<< 01-02-2014 >>

---

00:03:21	EXetoC	reactormonk: same result when introducing an additional parenthesis pair?
00:12:08	dom96	Varriount: I am now at your disposal until I fall asleep :P
00:12:14	dom96	Varriount: Sorry for being away for so long.
00:12:39	dom96	Varriount: Just read through the logs: it seems that you arrived at the exact same point that I was at? I.e. it works when you alloc0 the TOverlapped struct.
00:12:48	dom96	I'm still unsure about the reason for this?
00:13:07	dom96	Maybe I missed it in the logs.
00:13:20	EXetoC	dom96: is my 2-line jester patch good?
00:13:36	dom96	EXetoC: the {.nimcall.} -> {.closure.}?
00:14:57	dom96	EXetoC: I still need to review the full implications of this change.
00:15:37	dom96	Varriount: Have you read this? https://github.com/nimrod-code/nimbuild/blob/master/structure.markdown
00:16:43	dom96	Varriount: btw I am willing to help you understand the builder if you wish.
00:17:35	*	darkf joined #nimrod
00:21:34	EXetoC	dom96: anything specific?
00:22:43	*	LordAndrew joined #nimrod
00:24:41	EXetoC	nevermind
00:28:34	*	BitPuffin joined #nimrod
00:34:26	*	BitPuffin quit (Ping timeout: 264 seconds)
00:59:15	*	brihat quit (Ping timeout: 252 seconds)
01:46:45	Varriount	So what's this talk of procedure call unification?
01:51:55	*	aruniiird joined #nimrod
01:57:02	EXetoC	Varriount: these will be callable using the same syntax: "proc init(var x: T)" and "proc init(): T". something like that
02:15:16	EXetoC	"var x = init()" -> "var x: T; init(x); x (discardable)"? We might find out when he blogs about it
02:15:48	Varriount	I like reading Araq's blog posts, they are quite informative.
02:18:06	EXetoC	but I wonder if it doesn't need to be more special than that, because now you can't use 'let', right?
02:18:53	Varriount	Huh?
02:29:18	*	fowl quit (Quit: Leaving)
02:32:29	*	DAddYE quit (Remote host closed the connection)
02:33:02	*	DAddYE joined #nimrod
02:37:39	*	DAddYE quit (Ping timeout: 260 seconds)
02:48:54	EXetoC	Varriount: I'm trying to figure out if a constructor feature is necessary
02:49:20	EXetoC	what else? replace var with let in that imaginary rewrite and it won't work
02:54:42	*	brihat joined #nimrod
03:11:23	*	fowl joined #nimrod
03:13:02	*	aruniiird quit (Ping timeout: 253 seconds)
03:18:41	OrionPK	hola gents
03:19:14	EXetoC	lo
03:21:45	OrionPK	what's new
03:24:26	EXetoC	not much
03:49:26	*	brson quit (Quit: leaving)
04:00:31	*	aruniiird joined #nimrod
04:21:42	reactormonk	seq[set[TEdge]] gives me Error: ordinal type expected
04:22:46	Varriount	What is TEdge?
04:23:14	reactormonk	TEdge = object of TObject
04:23:15	*	delian66 quit (Read error: Operation timed out)
04:23:53	Varriount	reactormonk: Built in sets can only contain ordinal values. Use the sets module for set objects that can hold more.
04:24:47	reactormonk	... duh
04:24:56	Varriount	;3
04:25:49	Varriount	Oh hi aruniiird
04:28:34	Varriount	reactormonk: I'm curious; if you don't mind me asking, what are you working on?
04:34:24	*	DAddYE joined #nimrod
04:39:27	*	DAddYE quit (Ping timeout: 272 seconds)
04:51:35	*	DAddYE joined #nimrod
04:59:44	reactormonk	Varriount, eve online PI stuff. Want to know which combination brings most money for the factories you need.
04:59:45	*	brson joined #nimrod
04:59:52	reactormonk	s/PI/planetary interaction/
05:00:01	reactormonk	how would you translate a seq[seq[T]] ?
05:00:10	Varriount	Ah. That explains the need for product()
05:00:20	reactormonk	yup
05:00:20	Varriount	reactormonk: Translate it to what?
05:00:42	reactormonk	Varriount, ehh transpose
05:01:20	Varriount	What kind of transposition?
05:01:36	*	bbodi joined #nimrod
05:02:20	reactormonk	http://en.wikipedia.org/wiki/Transpose
05:02:41	reactormonk	@[@[1,2,3],@[4,5,6]] => @[@[1,4],@[2,5],@[3,6]]
05:02:56	Varriount	Oh that.
05:03:00	Varriount	Use zip
05:03:33	reactormonk	zip?
05:04:07	Varriount	http://build.nimrod-lang.org/docs/sequtils.html
05:04:33	reactormonk	how does zip help here?
05:05:00	Varriount	reactormonk: Zip the two sequences inside the containing seq?
05:05:17	reactormonk	Varriount, nope, transpose is for an undefined number of sequences
05:05:36	Varriount	Still easy. Are they all the same length?
05:06:46	reactormonk	yup
05:08:38	Varriount	reactormonk: Do you want it to be an iterator?
