#clojure log - Feb 10 2013

The Joy of Clojure
Main Clojure site
Google Group
IRC
List of all logged dates

0:01 yedi: nvm

0:50 how would i get the whole request map in compojure? do i _have_ to destructure?

0:53 * nightfly doesn't see how destructuring could ever be required, for anything

0:58 yedi: nvm, i was misunderstanding the compojure api

1:19 loliveira: hi, do I insert multiple records using cloure.jdbc? clojure.java.jdbc/insert-records, but it generates one insert command per row. How to avoid that?

1:50 callenbot: I am having a bizarre problem with Monger

1:51 the goddamn find-map-by-id isn't working

1:51 yedi: callenbot: i decided on clabango v_v

1:51 it's just too familiar

1:51 callenbot: Raynes: VICTORY IS MINE

1:52 yedi: s'all good. Learn one new thing at a time.

1:52 yedi: you can be hipster-hardcore with your HTML later.

1:53 Raynes: yedi: You're dead to me.

1:53 clojurebot: Cool story bro.

1:53 Raynes: ;)

1:53 (It's okay actually)

2:21 callenbot: alright, I need to lockdown routes

2:21 sandbar and friend look like boondoggles.

2:21 What are my options?

2:26 ah fuck it I'll write a wrapper.

2:42 michaelr525: heu

2:42 hey

2:43 idea+laclojure is actually pretty nice..

2:48 callenbot: michaelr525: oh yeah? cool.

2:49 michaelr525: callenbot: yup.. are you related to la clojure somehow?

3:03 re

3:04 callenbot: did you write something when I was fighting with the train wifi? :)

3:29 callenbot: https://www.refheap.com/paste/11138 this is clearly a dysfunctional pattern. Any advice?

3:31 amalloy: callenbot: https://github.com/flatland/useful/blob/develop/src/flatland/useful/debug.clj#L26

3:34 callenbot: amalloy: I don't mean the print, I mean the dysfunctional arg passing and stuff.

3:34 amalloy: thanks though.

3:35 amalloy: the only thing that jumped out at me as dysfunctional was your printing

3:35 callenbot: amalloy: it's not proper mate.

3:35 (posts 51173c4a3004b7cebbd09ea9)

3:35 *** call handler ***

3:35 #<admin$edit neubite.routes.admin$edit@77749c47>

3:35 ((posts 51173c4a3004b7cebbd09ea9))

3:36 see how it's doubly-wrapped in a list?

3:36 it's because it gets kicked through an & args twice.

3:36 amalloy: see the problem now?

3:36 I need to splice the argument list without using a macro.

3:37 I feel like I've run into before and I'm missing something critical.

3:39 amalloy: you...just need to call (apply call-handler fn args)?

3:39 also, the if clause in call-handler can just go away

3:41 callenbot: amalloy: thanks for the pointer on getting rid of the if clause. I take it apply takes care of it.

3:42 hrm yes. This cleared up a few things for me.

3:42 amalloy: danke!

3:55 Raynes: callenbot: What nationality are you, sir?

3:55 If you don't mind my asking.

3:55 Your dankes confuse me.

4:05 callenbot: Raynes: disgruntled midwestern German-American (native born to the US).

7:48 svedubois: How I can convert this for statement in java to clojure? https://www.refheap.com/paste/11141

8:14 AimHere: svedubois, there's a lot of ways of doing it - you could do it with map/reduce/for/loop + recur. This one should return a vector with min/max in it, respectively: https://www.refheap.com/paste/11145

8:17 svedubois, modulo missing parentheses

8:22 svedubois: AimHere: I have changed (image/dimension d) to (.dimension img d), and I obtain: clojure.lang.ArityException: Wrong number of args (1) passed to: core$reduce

8:23 AimHere: I missed a parentheses at the end of the function

8:23 There should be one at the end of the line that starts (conj maxes ...

8:26 svedubois: AimHere: With https://www.refheap.com/paste/11147 I obtain java.lang.UnsupportedOperationException: nth not supported on this type: Long

8:28 AimHere: Isn't that because you put the closing parentheses on the wrong line. There should be one at the end of the line with (conj maxes on it

8:28 '[[][]]' isn't part of the function

8:31 tgoossens: Hi. From now on i want to leave sublime behind me for clojure development and learn to work with emacs (hurray!)

8:31 emacs or xemacs

8:31 ?

8:31 or doesn't it matter

8:31 pimeys: emacs24 package at least in ubuntu uses gtk and therefore x

8:33 AimHere: xemacs is a fork of emacs by some people who were pissed off about copyright attribution or something

8:35 tgoossens: whats a good resource to start clojure development with emacs

8:35 svedubois: AimHere: it works

8:43 deg: Is it a bug that clojure.pprint/cl-format does not format objects the same as java format?

8:43 ,(let [f (clojure.java.io/as-file "foo")] [(format "%s" f) (clojure.pprint/cl-format nil "~A" f)])

8:43 clojurebot: #<ExecutionException java.util.concurrent.ExecutionException: java.lang.RuntimeException: java.lang.ClassNotFoundException: clojure.pprint>

8:43 deg: (oops???)

8:44 Looks like the robot doesn't handle pprint, I guess because of the eval escapes it has.

8:44 Anyway, that expression returns ["foo" "#<File foo>"]

8:47 svedubois: AimHere: I would like to use the vector of mins and maxes to create and interval, like (FinalInterval. mins maxes)

8:47 But I obtain this error: java.lang.ClassCastException: clojure.lang.PersistentVector cannot be cast to [J

8:47 The value of mins and maxes are [[-890 -458] [889 457]]

9:07 mdeboard: Anyone here dealt with "JNI wrapper classes" in Clojure? I'm trying to use a 3d-party library's SDK (Callas's pdfEngine lib), which exposes its C API in Java via the PDFEngine JNI wrapper class. I've proxied the class but when I invoke I get unhelpful exceptions. I was hoping someone might have general words of advice about this topic.

9:08 pepijndevos: can core.loic do interesting things to sets? Like, contrain one set to e the suset of another and stuff like that.

9:08 mdeboard: It's actually not even critical I succeed, it was more an experimentation to see if it would be useful to write our own software using the SDK or continue shelling out to use the CLI provided

10:55 keugaerg: ,(do (println "test") (for [x '(1 2 3)] (prn x)))

10:55 clojurebot: test\n(1\n2\nnil 3\nnil nil)

10:55 keugaerg: Hi

10:56 Sorry if it seems a newbie question ( I didn't wrote a line of Clojure since last summer .. )

10:56 hyPiRion: keugaerg: hello, ask away, we won't bite :)

10:57 keugaerg: huh ... no ok sorry I just get it

10:57 No I was wondering why the output in the `for' was .. returned in the list

10:57 I am just understanding it is actually well printed out

10:58 but it is just mixed with the return value output ..

