<<16-08-2012>>
00:54:00*Trixar_za is now known as Trix[a]r_za
01:06:52*q66 quit (Quit: Quit)
03:27:49*fowl joined #nimrod
03:29:01*dom96 quit (*.net *.split)
03:29:01*mal`` quit (*.net *.split)
03:34:08*dom96 joined #nimrod
03:34:08*mal`` joined #nimrod
03:43:18*zahary1 joined #nimrod
03:43:21*zahary quit (Read error: Connection reset by peer)
03:47:37*zahary1 quit (Ping timeout: 240 seconds)
03:47:43*zahary joined #nimrod
07:09:28*XAMPP quit (Ping timeout: 246 seconds)
07:09:45*Trix[a]r_za is now known as Trixar_za
07:44:37*Araq_ joined #nimrod
07:54:36*reactormonk quit (Ping timeout: 268 seconds)
08:09:08*Araq_ quit (Quit: ChatZilla 0.9.88.2 [Firefox 14.0.1/20120713134347])
09:28:33*q66 joined #nimrod
10:04:30shevyAraq, seen this http://tomasp.net/blog/type-first-development.aspx ?
10:10:36*Trixar_za is now known as Trix[a]r_za
10:42:52*Araq_ joined #nimrod
11:22:24Araq_shevy: I always use TFD :P
11:22:40Araq_but I had no name for that :-)
11:34:03*fowl quit (Quit: Leaving)
11:49:12*Araq_ quit (Quit: ChatZilla 0.9.88.2 [Firefox 14.0.1/20120713134347])
12:06:49*Amrykid quit (Ping timeout: 272 seconds)
12:06:49*Araq quit (Ping timeout: 272 seconds)
12:07:51*Araq_bnc joined #nimrod
12:08:21*Amrykid joined #nimrod
13:32:54*fowl joined #nimrod
14:22:53*XAMPP joined #nimrod
14:31:05*Amrykid quit (Changing host)
14:31:05*Amrykid joined #nimrod
14:55:22*zahary1 joined #nimrod
14:55:54*zahary quit (Read error: No route to host)
15:01:17*zahary1 quit (Read error: Connection reset by peer)
15:01:30*zahary joined #nimrod
15:09:15*reactormonk joined #nimrod
15:11:11*|apriori| joined #nimrod
15:51:11*Araq_bnc is now known as Araq
15:56:09*shevy quit (Ping timeout: 244 seconds)
16:00:09Araqso yes
16:00:15AraqI broke bootstrapping
16:00:17Araq:P
16:02:45fowlhey Araq what do you think: https://gist.github.com/3342882
16:08:11*shevy joined #nimrod
16:12:29Araqfowl: use 'varargs' instead of 'openarray' :D
16:12:41Araq':=' --> meh ...
16:12:50Araqand find a better name for 'und'
16:13:07Araqin fact, I guess macros.add could return the father
16:13:12Araqand be 'discardable'
16:14:23Araqbtw does your code actually work?
16:14:54fowlno
16:15:34fowlit derps on line 115 which is result.add(the type section)
16:16:44fowlError: internal error: semGenericParamList
16:23:09Araqhurray ...
16:23:25AraqI guess nnkNilLit is evil :D
16:23:38Araqand maps to some real 'nil' perhaps
16:23:55Araqor hrm
16:24:04Araqmaybe you construct an invalid AST ...
16:25:55fowltype XXX = tuple[..] should be just like that
16:26:10fowli dunno what the middle nil is
16:37:02Araqwell it's for generic params
16:37:13Araqtype XXX[T] = tuple[..]
16:37:23Araqyou need to use nnkEmpty, not nil
16:37:29Araqfor the missing [T]
16:40:40fowlah ok
16:48:27Araqhttp://www.reddit.com/r/programming/comments/xufn3/alan_kay_normal_considered_harmful/
16:48:42Araq"So I've tried this Lively Kernel he talks about and it's even shittier than I imagined." lol
16:48:48*Araq doesn't like Alan Kay
16:57:18fowli dont really understand lively kernel, draggable widgets dont excite me i guess
17:06:40Araqone simply can't write software like "biological beings" work
17:06:54Araqwhich is always what Alan Kay seems to imply
17:07:18Araqhis way of "software engineering" doesn't work and no other field of engineering works like that IMHO
17:07:33|apriori|Araq: a short resumé about the topic plz
17:08:27AraqAlan Kay rants about industry I guess
17:08:52AraqI can't really tell as I stopped listening to his BS years ago :P
17:14:51fowlcool
17:14:54fowlit works now :>
17:21:32*zahary quit (Read error: Operation timed out)
17:23:11*zahary joined #nimrod
17:24:02Araqin fact, why he even invented smalltalk when knowing lisp is beyond me
17:24:43Araqbut iirc he used smalltalk to teach children programming ...
