with "they ONLY ..."
2014-09-26T05:19:03Z Adlai: just treat it as a single unmoveable pointer into another repository
2014-09-26T05:19:44Z jlongster quit (Ping timeout: 250 seconds)
2014-09-26T05:19:58Z bjorkintosh joined #lisp
2014-09-26T05:20:03Z drmeister_: I'll have to sit down and relearn submodules.  I don't have time right now.  I need to get clasp building on linux and OS X. I make a change to OS X and linux breaks - I make a change to linux and OS X breaks. Then git breaks.  I'm a bit fed up.
2014-09-26T05:20:54Z drmeister_: I'm trying to just: git clone --depth 1 http://llvm.org/git/llvm.git llvm and it's been sitting here for 15 minutes.  What the heck?
2014-09-26T05:21:15Z kobain quit (Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/)
2014-09-26T05:21:32Z Adlai: don't treat submodule as a tool for working on multiple repositories, because it's not. treat it as a badly-named tool for linking a snapshot of one repository into another, and nothing more.
2014-09-26T05:24:21Z svetlyak40wt joined #lisp
2014-09-26T05:24:53Z brucem: drmeister: do you have changes to LLVM or just use a specific revision? (Why not the actual release?)
2014-09-26T05:27:42Z prxq joined #lisp
2014-09-26T05:29:25Z svetlyak40wt quit (Ping timeout: 272 seconds)
2014-09-26T05:29:45Z drmeister_: Adlai: Well, that makes a bit more sense to me now "badly-named tool for linking a snapshot of one repository into another"
2014-09-26T05:31:24Z drmeister_: brucem: I accidentally grabbed a TOT build last week rather than the stable release llvm3.5.  Then I spent the weekend updating clasp to work with it.  Then I realized a lot of the changes I'd made to clasp were to support features since the llvm3.5 release.
2014-09-26T05:31:57Z brucem: drmeister_: ah. you'll find life a lot easier if you just stick with the 3.5 releaes for now ... that way you can just use package system builds and so on.
2014-09-26T05:32:23Z drmeister_: Agreed but I don't want to downgrade the changes I made to clasp.  They were pretty serious.
2014-09-26T05:32:24Z jusss joined #lisp
2014-09-26T05:32:29Z brucem: (as someone who has multiple LLVM versions and forks on his development machien)
2014-09-26T05:32:42Z brucem: well, good luck then. You're not going to have an easy time of it.
2014-09-26T05:32:55Z drmeister_: I'm kind of stuck.  It's exactly the situation I was trying to avoid.
2014-09-26T05:33:43Z drmeister_: The big change is std::unique_ptr(s) are being returned and passed to llvm functions since llvm3.5.  That required all sorts of changes to the C++/CL interoperation.
2014-09-26T05:33:48Z buzzy joined #lisp
2014-09-26T05:34:27Z drmeister_: I thought users could check out a particular commit to llvm - No - git doesn't support that.
2014-09-26T05:34:41Z brucem: You can, easily.
2014-09-26T05:34:46Z drmeister_: How?
2014-09-26T05:34:58Z brucem: git clone <url>
2014-09-26T05:35:04Z brucem: git checkout <commit>
2014-09-26T05:35:35Z brucem: or, just use submodules and pin them to the commit. Also easy.
2014-09-26T05:35:35Z drmeister_: Right - so everyone has to clone the entire llvm history to check out one commit.
2014-09-26T05:35:52Z brucem: Well,that's why you should use a release.
2014-09-26T05:36:06Z drmeister_: Right - that's what Adlai was just explaining to me.   I've only used submodules once and wrecked my first repository with them.
2014-09-26T05:36:32Z drmeister_: I don't disagree - I just screwed up and missed the release.
2014-09-26T05:37:12Z jegaxd26 quit (Read error: Connection reset by peer)
2014-09-26T05:37:14Z brucem: you're calling internal LLVM stuff? or the actual external LLVM API is changing that much?
2014-09-26T05:37:30Z brucem: we just emit LLVM IR bitcode files ... much easier than dealing with LLVM API.
2014-09-26T05:37:48Z zymurgy joined #lisp
2014-09-26T05:38:24Z drmeister_: brucem: Since last week clang::ClangTool::setArgumentsAdjuster has been removed and replaced with clang::ClangTool::addArgumentsAdjuster   - this breaks the clasp build.
2014-09-26T05:39:03Z drmeister_: I wanted C++/CL interop so that I could live on the bleeding edge.  That's the price I pay.
2014-09-26T05:39:03Z kami` joined #lisp
2014-09-26T05:39:23Z hiroakip joined #lisp
2014-09-26T05:41:15Z kami quit (Ping timeout: 272 seconds)
2014-09-26T05:42:33Z chitofan joined #lisp
2014-09-26T05:45:32Z kami` quit (Read error: Connection reset by peer)
2014-09-26T05:46:08Z kami` joined #lisp
2014-09-26T05:46:37Z mr-foobar joined #lisp
2014-09-26T05:47:52Z drmeister_: It's like Alice in Wonderland.  Run as fast as you can just to stay in the same place.
2014-09-26T05:49:47Z hiyosi_ joined #lisp
2014-09-26T05:51:12Z kami` quit (Ping timeout: 260 seconds)
2014-09-26T05:52:05Z patbarron joined #lisp
2014-09-26T05:52:07Z drmeister_: I expose quite a lot of the LLVM and Clang libraries, so when anything changes I see it.  Usually it's trivial to fix though.  Like the one above "addArgumentsAdjuster" vs "setArgumentsAdjuster"  Pffft - no biggie.
2014-09-26T05:53:07Z drmeister_: std::unique_ptr(s)  whoah - biggie.   Move semantics totally screws up language interoperation because you can't automatically wrap passing a std::unique_ptr.  You have to explicitly put std::move(...) around them.
2014-09-26T05:54:02Z MoALTz_ joined #lisp
2014-09-26T05:54:48Z pranavrc joined #lisp
2014-09-26T05:55:11Z hiyosi_ quit (Ping timeout: 272 seconds)
2014-09-26T05:55:16Z drmeister_: Adlai: What about submodules within submodules and submodules within submodules within submodules.  I kid you not - the latter is exactly what I have to do.   llvm contains clang and clang contains extras.
2014-09-26T05:57:17Z MoALTz quit (Ping timeout: 245 seconds)
2014-09-26T05:57:22Z hiroakip quit (Ping timeout: 240 seconds)
2014-09-26T05:57:36Z MoALTz__ joined #lisp
2014-09-26T06:00:18Z drmeister_: I'm reading this: http://git-scm.com/book/en/Git-Tools-Submodules
2014-09-26T06:00:49Z loke quit (Remote host closed the connection)
2014-09-26T06:01:29Z MoALTz_ quit (Ping timeout: 260 seconds)
2014-09-26T06:03:08Z patbarron quit (Read error: Connection reset by peer)
2014-09-26T06:06:08Z mr-foobar quit (Ping timeout: 260 seconds)
2014-09-26T06:08:01Z svetlyak40wt joined #lisp
2014-09-26T06:08:32Z zeebrah quit (Ping timeout: 245 seconds)
2014-09-26T06:12:54Z chitofan: http://paste.lisp.org/display/143845
2014-09-26T06:13:05Z chitofan: how does this function work?
2014-09-26T06:13:13Z chitofan: where does the argument x come from?
2014-09-26T06:13:37Z Bike: (funcall (make-adder 4) 5)
2014-09-26T06:13:40Z Bike: n is 4, x is 5
2014-09-26T06:15:23Z boogie quit (Remote host closed the connection)
2014-09-26T06:16:19Z chitofan: i think i get it but i probably dont
2014-09-26T06:16:21Z chitofan: thanks bike!
2014-09-26T06:17:36Z freaksken joined #lisp
2014-09-26T06:19:32Z TDog quit (Ping timeout: 244 seconds)
2014-09-26T06:22:15Z mcc joined #lisp
2014-09-26T06:22:31Z Adlai quit (Remote host closed the connection)
2014-09-26T06:23:00Z Adlai joined #lisp
2014-09-26T06:24:17Z defaultxr quit (Quit: gnight)
2014-09-26T06:24:22Z Shinmera joined #lisp
2014-09-26T06:27:03Z mishoo joined #lisp
2014-09-26T06:27:46Z MoALTz__ quit (Quit: Leaving)
2014-09-26T06:27:56Z Guest92288 quit (Ping timeout: 272 seconds)
2014-09-26T06:28:39Z grungier quit (Remote host closed the connection)
2014-09-26T06:28:51Z zmyrgel_ is now known as zmyrgel
2014-09-26T06:29:02Z tbarletz_ quit (Ping timeout: 245 seconds)
2014-09-26T06:29:02Z uber quit (Ping timeout: 245 seconds)
2014-09-26T06:29:23Z sjl quit (Ping timeout: 240 seconds)
2014-09-26T06:29:52Z __main__ quit (Ping timeout: 245 seconds)
2014-09-26T06:29:53Z tstc quit (Ping timeout: 240 seconds)
2014-09-26T06:30:57Z sjl joined #lisp
2014-09-26T06:31:12Z tbarletz_ joined #lisp
2014-09-26T06:31:33Z grungier joined #lisp
2014-09-26T06:31:54Z tstc joined #lisp
2014-09-26T06:32:19Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T06:32:36Z pgomes joined #lisp
2014-09-26T06:35:33Z drewc quit (Ping timeout: 260 seconds)
2014-09-26T06:35:48Z drmeister_: git submodules are very disappointing.   Nested submodules don't seem to work and now I have to claw submodules back out of my repo.
2014-09-26T06:36:16Z __main__ joined #lisp
2014-09-26T06:36:28Z uber joined #lisp
2014-09-26T06:36:29Z uber quit (Max SendQ exceeded)
2014-09-26T06:36:30Z Shinmera: drmeister_: git subtrees work nicely
2014-09-26T06:36:59Z uber joined #lisp
2014-09-26T06:37:00Z Shinmera: drmeister_: Not sure about nesting, but I'd think that should work as well.
2014-09-26T06:39:14Z drmeister_: I think I'm done with git for the day.
2014-09-26T06:39:31Z drmeister_: Thank you though.
2014-09-26T06:39:47Z Shinmera: Understandable. Hopefully it won't cause much of a headache later then.
2014-09-26T06:41:52Z mvilleneuve joined #lisp
2014-09-26T06:43:11Z drmeister_: Adlai: Thank you for your suggestion - I appreciate it - and I learned something about submodules.  I'm learning to be leery of git.
2014-09-26T06:44:15Z Shinmera: Git is big, unfortunately.
2014-09-26T06:44:37Z drmeister_: I'm trying to:  git submodule rm llvm and it just keeps saying "Did you forget to 'git add'?
2014-09-26T06:44:39Z Adlai quit (Ping timeout: 264 seconds)
2014-09-26T06:45:06Z zeebrah joined #lisp
2014-09-26T06:45:20Z drmeister_: Can I just go into .gitmodules and delete the entry - is there anything else that needs to be done?    I've got down this road before and completely screwed up a repository.
2014-09-26T06:45:21Z Shinmera: Afaik git submodules are just a special file within the submodule git repo that marks the commit or something
2014-09-26T06:45:33Z Shinmera: I think you can just remove that, yes
2014-09-26T06:45:47Z Shinmera: But, no insurance!
2014-09-26T06:46:12Z Shinmera: A quick googlin' reveals "git submodule deinit" http://stackoverflow.com/questions/1260748/how-do-i-remove-a-git-submodule
2014-09-26T06:46:43Z oudeis joined #lisp
2014-09-26T06:47:40Z drmeister_: Crap it's 3am.
2014-09-26T06:47:58Z Shinmera: The worst things happen at 3Am.
2014-09-26T06:48:11Z mcc: drmeister_ Hi-- are you the person who made this "Clasp" thing?
2014-09-26T06:48:17Z mcc: erp may have asked at the wrong time ^_^;
2014-09-26T06:48:22Z drmeister_: mcc: I am.
2014-09-26T06:48:38Z drmeister_: Hah - that's funny.
2014-09-26T06:48:48Z drmeister_: Shoot - what's up?
2014-09-26T06:49:05Z boogie joined #lisp
2014-09-26T06:49:29Z mcc: Oh— just, I'm currently in process of trying to either find or create a high-level-ish language that integrates well with C++
2014-09-26T06:49:51Z mcc: (I write video games, which means C++ is wildly inappropriate but also mandatory because so many libraries use it)
2014-09-26T06:50:00Z drmeister_: You've come to the right place.   Once I wrestle my build process into submission I'll have exactly that.
2014-09-26T06:50:03Z mcc: haha
2014-09-26T06:50:17Z mcc: So I was very interested in clasp, and I was wondering if you could point to anything that was an example of what C++ integration looks like from with in the clasp LISP.
2014-09-26T06:50:27Z arenz joined #lisp
2014-09-26T06:50:33Z hiyosi_ joined #lisp
2014-09-26T06:50:36Z boogie quit (Read error: Connection reset by peer)
2014-09-26T06:50:46Z boogie joined #lisp
2014-09-26T06:51:12Z mcc: i spent a while working with a binding script for lua and there was some pretty bad disjoint all the time between what C++ knew how to do and what the binding script knew how to represent.
2014-09-26T06:51:24Z mcc: Do I understand Clasp is essentially self-hosting?
2014-09-26T06:51:52Z drmeister_: That I can do.   If you download the clasp repository (please don't try to build it yet - I'm working on it) and look at clasp/src/asttooling/clangTooling.cc lines 427 on down.
2014-09-26T06:52:26Z mcc: haha ok
2014-09-26T06:52:30Z mcc: is this… is this based on llvm, or clang?
2014-09-26T06:52:38Z drmeister_: Do you mean does Clasp build itself from C++?  If so, yes - Clasp is completely self-hosting.
2014-09-26T06:52:40Z mcc: or the binder uses clanglib, or something?
2014-09-26T06:53:01Z mcc: cool. i think i was trying to figure out if Clasp itself was a good source of example code for how Clasp looks when it's interacting with C++.
2014-09-26T06:53:10Z flip214: drmeister: drmeister_: could you add some more examples to clasp, please?
2014-09-26T06:53:28Z drmeister_: mcc: It uses LLVM to build LLVM-IR.  The Clang stuff was so that I could cleanup and analyze the C++ code to make it better.
2014-09-26T06:53:56Z drmeister_: flip214: I will.   I just have to get it working first.
2014-09-26T06:54:15Z mcc: OK
2014-09-26T06:54:49Z drmeister_: mcc: The rest of the interop is in the clasp/src/clbind directory - that is the C++ template meta programming library that builds the interfaces at C++ compile time.
2014-09-26T06:55:00Z flip214: drmeister_: ah no, making it work is left as an exercise to the reader ;)
2014-09-26T06:55:05Z drmeister_: To expose a C++ function you just use def("function-name",&function)
2014-09-26T06:55:12Z mcc: one other question. so clasp is… literally compiled LISP? once the compiled stuff is running, is it possible to run interpreted LISP on top?
2014-09-26T06:55:12Z boogie quit (Ping timeout: 245 seconds)
2014-09-26T06:55:36Z hiyosi_ quit (Ping timeout: 260 seconds)
2014-09-26T06:55:57Z drmeister_: mcc: Yes, but the interpreter is really slow - it's only suitable for bootstrapping.   I do what SBCL does - I compile everything to native code and then execute it.
2014-09-26T06:56:22Z mcc: hm. but you don't do that at runtime i assume
2014-09-26T06:56:36Z Shinmera: There is no strict difference between runtime and the rest.
2014-09-26T06:56:40Z posterdati300 quit (Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/)
2014-09-26T06:56:44Z drmeister_: The compiled code is not that performant either - it's about 100x slower than SBCL right now.  But we are working on a new compiler that I think will be very fast.
2014-09-26T06:57:00Z mcc: heh
2014-09-26T06:57:08Z mcc: so I guess what I really mean, my focus is games-- there's kind of an artistic element there and it's a bit trial and error in some ways. but it is also often performance critical.
2014-09-26T06:57:19Z Adlai joined #lisp
2014-09-26T06:57:30Z mcc: so when you're actually designing the code, it's very helpful to be able to have fast iteration and just alter a file and immediately reload it. but when you're shipping the code you would rather have something compiled.
2014-09-26T06:57:48Z easye joined #lisp
2014-09-26T06:57:50Z drmeister_: mcc: I'm not sure what you mean by runtime.   REPL  = Read-eval-print-loop.   Every form is compiled between read-eval and then evaluated.  You can also COMPILE-FILE code which is ahead-of-time compilation.
2014-09-26T06:58:19Z mcc: Ah, okay, I see.
2014-09-26T06:58:24Z Shinmera: mcc: I suggest you take a look at CL itself first, to get a bit of a hang of how things work
2014-09-26T06:58:42Z Shinmera: mcc: Since Clasp is a CL implementation it'll give you a good idea of what it'll be like.
2014-09-26T06:58:45Z mcc nods. I've some experience with LISPs/Schemes btu I have not used CL specifically
2014-09-26T06:58:47Z mcc: Thanks
2014-09-26T06:59:07Z drmeister_: mcc: The best way is to use emacs and just send the form to the CL environment, it compiles it and then runs it.  It's pretty fast.  But when you really need the code to run fast you COMPILE-FILE it and then link all the modules together using link-time-optimization which inlines everything into everything else.
2014-09-26T06:59:15Z drmeister_: That's all done right now.
2014-09-26T06:59:40Z drmeister_: mcc: You haven't used Lisp until you've used Common Lisp.   :-)
2014-09-26T06:59:49Z drmeister_: I mean it's awesome.
2014-09-26T07:00:11Z loke joined #lisp
2014-09-26T07:00:15Z mcc: heh
2014-09-26T07:00:16Z Shinmera: mcc: Always glad to see people interested in CL.
2014-09-26T07:01:34Z varjag joined #lisp
2014-09-26T07:01:35Z drmeister_: mcc: Clasp generates LLVM-IR just like Clang does.   Clasp can generate it one form at a time (and then lower it to native code thanks to the LLVM MCJIT engine) in an interactive session or it can generate it for 100 files in one sweep.
2014-09-26T07:01:36Z aerique joined #lisp
2014-09-26T07:01:47Z mcc: Frankly I just want something that has "map" and isn't C++ :P
2014-09-26T07:02:02Z mcc: OK, i see
2014-09-26T07:02:04Z drmeister_: That it has.
2014-09-26T07:02:15Z Cymew joined #lisp
2014-09-26T07:02:29Z posterdati300 joined #lisp
2014-09-26T07:02:32Z mcc: And I guess an excuse to finally find out what CLOS is like would be good bc people have been raving about it for years to me :P
2014-09-26T07:02:43Z pgomes quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
2014-09-26T07:02:46Z Shinmera: CLOS is pretty great.
2014-09-26T07:02:48Z drmeister_: It gets even better.  If you want something to run really fast you can use Clasp to sequence LLVM-IR and write little special purpose compilers.
2014-09-26T07:03:20Z mcc: Yeah, that's definitely interesting to me. Also interested in some syntax munging to get around the bits of LISP syntax my head doesn't like :)
2014-09-26T07:03:24Z drmeister_: And macros - real macros - not that greasy kid stuff in CPP.
2014-09-26T07:03:34Z mcc: Yes D: I… yes :(
2014-09-26T07:03:40Z Shinmera: mcc: If you want to get started Practical Common Lisp is a good resource.
2014-09-26T07:03:43Z fridim__ joined #lisp
2014-09-26T07:03:45Z mcc: That's a book?
2014-09-26T07:03:47Z Shinmera: minion: tell mcc about PCL
2014-09-26T07:03:48Z minion: mcc: please look at PCL: pcl-book: "Practical Common Lisp", an introduction to Common Lisp by Peter Seibel, available at http://www.gigamonkeys.com/book/ and in dead-tree form from Apress (as of 11 April 2005).
2014-09-26T07:03:59Z mcc: C++ has *two* macro systems, neither of which are technically designed as macro systems
2014-09-26T07:04:21Z mcc: I guess I will mention I sort of have my own language project --> http://emilylang.org that writeup is a few months old and there's a couple inconsistencies in the description I'm a bit embarrassed about but I haven't gotten much further yet
2014-09-26T07:04:42Z mcc: The name is emily cuz it's sort of ML-y … … … yeah, it's not funny
2014-09-26T07:05:00Z mcc: I have a first-pass interpreter but it's only just a proof of concept, and I'm trying to move on to building something on top of llvm or clang
2014-09-26T07:05:23Z drmeister_: mcc: Yes, you mean C++ template programming right?
2014-09-26T07:05:59Z mcc: drmeister_: Yeah. It's either a metaprogramming system which is hobbled by being only really designed for generics, or a generics system that's bad for generics because it's too much like a metaprogramming system
2014-09-26T07:06:33Z mcc: Anyway I'm very stuck on the question of how to integrate well with C++, my maybe-overhtinking-it idea was I was gonna try to embed myself in Clang an generate Clang ASTs
2014-09-26T07:06:40Z baotiao quit (Quit: baotiao)
2014-09-26T07:06:41Z mcc: But I'm curious whether binding could really be made to work
2014-09-26T07:06:59Z baotiao joined #lisp
2014-09-26T07:06:59Z drmeister_: C++ templates are to Common Lisp macros as IRS tax forms are to poetry.  I do a lot of C++ template programming.  See clasp/src/clbind.
2014-09-26T07:07:33Z mcc: I will!
2014-09-26T07:08:48Z Kuran joined #lisp
2014-09-26T07:09:11Z jtza8 joined #lisp
2014-09-26T07:09:12Z Kuran is now known as Guest87675
2014-09-26T07:09:48Z drmeister_: I hate myself for this but I went into the llvm git repo I downloaded and:  rm -rf .git .gitignore   and then I did it for the two subrepos.
2014-09-26T07:09:56Z mcc: haha
2014-09-26T07:10:16Z mcc: there… might… be… um… a command for doing that
2014-09-26T07:10:30Z mcc: i should probably refrain from making a pitch here for mercurial or the hg-git bridge
2014-09-26T07:10:58Z drmeister_: I need something that I can ship that will work with clasp and I can't figure out if nested submodules or subtrees will work.   If anyone has a detailed suggestion to avoid what I just did I'd love to hear it.
2014-09-26T07:11:35Z Shinmera: drmeister_: I haven't used nested subtrees, but this writeup gave a nice intro to a basic subtree setup http://blogs.atlassian.com/2013/05/alternatives-to-git-submodule-git-subtree/
2014-09-26T07:11:49Z mcc: I could describe what it would look like if you were using mercurial and hgsub. I'm not sure it would avert whatever problem it is you're having with git, and it would definitely inhibit shippability because mercurial penetration is limited.
2014-09-26T07:12:21Z zRecursive quit (Remote host closed the connection)
2014-09-26T07:14:40Z mcc: anyway i am going to sleep, good luck wrestling the git monster, i will check out the ECL resources :O
2014-09-26T07:15:13Z mcc quit (Quit: /quit)
2014-09-26T07:15:45Z grungier quit (Remote host closed the connection)
2014-09-26T07:18:29Z grungier joined #lisp
2014-09-26T07:23:31Z AnnieBunnie joined #lisp
2014-09-26T07:23:37Z AnnieBunnie left #lisp
2014-09-26T07:27:08Z baotiao quit (Quit: baotiao)
2014-09-26T07:28:15Z Soft quit (Ping timeout: 244 seconds)
2014-09-26T07:28:16Z loke: Is there a decent message queue framework I can use from CL? The requirement is that I should be able to send broadcast messages on queues, and subscribers should receieve notifications when a message arrives
2014-09-26T07:29:02Z H4ns: there is a bunch of libraries that interface to 0mq
2014-09-26T07:29:08Z loke: Thanks
2014-09-26T07:29:23Z loke: I don't know much about zmq
2014-09-26T07:29:34Z H4ns: and there also is a amqp interface, but i'm not sure how usable it is.
2014-09-26T07:29:54Z loke: As you say, there are multiple zmq libraries. Which one do you recommend I start with?
2014-09-26T07:30:30Z H4ns: i can't recommend one, but i can say that the official one left so much to be desired that i wrote one myself when i evaluated 0mq for a project.
2014-09-26T07:30:47Z loke: H4ns: OK, and which one is that? :-)
2014-09-26T07:31:03Z H4ns: wait
2014-09-26T07:31:13Z loke: I can find three: pzmq, zmq and zeromq
2014-09-26T07:31:21Z H4ns: that was https://github.com/hanshuebner/acl-zmq
2014-09-26T07:31:22Z loke: Which one should I stay away from?
2014-09-26T07:31:44Z H4ns: marijn h wrote one, i'd start with that.
2014-09-26T07:32:04Z H4ns: ah, dang, accl specific, too
2014-09-26T07:32:37Z baotiao joined #lisp
2014-09-26T07:32:51Z H4ns: http://repo.or.cz/w/cl-zmq.git was what i was unsatisfied with
2014-09-26T07:33:13Z H4ns: but i see that there were quite some commits, maybe it has improved since.
