| 00:03:39 | Araq | guys, can you please dom96's new closure macro? we like to showoff it on the news page |
| 00:03:57 | flaviu | It works for me |
| 00:08:12 | * | xenagi joined #nimrod |
| 00:09:07 | BitPuffin | is there a quick way to test it |
| 00:09:30 | BitPuffin | like ./cock test closure |
| 00:09:40 | BitPuffin | probably good if someone (me) tests it on mac |
| 00:09:57 | Araq | the point is to come up with a test for yourself |
| 00:10:10 | Araq | not test it whether it works on your particular machine |
| 00:10:19 | BitPuffin | ah |
| 00:10:31 | BitPuffin | was thinking if there was a test in the test suite or something |
| 00:10:40 | dom96 | yes, what Araq said. |
| 00:10:49 | dom96 | There is a test though: closure/tclosuremacro |
| 00:11:18 | dom96 | now what's a cool example showing off closures? |
| 00:11:31 | BitPuffin | dom96: quicksort |
| 00:13:53 | BitPuffin | lol can't get the executables to be equal |
| 00:16:27 | * | Demos quit (Ping timeout: 250 seconds) |
| 00:18:13 | Araq | wtf. now both macro docs are empty |
| 00:18:37 | * | psquid quit (Ping timeout: 250 seconds) |
| 00:19:00 | BitPuffin | wut |
| 00:19:15 | * | psquid joined #nimrod |
| 00:21:03 | BitPuffin | is there a way to set global imports in a nimrod.cfg file? |
| 00:21:17 | BitPuffin | because it would be nice to enable sugar such as the new closure macro in such a file |
| 00:21:36 | renesac | and then your code won't compile for anybody else? |
| 00:21:41 | BitPuffin | so you'd put that in the project config file |
| 00:21:51 | Araq | it's possible but horrible |
| 00:22:01 | BitPuffin | renesac: not at all the case if it's in the project config |
| 00:22:16 | BitPuffin | and/or specified in the build system |
| 00:23:29 | BitPuffin | Araq: for most things I agree but in the case of sugar the whole point of it is that during your code flow you can write something faster than you would be able to otherwise and stopping to import the sugar kinda breaks the flow |
| 00:23:30 | BitPuffin | I dunno |
| 00:23:40 | BitPuffin | I'm a bit torn |
| 00:23:47 | Araq | it will be added to system |
| 00:24:12 | EXetoC | BitPuffin: import=x I think (--import:PATH) |
| 00:24:17 | BitPuffin | Araq: ah |
| 00:24:20 | BitPuffin | kewl |
| 00:24:58 | BitPuffin | EXetoC: you mean --import:x ? |
| 00:25:34 | EXetoC | I just copied from the help output |
| 00:25:44 | BitPuffin | ah |
| 00:26:00 | renesac | is a centralized compilation cache, that would store precompiled stdlibs, planned? |
| 00:27:10 | * | Demos joined #nimrod |
| 00:27:45 | * | renesac left #nimrod ("Leaving") |
| 00:28:47 | xenagi | Araq, I submitted a pull request for bug #1081. Let me know if it's good enough |
| 00:29:08 | * | renesac joined #nimrod |
| 00:30:40 | Araq | xenagi: not really |
| 00:31:04 | xenagi | why not? |
| 00:32:24 | Araq | I think it's more complex than necessary |
| 00:34:59 | xenagi | hmm, i suppose some of those cases could be cascaded |
| 00:35:23 | flaviu | Couldn't it be done by just adding the underscore in all cases? |
| 00:35:46 | xenagi | yes |
| 00:35:51 | xenagi | should it? |
| 00:36:16 | Demos | ermmmm reserved names ahoy |
| 00:38:11 | * | DAddYE joined #nimrod |
| 00:38:33 | flaviu | Araq: What do you think about appending _ to every name? |
| 00:38:50 | Araq | it's crap. |
| 00:38:57 | dom96 | Why _? |
| 00:39:10 | dom96 | Append 'ID' if the identifier begins with a number. |
| 00:39:22 | xenagi | _ would not hinder readability of the variable name |
| 00:39:35 | * | Araq hates C's __ASCII_ART__ |
| 00:40:01 | Demos | but Araq names are a scarce resource and we must make sure we never ever conflict |
| 00:40:16 | Demos | :D |
| 00:40:33 | flaviu | Well, the generated code is already less than beautiful, and as dom96 says, any other letter would also work as a prefix |
| 00:41:05 | xenagi | So just pick some arbitrary letter for all cases? |
| 00:41:47 | xenagi | how about... "NIMVAR_" |
| 00:41:57 | Araq | why is `12` a valid identifier in the first place? |
| 00:42:04 | flaviu | xenagi: Some compilers only look at the first 6 chars |
| 00:42:09 | Araq | that is what we should fix |
| 00:42:17 | Demos | flaviu: get a new compiler |
| 00:42:26 | xenagi | oh, well then thats a different issue entirely Araq |
| 00:42:27 | Araq | dom96: I fixed the docgen issue, or rather found a workaround |
| 00:42:30 | Demos | it is not 1960 any more |
| 00:43:03 | Araq | Demos: but but but what if 'memcpy' is broken? autoconf can check for that |
| 00:43:16 | dom96 | Araq: pushhh ittt |
| 00:43:17 | flaviu | Araq: Did you see my usecase? ``let `1/2` = 1/2`` |
| 00:43:46 | Araq | we should care about portability to Barely Legal Unix 0.8 |
| 00:44:17 | Araq | but not about portability to windows of course as that lacks a posix-like shell ... |
| 00:45:04 | Araq | flaviu: that's not a use case, that's an academic exercise |
| 00:45:12 | BitPuffin | Araq: isn't powershell sorta kinda but not really posixy |
| 00:45:27 | Araq | BitPuffin: like your mum. |
| 00:45:32 | Araq | now see what you've done |
| 00:45:37 | flaviu | What if someone wants to use nimrod on a micro-controller with a crappy compiler or something? |
| 00:45:37 | flaviu | Araq: I've used that syntax before for some optimizations |
| 00:45:45 | BitPuffin | My mom is fully POSIX compliant |
| 00:46:01 | xenagi | lol |
| 00:46:02 | BitPuffin | Araq: carefull, I might ban you from uh.. my let's play?? I can't ban you from anything |
| 00:46:09 | BitPuffin | :/ |
| 00:46:31 | BitPuffin | I'm gonna have to create a kickass multiplayer RTS just so that I can ban you from it |
| 00:46:33 | BitPuffin | XD |
| 00:46:54 | xenagi | flaviu, like what use cases? |
| 00:49:14 | * | Demos quit (Ping timeout: 240 seconds) |
| 00:49:19 | flaviu | xenagi: Manual subexpression elimination |
| 00:51:28 | * | xenagi quit (Quit: Leaving) |
| 00:59:57 | * | xenagi joined #nimrod |
| 01:03:45 | flaviu | grep -r 'ord(\'A\')' ./compiler/ | wc -l |
| 01:03:45 | flaviu | >> 18 |
| 01:07:52 | Araq | so? afraid ord('A') changes its value? |
| 01:11:46 | flaviu | Sorry, I should have elaborated. Half of those are basically toLower reimplemented. |
| 01:11:58 | Araq | ofc |
| 01:12:17 | Araq | old pascal neither had 'inline' nor macros |
| 01:14:16 | Varriount | I'm back |
| 01:16:00 | renesac | <pengvado> windows is posix compliant |
| 01:16:01 | renesac | <pengvado> every single posix function returns ENOTIMPLEMENTED, which is one complying implementation :) |
| 01:16:02 | dom96 | Varriount: Good, I was starting to get worried you may sleep until tomorrow :P |
| 01:16:24 | EXetoC | "var cvar = c; put(cvar, T)" matches correctly, so will a rewrite for put(var c, T) be fine? |
| 01:16:33 | Varriount | dom96: It was tempting. |
| 01:16:54 | Araq | EXetoC: no that would be a hack |
| 01:22:42 | Araq | argh |
| 01:22:56 | Araq | proc pushFrame(s: PFrame) {.compilerRtl, inl, exportc: "nimFrame".} = |
| 01:22:57 | Araq | # XXX only for backwards compatibility |
| 01:22:59 | Araq | ... |
| 01:23:01 | Araq | proc nimFrame(s: PFrame) {.compilerRtl, inl, exportc: "nimFrame".} = |
| 01:23:02 | Araq | ... |
| 01:23:04 | Araq | spot the problem |
| 01:25:37 | dom96 | same exportc name? |
| 01:25:44 | Araq | yeah |
| 01:32:34 | * | DAddYE quit () |
| 01:39:48 | * | Demos joined #nimrod |
| 01:41:14 | * | Demos quit (Read error: Connection reset by peer) |
| 01:45:16 | NimBot | Araq/Nimrod devel 042ffed Araq [+0 ±2 -0]: some progress for #1082 |
| 01:45:16 | NimBot | Araq/Nimrod devel a868575 Araq [+0 ±2 -0]: doc2 likes future.nim now |
| 01:45:16 | NimBot | Araq/Nimrod devel 667b7d8 Araq [+0 ±1 -0]: Merge branch 'devel' of https://github.com/Araq/Nimrod into devel |
| 01:45:16 | NimBot | Araq/Nimrod devel b15f323 Araq [+0 ±1 -0]: updated news.txt |
| 01:45:16 | NimBot | 1 more commits. |
| 01:48:49 | dom96 | Araq: oh, you should add a link to the downloads page in the news |
| 01:52:09 | Araq | well I'm done |
| 01:52:35 | Araq | I'll do the rest "tomorrow" night |
| 01:52:55 | dom96 | bah |
| 01:53:01 | dom96 | So I stayed up for nothing? |
| 01:53:17 | Araq | you helped a great deal? |
| 01:53:19 | Varriount | What branch do I use when generating installers? |
| 01:53:59 | Araq | Varriount: master but I haven't merged yet. should I? |
| 01:54:06 | Varriount | It's up to you. |
| 01:54:20 | flaviu | Araq: Can you do a quick comment on https://github.com/Araq/Nimrod/pull/1113 before going to bed? |
| 01:55:58 | dom96 | Varriount: Practice generating on devel for now |
| 01:56:08 | dom96 | We'll probably make some other changes tomorrow |
| 01:57:01 | Varriount | Plus the build bots. |
| 01:58:00 | NimBot | Araq/Nimrod devel e74868f flaviut [+0 ±1 -1]: Delete dead file |
| 01:58:00 | NimBot | Araq/Nimrod devel fb91efd flaviut [+0 ±0 -1]: charsets is never used |
| 01:58:00 | NimBot | Araq/Nimrod devel d0f0076 flaviut [+0 ±0 -1]: parsecfg is never used and is also part of the stdlib |
| 01:58:00 | NimBot | Araq/Nimrod devel 41f82fb Andreas Rumpf [+0 ±1 -3]: Merge pull request #1113 from flaviut/del_dead... 2 more lines |
| 01:58:15 | Araq | flaviu: let's see what it breaks |
| 01:58:39 | Araq | Varriount: is the babel integration with the windows installer done? |