17:25:24fowlthe gui would be cool if it was usable in any way
17:25:35fowlnot all archaic
17:26:28AraqI have to go see you later
17:44:14dom96hello
17:47:29fowlsup
17:48:15dom96not much not much, just got back from work.
17:48:16dom96You?
17:49:26fowllooking for something cool to do so i can say im doing that
17:49:28fowl>_>
17:51:07dom96Make a robot using Nimrod
17:51:09dom96Send it to Mars.
17:55:59dom96I'm always too tired to work on Nimrod after work :(
19:36:06*|apriori| quit (Remote host closed the connection)
20:32:23shevyhmm
20:33:31shevyAraq Alan Kay stated that the central point about OOP are messages sent to objects. I don't know a single lisp implementation that really is OOP to this principle
20:34:08shevywhy use nimrod when one can have lisp!!!!
20:34:09shevy;P
20:34:30shevythe game haxima used to be written in scheme, it was quite interesting to read all the world files
20:34:37shevythe guy ~1 year ago switched to python
20:34:41shevysadly, stopped his work mostly :(
20:34:59shevy http://myweb.cableone.net/gmcnutt/nazghul.html
20:35:30shevy https://github.com/gmcnutt/hax2
20:39:41*XAMPP_ joined #nimrod
20:42:58*XAMPP quit (Ping timeout: 268 seconds)
20:44:27dom96hey shevy
20:44:31*reactormonk quit (Ping timeout: 252 seconds)
20:45:05fowldom96: hey try listening to this while you code i've been pretty productive on it :p http://www.youtube.com/watch?v=lP0D5u3EMfA
20:45:27dom96wow, 5 hours cool
20:46:07dom96Awesome. Thanks.
20:46:24shevydom96 teach me!
20:46:31shevyhehe
20:46:38dom96shevy: Try writing something!
20:46:45shevyI've mostly been trying to continue rewriting my largest ruby project
20:46:56dom96Start a new project in Nimrod!
20:46:59dom96Come on!
20:47:11fowlshevy: look at this its a giant macro that builds packing/unpacking functions: https://gist.github.com/3342882
20:48:05shevydom96 hmm
20:48:33dom96fowl: That is pretty cool
20:48:47shevydom96 I first need to be able to walk before I can run!
20:49:07dom96shevy: To learn how to program you need to actually program.
20:49:11shevyomg fowl what's this do?
20:49:16dom96So that you can remember it all
20:49:25dom96Having me tell you small things all the time won't work
20:49:31shevyhehe ok
20:52:11dom96shevy: So, what are you going to write?
20:53:04shevyI have absolutely no idea
20:53:11shevymost things seem to require a lot of knowledge
20:53:21shevyhmmm
20:53:21dom96IRC bot!
20:53:25shevyhehehe
20:53:29shevyI did this when I came to ruby
20:53:37shevyI dont want to re-do the same things again and again :\
20:53:43shevyperhaps something GUI related would be nice
20:55:06shevyin gtk :D
20:55:07dom96Make an IRC client :P
20:55:11shevylol
20:55:22dom96A simple one shouldn't be that difficult.
20:55:29shevyhmm
20:55:39dom96Well GTK will be a bitch
20:55:41shevyis there a nimrod IRC GUI client yet?
20:55:47dom96no
20:56:03dom96or not as far as I know at least
20:56:03shevyhmm what is wrong with GTK? I mean, the C sources are awful but the end result is not so bad in my opinion
20:56:17dom96It can be very annoying
20:56:28dom96Especially with scrolling
20:56:43dom96I had to strip down Aporia to its very core to track down an issue with scrolling
20:57:03dom96It turned out I was showing the main window to quickly lol
20:57:48dom96*too
20:59:12shevyhmm
20:59:43dom96I'm telling you.
