<< 21-05-2014 >>
00:01:29*noam quit (Ping timeout: 264 seconds)
00:02:01*springbok joined #nimrod
00:04:08*genivf quit (Ping timeout: 255 seconds)
00:27:01*noam joined #nimrod
01:01:27*boydgreenfield quit (Quit: boydgreenfield)
01:06:28*q66 quit (Quit: Leaving)
01:37:57*brson quit (Ping timeout: 252 seconds)
01:42:32*hoverbear joined #nimrod
01:44:31*brson joined #nimrod
01:56:34*brson quit (Quit: leaving)
02:07:37*bjz joined #nimrod
02:12:50*bjz quit (Ping timeout: 255 seconds)
02:19:04*bjz joined #nimrod
02:24:41*boydgreenfield joined #nimrod
02:30:39*freezerburnv joined #nimrod
02:30:40*freezerburnv quit (Client Quit)
02:30:58*freezerburnv joined #nimrod
02:34:28*boydgreenfield quit (Quit: boydgreenfield)
03:01:34*freezerburnv quit (Quit: freezerburnv)
03:07:57*nande_ joined #nimrod
03:11:18*BitPuffin quit (Ping timeout: 240 seconds)
03:38:53*xenagi quit (Quit: Leaving)
03:43:11*boydgreenfield joined #nimrod
03:51:18*boydgreenfield quit (Ping timeout: 276 seconds)
04:01:34*ehaliewi` joined #nimrod
04:04:38*ehaliewicz quit (Ping timeout: 255 seconds)
04:13:02*flaviu1 quit (Remote host closed the connection)
04:17:45*xtagon joined #nimrod
04:31:01*xtagon quit (Quit: Leaving)
04:32:17*nequitans joined #nimrod
04:35:35*erlnoob joined #nimrod
04:49:30*bjz_ joined #nimrod
04:50:08*bjz quit (Read error: Connection reset by peer)
05:37:02*ehaliewi` is now known as ehaliewicz
05:39:35*bjz_ quit (Ping timeout: 255 seconds)
05:40:23*bjz joined #nimrod
05:49:51*nequitans quit (Ping timeout: 265 seconds)
05:52:07*hoverbear quit ()
06:05:34VarriountOk. After being away for what seems like years, I am now back.
06:06:36fowlwb
06:44:36*fowl quit (Quit: Derogatory terms for gynecomastia can include moobs (for male boobs) and bitch tits.[34])
06:58:57*fowl joined #nimrod
07:02:06*kunev joined #nimrod
07:33:21Varriountdom96: I've hit a snag concerning cancelling the monitoring of a file
07:50:33Varriountdom96: There's the possibility that if someone somehow performs an asynchronous operation on a directory handle that they are watching, and stop watching the directory, the asynchronous operation will also be cancelled.
08:27:09*BitPuffin joined #nimrod
08:44:27*erlnoob quit (Quit: Leaving.)
08:55:31*darkf quit (Ping timeout: 240 seconds)
09:06:33*nande_ quit (Read error: Connection reset by peer)
09:26:43*springbok quit (Ping timeout: 240 seconds)
09:37:08*springbok joined #nimrod
09:37:32*springbok is now known as Guest56053
09:47:58AraqVarriount: http://forum.nimrod-lang.org/t/449
09:48:08Araqis that a win XP problem?
10:05:38Araqhmm we have XP users... interesting
10:08:41*Varriount|Mobile joined #nimrod
10:12:06Varriount|MobileAraq: Yes, it's a WinXP problem. How should this problem be solved? Compile-time or runtime checks?
10:12:55Varriount|MobileIs there even a compile time symbol available which tells the version of Windows a program is being run on?
10:34:14AraqVarriount|Mobile: there is no define for winxp and it would be fragile anyway
10:34:51Araqwe should move createLink into its own module to deal with the problem
10:35:16Araqalternatively we could make os.nim {.deadCodeElim:on.}
10:36:33AraqI leave the decision up to you
10:37:13Araq{.deadCodeElim:on.} for os and winlean is a good idea anyway though
10:39:35Araqbut hmm we should have -d:winxp that is an optional define
10:39:57Araqso people can check whether their code uses any dependency > XP
10:42:13*ehaliewicz quit (Remote host closed the connection)
11:31:04*OffTheRails joined #nimrod
11:57:41Araqhi OffTheRails welcome
12:13:03*untitaker quit (Ping timeout: 240 seconds)
12:18:33*untitaker joined #nimrod
12:51:58*nequitans joined #nimrod
13:23:36*freezerburnv joined #nimrod
13:56:54*nequitans quit (Ping timeout: 240 seconds)
14:21:42*kunev quit (Quit: leaving)
14:34:33Varriount|MobileAraq: I thought dead code elimination was on by default?
