<< 26-10-2013 >>
00:01:47Araqeven if not necessary (I think they are btw), sameTypeAux seems to have the bug, right?
00:02:56MFlamersameTypeAux, treats alot of types the same.... of tyGenericParam, tyGenericInvokation, tyGenericBody, tySequence,
00:02:56MFlamer tyOpenArray, tySet, tyRef, tyPtr, tyVar, tyArrayConstr,
00:02:56MFlamer tyArray, tyProc, tyConst, tyMutable, tyVarargs, tyIter,
00:02:56MFlamer tyOrdinal, tyTypeClass:
00:03:09MFlamerwhich seems a little suspicious
00:07:53Araqwell that's the power of generic programming
00:22:07Araqgood night
00:30:57MFlamerok, I think I might have it. You'll know in the morning! good night
00:32:44*DAddYE quit (Remote host closed the connection)
00:56:52*q66 quit (Quit: Leaving)
01:16:22*mflamer_ joined #nimrod
01:34:01*DAddYE joined #nimrod
01:38:20*DAddYE quit (Ping timeout: 245 seconds)
01:53:28*MFlamer quit (Ping timeout: 240 seconds)
02:17:27mflamer_Hey how do I create multiple pull requests?
02:17:46mflamer_Seems like it wants to pile them together
02:26:15*mflamer joined #nimrod
02:29:01*mflamer_ quit (Ping timeout: 272 seconds)
02:35:37*mflamer quit (Ping timeout: 272 seconds)
02:35:40*DAddYE joined #nimrod
02:40:19*DAddYE quit (Ping timeout: 272 seconds)
02:50:00*fowl joined #nimrod
03:10:46*mflamer joined #nimrod
03:14:02*mflamer_ joined #nimrod
03:16:09*mflamer quit (Ping timeout: 272 seconds)
03:36:36*DAddYE joined #nimrod
03:41:10*DAddYE quit (Ping timeout: 268 seconds)
03:45:03*brson quit (Quit: leaving)
03:57:51*fowl quit (Quit: Leaving)
04:27:11*jamil_1 joined #nimrod
04:37:44*DAddYE joined #nimrod
04:42:05*DAddYE quit (Ping timeout: 245 seconds)
04:47:38*OrionPK quit (Quit: Leaving)
05:29:17*xenagi quit (Quit: Leaving)
05:38:57*DAddYE joined #nimrod
05:43:20*DAddYE quit (Ping timeout: 245 seconds)
06:40:07*DAddYE joined #nimrod
06:44:35*DAddYE quit (Ping timeout: 245 seconds)
07:00:58*mflamer_ quit (Ping timeout: 268 seconds)
07:41:47*DAddYE joined #nimrod
07:46:15*DAddYE quit (Ping timeout: 245 seconds)
08:42:39*DAddYE joined #nimrod
08:47:06*DAddYE quit (Ping timeout: 252 seconds)
08:56:04*wlhlm joined #nimrod
09:43:53*DAddYE joined #nimrod
09:48:08*DAddYE quit (Ping timeout: 240 seconds)
10:37:27*gradha joined #nimrod
10:40:33gradhamflamer_ pull requests are just branches, if you commit to your branch and push it to the server the pull request will be updated, if you want to push separate changes, create a branch per pull request
10:45:32*DAddYE joined #nimrod
10:50:03*DAddYE quit (Ping timeout: 248 seconds)
10:51:39*q66 joined #nimrod
10:51:49gradhamflamer_ also, if you have already made a pull request and then push additional commits to your repo, the pull request will be updated but people watching it won't get a notification of this, so you should "bump" the issue if you want to bring attention to the changes
10:53:26*jamil_1 quit (Read error: Connection reset by peer)
10:53:35*jamil_1 joined #nimrod
11:23:48*gradha quit (Quit: bbl, need to watch https://www.youtube.com/watch?v=n7pXRdkdJxI again)
11:28:32dom96hello
11:32:07Araqhi dom96
11:33:23dom96So I spent my whole night trying to get a damn footer to stay at the bottom on my website, at 4am I gave up. CSS sure is great.