| 01:58:42 | flaviu | I don't think it breaks anything, it compiles for me. Grep turned up no usages either |
| 01:59:03 | Varriount | Araq: Yes. |
| 01:59:09 | Araq | flaviu: but only nimgrep can grep nim |
| 01:59:29 | Araq | Varriount: awesome! |
| 01:59:39 | Araq | how come I missed your PR? |
| 01:59:47 | Varriount | What PR? |
| 02:00:14 | Araq | compiler/nimrod.ini patches? |
| 02:02:50 | Araq | anyway, good night |
| 02:03:02 | dom96 | same, bye |
| 02:05:39 | Varriount | Goodnight |
| 02:08:10 | * | Jesin joined #nimrod |
| 02:22:35 | * | xenagi quit (Quit: Leaving) |
| 02:23:00 | fowl | BitPuffin, still around? |
| 02:26:00 | reactormonk | Araq, oh, sweet |
| 02:27:50 | * | ehaliewicz quit (Remote host closed the connection) |
| 02:28:17 | BitPuffin | fowl: no I'm dead |
| 02:29:13 | BitPuffin | fowl: are you asking if I'm around in the irc channel in general or if I'm awake lol |
| 02:30:10 | fowl | just making sure u still exist |
| 02:33:01 | BitPuffin | fowl: ah, I do, thanks. Do you? |
| 02:33:54 | fowl | all i'm sure of is that from my perspective, i might exist, and i cant even be sure of that |
| 02:35:25 | BitPuffin | hmm |
| 02:35:30 | BitPuffin | now you are making me think the same thing |
| 02:35:36 | BitPuffin | I retract my previous reply |
| 02:35:41 | BitPuffin | it is now "maybe" |
| 02:39:56 | fowl | im thinking about making nake read the first line of a nakefile for compiler options |
| 02:40:32 | BitPuffin | interesting |
| 02:40:58 | fowl | i have a task that downloads game data, need -d:ssl for it |
| 02:41:08 | fowl | btw BitPuffin u up for compiling more things for osx? |
| 02:43:18 | flaviu | Does dom do the build farm? |
| 02:43:33 | reactormonk | flaviu, yup |
| 02:45:30 | BitPuffin | fowl: hmm |
| 02:45:37 | BitPuffin | I didn't really compile anything last time |
| 02:45:40 | BitPuffin | lol |
| 02:45:43 | BitPuffin | I dunno |
| 02:45:53 | BitPuffin | as long as it doesn't require me to edit source files then maybe |
| 03:01:27 | * | q66 quit (Quit: Leaving) |
| 03:01:46 | fowl | BitPuffin, libenet and chipmunk? :> |
| 03:07:35 | BitPuffin | fowl: are they big? |
| 03:07:42 | BitPuffin | will I need to modify allegro? |
| 03:14:30 | * | BitPuffi1 joined #nimrod |
| 03:15:11 | fowl | BitPuffi1, no and no |
| 03:15:39 | * | BitPuffin quit (Read error: Connection reset by peer) |
| 03:17:02 | fowl | BitPuffi1, http://enet.bespin.org http://chipmunk-physics.net |
| 03:18:51 | BitPuffi1 | fowl: installing through brew :P |
| 03:19:50 | fowl | sfml is nice, they have binaries for the lame platforms |
| 03:20:27 | BitPuffi1 | installed |
| 03:22:38 | * | Varriount bangs head repeatedly on desk |
| 03:23:00 | Varriount | Does nimrod use some sort of compiler-specific stuff for exception handling? |
| 03:23:21 | flaviu | Varriount: The VM does fancy stuff for exceptions |
| 03:23:32 | fowl | Varriount, setjmp/longjmp in c |
| 03:23:37 | Varriount | Hm. |
| 03:24:04 | Varriount | gcc is linking a C++ exception handling dll to nimrtl.dll, which is causing a couple tests to fail. |
| 03:24:34 | * | flaviu quit (Quit: Night) |
| 03:24:42 | * | psquid quit (Ping timeout: 240 seconds) |
| 03:24:59 | * | psquid joined #nimrod |
| 03:25:01 | Varriount | Also, did someone spiff up the tester? I could swear that, until a week or so ago, it didn't have colored output. |
| 03:31:07 | fowl | Varriount, iirc c++ exceptions are used in cpp mode |
| 03:31:45 | Varriount | Is the runtime library, by default, compiled in C++ mode? |
| 03:33:28 | fowl | no |
| 03:33:39 | fowl | ? |
| 03:45:43 | Skrylar | Varriount: unittest module does that |
| 03:45:52 | Skrylar | though i've stopped using unittest and went to doasserts >_> |
| 03:46:45 | Skrylar | fowl: is htere a reason to care about chimpmonk instead of box2d? |
| 03:47:24 | fowl | there no c interface for box2d |
| 03:47:40 | fowl | also chipmunk is just as good |
| 03:48:09 | Skrylar | i have an automatic distrust of pro versions :\ |
| 03:48:32 | fowl | pro version is the obj c interface for ios |
| 03:48:41 | Skrylar | don't ask me what a better model is, i just don't trust people who do it |
| 03:48:45 | Skrylar | no, it also includes auto-boundaries |
| 03:48:48 | fowl | tax the rich |
| 03:49:01 | Skrylar | i just checked their feature matrix and its one of those "can you trust whatever improvements to not be deemed 'pro' or not" situations |
| 03:49:07 | Skrylar | anyway, not my project so i don't care |
| 03:49:29 | fowl | Skrylar, id write my own auto boundary anyway, |
| 03:49:57 | Skrylar | i'm tempted to hook the winapi for rendering font glyphs |
| 03:50:03 | Skrylar | mostly because it sheds a dependency on one platform |
| 03:53:21 | BitPuffi1 | fowl: so gief the oneliner? :P |
| 04:02:15 | Skrylar | i'm feeling omnicidal. ._. |
| 04:03:15 | BitPuffi1 | http://youtu.be/o4nTkMNVVdQ |
| 04:03:59 | BitPuffi1 | taht's what comes to mind when you say omnicidal |
| 04:08:44 | fowl | BitPuffi1, for what |
| 04:09:39 | BitPuffi1 | fowl: you wanted me to run something? |
| 04:09:52 | BitPuffi1 | or did I install chipmunk and shit for nothing |
| 04:09:57 | fowl | did u fix your allegro |
| 04:10:27 | fowl | oh can you send me the binaries? i will put them with the babel pkg |
| 04:12:13 | BitPuffi1 | no |
| 04:12:21 | BitPuffi1 | I did naught |
| 04:12:48 | fowl | ill take a look at updating this week |
| 04:13:07 | fowl | to allegro 5.1 |
| 04:14:08 | BitPuffi1 | ahh |
| 04:14:09 | BitPuffi1 | do et |
| 04:14:13 | BitPuffi1 | nyeaaaahh! |
| 04:15:26 | Skrylar | BitPuffi1: no, omni-cide is "to kill everything" |
| 04:15:53 | Skrylar | homicidal is specifically a person |
| 04:16:21 | fowl | BitPuffi1, ok but im probably going to use a -d because it sounds easier |
| 04:19:21 | Skrylar | i'll just keep using sdl2 |
| 04:19:54 | fowl | they have similar apis anyways |
| 04:20:00 | Skrylar | yeah but sdl is pretty stable |
| 04:20:27 | Skrylar | now i get to deal with getting font glyphs on a sprite sheet |
| 04:20:38 | Skrylar | being able to actually write GUI code is near, the yaks are almost all bald \o/ |
| 04:20:52 | fowl | why do you have to go so low level |
| 04:20:55 | fowl | use cairo or something |
| 04:21:04 | Skrylar | cairo doesn't do fonts |
| 04:21:12 | fowl | well use something that does |
| 04:21:36 | Skrylar | i already wrote a hashtable cache and i have some unicrap compatible formatting around here somewhere |
| 04:21:47 | Skrylar | compared to two months ago, just stamping glyphs on a texture atlas is easy |
| 04:27:02 | fowl | i feel ya |
| 04:27:40 | fowl | i think about quitting the cigarettes all the time but then im like why both, i already started |
| 04:27:52 | fowl | why botehr* |
| 04:31:03 | Skrylar | i donno, i could go for some heavy psychotropics |
| 04:41:24 | BitPuffi1 | fowl, Skrylar: allegro does fonts |
| 04:41:27 | BitPuffi1 | justasayin |
| 04:57:17 | renesac | Skrylar, you will handle kerning too? |
| 05:36:16 | Skrylar | renesac: maybe, donno |
| 05:37:40 | Skrylar | right now programming is the last thing on my mind |
| 05:42:35 | renesac | kerning seems a pretty hard thing... last time I saw OpenOffice had many troubles with it... |
| 05:42:54 | renesac | so you must basically reimplement freetype? |
| 05:44:40 | Skrylar | no? |
| 05:44:43 | Skrylar | freetype is a rasterizer |
| 05:44:47 | Skrylar | it doesn't do layout |
| 05:44:48 | renesac | well, I think it isn't the right time to make those questions |
| 05:45:12 | Skrylar | if there was anything i was 'reimplementing' it would be pango |
| 05:45:12 | renesac | layout isn't an orthogonal thing to rasterization? |
| 05:45:23 | renesac | so you aren't doing font rasterization? |
| 05:45:25 | Skrylar | no |
| 05:45:40 | Skrylar | windows will do it for you, osx has a system for it, and theres freetype |
| 05:45:50 | Skrylar | layout is a bit more problematic |
| 05:46:00 | Skrylar | as far as diacritic marks, kerning and all that shit |
| 05:47:02 | renesac | ok, so individual characters are rendered by an external library, and you place them with your code |
| 05:48:06 | renesac | that makes sense, but still, you will certainly have to work on kerning latter... but it is not a critical component to have a GUI runing at least |
| 06:10:17 | Skrylar | i'll worry about it some time when i give a fuck; right now i'm dealing with other problems |
| 06:30:55 | * | noam joined #nimrod |
| 06:39:39 | fowl | gn |
| 07:12:24 | * | OrionPK quit (Remote host closed the connection) |
| 08:09:41 | NimBot | nimrod-code/csources master 85caad2 Araq [+0 ±91 -0]: csources for version 0.9.4 |
| 08:10:54 | NimBot | Araq/Nimrod devel d22506c Araq [+0 ±1 -0]: fixes #847 |
| 08:10:54 | NimBot | Araq/Nimrod devel 80e377b Araq [+0 ±1 -0]: manual.txt updates |
| 08:10:54 | NimBot | Araq/Nimrod devel 7a55701 Araq [+0 ±3 -0]: version 0.9.4 |
| 08:10:54 | NimBot | Araq/Nimrod devel 687a1b7 Araq [+0 ±1 -3]: Merge branch 'devel' of https://github.com/Araq/Nimrod into devel |
| 08:20:35 | NimBot | Araq/Nimrod master 5da463e Jason Livesay [+0 ±1 -0]: Redis: optional pipelining and better tested transactions |
| 08:20:35 | NimBot | Araq/Nimrod master 9a728b1 Jason Livesay [+0 ±1 -0]: Don't need ref string; use PPipeline instead of ref TPipeline |