10:58 no it's ok I get it :-) Don't bother ! :-)

10:58 hyPiRion: thnx :)

10:59 hyPiRion: keugaerg: I didn't do anything, but you're welcome I suppose. It's because the for loop is lazy

11:00 abp`: hyPiRion: loop? :P

11:01 hyPiRion: abp`: dangit, I'm writing about while and for loops.

11:01 for comprehension

11:01 abp`: hyPiRion: I see

11:47 enquora: new to clojure ecosystem, trying to install https://github.com/fmw/vix results in lein deps error: Error: cannot resolve leiningen.cljsbuild/activate hook

11:48 can anyone explain error and solution?

11:52 ln22: I tried to get Vix working too.  Got an error during the user setup process.  Where are you stuck enquora?

11:53 enquora: I'm following the repo instructions. first time thru, starting the lein repl failed because of the hooks error

11:53 just ran again and the repl seems to be running though

11:53 dunno

11:54 ln22: Did you change directories to the vix folder?

11:55 enquora: yes

11:55 I'm in the repl now, making first user

11:55 will see if that works

11:57 ln22: If it does work let me know what you did!

11:57 I got an error during this step.

11:57 (add-user "vix" "my-username" "my-password" {:* [:GET :PUT :DELETE :POST]})

11:57 enquora: ln22: well, there's a user doc in couch now. seems to be working

11:57 ln22: The add-user function worked for you?

11:57 enquora: ln22: I ran lein deps again, several times

11:57 yes, indeed

11:58 ln22: Interesting.  Well at least I know its my system now.

11:58 enquora: oh, I also gave it my worst !@*()^& jvm dirty look. perhaps that put the fear into it

11:58 ln22: I'm going to wait a little bit anyways I'm really interested in the webshop capabilities.

11:58 Haha

12:00 enquora: I'm rewriting an ancient docs-sharing system. this isn't quite what I need, but I'm looking for an example of a real clojurescript application, and need to use both couch and elasticsearch on backend, so it seems like it's worth a look

12:01 ln22: Ah I see.

12:01 enquora: ln22: completely new to the clojure ecosystem, but not lisp - so this will be a bit of an adventure

12:01 ln22: I'm trying to learn Clojure at the moment (it's my first attempt at programming).  I really want to contribute to projects like Vix.

12:02 enquora: lein is both a dependancy and general project management tool, correct?

12:02 for task execution, etc?

12:02 what is Soy, and why doesn't lein handle installing this?

12:03 ln22: I have no idea what Soy is but from what I've seen Lein is for dependencies and Git is still a really popular tool for version handling ect.

12:03 Keep in mind that I've been a programmer for 8 days....

12:04 abp`: enquora: "neo.res.rr.com) has joined channel

12:04 #clojure [02:54]

12:04 *** ecmendenhall (~connormen@wsip-70-184-81-11.ph.ph.cox.net) has joined

12:04 channel #clojure

12:04 <Raynes> callenbot: hey [03:00]

12:04 <Raynes> callenbot: Walk off a balcony.

12:04 <callenbot> Raynes: bots can't walk [03:01]

12:04 <callenbot> Raynes: I'll take a flying leap with you in LA though :D

12:04 <Raynes> :p [03:03]

12:04 *** bkearns (~bkearns@75-101-54-23.dsl.static.sonic.net) has quit: Quit:

12:04 Leaving. [03:06]

12:04 *** h0bbit (~h0bbit@59.95.15.235) has joined channel #clojure

12:04 <amalloy> yedi: presumably Raynes's tutorial is assuming that you haven't

12:04 written non-p elements with class=meow [03:07]

12:04 *** tbaldridge (~tbaldridg@c-67-190-114-220.hsd1.co.comcast.net) has joined

12:04 channel #clojure

12:04 <Raynes> yedi: amalloy is correct. [03:08]

12:04 enquora: ln22: what's your usage case for something like vix?

12:04 abp`: <yedi> ok [03:10]

12:04 <cshell> If I have a vector that has alternating key and value (ie. [:a "a" :b

12:05 "b]) what would be the right function to use to update the :b key?

12:05 So I'd return [:a "a" :b "c"] [03:11]

12:05 <cshell> ?

12:05 *** tbaldridge (~tbaldridg@c-67-190-114-220.hsd1.co.comcast.net) has quit:

12:05 Ping timeout: 246 seconds [03:13]

12:05 *** hammer (~hammer@99-20-245-250.lightspeed.mssnks.sbcglobal.net) has joined

12:05 channel #clojure [03:14]

12:05 *** hammer (~hammer@99-20-245-250.lightspeed.mssnks.sbcglobal.net) has quit:

12:05 Client Quit

12:05 <callenbot> when does Clojure 1.5 go stable? [03:15]

12:05 <callenbot> I'd like to stop using my dissoc-in shim.

12:05 *** whamied (~whamied@24.30.125.73) has joined channel #clojure [03:17]

12:05 *** yogthos (~yogthos@li231-96.members.linode.com) is now known as

12:05 yogthos|away

12:05 *** craigbro (~craigbro@pool-64-222-126-127.burl.east.myfairpoint.net) has

12:05 quit: Ping timeout: 260 seconds

12:05 <amalloy> cshell: use a better data structure than that :P

12:05 *** loganlinn (~Adium@c-98-210-140-30.hsd1.ca.comcast.net) has joined channel

12:05 #clojure [03:18]

12:05 <cshell> amalloy: that's actually a good idea

12:06 <cshell> amalloy: I must be tired

12:06 <cshell> amalloy: thanks :)

12:06 *** francisl (~anonymous@69.157.143.178) has joined channel #clojure [03:19]

12:06 *** bosie (~bosie@91-119-78-115.dynamic.xdsl-line.inode.at) has quit: Ping

12:06 timeout: 245 seconds

12:06 *** bmaddy (~bmaddy@72.21.231.26) has joined channel #clojure [03:21]

12:06 *** hive-mind (pranq@unaffiliated/contempt) has joined channel #clojure

12:06 *** d2dchat (~d2dchat@c-68-61-5-136.hsd1.mi.comcast.net) has joined channel

12:06 #clojure [03:22]

12:06 *** hughfdjackson (~hughfdjac@unaffiliated/hughfdjackson) has quit: Quit: Lost

12:06 terminal [03:24]

12:06 *** ahokaomaeha (~suckmy@unaffiliated/dijkstragroupie) has quit: Quit: When I

12:06 come back, please tell me in what new ways you have decided to be

12:06 completely wrong. [03:25]

12:06 *** S11001001 (~sirian@fsf/member/S11001001) has joined channel #clojure

12:06 *** ahokaomaeha (~suckmy@unaffiliated/dijkstragroupie) has joined channel

12:06 #clojure

12:06 *** normanrichards (~normanric@70.114.215.220) has quit: [03:28]

12:06 keugaerg: abp`: I think you're flooding the channel ..