11:34:30dom96This is pretty cool: http://0hgame.eu/
11:35:23AraqCSS is an abomination
11:36:13Araqany "web standard" is, except JSON
11:41:42dom96yep
11:46:21*DAddYE joined #nimrod
11:50:51*DAddYE quit (Ping timeout: 248 seconds)
13:07:04*Associat0r quit (Quit: Associat0r)
13:49:19*DAddYE joined #nimrod
13:54:00*DAddYE quit (Ping timeout: 252 seconds)
14:07:12*io2 joined #nimrod
14:07:39*OrionPK joined #nimrod
14:13:48*Endy joined #nimrod
14:23:40*mflamer joined #nimrod
14:28:35*[1]Endy joined #nimrod
14:31:57*Endy quit (Ping timeout: 252 seconds)
14:31:57*[1]Endy is now known as Endy
14:51:16*q66 quit (Ping timeout: 246 seconds)
14:51:21*DAddYE joined #nimrod
14:56:41*DAddYE quit (Ping timeout: 245 seconds)
14:59:15mflamerok, I guess I'll have to create another branch. I'm used to perforce where a branch is much less granular
15:04:01*q66 joined #nimrod
15:17:37AraqMFlamer: what caused the bug?
15:28:27*xenagi joined #nimrod
15:43:45*mflamer quit (Ping timeout: 245 seconds)
15:55:59*DAddYE joined #nimrod
16:01:07*DAddYE quit (Ping timeout: 265 seconds)
16:27:04*hoverbear joined #nimrod
16:30:00dom96hey hoverbear, haven't seen you in a while.
16:30:24hoverbeardom96: Been beating my head against the wall with C
16:30:38dom96cool
16:30:42hoverbearNot really.
16:31:58dom96But at least that experience has brought you back here.
16:32:11hoverbearActually that was Araq's fault.
16:32:46dom96how so?
16:33:15hoverbearI dunno man it's saturday morning and I have to work on an SQL (Uck) assignment
16:35:49hoverbearI guess just being so goddamn nice.
16:35:56hoverbearAraq: You sexy beast.
16:38:13*dyu joined #nimrod
16:54:32hoverbeardom96 How's life treating you?
16:55:13dom96Good, but very busy :\
16:55:27hoverbeardom96 That's usually how it is
16:58:20*DAddYE joined #nimrod
16:58:38hoverbearHaha, making an E/R Diagram and for some reason I really, really want to make it totally symetric
17:02:31*DAddYE quit (Ping timeout: 245 seconds)
17:06:20*dyu quit (Quit: Leaving)
17:07:44*xenagi quit (Quit: Leaving)
17:33:32*gradha joined #nimrod
17:34:20Araqgradha: how can dom96's commit that's socket related make your code fail to compile?
17:35:00gradhaI'm not saying that commit makes it stop compile, but using the compiler at that version doesn't work
17:35:09gradhawant me to bisect a working version?
17:35:53gradhathe previously working version of the compiler I had was from around the end of august
17:36:17OrionPKcan I somehow put multiple sourcecode filter (procs) in 1 file?
17:37:03gradhaOrionPK: nimforum does so, main.tmpl has several procs
17:37:12OrionPKgradha thanks, I'll take a peak
17:37:45gradhayou need to use "end proc" and such https://github.com/nimrod-code/nimforum/blob/master/main.tmpl#L88
17:37:53OrionPKthats exactly what I was looking for
17:37:58OrionPKsome kind of 'end proc' marker
17:43:26*filwit joined #nimrod
17:43:31filwithi
17:43:44gradhayohoo
17:44:03filwithey gradha
17:44:15filwitgood to see you're still around, Dirkson :)
17:44:41filwithey, so i noticed "using" is a keyword in nimrod-git
17:45:00gradhadom96: did you test https://github.com/dom96/ipsumgenera/commit/9c449b10622cc16dde23f6766240a910cb4f023f? ipsum stops working for me
17:45:07OrionPKwould be nice if I could end the template itself and just start doing nimrod without the # preface
17:45:12filwitis it like zahary wanted, where it's basically the 'with' statement?
17:45:53Dirksonfilwit: Thanks! Though I'm not sure I like the language, honestly. The first non-trvial code example I tried to read was unreadable due to implicit typing. Makes coding a little faster, but means I can't read type from context.