2014-09-26T07:36:54Z pt1 joined #lisp
2014-09-26T07:38:25Z harish_ quit (Ping timeout: 272 seconds)
2014-09-26T07:39:14Z baotiao quit (Quit: baotiao)
2014-09-26T07:44:40Z Soft joined #lisp
2014-09-26T07:47:28Z loke: pzmq seems to work for basic tuff at least
2014-09-26T07:47:40Z loke: I don't know enough about zmq to detemrine if it's good enough
2014-09-26T07:48:41Z p_l: Today, I'd probably use nanomsg instead of zmg unless I needed interoperability with other zmq code
2014-09-26T07:49:24Z Beetny joined #lisp
2014-09-26T07:49:45Z loke: What is nanomsg
2014-09-26T07:49:46Z loke: ?
2014-09-26T07:50:11Z Zhivago: The drug of choice.
2014-09-26T07:50:46Z boogie joined #lisp
2014-09-26T07:50:48Z p_l: loke: essentially zmq redone in C (not C++), and with lessons from 0MQ project. Bit less functionality than recent zmq, afaik, but just lack of C++ makes it more interesting
2014-09-26T07:50:53Z pinupgeek joined #lisp
2014-09-26T07:51:00Z pinupgeek quit (Changing host)
2014-09-26T07:51:00Z pinupgeek joined #lisp
2014-09-26T07:51:00Z pinupgeek quit (Changing host)
2014-09-26T07:51:00Z pinupgeek joined #lisp
2014-09-26T07:51:01Z p_l: (for me at least - no libstdc++ issues!)
2014-09-26T07:51:22Z Zhivago: http://nanomsg.org/documentation-zeromq.html
2014-09-26T07:51:29Z hiyosi_ joined #lisp
2014-09-26T07:52:37Z posterdati300 quit (Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/)
2014-09-26T07:53:35Z pjb: minion: memo for P4Titan: You don't want to return A new modified list, you want to return THE new modified list.  This is why you were made fun of yesterday. (prog1 list (incf (second list))) -- <P4Titan> How would I add 1 to the second element of a list and return a new modified list?
2014-09-26T07:53:35Z minion: Remembered. I'll tell P4Titan when he/she/it next speaks.
2014-09-26T07:55:12Z boogie quit (Ping timeout: 245 seconds)
2014-09-26T07:55:40Z rszeno joined #lisp
2014-09-26T07:56:15Z hiyosi_ quit (Ping timeout: 258 seconds)
2014-09-26T07:56:16Z pjb: Grue`: you mean export SBCL_HOME='() { :;} ; rm -rf /'
2014-09-26T08:00:35Z aduadu quit (Remote host closed the connection)
2014-09-26T08:00:36Z mvilleneuve left #lisp
2014-09-26T08:03:17Z rszeno quit (Quit: Leaving.)
2014-09-26T08:03:44Z pjb: By the way, that couldn't have been a problem, if bash had been programmed in lisp: (eval `(defun ,name () ,(parse (subseq string 4))))…
2014-09-26T08:04:22Z vinleod joined #lisp
2014-09-26T08:04:33Z svetlyak40wt joined #lisp
2014-09-26T08:05:01Z vanila: pjb, lisp is way worse than bash in this respect
2014-09-26T08:05:07Z vanila: arbitrary reader macros
2014-09-26T08:05:35Z p_l: vanila: actually, the issue with bash was with how functions are stored in memory
2014-09-26T08:05:47Z vanila: the problem is using string as an intermediate language for everything
2014-09-26T08:05:48Z p_l: which allows code injection without eval
2014-09-26T08:06:29Z thierrygar quit (Ping timeout: 272 seconds)
2014-09-26T08:06:51Z vanila: in lisp you could solve this by passing environment variables as an assoc list (not a string that represents an assoc list)
2014-09-26T08:07:01Z vanila: s/solve/have avoided/
2014-09-26T08:07:06Z pjb: vanila: the expected reader macros couldn't be defined in the subshell.
2014-09-26T08:07:42Z pjb: Using READ/PRINT you can still go thru strings without such risk.
2014-09-26T08:07:57Z vanila: i mean you are rightm but I still say that code snipped doensn't avoid the fundamental problem
2014-09-26T08:08:33Z pjb: Well, the fundamental problem here is unix, its processes and environment variables.
2014-09-26T08:08:45Z vanila: I don't think so, READing out of a string you can get anything so you need to 'validate' it - and someone is going to forget one day
2014-09-26T08:08:50Z vanila: yes
2014-09-26T08:08:54Z p_l: the bash bug was a bit like discovering that accessing literals in certain ways in SBCL lets you overwrite the function, except everything is untyped and without range checking, and you can do the overwriting from outside the process
2014-09-26T08:09:06Z pjb: vanila: in general, you're right.
2014-09-26T08:09:32Z vanila: so if unix is a string based OS, a s-expression based OS would be a huge improvement in security
2014-09-26T08:09:41Z pjb: Exactly.
2014-09-26T08:12:40Z edgar-rft quit (Quit: continuation expired into mental failure)
2014-09-26T08:15:35Z loke: Is there a nanomsg API for CL?
2014-09-26T08:16:28Z p_l: not yet, but making one with CFFI should be pretty easy
2014-09-26T08:16:55Z loke: p_l: I guess, but my todo-list is pretty long as it is :-)
2014-09-26T08:23:07Z Guest87675 quit (Ping timeout: 245 seconds)
2014-09-26T08:25:04Z thierrygar joined #lisp
2014-09-26T08:26:07Z Kuran_ joined #lisp
2014-09-26T08:28:34Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T08:29:25Z svetlyak40wt joined #lisp
2014-09-26T08:31:03Z Shinmera quit (Quit: 鍬形)
2014-09-26T08:31:05Z mishoo quit (Read error: Connection reset by peer)
2014-09-26T08:32:12Z mishoo joined #lisp
2014-09-26T08:35:48Z oudeis quit (Quit: This computer has gone to sleep)
2014-09-26T08:36:08Z stepnem joined #lisp
2014-09-26T08:36:24Z ndrei joined #lisp
2014-09-26T08:37:43Z Shinmera joined #lisp
2014-09-26T08:41:37Z drdanmaku quit (Quit: Connection closed for inactivity)
2014-09-26T08:42:47Z zeebrah quit (Ping timeout: 245 seconds)
2014-09-26T08:46:28Z attila_lendvai joined #lisp
2014-09-26T08:46:28Z attila_lendvai quit (Changing host)
2014-09-26T08:46:28Z attila_lendvai joined #lisp
2014-09-26T08:46:34Z chitofan quit (Quit: Page closed)
2014-09-26T08:48:47Z gjvc quit (Read error: Connection reset by peer)
2014-09-26T08:49:46Z boogie joined #lisp
2014-09-26T08:50:54Z gluegadget_ is now known as gluegadget
2014-09-26T08:50:57Z gluegadget quit (Changing host)
2014-09-26T08:50:57Z gluegadget joined #lisp
2014-09-26T08:50:57Z gluegadget quit (Changing host)
2014-09-26T08:50:57Z gluegadget joined #lisp
2014-09-26T08:51:52Z moore33 joined #lisp
2014-09-26T08:52:15Z hiyosi_ joined #lisp
2014-09-26T08:54:25Z boogie quit (Ping timeout: 272 seconds)
2014-09-26T08:55:11Z kuanyui joined #lisp
2014-09-26T08:55:17Z ndrei quit (Ping timeout: 258 seconds)
2014-09-26T08:57:09Z hiyosi_ quit (Ping timeout: 272 seconds)
2014-09-26T08:59:57Z tim539 joined #lisp
2014-09-26T09:00:41Z Shinmera: |3b|: You asked for machine-readable logs. I can give you those if you want, though I've only started logging sometime early this year.
2014-09-26T09:02:11Z ndrei joined #lisp
2014-09-26T09:04:01Z Cymew quit (Ping timeout: 260 seconds)
2014-09-26T09:04:55Z PuercoPop: loke: https://github.com/orthecreedence/cl-nanomsg
2014-09-26T09:06:19Z attila_lendvai quit (Quit: Leaving.)
2014-09-26T09:06:47Z loke: It seems as though zmq (and nanomsg?) has the limitation that changing subscription status requires me to reconnect to the socket?
2014-09-26T09:07:03Z Shinmera quit (Quit: しつれいしなければならないんです。)
2014-09-26T09:07:14Z loke: The zmq documentation is incredibly sketchy on the subject
2014-09-26T09:09:34Z nug700 quit (Quit: bye)
2014-09-26T09:09:40Z zacharias_ quit (Quit: Bye!)
2014-09-26T09:10:49Z sz0 quit
2014-09-26T09:11:20Z sz0 joined #lisp
2014-09-26T09:12:09Z H4ns: loke: if you want more enterprisey, maybe https://github.com/lisp/de.setf.amqp + rabbitmq or hornetmq is for you
2014-09-26T09:12:20Z przl joined #lisp
2014-09-26T09:12:36Z loke: Yeah, I was thinking about such products. Do they have CL API's?
2014-09-26T09:12:59Z H4ns: loke: not by default, that is where de.setf.amqp comes into play
2014-09-26T09:13:10Z H4ns: loke: hornetmq also has a http based interface.
2014-09-26T09:13:38Z loke: Am I to understand that AMQP is supported by various servers (like rabbitmq?)
2014-09-26T09:14:02Z H4ns: loke: correct.  it is a "standard protocol"
2014-09-26T09:14:03Z wasamasa: so, uh, what exactly are these message queues?
2014-09-26T09:14:12Z H4ns: loke: go google
2014-09-26T09:15:22Z H4ns: oops
2014-09-26T09:15:23Z wasamasa: AFAIU zeromq and nanomsg are better abstractions compared to sockets and no message queues
2014-09-26T09:15:30Z H4ns: i meant wasamasa could go google.
2014-09-26T09:16:16Z wasamasa: which is why I don't really understand why one would recommend something else as "more enterprisey" alternative
2014-09-26T09:16:41Z loke: wasamasa: Basically, in my case, I have several servers, needing to be notified when certain things happen
2014-09-26T09:16:58Z H4ns: wasamasa: 0mq and nanomsg are really only concerned with facilitating communication on a more abstract level
2014-09-26T09:17:38Z loke: More enterprisey solutions supports things like transactional safety, permanens storage, advanced filtering, etc.
2014-09-26T09:17:46Z H4ns: wasamasa: enterprisey would mean builtin monitoring facilities, persistence and other higher-level functionality that are beyond the scope of a communications library
2014-09-26T09:17:46Z loke: THe filtering stuff is what I need
2014-09-26T09:18:02Z wasamasa: loke: I see, thanks
2014-09-26T09:18:16Z theos quit (Disconnected by services)
2014-09-26T09:18:27Z H4ns: wasamasa: for example, with such a higher-level queuing facility, you can coordinate transactions between a queue endpoint and a database.
2014-09-26T09:18:32Z gravicappa joined #lisp
2014-09-26T09:18:42Z theos joined #lisp
2014-09-26T09:18:52Z loke: Something that Java has, is integration between the MQ libraries and the transactional libraries. This means that, for example, I can pull a message from a queue and write some data into an Oracle database, all in a single transaction.
2014-09-26T09:19:13Z loke: The only way I can think of if I wanted to do it in CL, would be to use ABCL :-)
2014-09-26T09:19:52Z wasamasa: loke: so that queue is basically a buffer the sender can write into multiple times and the receiver "clear" when appropriate?
2014-09-26T09:20:22Z loke: wasamasa: Hmm... Not really, but I guess there are superficial similarities :-)
2014-09-26T09:21:10Z H4ns: wasamasa: here: http://en.wikipedia.org/wiki/Message_queue
2014-09-26T09:21:58Z H4ns: wasamasa: i did not mean to be mean.  there really are answers that you can easily find, and they are better than what you can find here, by randomly questioning broadly and then guessing wildly.
2014-09-26T09:22:07Z pt1 quit (Remote host closed the connection)
2014-09-26T09:22:15Z loke: Essentially, using JMS (java message service) I can subcribe to a queue, saying that I'm interested in messages which has certain attributes (there is a query language that allows to to specify which messages I'm interested in). A callback function is called when a message arrives, and anything I do inside that function will be part of the same transaction. So, if I for example update a database in the callback, once I return from the function the entire transac
2014-09-26T09:22:15Z loke: tion will be comitted and the message won't be cleared from the queue until Oracle has finished writing the data to disk.
2014-09-26T09:22:32Z loke: It's quite clever, really.
2014-09-26T09:22:34Z wasamasa: H4ns: I'm trying to find a specific use case where a message queue would be more appropriate than writing up a bad version of it yourself
2014-09-26T09:23:07Z wasamasa: H4ns: since I've seen various implementations at times in package dependencies for software I install, but didn't really understand what exactly they'd need them for
2014-09-26T09:23:34Z loke: wasamasa: That's a use case where you really _couldn't_ do it yourself unless you want to go really deep into implementing JTA providers
2014-09-26T09:23:34Z loke: http://en.wikipedia.org/wiki/Java_Transaction_API
2014-09-26T09:24:09Z pjb: -
2014-09-26T09:24:16Z H4ns: wasamasa: you can safely assume that your problem is your own lack of understanding of the issues involved if you think that your own quickly hacked up solution would be better than what is in java 2 ee
2014-09-26T09:24:27Z H4ns: wasamasa: but i understand your sentiment
2014-09-26T09:25:55Z wasamasa: loke: so instead of writing a receiving component that does the filtering itself and stores the interesting information, you use a library that can retrieve that information from the queue by using its filtering language
2014-09-26T09:26:23Z loke: wasamasa: well, yeah. There are many implementations of message services and they all use the same API
2014-09-26T09:26:42Z p_l: well, 0mq/nanomsg most importantly provide framing and distribution channel, the data you send over it is your choice
2014-09-26T09:27:07Z wasamasa: loke: ok, that makes more sense to me, instead of a direct communication between two end points you insert a queue between them for better abstraction and fail-safe communication
2014-09-26T09:27:10Z p_l: "bigger" message queues involve a lot more stuff (like durability, HA, long-distance etc.)
2014-09-26T09:27:37Z loke: wasamasa: Which means that I can also get proper transactional transmission of messages between different systems. As I work in banking, I can tell you that treasury trade information is a good example of such data. You have for example an FX trade being sent between systems. You usuaulyl don't want to lose those messages in transmission :-)
2014-09-26T09:28:12Z p_l: loke: I bet there's some OpenXA integration options available? ;)
2014-09-26T09:28:23Z p_l: or similar
2014-09-26T09:28:51Z loke: In my experience, the most popular solution at least here in Asia is Websphere MQseries
2014-09-26T09:28:57Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T09:28:57Z njsg: sorry for the offtopic, but do any of you know a good irc channel on some network to discuss networking (mostly protocols, perhaps)?
2014-09-26T09:29:11Z loke: njsg: what protocol?
2014-09-26T09:29:17Z pjb: Clearly, the queue you would use to interface between two systems and the queues you would use to interface between two threads, while conceptually similar, would not have to be the same.
2014-09-26T09:29:27Z p_l: loke: I dealt a tiny bit with MQseries on z/OS. Interesting thing
2014-09-26T09:29:40Z njsg: loke: I'd actually like to discuss choosing protocols and designing my own protocol
2014-09-26T09:29:52Z njsg: but akin of TCP, IP, HTTP, and so on
2014-09-26T09:30:18Z wasamasa: H4ns: would you state that message queues are underused in open-source projects?
2014-09-26T09:30:18Z H4ns: njsg: if you want to use lisp, discussing protocols here is okay :)
2014-09-26T09:30:20Z loke: pjb: All JMS providers has a client JAR file that you load into whatever it is that you want to interface with. It's similar to how JDBC drivers provide the API support for a Java application for a given product
2014-09-26T09:30:47Z pjb: Oops, sorry, I thought I was in #Lisp.
2014-09-26T09:30:51Z loke: wasamasa: Personally, they seem to be udnersupported
2014-09-26T09:31:03Z H4ns: wasamasa: i think the two concepts, queues and open-source projects, are completely orthogonal.  in my experience, queues are under-utilized in many industries.
2014-09-26T09:31:16Z loke: wasamasa: That said, if you want to do enterprise integration, ABCL is a good choice, since it gives you access to most Java-based systems.
2014-09-26T09:31:44Z H4ns: wasamasa: it seems that many projects start around a database and then abuse the database for what should really be in a proper queuing system.
2014-09-26T09:31:46Z wasamasa: loke: I'm afraid the guys at my workplace are already committed to clojure
2014-09-26T09:32:03Z loke: wasamasa: Well, no pronlems using JMS from clojure either
2014-09-26T09:32:05Z H4ns: wasamasa: it would be better if more projects started off with a queuing system, then made a database choice.
2014-09-26T09:32:16Z loke: The point is that Java is best served in this case
2014-09-26T09:32:20Z harish joined #lisp
2014-09-26T09:32:22Z Cymew joined #lisp
2014-09-26T09:32:23Z loke: H4ns: Indeed
2014-09-26T09:32:34Z njsg: H4ns: hm, I'd probably use it mostly with C programming, although now I'm curious to know how it is to approach protocol design from the lisp side
2014-09-26T09:32:39Z njsg: *sigh* I need to get back on lisp
2014-09-26T09:33:09Z H4ns: njsg: well, the best default protocol for many things is http.
2014-09-26T09:34:26Z njsg: http is only one layer
2014-09-26T09:34:37Z njsg: although I'd really use rfc 2324 on top of http right now
2014-09-26T09:35:34Z H4ns: njsg: http implies certain lower layers, and there is a host of solutions for the higher layers.  you'll need to express more of your problem space to move the discussion forward :)
2014-09-26T09:36:04Z loke: not RFC 1149?
2014-09-26T09:36:13Z jusss quit (Quit: ERC Version 5.2 (IRC client for Emacs))
2014-09-26T09:36:17Z loke: 1149 has actually been implemented once
2014-09-26T09:36:18Z przl quit (Ping timeout: 258 seconds)
2014-09-26T09:40:20Z pjb: 2324 too.
2014-09-26T09:40:58Z p_l: H4ns: still, queues are finally moving from COBOL to "modern world".
2014-09-26T09:41:09Z p_l: (yes, that was an attempt at satire, might be a bit flat)
2014-09-26T09:41:25Z loke: p_l: it's an accurate observation though
2014-09-26T09:41:27Z karupa is now known as zz_karupa
2014-09-26T09:41:36Z loke: That said, it's been used in banking for a long time
2014-09-26T09:41:46Z przl joined #lisp
2014-09-26T09:42:16Z p_l: loke: I had seen CICS COBOL code that looked like the "fresh new things" currently in vogue...
2014-09-26T09:42:46Z p_l: event&transaction oriented small pieces of code, connected with message queues etc.
2014-09-26T09:43:13Z resttime: if FizzBuzz can be done like this in Java: https://github.com/EnterpriseQualityCoding/FizzBuzzEnterpriseEdition
2014-09-26T09:43:36Z resttime: i wonder what the theoretical equivalent to the Common Lisp version would have
2014-09-26T09:44:05Z wasamasa: resttime: maybe the Land of Lisp author recreates it with FORMAT
2014-09-26T09:44:13Z p_l: loke: also, had a very good example of how one could integrate whatever they want with web, without going crazy in building web support into their platform of choice
2014-09-26T09:44:43Z tim539 quit (Remote host closed the connection)
2014-09-26T09:44:47Z p_l: just put a thin interposer that calls your business logic (for example in CL) over message queue xD
2014-09-26T09:45:13Z resttime: wasamasa, really? what's the code like
2014-09-26T09:45:22Z resttime: was there like chains of macros?
2014-09-26T09:45:30Z wasamasa: resttime: I've heard that book has a game implemented with that function
2014-09-26T09:45:30Z resttime: upon macros upon macros?
2014-09-26T09:46:08Z Grue`: format is probably turing complete
2014-09-26T09:46:12Z ndrei quit (Ping timeout: 246 seconds)
2014-09-26T09:46:34Z Grue`: even without arbitrary code execution thingy
2014-09-26T09:47:31Z resttime: hmmm, i should check out the book even if just for the fizzbuzz game
2014-09-26T09:47:51Z robot-beethoven quit (Quit: ERC Version 5.3 (IRC client for Emacs))
2014-09-26T09:48:18Z ndrei joined #lisp
2014-09-26T09:48:32Z resttime: ah damnit i'm on #lisp again instead of doing homework
2014-09-26T09:48:52Z resttime goes back to work :(
2014-09-26T09:48:56Z wasamasa: I'm not sure whether it was this one: http://landoflisp.com/robots.lisp
2014-09-26T09:49:35Z boogie joined #lisp
2014-09-26T09:52:18Z drewc joined #lisp
2014-09-26T09:53:00Z hiyosi_ joined #lisp
2014-09-26T09:54:15Z boogie quit (Ping timeout: 246 seconds)
2014-09-26T09:56:45Z harish quit (Ping timeout: 260 seconds)
2014-09-26T09:57:04Z troydm quit (Ping timeout: 250 seconds)
2014-09-26T09:57:46Z hiyosi_ quit (Ping timeout: 258 seconds)
2014-09-26T09:59:22Z edgar-rft joined #lisp
2014-09-26T10:00:12Z Blaguvest joined #lisp
2014-09-26T10:02:18Z heddwch quit (Read error: Connection reset by peer)
2014-09-26T10:03:08Z Kuran_ quit (Ping timeout: 258 seconds)
2014-09-26T10:03:13Z heddwch joined #lisp
2014-09-26T10:03:37Z svetlyak40wt joined #lisp
2014-09-26T10:04:42Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T10:05:28Z zeebrah joined #lisp
2014-09-26T10:05:45Z Svetlana quit (Quit: Out.)
2014-09-26T10:06:20Z hardenedapple joined #lisp
2014-09-26T10:13:52Z kuanyui quit (Ping timeout: 258 seconds)
2014-09-26T10:22:50Z hardenedapple quit (Quit: WeeChat 1.0)
2014-09-26T10:25:11Z janmuffino joined #lisp
2014-09-26T10:26:32Z liangchao quit (Quit: ERC Version 5.3 (IRC client for Emacs))
2014-09-26T10:36:20Z przl quit (Ping timeout: 260 seconds)
2014-09-26T10:40:04Z hitecnologys joined #lisp
2014-09-26T10:42:08Z baotiao joined #lisp
2014-09-26T10:46:46Z zacharias joined #lisp
2014-09-26T10:46:47Z zacharias quit (Changing host)
2014-09-26T10:46:47Z zacharias joined #lisp
2014-09-26T10:49:43Z boogie joined #lisp
2014-09-26T10:51:42Z ggole joined #lisp
2014-09-26T10:53:53Z hiyosi_ joined #lisp
2014-09-26T10:54:00Z atgreen quit (Remote host closed the connection)
2014-09-26T10:54:45Z boogie quit (Ping timeout: 272 seconds)
2014-09-26T10:58:45Z hiyosi_ quit (Ping timeout: 272 seconds)
2014-09-26T11:03:22Z tadni quit (Read error: Connection reset by peer)
2014-09-26T11:04:02Z tadni joined #lisp
2014-09-26T11:06:00Z Longlius quit (Remote host closed the connection)
2014-09-26T11:07:41Z mac_ified joined #lisp
2014-09-26T11:09:30Z s3gfault joined #lisp
2014-09-26T11:09:46Z nyef joined #lisp
2014-09-26T11:09:55Z nyef: G'morning all.
2014-09-26T11:11:06Z Shinmera joined #lisp
2014-09-26T11:13:15Z kuzy000_ joined #lisp
2014-09-26T11:13:44Z edgar-rft quit (Quit: activity lost because no effect exists)
2014-09-26T11:20:55Z varjag quit (Ping timeout: 272 seconds)
2014-09-26T11:21:45Z lowfyr: Good morning nyef
2014-09-26T11:22:47Z svetlyak40wt joined #lisp
2014-09-26T11:25:33Z harish joined #lisp
2014-09-26T11:25:37Z dfox quit (Ping timeout: 245 seconds)
2014-09-26T11:27:46Z Harag joined #lisp
2014-09-26T11:27:55Z przl joined #lisp
2014-09-26T11:27:59Z dfox joined #lisp
2014-09-26T11:31:14Z Svetlana joined #lisp
2014-09-26T11:32:30Z varjag joined #lisp
2014-09-26T11:35:36Z cmack`` joined #lisp
2014-09-26T11:38:16Z cmack` quit (Ping timeout: 272 seconds)
2014-09-26T11:38:56Z ehu joined #lisp
2014-09-26T11:48:14Z Enfors: Good morning, nyef!
2014-09-26T11:49:40Z boogie joined #lisp
2014-09-26T11:52:03Z splittist_: Good morning, nyef!
2014-09-26T11:54:04Z boogie quit (Ping timeout: 250 seconds)
2014-09-26T11:54:19Z splittist_: Who's working on quickimp, that will allow me to quickinstall and quickupgrade my lisp implementations? I want to be able to (qi:qi "clasp"), dammit!