| 08:20:35 | NimBot | Araq/Nimrod master 3f3014f flaviut [+0 ±1 -0]: Koch deletes the .gitignore while cleaning |
| 08:20:35 | NimBot | Araq/Nimrod master be02aae Jason Livesay [+0 ±1 -0]: factor per comments |
| 08:20:35 | NimBot | 102 more commits. |
| 08:27:37 | Araq | just fyi I'll be back tonight |
| 08:27:50 | Araq | please test the new release! |
| 08:27:57 | Araq | bbl |
| 08:42:11 | * | Matthias247 joined #nimrod |
| 10:17:06 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 11:35:48 | * | uvelichitel joined #nimrod |
| 12:00:16 | * | io2 joined #nimrod |
| 12:00:59 | * | BitPuffi1 quit (Ping timeout: 252 seconds) |
| 12:05:38 | Varriount | dom96: |
| 12:05:40 | Varriount | <datadigger> Varriount: Why no 64-bit? Well, there is hardly any audience for it, 32bit is fine for almost all applications. |
| 12:06:00 | Varriount | Is what I was told when asking about a 64 bit sqlite binary in #sqlite |
| 12:09:15 | dom96 | I can't say I disagree with him. |
| 12:12:08 | * | q66 joined #nimrod |
| 12:12:08 | * | q66 quit (Changing host) |
| 12:12:08 | * | q66 joined #nimrod |
| 12:13:52 | Varriount | Yes, but how are people ever going to adopt 64 bit if there's no support for it? |
| 12:14:07 | Varriount | Stupid chicken-and-egg problems... |
| 12:14:14 | EXetoC | maybe that's the case for some apps, but for db's, really? |
| 12:15:34 | Varriount | EXetoC: What do you consider a "binary"? |
| 12:18:47 | EXetoC | anything that contains executable code, I guess. does the memory cap apply in this case? because you want as much as possible to reside in RAM in this case I'd imagine |
| 12:21:54 | Varriount | EXetoC: Someone on #sqlite was confused because I considered the compiled, distributable package of sqlite.dll and it's files hosted on sqlite's website to be a "binary" |
| 12:23:27 | Varriount | I mean, I guess the proper term should have been something like a "binary package" or "the compiled and packaged dll", however I've always thought that, colloquially, a "binary" could refer to the ready and packaged distributable version of a computer program or library. |
| 12:24:01 | EXetoC | ok |
| 12:24:13 | EXetoC | dom96: should "echo(@[1, 2, 3, 4].map((x) => x * 2))" work? |
| 12:24:43 | EXetoC | wait |
| 12:25:28 | dom96 | I think so |
| 12:25:33 | dom96 | What's the error? |
| 12:26:51 | EXetoC | dom96: "test.nim(20, 6) Error: type mismatch: got (seq[int], proc (GenericParam): auto) but expected one of: system.map(data: var openarray[T], op: proc (var T){.closure.}) system.map(data: openarray[T], op: proc (T): S{.closure.}): seq[S]" |
| 12:27:17 | Varriount | I bet it's the {.closure.} pragma |
| 12:27:30 | dom96 | oh, you need to annotate the x type |
| 12:27:39 | dom96 | (x: int) => x * 2 |
| 12:28:26 | EXetoC | yes, and I can not annotate the return type at the same time |
| 12:28:56 | dom96 | you can |
| 12:29:02 | dom96 | But you shouldn't need to |
| 12:29:17 | EXetoC | it doesn't work in this case |
| 12:29:48 | EXetoC | you shouldn't have to annotate anything ideally, but it's an improvement either way. great work |
| 12:30:39 | dom96 | yeah, better type inference will hopefully happen |
| 12:33:37 | EXetoC | damn I got 19 open issues now ^_^ |
| 12:34:25 | Varriount | Gotta report them all! |
| 12:42:31 | * | Araq0 joined #nimrod |
| 12:42:52 | Araq0 | Boo |
| 12:43:00 | EXetoC | hi |
| 12:43:29 | Varriount | Araq0: Ah! |
| 12:43:34 | * | Varriount runs and hides |
| 12:43:44 | Araq0 | Did i break everything? |
| 12:43:57 | dom96 | Araq0: You didn't commit all the C sources |
| 12:44:26 | EXetoC | "release" as in master? |
| 12:44:27 | Araq0 | I did commit am |
| 12:44:51 | dom96 | Araq0: Yeah, and you only committed the files that changed. You forgot to add the new files. |
| 12:44:59 | Varriount | Araq0: I got gcc to stop crashing... now collect2 crashes. >_< |
| 12:45:05 | Araq0 | Argh |
| 12:46:05 | Araq0 | Dom96 did you fix it |
| 12:46:22 | dom96 | not yet |
| 12:46:30 | dom96 | Just finished watching GoT :P |
| 12:46:41 | Araq0 | Lol |
| 12:47:42 | Araq0 | dom96: pm me |
| 12:57:16 | * | BitPuffi1 joined #nimrod |
| 12:59:36 | EXetoC | BitPuffi1: morning |
| 13:02:24 | * | BitPuffi1 quit (Ping timeout: 276 seconds) |
| 13:02:45 | Araq0 | Is it too late to move spawn to its own module? |
| 13:06:18 | EXetoC | hasn't it been used in only a couple of places? I can deal with the refactoring if you want |
| 13:07:15 | Araq0 | It is completely new. So no |
| 13:12:31 | EXetoC | you might've missed the word 'only'. anyway, are you reluctant towards pushing more stuff to master? and is master the so-called "release"? |
| 13:12:49 | EXetoC | but I'll have to wait for the C sources issue to be fixed before I can try it |
| 13:13:48 | Araq0 | It is a release |
| 13:14:17 | Araq0 | No further pushes to master |
| 13:15:05 | EXetoC | right |
| 13:17:38 | EXetoC | so, let's deal with that after the release then? I wonder if module migrations can be compiler-assisted. the fact that system is imported implicitly might complicate things though |
| 13:20:29 | Araq0 | Oh well it is not important |
| 13:20:47 | EXetoC | I think I'm overcomplicating this. it might be fairly simple |
| 13:22:22 | EXetoC | just invoke a pragma in the involved modules |
| 13:22:33 | NimBot | nimrod-code/csources master e6e8462 Dominik Picheta [+39 ±7 -0]: Fixes #2. |
| 13:22:46 | Araq0 | Bbl |
| 13:22:48 | dom96 | EXetoC: Can you test the new c sources please? |
| 13:23:15 | EXetoC | on master? |
| 13:23:52 | dom96 | that's irrelevant |
| 13:24:23 | * | Araq0 quit (Quit: Bye) |
| 13:24:59 | EXetoC | I might as well if the release needs to be tested |
| 13:25:16 | dom96 | master and devel are currently the same |
| 13:26:00 | EXetoC | I will just to be safe |
| 13:28:17 | dom96 | ok, so 0.9.4 is tagged |
| 13:28:17 | EXetoC | ok there's a tag now |
| 13:28:22 | EXetoC | :p |
| 13:28:25 | * | dom96 waits patiently for Varriount to show up |
| 13:29:05 | EXetoC | build.sh needs the executable flag again |
| 13:29:23 | dom96 | ~1400 commits https://github.com/Araq/Nimrod/compare/v0.9.2...v0.9.4 |
| 13:29:23 | dom96 | woo |
| 13:31:12 | * | [1]Endy joined #nimrod |
| 13:31:57 | EXetoC | aporia error * 100: "Could not restore file from session.." :-) |
| 13:32:48 | EXetoC | what are some other programs that I can try? my rather featureless OpenGL program works |
| 13:33:35 | dom96 | cool, you use aporia? |
| 13:33:47 | * | darkf quit (Quit: Leaving) |
| 13:34:07 | EXetoC | no but I wanted to test some apps with the release build |
| 13:34:12 | dom96 | oh |
| 13:34:18 | dom96 | try nimkernel :P |
| 13:34:30 | dom96 | jester needs fixing I think |
| 13:34:33 | dom96 | you could fix it and make a PR |
| 13:34:51 | dom96 | do the C sources work now though? |
| 13:35:02 | * | wan quit (Quit: Lost terminal) |
| 13:35:49 | EXetoC | yes. I pulled in the csources directory and then followed the instructions |
| 13:36:03 | dom96 | good |
| 13:43:55 | EXetoC | dom96: is there a lack of i686 resources since i586 is used? or is it just easier to target? |
| 13:45:21 | dom96 | I have no idea. |
| 13:46:02 | EXetoC | I mean with regards to nimkernel |
| 13:46:45 | * | OrionPK joined #nimrod |
| 13:47:44 | dom96 | EXetoC: oh, er, can't even remember lol. I think it is easier. The OSDev wiki might have more info. |
| 13:48:15 | EXetoC | I assume it doesn't conflict with the gcc package |
| 13:48:17 | EXetoC | ok |
| 13:49:00 | dom96 | EXetoC: are you on linux 64bit? |
| 13:49:26 | EXetoC | nevermind. I didn't consider /usr/local |
| 13:49:28 | EXetoC | yes |
| 13:49:49 | dom96 | ahh, nvm |
| 13:49:50 | dom96 | brb |
| 13:55:09 | * | xenagi joined #nimrod |
| 13:59:14 | EXetoC | dom96: "/tmp/cc5lsXUC.s:221: Error: invalid instruction suffix for `push'" |
| 14:01:38 | EXetoC | I think that out-of-bounds error is triggered at compile-time |
| 14:02:10 | * | io2 quit (Quit: ...take irc away, what are you? genius, billionaire, playboy, philanthropist) |
| 14:16:21 | * | BitPuffi1 joined #nimrod |
| 14:21:08 | * | BitPuffi1 quit (Ping timeout: 252 seconds) |
| 14:30:13 | dom96 | EXetoC: Yeah, the compiler now catches the out-of-bounds error at compile-time |
| 14:30:25 | dom96 | never seen that invalid instruction error before though |
| 14:31:00 | * | Matthias247 joined #nimrod |
| 14:32:17 | dom96 | EXetoC: that happens for nimkernel right? |
| 14:32:44 | EXetoC | dom96: yes |
| 14:45:33 | dom96 | EXetoC: wrong assembler perhaps |
| 14:50:46 | * | wan joined #nimrod |
| 15:20:28 | * | bastian_ joined #nimrod |
| 15:21:26 | bastian_ | congrats on releasing 0.9.4! |
| 15:23:22 | dom96 | bastian_: thanks! |
| 15:23:39 | dom96 | Still working on getting the binaries on the website + some other adjustments |
| 15:30:09 | renesac | dom96, what about those name changes on the docs tab that we talked about? |
| 15:30:38 | renesac | I should have done a pull request yesterday... |
| 15:33:34 | dom96 | renesac: I can still do that |
| 15:33:51 | dom96 | renesac: make a PR now |
| 15:34:34 | renesac | ok |
| 15:37:28 | Varriount | dom96: Installers are being uploaded. I'll be uploading the zip files in a minute |
| 15:39:22 | renesac | the Source code filters and Term rewriting macro should be inside the manual |