17:47:02filwitDirkson, yeah, i had my problems with the language as well. I'm a C#/D guy originally, and it was really hard to adjust to Nimrod's very functional style
17:47:33filwitDirkson: luckily, Nimrod gives you a lot of control over it's style
17:47:52DirksonOh aye?
17:48:08filwitDirkson: for instance, i really don't like having to declare procedures outside the "body" of the type (like C#)
17:48:30Dirksonfilwit: I'm a C guy, myself. I'm not sure how that affects this, though.
17:48:34filwitDirkson: in Nimrod, I can just write a macro which pumps out the appropriate AST though
17:48:46DirksonAST?
17:48:54filwitDirkson: one sec, let me show you an example
17:48:57DirksonSure
17:49:42filwitDirkson: one i wrote a little while ago: https://gist.github.com/PhilipWitte/6775187
17:50:05Dirksonfilwit: I started defining this: http://www.orangehattech.com/language.png Likely won't ever go anyway, but just puddling around with it has taught me a good deal about why languages end up structured like they are.
17:50:09Araqfilwit: I don't think that's Dirkson's problem. He is a *C* guy
17:50:18AraqC has no classes either
17:50:25filwitah, i see
17:50:54DirksonIt does have structs, which I use VERY much like classes. Constructors, destructors, inheritance, the works.
17:51:31*hoverbear quit (Quit: Textual IRC Client: www.textualapp.com)
17:51:33OrionPKyou can do that in nimrod as well
17:51:44OrionPKtreat types like classes
17:52:26DirksonI'm not saying my way is *sane*, just that it's what I ended up doing :D I didn't have a good understanding of OOP before I started my project - I think if I did, I would have at least gone C++ from the start.
17:53:06filwitDirkson: what language is that? your language design?
17:54:50Dirksonfilwit: Yup. I starting trying to design the language I wanted, to see how it compared to languages that exist, and to get a better idea on why some language features exist. (Had a good discussion on Garbage Collection with a friend as a result of this) I'm not sure I've succeeded yet, but hey : )
17:56:30filwitDirkson: lol, it's interesting you mention language design. awhile ago i had an argument with Araq on language design. My wrote my initial design here: https://gist.github.com/PhilipWitte/4974439
17:56:36OrionPKyou require tabs & not spaces with that language? :p
17:56:53dom96gradha: oh, sorry. I need to go to the shop, but I will fix it once I get back.
17:57:03dom96gradha: btw, take a look at the new http://picheta.me
17:57:07DirksonOrionPK: I like tabs better ^^ But realistically, I'd allow people to use spaces instead, if they really wanted. Simple thing to replace.
17:57:55DirksonIf I ever do write a language, I think I'd definite take a page from Nimrod's book and compile it to C. That's a brilliant move.
17:58:05filwitDirkson: the thing is, writing a language is a lotalota work, and honestly, i think there is probably a way to make a language where everyone is happy, and I think with a little prodding, Nimrod could be that language
17:59:13filwitDirkson: yeah, i wasn't the biggest fan of "compile to C" originally (thought it would be slow). but it does have it's compatibility benefits.
17:59:16Dirksonfilwit: Your syntax seems like a C/Nimrod mix, which is interesting : ) Nimrod type "var Variable : type" constructs, but with brackets and such
17:59:27OrionPKI would compile to C++, dirkson :p
17:59:37DirksonOrionPK: That makes some sense : )
17:59:45*DAddYE joined #nimrod
18:00:08OrionPK99% of your generated code could be C, but then that 1% could take advantage of higher level stuff that's well optimized
18:00:35*hoverbear joined #nimrod
18:00:51Dirksonfilwit: I think it has a tendency to take advantage of 20+ years of optimizations on C compilers - Which is great for a new language. Nimrod is almost as performant as straight C, near as I can tell.
18:01:36filwitDirkson: yeah, that too
18:02:03*shodan45 's toy language would compile to COBOL & brainf*ck, randomly mixed up. :P
18:02:19Dirksonshodan45: *laughs*
18:03:00OrionPKactually LLVM might be better nowadays
18:03:15shodan45and the syntax would be xml mixed with javascript for macros xD
18:03:17gradhadom96: I would get rid of the width=device-width, without it the letter size is more readable on iphones
18:03:25DirksonOrionPK: There was another new language that used LLVM, and got really good results. I don't really understand what LLVM IS, though.