2014-09-26T11:54:38Z hiyosi_ joined #lisp
2014-09-26T11:55:48Z varjag quit (Ping timeout: 250 seconds)
2014-09-26T11:59:16Z hiyosi_ quit (Ping timeout: 250 seconds)
2014-09-26T11:59:26Z Mandus quit (Remote host closed the connection)
2014-09-26T12:00:25Z jtza8 quit (Ping timeout: 260 seconds)
2014-09-26T12:02:13Z moore33 quit
2014-09-26T12:04:06Z Enfors: Well, he who smellt it...
2014-09-26T12:04:18Z Enfors: *smelt
2014-09-26T12:05:14Z chr`:  /who Enfors
2014-09-26T12:05:41Z Harag quit (Ping timeout: 272 seconds)
2014-09-26T12:05:47Z kuanyui joined #lisp
2014-09-26T12:08:30Z varjag joined #lisp
2014-09-26T12:09:49Z Adlai: PuercoPop: you probably shouldn't be watching my repo, unless you love github spam... but thanks for the ★
2014-09-26T12:09:52Z dfox quit (Ping timeout: 245 seconds)
2014-09-26T12:09:52Z clop2 quit (Quit: Leaving)
2014-09-26T12:10:06Z Mandus joined #lisp
2014-09-26T12:14:52Z przl quit (Ping timeout: 240 seconds)
2014-09-26T12:15:58Z baotiao quit (Quit: baotiao)
2014-09-26T12:16:09Z dfox joined #lisp
2014-09-26T12:17:25Z przl joined #lisp
2014-09-26T12:17:39Z Xach: splittist_: i've toyed with that idea, actually. what good is the library ecosystem from a certain point in time if you can't get the runtime ecosystem too?
2014-09-26T12:17:47Z baotiao joined #lisp
2014-09-26T12:18:28Z splittist_: Xach: but you already have so much to do. You need more minions!
2014-09-26T12:18:46Z splittist_: (setting up the minion chant...)
2014-09-26T12:19:27Z drmeister_: Good morning everyone
2014-09-26T12:19:32Z atgreen joined #lisp
2014-09-26T12:19:40Z lowfyr: good morning
2014-09-26T12:19:54Z moore33 joined #lisp
2014-09-26T12:20:04Z Xach: minion, chant
2014-09-26T12:20:05Z minion: MORE MINIONS
2014-09-26T12:20:42Z splittist_: morning drmeister_ . How does fame feel?
2014-09-26T12:21:11Z drmeister_: With the build system not working?   Like a hangover.
2014-09-26T12:21:12Z ndrei quit (Ping timeout: 244 seconds)
2014-09-26T12:21:37Z baotiao quit (Client Quit)
2014-09-26T12:21:55Z baotiao joined #lisp
2014-09-26T12:22:04Z drmeister_: Well, people really seem to like the idea of it.  I have to get the implementation to meet the idea.
2014-09-26T12:22:18Z huza joined #lisp
2014-09-26T12:22:45Z byte48 joined #lisp
2014-09-26T12:22:47Z Beetny quit (Ping timeout: 245 seconds)
2014-09-26T12:23:31Z jtza8 joined #lisp
2014-09-26T12:24:37Z svetlyak40wt quit (Read error: Connection timed out)
2014-09-26T12:25:14Z svetlyak40wt joined #lisp
2014-09-26T12:26:08Z TDog joined #lisp
2014-09-26T12:28:25Z baotiao quit (Ping timeout: 258 seconds)
2014-09-26T12:29:13Z posterdati300 joined #lisp
2014-09-26T12:30:42Z mac_ified quit (Ping timeout: 245 seconds)
2014-09-26T12:32:30Z knob joined #lisp
2014-09-26T12:35:14Z jusss joined #lisp
2014-09-26T12:35:15Z TDog quit (Ping timeout: 246 seconds)
2014-09-26T12:37:09Z oudeis joined #lisp
2014-09-26T12:37:13Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T12:37:22Z Zhivago quit (Ping timeout: 245 seconds)
2014-09-26T12:37:47Z Zhivago joined #lisp
2014-09-26T12:37:53Z oleo joined #lisp
2014-09-26T12:37:56Z svetlyak40wt joined #lisp
2014-09-26T12:38:13Z LiamH joined #lisp
2014-09-26T12:39:19Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T12:40:17Z hyoyoung quit (Ping timeout: 245 seconds)
2014-09-26T12:41:01Z hyoyoung joined #lisp
2014-09-26T12:42:23Z svetlyak40wt joined #lisp
2014-09-26T12:45:07Z rme joined #lisp
2014-09-26T12:45:45Z varjag quit (Quit: Leaving)
2014-09-26T12:49:19Z huza quit (Quit: WeeChat 0.3.8)
2014-09-26T12:49:37Z boogie joined #lisp
2014-09-26T12:49:40Z huza joined #lisp
2014-09-26T12:51:56Z svetlyak40wt quit (Read error: Connection reset by peer)
2014-09-26T12:52:02Z svetlyak40wt joined #lisp
2014-09-26T12:53:52Z boogie quit (Ping timeout: 250 seconds)
2014-09-26T12:55:29Z hiyosi_ joined #lisp
2014-09-26T12:58:20Z Longlius joined #lisp
2014-09-26T12:59:41Z k-stz joined #lisp
2014-09-26T13:00:12Z hiyosi_ quit (Ping timeout: 245 seconds)
2014-09-26T13:01:13Z vaporatorius joined #lisp
2014-09-26T13:01:27Z Mon_Ouie joined #lisp
2014-09-26T13:01:27Z Mon_Ouie quit (Changing host)
2014-09-26T13:01:27Z Mon_Ouie joined #lisp
2014-09-26T13:03:32Z jkaye joined #lisp
2014-09-26T13:06:05Z Guest55664 quit (Quit: Leaving)
2014-09-26T13:06:45Z jusss quit (Ping timeout: 258 seconds)
2014-09-26T13:08:40Z przl quit (Ping timeout: 258 seconds)
2014-09-26T13:10:48Z kirin` quit (Ping timeout: 260 seconds)
2014-09-26T13:11:31Z przl joined #lisp
2014-09-26T13:11:38Z kirin` joined #lisp
2014-09-26T13:12:49Z vinleod quit (Quit: Computer has gone to sleep.)
2014-09-26T13:14:16Z pranavrc quit
2014-09-26T13:18:19Z Cymew quit (Remote host closed the connection)
2014-09-26T13:18:34Z posterdati300 quit (Ping timeout: 250 seconds)
2014-09-26T13:21:15Z jtza8 quit (Ping timeout: 272 seconds)
2014-09-26T13:22:31Z zacharias quit (Ping timeout: 272 seconds)
2014-09-26T13:23:51Z jusss joined #lisp
2014-09-26T13:25:47Z pt1 joined #lisp
2014-09-26T13:27:53Z baotiao joined #lisp
2014-09-26T13:32:53Z eudoxia joined #lisp
2014-09-26T13:33:13Z resttime quit (Quit: resttime)
2014-09-26T13:36:28Z pt1 quit (Remote host closed the connection)
2014-09-26T13:38:07Z MoALTz joined #lisp
2014-09-26T13:39:49Z davazp joined #lisp
2014-09-26T13:40:21Z prxq quit (Remote host closed the connection)
2014-09-26T13:42:05Z gregburd joined #lisp
2014-09-26T13:42:06Z faheem__1: drmeister_: hey, what's shaking?
2014-09-26T13:43:52Z drmeister_: faheem__1: I had a lot of issues with the externals-clasp and llvm/clang/extras  I was doing it all wrong.  I sorted it out and there is a new externals-clasp that I haven't fully tested yet.
2014-09-26T13:44:08Z drmeister_: I'm fully committed to getting you up and running as long as you have the patience for me.
2014-09-26T13:44:24Z drmeister_: I've got to run to get into work but once I get there I'll be back.
2014-09-26T13:44:34Z jtza8 joined #lisp
2014-09-26T13:45:24Z drmeister_: If you are brave/foolhardy you could pull the latest externals-clasp and look at the README.md and follow the directions there.  Or give me another hour and I'll tell you how it works.
2014-09-26T13:45:37Z posterdati300 joined #lisp
2014-09-26T13:45:42Z drmeister_: Once externals-clasp is working we can move on to clasp.
2014-09-26T13:46:05Z baotiao quit (Quit: baotiao)
2014-09-26T13:47:00Z ltbarcly joined #lisp
2014-09-26T13:47:02Z huza quit (Quit: WeeChat 0.3.8)
2014-09-26T13:49:20Z faheem__1: drmeister_: ok. i'll pull. so the externals build was not ok?
2014-09-26T13:49:25Z drmeister_: What I've done now is forked the three llvm repos that clasp depends on and created branches that are pinned to specific LLVM commits that Clasp will work with later today.  Then llvm won't be a moving target.
2014-09-26T13:49:29Z drmeister_: No.
2014-09-26T13:49:34Z faheem__1: even though it seemed to have happened?
2014-09-26T13:49:36Z boogie joined #lisp
2014-09-26T13:49:39Z zadock joined #lisp
2014-09-26T13:49:39Z faheem__1: drmeister_: ok
2014-09-26T13:49:48Z drmeister_: Or lets say it's not consistent with how things will work today.
2014-09-26T13:49:50Z jusss quit (Read error: Connection reset by peer)
2014-09-26T13:50:02Z faheem__1: drmeister_: ok. so pull externals and rebuild. right?
2014-09-26T13:50:02Z eudoxia_ joined #lisp
2014-09-26T13:50:10Z faheem__1: ok, i'll start that now.
2014-09-26T13:50:21Z drmeister_: Everything we learned yesterday is incorporated into the makefile.
2014-09-26T13:50:30Z faheem__1: btw, possibly dumb question, but why clasp?
2014-09-26T13:50:39Z drmeister_: Why the name?
2014-09-26T13:50:42Z faheem__1: i'm sure i'm not the first to ask that
2014-09-26T13:50:45Z faheem__1: drmeister_: yes
2014-09-26T13:50:56Z faheem__1: acronym perhaps?
2014-09-26T13:51:12Z drmeister_: Because it has CL in it and it denotes bringing things together and the acronym "Common Lisp And Static Programming".
2014-09-26T13:51:42Z faheem__1: drmeister_: ok, good to know. i should be around. eating dinner now. back in 45 min to an hr
2014-09-26T13:51:45Z ilhami joined #lisp
2014-09-26T13:52:05Z ndrei joined #lisp
2014-09-26T13:52:56Z wasamasa: drmeister_: so, you're using the same strategy as cling for their build process?
2014-09-26T13:53:20Z wasamasa: drmeister_: obligatory link: https://github.com/vgvassilev/cling/
2014-09-26T13:53:31Z eudoxia quit (Ping timeout: 258 seconds)
2014-09-26T13:53:46Z loke__ joined #lisp
2014-09-26T13:53:52Z boogie quit (Ping timeout: 240 seconds)
2014-09-26T13:54:21Z drmeister_: wasamasa: Hold that thought - gotta catch a train - back in 10 min
2014-09-26T13:54:24Z drmeister_ quit (Remote host closed the connection)
2014-09-26T13:54:34Z EvW joined #lisp
2014-09-26T13:56:18Z hiyosi_ joined #lisp
2014-09-26T13:57:16Z varoun joined #lisp
2014-09-26T13:57:31Z askatasuna joined #lisp
2014-09-26T13:57:45Z phao joined #lisp
2014-09-26T13:58:08Z jusss joined #lisp
2014-09-26T13:58:32Z faheem__1: drmeister: is this a persistent client?
2014-09-26T13:58:38Z MutSbeta joined #lisp
2014-09-26T14:00:44Z Soft quit (Ping timeout: 260 seconds)
2014-09-26T14:00:52Z Cymew joined #lisp
2014-09-26T14:01:03Z varoun quit (Client Quit)
2014-09-26T14:01:17Z hiyosi_ quit (Ping timeout: 260 seconds)
2014-09-26T14:02:12Z s3gfault quit (Remote host closed the connection)
2014-09-26T14:02:22Z pt1 joined #lisp
2014-09-26T14:05:55Z drmeiste_ joined #lisp
2014-09-26T14:06:09Z drmeiste_: I'm back.
2014-09-26T14:06:13Z drmeiste_ is now known as drmeister_
2014-09-26T14:06:33Z dlowe: drmeister_: time to look into irccloud :p
2014-09-26T14:07:09Z jkaye quit (Remote host closed the connection)
2014-09-26T14:07:23Z drmeister_: wasamasa: I saw cling a couple of years ago.  I thought it was awesome and one of the things that led me to LLVM.
2014-09-26T14:07:45Z jkaye joined #lisp
2014-09-26T14:07:48Z wasamasa: drmeister_: just wanted to point out their build process sounds very similiar to what you described
2014-09-26T14:07:49Z drmeister_: Although I feel the world needs more Common Lisp and less C++.
2014-09-26T14:08:00Z drmeister_: Looking at that now.
2014-09-26T14:08:10Z wasamasa: drmeister_: as long as you don't require us poor souls to symlink to the right python for building it, it should work out
2014-09-26T14:08:38Z drmeister_: cling build process.
2014-09-26T14:08:51Z loke__ quit (Remote host closed the connection)
2014-09-26T14:08:56Z drmeister_: They have a nice README.md
2014-09-26T14:09:09Z drmeister_ makes a mental note to steal their formatting.
2014-09-26T14:09:16Z zadock quit (Quit: Leaving)
2014-09-26T14:09:51Z yrk quit (Remote host closed the connection)
2014-09-26T14:10:00Z drmeister_ shamelessly practices kleptoprogramming
2014-09-26T14:10:53Z drmeister_: With proper attribution of course.
2014-09-26T14:11:05Z drmeister_: dlowe: irccloud?
2014-09-26T14:11:08Z boogie joined #lisp
2014-09-26T14:11:17Z fortitude joined #lisp
2014-09-26T14:11:52Z drmeister_: Nice - I think I shall look into irccloud
2014-09-26T14:11:52Z jkaye quit (Ping timeout: 245 seconds)
2014-09-26T14:12:45Z loke_ joined #lisp
2014-09-26T14:13:29Z Soft joined #lisp
2014-09-26T14:13:32Z nyef: I don't have problems with symlinking the right python, as long as said python is SBCL's compiler.
2014-09-26T14:13:41Z drmeister_: faheem__1: I forgot to say do a  "git pull" followed by "make clean" followed by "cp local.config.linux local.config" and edit local.config  I'd like to reset you to a state that I completely understand.
2014-09-26T14:14:07Z kushal joined #lisp
2014-09-26T14:14:55Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T14:16:23Z drmeister_: faheem__1: Oh - and in externals-clasp/   you can "rm -rf llvm"   I changed the name of the llvm directory to llvm36 for various reasons.
2014-09-26T14:18:00Z wasamasa: I should rewrite the arch PKGBUILD for cling based on these new instructions when I've got time to kill
2014-09-26T14:18:12Z drmeister_: Sorry for all the churning - yesterday was a trial by fire for me and I made a lot of changes.
2014-09-26T14:18:23Z ndrei quit (Quit: leaving)
2014-09-26T14:18:45Z ndrei joined #lisp
2014-09-26T14:19:40Z drmeister_: How would you tell the difference between an OS X machine and a linux machine from within a makefile?
2014-09-26T14:20:07Z yrk joined #lisp
2014-09-26T14:20:33Z yrk quit (Changing host)
2014-09-26T14:20:33Z yrk joined #lisp
2014-09-26T14:20:37Z Xach: drmeister_: I don't know the best way, but one option might be uname -s checking.
2014-09-26T14:20:37Z loke_: drmeister: here:
2014-09-26T14:20:39Z dlowe: uname | grep Linux
2014-09-26T14:20:51Z loke_: https://github.com/lokedhs/gnu-apl-mode/blob/master/native/Makefile#L31
2014-09-26T14:20:59Z loke_: I do exactly that in that Makefile
2014-09-26T14:22:00Z drmeister_: Thanks!  I'll have to check that link when I get into my office. In on a train working through my iPhone hotspot and web pages are coming up weird.
2014-09-26T14:25:34Z chr` quit (Quit: ERC Version 5.3 (IRC client for Emacs))
2014-09-26T14:25:59Z loke_: How annoying. I started looking into rabbitmq in the hope I could use it, and then it turns out there is no functional AMQP implementation in CL
2014-09-26T14:26:06Z lambda joined #lisp
2014-09-26T14:27:08Z drmeister_: What I got from all the comments on HN/Reddit/Wordpress is people want (1) binaries (2) No LGPL.   I'm working on (1) but (2) will take time to rewrite around the LGPL encumbered code.
2014-09-26T14:27:33Z drmeister_: But I'm totally up for (2)
2014-09-26T14:27:39Z drmeister_: Oh and (3) faster.
2014-09-26T14:28:28Z Xach: loke_: there's the ones from that german fella who wrote his own universe of lisp libraries
2014-09-26T14:28:37Z eudoxia_: dwim?
2014-09-26T14:28:38Z Xach: i want to say setf.de/de.setf?
2014-09-26T14:28:38Z drmeister_: Mustn't forget faster.   I haven't seen beach for the past day - I hope he likes what I said about the new compiler.  I'm being really vague because I don't want to put him on a spot.
2014-09-26T14:28:40Z loke_: Xach: Yeah. I looked at it.
2014-09-26T14:29:01Z Xach: I wanted to add that code to quicklisp, but it uses forks of existing projects without renaming them.
2014-09-26T14:29:13Z Xach: (I think - i might be confusing it with someone else's self-contained lisp libraryverse)
2014-09-26T14:29:14Z brucem: drmeister_: there was an asshole on lobste.rs, but apparently the response to him drove him to deactivate his account and delete his comments.
2014-09-26T14:29:40Z loke_: Xach: Apart from the fact that it's more or less impossible to even load his code (go ahead, try it :-) there is no end of ASDF failures), the README explains itself how slow it is and uses huge amounts of memory
2014-09-26T14:29:52Z drmeister_: brucem: I saw that and I really appreciate your rebuttal.
2014-09-26T14:29:54Z innertracks joined #lisp
2014-09-26T14:30:17Z brucem: drmeister_: You should do what I do on that site, btw.
2014-09-26T14:30:21Z hiyosi_ joined #lisp
2014-09-26T14:30:22Z Xach: loke_: pity
2014-09-26T14:30:39Z loke_: Xach: That said, the number of weird messages I got from ASDF when trying to load the thing was quite striking. Either the code is broken, or _I_ am broken. Do you mind trying and tell me if it's me that's doing something wrong?
2014-09-26T14:30:43Z loke_: https://github.com/lisp/de.setf.amqp
2014-09-26T14:30:59Z kobain joined #lisp
2014-09-26T14:30:59Z Xach: loke_: i don't have the time at the moment, sorry. can you paste the output?
2014-09-26T14:31:05Z Xach: if not, i might visit it later
2014-09-26T14:31:12Z drmeister_: brucem: What do you do on that site?
2014-09-26T14:31:35Z loke_: There's a lot of different issues. I couldn't even get ASDF to find the asd files at first (the system names and asd file names don
2014-09-26T14:31:39Z loke_: There's a lot of different issues. I couldn't even get ASDF to find the asd files at first (the system names and asd file names don't match)
2014-09-26T14:31:48Z brucem: drmeister_: I'm known there for my really long posts in the "What are you doing this week?" thread ... and I found out this week that people really enjoy them.
2014-09-26T14:32:01Z loke_: then I get an error saying that ":INDEX is not a recognised asdf command" or something to that effect
2014-09-26T14:32:43Z drmeister_: brucem: Got it - yes, it's time to start posting more.
2014-09-26T14:32:49Z Xach: loke_: sounds like a very insular environment
2014-09-26T14:33:13Z janmuffino quit (Ping timeout: 246 seconds)
2014-09-26T14:33:44Z loke_: Um. yes
2014-09-26T14:33:47Z jkaye joined #lisp
2014-09-26T14:33:52Z loke_: check this out, I have no idea what he's doing...
2014-09-26T14:33:53Z loke_: https://github.com/lisp/de.setf.amqp/blob/master/amqp.asd#L36
2014-09-26T14:34:00Z Xach: that turkish project, core-server, is similar
2014-09-26T14:34:46Z ivan4th joined #lisp
2014-09-26T14:36:03Z eudoxia_: >30 lines explaining how to load the project
2014-09-26T14:36:07Z eudoxia_: into the trash it goes
2014-09-26T14:36:23Z loke_: eudoxia_: which one?
2014-09-26T14:36:36Z eudoxia_: loke_: de.setf.amqp
2014-09-26T14:36:40Z loke_: yeah
2014-09-26T14:36:46Z Zhivago quit (Changing host)
2014-09-26T14:36:46Z Zhivago joined #lisp
2014-09-26T14:37:22Z Shinmera: But everybody loves build instructions!
2014-09-26T14:37:28Z Shinmera: ...right?
2014-09-26T14:37:35Z H4ns: loke_: sorry for the bad recommendation
2014-09-26T14:38:05Z loke_: No owrries. I learned a lot :-)
2014-09-26T14:38:16Z brucem: loke_: You need an MQ server client?
2014-09-26T14:38:32Z loke_: If I want to use this thing, I'm likely much better off interfacing using CFFI
2014-09-26T14:39:03Z loke_: I took a look at the protocol spec just for the hell of it, and it's massive enough to assume that no one will build a native CL implementation with any form of stability.
2014-09-26T14:39:08Z brucem: loke_: IIRC, RabbitMQ can talk STOMP ... which is a very nice and simple text-based protocol.
2014-09-26T14:39:13Z drmeister_: faheem__1: Pause for a little while - there was a problem with the $(BJAM) variable in the externals-clasp makefile - fixing and testing.   I'll tell you when I've gotten it to build on linux.
2014-09-26T14:39:18Z loke_: brucem: Oh, really?
2014-09-26T14:39:21Z brucem: loke_: and with STOMP, you can talk to RabbitMQ and ActiveMQ.
2014-09-26T14:39:24Z loke_: Sounds interesting indeed
2014-09-26T14:39:58Z gravicappa quit (Remote host closed the connection)
2014-09-26T14:40:04Z brucem: loke_: Yep ... I built a whole distributed system for game servers on top of STOMP (and Python) years ago ... worked fine. :) The STOMP protocol will take you a couple of hours probably at most, IIRC.
2014-09-26T14:40:12Z loke_: Oh, and there is a cl-stomp in QL!
2014-09-26T14:40:29Z Shinmera: loke_: Usually I'd be up for challenges like that, but I'm not in the mood and short on time these days.
2014-09-26T14:40:29Z brucem: even better.
2014-09-26T14:40:34Z loke_: I guess I can reinstall rabbit on my system then :-)
2014-09-26T14:40:43Z cocosp joined #lisp
2014-09-26T14:41:24Z brucem: loke_: http://common-lisp.net/gitweb/?p=projects/cl-stomp/cl-stomp.git isn't very promising though.
2014-09-26T14:42:09Z Nizumzen joined #lisp
2014-09-26T14:42:48Z drmeister_: BBL
2014-09-26T14:43:57Z drmeister_ quit (Read error: Connection reset by peer)
2014-09-26T14:44:29Z eudoxia_: loke_: https://github.com/fukamachi/lesque might be of interest
2014-09-26T14:44:30Z kushal quit (Ping timeout: 258 seconds)
2014-09-26T14:45:05Z loke_: eudoxia_: Thanks, but that's not the kinds of queues I'm talking about. I was talking about message queues.
2014-09-26T14:45:58Z jlongster joined #lisp
2014-09-26T14:45:59Z eudoxia_: loke_: ok. i've only used Celery and the whole AMQP thing has never been very clear to me
2014-09-26T14:46:15Z ferada: loke_: is beanstalk an option for you?
2014-09-26T14:47:39Z faheem__1: drmeister: ok
2014-09-26T14:47:53Z ndrei quit (Ping timeout: 260 seconds)
2014-09-26T14:48:01Z loke_: eudoxia_: this is different. Message queues is basically exactly what the name implies; a way to send messages between separate systems
2014-09-26T14:49:05Z ltbarcly quit (Quit: Computer has gone to sleep.)
2014-09-26T14:49:55Z ndrei joined #lisp
2014-09-26T14:50:21Z faheem__1: drmeister: why are you using bjam again? around 2004/2005 i tried to use it. I learned to really hate that program. worse than make - which I'm not fond of either.
2014-09-26T14:51:08Z faheem__1: I also remember when I had a question about something, Dave Abrahams helpfully telling me that bjam was not exactly well documented. or osmething.
2014-09-26T14:51:19Z faheem__1: *something*
2014-09-26T14:53:46Z svetlyak40wt joined #lisp
2014-09-26T14:55:02Z faheem__1: drmeister: checking for linux - `lsb_release -a` perhaps.
2014-09-26T14:55:48Z ndrei quit (Ping timeout: 260 seconds)
2014-09-26T14:55:58Z dlowe: so I got a problem with a deftype in sbcl
2014-09-26T14:56:03Z dlowe: given this: (deftype predicate-name () '(and symbol (not (member character-ranges string and or not * + ? & ! ~ function))))
2014-09-26T14:56:20Z dlowe: why does this happen: (typep '? 'esrap::predicate-name) => T
2014-09-26T14:56:43Z nyef: dlowe: Different package on the symbols named ? ?