20:59:46dom96Start with an IRC client
20:59:48dom96er.
20:59:52dom96IRC bot I mean
21:00:00dom96You get to have lots of fun with it.
21:00:08dom96Because you can add whatever features you want
21:00:28dom96Like getting the title of a web page when a URL is spoken in a channel
21:00:40dom96Integrating with twitter
21:00:53dom96using other famous website's APIs
21:01:01dom96Of course you will have to implement those in Nimrod :P
21:01:11dom96But that's part of the fun ;)
21:01:18dom96httpclient is already there
21:01:28dom96And I wrote it, so it's obviously awesome :P
21:01:43shevywell
21:01:49shevyI wanna concentrate on GUIs
21:01:49shevy:P
21:01:59fowltry sfml
21:02:06fowl:>
21:02:17shevyI think I know what I could do - I make a skeleton, just with fancy layout, and someone else can connect the base IRC things hehehe
21:02:41dom96shevy: Have you ever made GUIs with GTK?
21:03:09shevydom96 in ruby-gtk yeah. they worked but looked AWFUL
21:03:29shevyI made minesweeper, tic tac toe, some simple image displayer thing... a wrapper around ContextFree design
21:03:36shevylemme show you that latter program
21:03:56dom96And you found creating the GUI the fun part?
21:04:02shevyhttp://www.contextfreeart.org/gallery/ it is commandline, for linux, but look at those generated images
21:04:17shevydom96, hmm not sure. kinda, yeah. it required too much manual work though
21:04:33dom96I hate it :P
21:04:48dom96I will make a Glade to Nimrod converter someday
21:05:02AraqI
21:05:04AraqAM
21:05:06AraqBACK
21:05:08Araq:D
21:05:12dom96YAY
21:05:16shevyglade is that XML abomination isn't it?
21:05:38shevyI'd wish we could use .css files in GTK
21:06:36AraqI'm glad we can't :P
21:06:45Araqcss is incredible bloat
21:07:01shevyand very successful :>
21:07:07Araqif anything GTK should be stripped down
21:07:30Araqtry to do a resizable 3 column layout in css :P
21:07:43Araqwithout the evil tables of course
21:08:00dom96Araq: I can do it :P
21:08:03dom96Probably
21:08:05dom96Maybe
21:08:17Araqit's been proven to be impossible :P
21:08:22shevyI use tables
21:08:29dom96Araq: lies
21:08:31Araqok, with css v3.0 maybe it's possible
21:08:46Araqor v4?
21:08:50Araqwho knows ...
21:09:01shevyjust shows you that CSS gets better and better with every iteration :>
21:09:15Araqit shows that CSS's designer was clueless
21:09:34Araqin fact the "spec first" people always produce crappy designs
21:16:24Araqso dom96, I made it
21:16:30Araqno more openarray related bugs
21:16:46Araq"x" in "xyz" doesn't compile anymore if you don't import strutils
21:16:46dom96Really?
21:16:50Araqreally
21:16:52dom96Awesome
21:16:56Araqopenarray != varargs now
21:17:04Araqand I implemented:
21:17:06dom96Did you fix that ambiguous aporia delete Trix[a]r_za bug? :P
21:17:22dom96So you introduced 'varargs'?
21:17:27dom96what?
21:17:28Araqecho "$# $#".format(23, 4.9) # automatic $
21:17:35dom96What's the difference?
21:17:56Araq"$# $#" % [$23, $4.9]
21:17:58Araqvs.
21:18:05Araq"$# $#".format(23, 4.9)
21:18:16Araqunfortunately I couldn't do it with the % operator
21:20:00dom96^^
21:20:21fowlaraq now that i have this macro working i want to call it from another macro so i can inject some leet hax how do i do that
21:20:57Araqwhat's the deal?
21:21:08Araqmacro invokation is a nnkCall too
21:21:36fowli dont want to invoke it i want to get the nodes
21:21:51fowlwell i do want to invoke it, and get them nodes
21:21:54AraqgetAst or something
21:22:07Araqthere is some illdocumented built-in for that
21:22:10AraqI can't remember
21:22:14AraqI never use macros
21:22:45Araqthey are pretty useless to a compiler writer ;-) :P
21:23:57fowlgetAst(macro(body)) returns call(macro)
21:24:57Araqyes, dom96 I did implement 'varargs' finally
21:25:12Araqas 10 commit messages should have told you :P
21:25:26dom96what's the difference between 'varargs' and 'openarray' then?