14:38:53*io2 joined #nimrod
14:50:15Varriount|MobileAraq: http://msdn.microsoft.com/en-us/library/windows/desktop/ms724451(v=vs.85).aspx
14:50:33Varriount|MobileHi io2
14:55:58*bjz quit (Ping timeout: 240 seconds)
14:56:55io2lo
14:57:30*freezerburnv quit (Quit: freezerburnv)
15:07:03*kunev joined #nimrod
15:15:02*Jesin joined #nimrod
15:18:02*Jesin quit (Client Quit)
15:28:28*freezerburnv joined #nimrod
15:33:35*Jesin joined #nimrod
15:50:59*kunev quit (Quit: Lost terminal)
16:05:48*OffTheRails quit (Ping timeout: 276 seconds)
16:08:04*OrionPK quit (Remote host closed the connection)
16:08:21*OrionPK joined #nimrod
16:18:12*hoverbear joined #nimrod
16:20:18*OffTheRails joined #nimrod
16:31:05*Matthias247 joined #nimrod
16:35:57*brson joined #nimrod
16:41:06*OffTheRails quit (Read error: No route to host)
17:00:17*Guest56053 quit (Ping timeout: 264 seconds)
17:08:34*vendethiel joined #nimrod
17:16:40*brson quit (Quit: leaving)
17:32:18*flaviu1 joined #nimrod
17:36:38*untitaker quit (Ping timeout: 240 seconds)
17:38:26flaviu1Varriount: How about typeclasses on NEP 1? I think they should be capitalized and without a prefix, but the stdlib currently says they should be lowercase
17:40:11flaviu1dom96: Your "Hacking the Nimrod compiler" tutorial is way out of date and unlikely to be useful, things have changed too much over the last three years.
17:43:04flaviu1dom96: Apparently I have deletion privalges on the wiki somehow. I had no idea, if you want to restore it, I've saved it in a gist https://gist.github.com/flaviut/2c1c24c8d90580ed5503
17:52:30*brson joined #nimrod
17:54:07Araqflaviu1: really? I thought it's only about adding a pragma
17:54:20Araqwhich didn't change in the past 3 years that much
17:54:44flaviu1Araq: Yes, isn't it recommended to use macros instead of messing with the insides of the compiler?
17:55:01AraqI agree typeclasses should be uppercased
17:55:13Araq(except of course that I think it shouldn't matter ... )
17:55:53Araqflaviu1: well I guess ...
17:56:03AraqI always hack the compiler though ...
17:56:33Araqbut this is because the macro API lacks lots of useful things
17:57:19flaviu1There is some useful information in it, so it might be better if I added it back and put a warning that it isn't the best way to add a pragma.
17:57:33AraqI could expose what I need but then I'm designing an API and not dealing with the actual problem anymore
17:57:41Araqand people tend to dislike my APIs :P
17:58:56Demos_I think windows.nim may be broken
17:59:04Demos_oh wait
17:59:33Demos_yeah
17:59:55Araqthe "cast" in VM bug?
17:59:58EXetoCok, any news regarding partial/full case sensitivity then, since we're discussing conventions?
18:00:16Demos_a lot of windows functions take a param of type UINT, which is NOT defined in windows.nim and so you get nimrod's uint. this is a 64-bit type, windows UINT is c's unsigned int
18:00:43EXetoCotherwise, let's just keep everything as it is, and then deal with that after we've decided and a migration tool has been released etc
18:01:31AraqEXetoC: I like to add FULLUPPER to --cs:partial ...
18:01:42Demos_I am struggleing with the typechecker saying "Error type mismatch: got (HWND, uint, LPARAM, LPARAM) but expected one of windows.DefWindowProc(wnd: HWND, msg: UINT, wp: WPARAM, lp: LPARAM): LRESULT
18:01:50Araqso FOOBAR is FOO_BAR but not FooBar ...