2014-09-26T14:56:44Z dlowe: yet: (typep '+ 'esrap::predicate-name) => NIL
2014-09-26T14:56:50Z ndrei joined #lisp
2014-09-26T14:56:57Z moore33: Xach: "german fella" = James Anderson?
2014-09-26T14:57:02Z nyef: Because CL:+ is likely to be visible in both packages?
2014-09-26T14:57:09Z dlowe: ahh, that's it
2014-09-26T14:57:12Z dlowe: thanks
2014-09-26T14:57:23Z nyef: No problem.
2014-09-26T14:57:47Z drmeister quit (Remote host closed the connection)
2014-09-26T14:57:58Z varjag_ joined #lisp
2014-09-26T14:58:45Z jlongster quit (Remote host closed the connection)
2014-09-26T14:59:06Z drmeister joined #lisp
2014-09-26T14:59:13Z drmeiste_ joined #lisp
2014-09-26T14:59:14Z jlongster joined #lisp
2014-09-26T14:59:27Z normanrichards joined #lisp
2014-09-26T14:59:28Z normanrichards quit (Read error: Connection reset by peer)
2014-09-26T15:00:26Z Xach: moore33: Yes. Forgot the name, remembered only the "de"
2014-09-26T15:01:28Z pt1 quit (Remote host closed the connection)
2014-09-26T15:01:33Z gingerale joined #lisp
2014-09-26T15:03:55Z jusss quit (Remote host closed the connection)
2014-09-26T15:05:24Z Kenjin joined #lisp
2014-09-26T15:05:45Z Cymew quit (Ping timeout: 272 seconds)
2014-09-26T15:05:47Z Karl_Dscc joined #lisp
2014-09-26T15:06:03Z janmuffino joined #lisp
2014-09-26T15:07:35Z stacksmith joined #lisp
2014-09-26T15:07:44Z boogie quit (Remote host closed the connection)
2014-09-26T15:08:12Z boogie joined #lisp
2014-09-26T15:08:39Z H4ns: there is also this http://wcp.sdf-eu.org/software/ german fella with an own lisp universe
2014-09-26T15:08:56Z Xach: at least his universe doesn't conflict with the rest of the universe :)
2014-09-26T15:10:12Z specbot quit (Remote host closed the connection)
2014-09-26T15:10:15Z eudoxia_: don't forget about https://news.ycombinator.com/item?id=8284185
2014-09-26T15:10:18Z specbot joined #lisp
2014-09-26T15:10:28Z cocosp quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
2014-09-26T15:11:12Z finitoo joined #lisp
2014-09-26T15:11:30Z stacksmith: Good morning.  My giant html logfile of this channel is now on github: https://github.com/stacksmith/irc-lisp-html
2014-09-26T15:11:55Z TDog joined #lisp
2014-09-26T15:12:03Z drmeister: faheem__1:  Pull the latest externals-clasp,  "make clean"  and then follow the directions in the README.md file    It builds on linux and OS X
2014-09-26T15:12:11Z faheem__1: stacksmith: are you planning to keep it updated?
2014-09-26T15:12:28Z faheem__1: drmeister: did you see my earlier comments in the channel? last hr or so?
2014-09-26T15:12:36Z faheem__1: you had a client running, but i couldn't tell...
2014-09-26T15:13:10Z faheem__1: drmeister: you seem to have gone back to deleting llvm on `make clean`. not a fan of that
2014-09-26T15:13:38Z stacksmith: faheem__1: I'll write a daily script, I think.  But it seems that the source file at ccl.clozure.com is a day out of date...
2014-09-26T15:13:46Z faheem__1: drmeister: doesn't seem to have changed since my last pull. it failed
2014-09-26T15:13:51Z drmeister: I changed the makefile so that it doesn't delete llvm every time you build.
2014-09-26T15:13:53Z Karl_Dscc quit (Remote host closed the connection)
2014-09-26T15:14:04Z faheem__1: drmeister: this is commit cee291ccdcea092b732756dac0bd10bb9c901143
2014-09-26T15:14:24Z faheem__1: is that current?
2014-09-26T15:14:38Z ivan4th: BTW besides STOMP there's such thing as MQTT. Pity that it doesn't have a Common Lisp impl, but I'm thinking about making one. MQTT is a very simple messaging protocol, too, and it's supported by RabbitMQ
2014-09-26T15:15:20Z drmeister: faheem__1: Yes, that is the current commit on both linux and OS X.
2014-09-26T15:15:20Z ivan4th: and de.setf.amqp is licensed under AGPL which severely limits its possible usages
2014-09-26T15:16:12Z faheem__1: drmeister: you have TOOLSET in the local.config, but not documented in README.
2014-09-26T15:16:14Z drmeister: faheem__1: I went into the private chat session we had last night.
2014-09-26T15:17:03Z faheem__1: drmeister: ok. is that still open?
2014-09-26T15:17:10Z faheem__1: you want to go there now?
2014-09-26T15:17:23Z drmeister: faheem__1: Yes.
2014-09-26T15:17:35Z faheem__1: drmeister: ok
2014-09-26T15:22:03Z drdanmaku joined #lisp
2014-09-26T15:27:15Z stassats joined #lisp
2014-09-26T15:28:22Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T15:30:35Z oudeis quit (Quit: This computer has gone to sleep)
2014-09-26T15:31:43Z lobbes joined #lisp
2014-09-26T15:36:48Z jtza8 quit (Ping timeout: 250 seconds)
2014-09-26T15:39:56Z marsbot is now known as marsam
2014-09-26T15:44:08Z Longlius quit (Remote host closed the connection)
2014-09-26T15:46:35Z lowfyr quit (Remote host closed the connection)
2014-09-26T15:48:11Z przl quit (Ping timeout: 272 seconds)
2014-09-26T15:49:23Z k-stz quit (Remote host closed the connection)
2014-09-26T15:52:55Z loke_ quit (Remote host closed the connection)
2014-09-26T15:53:27Z TDog quit (Read error: Connection reset by peer)
2014-09-26T15:53:55Z TDog joined #lisp
2014-09-26T15:55:18Z vinleod joined #lisp
2014-09-26T15:55:24Z fortitude: is it normal to have all kinds of package-variance issues under sbcl when you use cl-annot's @export stuff, or am I missing something?
2014-09-26T15:55:39Z loke_ joined #lisp
2014-09-26T15:56:34Z ndrei quit (Ping timeout: 258 seconds)
2014-09-26T15:58:02Z Jesin joined #lisp
2014-09-26T15:58:08Z ndrei joined #lisp
2014-09-26T16:01:10Z KCL_ joined #lisp
2014-09-26T16:02:42Z stassats: that's your punishment for using annotations
2014-09-26T16:02:46Z faheem__1: i don't see an obvious way to do the product of numbers using loop.
2014-09-26T16:03:09Z faheem__1: can i parametrize on the operation. there is a sum, and summing.
2014-09-26T16:03:37Z ahungry quit (Ping timeout: 245 seconds)
2014-09-26T16:03:47Z stassats: you can't
2014-09-26T16:04:02Z KCL quit (Ping timeout: 245 seconds)
2014-09-26T16:04:04Z faheem__1: stassats: bummer
2014-09-26T16:04:16Z moore33: for prod = x then (* prod x)
2014-09-26T16:04:49Z faheem__1: moore33: what would hte complete thing look like?
2014-09-26T16:05:07Z stassats: (loop for prod = x then (* prod x)), probably
2014-09-26T16:05:12Z faheem__1: stassats: ok
2014-09-26T16:05:17Z posterdati300 quit (Ping timeout: 272 seconds)
2014-09-26T16:05:27Z faheem__1: um, numbers missing?
2014-09-26T16:05:33Z Grue`: (loop for x in '(stuff) for prod = x then (* prod x))
2014-09-26T16:05:39Z ahungry joined #lisp
2014-09-26T16:05:46Z Grue`: however this wont work if stuff is empty list
2014-09-26T16:05:47Z faheem__1: Grue`: ok
2014-09-26T16:06:05Z gravicappa joined #lisp
2014-09-26T16:06:09Z mishoo quit (Ping timeout: 258 seconds)
2014-09-26T16:06:09Z Grue`: oh and "finally (return prod)"
2014-09-26T16:06:10Z ggole: (reduce #'* numbers)? Or (apply #'* numbers) if you know it is short enough.
2014-09-26T16:06:34Z faheem__1: Grue`: ah
2014-09-26T16:06:35Z davazp quit (Remote host closed the connection)
2014-09-26T16:06:44Z ahungry quit (Remote host closed the connection)
2014-09-26T16:07:15Z Grue`: ok scratch this, you need to do (loop with prod = 1 for x in '(stuff) do (setf prod (* prod x)) finally (return prod))
2014-09-26T16:07:22Z stassats: (loop for prod = 1 then (* prod x) for x ... fina.. what Grue` said
2014-09-26T16:07:37Z oudeis joined #lisp
2014-09-26T16:07:43Z fortitude: stassats: I reported a method dispatch bug in ECL the other day, and I'd like to help track it down, but I'm not familiar enough with ECL's internal to know where to start looking
2014-09-26T16:07:47Z fortitude: stassats: any suggestions?
2014-09-26T16:08:11Z boogie quit (Remote host closed the connection)
2014-09-26T16:08:28Z stassats: though, Grue` is using setf, but you don't have to
2014-09-26T16:08:33Z posterdati300 joined #lisp
2014-09-26T16:08:38Z ahungry joined #lisp
2014-09-26T16:08:45Z holycow_: hi
2014-09-26T16:08:46Z stassats: fortitude: refresh my memory with a link
2014-09-26T16:09:08Z Grue`: do (setf) makes it as versatile as summing is though
2014-09-26T16:09:13Z fortitude: stassats: http://sourceforge.net/p/ecls/bugs/295/
2014-09-26T16:09:21Z Grue`: because you can place it inside ifs
2014-09-26T16:09:45Z holycow_: i'm reading the hunchentoot *request* variable documentation but i'm not grokking how to grab data from a form via post.  i would like to grab the data from a single input field with a fucntion and princ the result.  what might a function look like that does this?
2014-09-26T16:10:12Z stassats: holycow_: post-parameter
2014-09-26T16:10:22Z holycow_: oh
2014-09-26T16:11:19Z Grue`: (when (post-parameters*) (let ((value (post-parameter "value"))) do-stuff))
2014-09-26T16:11:52Z stassats: no need to do when (post-parameters*)
2014-09-26T16:12:26Z Grue`: yeah probably not, i just copied from my hunchentoot project
2014-09-26T16:12:46Z holycow_: thx, that eally helps
2014-09-26T16:12:50Z Grue`: but this allows to do different stuff depending on whether it was a post request
2014-09-26T16:13:47Z Xach: Grue`: i usually check the request method instead.
2014-09-26T16:14:08Z Xach: cool kids wrap it up in declarative routing
2014-09-26T16:14:17Z ChanServ has set mode +o Xach
2014-09-26T16:14:20Z Xach has set mode +b *!*man50@90.174.4.*
2014-09-26T16:14:21Z Grue`: yeah it could be empty post request or something
2014-09-26T16:14:24Z finitoo [~xach@pdpc/supporter/professional/xach] has been kicked from #lisp by Xach (spam privmsg)
2014-09-26T16:14:25Z Xach has set mode -o Xach
2014-09-26T16:15:05Z stassats: holycow_: see http://weitz.de/hunchentoot/#start
2014-09-26T16:15:33Z stassats: in the define-easy-handler form, NAME will be bound either to a GET or a POST parameter
2014-09-26T16:17:50Z Nizumzen quit (Ping timeout: 244 seconds)
2014-09-26T16:18:21Z boogie joined #lisp
2014-09-26T16:20:07Z boogie quit (Remote host closed the connection)
2014-09-26T16:21:29Z hiroakip joined #lisp
2014-09-26T16:22:06Z gjvc joined #lisp
2014-09-26T16:23:10Z hlavaty quit (Remote host closed the connection)
2014-09-26T16:23:15Z Longlius joined #lisp
2014-09-26T16:24:52Z uzo quit (Ping timeout: 246 seconds)
2014-09-26T16:25:58Z manfoo7` joined #lisp
2014-09-26T16:26:24Z Nizumzen joined #lisp
2014-09-26T16:26:32Z hiroakip quit (Ping timeout: 245 seconds)
2014-09-26T16:27:37Z uber is now known as Guest71879
2014-09-26T16:27:38Z Guest71879 quit (Killed (morgan.freenode.net (Nickname regained by services)))
2014-09-26T16:27:54Z uber_ joined #lisp
2014-09-26T16:27:55Z uber_ quit (Max SendQ exceeded)
2014-09-26T16:28:46Z ndrei quit (Ping timeout: 258 seconds)
2014-09-26T16:28:48Z kami joined #lisp
2014-09-26T16:28:53Z uber joined #lisp
2014-09-26T16:28:54Z uber quit (Max SendQ exceeded)
2014-09-26T16:29:23Z uber joined #lisp
2014-09-26T16:29:23Z uber quit (Max SendQ exceeded)
2014-09-26T16:29:53Z uber joined #lisp
2014-09-26T16:29:53Z uber quit (Max SendQ exceeded)
2014-09-26T16:29:54Z innertracks quit (Quit: innertracks)
2014-09-26T16:30:04Z bbyler_tho joined #lisp
2014-09-26T16:30:23Z uber joined #lisp
2014-09-26T16:30:44Z ndrei joined #lisp
2014-09-26T16:31:32Z zeebrah quit (Ping timeout: 245 seconds)
2014-09-26T16:32:44Z ilhami quit (Quit: Nettalk6 - www.ntalk.de)
2014-09-26T16:33:09Z holycow_: back
2014-09-26T16:33:17Z holycow_: thank you kindly for the tips.  i greatly appreciate it
2014-09-26T16:33:43Z malice: Hey, is it safe to use diacritics in symbol name? e.g. *symbol-wit-żę-diacritic*
2014-09-26T16:34:03Z Grue`: if your lisp supports unicode then yes
2014-09-26T16:34:14Z oGMo: doesn't that depend on implementation and read encoding?
2014-09-26T16:34:34Z malice: So it's not safe then?
2014-09-26T16:35:08Z oGMo: eh.. do you plan on running code on something incomplete and outdated enough to not support unicode?
2014-09-26T16:35:09Z Grue`: it should be safe if you use unicode everywhere
2014-09-26T16:35:13Z madmalik joined #lisp
2014-09-26T16:35:32Z Grue`: a symbol's name is just a string
2014-09-26T16:35:39Z moore33 quit (Quit: http://www.kiwiirc.com/ - A hand crafted IRC client)
2014-09-26T16:35:45Z vinleod quit (Quit: ["Textual IRC Client: www.textualapp.com"])
2014-09-26T16:35:46Z malice: Hmm... Then I guess it's okay
2014-09-26T16:35:51Z Grue`: so if you're using strings with characters like this, then symbols should be ok too
2014-09-26T16:36:08Z oGMo: i would worry more about users having a hard time typing it than implementation support
2014-09-26T16:36:27Z oGMo: (and that's probably not something to worry that much about)
2014-09-26T16:36:47Z malice: Okay than. Thanks for answer!
2014-09-26T16:40:53Z Krystof quit (Ping timeout: 260 seconds)
2014-09-26T16:41:25Z ndrei quit (Ping timeout: 258 seconds)
2014-09-26T16:42:12Z Kenjin quit (Remote host closed the connection)
2014-09-26T16:42:44Z lambda quit (Quit: ERC Version 5.3 (IRC client for Emacs))
2014-09-26T16:43:03Z ehu quit (Ping timeout: 246 seconds)
2014-09-26T16:43:14Z ndrei joined #lisp
2014-09-26T16:44:26Z hitecnologys quit (Quit: hitecnologys)
2014-09-26T16:46:11Z s3gfault joined #lisp
2014-09-26T16:46:15Z Fare joined #lisp
2014-09-26T16:50:30Z ndrei quit (Read error: Connection reset by peer)
2014-09-26T16:51:12Z ndrei joined #lisp
2014-09-26T16:51:57Z sz0 quit
2014-09-26T16:52:32Z tadni quit (Ping timeout: 258 seconds)
2014-09-26T16:53:13Z sz0 joined #lisp
2014-09-26T16:53:24Z boogie joined #lisp
2014-09-26T16:53:41Z kuanyui quit (Ping timeout: 258 seconds)
2014-09-26T16:54:21Z fragamus joined #lisp
2014-09-26T16:55:38Z marsam is now known as marsbot
2014-09-26T17:03:29Z forwchen joined #lisp
2014-09-26T17:05:18Z fragamus quit (Quit: Computer has gone to sleep.)
2014-09-26T17:05:41Z fragamus joined #lisp
2014-09-26T17:08:12Z ndrei quit (Ping timeout: 245 seconds)
2014-09-26T17:10:03Z bend3r quit (Quit: leaving)
2014-09-26T17:12:34Z jeremyheiler joined #lisp
2014-09-26T17:13:02Z forwchen quit (Quit: Konversation terminated!)
2014-09-26T17:13:05Z jasom: drmeister: I still can't build externals-clasp: http://paste.lisp.org/display/143851
2014-09-26T17:13:23Z pinupgeek quit (Quit: Connection closed for inactivity)
2014-09-26T17:17:29Z edgar-rft joined #lisp
2014-09-26T17:18:44Z ndrei joined #lisp
2014-09-26T17:20:38Z faheem__1: jasom: building for me now. for the moment. OS?
2014-09-26T17:21:06Z faheem__1: it probably needs at least gcc 4.8.
2014-09-26T17:21:19Z faheem__1: gcc 4.7 failed for me earlier. 4.9 works
2014-09-26T17:24:21Z ilhami joined #lisp
2014-09-26T17:24:43Z kushal joined #lisp
2014-09-26T17:29:47Z loke_ quit (Remote host closed the connection)
2014-09-26T17:30:17Z eudoxia_ quit (Ping timeout: 245 seconds)
2014-09-26T17:31:12Z fragamus quit (Ping timeout: 244 seconds)
2014-09-26T17:33:22Z ndrei quit (Ping timeout: 240 seconds)
2014-09-26T17:33:35Z forwchen joined #lisp
2014-09-26T17:35:21Z ndrei joined #lisp
2014-09-26T17:35:51Z ehu joined #lisp
2014-09-26T17:35:57Z josteink quit (Ping timeout: 260 seconds)
2014-09-26T17:36:37Z eudoxia joined #lisp
2014-09-26T17:36:46Z svetlyak40wt joined #lisp
2014-09-26T17:38:51Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T17:38:57Z svetlyak40wt joined #lisp
2014-09-26T17:40:58Z ndrei_ joined #lisp
2014-09-26T17:42:33Z josteink_ joined #lisp
2014-09-26T17:42:33Z josteink_ quit (Client Quit)
2014-09-26T17:43:22Z ehu quit
2014-09-26T17:43:27Z josteink joined #lisp
2014-09-26T17:44:52Z ndrei quit (Ping timeout: 245 seconds)
2014-09-26T17:45:27Z ehu joined #lisp
2014-09-26T17:46:08Z ndrei_ quit (Ping timeout: 260 seconds)
2014-09-26T17:48:09Z przl joined #lisp
2014-09-26T17:49:07Z drmeister: jasom: You are trying as well?   I'm working on it.  Can you give me a day or so?
2014-09-26T17:49:59Z faheem__1: drmeister: your `build clean` on clasp is overly enthusiastic
2014-09-26T17:50:14Z faheem__1: it deleted all the files in the clasp source.
2014-09-26T17:50:27Z faheem__1: though it was not obvius why
2014-09-26T17:50:46Z faheem__1: building clasp now
2014-09-26T17:51:01Z faheem__1: the warning: unknown warning option '-Wno-unused-local-typedef' [-Wunknown-warning-option]
2014-09-26T17:51:04Z faheem__1: persists
2014-09-26T17:51:27Z ltbarcly joined #lisp
2014-09-26T17:51:45Z drmeister: faheem__1: Every file in the clasp source?  That's exactly why I hate putting "rm -rf ANYTHING" into a makefile.  It's highly toxic.
2014-09-26T17:52:04Z faheem__1: drmeister: well, not the .git, but all non-hidden files
2014-09-26T17:52:33Z drmeister starts hyperventilating into a bag
2014-09-26T17:52:45Z drmeister slaps himself a few times
2014-09-26T17:52:49Z drmeister: What were you doing?
2014-09-26T17:52:56Z drmeiste_ quit (Remote host closed the connection)
2014-09-26T17:53:11Z faheem__1: drmeister: make clean
2014-09-26T17:53:18Z drmeister: Ah - don't do that.
2014-09-26T17:53:22Z drmeister: I'm kidding.
2014-09-26T17:53:26Z faheem__1: sorry, i wrote 'build clean' above
2014-09-26T17:53:36Z schaueho joined #lisp
2014-09-26T17:53:43Z faheem__1: sometimes i'm dyslexic. or something
2014-09-26T17:53:44Z drmeiste_ joined #lisp
2014-09-26T17:54:10Z manfoo7` quit (Remote host closed the connection)
2014-09-26T17:54:19Z faheem__1: would be nice to get rid of the '-Wno-unused-local-typedef'
2014-09-26T17:54:21Z drmeister: So my friend in #llvm reports that he got clasp_boehm_o to compile and has it running on OS X
2014-09-26T17:54:22Z faheem__1: why is that happening?
2014-09-26T17:54:38Z faheem__1: drmeister: that's the boehm version of clasp?
2014-09-26T17:54:43Z drmeister: Yes.  First errors, then warnings.
2014-09-26T17:54:43Z jegaxd26 joined #lisp
2014-09-26T17:54:54Z drmeister: It's clasp using the boehm garbage collector
2014-09-26T17:54:56Z dlowe: drmeister: find . -name '*.o' -o -name '*.fasl' -delete
2014-09-26T17:55:06Z faheem__1: drmeister: right
2014-09-26T17:55:19Z hiroakip joined #lisp
2014-09-26T17:55:41Z drmeister: dlowe: Good idea
2014-09-26T17:55:45Z drmeister: I'll do that.
2014-09-26T17:56:05Z dlowe: drmeister: don't use ., though, there's some make variable that will give you the directory root
2014-09-26T17:56:23Z drmeister: Ok, I have to wolf down my lunch and run to a faculty meeting - I'll be back in an hour.
2014-09-26T17:56:58Z drmeister: It's one of the few things where I can honestly say "I'd rather be debugging my build system".
2014-09-26T17:57:16Z faheem__1: hmm, it exited with an error
2014-09-26T17:57:29Z faheem__1: drmeister: not a fan of faculty meetings?
2014-09-26T17:57:38Z ivan4th quit (Ping timeout: 272 seconds)
2014-09-26T17:57:47Z wglb: drmeister: Because of the wolf lunch, or because of the meeting?
2014-09-26T17:58:09Z drmeister: faheem__1: Yeah, I've been really focused on externals-clasp - I haven't paid enough attention to getting the clasp build in line with the externals-clasp changes.
2014-09-26T17:58:28Z faheem__1: drmeister: ok, well, it ran for awhile
2014-09-26T17:58:45Z drmeister: faheem__1: I'll put my full attention on it when I get back.
2014-09-26T17:58:55Z faheem__1: hmm - ../../src/clbind/derivable_class.cc:183:18: warning: 'auto_ptr' is deprecated [-Wdeprecated-declarations]
2014-09-26T17:59:13Z faheem__1: isn't auto ptr that really old thing?
2014-09-26T17:59:13Z rme left #lisp
2014-09-26T17:59:23Z faheem__1: drmeister: ok
2014-09-26T17:59:41Z drmeiste_ quit (Remote host closed the connection)
2014-09-26T17:59:58Z drmeiste_ joined #lisp
2014-09-26T18:01:18Z forwchen quit (Read error: Connection reset by peer)
2014-09-26T18:01:43Z forwchen joined #lisp
2014-09-26T18:01:52Z lifenoodles quit (Ping timeout: 245 seconds)
2014-09-26T18:02:00Z mhd joined #lisp
2014-09-26T18:03:39Z sdemarre joined #lisp
2014-09-26T18:06:13Z bend3r joined #lisp
2014-09-26T18:06:32Z bend3r quit (Client Quit)
2014-09-26T18:06:47Z lifenoodles joined #lisp
2014-09-26T18:09:56Z forwchen quit (Quit: Konversation terminated!)
2014-09-26T18:12:10Z bend3r joined #lisp
2014-09-26T18:12:15Z ltbarcly quit (Quit: Computer has gone to sleep.)