12:06 abp`: *** h0bbit (~h0bbit@59.95.15.235) has quit: Quit: Computer has gone to sleep.

12:06 *** ecmendenhall (~connormen@wsip-70-184-81-11.ph.ph.cox.net) has quit: Quit:

12:07 ecmendenhall

12:07 *** infynyxx (~infynyxx@cpe-24-90-84-91.nyc.res.rr.com) has joined channel

12:07 #clojure [03:29]

12:07 *** sritchie (~sritchie@c-67-180-210-142.hsd1.ca.comcast.net) has quit: Quit:

12:07 sritchie [03:31]

12:07 <gfredericks> clojure.java.jdbc/with-connection is deprecated? [03:33]

12:07 *** freakazoid (~seanl@c-67-164-106-36.hsd1.ca.comcast.net) has joined channel

12:07 #clojure

12:07 S11001001: abp`: please kill your client instead of waiting for the paste to finish

12:07 abp`: *** looper (~looper@p5B323F93.dip.t-dialin.net) has quit: Ping timeout: 245

12:07 seconds [03:34]

12:07 * gfredericks huhs to himself

12:07 <amalloy> callenbot: your statement seems to imply that clojure 1.5 contains

12:08 abp: S11001001: It was still pasting?

12:08 keugaerg: It stoped when you disconnected

12:08 S11001001: abp: yeah, it buffered, ironically to avoid being killed for flood. Good now

12:08 enquora: ln22: just noticed that the vix readme instructions assume use of a debian OS

12:08 abp: S11001001: Didnt't even see that.. keugaerg: Yeah when S11001001 said it was still pasting..

12:09 S11001001: abp: your client probably showed the messages being all written right away

12:09 abp: Ok, sorry again. Screwed me with emacs keychords. Still a noob..

12:10 ln22: I'm using Ubuntu 12.10.  That's probably why it failed.

12:10 abp: S11001001: Some but not all.

12:14 enquora: ln22: ubuntu 12.10 should be just fine

12:15 Soy is a templating system. have another failure following instructions to install it and compile templates

12:15 ln22: In theory.  I'm going to put Arch on this system Monday and try Vix again.  I'm sure it will work.

12:15 A failure involving Vix or Soy?

12:15 enquora: Soy

12:15 this line: java -jar SoyToJsSrcCompiler.jar --shouldProvideRequireSoyNamespaces --shouldGenerateJsdoc --outputPathFormat resources/public/js/soy/{INPUT_FILE_NAME_NO_EXT}.soy.js soy/editor.soy soy/feed.soy

12:16 it's complaining about what looks to be an invalid template

12:23 ln22: did manage to at least launch the server and see web pages. Lots of warnings and errors though, and I don't see any functionality when I use the admin login. Need to browse the code a bit to see what it is supposed to do, and probably ping the author

12:31 ln22: Interesting.  Like I said I will probably be waiting until either there are WebShop capabilities or until I'm capable of actually contributing (which me be awhile.  Haha).

12:31 *might

12:41 enquora: ln22: just sent an email to the author. will browse the code a bit now, but mostly that's it for the day.

12:42 interesting that you should pick clojure as a first programming language, if I understood you correctly

12:42 ln22: I did.  It has been interesting... xD

12:44 enquora: I don't disagree with the choice, but that wouldn't be the conventional wisdom

12:44 but then, there were far fewer languages available to choose from when I learned ;-)

12:44 ln22: Lets hope that being unconventional pays off.  Haha

12:45 enquora: Well, you're less likely to learn bad habits this way!

12:45 ln22: I'm starting to see that now.  Like never using loops for example.

12:45 enquora: and it certainly is more productive than many other languages, once you get the hang of it

12:46 one of the interesting things about programming is that naming things (consistently and clearly) is one of the hardest things in programming

12:46 In procedural/imperative programming, that's a problem

12:47 here, not so much, because so many intermediate states that are exposed in imperative programming, and thus need to be named, don't exist here

12:48 ln22: Hopefully I never even have to worry about those things!  Haha

12:49 Hodapp: enquora: I've noticed this too.

12:49 enquora: so much of it is bookkeeping for ephemeral state.

12:49 ln22: One thing that I am really struggling with is the lack of "beginner" books / courses for Clojure.

12:49 There are books but none are truly "for dummies".

12:49 enquora: don't know the answer to that, unfortunately

12:50 the community here is active - that helps

12:51 ln22: The community has helped me massively!

12:51 enquora: I've been looking for examples of non-trivial web apps using both clojurescript and clojure, and haven't had much success

12:51 ln22: I am probably the least skilled programmer in the entire channel too.  xD

12:52 arrdem: enquora: the noir sample blog is out of date (noir is depricated) but it's a decent example

12:53 * arrdem goes to see if his blog's codebase is in a publishable state

12:53 enquora: arrdem: of clojurescript combined with clojure on the server?

12:53 arrdem: enquora: clojure/ring server...

12:53 no clojurescript tho

12:54 enquora: that's were my problem domain is - complex web-browser apps that interact with the server

12:54 a distributed system

12:54 have fully discounted node.js, but the sort of thing the meteor project appears to be attempting

12:55 will have a look, though

12:55 thks

13:22 gfredericks: somehow I've mostly missed the fact that defn and defmacro have explicit support for adding metadata via a map

13:36 TimMc: In several different places!

13:44 gfredericks: TimMc: yeah that part quite surprised me

14:12 Sgeo: I just realized that in my blog post I __________HAVE__________ to talk about flatten

14:12 flatten?

14:12 clojurebot: flatten is rarely the right answer. Suppose you need to use a list as your "base type", for example. Usually you only want to flatten a single level, and in that case you're better off with concat. Or, better still, use mapcat to produce a sequence that's shaped right to begin with.

14:14 gfredericks: (def unflatten #(map vector %))

14:14 Sgeo: flatten is the devil

14:14 It doesn't exist in Haskell, and for good reason

14:15 gfredericks: because it can't be typed?