05:08:55	reactormonk	Varriount, why not, but not needed
05:09:07	reactormonk	I'll just warp an accumulateResult around it
05:09:18	Varriount	"warp" sounds cool
05:09:39	reactormonk	too much eve
05:14:41	Varriount	reactormonk: https://gist.github.com/Varriount/8748313
05:15:44	Varriount	It says something about nimrod, that I can come up with a generic transposition iterator in less than 30 minutes, with only one file, less than 15 lines of code
05:16:00	Varriount	You'd never have it that easy in C++
05:16:28	Varriount	reactormonk: I used repr() because I'm too lazy to import sequtils.
05:19:48	Varriount	reactormonk?
05:20:22	EXetoC	I don't know what for. is this what you want? "when false: proc `$`*[T](a: openArray[T]): string"
05:20:51	EXetoC	it caused some problems, and has been disabled for ages
05:21:46	Varriount	no, I want " proc `$`*[T](a: seq[T]): string
05:22:14	Varriount	But repr achieves what I want, plus I get an interesting pointer address. :3
05:24:50	*	[1]Endy joined #nimrod
05:24:59	EXetoC	openarray means any array or sequence
05:26:00	Varriount	EXetoC: What kind of problems did it cause?
05:26:48	*	[2]Endy joined #nimrod
05:27:20	Varriount	Hi [1]Endy
05:28:21	Varriount	EXetoC: Do you know of any easy way to iterate over the fields of an object, and get their types?
05:29:07	*	[1]Endy quit (Ping timeout: 252 seconds)
05:35:03	*	[1]Endy joined #nimrod
05:38:15	*	[2]Endy quit (Ping timeout: 260 seconds)
05:40:54	EXetoC	Varriount: I can never remember
05:41:02	EXetoC	Varriount: get the types? system.fields maybe
05:41:37	Varriount	I've tried that, I just get an "Error: expression 'x' has no type (or is ambiguous)"
05:42:01	Varriount	Maybe it's because I'm using it in a template. :/
05:51:45	EXetoC	could be the documented bug
06:05:31	*	DAddYE quit (Remote host closed the connection)
06:09:21	*	xtagon quit (Ping timeout: 252 seconds)
06:18:53	*	Demos quit (Ping timeout: 245 seconds)
06:43:16	*	DAddYE joined #nimrod
07:12:14	*	aruniiird quit (Ping timeout: 264 seconds)
07:15:25	*	delian66 joined #nimrod
07:16:14	*	[1]Endy quit (Quit: HydraIRC -> http://www.hydrairc.com <- Now with extra fish!)
07:23:55	*	aruniiird joined #nimrod
07:31:58	*	DAddYE quit (Remote host closed the connection)
07:32:25	*	DAddYE joined #nimrod
07:33:20	delian66	I'm trying to build nimrod 0.9.2 from the .zip archive on the download page on Ubuntu 13.04, but I'm getting an error: http://pastebin.com/VqinXfqY
07:33:48	delian66	is this still supported, or should I instead just try the build from github thing ?
07:34:36	*	DAddYE_ joined #nimrod
07:34:56	*	DAddYE quit (Read error: Connection reset by peer)
07:35:33	fowl	delian66, known error with ubuntu
07:35:51	fowl	delian66, use the latest from github
07:35:55	delian66	btw, on Ubuntu, sh is a link to dash, and the build script should be run with bash
07:35:58	delian66	ok, 10x
07:36:48	*	DAddYE_ quit (Remote host closed the connection)
07:37:15	*	DAddYE joined #nimrod
07:41:29	*	DAddYE quit (Ping timeout: 252 seconds)
07:57:07	*	ddl_smurf joined #nimrod
08:07:33	*	DAddYE joined #nimrod
08:10:30	delian66	yay ... success !
08:10:34	delian66	Nimrod Compiler Version 0.9.3 (2014-02-01) [Linux: amd64]
08:10:35	delian66	:-)
08:44:38	*	aruniiird quit (Ping timeout: 264 seconds)
08:45:16	*	LordAndrew quit (Quit: Out.)
08:57:41	*	aruniiird joined #nimrod
08:58:19	*	io2 joined #nimrod
09:09:14	*	XAMPP_ joined #nimrod
09:11:34	*	XAMPP quit (Ping timeout: 260 seconds)
09:21:26	*	DAddYE quit (Remote host closed the connection)
09:21:52	*	DAddYE joined #nimrod
09:26:34	*	JStoker quit (Killed (hobana.freenode.net (Nickname regained by services)))
09:26:44	*	JStoker joined #nimrod
09:26:59	*	DAddYE quit (Ping timeout: 272 seconds)
09:30:28	*	brihat quit (*.net *.split)
09:30:28	*	EXetoC quit (*.net *.split)
09:30:29	*	zielmicha8 quit (*.net *.split)
09:30:29	*	[Pete_27] quit (*.net *.split)
09:30:30	*	iNode000 quit (*.net *.split)
09:34:36	*	brihat joined #nimrod
09:34:36	*	EXetoC joined #nimrod
09:34:36	*	iNode000 joined #nimrod
09:34:36	*	zielmicha8 joined #nimrod
09:34:36	*	[Pete_27] joined #nimrod
09:43:45	*	aruniiird quit (Ping timeout: 245 seconds)
09:53:12	*	BitPuffin joined #nimrod
09:56:47	*	aruniiird joined #nimrod
10:00:49	*	io2 quit (Ping timeout: 252 seconds)
10:19:05	*	_davidk_ joined #nimrod
10:19:59	*	shodan45 quit (Ping timeout: 240 seconds)
10:22:17	*	barry joined #nimrod
10:25:21	Araq	hi barry welcome
10:25:30	barry	hello Araq
10:25:44	Araq	delian66: the build script is supposed to be posix style, so whatever 'sh' points to should be fine
10:26:03	Araq	hi _davidk_ welcome
10:28:55	delian66	Araq: well, the build script did not work under sh linked to dash, I suppose there is some bashism in it, because it run fine, when I invoke it with bash; will try to determine what exactly messed it later today.