18:02:19Demos_Araq: is cs: partial global to the program, or can it be done per module
18:02:29Araqglobal to the program
18:03:51AraqEXetoC: there is no news really. I want the community to decide
18:03:57flaviu1dom96: I added it back with a disclaimer, I noticed several remarks are still useful, even if the tutorial itself is outdated
18:04:17Demos_oh it is because UINT is NOT exported from windows.nim
18:04:28Demos_also why the fuq is it UINT=int32
18:04:32Demos_that seems really wierd
18:04:35AraqDemos_: yeah yeah yeah we got it. please fix it
18:04:42Demos_OK on it
18:06:13EXetoCAraq: ok
18:11:18EXetoCcast[cstring] sure is a convenient way to access bytes compared to casting to ptr array
18:14:50EXetoCfowl: command-line reddit browser/monitor http://cortex.glacicle.org/
18:18:22*q66 joined #nimrod
18:18:22*q66 quit (Changing host)
18:18:22*q66 joined #nimrod
18:23:27AraqVarriount|Mobile: so we should check the win version at runtime and throw an exception if createSymLink is called?
18:27:10EXetoCwhy not when building the compiler?
18:29:44*untitaker joined #nimrod
18:32:09EXetoCcan it be queried then?
18:32:45Araqthat doesn't really work
18:32:51Araqwindows is not linux
18:33:13Araqpeople regularly build on vista for windows 8, or for win XP
18:33:40flaviu1Araq: You need a special toolchain IIRC to build for XP
18:33:53Araqno
18:34:53flaviu1You do with VC++ 2012, not with C, you're right
18:35:05Araqflaviu1: we have at least 2 people running 0.9.4 on win xp, check the forum
18:35:29Araqthe only problem is the new sym link feature in os.nim that requires vista
18:35:30EXetoCyou need some app just to check the version? oh well
18:35:38EXetoCthird-party app I mean
18:35:49AraqEXetoC: no? it's a win API call
18:36:31EXetoCyes but you implied that there's no portable app for querying that information
18:36:39EXetoCthat comes with windows
18:37:18Araqno, that's not at all what I'm talking about
18:38:34*Johz joined #nimrod
18:43:46EXetoCright
18:52:53Demos_what is app:gui supposed to do? should it like with user32?
18:54:19EXetoCno console window?
18:56:23*brson quit (Quit: Lost terminal)
18:56:40*brson joined #nimrod
19:03:59*hoverbear quit ()
19:12:39*Jesin quit (Quit: Leaving)
19:22:22*brson quit (Quit: leaving)
19:33:24Araqhmm
19:34:11Araqsometimes even I forget about the usefulness of templates
19:34:35Araq template rec(a,b): expr = ple(m,a,b) == impYes
19:34:52Araqand now I can use 'rec' for any recursive call
19:35:04Araqthis will be a new idiom in the compiler
19:37:31EXetoCright, type constraints have no effect still. that's not too good
19:38:48Araqwhat do you mean?
19:39:14EXetoCT:SomeTypeClass
19:39:50EXetoCI think you said that zahary broke it
19:40:06AraqI fixed the reported bugs
19:40:09Araqabout it
19:40:19EXetoCodd
19:41:33EXetoC"proc addMap[T:uint16|uint32](o: var TMsgPack, len: T) ... m.addMap(1)". T is now int
19:45:47Araqoh yeah, I think I know where that comes from ... -.-
19:46:02*Varriount|Mobile quit (Ping timeout: 255 seconds)
19:53:12*silven quit (Remote host closed the connection)
19:59:22*ehaliewicz joined #nimrod
20:07:37*hoverbear joined #nimrod
20:18:43*ehaliewi` joined #nimrod
20:19:57*ehaliewi` is now known as ehaliewicz`
20:20:39*ehaliewicz quit (Ping timeout: 240 seconds)
20:27:53*brson joined #nimrod
20:28:32*nande_ joined #nimrod
20:28:52flaviu1/irchelp/
20:29:14flaviu1sorry, it passed through for some reason
20:32:39*brson quit (Ping timeout: 240 seconds)
20:34:33*flaviu1 is now known as flaviut
20:35:53*flaviut is now known as flaviu1
20:38:09*brson joined #nimrod
20:42:30*enurlyx joined #nimrod
20:48:58EXetoCAraq: it's an int literal thing, right?