14:16 Sgeo: Pretty much, yeah. Although, I guess you could make a type that you could define a flatten operation on

14:16 * Sgeo would need to think about it a bit

14:16 gfredericks: I don't think this could work

14:16 I think you can only flatten a fixed number of levels

14:17 so you'd have a flatten1, flatten2, flatten3, etc

14:17 Sgeo: data Thing = One String | Many [Thing]

14:18 flatten :: [Thing] -> [Thing

14:18 erm

14:18 flatten :: [Thing] -> [Thing]

14:18 flatten' :: Thing -> Thing

14:18 gfredericks: I think I might be going crazy.

14:19 yeah totes crazy

14:19 Sgeo: flatten' (One string) = One string

14:19 gfredericks: I have only regrets.

14:19 Sgeo: flatten' ... hmm

14:19 Wait, I don't know if that's actually working

14:19 * Sgeo would need to think about it some more

14:21 gfredericks: I think it does

14:22 I was confusing the fact that you can't do runtime polymorphism with typeclasses with the fact that you _can_ do runtime polymorphism with case classes

14:30 ticking: I wonder why regex don't implement the fn protocol to do a re-matches by default.

14:40 atyz: hey guys, a little confused by an error i'm getting when typing "lein test" - could someone suggest what i'm doing wrong:

14:40 Exception in thread "main" java.io.FileNotFoundException: Could not locate rainbowtickets/test/handler__init.class or rainbowtickets/test/handler.clj on classpath:

14:41 deg: Is it a bug that clojure.pprint/cl-format does not format objects the same as java format?

14:41 lynaghk: atyz: it sounds like the namespace/filename don't match up

14:41 deg: (let [f (clojure.java.io/as-file "foo")] [(format "%s" f) (clojure.pprint/cl-format nil "~A" f)])

14:41 atyz: ok great - i'll look into it

14:41 thank you

14:41 lynaghk: atyz: no worries.

14:41 deg: returns ["foo" "#<File foo>"]

14:41 Sun 15:45: *** snake-john JOIN

14:41 Sun 15:46: *** loliveira QUIT Quit: loliveira

14:41 Sun 15:47: *** Snaffu JOIN

14:42 (oops, excuse cut-n-paste gubbish)

14:42 ticking: atyz: your folder/file hierarhy needs to match up the napespace hierarhy same should go for tests

14:42 atyz: ticking ahh that makes sense. i thought it would be able to infer it from the namespace

14:43 loliveira: how do I make a multi row insert using clojure.java.jdbc?

14:56 raxelo: Hi all, I am playing with seesaw and for some reason exceptions that happen don't fall into my REPL

15:01 thalassios_xelon: hello room :))

15:04 in a clojure program the garbage collector has much more work to do than in standar java program?

15:04 i get a gc overhead limit ... error

15:06 raxelo: thalassios_xelon, how this error looks like ?

15:09 thalassios_xelon: i have a toy program

15:10 raxelo: ok, but the error, stacktrace itself.

15:10 how does it look like ?

15:10 thalassios_xelon: ok wait a minute

15:12 is it true that in clojure we need the garbage collector much much more

15:12 ?

15:13 bawr: as usual with such questions, it depends ;)

15:14 amalloy: thalassios_xelon: generally yes, but gc overhead errors are much more severe than that

15:14 raxelo: thalassios_xelon, I think it is more yes than no =) due to immutability in the core

15:15 amalloy: they basically mean "you have a memory leak, but it's slow"

15:15 clojurebot: what the hell, just crash it

15:15 atyz: is there a way to run a specific test only with lein test?

15:15 amalloy: (inc clojurebot)

15:15 lazybot: ⇒ 19

15:16 [1]thalassios_xe: OutOfMemoryError GC overhead limit exceeded

15:16 clojure.lang.RT.cons (RT.java:559)

15:16 clojure.core/cons (core.clj:29)

15:16 clojure.core/list* (core.clj:590)

15:16 clojure.core/apply (core.clj:603)

15:16 dxeh: "Sets are collections of unique values. (= __ (clojure.set/union #{:a :b :c} #{:b :c :d}))" how can i fill in this blank and can someone explain how they got the answer? Trying to pick up clojure but im not familiar with any lisps :p

15:16 [1]thalassios_xe: clojure.core/partial/fn--4070 (core.clj:2343)

15:16 testing.parse/parse-filter (parse.clj:25)

15:16 testing.parse/parse-filter (parse.clj:25)

15:16 testing.parse/parse-filter (parse.clj:25)

15:16 testing.parse/parse-filter (parse.clj:25)

15:17 xeqi: atyz: lein test :only some.namepace/test-name

15:17 atyz: oh my thank you

15:17 ivan: dxeh: https://en.wikipedia.org/wiki/Set_union

15:17 [1]thalassios_xe: dxen you put in __ the value to make the (= .. ..) true

15:18 dxeh: [1]thalassios_xe, yes i know lol im not stupid, i just dont know clojure well enough to make the expression evaluate to true

15:19 ivan: much easier than you are imagining

15:19 [1]thalassios_xe: you put the value down in the answers box

15:20 and press run...

15:21 dxeh: [1]thalassios_xe, do you not understand that i just said i _don't_ know clojure enough to even make it work and thats why i am asking how it could be solved

15:22 ivan: &(clojure.set/union #{:a :b :c} #{:b :c :d})

15:22 dxeh: and ivan thanks i will read

15:22 lazybot: ⇒ #{:a :c :b :d}

15:22 WuHoUnited: #{:a :b :c} is a set containing :a, :b and :c #{:b :c :d} is the set containing :b :c :d. You need to write down the set containing :a :b :c :d

15:24 raxelo: this community is very patient, I'll stay here )

15:24 Pure_Loulou: (= #{:a :b :c :d} (clojure.set/union #{:a :b :c} #{:b :c :d})) thats true

15:26 do you know what arguments to put to vm arguments in eclipse,so i dont get memory errors?

15:27 i get OutOfMemoryError GC overhead limit exceeded

15:29 raxelo: Pure_Loulou, -XX:-UseGCOverheadLimit

15:29 but this won't solve your problem in the root as you understand

15:29 amalloy: Pure_Loulou: write programs without memory leaks

15:29 raxelo: mentioned option for java 6th version

15:30 Pure_Loulou: what do you mean memory leaks?

15:30 in clojure?

15:31 dxeh: Pure_Loulou, he's basically implying your code has a bug which needs fixing (its allocating more memory than its deallocating)

15:32 _severe_ bug

15:32 ordnungswidrig: it's about eclipse oom'ing!