| 15:39:28 | renesac | but I will not mess with it now |
| 15:39:38 | renesac | should I send a pull request to devel or master? |
| 15:40:12 | dom96 | devel |
| 15:41:38 | dom96 | Varriount: let me know when they upload |
| 15:42:23 | renesac | https://github.com/Araq/Nimrod/pull/1129 |
| 15:43:22 | renesac | "Compiler Options" was the other alternative I said here |
| 15:43:23 | NimBot | Araq/Nimrod devel c1c889b ReneSac [+0 ±1 -0]: Rearrangements on the webpage's docs tab. |
| 15:43:23 | NimBot | Araq/Nimrod devel e6675bd Dominik Picheta [+0 ±1 -0]: Merge pull request #1129 from ReneSac/devel... 2 more lines |
| 15:43:28 | dom96 | renesac: great, thanks |
| 15:44:17 | Varriount | dom96: Installers are done uploading |
| 15:44:41 | Varriount | dom96: What kinds of archives should I make? 7zip, zip, tar.gz? |
| 15:44:50 | dom96 | Varriount: zip |
| 15:44:55 | renesac | I think it was a zip last time |
| 15:49:58 | dom96 | Varriount: So "slim" is the "gamera" edition? |
| 15:50:07 | Varriount | I guess. |
| 15:50:18 | Varriount | It's the version without the bundled mingw |
| 15:50:27 | dom96 | Varriount: According to the old downloads page the gamera edition includes a minimal gcc |
| 15:50:45 | Varriount | I wasn't able to get the trimgcc executable working. |
| 15:51:47 | Varriount | dom96: Or rather, the trimgcc executable was a bit too... effective. It deleted the entire gcc folder, or threw an error. |
| 15:52:00 | dom96 | lol |
| 15:52:02 | dom96 | oh well |
| 15:52:17 | dom96 | What does 'full' include? |
| 15:52:29 | Varriount | the mingw distribution |
| 15:52:52 | dom96 | what about babel? |
| 15:53:11 | Varriount | Included in everything |
| 15:53:30 | Varriount | As are the dlls. |
| 15:53:34 | dom96 | ahh, even in slim? |
| 15:53:46 | Varriount | Yeah. It doesn't contribute greatly to file size. |
| 15:53:51 | dom96 | good |
| 15:54:06 | dom96 | Installing slim works |
| 15:54:08 | dom96 | Trying full now |
| 15:55:29 | dom96 | hrm, the rendered docs are missing |
| 15:55:51 | Varriount | I didn't know that I was supposed to generate them. |
| 15:56:07 | Varriount | Tell me what commands need to be run, and I'll do it. |
| 15:56:38 | dom96 | well, docgen on Windows is a bit broken |
| 15:57:10 | dom96 | where is babel? I don't see it in bin/? |
| 15:57:19 | Varriount | The exe should be in there. |
| 15:57:26 | dom96 | it's not |
| 15:57:32 | Varriount | Gah. |
| 15:57:52 | Varriount | dom96: Give me a bit to add some stuff to my batch files. |
| 15:58:06 | dom96 | ok, I need to get you the rendered docs anyway |
| 15:58:23 | dom96 | Are you sure you can't slim down mingw? |
| 15:58:30 | dom96 | The whole thing is 80MB |
| 15:58:42 | dom96 | And it seems to include ada and fortran compilers |
| 15:59:06 | dom96 | It also includes python for some reason... |
| 15:59:53 | Varriount | dom96: I'll try. |
| 16:00:32 | * | BitPuffi1 joined #nimrod |
| 16:00:43 | Varriount | dom96: It includes python headers and dll. |
| 16:01:11 | renesac | and how big the full version would be compressed with 7zip LZMA2 Ultra? |
| 16:01:56 | Varriount | renesac: Lemme see |
| 16:02:08 | * | Demos joined #nimrod |
| 16:03:21 | dom96 | Varriount: well, the old one seems to include full gcc too |
| 16:03:33 | dom96 | Probably safer to just leave it |
| 16:04:16 | dom96 | Varriount: oh, you need to change the config |
| 16:04:22 | dom96 | before you build the installers too |
| 16:04:28 | dom96 | uncomment line 88 |
| 16:05:58 | dom96 | Varriount: please try to make slim as it was in the last release though |
| 16:06:43 | dom96 | Varriount: Install this and see what the dist folder contains: http://nimrod-lang.org/download/nimrod_gamera_0.9.2.exe |
| 16:07:31 | Varriount | dom96: I've /tried/ |
| 16:08:33 | dom96 | Varriount: ok, here is what we'll do then. Make a slim installer with no mingw, i'll add it to the download page and then if Araq wants he can help you create a "gamera" edition with a minimal mingw. |
| 16:11:03 | Demos | yay! I just built nimrod using vcc |
| 16:11:52 | Demos | also, it looks like devel is versioned at 0.9.4 instead of 0.9.5 |
| 16:13:04 | dom96 | we have bigger fish to fry right now :P |
| 16:14:26 | * | q66 quit (Ping timeout: 252 seconds) |
| 16:14:43 | * | vbtt joined #nimrod |
| 16:15:23 | dom96 | You guys think .tar.gz is more appropriate for Linux? |
| 16:15:33 | Varriount | Yes. |
| 16:15:35 | vbtt | Hello friends |
| 16:15:44 | Varriount | Hi vbtt. |
| 16:15:51 | vbtt | congrats on the new release |
| 16:16:15 | Varriount | vbtt: Well, atm, we are generating the releases. |
| 16:16:28 | dom96 | I wish Araq didn't upload the new website... |
| 16:16:33 | vbtt | ok. I saw it on the website news |
| 16:18:21 | vbtt | the async stuff looks cool |
| 16:18:57 | dom96 | vbtt: I'm glad you think so :) |
| 16:19:22 | vbtt | does the calling thread just sleep in await? |
| 16:20:20 | dom96 | no, it yields |
| 16:20:39 | dom96 | it's all implemented using iterators |
| 16:21:39 | dom96 | yielding gives back control to the dispatcher which is basically an event loop |
| 16:21:51 | vbtt | So another async proc can run while one async proc is waiting? Using the same thread? |
| 16:22:19 | dom96 | yep |
| 16:22:27 | vbtt | cos that's awesome! |
| 16:22:56 | vbtt | it's kind of like co routines then |
| 16:23:03 | dom96 | yep |
| 16:23:12 | dom96 | The iterators are pretty much coroutines |
| 16:23:38 | vbtt | except the have to be tagged with the async pragma |
| 16:24:58 | BitPuffi1 | dom96: tar.gz/bz2/xz either of those is fine |
| 16:25:08 | dom96 | BitPuffi1: oh cool, you're here. |
| 16:25:16 | dom96 | BitPuffi1: Wanna provide me with mac binaries? |
| 16:25:21 | BitPuffi1 | whereelse would I be |
| 16:25:22 | BitPuffi1 | sure |
| 16:25:26 | dom96 | cool |
| 16:26:12 | BitPuffi1 | have we fixed enough bugs to release? |
| 16:26:32 | BitPuffi1 | slash Araq |
| 16:26:34 | dom96 | who knows :P |
| 16:26:42 | BitPuffi1 | >_> |
| 16:26:53 | * | q66 joined #nimrod |
| 16:26:53 | * | q66 quit (Changing host) |
| 16:26:54 | * | q66 joined #nimrod |
| 16:27:06 | dom96 | gah, i'm copying binaries around. I hope I don't mix up the platforms for some of them... |
| 16:27:16 | BitPuffi1 | lol |
| 16:27:28 | BitPuffi1 | I think there is commands to tell which platform they are for |
| 16:27:30 | BitPuffi1 | not sre |
| 16:27:34 | dom96 | yeah |
| 16:27:35 | dom96 | file |
| 16:27:47 | BitPuffi1 | yeah I was thinking of file |
| 16:27:53 | BitPuffi1 | was wondering if it knew about binaries tho |
| 16:28:41 | dom96 | I wonder how many people out there actually use ppc64 |
| 16:29:55 | BitPuffi1 | I do, don't you? |
| 16:29:58 | BitPuffi1 | thought everybody did |
| 16:30:12 | dom96 | nah, i'm stuck on arm |
| 16:30:19 | dom96 | hrm |
| 16:30:25 | BitPuffi1 | since when do you use arm |
| 16:30:33 | BitPuffi1 | other than on phone |
| 16:30:37 | dom96 | Can I be bothered creating an arm binary |
| 16:30:47 | Varriount | dom96: https://gist.github.com/Varriount/11147952 |
| 16:31:01 | Varriount | Imagine if I had to do that by hand. |
| 16:31:32 | dom96 | Varriount: nice |
| 16:31:45 | dom96 | robocopy sounds cool |
| 16:31:49 | BitPuffi1 | is koch install fixed yet btw? |
| 16:31:51 | * | brihat left #nimrod (#nimrod) |
| 16:31:59 | dom96 | BitPuffi1: was it broken? |
| 16:32:10 | BitPuffi1 | yes it was extremely broken |
| 16:32:23 | BitPuffi1 | ./koch install /usr/local |
| 16:32:30 | BitPuffi1 | will create /usr/local/nimrod |
| 16:32:33 | Varriount | dom96: It breaks the Windows tradition of boring names. |
| 16:32:33 | BitPuffi1 | it should work like prefix |
| 16:32:43 | dom96 | oh shit |
| 16:32:46 | Varriount | dom96: It stands for Robust File Copy |
| 16:33:01 | dom96 | We should have changed where it puts configs in /etc/ |
| 16:33:23 | dom96 | BitPuffi1: Is there an issue on github for this? |
| 16:33:38 | BitPuffi1 | I don't know, it's been around since I started using nimrod |
| 16:33:48 | BitPuffi1 | Demos is aware of it aswell |
| 16:33:55 | Varriount | dom96: What about the docs? |
| 16:33:57 | BitPuffi1 | we were talking about it on the VNUG the other day |
| 16:34:04 | BitPuffi1 | so there is a good reason to be in there ;) |
| 16:34:28 | dom96 | Varriount: working on it |
| 16:34:31 | EXetoC | as well |
| 16:35:49 | Varriount | dom96: Personally, if it was up to me, I would just have a link to Mingw-w64 on the downloads page, and do without the included distribution altogether. But w/e |
| 16:36:38 | dom96 | Varriount: http://picheta.me/private/docs.zip |
| 16:36:45 | dom96 | Varriount: The lower the barrier to entry the better. |
| 16:36:54 | dom96 | Hunting down dependencies is not fun. |
| 16:37:18 | Demos | wow, so asismp has a "boost workaround mode" if you do not want to use boost, I tried it and it will not compile because they forgot to include a header that happens to be included by boost |
| 16:38:04 | * | shodan45 joined #nimrod |
| 16:38:22 | BitPuffi1 | dom96: I guess we'll have to make the command like ./koch install -d:installConfig /usr/local |
| 16:38:34 | BitPuffi1 | since maybe you are installing it in ~/nimrod or something |
| 16:38:39 | dom96 | BitPuffi1: Where are we with those binaries? |
| 16:38:43 | BitPuffi1 | then it would be weird if it installed the config in /etc |