10:29:13	Araq	great, thanks
10:31:55	*	ninjashogun joined #nimrod
10:32:10	ninjashogun	hey there
10:32:29	ninjashogun	does, uh, 'girvo' hang out here?
10:35:23	*	aruniiird quit (Ping timeout: 265 seconds)
10:35:37	Araq	hi ninjashogun
10:35:56	Araq	yes, though I haven't seen him for a while
10:36:05	Araq	!seeen girvo
10:36:14	Araq	NimBot: !seen girvo
10:36:33	*	Araq can't remember the syntax ...
10:36:45	Araq	!seen girvo
10:36:45	NimBot	girvo was last seen on Mon Jan 13 14:48:32 2014 quitting with message: Quit: Leaving
10:41:22	Araq	ninjashogun: any idea what happened to him?
10:45:54	ninjashogun	ah okay
10:46:07	ninjashogun	I don't know, he justmentioned this channel on Hacker News
11:01:13	*	brson quit (Ping timeout: 252 seconds)
11:02:49	*	brson joined #nimrod
11:14:18	*	brson quit (Ping timeout: 245 seconds)
11:28:32	*	resure joined #nimrod
11:34:54	*	luyang joined #nimrod
11:36:32	*	clsterfks joined #nimrod
11:39:22	*	NimBot joined #nimrod
11:47:01	*	ninjashogun quit (Read error: Connection reset by peer)
11:47:31	*	ninjashogun joined #nimrod
11:50:14	*	[1]Endy joined #nimrod
11:57:23	*	luyang quit (Quit: Leaving.)
11:58:37	*	ninjashogun quit (Ping timeout: 272 seconds)
12:00:41	Araq	hi resure welcome
12:03:51	*	luyang joined #nimrod
12:03:51	*	luyang quit (Changing host)
12:03:51	*	luyang joined #nimrod
12:04:07	luyang	what's up
12:05:22	Araq	hi luyang welcome
12:05:43	Araq	not much "up", fixing bugs for 0.9.4
12:05:50	*	vendethiel quit (Read error: Connection reset by peer)
12:25:06	*	vendethiel joined #nimrod
12:31:59	*	vendethiel quit (Ping timeout: 240 seconds)
12:32:27	*	vendethiel joined #nimrod
12:49:39	*	failover joined #nimrod
12:49:45	*	failover left #nimrod (#nimrod)
12:51:09	*	epsylon joined #nimrod
12:51:53	*	resure left #nimrod (#nimrod)
12:59:19	*	enurlyx joined #nimrod
13:00:36	Araq	hi epsylon welcome
13:04:02	*	luyang quit (Quit: Leaving.)
13:05:27	enurlyx	Hi Araq
13:05:38	enurlyx	Do you have plans for a gui? Claro?
13:05:45	enurlyx	Or is someone else working on a gui?
13:06:05	Araq	from time to time I try to sell Claro but nobody is interested
13:06:35	Araq	UIs are complex, I guess we should wrap webkit or similar and use that ...
13:06:38	enurlyx	Claro source is in c?
13:06:52	Araq	yes
13:08:53	Araq	exhu is working on a native UI library but I don't know its status
13:09:28	Araq	it uses x11 or the winapi directly, I think
13:09:51	enurlyx	That sounds good. I cant think of an CAD-App with WebKit
13:10:09	Araq	he might even have a wayland port by now
13:11:34	*	darkf quit (Quit: Leaving)
13:12:04	Araq	enurlyx: steam uses webkit, I think
13:12:32	Araq	and in general these days most "software portals" are browser based
13:13:11	dom96	Indeed. Steam's UI uses libcef. Battle.net's too.
13:14:23	*	io2 joined #nimrod
13:16:14	Araq	oh yeah
13:16:27	Araq	we should wrap that
13:16:39	dom96	Varriount already has.
13:16:52	dom96	I think it's untested though...
13:16:53	Araq	what? o.O
13:19:28	Araq	Varriount, dom96 just alloc0 and dealloc, no need to keep a seq of refs around
13:20:00	*	noam quit (Ping timeout: 245 seconds)
13:20:31	dom96	Ahh so my instinct was correct. Good.