18:03:31*shodan45 slaps himself
18:03:39*io2 quit ()
18:03:41filwitlol, shodan45
18:03:55OrionPKdirkson it's basically an intermediate language that can be used to create binary on several platforms
18:04:10OrionPKso clang takes C -> LLVM -> binary (as I understand it)
18:04:21*DAddYE quit (Ping timeout: 272 seconds)
18:04:34DirksonOrionPK: Ah! c-- was an attempt at that.
18:04:48shodan45I think technically "LLVM" should be "LLVM IR" (intermediate representation, iirc?)
18:04:55OrionPKright
18:05:15OrionPKLLVM is the tool that converts the IR
18:05:25shodan45and yes, you can write LLVM IR by hand :)
18:05:33OrionPKbut shouldnt :P
18:05:40*Dirkson was TOTALLY not thinking that......
18:05:41OrionPKit's readable though
18:06:03*gradha foresees a future where people will discourage writing nimrod directly by hand
18:06:13OrionPKafk
18:06:52shodan45gradha: or people would do it "for fun", like coding in asm is today, or walking places
18:06:59DirksonMostly afk myself. Europe needs apples.
18:07:27shodan45Dirkson: so you're a European apple farmer? o_O
18:12:39*Dirkson bought Eruo Truck Simulator last night. Work will not be getting accomplished today.
18:12:49shodan45ohhhh
18:13:58shodan45might get that if it goes on sale for stupidly cheap prices (already have way too many games to play)
18:35:07filwitwhat is the status of the SDL lib?
18:35:21filwitmeaning, what version of SDL does it bind too?
18:36:40filwitSDL 1.2 or SLD 2.0?
18:37:09filwitand if SDL 1, is there an SDL2 binding anywhere?
18:38:26AraqSDL 2 is in babel afaik
18:39:17filwitaha! thanks
18:39:32filwiti thought i had seen someone mentions and SDL2 before
18:41:04AraqDirkson: soon your cursor will be able to tell you a variable's type :P
18:42:15Araqbut hey, lets cripple the PL instead of acknowledging PLs need tool support
18:47:31Araqsee you later
18:47:40gradhabye
19:00:01filwitbye
19:00:13gradhasee you later
19:00:21filwitno, lol
19:00:29filwiti was responding to Araq
19:00:38filwitbut was probably way to late
19:00:50gradhaok, no seing you later then
19:00:56*DAddYE joined #nimrod
19:05:11*DAddYE quit (Ping timeout: 245 seconds)
19:05:24filwithrm. somehow, there is already a "babel" command on my machine
19:05:40filwitsay's it's for molecular model building or some odd thing
19:05:48gradhasounds like a virus
19:05:55filwitlol
19:06:03filwitthis is arch linux..
19:08:09filwitit's an easy work around, but just though dom96 should know there is a conflict somewhere
19:16:01dom96filwit: I think you told me that before.
19:19:27Dirksonshodan45: Well, it's currently 66% off on steam. That's why I bought it ^^
19:19:59DirksonAraq: *considers* That's potentially a good point.
19:21:07gradhaDirkson: have you been itroduced to nimrod's idetool command?
19:28:21Dirksongradha: I have not, I think?
19:29:01gradhait's part of the plan on making better tools around nimrod than simple text editors
19:30:27gradhathere's a description at http://build.nimrod-code.org/docs/idetools.html but the short story is you invoke the compiler and it provides you info for syntax highlighting, completion, etc
19:33:51gradhaI was meaning to use idetools to build a source/documentation hyperlinker (https://github.com/gradha/the_hyperlink_vs_nimrod) but got distracted
19:34:50*Associat0r joined #nimrod
19:34:50*Associat0r quit (Changing host)
19:34:50*Associat0r joined #nimrod
19:35:53gradhayou can see an old example of the hyperlinker at http://dl.dropboxusercontent.com/u/145894/t/weird.proc.th.html
19:36:04gradhanote how you can click on procs and you are redirected to their documentation
19:36:49gradhaunfortunately the documentation uses numeric identifiers which change when you shuffle procs, so that's one of the distractions I want to tackle first
19:57:56OrionPKgradha any ideas on using a source code filter template to wrap the body of a source code filter
19:58:59gradhano idea sorry
19:59:05OrionPKhttps://gist.github.com/onionhammer/7173805
19:59:49OrionPKit's close to working, I can get it to inject header info, but I can't get it to wrap around the content, yet anyway
20:00:29*hoverbear quit (Quit: Textual IRC Client: www.textualapp.com)
20:02:10*DAddYE joined #nimrod
20:02:55*Raynes quit (Quit: ZNC - http://znc.sourceforge.net)
20:03:33dom96OrionPK: what error do you get?