21:25:54Araqproc o(x: openarray[int])
21:26:02Araqproc v(x: varargs[int])
21:26:09Araqo([1, 2, 3]) #ok
21:26:16Araqo(1, 2, 3) # invalid
21:26:23Araqv([1, 2, 3]) #ok
21:26:28Araqv(1, 2, 3) #ok
21:26:45Araq"openarray" now only means "seq or array"
21:27:00Araqit doesn't imply "implicit array construction" anymore
21:27:07Araqplus you can do this:
21:27:23Araqproc v(x: varargs[string, `$`])
21:27:35Araqv(1, 2, 3) # transformed into
21:27:45Araqv([$1, $2, $3])
21:28:03Araq== v("1", "2", "3")
21:28:08dom96nice
21:28:37Araqit introduces local type converters
21:28:45Araqreally good for formatted output
21:32:28Araqfowl: getAst should work, are you sure you use it in a macro with a macro invokation as argument?
21:34:00Araqand no, dom96, I didn't fix the ambiguous identifier bug yet
21:34:07Araqwe talked twice about it
21:34:15Araqis it important?
21:34:59*XAMPP_ quit (Quit: Leaving)
21:36:24fowlAraq: yea i think it will work i just have to change the body i send it
21:38:45dom96Araq: Well. As long as Aporia works for Trix[a]r_za I don't really care
21:42:23Araqcan't you just replace DELETE with module.DELETE everywhere and be done with it?
21:47:05dom96I could yes
21:48:34Araqwell? do it
21:49:02shevylazy dom!
21:49:17shevywork sucks his life essence away
21:49:48shevyas it does with me, I am watching The Good, the bad and the ugly, and doing some other easy stuff... non-coding stuff. my brain is sleeping already :P
21:50:09dom96yes... :(
21:52:25AraqI should stop working on nimrod too then I guess?
21:52:32Araq:P
21:53:56dom96Don't you dare
21:54:03dom96I'm actually working on Aporia right now :P
21:56:54Araqyay
21:57:16shevywhatcha fixing in aporia right now?
21:57:48dom96Currently I am making a color scheme for fun :P
21:58:11Araqlol
21:58:19Araqthat hardly qualifies as "work"
21:58:59dom96It totally does :P
22:01:09Araqnow ... fix 'final' for objects? or implement hygienic templates?
22:01:15Araqwhat should it be?
22:01:25dom96templates
22:01:36Araqalright
22:01:40Araqthat's more fun anyway
22:01:49Araqand jester will kick ass afterwards
22:02:22dom96:)
22:17:27Araqpragma parsing? cool
22:17:41AraqI hope you thought about that they can end with } too
22:17:52Araq.} is more common though
22:19:18dom96nah.
22:19:24dom96Just an experiment
22:19:31dom96Too tired to make it work for all cases :P
22:19:41dom96Also gtksourceview gives some warning
22:19:45Araqwell it's .} or }
22:19:52Araq2 cases
22:19:56dom96Because I made the style "nimrod:pragma"
22:30:55Araqhrm logging levels don't make much sense
22:31:23Araqbut not having them will annoy people
22:31:59Araqinstead it should be set-based
22:32:29Araq"this message is logged into the file and on console"
22:32:40Araq"this message is also sent per email"
22:32:42Araqetc.
22:33:12Araqthese things never fit a hierarchy
22:47:40dom96And with that
22:47:42*dom96 -> sleep
22:47:50Araqme too
22:47:52Araqgood night
22:47:54dom96Araq: IIRC I've looked at the logging module
22:48:01dom96And found some things I didn't like :P
22:48:07dom96And decided to fix it :P
22:48:13Araqlike what?
22:48:17dom96I can't remember
22:48:26dom96let me check
22:48:53dom96Araq: 91 stargazers!
22:49:35AraqI know
22:50:40dom96I think it was the use of global variables
22:50:50dom96but meh
22:50:55*dom96 shrugs
22:51:06dom96I'll look at it again tomorrow in the train
22:51:10dom96Good night
22:51:16Araqbye
23:25:39*Trix[a]r_za is now known as Trixar_za