13:21:12	Araq	bbl
13:23:05	*	randallsquared joined #nimrod
13:23:17	dom96	hello randallsquared
13:24:16	*	[2]Endy joined #nimrod
13:24:51	randallsquared	Hi dom96.
13:24:53	enurlyx	Where can i find exhu's project?
13:25:00	enurlyx	What is outstanding about Claro?
13:25:21	dom96	enurlyx: https://bitbucket.org/exhu/paulina
13:26:55	*	[1]Endy quit (Ping timeout: 272 seconds)
13:29:16	enurlyx	Did you know a group at Microsoft works at a new Systemsprogramming language (M#)?
13:31:06	*	[1]Endy joined #nimrod
13:31:12	*	aruniiird joined #nimrod
13:34:51	*	[2]Endy quit (Ping timeout: 252 seconds)
13:39:49	dom96	enurlyx: I heard about it a while back. Haven't heard any news about it recently though.
13:42:46	enurlyx	Yeah, not so much infos. I just wanted to mention.
13:44:51	*	aruniiird quit (Ping timeout: 265 seconds)
14:01:59	bbodi	Hi all!
14:02:20	dom96	hi bbodi!
14:02:46	bbodi	A simple console application at startups consume 50 MB memory. What is the reason behind it? Is it configurable?
14:02:49	bbodi	hi dom96!
14:04:20	dom96	Can I see your code?
14:05:00	bbodi	yes: echo GC_getStatistics()
14:05:02	bbodi	:D
14:05:13	bbodi	thats the whole program
14:05:32	bbodi	"[GC] occupied memory: 40960"
14:05:51	dom96	Interesting. I have no idea heh.
14:22:10	dom96	Actually, I think that value is in bytes.
14:22:16	bbodi	yes it is
14:28:43	reactormonk	Varriount, sorry, went to sleep
14:29:45	reactormonk	Varriount, looks good. Araq, https://gist.github.com/Varriount/8748313 to sequtils?
14:31:14	*	decoa joined #nimrod
14:31:24	reactormonk	Araq, btw, what's the distinction between algorithm and sequtils?
14:31:25	dom96	hello decoa
14:31:44	*	decoa quit (Quit: AndroIRC - Android IRC Client ( http://www.androirc.com ))
14:37:08	NimBot	nimrod-code/nimbuild master 04793c7 Dominik Picheta [+0 ±1 -0]: A single thread is now created removing the need for joinThreads.... 3 more lines
14:46:53	reactormonk	http://dpaste.com/1580454/ got this test failing - how do I annotate accumulateResult?
14:48:55	reactormonk	uhm, is an iterator first-class yet?
14:50:03	reactormonk	oh, it quite literally accumulate the result :-/
14:52:03	dom96	Perhaps you want sequtils.toSeq?
14:53:41	*	enurlyx quit (Quit: Leaving)
14:55:27	*	io2 quit (Ping timeout: 260 seconds)
14:56:54	NimBot	Araq/Nimrod devel ea08ebc Dominik Picheta [+0 ±1 -0]: Fixes wrong slurp path in tester.
15:07:27	*	ceptr joined #nimrod
15:08:45	dom96	hi ceptr
15:09:06	ceptr	hi dom96
15:11:05	reactormonk	dom96, most likely
15:19:08	reactormonk	lib/pure/collections/sets.nim(132, 2) Error: internal error: for statement not eliminated
15:19:10	reactormonk	O.o
15:25:42	*	ahhdem joined #nimrod
15:25:46	*	ahhdem left #nimrod (#nimrod)
15:27:14	Araq	reactormonk: your PR is nice except that use unittest.nim for the test and this is not really supported by the tester
15:27:55	reactormonk	Araq, muh
15:28:43	*	panz53 joined #nimrod
15:28:47	Araq	dom96: what's the slurp path in the tester?
15:28:50	Araq	hi panz53 welcome
15:29:00	Araq	how can it be wrong?
15:29:10	dom96	look at the diff
15:31:19	Araq	you should remove tests/tester instead
15:31:29	Araq	and use the new tester
15:31:52	reactormonk	Araq, how complicated would it be to add unittest support to the tester?
15:32:47	Araq	not very, but it sucks
15:32:57	Araq	unittest is a quite complex meta program itself
15:33:14	Araq	so I never know if I break the unittest module or the underlying test
15:33:24	reactormonk	hmmm
15:33:51	reactormonk	so I just should replace check with echo and go true/true/true/ etc?
15:34:02	Araq	or use doAssert
15:34:09	Araq	which is not nearly as complex
15:35:06	Araq	the unittest module is loved by many, but for me it is a maintainance pita
15:35:25	dom96	Araq: Why is it still there? And why does koch tests execute it?
15:35:46	Araq	*maintenance
15:35:55	*	io2 joined #nimrod
15:36:02	Araq	dom96: my koch doesn't execute it
15:36:22	reactormonk	Araq, write a little unittest-koch? ^^
15:37:20	Araq	hi ceptr welcome
15:37:27	*	psquid quit (Quit: work)
15:38:29	dom96	Araq: ahh. Nimbuild didn't rebuild koch.