15:32 Pure_Loulou: -Xmx2G

15:33 raxelo: ordnungswidrig, I am not sure Eclipse usually launches application in separate process if it is not about REPL

15:33 amalloy: ordnungswidrig: -Xmx∞G for a long-term fix

15:35 ordnungswidrig: raxelo: it launches in a separate vm

15:35 raxelo: ordnungswidrig, right - separate vm/process

15:36 dxeh: lol infinity

15:36 raxelo: I just mean that it is not related to eclipse configuration itself but configuration of this particular run-configuration

15:36 ordnungswidrig: I understand Pure_Loulou that eclipse is oo'ing

15:36 but seems I'm wrong

15:36 Pure_Loulou: dxeh can someone program in clojure and control memory deallocating?

15:37 ordnungswidrig: Pure_Loulou: you can delay deallocation

15:37 ivan: amalloy: funny how that actually works

15:38 dxeh: Pure_Loulou, probably not but that doesnt mean you cant stop memory leaks

15:38 amalloy: does it?

15:38 Pure_Loulou: dxeh are there memory leaks in clojure?

15:38 ivan: -XX:+PrintFlagsFinal indicates that it sets MaxHeapSize to 0

15:38 * ordnungswidrig thinks Pure_Loulou is a bot

15:38 dxeh: no there are memory leaks in your code

15:39 Pure_Loulou: my code is in clojure not c,c++

15:39 dxeh: ok

15:39 metellus: Pure_Loulou: your problem is that the system is spending so much time trying to do garbage collection that it can't do other stuff

15:39 brehaut: http://blog.ezyang.com/2011/05/space-leak-zoo/

15:39 Pure_Loulou: i know thats my problem....

15:39 dxeh: Pure_Loulou, http://en.wikipedia.org/wiki/Memory_leak

15:40 metellus: which means that you're creating so many objects that something bad is probably happening

15:40 dxeh: read up on what a memory leak is maybe that will help you understand

15:40 Pure_Loulou: ok thx guys :) good night i am newbie

15:40 brehaut: Pure_Loulou, dxeh: i like the haskell peoples definition of a memory leak as a specific type of space leak

15:40 raxelo: Pure_Loulou, maybe you can share your code somewhere, if it is possible to glance what can be the problem

15:41 dxeh: brehaut, i dislike seeing any definition of memory leak anywhere in my code :) annoying to fix xd

15:43 &(hash-map :a 10, :b 20, :c 30) :b)

15:43 lazybot: ⇒ {:a 10, :c 30, :b 20}

15:44 ivan: &((hash-map :a 10, :b 20, :c 30) :b)

15:44 lazybot: ⇒ 20

15:44 dxeh: o

15:44 lol

15:44 metellus: &(+ 1 2) lazybot doesn't care about the rest of the line?

15:44 lazybot: ⇒ 3

15:44 dxeh: ivan, why 20

15:45 ivan: that's the value for the :b key

15:45 dxeh: because :b is 20 in the map and it wants the value of :b

15:45 yeah ok

15:45 thalassios_xelon: dexh that memory leak you said,comfused me,when you program in clojure you take care for "memory leaks"?

15:45 dxeh: thalassios_xelon, to be blunt if you dont write bad code you dont get memory leaks

15:45 just be careful

15:46 the problem is not with clojure, but with something you've written and you have to now go back and debug to find out what the problem is :p

15:46 thalassios_xelon: maybe i wrote bad code that consumes much memory...

15:46 amalloy: if you don't write bad code you're also a heffalump

15:46 everyone writes bad code

15:47 thalassios_xelon: it is my first clojure program...

15:47 metellus: what makes you think it has a memory leak?

15:47 dxeh: oom exception sums it up

15:48 raxelo: thalassios_xelon, maybe you use 3rd party library that may leak too

15:48 thalassios_xelon: i dont think it has memory leak,i think that it allocates memorie faster than it is release it

15:48 dxeh: ... thats what a memory leak is

15:48 WuHoUnited: can you show us the program?

15:48 metellus: oh, the nick change threw me off

15:48 thalassios_xelon: WuHoUnited, its copyright:P

15:48 its a toy program

15:49 anyway i will read more thx guys goodnight :)

16:02 Frozenlo`: Is there an emacs add-on to get the docs from clojuredocs?

16:08 ivan: typing " inside a string in paredit -> conversion experience

16:08 Frozenlock: ivan: does it escape it automagically?

16:08 ivan: yep. and when you backspace, it deletes both \ and "

16:09 bawr: ivan: *_*

16:09 ivan: Frozenlock: I use cdoc from (require '[cd-client.core :refer [cdoc cdir]])

16:09 Frozenlock: your REPL may already have it

16:18 Frozenlock: ivan: looks promising! I'll add it to my leiningen profile, thanks!

16:27 raxelo: Frozenlock, I am playing with seesaw now and using https://github.com/franks42/clj-ns-browser intensively

16:27 it is out of emacs but good for investigation

16:31 ivan: I would use that more if it were not for https://github.com/franks42/clj-ns-browser/issues/51

16:31 I suppose I should just fix that

16:35 Frozenlock: Is there a way to make my repl refer a namespace everytime I switch to a new one?

16:36 In this case I would like the clojuredocs-client to be always available.

16:36 ivan: when I asked a month ago, the best suggestion was to bind some Emacs key to insert the right symbol, e.g. cd-client.core/cdoc

16:37 raxelo: ivan, I see... it is not relevant for me so far, linux user

16:37 ivan: there was also talk of a weird REPLy feature that expands just one symbol, no matter what namespace you're in

16:38 Frozenlock: Well the `repl' namespace seems to always be available (the `doc' macro)

16:42 ivan: in a vanilla nREPL, clojure.core> (doc +)

16:42 CompilerException java.lang.RuntimeException: Unable to resolve symbol: doc in this context, compiling:(NO_SOURCE_PATH:1:1)

16:42 I've lost track of all the extra stuff lein and reply do

16:45 http://dev.clojure.org/jira/browse/NREPL-31 ah, maybe because you're on Clojure 1.4?

16:45 anyway, the helpful behavior you are experiencing is a bug

16:46 Frozenlock: Isn't it a feature? :P

16:46 ivan: until you paste in some code where `doc` is referring to something else

16:47 black_joe: I have consulted documentation and can't find anything. How do you apply a variable function to some parameters?

16:47 https://www.refheap.com/paste/fc8e3586eb7977e86bedb71af Like this.

16:47 Here I am trying to use spit as a default function, but allowing a new function in the &rest

16:48 amalloy: black_joe: that looks like the right way to do it

16:48 black_joe: I thought so too, but it throws a NullPointerException.

