<< 22-12-2013 >>
00:27:15ddl_smurfis that stack bottom thing fixed yet ?
00:29:12OrionPKMstack bottom girls make the world go 'round
00:29:17*io2 quit ()
00:37:14VarriountAraq, did that stack trace gathered from the segfaulting builder do anything to help?
00:41:38*psquid joined #nimrod
00:48:24*jenjimm quit (Ping timeout: 252 seconds)
00:59:21*jimmt joined #nimrod
01:09:58fowlhola
01:10:01*brson quit (Ping timeout: 250 seconds)
01:13:29Varriountfowl, hi
01:14:42*brson joined #nimrod
01:18:57Demoshey fowl, I looked at your entity code on github and it helped me. I thought you should know that your work helped someone make his code better
01:21:04fowlreally?
01:21:22OrionPKMlol
01:21:30OrionPKMfowl is surprised?
01:22:09fowlyes lol
01:29:59Demosyeah, you were storing stuff all in one big hunk of memory
01:30:20Demoswas looking to see how you dealt with construction, destruction and copying of things with pointers in em
01:30:51Demosalthough I kinda suspect that following a function pointer while doing a copy could really murder perf
01:32:35Demosanyway I was storing size and a type value inside the component, saw you doing it in another structure and found that was a better idea
01:35:01VarriountDemoes, fowl, entities as in entity processing systems?
01:35:32Demosentity component systems, no idea what an entity processing system is
01:35:47VarriountAh, yes, that 's what I meant
01:36:33VarriountDemos, fowl, like this -> http://gamadu.com/artemis/
01:37:42VarriountAlthough, come to think of it, nimrod's separation of types and the procedures that can act upon them seems like it would lend itself well the such paradigms
01:38:33DemosI am actually wrting this in c++, and I can not really use types much in the internals since my world would be a std::vector<std::vector<component>> and you can not have value types in something like that
01:38:51Varriount*shrug*
01:38:58DemosI need to use singletons (types not OOP globals) to make a container that holds whatever type you put in first
01:40:25fowlDemos, for constructors i allow each component to have an initializer, then the typeinfo's initializer is just a collection of component initializers that it runs on the entity, i never got around to destructors but they would work the same way
01:41:06fowlDemos, entitty is based off zahary 's design for a component system, he originally used c++ for it
01:42:51Demosyea, I store components seperately from entities themselves so I only need the one constructor, but yeah I do the same. I have a default one that calls ::new(pos) T() and pos->~T() and so on
02:38:59*MFlamer joined #nimrod
03:23:01*Demos quit (Quit: Konversation terminated!)
03:33:34*Mordecai joined #nimrod
03:34:59*psquid quit (Ping timeout: 246 seconds)
03:43:20*brson quit (Ping timeout: 252 seconds)
03:50:04*EXetoC quit (Quit: WeeChat 0.4.2)
04:33:36shodan45still trying to understand OOP stuff in nimrod... TObject is a built-in root-of-all-objects (-classes?)?
04:34:58*BitPuffin quit (Read error: Operation timed out)
04:36:24shodan45hm, that doesn't look right :/
04:37:42fowlno TObject is just a base object you can inherit from
04:37:57fowlyou can make your own base object using object{.inheritable.}
04:39:19shodan45fowl: ok, so what does TObject give you, then?
04:39:59fowla base type to inherit from?
04:42:35shodan45right, but it isn't required - you can just have "type TFoo = object" (right?)
04:43:32fowlyes but thats not inheritable unless you add {.inheritable.}
04:46:17shodan45ok, so then what's the difference between "TFoo = object of TObject" and "TFoo {.inheritable.} = object"?
04:47:10*brson joined #nimrod
04:47:15fowlthe difference is that in the second one TFoo doesnt inherit from TObject
04:48:48shodan45:)
04:49:15shodan45fowl: .... ok, but why would you want to do that?