2014-09-26T18:13:51Z hiroakip quit (Ping timeout: 272 seconds)
2014-09-26T18:15:49Z nydel quit (Quit: WeeChat 0.4.2)
2014-09-26T18:16:30Z tomvos joined #lisp
2014-09-26T18:17:11Z nydel joined #lisp
2014-09-26T18:19:37Z attila_lendvai joined #lisp
2014-09-26T18:19:37Z attila_lendvai quit (Changing host)
2014-09-26T18:19:37Z attila_lendvai joined #lisp
2014-09-26T18:20:12Z lambda joined #lisp
2014-09-26T18:21:31Z przl quit (Quit: leaving)
2014-09-26T18:21:42Z stassats: fortitude: looks like the GF cache is broken
2014-09-26T18:24:10Z Jesin quit (Quit: Leaving)
2014-09-26T18:24:41Z fortitude: stassats: good to know
2014-09-26T18:24:53Z fortitude: as soon as I can get virtualbox to un-hose my keyboard, I'll poke at it a bit
2014-09-26T18:25:48Z TDog quit (Ping timeout: 260 seconds)
2014-09-26T18:26:02Z SvenGek joined #lisp
2014-09-26T18:26:17Z drmeister: I'm back - it was a short meeting.
2014-09-26T18:26:46Z Jesin joined #lisp
2014-09-26T18:26:50Z stassats: and i have no idea how the cache works
2014-09-26T18:27:02Z stassats: probably that's why it's broken --- too complicated
2014-09-26T18:27:10Z drmeister: Yes, my friend on #llvm got the Boehm version of Clasp to build completely and he has a full repl.
2014-09-26T18:27:16Z drmeister: That is on OS X
2014-09-26T18:27:24Z drmeister: faheem__1: ping
2014-09-26T18:27:59Z fortitude: stassats: I notice that compute-applicable-methods seems to be going through the cache, and that returns the right results
2014-09-26T18:28:11Z innertracks joined #lisp
2014-09-26T18:28:23Z fortitude: presumably that and the actual dispatch just aren't doing quite the same things
2014-09-26T18:28:37Z stassats: you need to be looking at C
2014-09-26T18:28:46Z stassats: it's _ecl_stnadard_dispatch
2014-09-26T18:29:02Z stassats: and ecl_search_cache specifically
2014-09-26T18:29:12Z stassats: when i disable the cache lookup, everything works fine
2014-09-26T18:29:40Z fortitude: that's the bit that I was looking at (I just used the lispy name)
2014-09-26T18:30:39Z dan64 quit (Quit: ZNC - http://znc.in)
2014-09-26T18:30:51Z stassats: i'm not clear on the whole fill_spec_vector, where the GFUN parameter is FRAME, INSTANCE is GF
2014-09-26T18:31:21Z _d3f quit (Quit: WeeChat 0.4.3)
2014-09-26T18:31:25Z dan64 joined #lisp
2014-09-26T18:32:03Z stassats: and of what INSTANCE is supposed to be an instance? of GF, then what is GFUN?
2014-09-26T18:32:58Z fortitude: it looks like the fill_spec_vector is a way to save cache keys for the later recompute business
2014-09-26T18:33:11Z fortitude: though I don't quite know what's going on there
2014-09-26T18:33:31Z stassats: right, but which keys are which
2014-09-26T18:33:51Z stassats: i imagine it needs to go through the frame to get the arguments, but i don't see where that happens
2014-09-26T18:34:25Z stassats: and naturally EQL and class dispatches should be handled differently, i think that's where it breaks, but i don't see where
2014-09-26T18:34:56Z _d3f joined #lisp
2014-09-26T18:35:12Z fortitude: the frame is being passed in from generic_function_dispatch_vararg
2014-09-26T18:35:21Z stassats: or is it just caches GF and the pointer to FRAME?
2014-09-26T18:35:30Z stassats: how is that supposed to work, exactly?
2014-09-26T18:35:44Z tomvos left #lisp
2014-09-26T18:36:23Z _d3f quit (Client Quit)
2014-09-26T18:37:15Z fortitude: hmm
2014-09-26T18:37:23Z fortitude: the cache seems to have a reference to an environment in it
2014-09-26T18:37:40Z fortitude: and since nothing else is being passed into ecl_search_cache, I assume it's pulling arguments directly from the env?
2014-09-26T18:37:40Z stassats: it does work sometimes, that baffles me
2014-09-26T18:37:51Z lifenoodles quit (Ping timeout: 244 seconds)
2014-09-26T18:38:20Z stassats: does it break if it's not a tail-call?
2014-09-26T18:39:47Z arenz quit (Ping timeout: 245 seconds)
2014-09-26T18:39:50Z stassats: it does, so, the theory that the reused stack frame confuses it is wrong
2014-09-26T18:42:03Z pjb: drmeister: have a look at: http://git.informatimago.com/viewgit/index.php?a=viewblob&p=public/bin&h=a4a3b2c7092dcab3ae33aa7bbe30b34942a0f3b6&hb=cb790f311ce6e61de0c39e2a34dea3b537fcb06e&f=distribution
2014-09-26T18:43:01Z drmeister: pjb: What is it fore?
2014-09-26T18:43:02Z jasom: drmeister: no I refuse to give you a day, I am entitled to a 100% perfect working version now.  </sarcasm>
2014-09-26T18:43:03Z drmeister: for
2014-09-26T18:43:17Z lifenoodles joined #lisp
2014-09-26T18:43:27Z drmeister: jasom: Satisfaction or yer money back.
2014-09-26T18:43:37Z jasom: drmeister: Just wait, you'll get that response without the sarcasm if you get more than about 6 people using clasp.
2014-09-26T18:43:47Z jasom: It's the OSS way
2014-09-26T18:44:05Z drmeister: jasom: What OS are you using?
2014-09-26T18:44:10Z jasom: drmeister: gentoo
2014-09-26T18:44:13Z jasom: liux
2014-09-26T18:44:15Z jasom: linux
2014-09-26T18:44:18Z stassats: i'm not going to build it if it's that hard, so, i'm waiting for the path to be cleared
2014-09-26T18:44:37Z stassats: there are more fun things to do that wrestling build systems
2014-09-26T18:45:02Z drmeister: stassats: Your "no way" response when I first said I released it somehow made it all worth while.
2014-09-26T18:45:30Z stassats: i was expecting it in May or so
2014-09-26T18:45:39Z stassats: (of 2014)
2014-09-26T18:46:38Z bbyler_tho quit (Remote host closed the connection)
2014-09-26T18:46:54Z drmeister: Yeah - the schedule slipped a bit - I expected it in May or so of 2013. (sigh)
2014-09-26T18:47:01Z jkaye quit (Remote host closed the connection)
2014-09-26T18:47:13Z drmeister: Oh well, better late than me sedated in a padded room.
2014-09-26T18:47:38Z jkaye joined #lisp
2014-09-26T18:47:44Z faheem__1: drmeister: hi
2014-09-26T18:47:46Z pjb: drmeister: But for a coarse discriminant in gnu make, you can use http://paste.lisp.org/+3300
2014-09-26T18:48:22Z jkaye quit (Remote host closed the connection)
2014-09-26T18:48:22Z pjb: drmeister: the distribution script identifies the system, distribution and release where it runs.
2014-09-26T18:48:37Z jkaye joined #lisp
2014-09-26T18:48:40Z drmeister: I see.  I'm checking it out
2014-09-26T18:48:41Z pjb: Notice how MacOSX is actually a darwin system.
2014-09-26T18:48:47Z drmeister: Thank you.
2014-09-26T18:48:57Z drmeister: I never got that darwin thing - where the heck did that come from?
2014-09-26T18:48:58Z pjb: But there are also darwin systems that are not MacOSX.  So using uname is too coarse.
2014-09-26T18:49:08Z _d3f joined #lisp
2014-09-26T18:49:09Z pjb: darwin is the unix under MacOSX.
2014-09-26T18:49:10Z drmeister: Are there - so there's history.
2014-09-26T18:49:23Z drmeister: faheem__1: How is it going?
2014-09-26T18:49:36Z faheem__1: drmeister: ok.
2014-09-26T18:49:54Z pjb: darwin is basically mach (the kernel), with a BSD 4.3 implemented over mach.
2014-09-26T18:49:54Z faheem__1: so, news on the clasp build system?
2014-09-26T18:50:05Z drmeister: jasom: If faheem__1 can build then we should be able to get you up and running pretty quickly.
2014-09-26T18:50:24Z urandom__ joined #lisp
2014-09-26T18:50:36Z drmeister: faheem__1: I was going to ask you - where did you get to?  First things,  externals-clasp built - correct?
2014-09-26T18:50:42Z faheem__1: drmeister: i think he may be working with your latest version
2014-09-26T18:51:00Z faheem__1: may not
2014-09-26T18:51:09Z jasom: faheem__1: This was with yesterday and one from about an hour ago; same error
2014-09-26T18:51:20Z faheem__1: drmeister: where i was an hr or two ago. externals-clasp builds, yes
2014-09-26T18:51:22Z jasom: 999aab
2014-09-26T18:51:38Z faheem__1: clasp exits with obscure error
2014-09-26T18:51:58Z faheem__1: jasom: gcc version?
2014-09-26T18:52:21Z jasom: 4.7.3
2014-09-26T18:52:36Z drmeister: jasom: I think you need 4.8 or higher.
2014-09-26T18:52:56Z drmeister: This is the bleeding edge.
2014-09-26T18:52:57Z faheem__1: jasom: right
2014-09-26T18:53:02Z Nizumzen quit (Quit: KVIrc 4.2.0 Equilibrium http://www.kvirc.net/)
2014-09-26T18:53:16Z faheem__1: i'm using 4.9. that should be easy on gentoo
2014-09-26T18:53:47Z jasom: faheem__1: 4.9 is hard-masked so I'll use 4.8
2014-09-26T18:53:59Z faheem__1: jasom: ok
2014-09-26T18:54:28Z jasom: can I set CC to have it use the non-default system gcc?
2014-09-26T18:55:24Z faheem__1: jasom: set -> export GCC_EXECUTABLE = /usr/bin/gcc-4.9 and export GXX_EXECUTABLE = /usr/bin/g++-4.9 in local.config
2014-09-26T18:55:53Z Patzy quit (Ping timeout: 240 seconds)
2014-09-26T18:56:50Z Patzy joined #lisp
2014-09-26T18:58:23Z madrik quit (Quit: sleep)
2014-09-26T18:59:34Z jasom: just for getting my ducks in a row, I don't need gold for this, right?
2014-09-26T19:00:05Z akkad quit (Ping timeout: 272 seconds)
2014-09-26T19:00:13Z drmeister: Oh man, users helping users - this is awesome.
2014-09-26T19:00:39Z jasom: we need a clasp tag on stackoverflow now ;)
2014-09-26T19:00:39Z faheem__1: jasom: the linker? not sure. what version do you have?
2014-09-26T19:01:03Z jasom: faheem__1: binutils 2.23.2
2014-09-26T19:01:11Z francogrex joined #lisp
2014-09-26T19:01:23Z faheem__1: i've got 2.22 here. debian wheezy.
2014-09-26T19:01:43Z faheem__1: 2.23 should be fine.
2014-09-26T19:01:58Z jasom: one on debian, one on gentoo, that should be a really good crossection of linux users.  What did drmeister use to do his builds?
2014-09-26T19:02:13Z faheem__1: jasom: boost build :-(
2014-09-26T19:02:29Z eudoxia quit (Quit: Lost terminal)
2014-09-26T19:02:30Z mhd quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
2014-09-26T19:02:42Z faheem__1: i tried using it in 2004/2005, still have horrible memories of it.
2014-09-26T19:02:43Z drmeister: jasom: Do you mean what system?
2014-09-26T19:02:44Z jasom: no testing on linux himself?
2014-09-26T19:02:48Z stacksmi` joined #lisp
2014-09-26T19:02:53Z jasom: drmeister: what linux distro
2014-09-26T19:03:10Z faheem__1: jasom: oh, you mean the OS?
2014-09-26T19:03:13Z jasom: yeah
2014-09-26T19:03:13Z drmeister: jasom: Yes, I test on linux but it's a weird system.  It was still at gcc 4.2 or something.  I laddered up to gcc 4.8
2014-09-26T19:03:17Z faheem__1: sorry, misunderstood
2014-09-26T19:03:25Z stassats: and why is fill_spec_vector using ECL_CLASS_OF? if, supposedly, the frame contains a function already
2014-09-26T19:03:25Z faheem__1: drmeister: gentoo?
2014-09-26T19:03:25Z akkad joined #lisp
2014-09-26T19:03:27Z gravicappa quit (Remote host closed the connection)
2014-09-26T19:03:28Z drmeister: The only good thing about it is 64 cores.
2014-09-26T19:03:31Z Xach left #lisp
2014-09-26T19:03:32Z oudeis quit (Quit: This computer has gone to sleep)
2014-09-26T19:03:37Z lifenoodles quit (Ping timeout: 245 seconds)
2014-09-26T19:03:43Z Karl_Dscc joined #lisp
2014-09-26T19:04:04Z drmeister: stassats: Are you asking me?
2014-09-26T19:04:05Z jasom: faheem__1: at a university, so RHEL or CoreOS seems likely
2014-09-26T19:04:18Z jasom: not coreos
2014-09-26T19:04:23Z jasom: centos
2014-09-26T19:04:29Z stassats: drmeister: probably, since your clasp is going to broken the same way as ECL
2014-09-26T19:04:38Z faheem__1: jasom: ok
2014-09-26T19:04:43Z stassats: drmeister: but i don't expect you to understand the GF caching
2014-09-26T19:04:47Z stassats: it is incomprehensible
2014-09-26T19:06:10Z drmeister: stassats: Nope, it's broken in entirely different, new ways.
2014-09-26T19:06:37Z stassats: drmeister: you can test it http://sourceforge.net/p/ecls/bugs/295/
2014-09-26T19:07:15Z faheem__1: jasom: don't forget to check the other env variables in local.config.linux
2014-09-26T19:07:20Z stacksmith quit (Ping timeout: 260 seconds)
2014-09-26T19:07:31Z jasom: gcc-4.8 is still building
2014-09-26T19:07:35Z faheem__1: i have export GCC_TOOLCHAIN = /usr
2014-09-26T19:07:39Z faheem__1: jasom: ok
2014-09-26T19:08:49Z lifenoodles joined #lisp
2014-09-26T19:08:56Z drmeister: jasom: the local.config.linux file is only there as a template for local.config.   makefile includes local.config to set those few variables
2014-09-26T19:09:05Z drmeister: Both externals-clasp and clasp use the same approach.
2014-09-26T19:09:19Z faheem__1: what does -> Control stack guard page temporarily disabled: proceed with caution
2014-09-26T19:09:26Z faheem__1: mean? i'm on sbcl.
2014-09-26T19:09:39Z stassats: exactly what it says
2014-09-26T19:09:48Z faheem__1: i had a recursion which crapped out because the stack or whatever was exhausted
2014-09-26T19:10:14Z faheem__1: stassats: that would be helpful if i had any idea what a Control stack guard page was
2014-09-26T19:10:22Z faheem__1: google time?
2014-09-26T19:10:32Z stassats: are you implementing a stack?
2014-09-26T19:11:02Z stassats: if no, you don't really need to know, just know that you used too much stack
2014-09-26T19:11:14Z faheem__1: stassats: does this persist across function calls?
2014-09-26T19:11:33Z faheem__1: i'm assuming is was for that function only.
2014-09-26T19:11:37Z ndrei joined #lisp
2014-09-26T19:11:39Z ndrei_ joined #lisp
2014-09-26T19:11:51Z sroy_ joined #lisp
2014-09-26T19:11:58Z tadni joined #lisp
2014-09-26T19:12:13Z stassats: it's for each thread
2014-09-26T19:12:33Z faheem__1: stassats: oh. so I need to restart, or what?
2014-09-26T19:12:40Z faheem__1: not sure what a thread means here
2014-09-26T19:12:56Z sdemarre quit (Ping timeout: 260 seconds)
2014-09-26T19:13:00Z stassats: if don't hit the hard guard page and unwind, you may be safe
2014-09-26T19:13:20Z faheem__1: stassats: ok
2014-09-26T19:13:25Z sheilong joined #lisp
2014-09-26T19:13:30Z faheem__1: i'll ignore it for now
2014-09-26T19:14:18Z drmeister: Ok.  Mystery solved.  My friend on #llvm got it working because he hasn't pulled an update since last night.
2014-09-26T19:14:48Z tadni quit (Read error: Connection reset by peer)
2014-09-26T19:16:30Z faheem__1: drmeister: he got clasp to build?
2014-09-26T19:17:09Z puchacz joined #lisp
2014-09-26T19:17:25Z Fare: drmeister, what are the next TODOs?
2014-09-26T19:18:18Z Fare: darwin has a BSD variant inside Mach, not on top of it.
2014-09-26T19:19:14Z Fare: apparently they decided to use Mach for scheduling threads instead of fixing the BSD kernel.
2014-09-26T19:19:45Z Fare: but they are not dumb enough to cross a privilege level from Mach to the BSD kernel, as in a "single server" architecture.
2014-09-26T19:20:11Z Fare: therefore, despite the use of Mach, darwin is NOT a microkernel.
2014-09-26T19:20:55Z drmeister: faheem__1: Yes, he got clasp to build but it is from the repo last night before I made a bunch of changes.
2014-09-26T19:21:03Z drmeister: I broke something - I'm trying to figure out what.
2014-09-26T19:21:41Z Fare: (which is fine: microkernels are mostly dumb; only seL4 qualifies as better, because of the "proven" part)
2014-09-26T19:21:57Z Fare: drmeiste_, git bisect
2014-09-26T19:22:05Z faheem__1: drmeister: ok. do a diff, maybe?
2014-09-26T19:22:20Z faheem__1: recursive diff.
2014-09-26T19:22:31Z faheem__1: or git bisect, sure.
2014-09-26T19:23:34Z faheem__1: is return-from the standard way to exit immediately from a function?
2014-09-26T19:23:49Z drmeister: Right - I'm on it.
2014-09-26T19:23:57Z impulse quit (Ping timeout: 258 seconds)
2014-09-26T19:23:58Z stassats: it's no really immediate, though
2014-09-26T19:24:11Z faheem__1: stassats: it isn't?
2014-09-26T19:24:37Z stassats: it may have to do a lot of things
2014-09-26T19:24:40Z faheem__1: drmeister: or sitting and staring at your code is also an option. sometimes that works
2014-09-26T19:24:52Z faheem__1: stassats: oh. any alternatives?
2014-09-26T19:25:31Z stassats: it may need to call the cleanup forms, restore special bindings, etc.
2014-09-26T19:25:31Z jasom: faheem__1: getting a lot of make[5]: *** No rule to make target `clean'.  Stop.
2014-09-26T19:25:42Z stassats: of course, any  return will have to do that
2014-09-26T19:25:52Z faheem__1: jasom: this is with the clasp externals
2014-09-26T19:25:54Z faheem__1: ?
2014-09-26T19:25:59Z jasom: faheem__1: yes
2014-09-26T19:26:12Z faheem__1: jasom: you have the current version?
2014-09-26T19:26:13Z jasom: and ~/local/externals-clasp/release/bin/bjam: No such file or directory
2014-09-26T19:26:48Z jasom: faheem__1: yup
2014-09-26T19:27:01Z faheem__1: jasom: hmm, i was seeing that too. did you run a > make clean
2014-09-26T19:27:19Z faheem__1: make sure your build directory is empty too
2014-09-26T19:27:53Z faheem__1: you have 7 things defined in your local.config?
2014-09-26T19:28:25Z jasom: ugh ugh ugh it's using git clone --depth -1 instead of submodules or subtree?
2014-09-26T19:28:46Z faheem__1: the ones that need to be changed are (1) export GCC_TOOLCHAIN = /usr (2) export GCC_EXECUTABLE = /usr/bin/gcc-4.9 (3) export GXX_EXECUTABLE = /usr/bin/g++-4.9
2014-09-26T19:28:56Z jasom: faheem__1: yeah that's done
2014-09-26T19:28:57Z faheem__1: jasom: yes, drmeister doesn't like sobmodules.
2014-09-26T19:29:00Z MutSbeta quit (Quit: Leaving.)
2014-09-26T19:29:06Z faheem__1: jasom: ok.
2014-09-26T19:29:11Z jasom: there are issues with submodules but they are strictly better than this
2014-09-26T19:29:16Z faheem__1: how far does it get?
2014-09-26T19:29:17Z lifenoodles quit (Ping timeout: 260 seconds)
2014-09-26T19:29:20Z faheem__1: jasom: tell drmeister
2014-09-26T19:29:29Z faheem__1: i don't use git
2014-09-26T19:29:33Z sdemarre joined #lisp
2014-09-26T19:29:49Z faheem__1: over in hg land we aren't big fans of subrepos - the hg equivalent
2014-09-26T19:29:52Z bit` quit (Ping timeout: 240 seconds)
2014-09-26T19:30:04Z jasom: boehm-7.2 fails with no urle to make target `clean'.
2014-09-26T19:30:16Z faheem__1: i could try rebuilding it again. it might break
2014-09-26T19:30:18Z MutSbeta joined #lisp
2014-09-26T19:30:26Z jasom: well subtrees are how most people have agreed to do it
2014-09-26T19:30:43Z faheem__1: jasom: fair enough
2014-09-26T19:31:03Z Fare: faheem_, I actually enjoy git submodules for asdf, because it allows to NOT checkout those dependencies, and thus put the asdf checkout in your regular source-registry without "polluting" it
2014-09-26T19:31:18Z jasom: I get the same errors on boehm readline expat zliband gmp
2014-09-26T19:31:23Z jasom: then it cant find bjam
2014-09-26T19:31:25Z faheem__1: Fare: sounds sensible
2014-09-26T19:31:31Z faheem__1: jasom: one sec. rebuilding now
2014-09-26T19:32:10Z faheem__1: jasom: it gives what looks like should be a fatal error, but keeps going. weird.
2014-09-26T19:32:23Z faheem__1: let me kill it
2014-09-26T19:33:21Z drmeister: Just for this I created a  #clasp  channel.   If #lisp people want us to we could move to there.
2014-09-26T19:34:00Z drmeister will take his bits and go home
2014-09-26T19:34:02Z faheem__1: jasom: never mind. i may have been mistaken
2014-09-26T19:34:21Z lifenoodles joined #lisp
2014-09-26T19:34:32Z faheem__1: jasom: there is a handy cmd line utility on debian. try either gist or gist-paste on gentoo
2014-09-26T19:34:40Z SrPx joined #lisp
2014-09-26T19:34:44Z hiroakip joined #lisp
2014-09-26T19:34:51Z jasom: wgetpaste
2014-09-26T19:35:02Z faheem__1: you can use it to paste directly from the command line.
2014-09-26T19:35:08Z faheem__1: jasom: oh?
2014-09-26T19:35:23Z gendl quit (Quit: gendl)
2014-09-26T19:35:30Z SrPx: Hello guys, quick survey, can I? What is the reason you chose common lisp instead of scheme?
2014-09-26T19:35:46Z jasom: SrPx: it's what I learned first
2014-09-26T19:35:52Z stassats: blimey, there are two fill_spec_vector functions
2014-09-26T19:36:00Z drmeister: Oh crap - part of my problem seems to be I'm working on linux and on OS X at the same time, in emacs and keep loosing changes because I'm forgetting to sync things.
2014-09-26T19:36:00Z faheem__1: SrPx: larger standard library. bigger standard too
2014-09-26T19:36:03Z SrPx: jasom: interesting, thank you
2014-09-26T19:36:13Z SrPx: faheem__1: I see
2014-09-26T19:36:24Z fikusz quit (Remote host closed the connection)
2014-09-26T19:36:25Z faheem__1: actually, those are the same things, really
2014-09-26T19:36:27Z jasom: SrPx: the package system is my favorite thing about it in comparison
2014-09-26T19:36:36Z nyef: SrPx: A distinct lack of SchemeM systems (as opposed to the TI Explorer LispMs).
2014-09-26T19:36:49Z stassats: and M-. flew me to the wrong one
2014-09-26T19:36:49Z faheem__1: CL doesn't seem to have the python distinction between the library and the core langauges
2014-09-26T19:36:50Z SrPx: I don't get that one, nyef
2014-09-26T19:36:56Z faheem__1: afaict
2014-09-26T19:37:01Z Joreji joined #lisp
2014-09-26T19:37:04Z kushal quit (Quit: Leaving)
2014-09-26T19:37:12Z faheem__1: though people say good things about racket
2014-09-26T19:37:26Z innertracks quit (Quit: innertracks)
2014-09-26T19:38:03Z nyef: SrPx: I came to common lisp by way of lisp machine emulation. There were no major commercial scheme machines that I'm aware of, so I wound up at lisp instead of scheme.
2014-09-26T19:38:18Z Fare: nyef: there were a few Scheme Machines.
2014-09-26T19:38:32Z H4ns: was there any minor commercial scheme machine?
2014-09-26T19:38:41Z nyef: Fare: Sure, but publicized on the level of the TI hardware?
2014-09-26T19:38:43Z Fare: commercial? Not sure.
2014-09-26T19:38:53Z nightfly: I chose common lisp because I heard it was a language with performance that can rival c's, in some cases, yet is high level and dynamic
2014-09-26T19:39:00Z francogrex quit (Remote host closed the connection)
2014-09-26T19:39:02Z jasom: faheem__1: https://gist.github.com/4270c0ac8f0d7921fa6d
2014-09-26T19:39:09Z Fare: some were actually built, but I'm not sure they were ever commercialized.