16:49 amalloy: black_joe: you're coming from...what, common lisp?

16:49 () is not nil

16:49 ivan: (rest list) when rest is a function argument?

16:49 amalloy: your &rest argument is not (), it is nil

16:50 black_joe: Isn't it both, since the variable arguements in &rest is a list?

16:50 amalloy: nothing can ever be both () and nil. they are different things

16:50 ivan's point is good as well; if that hasn't caused problems yet, it will

16:50 black_joe: Okay, I'll change that. Nil is more readable anyway.

16:50 amalloy: black_joe: no, change it to something better

16:51 (if (seq list) (...list is not empty...) (...list is empty...))

16:51 clojure-flavored nil punning

16:52 black_joe: Okay. I had seen seq before, but never knew it was a predicate. They usually have ?'s.

16:52 dnolen: black_joe: seq is not a predicate

16:52 black_joe: it will however return nil on empty list making it useful as a predicate since nil is false-y

16:52 abp: ,(seq '())

16:52 clojurebot: nil

16:52 abp: ,(seq nil)

16:52 clojurebot: nil

16:53 abp: ,(seq '(1 2 3))

16:53 clojurebot: (1 2 3)

16:53 black_joe: So it's kind of like using "touch" to create a file.

16:53 WuHoUnited: also, it appears you're mapping something across lisp, so you could use map

16:53 abp: (doc seq)

16:53 clojurebot: "([coll]); Returns a seq on the collection. If the collection is empty, returns nil. (seq nil) returns nil. seq also works on Strings, native Java arrays (of reference types) and any objects that implement Iterable."

16:54 black_joe: I was originally going to use that, yeah, but I was getting an error I couldn't correct. This works for my intended purposes, but I was just trying to add variable functions.

16:54 For now I'll just leave that out.

16:55 WuHoUnited: yeah, i guess you're not done with it anyway, since you aren't using rest as a vararg anyway

16:57 abp: black_joe: Probably you want to have a look at: https://github.com/bbatsov/clojure-style-guide

16:58 black_joe: Thank you! I've been kind of melding together C++ and Lisp standards for Clojure.

16:59 abp: black_joe: np, hope it helps you

17:12 VladimirFromUa: have a good day/night to all

17:40 tgoossens: wow paredit is great

17:40 abp: tgoossens: Yea. Emacs?

17:44 tgoossens: abp: first time i'm using emacs

17:44 still screwing everything up after a while

17:44 its not that important (but still)

17:44 i'm trying to figure out how i can get clojure syntaxhighlighting

17:44 the same way as in sublimetext2

17:45 hyPiRion: M-x package-list-packages

17:45 Oh wait, you need marmalade.

17:45 tgoossens: i have

17:45 that

17:45 abp: tgoossens: Hrhr, I'm using Emacs for a week now. Just pasted part of the irc log due to a awfully wrong gone keychrod in ERC this afternoon. :D

17:45 tgoossens: abp: cheers!

17:46 abp: tgoossens: I just grabbed emacs live after reading the tutorial and looking through a few .emacs.d repos on github.

17:46 tgoossens: emacs live?

17:47 abp: tgoossens: Sam Aarons Emacs-package bundle. https://github.com/overtone/emacs-live

17:48 tgoossens: Look through the key-rebindings if you use that..

17:49 tgoossens: looks flashy :p

17:49 abp: tgoossens: But despite those it's pretty great. You could even disable the rebindings, but they are ok for me.

17:49 tgoossens: hmm

17:50 abp: tgoossens: Probably, but quite some cool packages in there and I don't need to configure anything. :x

17:50 tgoossens: abp: i'll take a look at it. first i'm going to play a bit

17:51 i don't want to make the same error as with sublime

17:51 ivan: heh, -noverify is really good at reducing startup times from tens of seconds to seconds

17:51 abp: tgoossens: Sure, but if you don't install that you want to get ace-jump mode for sure.

17:51 tgoossens: three keystrokes to every symbol on screen.

17:52 tgoossens: ST got slow through plugins?

17:52 tgoossens: i had been using it for a few months. and I never really explored its possibilities etc. Now I really want an editor I understand, control completely

17:52 so it has been my own laziness

17:52 also

17:53 i never had some real projects with clojure

17:53 so it didn't really matter

17:53 but a month ago. i started a small projects (strips planning) in clojure and i felt the pain badly

17:53 abp: tgoossens: Oh ok. Yeah I've been reading through Emacs modes and packages relentlessly.

17:54 tgoossens: productivity nihil (compared to what it could be)

17:54 abp: tgoossens: I was using CCW, but Emacs just blows it out of the water.

17:54 tgoossens: and after reading the chapter "power editing" in "the pragmatic programmer". I realized I really had to do something about it

17:55 abp: but it's hard (perhaps even boring) to just read about a tool if you are not using it a lot

17:55 brehaut: abp: that depends a little on how much javaland you are doing

17:55 amalloy: agreed, eclipse blows emacs out of the water for java

17:56 abp: tgoossens: I did the tutorial and started reading through modes etc. Then picked up Emacs Live and got going, reading on about installed modes.

17:57 tgoossens: abp: i might just do the same

17:57 abp: brehaut: amalloy: Sure. For Java I used NetBeans. So CCW was a real painspot. Last time I used Eclipse was for Android dev quite some time ago..

17:58 No one cares about doing Java in Emacs?

17:58 callenbot: abp: java programmers aren't people

17:59 abp: callenbot: What if they evolve?

18:00 callenbot: I were doing Java mostly, while stumbling through Scala to Clojure. :P

18:01 powr-toc: Does anyone know how project works in core.logic?

18:14 Frozenlock: I must use a datastructure for a keyword and an integer. Is there some obvious advantages to either use maps or vectors? [[:some-key 10][:some-other-key 20]] vs [{:some-key 10} {:some-other-key 20}].

18:15 danlarkin: this is an impossible question to answer in the abstract

18:26 callenbot: danlarkin: dude, I beg of you.

18:26 danlarkin: hack on clabango more. I use it a ton and love it.

18:27 danlarkin: do you still use it?

18:27 danlarkin: I do

18:27 callenbot: danlarkin: thank the gods

18:27 danlarkin: but it does everything I need right now

18:27 patches welcome!

18:27 callenbot: danlarkin: so you're open to PRs? Sweet.