04:51:58shodan45I'm trying to find where TObject is defined in the nimrod source, but not finding anything
04:53:19fowlsystem.nim probably
04:53:58fowlas for why use TObject or dont use it, its preferecne
04:59:12shodan45heh, github search for system.nim didn't find it... I guess it doesn't index file names? o_O
04:59:59fowlshodan45, its in nimrod's lib/
05:01:56shodan45fowl: yeah, I know, I was trying to use search as a faster way to get there
05:03:07shodan45and it looks like aside from {.inheritable.}, it also {.exportc: "TNimObject".}
05:03:19shodan45eh, I'll stop digging there :)
05:04:10*BitPuffin joined #nimrod
05:09:15*BitPuffin quit (Ping timeout: 272 seconds)
05:10:31shodan45fowl: thanks for the help
05:10:38fowlnp
05:10:40fowlgood night
05:15:54*xenagi joined #nimrod
05:29:20*MFlamer quit (Ping timeout: 240 seconds)
06:04:30*BitPuffin joined #nimrod
06:09:14*BitPuffin quit (Ping timeout: 264 seconds)
06:10:09*brson quit (Read error: Operation timed out)
06:12:48*brson joined #nimrod
06:17:23*xenagi quit (Quit: Leaving)
06:19:02*brson quit (Ping timeout: 240 seconds)
06:20:06*brson joined #nimrod
06:56:17*psquid joined #nimrod
06:57:42*Mordecai quit (Ping timeout: 240 seconds)
07:05:13*BitPuffin joined #nimrod
07:09:51*BitPuffin quit (Ping timeout: 240 seconds)
07:22:03*brson quit (Ping timeout: 240 seconds)
07:51:10*zahary quit (Read error: Connection reset by peer)
07:51:41*zahary joined #nimrod
08:05:57*BitPuffin joined #nimrod
08:10:47*BitPuffin quit (Ping timeout: 250 seconds)
08:35:49*zielmicha joined #nimrod
09:06:44*BitPuffin joined #nimrod
09:11:19*BitPuffin quit (Ping timeout: 246 seconds)
10:30:25*shodan45 quit (Quit: Konversation terminated!)
10:35:17Araqddl_smurf: I pushed my 'fix' for the stack bottom setting, but now it fails on my linux machine ... -.-
10:37:11ddl_smurfare there platforms where the stack grows upwards ?
10:37:22Araqyeah but they are not common
10:38:02*io2 joined #nimrod
10:40:08ddl_smurfi haven't been able to reproduce the corruption thing with anything else than nimbuild
10:40:37ddl_smurf(though admittedly, having to go into the C to "fix" the initStackBottom call is so tedious i didn't insist too much)
10:41:00Araqwell use master and it shouldn't be necessary anymore
10:41:20ddl_smurfhttps://github.com/Araq/Nimrod/commits/master ?
11:16:40*Mordecai joined #nimrod
11:18:46*psquid quit (Ping timeout: 272 seconds)
11:33:03*brihat quit (Quit: Leaving.)
11:34:00dom96hi
11:35:37*gradha joined #nimrod
11:35:40NimBotdom96/jester master 8d445b8 onionhammer [+0 ±1 -0]: Export strtabs from jester.nim
11:35:40NimBotdom96/jester master d405529 Dominik Picheta [+0 ±1 -0]: Merge pull request #7 from onionhammer/patch-1... 2 more lines
11:36:45gradhadom96: do you win when playing against Bitpuffin?
11:37:03dom96gradha: we play together
11:37:24gradhawhat do you play against then?
11:38:02dom96other random people
11:38:41gradhafascinating
11:40:35Araqddl_smurf: yes
11:43:20*brihat joined #nimrod
12:14:02*q66_ joined #nimrod
12:14:43*q66 quit (Disconnected by services)
12:14:45*q66_ is now known as q66
12:19:42*BitPuffin joined #nimrod
12:28:46dom96hello BitPuffin!
12:28:59dom96We should do some coding today instead of playing Dota 2 I think :P
12:29:26Araqindeed
12:29:31Araqdiff feature for nimbuild!
12:29:41Araqthe tester needs to know about babel!
12:30:17dom96That's a bit ambitious.
12:30:25dom96And the nimbuild corruption still hasn't been fixed...