20:51:30EXetoCI guess
20:52:31EXetoCreporting it soon
20:52:55*freezerburnv quit (Quit: freezerburnv)
20:53:27Araqyup, thanks
21:01:15enurlyxDemos_: also why the fuq is it UINT=int32
21:01:21enurlyxI think it is correct
21:01:41enurlyxSee: http://msdn.microsoft.com/en-us/library/windows/desktop/aa383751%28v=vs.85%29.aspx#UINT
21:01:51enurlyxAnd: http://msdn.microsoft.com/en-us/library/s3f49ktz.aspx
21:04:26NimBotnimrod-code/babel master 83d816e Grzegorz Adam Hankiewicz [+0 ±1 -0]: Updates version numbers for required nimrod compiler.
21:04:26NimBotnimrod-code/babel master 5322268 Dominik Picheta [+0 ±1 -0]: Merge pull request #40 from gradha/pr_updates_readme... 2 more lines
21:18:02*Matthias247 quit (Read error: Connection reset by peer)
21:21:21Demos_enurlyx: yeah it is the right size
21:22:31enurlyxDemos_: I answered some of your questions yesterday. Maybe you can check the logs
21:22:40enurlyxWhat wrappers are you working on?
21:22:52Demos_directX
21:23:43enurlyxhow do you handle enumerations if there are two entrys with the same value?
21:23:46enurlyxIgnore one?
21:24:01Demos_I split them into two enumerations
21:24:46Demos_the wrapper is "done" but I have not tested it, I am having trouble createing a render target for some reason
21:25:20enurlyxDefWindowProc?
21:25:23Demos_no
21:25:27Demos_I got that working
21:26:09dom96Varriount: Can't you have multiple handles to the same directory?
21:27:17enurlyxDemos_: How did you handle the dependencies? Wrap them also or did you only pick the needed pieces?
21:27:38Demos_which dependencies, I wrapped dxgi as well
21:27:53Demos_I have not wrapped rpc.h though
21:28:09enurlyxah yes. rpc is where i stopped
21:28:48enurlyxWould a idl2nim help?
21:29:07flaviu1Araq: There's a backlog of pull requests. Looking over them, #1207, #1206, #1191, #1169, #1114, #1210 are non-major changes and can probably be merged without much hassle. I'd also like my own #1204 (`` cgen), and #1190 (better int parsing) merged, although you might want to hold off on #1190 until I write a few tests
21:29:32Demos_yes, however looking that the dx idl files it looks like they go and emit a lot of C code from strings
21:29:48Demos_I do not think it is really worth it to write an idl2nim generator
21:29:55flaviu1dom96: What do you think about PR #1209?
21:30:51Demos_do you really need rpc.h? I don't think anyone uses the really insane COM "features" with directX
21:32:02enurlyxMh not really at the moment. I just have no experience in wrapping things and was thinking that it would make sense to wrap all dependent parts.
21:32:04dom96flaviu1: I think fowl shouldn't simply remove that, but loosen it a bit.
21:32:26enurlyxAt the moment i want to have direct2d and directwrite. I want to do some gui thinks.
21:32:55flaviu1dom96: Post a comment then, he can't read your mind :P
21:33:00Demos_also rpc seems to be almost all macros
21:33:22dom96flaviu1: I know. I was busy. And you asked me what I thought so I told you.
21:33:27Demos_and d2d needs a D3DDevice I think
21:35:53*boyd joined #nimrod
21:35:53*flaviu1 quit (Read error: Connection reset by peer)
21:36:14enurlyxYes it builds on direct3d
21:36:17*boyd is now known as Guest86454
21:36:57enurlyxSeems microsoft deprecated gdi+
21:37:35Demos_yeah, but seeing as I can not even create a render target a d2d wrapper would not be useful
21:37:52Demos_d2d is pretty cool in that it can SHARE a swap chain with d3d
21:37:55*flaviu1 joined #nimrod
21:38:07Demos_as in the very same memory is shared between d3d, d2d and dwm
21:38:27VarriountMeep. I'm here.
21:38:32Araqmeh, I'm not the only one with write access
21:39:21Guest86454Can anyone point me to an example using marshal and a table? I can't seem to figure out how to properly initialize it / not get a segfault.