15:38:44	Araq	yeah and it shouldn't
15:38:51	Araq	koch changes rarely
15:39:01	dom96	It does when github tells it that it was modified.
15:39:26	Araq	can github do that?
15:39:35	Araq	koch.nim is part of the compiler
15:39:48	dom96	yes.
15:40:07	Araq	it can tell us a single file changed? awesome
15:40:08	dom96	When you commit, github sends a list of modified files to nimbuild
15:40:18	Araq	ah I see
15:41:04	dom96	Araq: Does the new tester still not generate testresults.json?
15:41:25	Araq	apparently not
15:41:48	Araq	outputJson is never called ... -.-
15:42:22	Araq	ah I see
15:42:36	Araq	it was generated in the 'merge' step which do not have anymore
15:45:26	reactormonk	Araq, kk, fixed all tests. The PRs are otherwise fine?
15:45:57	Araq	the *pr*, I only looked at one
15:46:36	reactormonk	take a look at the rest too
15:46:52	Araq	later
15:47:40	*	phI||Ip joined #nimrod
15:47:43	Araq	dom96: you only need the 3 numbers, right?
15:47:51	Araq	hi phI||Ip welcome
15:47:56	phI||Ip	thanks
15:48:56	dom96	Araq: hrm, you probably don't have the concept of the reject, compile and run categories right?
15:49:14	dom96	Araq: That's how the JSON used to be split up like.
15:49:23	Araq	dom96: not really, but I can recompute it
15:49:30	dom96	Araq: Just change it to whatever works and i'll adapt the builder.
15:49:46	dom96	All the builder needs is the number of tests passed, skipped and the total.
15:49:58	Araq	ok, you'll get that instead then
15:50:13	dom96	testresults.html is also required
15:50:24	Araq	that should be generated already
15:50:37	dom96	ok. In the same dir as koch?
15:55:14	Araq	yes
15:57:55	Araq	have our test machines sqlite installed?
15:59:17	dom96	yep
16:00:55	dom96	Lucky for you :P
16:02:09	Araq	well firefox uses it for its configuration
16:02:36	dom96	Yeah. All servers have firefox installed by default :P
16:02:57	Araq	it is the most deployed database in the world :P
16:03:09	Araq	my smartphone has it too
16:04:41	*	codehotter joined #nimrod
16:04:43	codehotter	#haskell
16:04:54	*	codehotter left #nimrod ("WeeChat 0.3.7")
16:05:15	Araq	that was ... impolite
16:06:16	*	truff_ joined #nimrod
16:07:10	dom96	hello truff_
16:07:30	truff_	hello
16:07:45	truff_	funny, you sent me here from #haskell
16:07:54	dom96	hah
16:08:05	Araq	codehotter sent me to #haskell
16:08:15	dom96	codehotter joined here and said "#haskell"
16:08:38	dom96	I figured I would do something similar in #haskell :P
16:08:52	*	panz53 quit (Quit: Leaving)
16:08:57	truff_	i guess it worked :)
16:09:00	Araq	yeah and that was nice because I have never heard of the Haskell programming language ...
16:18:30	*	achim joined #nimrod
16:28:03	*	ceptr quit (Ping timeout: 272 seconds)
16:34:30	*	Clinteger\c joined #nimrod
16:40:02	dom96	Araq: Actually, I think you should keep these categories. Otherwise testing other branches will fail.
16:41:25	dom96	Araq: Just make it generate: {"run": {"passed": ..., "total": ..., "skipped": ...}, "reject": ..., "compile": ..., "passed": ..., "total": ..., "skipped": ...}
16:41:53	dom96	hi Clinteger\c
16:42:36	dom96	Araq: or hrm, nah. I can just get the builder to work with both formats.
16:45:17	Clinteger\c	hello :)
17:08:41	*	aruniiird joined #nimrod
17:20:05	*	BitPuffin quit (Ping timeout: 248 seconds)
17:37:24	*	aruniiird quit (Ping timeout: 252 seconds)
17:44:56	reactormonk	Clinteger\c, sup
17:46:12	*	BitPuffin joined #nimrod
18:02:14	*	[2]Endy joined #nimrod
18:04:37	*	[3]Endy joined #nimrod
18:04:41	*	[1]Endy quit (Ping timeout: 272 seconds)
18:06:07	*	brson joined #nimrod
18:07:00	Araq	hi Clinteger\c welcome
18:07:51	*	[2]Endy quit (Ping timeout: 272 seconds)
18:12:13	*	BitPuffin quit (Ping timeout: 245 seconds)
19:04:57	*	bbodi quit (Ping timeout: 252 seconds)
19:14:04	*	ninjashogun joined #nimrod
19:15:30	*	luyang joined #nimrod
19:15:30	*	luyang quit (Changing host)
19:15:30	*	luyang joined #nimrod
19:16:23	NimBot	nimrod-code/nimbuild master a774fff Dominik Picheta [+0 ±1 -0]: Add support for new testresults.json.
19:16:23	NimBot	nimrod-code/nimbuild master fadfd3b Dominik Picheta [+0 ±1 -0]: Get rid of dependency on koch clean.