12:33:07*ddl_smurf quit (Quit: ddl_smurf)
12:37:53Araqso? waiting for that fix isn't wise
12:45:43gradhagood luck
12:45:47*gradha quit (Quit: bbl, need to watch http://www.youtube.com/watch?v=dEf4PJZXBxA again)
13:15:53*Endy joined #nimrod
13:23:12*darkf quit (Quit: Leaving)
14:02:00*q66 quit (Ping timeout: 246 seconds)
14:02:22*q66 joined #nimrod
14:19:06BitPuffiney dom96!
14:19:10BitPuffinwanna play dota xD
14:19:12dom96hah
14:19:18dom96well...
14:20:06dom96I need to eat something first
14:20:15BitPuffinwell I can't right now anyway
14:20:19dom96ok
14:20:22dom96when can you?
14:20:23BitPuffinthere's too many people here
14:20:26BitPuffinhmm
14:20:28BitPuffina bit later
14:20:31BitPuffinnot way later I guess
14:20:34BitPuffinjust a bit xD
14:20:40dom96lol
14:20:41dom96ok
14:20:41BitPuffinpuffin
14:21:05BitPuffinAraq: wanna join us? Or are you still updating lol
15:08:49*[1]Endy joined #nimrod
15:12:19*Endy quit (Ping timeout: 260 seconds)
15:12:20*[1]Endy is now known as Endy
15:37:43BitPuffinis it even legal to make games that requires ~15 gb on the disk
15:40:02*Hannibal_Smith joined #nimrod
15:43:16*io2 quit ()
15:52:24*EXetoC joined #nimrod
16:07:45BitPuffindom96: are you ready?
16:07:50BitPuffindom96: wee yu aaat
16:19:45*Hannibal_Smith left #nimrod ("Sto andando via")
16:22:08dom96BitPuffin: I'm only making food now lol
16:22:19BitPuffindom96: damn it
16:28:30OrionPKMfuck yeah
16:28:34OrionPKMscrew sourcecode filters
16:28:34OrionPKMhttps://dl.dropboxusercontent.com/u/417554/jestertemplates.png
16:29:12dom96niceee
16:29:16dom96How does that work?
16:29:25OrionPKMmacro magic
16:29:43OrionPKMparses through the string at compile time looking for non-repeating $'s
16:30:48OrionPKM$a-z0-9_ are treated as variables, ${ independent expressions }, $( ) are simple statements with .'s or []'s
16:32:55dom96only problem is that it won't look as great on github.
16:33:10OrionPKMit'll still look better than a sourcecode filter :P
16:34:19dom96so where is the package? :P
16:34:51OrionPKMheh the template stuff is up on my onion-nimrod repo, but i'm not satisfied with the output yet
16:35:07OrionPKMthe indentation gets kinda fucked up still
16:35:14OrionPKMfor the rendered HTML/js/css/whatever
16:36:07OrionPKMhttps://github.com/onionhammer/onion-nimrod/tree/master/templates
16:36:11OrionPKMafk a bit
16:36:16BitPuffinmake your damn food dom96 !
16:36:17dom96I don't think you should worry about that too much.
16:36:24dom96BitPuffin: I'm eating right now!
16:44:25BitPuffindom96: eat faster!½
16:46:23fowlhola
16:47:08fowlOrionPKM, very cool
16:48:09OrionPKMthanks
16:51:35dom96BitPuffin: ok ok
16:51:38dom96BitPuffin: rebooting now
16:52:52OrionPKMcome on guys, less playing games, more making them
16:58:19BitPuffindom96: sec
16:58:30VarriountGood morning guys!
16:58:43fowli had a dream that i stole gta5 from someone at a store
16:59:18OrionPKMhaha, thanks fowl
16:59:21OrionPKMmorning varriount
17:01:06*gradha joined #nimrod
17:01:47BitPuffindom96: now!