20:03:38OrionPKinvalid indentation
20:03:50OrionPKon last last in the gist
20:04:08filwitdom96: i doubt it. pretty sure this is the first time i've tried to use babel
20:04:12dom96why don't you just make the 'template view' a proc?
20:04:22dom96and pass a string to it
20:04:23OrionPKbecause I'm trying to wrap the statement
20:04:59OrionPKI suppose I could do it like that though
20:05:45dom96just do master.view(result) at the end of that last proc.
20:06:02OrionPKah, thats a thought
20:06:10dom96filwit: Maybe someone else told me about it then.
20:06:40*DAddYE quit (Ping timeout: 265 seconds)
20:11:03*Raynes joined #nimrod
20:12:16OrionPKdom96 ugly but effective ;)
20:12:44OrionPKwish sourcecode filter syntax was a little prettier somehow
20:12:50*Raynes quit (Client Quit)
20:13:07*Raynes joined #nimrod
20:13:54dom96yeah, it could use some prettifying
20:14:17OrionPKmaybe take some inspiration from razor syntax in asp.net mvc
20:14:18filwithrm.. though sdl2 was in babel..
20:14:29filwitbut says failure not found
20:14:33OrionPKhttp://weblogs.asp.net/blogs/scottgu/image_2EE4DDA5.png
20:15:53dom96fowl should really separate it out of his nimlibs repo
20:16:15dom96filwit: it's in the 'fowltek' package
20:21:07filwitthanks, dom96
20:25:39filwitdamn, "Unsatisfied dependency: nimrod (>= 0.9.3)"
20:29:31dom96Not to worry, one of these days I will add a "--just-install-the-fucking-thing" flag :P
20:31:25filwitLOL
20:32:10filwitso what should i do? i'm thinking about just uninstalling the Nimrod from the arch-repo, and just using git-head
20:32:30dom96yeah. You should be using git-head anyway.
20:32:47gradhamaybe update arch? is it updatable?
20:32:50*brson joined #nimrod
20:33:42dom96gradha: You shouldn't change the names in the .ini file.
20:33:47filwitgradha: i think it's just at the last release (like it probably should be)
20:34:14filwitgradha: then again.. this IS Archlinux we're talking about :D
20:34:42dom96gradha: 'title' in .ini file is '${blog_title}' in the layouts.
20:39:55filwitkinda sux, i liked having the "official arch-linux nimrod release" installed :(
20:40:21gradhadom96: that was a try to get past the authors error
20:40:47dom96If you really want to you can clone the fowltek repo manually, edit fowltek.babel and then 'babel install'
20:41:40filwitnah, it's cool dom96, i like the idea of being on the git branch anyways
20:42:10filwiti noticed there's been a few changes, and i like messing with new stuff (and i've never had anything break on me before)
20:42:31dom96gradha: well does it work now?
20:44:13*Endy quit (Ping timeout: 268 seconds)
20:44:38gradhadom96: why would it? without changes I get the first error
20:45:59dom96update ${author} to ${blog_author} and it should work
20:49:01gradhaok, it works
20:49:20dom96good
20:54:08*brson quit (Ping timeout: 240 seconds)
20:54:58dom9612 issues to go... :P
20:55:36gradhamore sooner than later you'll learn to ignore the count
20:56:21*brson joined #nimrod
20:57:52dom96A better template engine would be nice... there is too much repetition between my layouts currently :\
21:03:23*DAddYE joined #nimrod
21:03:33dom96What's a good templating engine I could maybe wrap?
21:05:33gradhamaybe start with https://www.gnu.org/software/m4/
21:06:03gradhaquote: "Some people find m4 to be fairly addictive."