19:16:23	NimBot	nimrod-code/nimbuild master 4b12f98 Dominik Picheta [+0 ±1 -0]: Changed the way the builder uses git.
19:16:23	NimBot	nimrod-code/nimbuild master 3184baf Dominik Picheta [+0 ±1 -0]: Nimrod/koch binaries are now tracked between branches.
19:16:34	dom96	hello luyang and ninjashogun
19:17:13	ninjashogun	hi dom96
19:17:18	ninjashogun	why did you say hi to just us two?
19:17:38	dom96	Because I like to say hi to new people :)
19:17:46	ninjashogun	nice of you :)
19:18:03	ninjashogun	do you follow startups at all dom96?
19:19:54	dom96	Not really. I look at HN if that counts heh.
19:22:19	luyang	dom96 your welcoming gives me an impression of a good and friendly channel
19:22:47	dom96	luyang: That's good. It is the impression we strive for :)
19:22:50	luyang	I keep reading about nimrod on Hacker News
19:22:53	luyang	good
19:23:27	luyang	I mean it's been mentioned in comments
19:23:30	ninjashogun	so none of you have done a startup?
19:24:19	dom96	ninjashogun: There probably are a couple in here that have. But this channel is mainly about the Nimrod programming language.
19:24:35	*	[3]Endy quit (Ping timeout: 245 seconds)
19:24:52	dom96	luyang: Yeah, I noticed some people commenting about this channel in that Ask HN thread.
19:25:09	luyang	yeah
19:25:30	NimBot	nimrod-code/nimbuild master f86a6c5 Dominik Picheta [+0 ±1 -0]: Removed 'refs/' from branch name passed to git checkout.
19:26:27	dom96	luyang: Have you tried Nimrod at all yet?
19:27:50	dom96	ninjashogun: I am actually considering creating some sort of startup, still need to do a lot of research on it though. Have you done a startup?
19:30:12	*	Mat3 joined #nimrod
19:30:16	Mat3	hi all
19:30:24	dom96	hey Mat3
19:34:48	Mat3	hi dom96
19:39:35	*	BitPuffin joined #nimrod
19:39:50	dom96	Mat3: what's new?
19:44:08	*	Demos joined #nimrod
19:46:56	Mat3	not much, preparing a blog, new website and finsihing my VM environment
19:47:16	Mat3	^finishing
19:47:29	dom96	Cool. Will you be blogging about Nimrod?
19:47:38	Demos	what VM envirenment. I was setting something up with VMs just last week
19:48:04	Mat3	my bare-metal VM designa as base for these Nimrod subset
19:48:09	Mat3	^design
19:48:55	Mat3	(you now, MISC style, support for parallel processing and AOT compiler)
19:49:51	Demos	oh dear, are we getting a third vm?
19:50:48	dom96	What are the other 2?
19:50:54	luyang	dom96: No I haven't tried it yet. I need to learn more about it
19:51:34	dom96	luyang: I see. Well if you have any questions feel free to ask them here, or even highlight me if no one answers.
19:52:04	Demos	dom96, I thought a big feature of 0.9.4 was going to be a new vm for compile time FFI and nimrod i
19:53:13	*	zahary joined #nimrod
19:54:01	dom96	Demos: Yeah. That's only one though, what's the second VM?
19:54:23	dom96	I suppose you are referring to the AST interpreter in 0.9.2?
19:55:28	Demos	erm the one nimrod i uses right now...
19:56:12	Mat3	no, its an bare-metal environment for systems with restricted ressources (like AVR microcontrollers)
19:56:17	dom96	Yeah, that's an AST interpreter. Not sure if you can call that a VM.
19:57:02	Araq	dom96: VM is not strictly defined, an AST interpreter can count as a VM
19:57:28	Mat3	I do not know what you mean with AST. Anyhow I call that a soft-core to be precise
19:57:33	dom96	Araq: I see.
19:57:47	Mat3	ciao
19:57:51	*	Mat3 quit (Quit: Verlassend)
19:57:55	NimBot	nimrod-code/nimbuild master 61aaf90 Dominik Picheta [+0 ±1 -0]: Git checkout should now be fixed.
19:58:56	*	Icefoz joined #nimrod
20:06:15	*	iNode000 quit (Ping timeout: 260 seconds)
20:17:16	*	Icefoz quit (Quit: leaving)
20:18:28	*	waa joined #nimrod
20:19:55	*	waa left #nimrod (#nimrod)
20:22:54	*	luyang quit (Quit: Leaving.)
20:30:56	*	zahary quit (Quit: Leaving.)
20:44:42	*	luyang joined #nimrod
20:44:42	*	luyang quit (Changing host)
20:44:42	*	luyang joined #nimrod
20:49:43	luyang	I just installed nimrod via homebrew (Mac OS X package manager)
20:50:58	luyang	and nimrod is not in the path… Nimrod has been installed to /usr/local/Cellar/nimrod/0.9.2/libexec. The compiler will currently fail to find system.nim if called through a ymlink. To compile nim files, specify the full path to the compiler: /usr/local/Cellar/nimrod/0.9.2/libexec/bin/nimrod compile --run hello.nim
20:51:03	luyang	that's just a shame
20:51:23	Demos	hey is code like this supposed to be invalid https://gist.github.com/barcharcraz/ac5c995322206dfbb456?