21:40:15*flaviu1 quit (Read error: Connection reset by peer)
21:41:00*flaviu1 joined #nimrod
21:41:23Guest86454Ok... fixed. Apparently an invalid filename to newFileStream won't cause an exception and thus the segfault.
21:42:41VarriountGuest86454: Could you report that as a bug?
21:43:56Guest86454Will do.
21:45:40enurlyxDemos_: For the export of UINT i made a PR some time ago
21:45:53Demos_I decided not to export it
21:46:08Demos_but I did change it to int32
21:46:28enurlyxalso works
21:46:31Demos_if it is exported than you would have to qualify int all the time
21:46:36enurlyxi use windows.UINT
21:46:47Demos_that will not work if it is not exported
21:46:58Demos_and the issue is more with accessing in
21:47:00Demos_*int
21:47:03enurlyxyes. i do #import windows except UINT
21:47:17Demos_heck I dont even know if system.int will work
21:47:45flaviu1Varriount: Can you go through the pull requests and merge the ones I've pointed out?
21:48:02Varriountflaviu1: Which ones have you pointed out?
21:48:06enurlyxThe idea was to rename UINT to WINUINT :D
21:48:35enurlyxBut just using int32 would work nice
21:48:43flaviu1Varriount: #1207, #1206, #1191, #1169, #1114, #1210, #1204
21:49:02VarriountWow, 25 pull requests. *cracks fingers*
21:49:11VarriountIt's time to *review*
21:50:52flaviu1It turns out the only way to get a list of collaborators is through https://api.github.com/repos/Araq/Nimrod/collaborators, github doesn't seem to expose it on the page
21:50:56enurlyxDemos_: Did you upload your wrappers somewhere?
21:51:42VarriountHm. Perhaps we should have some sort of official pull request merging practice.
21:52:02VarriountLike, where a pull request is marked ok by two people (isn't that how BSD does things?)
21:53:01Demos_https://bitbucket.org/barcharcraz/directnimrod
21:53:17EXetoCnot using aliases is better
21:53:34NimBotAraq/Nimrod devel e59ba5d Charlie [+0 ±1 -0]: added note about `$` for cstrings to the manual
21:53:34NimBotAraq/Nimrod devel 6b6262b Varriount [+0 ±1 -0]: Merge pull request #1207 from barcharcraz/cstringManual... 2 more lines
21:53:57Demos_EXetoC: clearly you have never seen any windows code :D
21:54:08EXetoC?
21:54:22Demos_In principle I agree, but when wrapping a windows header it is MUCH eaiser if the windows typedefs work
21:54:24NimBotAraq/Nimrod devel 8a183da Charlie [+0 ±1 -0]: added random(max: float): float support to windows
21:54:24NimBotAraq/Nimrod devel d32b427 Charlie [+0 ±1 -0]: added a note about random(max: float): float's resolution on windows
21:54:24NimBotAraq/Nimrod devel 2fa739a Varriount [+0 ±1 -0]: Merge pull request #1206 from barcharcraz/windowsRandom... 2 more lines
21:54:47enurlyxI like to stay with the official types as close as possible.
21:54:53Demos_so do I
21:54:56enurlyxIts easyer to use documentation
21:54:56NimBotAraq/Nimrod devel 9428bed Simon Hafner [+0 ±2 -0]: Fixes #1168
21:54:56NimBotAraq/Nimrod devel c40f2d6 Varriount [+0 ±2 -0]: Merge pull request #1169 from Araq/sigpipe... 2 more lines
21:55:07Demos_and eaiser to wrap in the first place
21:55:08NimBotAraq/Nimrod devel 35e603b Nick Greenfield [+0 ±1 -0]: Fix nimprof import error when --threads:on.
21:55:08NimBotAraq/Nimrod devel f05ba51 Varriount [+0 ±1 -0]: Merge pull request #1210 from boydgreenfield/nimprof-lock-fix... 2 more lines
21:55:42enurlyxyes
21:55:44Varriountflaviu1: You've made sure https://github.com/Araq/Nimrod/pull/1204/files works, right?
21:55:45Demos_anyway I gotta go, checkout swappingTest.nim if you think you can figure out what is preventing CreateRenderTargetView from working
21:55:52EXetoCsearch/replace?
21:55:58enurlyxThanks. i go too.