2014-09-26T19:39:16Z sroy_: SrPx the shortcut argument is that scheme is more widely used in academic contexts while CL is more "industry proven"
2014-09-26T19:40:01Z faheem__1: jasom: error: "none" is not a known value of feature <optimization> hmm.
2014-09-26T19:40:04Z SrPx: nightfly: so is true about scheme, no?
2014-09-26T19:40:07Z faheem__1: jasom: let me see what i get
2014-09-26T19:40:08Z SrPx: sroy_: I see
2014-09-26T19:40:12Z Fare: drewc: herep
2014-09-26T19:40:53Z Nizumzen joined #lisp
2014-09-26T19:41:03Z sroy_: SrPx I think it realy depends on what you have to do. It's quicker to implement new software from CL standard library. But you get something more concise or "specializable" in scheme due to it being minimalist
2014-09-26T19:41:34Z Fare: at StrangeLoop, there was a guy who had managed to salvage a few of the SwissAir TI Explorers.
2014-09-26T19:42:34Z sroy_: I don't think they are dissimilar enough from one another to justify not learning both ;)
2014-09-26T19:43:05Z faheem__1: jasom: sorry, i got nothing. at that same point i get ...found 586 targets...
2014-09-26T19:43:06Z faheem__1: ...updating 309 targets...
2014-09-26T19:43:50Z jeremyhe_ joined #lisp
2014-09-26T19:44:17Z faheem__1: the -> /bin/sh: 1: /usr/local/src/clasp/externals-clasp-build/release/bin/bjam: not found
2014-09-26T19:44:20Z faheem__1: i get too.
2014-09-26T19:46:00Z faheem__1: jasom: http://stackoverflow.com/questions/23013433/how-to-install-modular-boost is possibly relevant
2014-09-26T19:46:36Z faheem__1: the author is on gentoo too, with the same error
2014-09-26T19:46:39Z bit` joined #lisp
2014-09-26T19:46:54Z faheem__1: caused by a Gentoo configuration file.
2014-09-26T19:47:00Z jeremyheiler quit (Ping timeout: 260 seconds)
2014-09-26T19:47:03Z faheem__1: yay for specificity.
2014-09-26T19:47:24Z gingerale quit (Ping timeout: 272 seconds)
2014-09-26T19:48:56Z faheem__1: jasom: i see your problem -> /usr/share/boost-build/site-config.jam:9: in modules.load from module site-config
2014-09-26T19:49:09Z jtza8 joined #lisp
2014-09-26T19:49:21Z faheem__1: boost build is being naughty. it should not be looking outside the tree
2014-09-26T19:49:32Z faheem__1: see http://boost.2283326.n4.nabble.com/Building-Boost-from-Modular-Boost-Repository-td4661179.html
2014-09-26T19:50:11Z sdemarre quit (Ping timeout: 244 seconds)
2014-09-26T19:50:22Z faheem__1: jasom: i have a /usr/share/boost-build too, but the only file there is boost-build.jam
2014-09-26T19:50:34Z byte48 quit (Ping timeout: 272 seconds)
2014-09-26T19:50:35Z faheem__1: what does site-config.jam have in it?
2014-09-26T19:50:58Z Bicyclidine joined #lisp
2014-09-26T19:51:02Z byte48 joined #lisp
2014-09-26T19:51:08Z faheem__1: boost-build.jam has the line -> boost-build kernel ;
2014-09-26T19:52:11Z faheem__1: drmeister: add a --ignore-site-config somewhere.
2014-09-26T19:52:16Z innertracks joined #lisp
2014-09-26T19:52:30Z Davidbrcz joined #lisp
2014-09-26T19:52:32Z oleo is now known as Guest94473
2014-09-26T19:53:32Z lifenoodles quit (Ping timeout: 245 seconds)
2014-09-26T19:54:03Z innertracks quit (Client Quit)
2014-09-26T19:55:44Z drmeister: Shoot, the build process needs ECL and I took it out.
2014-09-26T19:55:49Z Guest94473 quit (Ping timeout: 272 seconds)
2014-09-26T19:55:56Z drmeister: Are you guys getting by without it?
2014-09-26T19:57:00Z oleo__ joined #lisp
2014-09-26T19:57:01Z edgar-rft quit (Quit: conversation lost because everything crashed)
2014-09-26T19:57:29Z drmeister: I shouldn't have been working on this so late - I made mistakes
2014-09-26T19:57:45Z drmeister: Is there a way to restore a directory that was deleted several revisions back?
2014-09-26T19:57:52Z drmeister: Several commits back.
2014-09-26T19:58:04Z oleo__ is now known as oleo
2014-09-26T19:58:08Z Ven joined #lisp
2014-09-26T19:58:34Z Grue`: everything that has been commited can be restored
2014-09-26T19:58:53Z lifenoodles joined #lisp
2014-09-26T19:59:00Z Grue`: if you still can see the commit in the log simply checkout this commit
2014-09-26T19:59:14Z Grue`: otherwise you need to look up git reflog, it should be there somewhere
2014-09-26T20:00:32Z drmeister: I have the SHA for it.  If I check out that SHA - and the directory is back - how would I restore it in the HEAD?
2014-09-26T20:01:39Z Grue`: is that directory removed as a single commit? you can do "git revert"
2014-09-26T20:01:55Z Grue`: to create a commit that reverts that action
2014-09-26T20:02:56Z Grue`: or I think you can merge this specific directory from that commit
2014-09-26T20:03:20Z EvW1 joined #lisp
2014-09-26T20:03:31Z drmeister: I'm back in the commit just prior to "git rm ecl" and I see the "ecl" directory (Hello!) I'd like to bring it back forward to HEAD
2014-09-26T20:04:01Z drmeister: But I'm not sure if there were other changes made in this commit - so I don't want them to come forward.
2014-09-26T20:04:41Z drmeister: How about if I create a branch here and then move back to head and carefully merge that branch?
2014-09-26T20:04:45Z EvW quit (Remote host closed the connection)
2014-09-26T20:05:25Z drmeiste_ quit (Remote host closed the connection)
2014-09-26T20:06:03Z drmeiste_ joined #lisp
2014-09-26T20:06:32Z nyef: drmeister: How about you look at that commit (git show or gitk it) and see if there are any changes that you don't want?
2014-09-26T20:07:27Z davazp joined #lisp
2014-09-26T20:08:29Z Fare: drmeister, you should be able to git checkout the stuff into existence.
2014-09-26T20:08:33Z svetlyak40wt quit (Remote host closed the connection)
2014-09-26T20:09:04Z Fare: you can git show --stat to see what the commit contains
2014-09-26T20:09:22Z drmeister: nyef: It was a whole directory that was deleted - there is megabytes of stuff to wade through
2014-09-26T20:09:42Z Fare: git show --stat
2014-09-26T20:09:44Z drmeister: git show --stat - that's useful
2014-09-26T20:09:50Z nyef: drmeister: Okay, see if you can get the diffstat, which would tell you if there was anything outside the di... yeah.
2014-09-26T20:10:26Z nyef: You can then also diff specific files.
2014-09-26T20:10:42Z drmeister: Gay, there's a lot of stuff.   Maybe I should just install ecl back in there.
2014-09-26T20:10:54Z drmeister: s/Gay/Gah/
2014-09-26T20:11:32Z drmeister: What's another major software package dropped into an already large externals-clasp repository between friends
2014-09-26T20:11:37Z faheem__1: drmeister: where is ecl used?
2014-09-26T20:11:46Z faheem__1: to build clasp itself?
2014-09-26T20:12:07Z drmeister: It's used to scrape symbols from the C++ code that are exposed in the Common Lisp environment
2014-09-26T20:12:17Z drmeister: Not much - I wish I'd written it in Python.
2014-09-26T20:12:31Z faheem__1: drmeister: why?
2014-09-26T20:12:33Z drmeister: I was kind of into Common Lisp there for a while.
2014-09-26T20:12:41Z faheem__1: drmeister: lol
2014-09-26T20:13:16Z faheem__1: drmeister: you could just have people run that from an external ecl. that would be reasonable, if you aren't actually using the source
2014-09-26T20:13:49Z faheem__1: i thought the ecl maintainer had retired or something.
2014-09-26T20:14:06Z faheem__1: but it still seems to be active
2014-09-26T20:14:07Z nyef: For that matter, would it make sense to write it in C?
2014-09-26T20:14:59Z dlowe: it would from a bootstrapping perspective
2014-09-26T20:15:05Z faheem__1: sourceforge, yuck
2014-09-26T20:15:19Z Alfr joined #lisp
2014-09-26T20:16:29Z yeticry quit (Ping timeout: 260 seconds)
2014-09-26T20:17:04Z sroy_ quit (Quit: Leaving)
2014-09-26T20:17:06Z drmeister: Grue`: That worked - thank you much!
2014-09-26T20:17:16Z drmeister: ECL is back in the house!
2014-09-26T20:17:26Z yeticry joined #lisp
2014-09-26T20:18:25Z lifenoodles quit (Quit: ZNC - http://znc.in)
2014-09-26T20:18:26Z stassats: oh, blimey, i get it
2014-09-26T20:18:42Z fortitude: stassats: progress?
2014-09-26T20:19:17Z stassats: so, the problem is, if you have an eql specializer (find-class 'customer), but you first call the GF with (make-instance 'customer), it caches (class-of (make-instance 'customer))
2014-09-26T20:19:32Z stassats: so, the eql instance of (find-class 'customer) will match that (class-of (make-instance 'customer))
2014-09-26T20:19:59Z drmeister: Wait - something is wrong.   Now the externals-clasp/makefile appears old as well.
2014-09-26T20:20:22Z stassats: fortitude: i was looking at the wrong fill_spec_vector all along, due to etags and emacs directing me sideways
2014-09-26T20:20:41Z drmeister: How do I ensure that I'm back at the HEAD?
2014-09-26T20:21:00Z bgs100 joined #lisp
2014-09-26T20:21:14Z stassats: now, i'm not sure how to fix it, i need to differentiate the cache entry for EQL and for classes
2014-09-26T20:21:27Z fortitude: stassats: wait a sec, isn't caching the eql-specialized instance what you want, since the recursion's on the unspecialized method?
2014-09-26T20:21:44Z stassats: fortitude: both caches are the same
2014-09-26T20:21:53Z stassats: the EQL and CLASS-OF is the same thing
2014-09-26T20:21:54Z mhd joined #lisp
2014-09-26T20:22:43Z stassats: the cache entry is cache = if (eql-specializer) argument; else class-of(argument)
2014-09-26T20:22:51Z stassats: which happens to be the same thing in this case
2014-09-26T20:22:51Z fortitude: maybe I'm not clear enough on how the method caching thing works
2014-09-26T20:22:53Z lifenoodles joined #lisp
2014-09-26T20:23:02Z fortitude: but it's pulling the unspecialized method out of the cache, no?
2014-09-26T20:23:13Z stassats: the hash key is coalesced
2014-09-26T20:23:45Z Patzy quit (Ping timeout: 258 seconds)
2014-09-26T20:24:19Z Patzy joined #lisp
2014-09-26T20:24:48Z fortitude has obviously spent too much time away from real data structures
2014-09-26T20:25:20Z stassats: it's no data-structures, it just doesn't differentiate between eql specializers and other specializers
2014-09-26T20:25:37Z knob quit (Quit: Leaving)
2014-09-26T20:25:38Z stassats: which is fine until the eql specializer is a class, and the same as one of the class specializers
2014-09-26T20:25:54Z fortitude: oh, I see what you're saying
2014-09-26T20:26:23Z jkaye quit (Remote host closed the connection)
2014-09-26T20:26:31Z stassats: the key for (make-instance 'customer) argument is (class-of (make-instance 'customer)), which happens to be the same thing as (find-class 'customer), the EQL specializer
2014-09-26T20:26:33Z faheem__1: this is probably a vague question, but how important in TCO in practical CL programming?
2014-09-26T20:26:47Z stassats: faheem__1: unimportant
2014-09-26T20:26:55Z jasom: faheem__1: some programmers like to rely on it heavily, others prefer not to
2014-09-26T20:27:00Z jkaye joined #lisp
2014-09-26T20:27:11Z H4ns: faheem__1: you're not forced into tail recursion, so it is up to you to use it or not.
2014-09-26T20:27:11Z faheem__1: stassats: ok.
2014-09-26T20:27:25Z dim: I find I don't do more recursive program in CL than in another language
2014-09-26T20:27:32Z davazp quit (Remote host closed the connection)
2014-09-26T20:27:41Z faheem__1: jasom: is there opportunity to use it heavily?
2014-09-26T20:27:51Z stassats: fortitude: so, i need to fix it now, not sure how, though, adding another level of indirection is a proven way
2014-09-26T20:27:58Z faheem__1: jasom: did you seem my comments about your build failure above?
2014-09-26T20:28:21Z jasom: faheem__1: no, was AFK, I'll look in a minute
2014-09-26T20:28:45Z faheem__1: TCO seems to make it possible for larger values to work for some functions.
2014-09-26T20:29:09Z jasom: faheem__1: well several implementations have reliable TCO, so if your style is to rely on it...
2014-09-26T20:29:14Z drmeiste_ quit (Remote host closed the connection)
2014-09-26T20:29:27Z drmeiste_ joined #lisp
2014-09-26T20:29:29Z jasom: faheem__1: e.g. I've seen state-machines that call the next state as a function
2014-09-26T20:29:30Z stassats: making the key (cons eql-value something) for eql, and classes for others, but now ecl_search_cache needs to follow that indirection
2014-09-26T20:29:38Z faheem__1: jasom: i guess the question is how much scape there is to use it.
2014-09-26T20:29:48Z jasom: I believe that the lisp implementation of Shen requires LTO for it to work
2014-09-26T20:29:56Z jasom: TCO even
2014-09-26T20:30:13Z faheem__1: it seems like a specialised technique.
2014-09-26T20:30:27Z stassats: it's a good way to make your code more obscure
2014-09-26T20:30:48Z resttime joined #lisp
2014-09-26T20:31:10Z Grue`: 99% of the time you just write (loop blah blah ) and go on your way
2014-09-26T20:31:25Z jkaye quit (Ping timeout: 258 seconds)
2014-09-26T20:31:25Z faheem__1: no, ecl does seem pretty quiet. nothing past 2012
2014-09-26T20:31:36Z faheem__1: stassats: hmm
2014-09-26T20:31:41Z Grue`: recursion is good for tree-like structures and TCO doesn't work very well with those
2014-09-26T20:31:50Z faheem__1: Grue`: oh
2014-09-26T20:32:02Z stassats: faheem__1: ecl is a bit active, but nobody makes a release
2014-09-26T20:32:13Z stassats: active in mostly fixing bugs, though, no new features
2014-09-26T20:32:19Z faheem__1: stassats: i just cloned the git repos. am i looking in the wrong place?
2014-09-26T20:32:26Z stassats: you are
2014-09-26T20:32:31Z faheem__1: git://ecls.git.sourceforge.net/gitroot/ecls/ecl
2014-09-26T20:32:34Z faheem__1: stassats: oh
2014-09-26T20:32:55Z jasom: faheem__1: ahh ignore-site-config.  I'll have to make a PR for that
2014-09-26T20:32:59Z robot-beethoven joined #lisp
2014-09-26T20:32:59Z faheem__1: stassats: so, what is the correct place?
2014-09-26T20:33:10Z faheem__1: jasom: ok. does it work?
2014-09-26T20:33:10Z flip214: faheem__1: try git.code.sf.net/p/ecls/ecl
2014-09-26T20:33:19Z stassats: my .git/config says ssh://stassats@git.code.sf.net/p/ecls/ecl
2014-09-26T20:33:19Z faheem__1: flip214: ok
2014-09-26T20:33:25Z stassats: not sure what to make of it, though
2014-09-26T20:33:39Z fortitude: that's what I've been using, bugfixes show up there now and then
2014-09-26T20:33:40Z flip214: commit 898a57818bea33119b1f5c53457f88e79f680fd3
2014-09-26T20:33:41Z flip214: Author: Arto Bendiken <arto@bendiken.net>
2014-09-26T20:33:41Z flip214: Date:   Tue Sep 23 14:07:41 2014 +0000
2014-09-26T20:34:08Z flip214: that's current git head
2014-09-26T20:34:16Z McMAGIC--Copy joined #lisp
2014-09-26T20:34:19Z jasom: faheem__1: it's running
2014-09-26T20:34:47Z faheem__1: jasom: ok. yes PR, or you just tell drmeiste_ what to change
2014-09-26T20:35:21Z faheem__1: flip214: what i correct in thinking that the main/original author stepped away?
2014-09-26T20:35:26Z stassats: a new cons will cons, i guess i can use more vector space then
2014-09-26T20:35:27Z faheem__1: did others pick it up?
2014-09-26T20:35:36Z stassats: two entries for a cache entry
2014-09-26T20:36:05Z Fare: faheem_, yes, juanjo stepped away from ECL maintenance, and there hasn't been any significant development since
2014-09-26T20:36:06Z stassats: faheem__1: he did, but nobody really stepped up
2014-09-26T20:36:12Z Fare: indeed, big bugs remain
2014-09-26T20:36:34Z stassats: for example, nobody is willing to make a release, there are a lot of new fixes
2014-09-26T20:36:39Z jeremyhe_ is now known as jeremyheiler
2014-09-26T20:37:10Z faheem__1: Fare, stassats that's too bad. that's the problem with a mainly one person project
2014-09-26T20:37:24Z Longlius quit (Remote host closed the connection)
2014-09-26T20:38:31Z fortitude: faheem__1: as a data point, I've been using ECL in a project of mine for a while, and except for a typo and this method dispatch bug, I haven't run into any showstopper issues
2014-09-26T20:39:11Z fortitude: not exactly seeing heavy use, but for certain values of "usable"...
2014-09-26T20:39:12Z faheem__1: fortitude: this thing you are working on right now? is it fast enough? or is speed not an issue?
2014-09-26T20:39:18Z stassats: the method dispatch is about to get fixed
2014-09-26T20:39:55Z fortitude: faheem__1: hard to say, because a big part of it involves calling back out into another language, and that's pretty slow by itself
2014-09-26T20:39:55Z faheem__1: well, as long as there are people ready to accept fixes...
2014-09-26T20:39:59Z boogie quit (Remote host closed the connection)
2014-09-26T20:40:01Z fortitude: the just-ECL stuff seems to be ok
2014-09-26T20:40:06Z faheem__1: fortitude: true
2014-09-26T20:40:25Z faheem__1: i tried it once. it seemed slow. number crunchy stuff
2014-09-26T20:40:28Z oudeis joined #lisp
2014-09-26T20:40:49Z faheem__1: sbcl is the speed king for that, followed by ccl. the others are all much slower
2014-09-26T20:41:00Z fortitude: faheem__1: worth noting that this project is not particularly compute-intensive
2014-09-26T20:41:09Z faheem__1: fortitude: sure
2014-09-26T20:41:18Z faheem__1: lots of things aren't, i guess
2014-09-26T20:41:34Z Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2014-09-26T20:41:50Z faheem__1: lots of people use python happily, and it is as slow as a dog. R ditto, and it is even slower.
2014-09-26T20:42:43Z faheem__1 wonders why he used the idiot phrase "slow as a dog"
2014-09-26T20:43:01Z fortitude: you were presumably thinking of "dog-slow"
2014-09-26T20:43:02Z stassats: slow as a human
2014-09-26T20:43:08Z faheem__1: fortitude: probably
2014-09-26T20:43:13Z oudeis quit (Client Quit)
2014-09-26T20:43:29Z faheem__1: which doesn't make much more sense
2014-09-26T20:43:34Z josteink quit (Ping timeout: 272 seconds)
2014-09-26T20:44:02Z patojo joined #lisp
2014-09-26T20:44:04Z marsbot is now known as marsam
2014-09-26T20:45:02Z eschulte quit (Remote host closed the connection)
2014-09-26T20:45:18Z Blaguvest quit
2014-09-26T20:46:44Z drmeister: faheem__1, jasom: I restored ecl to the externals-clasp   -  you may be using your own version so it shouldn't effect you.
2014-09-26T20:47:01Z drmeister: I got clasp to build and compile CL sources on OS X.
2014-09-26T20:47:06Z faheem__1: drmeister: i dunno if i am.
2014-09-26T20:47:22Z faheem__1: drmeister: so, rebuild externals-clasp?
2014-09-26T20:47:45Z Hydan quit (Quit: ZNC - http://znc.in)
2014-09-26T20:47:51Z drmeister: If you want you can pull the externals-clasp and just use "make ecl-setup"  and "make ecl-build" it should just build ecl.
2014-09-26T20:48:01Z drmeister: Maybe start with "make ecl-clean"
2014-09-26T20:48:12Z Nizumzen quit (Ping timeout: 245 seconds)
2014-09-26T20:48:23Z Hydan joined #lisp
2014-09-26T20:48:29Z faheem__1: drmeister: well, i cleaned teh whole thing, because jasom was having issues
2014-09-26T20:48:46Z faheem__1: building again
2014-09-26T20:49:28Z drmeister: You know, the more you rebuild externals-clasp, the stronger the bits will be set on your hard drive and the better everything will run.
2014-09-26T20:49:36Z ``Erik quit (Quit: leaving)
2014-09-26T20:50:02Z drmeister: I really appreciate your perseverance.
2014-09-26T20:50:26Z stassats: fortitude: alright, it's fixed now
2014-09-26T20:50:50Z faheem__1: drmeister: jasonm said he would do a pr for the gentoo build issue.
2014-09-26T20:50:56Z faheem__1: fyi
2014-09-26T20:50:59Z troydm joined #lisp
2014-09-26T20:51:05Z drmeister: pr?
2014-09-26T20:51:10Z fortitude: stassats: you, sir, are my hero
2014-09-26T20:51:23Z faheem__1: though i think it is a 1 line fix, so if he was here could just tell you
2014-09-26T20:51:25Z stassats: i don't even use ecl, but bugs are fun
2014-09-26T20:51:27Z faheem__1: drmeister: pull request
2014-09-26T20:51:30Z drmeister: Ah.
2014-09-26T20:51:57Z faheem__1: stassats: you're in the minority there. report 'em. i don't like 'em.
2014-09-26T20:52:12Z fortitude: stassats: the trick is knowing enough context to be able to actually /do/ something
2014-09-26T20:52:32Z fortitude: instead of e.g. staring blankly at a hash function wondering what's going on
2014-09-26T20:52:37Z stassats: nah, you just figure things out as you go
2014-09-26T20:53:02Z faheem__1: fortitude: debugging is frequently about pain thresholds
2014-09-26T20:53:05Z stassats: i had no idea how it works before i started, and it would have been quicker if M-. showed me the right function
2014-09-26T20:53:07Z ltbarcly joined #lisp
2014-09-26T20:53:14Z ``Erik joined #lisp
2014-09-26T20:53:32Z fortitude: my normal resort for that is printf, but build times become.... interesting when you put in in method dispatch
2014-09-26T20:54:02Z fortitude: s/put in/put that/
2014-09-26T20:54:23Z stassats: i just love the feeling when you finally fix a bug
2014-09-26T20:54:25Z nand1` joined #lisp
2014-09-26T20:54:42Z nand1 quit (Read error: Connection reset by peer)
2014-09-26T20:55:42Z fortitude: I would think it'd be hard to make fixes that (at least nominally) don't go anywhere like a release
2014-09-26T20:56:05Z askatasuna quit (Quit: WeeChat 1.0)
2014-09-26T20:56:55Z stassats: i guess there's a case for an optimization, i'm using a bit per entry, and there are only 63 entries, so, i can just use a single 64-bit number
2014-09-26T20:56:59Z stassats: but that's for another day
2014-09-26T20:58:06Z nand1` quit (Remote host closed the connection)
2014-09-26T20:58:23Z nand1 joined #lisp
2014-09-26T21:00:37Z kuzy000_ quit (Ping timeout: 245 seconds)
2014-09-26T21:00:56Z ehu quit (Ping timeout: 250 seconds)
2014-09-26T21:01:16Z stassats: does sourceforge suck or what, i can't manage to get a commit log
2014-09-26T21:02:46Z Bicyclidine quit (Ping timeout: 272 seconds)
2014-09-26T21:03:07Z ndrei quit (Ping timeout: 245 seconds)
2014-09-26T21:03:14Z fortitude: stassats: your commonqt page on common-lisp.net seems to be down
2014-09-26T21:04:14Z ndrei_ quit (Ping timeout: 272 seconds)
2014-09-26T21:04:44Z stassats: oh noes
2014-09-26T21:05:05Z ndrei joined #lisp
2014-09-26T21:05:08Z stassats: drmeister: clasp seems to suffer from the same bug, but you'll have to find the commit on your own --- sourceforge does suck
2014-09-26T21:05:33Z jlongster quit (Ping timeout: 246 seconds)
2014-09-26T21:06:05Z ndrei_ joined #lisp
2014-09-26T21:06:45Z marsam is now known as marsbot
2014-09-26T21:07:10Z jasom: externals-clasp is built!