20:51:55	Demos	we should fix the symlink thing.
20:52:07	dom96	Isn't it already fixed in master?
20:52:25	Varriount	'symlink thing'?
20:52:25	Demos	OK if it is than we should fix that message and the brew install files.
20:52:33	dom96	luyang: I suggest simply cloning the repo and adding where the nimrod binary is to your path.
20:52:51	dom96	Demos: We need to wait until 0.9.4 is released then
20:53:05	Demos	luyang, you could also use brew diy to let homebrew managed the compiler produced from master
20:53:15	Demos	s/managed/manage
20:53:18	luyang	brew dyi? :)
20:53:52	luyang	sounds like a good solution but I haven't done it myself before with brew
20:54:11	luyang	I already have brew, it's just a matter of setting the path
20:54:21	luyang	s/brew/nimrod/
20:54:41	Demos	https://github.com/Homebrew/homebrew/wiki/FAQ down under the "Can I install my own stuff to /usr/local"
20:55:31	Demos	tldr, clone nimrod, cd to the clone, brew diy, build nimrod
20:55:50	Demos	install nimrod to /usr/local/Cellar/nimrod/0.9.3
20:55:57	Demos	brew link nimord
20:56:17	Araq	Demos: no, it's not supposed to be valid
20:56:23	Demos	darn
20:56:39	Demos	I guess it is a edge case, is it even valid in C?
20:58:58	*	brson quit (Quit: leaving)
20:59:39	luyang	Demos thanks for the explanation
21:00:11	Demos	no problem
21:02:01	luyang	I tried to compile aporia and got command line(0, 0) Error: internal error: expr(nkNone); unknown node kind
21:02:02	luyang	No stack traceback available
21:02:32	*	zahary joined #nimrod
21:07:02	dom96	Yeah, I think that's this issue: https://github.com/Araq/Nimrod/issues/342
21:07:41	luyang	ah
21:08:14	luyang	koch?
21:09:10	luyang	(the solution mentions ./koch)
21:09:23	luyang	./koch boot -d:release --gc:markAndSweep
21:09:23	luyang	but I'm using nimcor
21:09:26	luyang	nimrod c
21:09:41	dom96	That's the tool used for bootstrapping the compiler.
21:10:31	luyang	ok I feel I have to read up on the nimrod tools
21:11:38	dom96	There isn't many you need to know about.
21:12:04	dom96	I would give you a list but I don't think one exists.
21:12:55	Araq	http://nimrod-lang.org/tools.html ?
21:13:08	Araq	but that's not complete
21:24:45	*	luyang quit (Quit: Leaving.)
21:29:59	*	BitPuffin quit (Ping timeout: 240 seconds)
21:36:56	*	shodan45 joined #nimrod
21:37:39	*	shodan45 quit (Read error: Connection reset by peer)
21:50:58	*	luyang joined #nimrod
21:50:58	*	luyang quit (Changing host)
21:50:58	*	luyang joined #nimrod
22:02:23	*	iNode000 joined #nimrod
22:02:47	*	ninjashogun quit (Quit: Leaving)
22:18:48	*	mblais5 joined #nimrod
22:19:09	*	mblais5 left #nimrod (#nimrod)
22:23:00	luyang	my ./koch command finishes with FAILURE
22:23:06	luyang	lib/system/gc_ms.nim(62, 2) Error: undeclared identifier: 'InstantiateForRegion'
22:31:48	dom96	Seems someone broke the mark and sweep GC>
22:35:46	luyang	does nimrod have anything similar to junit?
22:35:49	*	io2 quit (Ping timeout: 248 seconds)
22:35:50	luyang	or nunit
22:37:32	luyang	that could probably prevent such breaks, right?
22:38:23	EXetoC	'unittest' maybe, but many people do unit testing without it
22:38:26	EXetoC	luyang: I suppose
22:38:57	luyang	I tried to rerun it without gc and it seemed to be successful ./koch boot -d:release -d:nativeStacktrace
22:38:59	EXetoC	I tried to fix the errors in gc_ms but it got too complicated for me. changing the case of some of the symbols was the easy part
22:39:42	EXetoC	ok, but --gc:markAndSweep was the supposed fix for Aporia
22:39:42	luyang	My goal is to be able to create a binary of Aporia
22:40:07	luyang	When I try to compile Aporia I get aporia.nim(11, 7) Error: cannot open 'glib2'
22:40:43	luyang	EXetoC: yeah, any idea how that gc could be fixed?
22:42:42	Demos	there may be dependencies, are you building with babel build?
22:42:44	EXetoC	the babel package manager can take care of dependencies. you can also install the gtk2 package manually if you want
22:43:36	luyang	Um, I read about installing thre gtk2 package using MacPorts but I have Homebrew and MacbPorts doesn't work together with it
22:43:44	Demos	and you will find that a lot of tools that are seperate programs for other languages are just libraries in nimrod.
22:44:10	Demos	is gtk not avalible via homebrew?