17:15:57*Mordecai is now known as psquid
17:22:17*ics quit (Ping timeout: 272 seconds)
17:22:23OrionPKMholy shit
17:22:48OrionPKMmy company sent me some cookies and they're actually huge and giant and soft and delicious
17:24:42*ics joined #nimrod
17:25:03OrionPKMnow i have diabetes
17:34:04fowlhuge AND giant? :o
17:36:07*xenagi joined #nimrod
17:36:35*zielmicha quit (Ping timeout: 272 seconds)
17:51:03*zielmicha-cloud quit (Read error: Connection reset by peer)
17:51:21*zielmicha-cloud joined #nimrod
18:02:34OrionPKMyope. both
18:02:52OrionPKMhuge and ginormous
18:15:28*zielmicha-cloud quit (Read error: Connection reset by peer)
18:15:46*zielmicha-cloud joined #nimrod
18:26:15*zielmicha joined #nimrod
18:27:56Araqzielmicha: bootstrapping on win works like on lin
18:28:17zielmichaI finally managed to do it
18:28:22zielmichaafter installing virtio drivers
18:29:06Araqwhat?
18:32:10zielmichaI thought that bootstraping doesn't work, but Windows was just unbelievably slow on KVM without virtio
18:35:58Varriountzielmicha, if bootstrapping didn't work on windows, we wouldn't have windows builder.
18:36:03Varriount*builders
19:04:59OrionPKMcourse bootstrapping on windows works
19:05:12OrionPKM90% of the time i do dev im on windows
19:05:46OrionPKMvarriount forget about the source code filter stuff for the ST plugin
19:21:04AraqVarriount: your fix is wrong
19:21:17Araqjust noticed it :P
19:21:29Araq result = f.cFilename[0].int == dot
19:21:30Araq if result:
19:21:32Araq result = f.cFilename[1].int in {dot, 0}
19:21:33Araq if result:
19:21:35Araq result = f.cFilename[2].int == 0
19:21:53Araqbut if filename[1] == 0 you shouldn't touch filename[2]
19:29:55*zielmicha-cloud quit (Ping timeout: 260 seconds)
19:30:03*OrionPKM quit (Remote host closed the connection)
19:31:29*zielmicha-cloud joined #nimrod
19:38:22*OrionPKM joined #nimrod
19:38:36OrionPKMahh there we go
19:39:38OrionPKMtransitioned my irc app off of sourcecode filters :)
19:41:44zielmichagradha: did code you reported to have stopped working ever worked on Windows (https://github.com/Araq/Nimrod/issues/770)?
19:49:37gradhazielmicha: I don't use windows, don't think windows users have a shell anyway
19:57:24*zielmicha quit (Quit: Leaving.)
19:57:35*zielmicha joined #nimrod
20:06:41*Endy quit (Ping timeout: 272 seconds)
20:12:24VarriountAraq, you want me to fix it, or are you going to fix it?
20:19:47AraqVarriount: already fixed it
20:20:04Varriountgradha, cmd.exe
20:20:57Varriountgradha, As far as I know, neither Linux nor Windows is a kind of turtle.
20:22:32Araqzielmicha: I think it's perfectly fine to break "foo >file", it was never officially supported
20:28:18Araqhmm things like these really turn against us for the FFI support:
20:28:24Araqvar SIGINT {.importc: "SIGINT", nodecl.}: cint
20:51:49VarriountAraq, howso?
20:52:27Araqwell you can't import SIGINT from a DLL, it's a constant
20:52:44VarriountOh. Ouch.
20:53:44VarriountHm. I wonder why constants aren't included (besides the sake of efficiency/memory)
20:53:58Araqbecause they are #defines?
20:54:23VarriountOh. Right.
20:55:13gradhaobjc guys spend a good deal wrapping defines around "extern cons" variables so they can change the value of the constant at runtime, defines are rare
20:55:43Varriountgradha, doesn't that defy the idea of a "constant"?
20:55:59Araqomg objc guys are misguided :P
20:56:24Araqthe real solution is to notice constants affect binary compatibility and never ever change them
20:56:26gradhaVarriount: the alternative is directx style "please fill in this structure the sizeof whatever you think it should be"
20:56:43VarriountOr do they only change the constant once, at the beginning of program setup?
20:57:01Araqhow can you use those fake constants in a switch statement?
20:57:07Araqright you can't. well done
20:58:00gradhanotifications in objc are even better, you have to pack objects, and those are runtime NSString ones, so every time you have a notification you strcmp stuff rather than compare to some integer
20:58:31Araqnow of course that's not so bad in C as the switch statement is retarded anyway. Sucks for Nimrod though.