18:27 danlarkin: yeah just follow the style I've establish as best you can and I'd be very happy to accept PRs

18:28 s/establish/established

18:31 I'm pleased you like it

18:31 powr-toc: Frozenlock, The main trade offs are probably in readability and extendability... maps aren't positional, vectors are... so if you're sure there will always be two values then pick a vector if you favour brevity... otherwise a hash-map

18:32 callenbot: danlarkin: it's saved me and my frontend guy's productivity in Clojure

18:33 danlarkin: we're extremely fond of it. Except for pathing and some form helper macros, we were able to 1:1 port old templates from Flask over

18:34 danlarkin: cool

18:34 callenbot: danlarkin: is there a way to perform a get on a map in a {{ }}?

18:34 danlarkin: I've got a config map I need to kick down and the . syntax isn't working.

18:35 danlarkin: {{foo.bar}} should do it

18:35 callenbot: I don't think it works if you treat it like a map

18:35 I think I have to pass it the keyword like a function arg

18:35 it's carica's config magic.

18:36 I basically want to do: {{(config :stripe_pk)}}

18:36 yeah it's an fn.

18:36 danlarkin: carica.core/config is not a map

18:36 callenbot: yeah it's an fn

18:36 can I call a function in {{ }}

18:36 ?

18:36 danlarkin: nope

18:36 callenbot: also registering template filters has never worked for me.

18:37 it ends up saying the name of the template filter doesn't exist.

18:37 danlarkin: that's odd, it's worked for me

18:37 callenbot: danlarkin: I figured it had. I'll see if I can snag a repro for you.

18:45 hiredman: powr-toc: project is not something you want to use

18:45 * devth thinks it'd be sweet if `get-in` could take regexex in addition to keywords in `ks`

18:48 amalloy: devth: it can take any object at all

18:48 hyPiRion: ,(get-in {#"foo" :bar} [#"foo"])

18:48 clojurebot: nil

18:49 hyPiRion: ,(= #"foo" #"foo")

18:49 clojurebot: false

18:49 hyPiRion: Ah!

18:49 amalloy: &(let [k #"foo"] (get-in {k :bar} [k]))

18:49 lazybot: ⇒ :bar

18:49 devth: amalloy: right, i meant something like: (get-in {:foo {"plain iso/12345" "this is the body"}} [:foo #"plain"]) => "this is the body"

18:50 ivan: what would [:foo #"plain" :child] do if there are multiple `plain` matches?

18:51 amalloy: i figured. that's really a disastrous feature that goes against what maps are for

18:52 devth: i supposed that too. i guess it deserves a custom fn

18:53 specifically, i want to parse out the "TEXT/PLAIN" from something like: {:from #<IMAPAddress foo@gmail.com>, :subject "Re: A friendly message", :sender #<IMAPAddress foo.com>, :multipart? true, :content-type "multipart/ALTERNATIVE; boundary=bcaec5524106840d9e04d4ee1246", :body [{"TEXT/PLAIN; charset=ISO-8859-1" "Lol.\r\n\r\n\r\nOn"} {"TEXT/HTML; charset=ISO-8859-1" "<div dir=\"ltr\">Lol.</div>"}]}

18:54 ivan: take the first is what i need, but good question

18:54 ivan: you may find that returning unpredictable results, given that maps lack order

18:55 devth: ivan: i'd never expect it to contain multiple keys with "TEXT/PLAIN"

18:59 gross https://gist.github.com/devth/4751606

19:04 ivan: perhaps (val (first (filter #(re-find #"e" (key %)) your-map)))

19:14 devth: ivan: better. let me try that

19:29 yedi: callenbot: does clabango not support else?

19:29 java.lang.Exception: unknown tag: {:type :tag, :body {:token " else ", :offset 10, :line 15, :file #<URL file:/Users/yedianyansi/Documents/dev/skateproj/resources/templates/dress-page.html>}}

19:33 for some reason git add . is taking forever (i have less than 10 changed files), does anyone know what might be happening?

19:33 nvm

19:35 ivan: what was it? was going to suggest dtruss -f git add . or similar

19:37 callenbot: yedi: no, it doesn't.

19:37 danlarkin: why doesn't clabango support else?

19:37 yedi: I've already factored it out of my templates (else, that is)

19:38 Raynes: Support it, or else!

19:40 callenbot: Raynes: you can't say that, there is no else.

19:56 tomoj: hmm, there's no unmunge?

19:56 guess I don't need it

20:00 Frozenlock: Two java methods with the same name, in the same library... but they don't take the same arguments because they are not in the same object. This is madness!

20:01 amalloy: tomoj: by any sensible definition of munge (eg http://en.wikipedia.org/wiki/Mung_(computer_term)), an unmunge operation cannot exist :P

20:01 Frozenlock: (conj [] 1) works and (conj {} 1) doesn't. madness? no; {} just requires different arguments for its conj

20:04 Frozenlock: MADNESS!!

20:05 amalloy: but in this case I can at least understand: conj requires an element that composes the target.

20:05 hiredman: https://blogs.oracle.com/jrose/entry/symbolic_freedom_in_the_vm

20:05 Frozenlock: [:a 1] would be an 'element' of a map.

20:05 bbloom: Frozenlock: I think it's called an "Entry"

20:08 amalloy: hiredman: i'm getting a connection timeout on that page, sadly

20:08 hiredman: huh, I am looking at it right now

20:09 tmciver: amalloy: that page worked for me.

20:09 amalloy: well, close the tab! you must be consuming all their server resources, you greedy jerk

20:09 seriously though, weird. i consistently get a timeout. wonder why, if it works for everyone else

20:26 tomoj: amalloy: was the namer of clojure.core/munge using a sensible definition?

20:26 I guess the problem is obvious

20:26 amalloy: tomoj: i think they've done what that wikipedia article describes. clojure's name munger is certainly a one-way operation

20:28 tomoj: does it work if you restrict the domain of unmunge to things returned by munge, and the domain of munge to reasonable namespace names?

20:30 hmm, is c.c/munge even used for namespace name -> filename?

20:31 no, it's just https://www.refheap.com/paste/57902e60c69e10e805d8e3240

20:43 ivan: Raynes: "Older" on https://www.refheap.com/pastes?page=2 links to the same page

20:43 Raynes: ivan: I know.

20:43 Thanks for letting me know though.

20:43 ivan: np!

20:43 Raynes: I'm in the middle of moving to laser for the templates and I need to finish that before I can fix the bug.

20:44 I mean, I could hotpatch the existing stuff, but given that you're the first person to notice it in at least 3 months of being broken… :

20:44 :P

20:44 s/hotpatch/hotfix

20:44 Whatever the buzzword for that is

20:45 Frozenlock: Raynes: it's probably faster to just tell you instead of making a pull request... it would be great if refheap.el had clojurescript associated with clojure.