| 16:38:47 | BitPuffi1 | dom96: ? |
| 16:38:49 | dom96 | BitPuffi1: Could I also get c2nim, babel and nimgrep please? |
| 16:38:56 | BitPuffi1 | ah |
| 16:38:58 | BitPuffi1 | I'm eating breakfast |
| 16:39:04 | dom96 | at 6pm? |
| 16:39:07 | BitPuffi1 | and I didn't know you wanted me to start |
| 16:39:13 | dom96 | you're living on the edge bro |
| 16:39:21 | BitPuffi1 | it's actualy 6:39 pm get it right |
| 16:39:32 | EXetoC | burgers probably |
| 16:39:35 | BitPuffi1 | well you know me |
| 16:39:38 | BitPuffi1 | I'm innovative |
| 16:41:47 | BitPuffi1 | dom96: which branch should I build from |
| 16:42:05 | dom96 | BitPuffi1: devel or master, it doesn't matter, they're both the same |
| 16:42:27 | BitPuffi1 | I see |
| 16:43:27 | BitPuffi1 | does it matter if I can't get the executables to be equal dom96? |
| 16:43:32 | dom96 | no |
| 16:43:44 | BitPuffi1 | why does that happen btw |
| 16:43:46 | dom96 | but test them to see if they can compile something :P |
| 16:43:48 | BitPuffi1 | what is it supposed to mean |
| 16:43:55 | BitPuffi1 | ah |
| 16:43:57 | BitPuffi1 | got them equal |
| 16:44:04 | dom96 | it's likely because it includes the compile time in the binary |
| 16:44:05 | BitPuffi1 | dom96: well they can build the compiler |
| 16:44:07 | BitPuffi1 | :P |
| 16:44:21 | dom96 | Sometimes that gets optimised away, sometimes not or something |
| 16:44:29 | dom96 | BitPuffi1: good |
| 16:46:59 | * | vbtt quit (Quit: Bye) |
| 16:47:06 | Varriount | The executables are supposed to be equal for safety and security |
| 16:47:50 | bastian_ | if t is a TTable[int, seq[int]]: any shorthand for doing add(t[a], b)? atm. i'm using: var x = t[a]; add(x, b); t[a] = x |
| 16:47:55 | Varriount | For safety, to catch any malformed executable code, and for security, to prevent polymorphic additions. |
| 16:48:37 | dom96 | bastian_: t.mget(a).add(b) |
| 16:48:49 | BitPuffi1 | dom96: what about nimweb and nimrepl etc |
| 16:48:57 | dom96 | BitPuffi1: don't need them |
| 16:49:14 | BitPuffi1 | dom96: I'll just babel install babel and copy the binary |
| 16:49:22 | bastian_ | dom96: cool, thanks, didn't see that one |
| 16:49:33 | renesac | Varriount, you should use maximum compression options in your script, not the default ones |
| 16:49:35 | renesac | :P |
| 16:49:52 | BitPuffi1 | ls rel |
| 16:49:54 | BitPuffi1 | babel c2nim nimgrep nimrod |
| 16:49:56 | BitPuffi1 | dom96: misssing anything ? |
| 16:50:14 | renesac | for zip and 7z it can only make the decompression faster |
| 16:50:14 | dom96 | nope |
| 16:50:59 | renesac | but in 7z you just have to take care of the size of the dictionary |
| 16:51:19 | renesac | but everybody has at least 64mb free ram nowadays |
| 16:51:59 | Varriount | renesac: Give me the correct command line then. |
| 16:55:00 | renesac | I think you need to add -mx9 to the 7z command line |
| 16:55:03 | BitPuffi1 | dom96: sending on skype |
| 16:55:03 | renesac | http://sevenzip.sourceforge.jp/chm/cmdline/switches/method.htm |
| 16:55:10 | dom96 | BitPuffi1: wait |
| 16:55:13 | BitPuffi1 | dom96: you should re-package it though to be consistent with the other releases |
| 16:55:25 | dom96 | BitPuffi1: of course |
| 16:55:40 | BitPuffi1 | well |
| 16:55:42 | BitPuffi1 | you'd have to |
| 16:55:46 | BitPuffi1 | to include the stdlib and stuff :P |
| 16:55:49 | BitPuffi1 | which I don't |
| 16:56:04 | BitPuffi1 | don't know why I sent them as a zip though, could have done a tar jizz |
| 16:56:20 | dom96 | what's preferred on mac, tar or zip or something else? |
| 16:57:29 | dom96 | BitPuffi1: what's your platform? |
| 16:58:09 | xenagi | PDP :P |
| 16:58:53 | dom96 | xenagi: PDP? |
| 16:59:04 | dom96 | BitPuffi1: And by platform I mean CPU. |
| 16:59:25 | xenagi | dom96, http://en.wikipedia.org/wiki/Pdp8 |
| 16:59:30 | BitPuffi1 | dom96: x86_64 |
| 16:59:36 | BitPuffi1 | dom96: zip |
| 16:59:45 | dom96 | BitPuffi1: k |
| 16:59:49 | BitPuffi1 | well |
| 16:59:55 | BitPuffi1 | unless I have to be explicit to get 64bit |
| 16:59:57 | BitPuffi1 | but I doubt it |
| 17:00:06 | dom96 | BitPuffi1: run nimrod -v |
| 17:00:20 | BitPuffi1 | yeah |
| 17:00:22 | BitPuffi1 | amd64 |
| 17:00:30 | BitPuffi1 | imrod Compiler Version 0.9.4 (2014-04-21) [MacOSX: amd64] |
| 17:00:38 | dom96 | BitPuffi1: good |
| 17:01:39 | dom96 | BitPuffi1: According to wikipedia the default archive utility in mac os supports tar.gz |
| 17:01:42 | dom96 | So i'll go with tht |
| 17:01:44 | dom96 | *that |
| 17:01:54 | BitPuffi1 | dom96: I'm pretty sure i had to install something for it so no |
| 17:01:57 | BitPuffi1 | dom96: go with zip |
| 17:02:04 | BitPuffi1 | dom96: every mac user uses zip |
| 17:02:05 | BitPuffi1 | fucking retards |
| 17:02:15 | xenagi | lol |
| 17:02:17 | dom96 | hrm "It works as a front-end for command-line programs that are part of OS X." |
| 17:02:20 | dom96 | I guess you're right |
| 17:02:25 | Varriount | dom96: Documentation? |
| 17:02:26 | BitPuffi1 | yeah |
| 17:02:31 | BitPuffi1 | the tar command is available |
| 17:02:32 | dom96 | Varriount: I gave you a link |
| 17:02:43 | dom96 | <dom96> Varriount: http://picheta.me/private/docs.zip |
| 17:05:31 | * | dom96 is even including SHA256 checksums |
| 17:05:40 | BitPuffi1 | that's good |
| 17:05:44 | BitPuffi1 | why not sha512? lol |
| 17:05:49 | BitPuffi1 | WHILE WE'RE AT IT |
| 17:05:55 | dom96 | sha2048 |
| 17:06:04 | dom96 | sha2^2048 |
| 17:06:07 | dom96 | YEAH |
| 17:06:17 | BitPuffi1 | yeah!!! |
| 17:07:05 | BitPuffi1 | or why not SHA4710400??!??!?! |
| 17:07:14 | BitPuffi1 | that's my favorite |
| 17:07:21 | dom96 | awwk, rust has Mac OS X binary installers |
| 17:07:29 | BitPuffi1 | :o |
| 17:07:44 | BitPuffi1 | you know what rust also has? |
| 17:07:55 | dom96 | Full-time devs? :( |
| 17:08:08 | BitPuffi1 | a shitty language!! |
| 17:08:16 | BitPuffi1 | where da hood where da hood where da hood at |
| 17:09:47 | Varriount | dom96: Anything else I need to include/tweak? |
| 17:10:19 | dom96 | Varriount: Let's go through a checklist: html docs? tweaked the config? |
| 17:10:32 | dom96 | babel? |
| 17:11:08 | dom96 | argh |
| 17:11:10 | dom96 | damn |
| 17:11:16 | dom96 | I guess I should include koch |
| 17:11:16 | Varriount | Got all those, except for the config. |
| 17:11:23 | dom96 | BitPuffi1: Sorry, I need koch. |
| 17:11:30 | BitPuffi1 | ._. |
| 17:12:08 | dom96 | Varriount: I wonder if it wouldn't make more sense for you to put koch in bin/? |
| 17:13:12 | dom96 | actually nah |
| 17:13:15 | dom96 | let's leave it as it is |
| 17:13:20 | dom96 | I bet koch makes some path assumptions |
| 17:13:28 | BitPuffi1 | dom96: you got it |
| 17:15:15 | dom96 | thx |
| 17:15:48 | * | icebattle quit (Remote host closed the connection) |
| 17:16:22 | * | brson joined #nimrod |
| 17:18:29 | dom96 | well, my dog wants out |
| 17:18:30 | dom96 | brb |
| 17:20:18 | * | Varriount quit (Ping timeout: 240 seconds) |
| 17:22:20 | * | Varriount joined #nimrod |
| 17:25:19 | * | DAddYE joined #nimrod |
| 17:28:50 | Demos | can I tell babel to use vcc? |
| 17:35:52 | dom96 | back |
| 17:36:11 | dom96 | Demos: Why would you need to? babel doesn't compile C sources |
| 17:36:53 | Demos | I was an idiot, do not worry about it |
| 17:39:23 | Demos | hm I am having trouble with passl when using MSVC |
| 17:39:26 | Demos | strange |
| 17:46:07 | Araq | hmm the Copyright note is stuck in 2013 |
| 17:46:59 | renesac | "There are no real uninitialized variables in Nimrod as they are initialized to binary zero" <-- unless you use the {.noInit.} pragma |
| 17:47:34 | Demos | well if you used the noinit pragma then you "know what you are doing" |
| 17:48:15 | renesac | that is a comment for the news piece |
| 17:48:53 | renesac | people may get the wrong idea by this |
| 17:49:38 | dom96 | Araq: If you want a slim installer with a slim gcc now would be the time to tell Varriount |
| 17:52:31 | * | Demos quit (Read error: Connection reset by peer) |
| 17:56:38 | Araq | dom96: oh yeah. can't ship without Gamera |
| 17:57:25 | dom96 | but we can ship without any binaries at all huh? |
| 17:59:22 | Araq | sure |
| 17:59:28 | Araq | it's a "beta release" |
| 17:59:42 | Araq | we should call it the "heart beat release" |
| 18:00:13 | dom96 | Araq: Help Varriount and let's finish this. |
| 18:07:59 | EXetoC | Araq: I'm trying to print a node tree while including flags. tricky |
| 18:08:10 | EXetoC | maybe I also need to try renderTree again |
| 18:08:32 | Araq | the node flags are irrelevant for most things |
| 18:08:57 | EXetoC | I'm looking for the dispatcher flag |
| 18:09:20 | Araq | trying to fix the method forwarding bug? |
| 18:09:27 | EXetoC | method recursion |
| 18:10:06 | EXetoC | "Error: unhandled exception: sym is not accessible" dunno where that comes from :E |
| 18:12:21 | Varriount | Back |
| 18:12:53 | reactormonk | wow, the github notifications thingy is flaring up |
| 18:14:09 | * | uvelichitel left #nimrod ("Textual IRC Client: www.textualapp.com") |
| 18:16:51 | EXetoC | traversing is tricky because you have to figure out if 'sons' is accessible |
| 18:19:21 | Araq | for i in 0 .. < n.safeLen: traverse(n[i]) |
| 18:19:55 | Araq | tricky if you don't know the ways |