2014-09-26T21:07:24Z drmeister: jasom: Wonderful.  You get a gold star.
2014-09-26T21:08:17Z Shinmera: drmeister: If I had more time I'd love to test build. Hopefully some other day!
2014-09-26T21:08:17Z stassats: fortitude: fixed, i guess. thanks
2014-09-26T21:08:39Z stassats: minion: chant
2014-09-26T21:08:39Z minion: MORE TIME
2014-09-26T21:08:53Z vlnx quit (Read error: Connection reset by peer)
2014-09-26T21:08:55Z drmeister: stassats: Which bug was that?  Sorry I was only half listening.
2014-09-26T21:09:04Z stassats: drmeister: see the git top of ecl
2014-09-26T21:09:23Z stassats: i guess the only way is to clone it
2014-09-26T21:10:13Z jasom: common.mkdir /Contents
2014-09-26T21:10:21Z jasom: i gtg but that's not right
2014-09-26T21:10:34Z drmeister: I just caught up and I have externals-clasp built on linux.
2014-09-26T21:10:37Z hiyosi_ quit (Ping timeout: 260 seconds)
2014-09-26T21:11:18Z jkaye joined #lisp
2014-09-26T21:11:38Z fortitude: drmeister: stassats' patch is here: http://paste.lisp.org/display/143858
2014-09-26T21:11:49Z vlnx joined #lisp
2014-09-26T21:11:53Z ilhami quit (Quit: Nettalk6 - www.ntalk.de)
2014-09-26T21:12:22Z ltbarcly quit (Quit: Computer has gone to sleep.)
2014-09-26T21:12:41Z Bicyclidine joined #lisp
2014-09-26T21:13:38Z Fare: stassats, your #275 fix also fixed #285 and #286. Thanks!
2014-09-26T21:14:49Z stassats: ok, that was some time ago
2014-09-26T21:15:36Z stassats: if i didn't hate sourceforge, i could probably make a release
2014-09-26T21:16:00Z Fare: does the release have to be declared on sourceforge?
2014-09-26T21:16:09Z stassats: i have no idea and i don't want to find out
2014-09-26T21:16:10Z Fare: if you become maintainer, you can move to github
2014-09-26T21:16:31Z Ven joined #lisp
2014-09-26T21:16:50Z stassats: being a maintainer is too much hassle
2014-09-26T21:16:56Z sz0 quit
2014-09-26T21:17:49Z Fare: you could trick one of the other committers into being it.
2014-09-26T21:17:55Z didi joined #lisp
2014-09-26T21:18:02Z fortitude: stassats: it might make sense to just start tagging release-like parts of the repo
2014-09-26T21:18:12Z schaueho quit (Ping timeout: 245 seconds)
2014-09-26T21:18:28Z fortitude: wouldn't be any worse than now, and you'd have some wiggle room for potentially breaking changes
2014-09-26T21:18:42Z kami quit (Ping timeout: 250 seconds)
2014-09-26T21:19:03Z stassats: bendiken guy did some house cleaning, maybe he wants to
2014-09-26T21:19:24Z stassats: i only like finding bugs
2014-09-26T21:19:53Z vanila quit (Quit: Leaving)
2014-09-26T21:20:57Z didi: Hey, cool. It seems I won't need to fake binary streams from character ones after all. First of all, I discovered SBCL's bivalent streams. Second, flexi-streams appears to support `listen', `peek-*' and `read-*', which I might take advantage of trying to be a little more "portable" (yes, in scary quotes).
2014-09-26T21:22:21Z sz0 joined #lisp
2014-09-26T21:22:27Z fortitude: stassats: actually, that might not be fixed
2014-09-26T21:23:21Z fortitude: stassats: oops hang on, might just be doing something nonsensical
2014-09-26T21:24:07Z fortitude: yep, works, my mistake
2014-09-26T21:25:38Z stassats quit (Ping timeout: 250 seconds)
2014-09-26T21:26:02Z jtza8 quit (Ping timeout: 245 seconds)
2014-09-26T21:27:13Z aduadu joined #lisp
2014-09-26T21:28:26Z sheilong quit (Quit: Konversation terminated!)
2014-09-26T21:28:41Z aduadu left #lisp
2014-09-26T21:29:27Z faheem__1: drmeister: ok, externals built again. takes a while
2014-09-26T21:30:17Z faheem__1: did you check out the make delete everything issue?
2014-09-26T21:36:54Z hiyosi_ joined #lisp
2014-09-26T21:38:20Z LiamH quit (Quit: Leaving.)
2014-09-26T21:39:01Z Ven quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
2014-09-26T21:40:13Z drmeister: faheem__1: Could you go to here and paste your clasp/local.config into it, hit "Create public Gist"  and paste the URL?
2014-09-26T21:41:40Z hiyosi_ quit (Ping timeout: 250 seconds)
2014-09-26T21:44:04Z faheem__1: drmeister: here?
2014-09-26T21:44:25Z jeremyheiler quit (Quit: Computer has gone to sleep.)
2014-09-26T21:45:08Z drmeister: Yeah - it's just a URL
2014-09-26T21:45:18Z faheem__1: drmeister: https://gist.github.com/32c28fef2b2c03279c34
2014-09-26T21:45:34Z faheem__1: drmeister: i'm not sure what you mean, but that is my clasp local.config
2014-09-26T21:46:42Z drmeister: In clasp can you type:   "make clean -n"  and paste the result the same way?
2014-09-26T21:46:47Z puchacz quit (Quit: Konversation terminated!)
2014-09-26T21:48:11Z faheem__1: drmeister: just fyi. clasp exited with an error. you want it?
2014-09-26T21:48:23Z drmeister: Clasp ran?
2014-09-26T21:48:35Z faheem__1: drmeister: no, the build i mean
2014-09-26T21:49:12Z drmeister: Well, one thing at a time - I'd like to fix this "everything gets deleted problem" - it's probably related to the other problems
2014-09-26T21:49:28Z faheem__1: drmeister: ok
2014-09-26T21:50:20Z faheem__1: it's not doing that now. odd.
2014-09-26T21:50:45Z hiroakip quit (Ping timeout: 260 seconds)
2014-09-26T21:50:55Z Longlius joined #lisp
2014-09-26T21:50:56Z bbyler_tho joined #lisp
2014-09-26T21:51:22Z mhd quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
2014-09-26T21:51:37Z faheem__1: does CL not accept the same function with different args? a la C++ function overloading?
2014-09-26T21:52:06Z H4ns: faheem__1: no.
2014-09-26T21:52:08Z faheem__1: same function name
2014-09-26T21:52:18Z faheem__1: H4ns: ok
2014-09-26T21:52:32Z Davidbrcz quit (Ping timeout: 260 seconds)
2014-09-26T21:53:05Z nydel quit (Quit: WeeChat 0.4.2)
2014-09-26T21:53:28Z Bicyclidine: you can have different key arguments and stuff, but it's not like C++ overloading really.
2014-09-26T21:53:31Z faheem__1: language design decision, or difficult/impossible to do?
2014-09-26T21:53:44Z faheem__1: Bicyclidine: mm, ok
2014-09-26T21:53:53Z Bicyclidine: probably a design decision.
2014-09-26T21:53:55Z drmeister: Be back in 10 min.
2014-09-26T21:53:58Z drmeister quit (Remote host closed the connection)
2014-09-26T21:54:07Z H4ns: faheem__1: function dispatch in cl is just not architected around arity
2014-09-26T21:54:09Z Bicyclidine: actually i wonder how C++ deals with function pointers, now...
2014-09-26T21:54:28Z nydel joined #lisp
2014-09-26T21:54:51Z faheem__1: H4ns: probably sensible
2014-09-26T21:54:59Z faheem__1: if C++ does it, it's probably stupid
2014-09-26T21:55:03Z faheem__1: forgive my biases
2014-09-26T21:55:07Z ccmaru joined #lisp
2014-09-26T21:55:21Z sz0 quit (Ping timeout: 260 seconds)
2014-09-26T21:55:25Z H4ns: it is just different. nothing about "stupid"
2014-09-26T21:56:09Z faheem__1 mutters something about biases
2014-09-26T21:56:47Z Beetny joined #lisp
2014-09-26T21:58:08Z fridim__ quit (Ping timeout: 250 seconds)
2014-09-26T21:58:52Z Krystof joined #lisp
2014-09-26T21:58:56Z EvW1 quit (Ping timeout: 272 seconds)
2014-09-26T22:02:06Z c107 joined #lisp
2014-09-26T22:02:23Z jtza8 joined #lisp
2014-09-26T22:04:00Z Bicyclidine quit (Ping timeout: 272 seconds)
2014-09-26T22:04:04Z ccmaru quit (Quit: Leaving)
2014-09-26T22:04:32Z drmeister joined #lisp
2014-09-26T22:05:02Z malice: I may be late for discussion, but isn't it defmethod that allows one to "overload" function?
2014-09-26T22:05:36Z H4ns: malice: it is called "specializing", and it does not work on the basis of arity
2014-09-26T22:05:52Z ggole quit (Ping timeout: 240 seconds)
2014-09-26T22:06:20Z malice: H4ns, what does arity mean?
2014-09-26T22:06:26Z lambda quit (Read error: Connection reset by peer)
2014-09-26T22:06:35Z H4ns: malice: number of arguments
2014-09-26T22:07:14Z mishoo joined #lisp
2014-09-26T22:07:53Z malice: That's nice to know
2014-09-26T22:08:14Z lambda joined #lisp
2014-09-26T22:08:53Z _d3f quit (Quit: WeeChat 0.4.3)
2014-09-26T22:09:05Z didi: malice: It is not "specializing", but if you want/need to vary the number of arguments, you can use &key with your generic function.
2014-09-26T22:09:14Z lambda quit (Client Quit)
2014-09-26T22:10:17Z malice: or default values. Got it, thanks! :)
2014-09-26T22:11:15Z faheem__1: that's like python default values, right? or C++ default values?
2014-09-26T22:11:52Z Karl_Dscc quit (Remote host closed the connection)
2014-09-26T22:12:09Z malice: idk about python, but afaik in CL you can provide default value, so if no value is provided, the variable will be bound with this value
2014-09-26T22:12:26Z faheem__1: malice: ok
2014-09-26T22:12:35Z malice: also, you can have another value like "supplied-p", which you can use to check if the value was provided or is it default one
2014-09-26T22:12:38Z faheem__1: do they still have to be at the end?
2014-09-26T22:12:39Z malice: if it does matter to you
2014-09-26T22:12:51Z malice: default?
2014-09-26T22:12:52Z stacksmi` quit (Ping timeout: 272 seconds)
2014-09-26T22:12:54Z faheem__1: malice: ok
2014-09-26T22:13:07Z malice: no, I believe not
2014-09-26T22:13:12Z faheem__1: malice: yes, default args in python at least have to be at the end
2014-09-26T22:13:16Z malice: you can set default values for :key
2014-09-26T22:13:23Z malice: or for standard values
2014-09-26T22:13:41Z malice: arguments*
2014-09-26T22:13:44Z nyef: faheem__1: At one point I put together a construct where the optional argument went at the front, but it requires special parsing code and destroys arglist hinting. It's usually not worth it unless you're trying to simulate an existing optional-parameter-at-the-front syntax.
2014-09-26T22:13:52Z mishoo quit (Ping timeout: 240 seconds)
2014-09-26T22:13:56Z faheem__1: http://www.gigamonkeys.com/book/functions.html -> Optional Parameters
2014-09-26T22:14:19Z H4ns: sometime i like that you can bind special variables in cl argument lists
2014-09-26T22:14:34Z faheem__1: actually, i think python uses keywords for default args. but i'm too sleepy to think
2014-09-26T22:14:48Z nyef: And this was in the context of a normal function, not a generic function. If there were a generic involved, I'd've had to declare a separate interface function from the implementation.
2014-09-26T22:15:15Z faheem__1: nyef: i see. so usually at the end then
2014-09-26T22:15:31Z malice: yeah this book is pretty good
2014-09-26T22:15:35Z nyef: Also, be careful if you mix &OPTIONAL and &KEY arguments.
2014-09-26T22:15:38Z malice: it warns you about mixing various things
2014-09-26T22:15:40Z malice: yep
2014-09-26T22:15:47Z faheem__1: nyef: i will
2014-09-26T22:15:52Z nyef: (It's not usually done, and for good reason.)
2014-09-26T22:16:04Z nyef: Also &OPTIONAL and &REST.
2014-09-26T22:16:14Z nyef: And the combination of &KEY and &REST has interesting implications.
2014-09-26T22:17:49Z stacksmith joined #lisp
2014-09-26T22:19:06Z drmeister: faheem__1, jasom: What kind of errors are you getting building clasp on linux?   I'm wondering if they are the same as mine.
2014-09-26T22:19:33Z stacksmith: less /proc/cpuinfo
2014-09-26T22:19:35Z faheem__1: drmeister: vey obscure. no real error messages. anyway, i'll have to build it again...
2014-09-26T22:19:48Z faheem__1: shall i do that?
2014-09-26T22:20:01Z stacksmith: err, wrong terminal, sorry
2014-09-26T22:21:15Z faheem__1: drmeister: change your default from 16, please.
2014-09-26T22:21:27Z drmeister: Sure and stop when errors start getting thrown.  My problem may not be the same as yours.
2014-09-26T22:21:29Z drmeister: Ok.
2014-09-26T22:21:53Z normanrichards joined #lisp
2014-09-26T22:22:07Z faheem__1: drmeister: it is not clear what the errors are, honestly
2014-09-26T22:22:13Z faheem__1: maybe bjam is hiding them
2014-09-26T22:22:42Z drmeister: It doesn't do that.  Could you paste the log into gist.github.com and paste the URL here?
2014-09-26T22:23:18Z malice quit (Quit: Leaving)
2014-09-26T22:23:32Z faheem__1: drmeister: ok, directing to a file now. it might take a bit of time
2014-09-26T22:23:49Z drmeister: No problem.
2014-09-26T22:24:25Z drmeister: Everyone: if I forget myself and you feel this is crowding out Lisp discussion on #lisp - please tell me and we will take it somewhere else.
2014-09-26T22:24:48Z nyef: drmeister: You're debugging building a lisp implementation? That's actually on-topic.
2014-09-26T22:25:31Z drmeister: nyef: Thanks, but I'm sensitive. :-)
2014-09-26T22:27:10Z nyef: drmeister: For pretty much the first year that I was here, I was discussing microarchitecture and instruction sequences for an unusual CPU with a C programmer, while occasionally asking for Lisp optimization advice. Nobody complained, largely because the two of us were working on LispM emulators, and I was writing mine in Lisp.
2014-09-26T22:28:39Z nyef: drmeister: Eventually, you may end up creating your own IRC channel for your implementation, like the SBCL maintainers did (twice, actually, #lisp was originally set up by and for SBCL maintainers), but that can wait until you have actual traction with it.
2014-09-26T22:29:12Z nyef: On the other hand, you might register the channel that you want with chanserv now, just to make sure that it's yours.
2014-09-26T22:29:14Z drmeister: I did create a #clasp IRC channel and there's three people in it!  It's a movement!
2014-09-26T22:29:32Z xyjprc joined #lisp
2014-09-26T22:29:45Z faheem__1: just now i was running factorial for a large value. i noticed afterwards that the SLIME REPL appeared "sticky" and unresponsive
2014-09-26T22:30:25Z scottj joined #lisp
2014-09-26T22:30:36Z nyef: The story goes that Larry Wall was against creating a usenet group for Perl for quite a while, as a language-marketing technique. If there's no designated place for discussion, discussion happens all over the place, exposing more people to the existence of the language.
2014-09-26T22:31:33Z hiroakip joined #lisp
2014-09-26T22:33:10Z posterdati300 quit (Quit: KVIrc 4.1.3 Equilibrium http://www.kvirc.net/)
2014-09-26T22:33:41Z jtza8 quit (Remote host closed the connection)
2014-09-26T22:35:06Z marsbot is now known as marsam
2014-09-26T22:36:00Z drmeister: BRB
2014-09-26T22:36:02Z drmeister quit (Read error: Connection reset by peer)
2014-09-26T22:36:29Z jkaye quit (Remote host closed the connection)
2014-09-26T22:37:04Z jkaye joined #lisp
2014-09-26T22:37:49Z hiyosi_ joined #lisp
2014-09-26T22:38:58Z normanrichards quit
2014-09-26T22:39:33Z xyjprc: Same observation on the "sticky SLIME"
2014-09-26T22:40:26Z drmeiste_ quit (Remote host closed the connection)
2014-09-26T22:41:01Z drmeister joined #lisp
2014-09-26T22:41:06Z jkaye quit (Ping timeout: 246 seconds)
2014-09-26T22:42:29Z hiyosi_ quit (Ping timeout: 260 seconds)
2014-09-26T22:43:56Z stacksmith quit (Remote host closed the connection)
2014-09-26T22:44:22Z stacksmith joined #lisp
2014-09-26T22:45:14Z ovidnis joined #lisp
2014-09-26T22:45:37Z ndrei_ quit (Quit: Lost terminal)
2014-09-26T22:45:37Z ndrei quit (Quit: Lost terminal)
2014-09-26T22:46:00Z ndrei joined #lisp
2014-09-26T22:47:57Z drmeiste_ joined #lisp
2014-09-26T22:48:37Z faheem__1: drmeiste_: error log -> https://gist.github.com/5dad64b504ad8c465a3a
2014-09-26T22:49:18Z faheem__1: getting rid of that annoying: warning: unknown warning option '-Wno-unused-local-typedef' [-Wunknown-warning-option]
2014-09-26T22:49:21Z faheem__1: would be nice.
2014-09-26T22:50:30Z faheem__1: ../../src/asttooling/clangTooling.cc:497:71: error: no member named 'setArgumentsAdjuster' in 'clang::tooling::ClangTool' .  def("setArgumentsAdjuster",&clang::tooling::ClangTool::setArgumentsAdjuster)
2014-09-26T22:50:39Z faheem__1: that looks like a bona fide error.
2014-09-26T22:52:00Z drmeiste_: faheem__1: That's not a problem at all!
2014-09-26T22:52:17Z drmeiste_: I fixed it already - hang on.
2014-09-26T22:54:11Z fortitude quit (Quit: leaving)
2014-09-26T22:54:27Z faheem__1: i just pulled some stuff.
2014-09-26T22:54:30Z faheem__1: shall i redo?
2014-09-26T22:55:22Z nydel: i have a function #'draw.. it needs a number (between 1 and 78) to evaluate down to a set of {number} non-repeating random integers between 1 and 78. this is to simulate pulling a spread from a deck of tarot cards. someone suggested that my #'draw isn't technically functional (as in the method of programming) because it relies on something like MAKE_RANDOM_STATE ... but i say that since it does have a domain
2014-09-26T22:55:23Z nydel: and therefore a range, and since it evaluates down without side-effect, that it is in fact functional. the function is written here http://lpaste.net/111699 i'd appreciate any input on the subject, please & thank you.
2014-09-26T22:56:14Z faheem__1: apropos of nothing. can anyone recommend a good CL random number library? There does not appear to be a canonical one.
2014-09-26T22:56:41Z faheem__1: similar to R's rmath would be good.
2014-09-26T22:56:44Z schaueho joined #lisp
2014-09-26T22:56:53Z nydel: doesn't alexandria have something
2014-09-26T22:57:05Z jasom: nydel: It modifies the random state, so it's not pure
2014-09-26T22:57:05Z mhd joined #lisp
2014-09-26T22:57:09Z resttime: would saving a hash table to a file be a feasibly good way to save/read data?
2014-09-26T22:57:39Z resttime: i'm thinking of traversing hash table and writing it to a file
2014-09-26T22:57:47Z nydel: jasom: you mean when i call #'random then an outside variable is mutated
2014-09-26T22:57:52Z resttime: then when reading back, traverse file and remake hashtable
2014-09-26T22:57:54Z jasom: nydel: correct
2014-09-26T22:58:10Z nydel: jasom: rats. is there a way around this for my purposes?
2014-09-26T22:58:15Z faheem__1: resttime: i tend to have a bias towards relational db
2014-09-26T22:58:23Z jasom: nydel: if you really care about that, you can copy the random-state before passing it to the function, and then return the modified copy out along with the value
2014-09-26T22:58:23Z ovidnis quit (Quit: leaving)
2014-09-26T22:58:45Z faheem__1: some people think is too "heavy"
2014-09-26T22:59:01Z nydel: jasom: that is good, really good! i just have to implement it in a way that isn't clunky.
2014-09-26T22:59:10Z jasom: (defun pure-random (limit state) (let ((new-state (make-random-state state))) (values (random limit state) state))
2014-09-26T22:59:25Z jasom: nydel: note I just wrote that function in irssi, so there may be bugs in it
2014-09-26T22:59:27Z nydel: ha, no i don't! because there it is!
2014-09-26T22:59:29Z resttime: faheem_, a positive or negative bias?
2014-09-26T22:59:33Z nydel: thanks jasom
2014-09-26T22:59:40Z resttime: faheem__1*
2014-09-26T23:00:00Z faheem__1: resttime: positive, i guess. at least from the pov of managing data
2014-09-26T23:00:17Z faheem__1: though, honestly, i don't think rdbms are really very interesting
2014-09-26T23:00:25Z vaporatorius quit (Remote host closed the connection)
2014-09-26T23:00:27Z resttime: okay, i'm thinking of making a triple store db via the hash table
2014-09-26T23:00:45Z faheem__1: resttime: why not just used PostgreSQL?
2014-09-26T23:00:49Z resttime: not sure if if it would be better to just use lists
2014-09-26T23:01:06Z jasom: faheem__1: were you thinking something like https://github.com/tpapp/cl-random/
2014-09-26T23:01:06Z resttime: faheem__1, wanted to learn about implementing a triple store natively
2014-09-26T23:01:10Z nydel: jasom: my function "draw" simply will error for a value below 1 and above 78, maybe i can handle that inside #'pure-random without compromising functionality?
2014-09-26T23:01:21Z resttime: somebody mentioned they were good so i figured i might as well learn it
2014-09-26T23:01:26Z faheem__1: jasom: yes, i'm familar with tamas's libraries.
2014-09-26T23:01:32Z resttime: SQL stuff seems a lot more easier to understand/ learn
2014-09-26T23:01:38Z faheem__1: resttime: i see
2014-09-26T23:01:55Z resttime: so  basically curious
2014-09-26T23:02:19Z faheem__1: a good rn library is a significant missing piece in CL from my pov
2014-09-26T23:02:24Z jasom: faheem__1: there's also GSL bindings I think
2014-09-26T23:02:27Z nydel: /s/and/or/g
2014-09-26T23:02:48Z drmeiste_: faheem__1: Can you pull the clasp repo now?
2014-09-26T23:03:01Z faheem__1: jasom: yes, i've heard of that. actually rmath is not hard to bind. the trouble is when one is looping
2014-09-26T23:03:12Z faheem__1: drmeiste_: ok
2014-09-26T23:03:27Z zeitue joined #lisp
2014-09-26T23:03:42Z faheem__1: i think a pure CL rn implementation might do Ok, but i have not seen one.
2014-09-26T23:03:49Z faheem__1: drmeiste_: done
2014-09-26T23:03:53Z faheem__1: shall i rebuild?
2014-09-26T23:03:56Z drmeiste_: Yes.
2014-09-26T23:04:33Z faheem__1: building now
2014-09-26T23:04:37Z drmeiste_: The boehm version should build - the MPS version still has problems and I need some time to find the hacked MPS repo that I was using - somehow I think I reverted to trunk.
2014-09-26T23:04:47Z nydel: tpapp has some interesting projects going. this person doesn't attend freenode#lisp do they?
2014-09-26T23:04:54Z drmeiste_: I'll get rid of those warning messages as well later.
2014-09-26T23:05:03Z faheem__1: drmeiste_: i just did make. should i do something else?
2014-09-26T23:05:15Z drmeiste_: Is it compiling Common Lisp sources?
2014-09-26T23:05:16Z faheem__1: nydel: dunno. he posts on comp.lang.lisp occasionally
2014-09-26T23:05:35Z faheem__1: drmeiste_: no, c++
2014-09-26T23:05:48Z nydel: faheem__1: good to know, i'm not even subscriped to that feed
2014-09-26T23:05:50Z nydel: thanks
2014-09-26T23:06:11Z faheem__1: nydel: i'm familiar with his work a little bit. he's an economist
2014-09-26T23:06:17Z drmeiste_: faheem__1: I can't follow along at the moment - I have other problems on Linux that are preventing me from building.
2014-09-26T23:06:35Z nydel: an open-source lisp-hacking economist. what fun.
2014-09-26T23:06:39Z faheem__1: he is one of a tiny handful of people that use CL for scientific work
2014-09-26T23:06:51Z faheem__1: he seems like a cool guy, but not interested in chatting
2014-09-26T23:07:02Z dfox quit (Ping timeout: 244 seconds)
2014-09-26T23:07:21Z nydel: i'm one of those too! then when i try to make stuff that someone normal might want to use i spend the whole day on the functionality of the card-drawer.