20:59:02gradhaI've always wondered why API designers fear integers so much
21:00:17VarriountI thought the only big difference between Nimrod's switch/case and C's was the in/ability to 'fall through' the statements.
21:01:22Araqnimrod's does exhaustive checking and has a sane syntax
21:05:26gradhaa dumptree of a comment shows CommentStmt without the data, does that mean you can't generate docstrings from a macro?
21:06:09AraqI dunno, it could also mean dumptree doesn't understand comments
21:08:34gradhainteresting, the docstrings for a module seem to be all global comments concatenated together
21:15:40*io2 joined #nimrod
21:15:42*io2 quit (Changing host)
21:15:42*io2 joined #nimrod
21:17:14VarriountAre ast trees passed to a macro immutable?
21:17:50Araqnope
21:18:17Araqimmutability tends to suck for construction
21:18:43VarriountHrm. Then why would adding a colonExpr node to a procedure's pragma node not show up?
21:19:03Araqtry with vm2 please
21:19:11Araqthe old vm sucks for macro evaluation
21:19:32VarriountAraq, when are you going to merge vm2?
21:20:07Araqwell type traits are still not on par with master
21:20:16VarriountAh
21:29:39VarriountAraq, is it me, or does vm2 call gcc differently from master?
21:30:23AraqI didn't notice any difference. what's different?
21:31:48VarriountNevermind, my error
21:32:51VarriountWtf? -> test.nim(28, 28) Error: invalid pragma: cimport: "GLEW"
21:33:38VarriountOh, importc, not cimport
21:39:09*zielmicha quit (Ping timeout: 240 seconds)
21:57:13OrionPKMvarriount do you mind if i delete the sourcecode filter stuff
21:57:24VarriountOrionPKM, not at all.
21:57:49VarriountOrionPKM, I must ask though, why?
21:59:55OrionPKMit just seems like more trouble than it's worth.. unless you want to keep plugging away on it; i have no intention of updating it
22:00:03OrionPKMhow far did you even get?
22:00:27VarriountOrionPKM, it works exactly as well as the regular syntax highlighting
22:00:52OrionPKMbut w/ a diff file extension?
22:01:20VarriountAnd the comment marks taken into account.
22:01:24OrionPKMdid you figure out a way to automatically update the YAML from the normal YAML?
22:01:46VarriountNot in any easy fashion
22:02:10OrionPKMmm, well, I've built something that fills the same kind of purpose as the source code filters
22:02:18OrionPKMand it's a lot easier to highlight
22:02:19OrionPKMhttps://dl.dropboxusercontent.com/u/417554/jestertemplates.png
22:03:18VarriountOrionPKM, 'tmpl'?
22:03:54OrionPKMtmpl and html are macros
22:04:20OrionPKMhtml/js/css/xml/glsl (so far) tell the syntax highlighter what syntax definition to use
22:04:53OrionPKMtmpl takes that 1 step further and breaks down the string into a series of statements
22:05:16fowlawesome work OrionPKM
22:05:46OrionPKMthe html/etc annotations also re-indent the input string at compile-time
22:06:30VarriountOrionPKM, and the syntax highlighting?
22:06:34Araqindeed impressive work, OrionPKM
22:06:48OrionPKMvarriount look at your repo
22:06:54OrionPKMthanks guys
22:07:10OrionPKMit's be nice if we had a way to automatically create the syntax highlighting for the sub-selectors
22:07:16VarriountOrionPKM, sorry I haven't worked on the highlighting in a bit
22:07:28OrionPKMas it is right now I have to copy-paste the YAML for each subselector I want to support
22:08:21OrionPKMtmpl isn't done yet, as I was telling dom earlier, the output indentation gets a little wacky
22:08:38OrionPKMthat only really hurts human readability though
22:08:40VarriountOrionPKM, remember what I told you about dynamically generating a new syntax file?
22:08:46OrionPKM:D
22:08:54OrionPKMthat YAML sucks?