22:45:03	luyang	I find these: gtk+ gtk-gnutella gtkmm3 gtkspell3 pygtksourceview gtk+3
22:45:15	EXetoC	gtk+ should be it
22:45:17	luyang	so I should install gtk+
22:45:35	EXetoC	but you also need the wrapper written in nimrod. here's the package manager for nimrod https://github.com/nimrod-code/babel
22:49:40	luyang	I am looking forward to getting all this up and running
22:49:53	luyang	installing gtk+ and babel is installed and PATH'd
22:50:11	Demos	do note that Aporia and nimrod's IDE integration are a little lacking right now.
22:50:45	luyang	ok :)
23:07:39	*	luyang quit (Quit: Leaving.)
23:08:02	*	luyang joined #nimrod
23:08:03	*	luyang quit (Changing host)
23:08:03	*	luyang joined #nimrod
23:12:23	*	luyang quit (Ping timeout: 252 seconds)
23:12:55	*	psquid joined #nimrod
23:13:59	*	zahary quit (Quit: Leaving.)
23:21:47	*	brson joined #nimrod
23:21:47	Demos	does nimrod have any notion of const at all? if I am writing declerations for C functions that take const params what do I use?
23:22:00	Araq	no
23:22:12	Araq	ignore the 'const'
23:22:53	Araq	it's absurd crap anyway
23:23:06	Demos	yeah, it is pretty pointless
23:23:22	Araq	the optimizer cannot do anything with it, since mutable aliases can exist
23:23:39	Demos	yeah, it can be nice in c++ because of the way params work
23:23:48	Demos	but in nimrod it would be pointles
23:23:50	Demos	just asking
23:23:54	Araq	and it doesn't compose, you cannot pass a vector<const string> to a vectory<string>
23:27:39	Demos	you mean the other way round right? vector<string> to vector<const string>
23:30:21	Demos	why did you decide to add unsigned types as opposed to keeping the % operators?
23:31:50	Araq	because it was too hard to constuct integers from signed int8, even for me :-)
23:32:07	Araq	unsigned sucks, but signed int8 instead of byte sucks even more
23:34:12	Demos	right, I guess that is reasonable
23:34:53	reactormonk	dom96, kk, updated
23:37:05	*	aruniiird joined #nimrod
23:42:30	EXetoC	Araq: did I get this part of the syntax unification right? "proc init(x: var T); var x = init()"
23:42:50	Araq	EXetoC: no
23:43:10	Araq	or yes
23:43:18	Araq	I dunno yet
23:43:33	Araq	in any case it'll require a pragma
23:43:37	Araq	most likely is:
23:44:00	Araq	proc foo(result: var T) {.func.}
23:44:17	Araq	# compiler also supports: let x = foo()
23:44:27	Araq	and generates this:
23:44:40	Araq	foo(freshTemp); let x = freshTemp
23:45:58	NimBot	Araq/Nimrod devel 1b8b276 Araq [+0 ±1 -0]: bugfix: object constructor doesn't allow 'distinct' types
23:45:58	NimBot	Araq/Nimrod devel c7f04ad Araq [+0 ±1 -0]: documented new symbol binding rules for templates
23:45:58	NimBot	Araq/Nimrod devel 7d45780 Araq [+0 ±2 -0]: case consistency for evalffi
23:45:58	NimBot	Araq/Nimrod devel 9919f31 Araq [+0 ±1 -0]: case consistency for evalffi
23:45:58	NimBot	12 more commits.
23:46:18	Araq	dom96: here you go the new tester supports json output
23:46:34	Araq	and 'koch' should produce it
23:47:50	dom96	thanks
23:48:18	Araq	also fyi the threading tests all work on win32, including 'joinThreads'
23:51:11	Araq	reactormonk: why do we need 'transpose' for seqs in the algorithm.nim?
23:51:39	Araq	that is only necessary for matrixes, right?
23:52:24	EXetoC	Araq: ok so the semantical difference is that one copy
23:53:13	Araq	EXetoC: no, it's a profound feature
23:53:44	EXetoC	I mean in that isolated case
23:53:57	Araq	oh ok
23:54:23	Araq	well the copy is not necessary, 'let x= freshtemp' could be a move too
23:55:57	NimBot	Araq/Nimrod devel c82135c Grzegorz Adam Hankiewicz [+0 ±1 -0]: Adds note about manual memory handling to GC doc.
23:55:57	NimBot	Araq/Nimrod devel d12652e Andreas Rumpf [+0 ±1 -0]: Merge pull request #843 from gradha/pr_gc_docs... 2 more lines
23:56:20	NimBot	Araq/Nimrod devel 3b10841 Grzegorz Adam Hankiewicz [+0 ±1 -0]: Avoids raising exceptions in tryInsertID.
23:56:20	NimBot	Araq/Nimrod devel eac992f Andreas Rumpf [+0 ±1 -0]: Merge pull request #855 from gradha/pr_fixes_try_insert_id... 2 more lines
23:57:28	EXetoC	I was trying to figure out how that could be done without introducing additional language features. Do you have to transform the C code rather than just the AST?
23:58:21	Araq	the compiler will transform just the AST
23:58:58	Araq	but it's tedious/slow/causes code bloat to do with a macro