| 18:20:02 | reactormonk | Araq, got another idea for float `$`... lemme test |
| 18:21:11 | EXetoC | will try that later. will have to look for something simpler again if I don't get anywhere. I'll look around a little, but maybe we could estimate the difficulty of issues |
| 18:21:45 | NimBot | Araq/Nimrod devel bc79f40 Simon Hafner [+0 ±1 -0]: changed pointer to array in sqlite callback |
| 18:29:26 | Varriount | dom96: It might be prudent to add to the website that Mingw-w64 is recommended over plain Mingw when on Windows. |
| 18:30:42 | reactormonk | how do I compile the tester again? |
| 18:31:08 | Araq | nimrod c tests/testament/tester |
| 18:33:20 | dom96 | Varriount: alright |
| 18:35:30 | reactormonk | Araq, are 10 numbers plenty? |
| 18:35:39 | reactormonk | or 16 |
| 18:36:30 | Araq | reactormonk: IMO we should port what ruby or python use |
| 18:36:43 | Araq | and not use the freaking ansi C stuff |
| 18:36:54 | renesac | I like the sound of 'heart beat release' |
| 18:37:28 | renesac | "Lots of new features have been implemented but most do not fullfill our quality standards." <-- really most? |
| 18:37:39 | reactormonk | Araq, I'd settle for %.16g and then port something more intersting |
| 18:37:42 | renesac | that sounds bad |
| 18:37:59 | reactormonk | renesac, where's that from? |
| 18:38:01 | renesac | and I don't think it is so bad |
| 18:38:09 | renesac | the top of the 0.9.4 announcement |
| 18:38:12 | renesac | in the news |
| 18:38:14 | renesac | tab |
| 18:38:15 | reactormonk | muh |
| 18:38:33 | reactormonk | that sounds REALLY bad |
| 18:38:56 | Araq | renesac: well our quality standard have raised :P |
| 18:39:18 | reactormonk | whoever has access, rewrite that to "Lots of new features have been implemented but most of them are not considered stable yet." |
| 18:39:34 | reactormonk | ... or something better. |
| 18:39:49 | reactormonk | because "not stable yet" says they're rather buggy |
| 18:40:48 | fowl | replace it with "xxx issues have been resolved since 0.9.2" |
| 18:41:08 | renesac | fowl, that is misleading |
| 18:41:13 | fowl | why |
| 18:41:19 | renesac | as many of those issues were due to bugs introduced in the mean time |
| 18:41:54 | renesac | and issues count pull requests and feature requests as well |
| 18:41:58 | dom96 | we already have "1400 changes" |
| 18:42:13 | renesac | that is more abstract, so it is ok |
| 18:42:16 | renesac | :P |
| 18:42:55 | NimBot | Araq/Nimrod float_$ 77a15dc Simon Hafner [+0 ±2 -0]: use %.16g for `$` float |
| 18:43:13 | fowl | reactormonk, why not do what ruby does? |
| 18:43:58 | fowl | 0.51 instead of 5.100000000e-1 |
| 18:45:31 | renesac | Araq, an idea: how about implement slices limited to borrowed pointers? |
| 18:45:39 | renesac | then the GC don't have to care about them |
| 18:45:55 | renesac | and most uses should be covered |
| 18:46:08 | Araq | sure that's nice |
| 18:46:23 | Araq | oh wait, we don't have borrowed pointers |
| 18:46:28 | renesac | yet! |
| 18:46:28 | fowl | reactormonk, you ashould use %.8g for 32bit floats |
| 18:47:16 | renesac | in your tagged pointers documentation you said they could be used to implement this |
| 18:47:35 | Araq | yes |
| 18:48:07 | Varriount | renesac: What do you mean, slicing borrowed pointers? |
| 18:48:16 | renesac | another option, less performant, is to store the base address and a start offset, plus the length of course |
| 18:48:47 | renesac | Varriount, the issue araq has with slices is that they add the problem of tracking interior pointers to the gc logic |
| 18:48:47 | Araq | slices are easy to do anyway, even keeping GC efficiency in mind |
| 18:49:06 | Araq | no, that's an oversimplification |
| 18:50:20 | renesac | what is an oversimplification? |
| 18:50:35 | Araq | "the issue araq has with slices is that they add the problem of tracking interior pointers to the gc logic" |
| 18:51:00 | renesac | I remember you saying that slices in D preclude the gc from being efficient |
| 18:51:07 | renesac | or make it much harder |
| 18:51:37 | Araq | that's correct but only because D and Golang picked a braindead representation |
| 18:52:06 | renesac | a pointer plus a lenght |
| 18:52:16 | Araq | you need to keep 3 words instead of 2. Problem of finding the root solved. |
| 18:52:34 | renesac | yeah, that is the less performant option that I said |
| 18:53:11 | Araq | it is not less performant really. |
| 18:53:46 | Araq | it's simply something people coming from C don't consider. |
| 18:54:13 | renesac | hum, I guess the compiler can optimize the constant add to the base pointer |
| 18:54:32 | Araq | you don't have to do that |
| 18:54:47 | Araq | you can keep (root, interiorPtr, length) |
| 18:55:00 | renesac | oh, of course |
| 18:55:11 | renesac | I wasn't thinking that option |
| 18:55:58 | renesac | yeah, so the only down side is a slightly higher memory overhead, but the GC would have that overhead anyway |
| 18:56:06 | Araq | but yeah, I like slicing restricted to borrowed pointers even better |
| 18:56:07 | * | Varriount wishes his college had a course on gc stuff |
| 18:56:24 | renesac | can't we have both? |
| 18:56:35 | renesac | and a type class that join them? |
| 18:56:54 | fowl | Araq, i see you can now do type X[T] = ref object of TObject but methods still do not work with these |
| 18:57:06 | Araq | "can't we have both?" -- new nimrod slogan? |
| 18:57:25 | renesac | or the borrowed pointers be an compiler optimization? (though it would mess up with the struct organization) |
| 18:57:27 | renesac | lol |
| 18:58:36 | Matthias247 | most times making one thing right is better than having both ;) |
| 18:58:48 | * | brson quit (Ping timeout: 265 seconds) |
| 18:58:49 | BitPuffi1 | Araq: thought that already was the slogan |
| 18:59:09 | BitPuffi1 | in regards to productivity and performance |
| 19:00:45 | * | BitPuffi1 is now known as BitPuffin |
| 19:01:12 | renesac | hum |
| 19:01:34 | renesac | how would a typeclass work, if structs of different sizes can be the same typeclass? |
| 19:01:46 | renesac | like, just a pointer, like the current string |
| 19:01:56 | Araq | a typeclass is just a compile time thing |
| 19:02:13 | Araq | it doesn't allow you to create polymorphic arrays |
| 19:02:25 | renesac | yeah, that is what I was going to ask |
| 19:02:36 | BitPuffin | Araq: like your mum then (revenge) |
| 19:02:46 | dom96 | Varriount: ETA on installers? |
| 19:03:11 | renesac | so the compiler forbids me creating an array with them? |
| 19:03:20 | * | brson joined #nimrod |
| 19:03:43 | Araq | proc foo[T: TypeClass](x: sesq[T]) |
| 19:03:51 | Araq | that's what it comes down to |
| 19:04:01 | Araq | it's eventually still a seq of a single type |
| 19:04:50 | renesac | but if I make a proc that accepts that TypeClass to add things to that array |
| 19:05:00 | renesac | what happens? |
| 19:05:21 | renesac | after it has been added to the array, the compiler can't keep track of the type anymore |
| 19:05:31 | renesac | it would be like an union, w/o a tag |
| 19:05:57 | fowl | renesac, type class is not a c++ interface, its a relation between types |
| 19:06:44 | fowl | either TX is MyTypeClass or TX isnot MyTypeClass |
| 19:07:11 | Araq | fowl: please do me a favour and git bisect the compiler until you find the commit that broke keineschweine |
| 19:07:23 | Araq | I know it runs with the new VM, so that's not the problem |
| 19:07:32 | Araq | *it used to run with the new VM |
| 19:07:36 | Varriount | dom96: Well, Araq wants the minimal installers |
| 19:08:13 | renesac | fowl, so the compiler would reject that code? |
| 19:08:15 | Varriount | But I can give you the other installers right now. |
| 19:08:49 | dom96 | Varriount: ok, are you working on the minimal ones now? |
| 19:09:05 | Varriount | Yes. |
| 19:09:41 | dom96 | it seems Elixir got a new release today too |
| 19:10:01 | fowl | renesac, tr yit |
| 19:10:43 | dom96 | and Scala 2.11.0. |
| 19:11:07 | BitPuffin | probably better to release tomorrew then |
| 19:11:15 | Matthias247 | do they contain eastereggs? ;) |
| 19:11:25 | dom96 | BitPuffin: you're right |
| 19:11:48 | BitPuffin | dom96: of course |
| 19:11:50 | BitPuffin | xD |
| 19:14:18 | dom96 | Varriount: If you have skype you can add me and send me the installers if that's easier for you |
| 19:16:31 | fowl | http://sqlkorma.com/ |
| 19:18:08 | EXetoC | but is tomorrow the optimal announcement day? don't forget to announce at the right time too! |
| 19:18:32 | renesac | fowl, for a seq it gives an instantiation error |
| 19:18:35 | renesac | for an array: |
| 19:18:36 | renesac | Error: internal error: GetUniqueType |
| 19:18:36 | renesac | No stack traceback available |
| 19:18:37 | EXetoC | reddit science |
| 19:18:55 | renesac | I will open an issue on this second one |
| 19:19:39 | Araq | Varriount: I'm failing at producing a small mingw |
| 19:20:13 | Araq | so ... no gamera edition anymore unless you succeed where I am failing |
| 19:24:23 | * | brson quit (Ping timeout: 252 seconds) |
| 19:29:38 | Varriount | dom96: The installers directory now has the updated installers, zip files, and 7z files. |
| 19:29:48 | Varriount | I'm working on the gamera edition at present. |