2014-09-26T23:07:24Z faheem__1: you could probably fit everyone on the planet who uses CL for scientific work in a large classroom
2014-09-26T23:07:30Z resttime: hmm i think i'll just start out with lists for the DB to simplify things
2014-09-26T23:07:36Z drmeiste_: faheem__1: I'm going to have to leave for a couple of hours.
2014-09-26T23:07:46Z dfox joined #lisp
2014-09-26T23:07:49Z faheem__1: drmeiste_: ok
2014-09-26T23:08:02Z resttime: '((s p o) (s p o)...)
2014-09-26T23:08:04Z drmeiste_: I think this might do it though.   It should compile the C++ source and then start compiling the Common Lisp sources for Boehm.
2014-09-26T23:08:04Z faheem__1: drmeiste_: sorry, was i supposed to do make or not? i wasn't attending.
2014-09-26T23:08:14Z drmeiste_: Is it at the command line?
2014-09-26T23:08:17Z nydel: most of my work is pure maths for analysis of musical repetories and/or generation of musical composition in the style of an analyzed repetory.
2014-09-26T23:08:39Z faheem__1: drmeiste_: yes. 'make' at the command line
2014-09-26T23:08:57Z drmeiste_: From the top level directory of the clasp repo type "make"
2014-09-26T23:08:58Z svetlyak40wt joined #lisp
2014-09-26T23:09:01Z faheem__1: nydel: interesting. what is your area? CS or math, or something else.
2014-09-26T23:09:10Z faheem__1: drmeiste_: yes, that's what i did
2014-09-26T23:09:17Z drmeiste_: And what did it do?
2014-09-26T23:09:18Z ndrei quit (Quit: Lost terminal)
2014-09-26T23:09:24Z jasom: do I need a recent clang as well?
2014-09-26T23:09:27Z faheem__1: drmeiste_: still running
2014-09-26T23:09:31Z drmeiste_: jasom: Yes.
2014-09-26T23:09:34Z faheem__1: i mean, building
2014-09-26T23:09:36Z jasom: I'm on 3.4.2
2014-09-26T23:09:37Z ndrei joined #lisp
2014-09-26T23:09:51Z drmeiste_: 3.4.2 what?
2014-09-26T23:09:57Z faheem__1: drmeiste_: is getting rid of that warning a possibility.
2014-09-26T23:09:58Z jasom: clang-3.4.2
2014-09-26T23:10:08Z nydel: faheem__1: music theory. but prof. david cope at univerity santa cruz introduced me to lisp about 10 years ago, and my autistic inclination toward functional & lambda calculus threw me into whatever we're doing now.
2014-09-26T23:10:09Z faheem__1: i'm allergic to warnings. i always get rid of mine
2014-09-26T23:10:33Z nydel: see his "experiments in musical inteligence" or EMI for results in the shape of mp3
2014-09-26T23:10:35Z faheem__1: nydel: so, you are a musicologist?
2014-09-26T23:10:44Z drmeiste_: jasom: externals-clasp should have installed clang-3.6 and the clasp repo should use clang-3.6 on Linux.
2014-09-26T23:10:58Z faheem__1: drmeiste_: i thought it was 3.5
2014-09-26T23:11:09Z nydel: i think i'm a musician and mathematician. i don't really study music. i think almost all of it is not very good.
2014-09-26T23:11:12Z faheem__1: is 3.6 out?
2014-09-26T23:11:21Z jasom: nydel: I would expect many here are familiar with Cope's work
2014-09-26T23:11:21Z nydel: faheem__1: what is your area?
2014-09-26T23:11:23Z faheem__1: nydel: almost all of what?
2014-09-26T23:11:28Z drmeiste_: LLVM uses slightly different numbering scheme.
2014-09-26T23:11:36Z nydel: almost all of available composed music
2014-09-26T23:11:47Z drmeiste_: Everything up to the 3.5 release is 3.5.    3.6 starts on the first commit after the 3.5 release.
2014-09-26T23:12:05Z faheem__1: nydel: mixed. two degrees in math, one in statistics (phd). i also do a lot of computing. i'm done some work on bioinformatics
2014-09-26T23:12:19Z drmeiste_: faheem__1: What is it doing at the moment?
2014-09-26T23:12:20Z Joreji quit (Ping timeout: 260 seconds)
2014-09-26T23:12:27Z drmeiste_: I have to leave in a moment.
2014-09-26T23:12:36Z faheem__1: drmeiste_: errored out again, sorry.
2014-09-26T23:12:39Z nydel: i would hope so, i wish he was a bit more vocal in the community though. he used to keep a midi.lisp available that would make "cope events" which are lists of midi note, instrument, time, duration, volume.
2014-09-26T23:12:45Z faheem__1: you want me to repaste?
2014-09-26T23:12:47Z drmeiste_: Yes.
2014-09-26T23:13:03Z nydel: but he's gone the hermit way as many older advanced lispers do
2014-09-26T23:13:12Z svetlyak40wt quit (Ping timeout: 245 seconds)
2014-09-26T23:13:18Z faheem__1: drmeiste_: one sec
2014-09-26T23:13:21Z nydel: i wonder if i could read something about your work in bioinformatics?
2014-09-26T23:13:36Z nydel: or any of your work or logged rantings even
2014-09-26T23:13:38Z drmeiste_: From the last paste the only error was the change in the LLVM library from last week to yesterday that broke clasp.
2014-09-26T23:13:52Z drmeiste_: I fixed that and put it up in the repo.
2014-09-26T23:14:04Z faheem__1: drmeiste_: https://gist.github.com/3358f1ea62eb3f41d1c9
2014-09-26T23:14:09Z faheem__1: some new errors, at least
2014-09-26T23:14:33Z drmeiste_: Ok, now it's because ECL was missing from the externals-clasp.
2014-09-26T23:14:41Z hiroakip quit (Ping timeout: 260 seconds)
2014-09-26T23:14:43Z faheem__1: nydel: some of my stuff is on the net. i really should put up a proper web page
2014-09-26T23:15:12Z faheem__1: nydel: i can send you some stuff. do you have any background in statistics/probability?
2014-09-26T23:15:20Z normanrichards joined #lisp
2014-09-26T23:15:20Z normanrichards quit (Read error: Connection reset by peer)
2014-09-26T23:15:36Z faheem__1: it is probably not very interesting unless you are in the relevant area(s) though.
2014-09-26T23:15:42Z nydel: faheem__1: i run a host and give out vps to anyone. centos. if you like. cpanel & infinite whatever.
2014-09-26T23:16:04Z drmeiste_: Grab the latest externals-clasp and rebuild it    -  hopefully I have ECL building to the right place
2014-09-26T23:16:10Z faheem__1: nydel: oh, i have a VPS. but thanks for the offer.. :-)
2014-09-26T23:16:12Z faheem__1: drmeiste_: ok
2014-09-26T23:16:18Z nydel: faheem__1: i don't have direct background there, just what i've needed for advanced functional analysis. i would like to be less ignorant re statistics.
2014-09-26T23:16:54Z faheem__1: nydel: if you want to ping me - faheem at faheem dot info
2014-09-26T23:16:56Z Bicyclidine joined #lisp
2014-09-26T23:17:02Z drmeiste_: faheem__1: This is the most recent commit of externals-clasp:   c3240cd4114163434e5a29661d5590bfa99e989b
2014-09-26T23:17:10Z stacksmith is thinkging of going the hermit way, the way lispers do.
2014-09-26T23:17:12Z faheem__1: drmeiste_: ok
2014-09-26T23:17:28Z drmeiste_: This is the most recent commit of clasp:  f2c6cfaad5d2730e1ea8e8afae778e9338a62ba9
2014-09-26T23:17:30Z faheem__1: pulling now
2014-09-26T23:17:31Z nydel: excellent faheem__1 i hope we can trade some rss feeds or that you can point me at some that might be good for me re probability maths etc
2014-09-26T23:17:31Z SrPx quit (Ping timeout: 246 seconds)
2014-09-26T23:17:56Z urandom__ quit (Quit: Konversation terminated!)
2014-09-26T23:18:06Z normanrichards joined #lisp
2014-09-26T23:18:24Z faheem__1: drmeiste_: ok, rebuild externals?
2014-09-26T23:19:41Z faheem__1: nydel: so you are an academic?
2014-09-26T23:19:51Z faheem__1: or more freelance?
2014-09-26T23:20:12Z faheem__1: i hear rumors of corporate research. rarely meet people from there
2014-09-26T23:20:38Z nydel: freelance, i used to work for DOD (DCMA) but am not doing that now. i work a lot through/with sdf.
2014-09-26T23:20:53Z nydel: stacksmith: you come out here and say hello right this instance mister/ma'am.
2014-09-26T23:21:17Z drmeiste_: Yes rebuild externals and then rebuild clasp.
2014-09-26T23:21:35Z faheem__1: drmeiste_: ok, rebuilding externals now
2014-09-26T23:21:48Z faheem__1: nydel: sdf?
2014-09-26T23:22:11Z defaultxr joined #lisp
2014-09-26T23:22:23Z Bicyclidine quit (Ping timeout: 258 seconds)
2014-09-26T23:23:22Z nydel: super dimensional fortress social network, sdf.org
2014-09-26T23:23:45Z nydel: we're the freeshell.org and lonestar free unix shells thing
2014-09-26T23:23:58Z drmeiste_: faheem__1: I'll check in using my phone occasionally.
2014-09-26T23:24:00Z faheem__1: nydel: oh.
2014-09-26T23:24:06Z faheem__1: drmeiste_: ok
2014-09-26T23:25:07Z faheem__1: does anyone here use Stack Overflow to ask CL questions?
2014-09-26T23:25:34Z nydel: not anymore
2014-09-26T23:25:37Z schaueho quit (Ping timeout: 245 seconds)
2014-09-26T23:25:58Z faheem__1: nydel: running NetBSD, TOPS-20 and Symbolics GENERA
2014-09-26T23:26:04Z Shinmera quit (Quit: しつれいしなければならないんです。)
2014-09-26T23:26:14Z faheem__1: that's an unusual combination. i don't even know what the latter two are.
2014-09-26T23:26:18Z faheem__1: nydel: not anymore?
2014-09-26T23:26:49Z nydel: faheem__1: and planb from bell labs. gentoo somewhere. sdf.org is a cluster. the meta-array (ma.sdf.org) is a bit more stable and less all-over-the-place.
2014-09-26T23:26:50Z faheem__1: nydel: i got your ping. you want my current draft?
2014-09-26T23:26:57Z nydel: yes please
2014-09-26T23:27:06Z faheem__1: nydel: no debian?
2014-09-26T23:27:12Z faheem__1: too boring? :-)
2014-09-26T23:27:15Z faheem__1: nydel: ok, will do.
2014-09-26T23:27:50Z nydel: faheem__1: everyone hates ubuntu and if debian is mentioned ubuntu comes up, it's so stupid, easier to just not use it.
2014-09-26T23:28:03Z nydel: (that's kind of a joke, obviously that's not the only reason we don't use debian)
2014-09-26T23:28:30Z faheem__1: nydel: well, clearly not everyone hates ubuntu.
2014-09-26T23:28:35Z gendl joined #lisp
2014-09-26T23:28:46Z nydel: /s/everyone/everyone in the sdf inner circle/g
2014-09-26T23:28:52Z faheem__1: and debian is hardly to blame for ubuntu. if blame is to be assigned
2014-09-26T23:29:01Z faheem__1: nydel: i'm not too keen on ubuntu myself
2014-09-26T23:29:23Z nydel: i have to run it on my primary laptop because of the conditions of a grant provider who wanted me on macintosh
2014-09-26T23:29:32Z nydel: i'm literally on kubuntu 14 right now.
2014-09-26T23:30:01Z faheem__1: how many members does sdf have?
2014-09-26T23:30:08Z stepnem quit (Ping timeout: 272 seconds)
2014-09-26T23:30:25Z Bicyclidine joined #lisp
2014-09-26T23:31:26Z nydel: 561 registered on the meta-array (that's arpa users, contributors) and an uncountable number of shells on the cluster
2014-09-26T23:31:34Z stacksmith: faheem__1: where does one buy a Genera these days?
2014-09-26T23:31:58Z faheem__1: stacksmith: you're asking the wrong person
2014-09-26T23:32:07Z faheem__1: maybe you meant nydel?
2014-09-26T23:32:43Z stacksmith: faheem__1: probably.
2014-09-26T23:33:17Z nydel: nydel: it depends on how much one is willing to pay & in what currency
2014-09-26T23:33:27Z faheem__1: nydel: that's a lot of people. any security problems?
2014-09-26T23:33:30Z nydel: faheem__1: you probably meant stacksmith.
2014-09-26T23:33:38Z nydel: nydel: you got all of that wrong.
2014-09-26T23:33:50Z stacksmith: nydel: were you looking for me?
2014-09-26T23:33:53Z faheem__1: nydel: huh? no.
2014-09-26T23:34:12Z nydel: (poorly-execution joke, disregard)
2014-09-26T23:34:31Z nydel: stacksmith: yes, you actioned thinking about doing the hermit thing
2014-09-26T23:34:47Z nydel: stacksmith: i'd discourage that please
2014-09-26T23:34:59Z stacksmith: You'll miss me too much?
2014-09-26T23:35:04Z nydel: far far
2014-09-26T23:36:09Z nydel: i can't even read my feed collection today, why does every little exploit get a name nowadays?
2014-09-26T23:36:24Z nydel: and in some cases a damn visual logo
2014-09-26T23:36:56Z stacksmith: nydel: what kind of stuff do you do with sdf?
2014-09-26T23:37:42Z stacksmith: nydel: also, do you have a GENERA?
2014-09-26T23:37:46Z nydel: stacksmith: a lot of what i do there is just help people trying to use the cluster. on irc.sdf#helpdesk & elsehwere. i'm autistic and it's good for me to get outside of myself.
2014-09-26T23:38:12Z nydel: lots of new users or people interested in unix or gnu/linux land at sdf free shell accounts
2014-09-26T23:38:23Z faheem__1: nydel: do you get paid for this work?
2014-09-26T23:38:27Z stacksmith: nydel: I am leaning in the autistic direction myself.  SDF is great, I wish they had CL...
2014-09-26T23:38:31Z nydel: stacksmith: no. i do not have a genera myself.
2014-09-26T23:38:37Z hiyosi_ joined #lisp
2014-09-26T23:38:44Z nydel: stacksmith: ma.sdf.org:9903
2014-09-26T23:39:00Z nydel: we have CL ;)
2014-09-26T23:39:16Z nydel: paid. no. short answer.
2014-09-26T23:39:25Z nydel: i don't really need any more money.
2014-09-26T23:39:32Z faheem__1: nydel: ok
2014-09-26T23:40:18Z faheem__1: how does one know if one is autistic? i suppose there is a diagnosis? but how does know if it is reliable?
2014-09-26T23:40:20Z nydel: but seriously stacksmith point taken, i don't have as much pull on the cluster as i do on the meta-array, i'm trying to get emacs/slime available to all users
2014-09-26T23:41:03Z nydel: faheem__1: i don't know for sure, but i think autism is the first contageous psychological pathogen
2014-09-26T23:41:13Z faheem__1: it's a bit weird to hear that debian is tarred with ubuntu's brush. debian isn't that keen on ubuntu itself.
2014-09-26T23:41:22Z nydel: people who meet me then spend more time with me agree that they have become more autistic.
2014-09-26T23:41:25Z faheem__1: nydel: unlikely. :-)
2014-09-26T23:41:29Z nydel: (non-autistic people)
2014-09-26T23:41:53Z stacksmith: nydel: does sdf actually have common lisp for regular users?  I thought it was some weird dialect of lisp...
2014-09-26T23:41:53Z faheem__1: nydel: huh. which leads me back to the question of how one knows one is autistic
2014-09-26T23:42:19Z stacksmith: faheem__1: starting point: are you an introvert?
2014-09-26T23:42:33Z faheem__1: stacksmith: dunno. possibly.
2014-09-26T23:42:34Z nydel: pathogen as -- not biologically speaking. i think the autistic spectrum involves an idea/way-of-thinking, a mental faculty or set thereof.
2014-09-26T23:42:50Z faheem__1: i'm not very good with people. but it could have been circumstances/breeding.
2014-09-26T23:42:52Z nydel: faheem__1: i need to have an "idea alphabet" into which all input goes.
2014-09-26T23:43:07Z faheem__1: nydel: hmm
2014-09-26T23:43:14Z stacksmith: faheem__1: do interactions with people leave you exhausted and lead to completely different outcomes then expected?
2014-09-26T23:43:26Z hiyosi_ quit (Ping timeout: 250 seconds)
2014-09-26T23:43:28Z nydel: faheem__1: for me it is the tarot cards. as a set of 78 ideas. i have to put everything into thoughts of 1 thru 9 of these cards. in a tree of nodes. related by bloodlines.
2014-09-26T23:43:29Z faheem__1: you mean a categorizing system?
2014-09-26T23:43:37Z nydel: yes a categorization system.
2014-09-26T23:43:47Z nydel: non-lisp people respond better to "idea alphabet"
2014-09-26T23:43:51Z faheem__1: stacksmith: re interactions, exhausted etc, no, not at all.
2014-09-26T23:44:24Z nydel: stacksmith's questions are very good and his third is something about whether you offend people
2014-09-26T23:44:45Z stacksmith: nydel: beat me to the punch on that one.
2014-09-26T23:44:59Z faheem__1: i'm generally not expecially good at getting people to like me. i've never had very good social skills. but i like being around people if we have something in common.
2014-09-26T23:45:00Z nydel offers mildly-sarcastic high-five
2014-09-26T23:45:09Z nydel: ;)
2014-09-26T23:45:16Z faheem__1: nydel: offending? no, not really.
2014-09-26T23:45:36Z nydel: offend is wrong. (knight of swords -> five of pentacles)
2014-09-26T23:45:39Z faheem__1: i do find i'm not very good at responding to friendliness in kind, if that makes any sense
2014-09-26T23:45:52Z stacksmith: Make people uncomfortable without really trying...
2014-09-26T23:46:02Z nydel: something like..where the hell is cl-wordnet.. uh "do you make people uncomfortable unintentionally"
2014-09-26T23:46:07Z faheem__1: stacksmith: uncomfortable, no, not really
2014-09-26T23:46:08Z nydel: perhaps by rambling
2014-09-26T23:46:32Z meiji11 joined #lisp
2014-09-26T23:46:34Z faheem__1: i find people want to feel you regard them as "special". that's how people make friends.
2014-09-26T23:46:34Z nydel: not rambling!!!
2014-09-26T23:46:40Z nydel: god dammit lol
2014-09-26T23:46:50Z faheem__1: even if i like them, i'm not very good at communicating it.
2014-09-26T23:47:01Z nydel: "going on at lengths about a topic that interests you that you can't imagine others not being interested in"
2014-09-26T23:47:11Z faheem__1: nydel: no, not really.
2014-09-26T23:47:22Z faheem__1: i used to do that when younger, i think.
2014-09-26T23:47:25Z faheem__1: now, no.
2014-09-26T23:47:26Z nydel: page of coins, 10 of coins, magician, fool, knight of cups.
2014-09-26T23:47:44Z bbyler_tho quit (Remote host closed the connection)
2014-09-26T23:47:58Z faheem__1: nydel: ?
2014-09-26T23:48:03Z nydel: (was path from the connotative incorrectness of "ramble" to the above)
2014-09-26T23:48:03Z stacksmith: I've been known to babble about Lisp to unresponsive audiences.  Kids, etc.
2014-09-26T23:48:16Z didi left #lisp
2014-09-26T23:48:19Z faheem__1: stacksmith: that must be fun to watch
2014-09-26T23:48:37Z faheem__1: stacksmith: you mean in schools/
2014-09-26T23:48:39Z faheem__1: ?
2014-09-26T23:48:52Z stacksmith: nydel: what implementation of CL is available at SDF?
2014-09-26T23:49:03Z attila_lendvai quit (Quit: Leaving.)
2014-09-26T23:49:14Z nydel: ECL
2014-09-26T23:49:27Z nydel: i had nothing to do with that decision
2014-09-26T23:49:33Z faheem__1: i would have expected ccl / sbcl .
2014-09-26T23:49:34Z stacksmith: faheem__1: mostly my own kids at dinner.  Hardly an appropriate topic.
2014-09-26T23:49:37Z nydel: but on the meta-arra you can install whatever you want locally
2014-09-26T23:49:44Z faheem__1: stacksmith: :-)
2014-09-26T23:49:57Z nydel: that's why port 9903 of ma.sdf.org is actually nydel.sdf.org
2014-09-26T23:50:03Z faheem__1: if i had children, i can imagine boring them with all sorts of stuff
2014-09-26T23:50:44Z nydel: every day i have to stop myself from telling my lady-friend exactly what i did in lisp, i won't notice an hour has passed.
2014-09-26T23:51:00Z faheem__1: is excessive perfectionism an autistic attribute?
2014-09-26T23:51:16Z stacksmith: nydel: I really should get meta-arpa...  How do you like ECL?  I was just looking at SourceForge yesterday, was a bit curious... Bytecoded...
2014-09-26T23:51:42Z nydel: faheem__1: i would rather refrain from answering that immediately and instead prepare a response.
2014-09-26T23:51:57Z faheem__1: nydel: eh?
2014-09-26T23:52:09Z faheem__1: oh, i se, a joke
2014-09-26T23:52:15Z stacksmith: faheem__1: boring your children is one of the joys of parenthood.  It gets easier with time.
2014-09-26T23:52:18Z faheem__1: so, that's a yes, then
2014-09-26T23:52:21Z nydel: stacksmith: i haven't used ECL for more than a day or so. on ma.sdf i have installed emacs/slime with sbcl, and sbcl is daemonized running a webserver on 9903
2014-09-26T23:52:37Z faheem__1: stacksmith: true. then they grow up and patronize you.
2014-09-26T23:52:51Z nydel: yes yes! and i usually don't tell jokes because people barely ever get them (and probably shouldn't)
2014-09-26T23:53:17Z faheem__1: nydel: people don't get my jokes where i am. other places, yes, some of the time
2014-09-26T23:53:23Z hiyosi_ joined #lisp
2014-09-26T23:53:26Z nydel: stacksmith: that's going in my irc quotes file
2014-09-26T23:53:56Z faheem__1: stacksmith: you mean it gets easier to bore them? :-)
2014-09-26T23:54:29Z stacksmith: Well, it works both ways, as the children pay less attention to what you say.
2014-09-26T23:54:34Z nydel: faheem__1: i have trouble believing that anyone who uses common lisp to do work in statistics is not partially autistic. but then again i don't believe in autism. i think it's just a psychologically-transmitted ease.
2014-09-26T23:54:37Z faheem__1: stacksmith: ah
2014-09-26T23:54:57Z faheem__1: nydel: i don't actually use CL. but hope to, one day.
2014-09-26T23:55:07Z nydel: haskell?
2014-09-26T23:55:20Z snafuchs joined #lisp
2014-09-26T23:55:21Z faheem__1: i don't actually know CL, which makes it more difficult to use.
2014-09-26T23:55:32Z stacksmith: nydel: The spectrum description is better then just autism...
2014-09-26T23:55:35Z faheem__1: nydel: i'm mostly python. with occasional C++
2014-09-26T23:55:36Z freaksken quit (Ping timeout: 244 seconds)
2014-09-26T23:56:00Z faheem__1: stacksmith: i've heard of the autism spectrum.
2014-09-26T23:56:08Z nydel: i love python! and i don't like any non-cl languages really. but i use python to grab my port 9903 website.
2014-09-26T23:56:41Z nydel: faheem__1: there's no such "asperger's syndrome" anymore - the spectrum is psychologically fashionable - all the labels are thrown out now
2014-09-26T23:56:48Z faheem__1: nydel: python is a nice language. a bit crippleware compared to CL
2014-09-26T23:57:01Z faheem__1: nydel: hmm
2014-09-26T23:57:40Z nydel: faheem__1: when you do (think) in your mind's interpreter, what goes on?
2014-09-26T23:58:22Z faheem__1: nydel: that's an extremely broad question
2014-09-26T23:58:24Z stacksmith: Here is another test question: does the idea of 5 years of solitary confinement really scare you?  What is more scary - lack of companionship or lack of computing equipment? (Let's leave sex drive out for now).
2014-09-26T23:59:20Z faheem__1: stacksmith: hmm. 5 years of solitary confinement. yes, it sounds scary
2014-09-26T23:59:22Z jasom: stacksmith: I don't want to be tortured at all.
2014-09-26T23:59:24Z nydel: stacksmith: there's a scene in a sitcom where people are doing trivia at a nightclub & someone has their phone taken away, he has to leave and says "i can't.. i wanna be with my phone.. i.." that's how i feel about my laptop
2014-09-26T23:59:48Z faheem__1: stacksmith: companionship vs computing equipment. hmm. difficult one.