21:06:38dom96lol, nah, i'm alright
21:07:30*DAddYE quit (Ping timeout: 245 seconds)
21:11:41dom96meh, reinventing the wheel is always good.
21:14:12dom96gradha: btw are you alright with me removing '.' from the url of articles? I noticed that one of your article's title contains an ellipsis which is then present in the url.
21:14:25shodan45someone make a PHP backend for nimrod so I don't have to use this POS language any more ;_;
21:15:03gradhadom96: I don't understand the implications of that change
21:15:10gradhathe ellipsis is an unicode char
21:15:33dom96gradha: If someone bookmarked that article they will get a 404 page.
21:15:43dom96By ellipsis I mean 3 periods.
21:16:02dom96shodan45: No need for a PHP backend. You can stop using PHP already.
21:16:36gradhadom96: I did use the unicode char for ellipsis precisely to test broken link implementations and hyperlink highlighters
21:16:48gradhabut AFAICS removing dots should not affect it, should it?
21:16:57gradhathe dot '.' is not the same as '…'
21:17:11dom96oh
21:17:18dom96I thought you just used 3 dots.
21:17:24dom96Not the actual unicode char.
21:17:34gradhaI was tempted to use unicode chars with hearts and unicorns
21:17:53dom96in that case this change won't affect you
21:18:04dom96But perhaps I should remove unicode chars from the URLs too?
21:18:28gradhanon latin users will love that decission
21:18:48shodan45dom96: no I can't, I have to work in (several) existing medium-to-almost-huge PHP code bases :(
21:18:59dom96gradha: lol
21:19:12dom96gradha: You're right, that's a bad idea. Browsers can handle unicode nowadays.
21:19:38dom96shodan45: Lies. Just delete them and rewrite them in Nimrod.
21:20:38gradhaso talking about broken link implementations, Safari doesn't highlight the article with ellipsis as seen
21:20:59gradhachrome and firefox do work properly
21:21:16gradhaquality Apple software, as usual
21:22:19dom96Apple can't even get a todo list app right... how they ever managed to get Safari to work at all will remain a mystery.
21:22:33OrionPKdom96 CTemplate is simple
21:22:35OrionPKbuilt by google
21:22:48gradhadom96: so true
21:23:52gradhaI believe the last iOS incarnation was not only designed by Ive, he also implemented it
21:24:07dom96OrionPK: That looks pretty much identical to my current "templating engine" in Ipsum.
21:26:17filwitdom96: i updated my hymn diagram a little (started to a modes example). see: http://reign-studios.net/philipwitte/hymn/
21:26:46dom96filwit: When will you start implementing this in Nimrod?
21:26:50Araqdom96: nimrod's templating engine allows for as much code re-use as you can handle
21:26:55filwitdom96: i've already started :)
21:27:16filwitdom96: i've got the part syntax working (building out to nimrod types correctly)
21:27:24filwitdom96: but i don't have a ton of time right now
21:27:35dom96Araq: Yes, but it also requires me to recompile my app.
21:28:21dom96filwit: cool
21:28:29dom96filwit: Where is the source code? :P
21:28:32Araqyou can always read in stuff at runtime
21:28:36filwitdom96: it's basic right now, working on getting SDL working, but i'll post code onto github when it's in a better state
21:28:49dom96Araq: That's precisely what I am doing.
21:29:03Araqsounds like the perfect templating system to me :P
21:29:06dom96Araq: Ipsum is currently a combination of compile-time templates and runtime templates.
21:29:21dom96Doesn't change the fact that I still need to come up with my own runtime templates.
21:29:38dom96filwit: alright
21:29:39Araqdom96: runtime templates are a design mistake anyway
21:29:40filwitdom96: i just wanted to show you my modes (state machine) design. in theory, it should scale to fit a UI system, to a editor-defined character control scheme
21:29:45dom96filwit: let me know when you do
21:29:57filwitdom96: will do
21:31:53dom96Araq: You're right, i'll just use Hakyll instead since Ipsum is a design mistake :P
21:32:13gradhaHakyll sounds fun
21:32:47Araqdom96: more seriously though, would faster compile times change your design?
21:33:25dom96possibly.
21:34:20dom96I also wish, along with OrionPK, that the source code filter syntax would be a bit prettier.