| 19:30:33 | Araq | Varriount: better make an installer without any GCC and lets update the download section and tell people where to find it |
| 19:30:43 | Varriount | Araq: Already did. |
| 19:31:02 | Araq | we should also host a copy of this |
| 19:31:10 | Varriount | "This"? |
| 19:31:31 | Araq | the recommended mingw version |
| 19:31:51 | Varriount | Why host a copy? |
| 19:32:46 | Araq | the mising SDL_ttf.dll made me nervous |
| 19:33:43 | Varriount | Araq: What? I haven't included SDL with the installers. |
| 19:34:03 | Araq | yeah I know, no worries |
| 19:34:31 | Araq | but that made me more nervous about dependencies :P |
| 19:35:24 | Varriount | Araq: That's what dependancy walker is for. |
| 19:36:06 | dom96 | Araq: Why host it when we have an installer which includes it? |
| 19:36:52 | Varriount | Also, we could just link to the mingw-w64 download on sourceforge. |
| 19:37:30 | Araq | dom96: because later installers won't include it anymore, perhaps |
| 19:37:34 | dom96 | Are we providing 7z archives too? |
| 19:37:57 | Araq | if you want to, I don't mind |
| 19:38:04 | dom96 | not really |
| 19:38:51 | Varriount | dom96: After all the hard work I put into compressing them? Finding each substring by hand? :p |
| 19:39:07 | dom96 | Varriount: Do you want them on the downloads page? |
| 19:39:54 | Varriount | I don't care whether they are there or not. |
| 19:39:54 | EXetoC | Araq: you didn't want to merge my spawn? |
| 19:39:55 | dom96 | I don't want to overload the user with too many options |
| 19:39:58 | EXetoC | +test |
| 19:41:04 | EXetoC | since sysspawn.nim is on master already |
| 19:41:05 | NimBot | Araq/Nimrod devel 03ffc34 Grzegorz Adam Hankiewicz [+0 ±3 -0]: Version switch displays options used during `koch boot` |
| 19:41:05 | NimBot | Araq/Nimrod devel c0338ea Varriount [+0 ±3 -0]: Merge pull request #1118 from gradha/pr_show_boot_options... 2 more lines |
| 19:42:30 | Araq | who is so merge happy here? |
| 19:42:41 | * | Varriount raises his hand |
| 19:42:59 | Araq | "version switch displays options used during 'koch boot'" |
| 19:43:05 | Araq | so tell me |
| 19:43:17 | Araq | do you understand gradha's patch? |
| 19:43:22 | Varriount | Yes. |
| 19:44:04 | Araq | and is it the easiest way to do it? :P |
| 19:44:46 | Varriount | I don't know of any simpler way. |
| 19:54:33 | dom96 | I swear UAC is making these installers start up 10 times slower. |
| 19:55:57 | BitPuffin | UAC<3 |
| 19:56:34 | * | Demos joined #nimrod |
| 19:57:30 | dom96 | Varriount: It looks like you didn't include the html docs... |
| 20:01:11 | dom96 | or babel? |
| 20:07:43 | Varriount | dom96: It must be a inno configuration bug - They're in the dirs on my computer |
| 20:08:18 | dom96 | please fix it |
| 20:10:33 | * | flaviu joined #nimrod |
| 20:16:27 | Araq | fyi: http://dreid.org/2012/03/30/deferreds-are-a-dataflow-abstraction/ |
| 20:16:52 | Araq | dom96: in particular of interest for you |
| 20:19:14 | Varriount | Araq: Could you explain where the installer generator get's its list of files to include? |
| 20:19:29 | Araq | compiler/nimrod.ini |
| 20:19:59 | Varriount | Yeah, but I don't see where that adds the paths, like the bin path |
| 20:23:52 | Araq | ah |
| 20:24:22 | Araq | that's tools/niminst/inno.tmpl |
| 20:24:52 | Araq | but nimrod.ini there is: BinPath: r"bin;dist\mingw\bin;dist" |
| 20:28:10 | Varriount | Araq: I was looking at the wrong file. |
| 20:30:25 | flaviu | Besides performance, wouldn't interpreting the AST raw also work? |
| 20:30:56 | Araq | flaviu: that's what 0.9.2 does |
| 20:33:34 | Varriount | dom96: Is there anything else missing? |
| 20:33:55 | flaviu | What motivated the switch to the register vm? |
| 20:34:14 | dom96 | Varriount: no, babel and docs. Test one of the installers before you upload please. |
| 20:36:20 | Araq | when I started working on the new VM I thought I had nothing more important to do ... |
| 20:36:57 | Araq | also I underestimated the effort it took by about 5 months |
| 20:37:35 | Araq | the old interpreter also had many bugs |
| 20:37:49 | Araq | the new VM has ... different bugs. yay. |
| 20:38:01 | renesac | Araq, now for 0.9.6 you will do a JIT compiler? |
| 20:38:05 | renesac | :D |
| 20:38:12 | Araq | no. |
| 20:38:19 | renesac | aww |
| 20:38:21 | renesac | XD |
| 20:38:26 | flaviu | I'm considering making a VM with code generation to keep the code simple. |
| 20:38:45 | Araq | but for nimrod 2 I am considering to simply reuse the C backend and invoke the C compiler at compile-time |
| 20:38:55 | Araq | with some aggressive caching of everything |
| 20:39:07 | flaviu | But like most things I consider doing, I probably won't ever finish it. |
| 20:39:08 | Varriount | Nimrod 2? |
| 20:39:41 | Araq | Varriount: what do you think? Nimrod 1 will be the end of all things? |
| 20:41:45 | Araq | flaviu: however, the new VM is much faster than the old one |
| 20:42:22 | Araq | I got a report that's like "the 1 minute compile time pause went away" |
| 20:43:55 | renesac | the new command syntax is very nice for using 'nimrod i' |
| 20:44:10 | renesac | too bad 'nimrod i' still isn't very nice to use |
| 20:45:30 | Araq | too bad nobody cares enough to fix it |
| 20:45:44 | renesac | yeah.. not the highest priority |
| 20:47:25 | renesac | oh, 'for' loops are now working |
| 20:47:35 | renesac | too bad directional arrows aren't |
| 20:47:57 | Araq | build with gnuReadline support |
| 20:48:06 | Araq | to get the arrow support |
| 20:48:16 | * | brson joined #nimrod |
| 20:48:27 | Araq | and wether for loops work or not depends on the phase of the moon |
| 20:48:40 | renesac | hum |
| 20:49:02 | flaviu | Araq: Will you be deprecating the TinyC interpreter? |
| 20:49:22 | Araq | that's effectively dead, flaviu |
| 20:49:34 | Araq | TinyC doesn't run properly anywhere afaict |
| 20:49:56 | Varriount | Why did you include it in the first place? |
| 20:50:04 | flaviu | So a pull request removing it would be accepted? |
| 20:50:05 | Demos | kinda a neat idea |
| 20:50:32 | Demos | but I guess if you have something that implements the C standard it does not matter, since you gotta copy at least the intersection of MSVC and GCC |
| 20:50:48 | Araq | flaviu: I prefer you test it first, spend 15 minutes to try to get it to work |
| 20:50:55 | Araq | and then remove it |
| 20:51:09 | flaviu | Ok |
| 20:51:19 | Varriount | Araq: What was the motive to include TinyC in the first place? |
| 20:51:36 | Araq | Varriount: it used to work rather nicely |
| 20:53:54 | renesac | see the features section: http://bellard.org/tcc/ |
| 20:54:40 | dom96 | Varriount: One of the main motives AFAIK was its speed. |
| 20:55:14 | dom96 | A long time ago I wanted a REPL really badly, and Araq told me to use TinyC as a compromise. |
| 20:55:27 | fowl | Araq, issue 907 keeps keineschweine from compiling for me |
| 20:55:29 | renesac | and a 100kb executable |
| 20:56:27 | Araq | fowl: I fixed that, should be in master even |
| 20:57:27 | renesac | oh, much better with gnuReadline |
| 20:59:17 | Araq | you know, on windows the arrow keys simply work these days. |
| 20:59:42 | renesac | why gnuReadline isn't default for linux? |
| 21:00:08 | Araq | SurrealLinux version 8 doesn't include GNU readline |
| 21:00:17 | renesac | hum |
| 21:00:25 | * | flaviu quit (Read error: Connection reset by peer) |
| 21:00:46 | dom96 | hah |
| 21:01:21 | Araq | and if you think I'm kidding, read this: http://forum.nimrod-lang.org/t/420 |
| 21:01:55 | Araq | "received a problem report stating that Nimrod misbehave when procfs is not present" |
| 21:02:06 | Araq | I thought procfs is a feature of the OS |
| 21:02:16 | Araq | but apparently people like to disable it |
| 21:02:32 | * | [1]Endy quit (Ping timeout: 255 seconds) |
| 21:02:45 | renesac | yeah... |
| 21:04:58 | renesac | ok, trying to import 'math' and using 'sum' crashed nimrod i |
| 21:05:34 | renesac | well, importing math raised lots of errors anyway |
| 21:06:01 | renesac | at the start it also raises lots of errors about 'importc' |
| 21:06:10 | reactormonk | fowl, because it's slightly more complicated |
| 21:06:28 | * | flaviu joined #nimrod |
| 21:09:55 | Araq | EXetoC: your test is risky, I am thinking of a solution. when I break spawn, things can deadlock and so the whole build process. |
| 21:10:13 | dom96 | Araq: Add a timeout to the tester |
| 21:10:37 | Araq | I guess we should do that anyway |
| 21:10:44 | Varriount | dom96: Installers updated. |
| 21:10:48 | renesac | I wanted a verbosity mode that only gave me the compilation statistics but not lots of hints about paths that babel adds |
| 21:10:52 | dom96 | Varriount: yay |
| 21:11:04 | renesac | is there someway to disable those path hints? |
| 21:11:11 | dom96 | Araq: yes, but enable it only if a test specifies it |
| 21:11:40 | Araq | dom96: why? in theory the compiler itself can go into an endless loop |
| 21:11:48 | Araq | so even a 1 line test is not safe |
| 21:12:13 | dom96 | Araq: How can you reliably determine that you are not terminating the test too soon due to a timeout? |
| 21:12:36 | flaviu | If a test is hitting a timeout, it needs to be reworked |
| 21:12:51 | Araq | yeah what flaviu says |
| 21:14:03 | dom96 | Varriount: noooo |