22:08:55OrionPKMidk
22:09:16VarriountWell, that bit would be generating xml
22:09:34OrionPKMI was actually thinking about *using* tmpl to make the YAML :D
22:10:30VarriountOrionPKM, when you asked if I had figured out a way to automatically update the yaml, I wasn't talking about the bit concerning syntax file generation and the other language
22:10:47OrionPKMhttps://github.com/Varriount/Nimrod-Sublime/blob/master/nimrod.YAML-tmLanguage#L107
22:12:36VarriountOrionPKM, as soon as I get my stuff for os.nim finished, Ill start work again on the syntax extension.
22:12:45OrionPKMcool
22:14:15OrionPKMi want to add in sublimeREPL support
22:17:42gradhaAraq: do you mind if I replace the hardcoded rstgen.defaultConfig with a macro which reads config/nimdoc.cfg?
22:18:49Araqyes I do
22:18:52*boydgreenfield joined #nimrod
22:20:08gradhaok, won't do it then
22:20:18Araqbtw you made highlite.nim not read doc/keywords.txt for the same reasons
22:21:33gradhaah, the relative path slurp bug
22:23:44AraqI'm not even sure it's a bug
22:24:24Araqslurp is not like import but more like readFile; readFile doesn't give a fuck about the compiler's notion of the current directory either
22:25:11Araqbut I guess people think it's more like import ...
22:25:38gradhaif you want people using static compilation there has to be a balance between not giving a fuck and having 3rd party library imports work
22:26:51gradhaslurp could maintain the current behaviour, and use a prefix like "$mod" and "$pro" for paths, which would be replaced by the absolute path to the imported file and project
22:26:56*ics quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
22:27:11gradhathen people could write slurp"$mod/myfile.txt" and be sure it works
22:27:45Araqsure and then the VM that also supports 'slurp' depends on the compiler's path notion
22:27:58Araqbut I guess the VM already depends on that anyway. *sigh*
22:29:40gradhaI'm not so interested in $pro, but a $mod with the absolute path of the improted file being processed would help a lot in this case, does that break too much?
22:30:12gradhait seems like the compiler importing the module could define a constant "module_path" or something
22:30:26gradhaand that would wonderfully break includes too
22:30:29Araqthat sounds like a recipe for desaster
22:30:42Araqabsolute paths are not the solution
22:35:01*Mordecai joined #nimrod
22:37:00*psquid quit (Disconnected by services)
22:37:04*Mordecai is now known as psquid
22:37:24*ddl_smurf joined #nimrod
22:41:24gradhathe problem of bundling resources along code is not something new, and for some reason macosx seems to have solved it with bundles but the rest of the world refuses to catch on
22:42:31Araqwhenever I have an icon on some .exe it's some working resource file mechanism going on
22:42:52Araqso exlcude windows from your rest of the world :P
22:43:35gradhaI'm talking about 3rd party libraries, not binaries
22:44:20gradhabundles for libraries include the .a and also other resources loaded at runtime, the toolchain knows how to deal with that and include it into your program
22:44:51gradhathe directory subdivision allows a library to have a Localized.strings and your main program also a Localized.strings, and they don't fight each other
22:46:57gradhaso it's like a .dll + jpg + txt + whatever, and it goes in a single package for convenience
22:47:30gradhaof course exes are wonderful, we still celebrate their naming convention
22:48:34gradhabtw, rst* modules won't go into babel because it's used by the compiler, right?
22:49:46Araqright
22:52:17gradhankCommentStmt are weird, no children, apparently no other useful values, and macros don't seem to run for "nimrod doc" anyway
22:54:02Araqyes
22:54:10Araqthat's why we have doc2 ...
22:54:49fowlgradha, it should have a strval with the comment in it
22:55:42gradhaError: field 'strVal' cannot be found
22:57:51Araqno, comments are special
22:58:24AraqI think macros.nim supports getting the comment value though
23:12:24gradhaoh, I know how to solve the 3rd party module static resource problem
23:12:43gradhathe module uses a prefix for slurps
23:12:54gradhaunless a specific constant is defined, this prefix is the empty string
23:13:11Araq?
23:13:16gradhawhen the constant is defined, is used as the prefix path
23:13:34gradhaso you can pass `babel path module` as the constant during compilation
23:13:59Varriountgradha, what if you have multiple 3rd party static modules?