21:35:07Araqok well make a suggestion then
21:35:47gradhaunicode is the way to go
21:36:10dom96Araq: *shrug*
21:36:17dom96There are way more important things to do anyway.
21:36:30*gradha escaping variables with π like a sir
21:36:52OrionPKre syntax filter, I can work on a proposal for how it might look, get some feedback on the forums
21:37:18AraqOrionPK: ok, sounds good
21:37:37Araqwe can always create new filters, will break nothing
21:38:38Araqwell ... in fact we can introduce reader macros instead of source code filters
21:38:43dom96Having to recompile ipsum and then regenerate my blog with every template update would get pretty annoying though so i'm not sure prettier templates will help all that much.
21:39:40OrionPKcompiling the templates improves performance though
21:40:16OrionPKalthough I suppose pre-parsing a template on initialization of a server is probably almost as fast
21:40:23Araqdom96: tried clang and -d:quick?
21:40:32Araqthat improves compile times
21:41:31dom968 seconds
21:41:35dom96Way too long
21:41:52OrionPKget a faster box ;)
21:42:24dom96That's what the JVM people say when I complain about it being a memory hog :P
21:42:33OrionPKlol
21:44:21gradhahmmm... faster box
22:01:58filwittrying to use fowls SDL2, but saying libSDL2.so not found..
22:02:05filwitcompiles though
22:04:37*DAddYE joined #nimrod
22:08:45*DAddYE quit (Ping timeout: 245 seconds)
22:10:07gradhamaybe you need to find your libSDL2.so in your system and force a linker path to it
22:11:43Araqon windows you could copy the .dll into the .exe's dir and it would work
22:13:02gradhagood night
22:13:17*gradha quit (Quit: bbl, need to watch https://www.youtube.com/watch?v=n7pXRdkdJxI again)
22:13:20OrionPKAraq dom96 http://forum.nimrod-code.org/t/289
22:14:05OrionPK(note that I have absolutely no idea how technically feasible this is to integrate into the compiler)
22:19:10AraqOrionPK: it requires lexer-parser-interactions so it's infeasible
22:19:36OrionPK:\
22:21:10OrionPKwhat parts specifically make it infeasible, or is it just the whole thing?
22:21:19OrionPKwhat's the least that can be done to make it more feasible
22:21:56dom96IIRC I tried to change the default prefix char for Nimrod code in standard templates once and that fails. Fixing this would be a good start.
22:26:07OrionPKit'd be nice if you could start the template inside of procs
22:26:17OrionPKand not need prefix chars except within the template
22:29:10Araqdom96: niminst changes the default prefix char
22:29:18Araqbut that's the $ not the #
22:29:32Araqperhaps you wanted to change the #
22:30:15AraqOrionPK: some """ ... """ are necessary so that the lexer accepts it
22:35:09OrionPKAraq okay.. perhaps after the filter: bit?
22:35:25Araqyeah
22:35:37OrionPKand maybe have the end """ instead of the "end filter"
22:36:30dom96I think you might as well just use htmlgen
22:36:31OrionPKmore advanced syntax highlighting in Aporia/et al will also help this
22:37:01OrionPKdom96 this isnt necessarily just html of course..
22:38:00dom96then isn't it just glorified string interpolation? :P
22:39:10OrionPKthe idea is only to improve on the syntax for what exists
22:39:22OrionPKhaving to preface every line with # stinks
22:39:34OrionPKwriting code in what effectively looks like comments
22:44:25Araqwell actually ... nimrod's DSL capabilities are so good we don't need source code filters
22:47:47OrionPKhttp://forum.nimrod-code.org/t/289
22:48:25OrionPKdont like that it's basically inside of a string :\
23:01:59filwitlol, gdamnit
23:02:29filwitall this time trying to figure out why libSDL2.so wont load, and turns out to be a recent ArchLinux bug
23:02:43filwitwell that's good though i guess
23:02:52filwitjust have to upgrade prob
23:05:50*DAddYE joined #nimrod
23:10:00*DAddYE quit (Ping timeout: 245 seconds)
23:12:35*filwit quit (Quit: Leaving)
23:27:19*wlhlm quit (Ping timeout: 260 seconds)