20:46 clojurescript-mode that is.

20:49 Raynes: Frozenlock: Released 0.0.3

20:49 Frozenlock: /90s on

20:49 wicked!

20:49 /90s off

20:50 rplaca: Raynes: don't get too aggressive with those version numbers!

20:52 Hodapp: don't get Donald E. Knuth syndrome either

20:52 rplaca: I think that my next project will just start with v1e-22

20:53 just to get out ahead of the trend

21:01 tomoj: you could digest edn reliably assuming no funky tagged literals, right?

21:02 I guess you have to invent some comparator for map keys and set elements of different types

21:02 callenbot: Raynes: does wrap-force-ssl work on Heroku?

21:11 Raynes: callenbot: I wrote it specifically for that purpose, so I hope so.

21:13 dxeh: any of you guys use elein emacs package

21:13 if so, worth it?

21:14 danlarkin: callenbot because I never got around to it

21:14 it'd be a good contribution though!

21:21 gfredericks: tomoj: I wouldn't think that would be a problem

21:21 tomoj: total ordering over all types isn't that tricky; couchdb does it for json

21:22 tomoj: not so easy for edn

21:22 well

21:22 I guess you can sort by tag for tagged literals

21:23 so maybe :)

21:23 gfredericks: sure

21:23 you don't have to care what the data-readers are

21:23 hell just assume everything is a tagged literal where the tag might be nil :)

21:23 maybe that's weird

21:24 what happens if you do nested literals?

21:24 read-string allows it

21:26 tomoj: shouldn't be a problem, you have to digest the tagged form too

21:26 gfredericks: sure

21:26 I wonder if the spec mentions that possibility

21:26 tomoj: db/fn is a good example

21:27 I feel there should be a print-edn

21:27 er, edn/pr or something

21:28 or fipp should have an ugly mode

21:29 gfredericks: tomoj: sort by (juxt (comp str class) identity) :)

21:29 that ignores data readers I guess

21:29 man your use case kind of demands an intermediate form where you have the tags and the unprocessed literals O_O

21:29 tomoj: that's easy

21:30 just (read-edn {:readers {} :default identity} ..) or whatever

21:30 gfredericks: I was just wondering that

21:30 tomoj: er, not identity

21:30 gfredericks: you'd probably want a defrecord for the tagged things

21:30 tomoj: yeah

21:30 gfredericks: wait does the function get the tag as well?

21:30 tomoj: s#read-edn#edn/read#

21:31 it gets the tag and the unprocessed value

21:31 so :default ->TaggedElement

21:31 gfredericks: okay then we're set

21:33 tomoj: I wonder why no 'uri in default-data-readers

21:35 gfredericks: or 'bytes

21:35 does edn have ratios?

21:36 tomoj: clojure.edn/read-string accepts it

21:36 gfredericks: which might be unrelated to whether the spec mentions them :)

21:37 tomoj: I don't see it in the spec

21:40 gfredericks: if I were a crochety old man I'd say that deserves a jira ticket

21:40 but I am a crochety young man, so I won't.

21:48 I knew core.logic would deliver me an example of a macro-writing macro :D

21:54 Raynes: gfredericks: heh

21:55 alex_baranosky: does anyone have a tool to reformat all the CLJ files in a project folder?

21:56 seems like you could hijack Emacs' clojure-mode formatting to do it, but I'm not very knowledgable in ELisp

22:34 ivan: how would you solve this: https://twitter.com/isaach/status/299595370760130561/photo/1 ? implement a subset of regex in core.logic?

22:35 Raynes: I'd start by telling MIT to eat a bag of phalluses.

22:47 SegFaultAX: Raynes: Not just suck, but eat?

22:50 gfredericks: explaining macros is hard.

22:52 ivan: these regexes look like proper CS regexes, no?

22:52 oh nevermind I see backreferences

22:53 * gfredericks throws a pile of theorems out the window

22:56 gfredericks: the backrefs shouldn't be hard for core.logic though

23:18 alex_baranosky: I'm trying to fix a bug in Slamhound that loses the second half of large files, but I am not knowledgable enough to pinpoint what could be causing it

23:19 anyone up for debugging it? It'd be nice to get a release out to fix kind of an egregious bug

23:20 I've even got a failing test case :)

23:32 tomoj: presumably the file doesn't have a :slam.hound.asplode/done sitting at the top-level half-way down

23:33 alex_baranosky: tomoj: nope :)

23:34 tomoj: well

23:34 it looks like it's copying and writing to the same file at the same time?

23:34 that can't be good

23:34 I'd try using a temp file

23:35 or.. the whole file really should be able to fit in memory, yes?

23:35 s/copying/reading/

23:35 alex_baranosky: there's something I don't understand about how read on a PushBAckerREader works

23:35 tomoj: io/reader is a BufferedReader, so presumably you lose whatever comes after the first buffered bit

23:37 alex_baranosky: will try the temp file approach

23:40 tomoj: hmm, if that were the explanation, I'd expect it to more likely blow up than truncate

23:43 alex_baranosky: tomoj: the files aren't really that big

23:43 just say 600 lines of Clojure

23:45 tomoj: apparently the default buffer size is 8192 bytes

23:45 is that about how big the truncated files are?

23:46 I don't have any sense for lines <-> bytes

23:51 technomancy: don't you hate it when you come to your senses in front of a jar of pistachios and have no idea how long you've been standing there eating them?

23:52 alex_baranosky: tomoj: the temp file fixed it!!!

23:53 xeqi: technomancy: at least they're somewhat healthy, not a bag of potato chips

23:53 alex_baranosky: technomancy: Fixed a bug in slam hound, would love to see the recent three features released. LEt me finish up the pull request though

23:53 technomancy: xeqi: plus the shells act as built-in rate limiters

23:53 alex_baranosky: cool; sure thing

23:54 noncom: i'm reading cojure the clojure documentation. and it says, for refs: "Refs are bound to a single storage location for their lifetime", and then, for agents, it says: "Where Refs support coordinated, synchronous change of multiple locations, Agents provide independent, asynchronous change of individual locations"

23:54 so are refs bound to a single or multiple locations?

23:55 technomancy: noncom: a ref is a single location, but refs support multiple refs being changed in a synchronous way

23:55 alex_baranosky: tomoj: the fix - https://github.com/technomancy/slamhound/pull/30/files#L2L16

23:56 noncom: ummm.... i don't get it. is it about using plural?

23:57 technomancy: noncom: read up on dosync

23:57 noncom: ok i wil!

Logging service provided by n01se.net