23:14:23*io2 quit ()
23:14:23*darkf joined #nimrod
23:14:23gradhaeach can use a different set of constants
23:14:32gradha-d:genieosRsrPath
23:14:44gradha-d:fowltekRsrPath
23:14:48gradhaetc
23:15:01Araqgradha is kidding obviously
23:15:40fowldocumentation is for newbs
23:15:53fowlif you want to see how to use it, look at the source code
23:16:22VarriountSince I do not have any direct experience with this kind of problem, all I can say is these: Find out the various ways that this problem has been solved before, list the pro's and con's of each, pick the best, and improve upon it.
23:17:06Araqalright, lets look at the languages which support 'slurp' at compile time. any suggestions?
23:17:11*boydgreenfield quit (Quit: boydgreenfield)
23:17:12VarriountTo many problems of these kinds of problems are solved by 'new' solutions which completely miss taking into account their previous solutions.
23:18:42gradhaVarriount: don't worry, Araq is right, I'm kidding, I don't have any of these problems I just invented. Now, if you could please stare into this light for a moment… thank you.
23:18:52AraqVarriount: that's a good point. However as long as all this knowledge doesn't end up as rants on reddit it's useless knowledge as it doesn't spread.
23:19:02dom96I wonder why unicode exception messages are so messed up on Windows.
23:19:13dom96Is that worth a bug report?
23:19:24Varriountdom96, howso? (I haven't encountered them)
23:19:42dom96or maybe cmd.exe can't display unicode.
23:20:00Araqgradha: please fix 'slurp' so that it behaves like 'import' and document it.
23:23:04dom96yeah, it's cmd.exe's fault.
23:23:20dom96Have to change to unicode with 'chcp 65001'
23:23:30Varriountdom96, I think you have to enable command extensions
23:23:43Araquse a real OS. I heard Haiku is a good choice.
23:24:10Varriounthttp://stackoverflow.com/questions/14109024/how-to-make-unicode-charset-in-cmd-exe-by-default
23:24:23Araqit doesn't run any software, that too solves lots of problems
23:26:57VarriountAraq, it's really fun when you try to write a raw stream of random data to stdout in windows. You get fun symbols
23:28:55Araqyeah but that's your problem. should have used stderr instead.
23:29:54AraqI suggested 'stdbanana' but got outvoted
23:30:06Varriount:3
23:30:33Araqat least they picked my suggestion to use "kill" for "send message"
23:30:41NimBotnimrod-code/babel master 422f247 Dominik Picheta [+0 ±1 -0]: Git checkout is now forced.... 4 more lines
23:32:39VarriountWait, what?
23:34:57Araqwell in my dreams I helped them to design unix
23:35:09dom96you mean nightmares right?
23:35:36Araqno
23:36:05gradhastderr and stdout screams for a stdnormalin and stderrin, because you want to pipe between programs, the cornerstone of unix
23:36:43Araqunix is a very nice OS when you have no idea what the word "operating" means
23:38:29gradhadom96: doesn't chcp have a "chcp unicode" alias?
23:38:43dom96no idea
23:48:19brihatI was looking at the doc page of "channels", but it doesn't mention how to create a new TChannel. Is there a spwan() function or somweething?
23:48:53Araqthere is actors.spawn
23:49:10Araqyou need to open a channel
23:49:14fowlbrihat, declare it, then open()
23:49:35brihatah, i see
23:54:37VarriountGah. I wish there was some way to magically enhance point and click games made 15+years ago.
23:55:13brihatthere are two 'nimrod' executables. one in compiler/ and another in bin/ . Was this changed recently or what?
23:55:51Varriountbrihat, the one in compiler is the first version generated by koch boot
23:56:20brihathow do i make it replace the on in bin/ ?
23:56:41Varriountbrihat, it does, after the boot process is done.
23:57:24Varriountbrihat, koch generates two versions of nimrod in succession, in order to make sure the compilation process is generating the correct code.
23:58:07VarriountThe final built executable ends in bin, the temporary versions up in compiler/
23:58:18brihatok, i see now.. the last line says "executables are equal: SUCCESS!"