21:56:32flaviu1Varriount: Yes, it does. I also made a test, passes
21:56:39*enurlyx quit (Quit: Nettalk6 - www.ntalk.de)
21:56:44NimBotAraq/Nimrod devel a568c61 flaviut [+0 ±3 -0]: Make codegen for `1` and similar valid
21:56:44NimBotAraq/Nimrod devel 2026137 flaviut [+1 ±0 -0]: Add test for #1081
21:56:44NimBotAraq/Nimrod devel e30294c flaviut [+0 ±5 -0]: Merge https://github.com/Araq/Nimrod into quotedidentescape
21:56:44NimBotAraq/Nimrod devel 8581fb7 Varriount [+1 ±3 -0]: Merge pull request #1204 from flaviut/quotedidentescape... 2 more lines
21:56:56NimBotAraq/Nimrod devel a16f762 Grzegorz Adam Hankiewicz [+0 ±1 -1]: Moves abstypes content into manual.
21:56:56NimBotAraq/Nimrod devel 2232c18 Varriount [+0 ±1 -1]: Merge pull request #1191 from gradha/pr_removes_abstypes... 2 more lines
21:57:07*vegai quit (Ping timeout: 252 seconds)
21:57:14*vegai joined #nimrod
21:58:23flaviu1Varriount: Also close #1125 without merging, it does almost the same thing that mine does, but it does it to every identifier, which Araq doesn't really like
21:59:30Araqyeah "doesn't really like". Years of experience tell me it WILL break code. Stop pretending I'm subjective please.
22:01:35VarriountAll hail the grand leaders ineffable wisdom!
22:01:39Varriount*leader's
22:02:32Araqeven the stuff that I review and merge breaks code
22:03:28flaviu1Araq: Sorry, poor wording. I understand and agree with why you didn't merge it.
22:03:36AraqVarriount's additions break things for win xp, gradha's docgen improvements broke some things, my new VM broke lots of code.
22:04:12Araq0.9.4 wasn't generated for powerpc 32bit and people complained
22:05:28Araqnobody uses Nimrod, but those who do use it on every OS out there ... ;-)
22:05:55EXetoCmy language will be linux only :p
22:06:37flaviu1Araq: Is my pull request on removing references to LLVM ok, or would you rather me close it since it doesn't really add anything new?
22:07:16*Varriount_ joined #nimrod
22:07:29*Varriount quit (Ping timeout: 264 seconds)
22:07:34*Varriount_ is now known as Varriount
22:08:01Araqflaviu1: don't know, didn't review it, I guess it's just fine
22:10:48*Guest86454 quit (Ping timeout: 240 seconds)
22:11:48EXetoCok other platforms too upon request
22:13:38AraqEXetoC: I request a port for SurrealUnix
22:13:57Araqit's my favourite OS
22:14:08VarriountDon't forget TempleOS
22:14:36flaviu1EXetoC: TempleOS means you need another backend for HolyC
22:14:39*brson quit (Ping timeout: 240 seconds)
22:14:52Araqthey took the core of BSD and replaced the B with an L
22:15:27VarriountAraq: Stop making me laugh so hard I choke on my food.
22:15:37Araqalso they truely follow the standards, when you trigger undefined behaviour, your computer literally explodes
22:17:26flaviu1EXetoC: I also want to use it on Parallella, so make sure that it compiles to Forth too
22:17:50NimBotAraq/Nimrod devel c210e12 boydgreenfield [+0 ±1 -0]: Clarify newFileSize & mappedSize params in memfiles.open() docs
22:17:50NimBotAraq/Nimrod devel a309a5f boydgreenfield [+0 ±1 -0]: Update posix open() call to incl. permissions... 2 more lines
22:17:50NimBotAraq/Nimrod devel 876cad3 boydgreenfield [+0 ±1 -0]: Fix missing import in nimprof.nim when --threads:on
22:17:50NimBotAraq/Nimrod devel fd352cc boydgreenfield [+0 ±1 -0]: Revert 876cad3a - making nimprof fix on a new branch
22:17:50NimBot1 more commits.
22:18:36Varriountflaviu1: forth is a very... interesting language.
22:19:08flaviu1Varriount: I know, but have you seen Parallella? It has 64 core with a target price of $99
22:21:43*brson joined #nimrod
22:24:07*JStoker quit (Ping timeout: 265 seconds)
22:25:00*JStoker joined #nimrod
22:25:12*OrionPK quit (Write error: Broken pipe)
22:31:20*TylerE_ joined #nimrod
22:33:01*armin1 joined #nimrod
22:44:32*NimBot joined #nimrod
22:49:33*JStoker quit (*.net *.split)
22:49:34*noam quit (*.net *.split)
22:51:31*JStoker joined #nimrod
22:54:46*Xuerian quit (Excess Flood)
22:55:45*Johz quit (Quit: Leaving)
22:56:53*Xuerian joined #nimrod
23:00:36*armin1 is now known as reloc0
23:02:59Demos_OK back
23:04:20Demos_we should support PonyOS (http://www.ponyos.org/)
23:06:19*brson quit (Ping timeout: 240 seconds)
23:08:01flaviu1Haha, it actually runs its own kernel
23:08:54*brson joined #nimrod
23:13:55*noam joined #nimrod
23:17:25*darkf joined #nimrod
23:18:26VarriountDemos_: DId you know that VS has a way to see cache misses?
23:19:08*io2 quit (Quit: ...take irc away, what are you? genius, billionaire, playboy, philanthropist)
23:19:14Demos_Varriount: I did not
23:19:30Demos_is it just part of the profiler?
23:19:41Demos_I thought the profiler was purely software
23:19:56VarriountDemos_: There are special registers for profiling
23:20:02VarriountLemme dig up the link
23:20:29flaviu1Demos_: http://www.slideshare.net/kulov/demystifying-visual-studio-2012-performance-tools
23:20:31Demos_I know hardware performence stuff exists on x86, I did not know VS used it
23:21:57VarriountDemos_: http://msdn.microsoft.com/en-us/library/bb385772.aspx
23:22:55VarriountI was reading through reddit yesterday, and found that comment, and thought "That sounds like something Demos_ would appreciate"
23:23:07Varriount*found a comment linking to that page
23:23:44Demos_heh, I have used vtune to do that kind of stuff, and it has a totally obscene array of data it can show you
23:23:57Varriountvtune?
23:24:03Demos_intel's profiler
23:25:13flaviu1"From $899" :( Looks really nice though
23:25:27VarriountOk, all that's left to do for file watching is implementing filtering.
23:25:52flaviu1Varriount: I'm curious why you had to lock files if you're implementing watching
23:26:03Varriountflaviu1: I don't
23:26:19VarriountWhere did I say I was going to lock files?
23:26:48flaviu1You asked about locking behavior on posix, I guess that was for something else
23:27:20Varriountflaviu1: I was probably just curious about how one would lock a file.
23:28:44VarriountOk, implementation question: If a user watches a file named "file.nim" and that file is renamed/moved, should the watch transfer to watching the new file, or keep monitoring the old file?
23:28:50VarriountEr, the old file na,e
23:28:56Varriount*name
23:29:38*hoverbear quit ()
23:29:55flaviu1Varriount: It shouldn't do anything fancy, just trigger a FileMoved or FileDeleted FileCreated event
23:30:24flaviu1Since any watching is likely to be performed on whole directories anyway
23:36:18njoejoeis there a plan to add a "list" command to ENDB which lists the lines of code around the current execution point? to aid in setting breakpoints. also ability to "watch" variables so their current value is shown for every line executed without having to add watchpoint pragma in the code?
23:37:03Demos_well ENDB does not really work right now
23:37:27Demos_just use --debugInfo --lineDir:on and debug using VisualStudio/windbg/gdb/lldb/whatever
23:38:26njoejoeDemos_: oh, didn't realize that. ok. I will try the gdb way then assuming that is the best for linux.
23:41:20Varriountflaviu1: Actually, on Windows, it's easier to monitor directories than individual files
23:41:40VarriountOr rather, the only way to monitor individual files is to monitor a directory and filter the results.
23:42:11flaviu1Not surprising, thats the most common usecase anyway
23:42:38VarriountAlso, what's odd is that the file monitoring mechanisms on Windows and Linux systems vary wildly in strengths and weakneses.
23:43:33VarriountLike, Linux's inotify allows individual file monitoring and more fine grained event notifications, but doesn't do well monitoring subdirectories containing lots of files.
23:43:35Demos_windows does not seem big on having a small API ....