| 21:14:13 | dom96 | Varriount: Now the config file is wrong! |
| 21:15:41 | Araq | and now you know why we never release. it's lots of hard work to even build an installer. |
| 21:15:58 | Araq | in theory it's all automated. |
| 21:16:08 | dom96 | we need to automate it |
| 21:16:11 | Araq | in practice every single release is different |
| 21:16:18 | dom96 | and create tests to verify it |
| 21:16:36 | reactormonk | Araq, windows installer? |
| 21:16:57 | Araq | reactormonk: not only. producing a .zip for linux has many of the same problems |
| 21:18:42 | * | renesac quit (Ping timeout: 245 seconds) |
| 21:18:48 | reactormonk | :-( |
| 21:19:33 | dom96 | well, I'm going to upload the linux and mac os X binaries |
| 21:19:37 | dom96 | Somebody should check them out |
| 21:19:40 | dom96 | in case I made some mistake |
| 21:20:23 | flaviu | `nimrod run` fails, it needs a filename. `nimrod run ./test1.nim` also fails, but because it can't find the run command |
| 21:21:20 | Araq | flaviu: did you notice how we produce full debug information in the VM? |
| 21:21:59 | flaviu | No, I saw it, but I didn't look into it |
| 21:22:14 | Araq | that feature alone is enough of a reason to make register based VMs superior and much easier to implement, IMO |
| 21:22:43 | * | renesac joined #nimrod |
| 21:23:07 | Araq | full debug information without slowing down execution |
| 21:23:34 | flaviu | Well, in my hypothetical VM, registers would still be used, but the stack would be the scratch-pad for calculations. |
| 21:23:34 | flaviu | Wow, thats awesome |
| 21:27:23 | dom96 | http://nimrod-lang.org/download/nimrod_0.9.4_linux_amd64.tar.gz |
| 21:27:39 | dom96 | http://http://nimrod-lang.org/download/nimrod_0.9.4_linux_i386.tar.gz |
| 21:27:43 | dom96 | er oops |
| 21:27:47 | dom96 | http://nimrod-lang.org/download/nimrod_0.9.4_linux_i386.tar.gz |
| 21:27:50 | dom96 | please test |
| 21:27:53 | Araq | .tar.gz ? |
| 21:27:59 | dom96 | yes. |
| 21:28:06 | dom96 | That's what Linux users like |
| 21:28:30 | Araq | I'm a Linux user and don't like tar.gz ... |
| 21:28:33 | Araq | :P |
| 21:28:40 | dom96 | why? |
| 21:28:52 | Araq | because it's stupid |
| 21:29:21 | Araq | and some tools produce a .tar from the .tar.gz |
| 21:29:30 | Araq | and then you have to untar it again. pathetic. |
| 21:29:40 | reactormonk | Araq, tar xf |
| 21:30:19 | Araq | nobody gives a fuck about .tar, it's an academic exercise to split archiving and zipping |
| 21:30:24 | dom96 | Araq: It's easier for package managers to use the .tar.fz |
| 21:30:31 | dom96 | *gz |
| 21:30:40 | Araq | dom96: now that is at least a reason |
| 21:30:41 | dom96 | tar is more common than unzip AFAIK |
| 21:32:32 | dom96 | Araq: Don't worry. Windows and Mac OS X is a zip. |
| 21:33:09 | Araq | I think Mac OS X wants a .dmg instead |
| 21:33:48 | reactormonk | yup |
| 21:34:23 | Araq | but maybe they found a way to make iTunes read zip files |
| 21:39:19 | flaviu | Araq: aunpack. I don't care what format my archives are |
| 21:45:25 | EXetoC | Araq: "echo int16(40000)". I'm trying to get this to generate an overflow error. where do I look? |
| 21:46:17 | Araq | EXetoC: most likely semfold |
| 21:46:22 | EXetoC | that just prints the 'int' value 40000 |
| 21:46:24 | EXetoC | right |
| 22:01:09 | Araq | dom96: I tried the 64bit linux tar.gz |
| 22:01:19 | Araq | koch says it's Version 0.9.3 |
| 22:01:23 | Araq | not good. |
| 22:01:34 | dom96 | koch? |
| 22:01:42 | Araq | yes? |
| 22:01:53 | Araq | oh wait a sec |
| 22:02:22 | Araq | ok, nimrod says it's 0.9.4 |
| 22:02:32 | Araq | so no showstopper |
| 22:02:42 | dom96 | good |
| 22:04:27 | Araq | well it works for me |
| 22:06:14 | dom96 | Varriount: I moved the zips so don't panic that they're gone |
| 22:11:38 | Skrylar | meep |
| 22:17:30 | dom96 | Araq: The rst cheatsheet has been overwritten by the rst module docs... |
| 22:17:53 | renesac | EXetoC, https://github.com/Araq/Nimrod/issues/936 |
| 22:18:19 | dom96 | Varriount: You still alive? |
| 22:21:12 | fowl | hay |
| 22:21:32 | fowl | will this work right: when nimrodversion < "0.9.3" |
| 22:22:07 | Araq | dom96: ok I have a backup |
| 22:22:13 | Araq | let me revert that file |
| 22:22:37 | dom96 | is the rst module listed in lib.html? |
| 22:23:11 | Araq | yes. damn |
| 22:23:30 | fowl | lol |
| 22:30:53 | Araq | well I can fix that |
| 22:31:27 | fowl | hrm i wonder what i use 'gl' for in keineschweine |
| 22:33:47 | Araq | dom96: I fixed it |
| 22:33:57 | Araq | but create an issue for nimforum to rename rst.html |
| 22:34:11 | Araq | to rst_cheatsheet or something |
| 22:36:23 | fowl | Araq, ks compiles now |
| 22:40:01 | Araq | fowl: I'm talking about the keine_schweine test in tests/manyloc |
| 22:40:14 | Araq | and that doesn't work anymore, at least on my computer |
| 22:49:32 | flaviu | Araq: You're planning a llvm backend? |
| 22:50:00 | Araq | no, that's old code you're looking at |
| 22:50:18 | Araq | llvm is lots of work and doesn't solve many problems |
| 22:50:19 | flaviu | Old docs actually, docs/nimrodc |
| 22:50:36 | Araq | oh that one |
| 22:50:52 | Araq | well it helps to keep it in mind when designing the FFI |
| 22:52:04 | flaviu | Well, the llvm solves the whole interpreter issue, and sidesteps most of undefined c beheivor |
| 22:53:18 | Araq | linking against a massive C++ project introduces its own set of problems |
| 22:54:09 | Araq | undefined C behaviour is a problem but the C spec is too insane and people noticed so I'm positive this will improve |
| 22:56:14 | * | Matthias247 quit (Read error: Connection reset by peer) |
| 22:58:01 | * | darkf joined #nimrod |
| 23:16:56 | Skrylar | hrm |
| 23:17:15 | Skrylar | seeing this notecard program reminds me of how weird it is that certain data trees are really simple yet they're not put together |
| 23:17:19 | dom96 | Shall we post to reddit and HN guys? |
| 23:17:40 | flaviu | dom96: I'm ready |
| 23:19:15 | Skrylar | i've been thinking about a data structure which basically revolved around having a single 'notecard' record and then all the other data related to it is bolted on separately, like a.. i donno |
| 23:19:19 | Skrylar | loosely relational something |
| 23:19:41 | flaviu | A tree? |
| 23:19:49 | Demos | mongoDB? |
| 23:20:28 | Skrylar | Demos: somewhat |
| 23:21:30 | Skrylar | Ecco Pro had it so you could edit the fields for the database, which they used to hack in things like making a boolean column to represent a tag/folder; articydraft has something similar where you define a 'schema' which is something like "research item" for an rts or "main character" and it bolts those fields on separately |
| 23:22:13 | flaviu | Prototypes? |
| 23:22:15 | dom96 | flaviu: Would you mind submitting it to reddit and HN? |
| 23:22:25 | flaviu | I only have a reddit account |
| 23:22:27 | Skrylar | flaviu: yeah, i was thinking something like those |
| 23:22:48 | dom96 | flaviu: That's ok, can you submit it to reddit? |
| 23:23:40 | dom96 | If so, use the following URL: http://nimrod-lang.org/news.html#Z2014-04-21-version-0-9-4-released |
| 23:23:40 | Skrylar | the whole idea was that it would be able to store an item, then have a bibliography tag attached so it could show up in a reference manager view, but then you could also put it inside a concept map so it had a separate map tag, so basically you can organize this same bullet point in a few different folders without leaving the program |
| 23:24:28 | Skrylar | mongodb-lite (does one exist?) or janking prototypes on to sqlite would probably do it |
| 23:25:34 | flaviu | I'm trying to, but I get an error. Give me a few minutes. |
| 23:25:53 | dom96 | ok. I'll submit it to HN |
| 23:27:17 | dom96 | https://news.ycombinator.com/item?id=7624689 |
| 23:29:25 | flaviu | Hmm, I guess reddit doesn't like logging in with ssl |
| 23:32:29 | flaviu | http://www.reddit.com/r/programming/comments/23mt34/nimrod_094_released_includes_new_async_io_module/ |
| 23:32:57 | dom96 | cool thanks |
| 23:34:34 | * | jturner joined #nimrod |
| 23:34:47 | dom96 | hello jturner! |
| 23:34:52 | jturner | hi |
| 23:35:00 | NimBot | Araq/Nimrod devel 9ea55bf Dominik Picheta [+0 ±3 -0]: Website updates. |
| 23:35:16 | jturner | will there no longer be zip/tar releases that contain csources like 0.9.2 did? |
| 23:35:57 | jturner | the 0.9.4 tar release on github doesn't seem to contain the c sources |
| 23:36:21 | dom96 | The C sources have been moved here: https://github.com/nimrod-code/csources/ |
| 23:37:23 | jturner | yup, no release archive tho, i ask cause i maintain a nimrod port on OpenBSD and having release archives is important |
| 23:37:43 | jturner | we can't just checkout from git for our builds like some linux distros do |
| 23:38:06 | jturner | i can host my own, but having an offical one is always nice |
| 23:39:10 | flaviu | dom96: Why not add the latest c sources to csources and add a release? |
| 23:39:22 | dom96 | Yeah. |
| 23:39:33 | dom96 | jturner: I can tag the latest commit on the C sources repo is that helps. |
| 23:39:34 | flaviu | That way we have an official 9.4 tarball |
| 23:39:47 | dom96 | Github will then provide a download link IIRC |
| 23:39:54 | jturner | yup a tag would help |
| 23:39:55 | jturner | thanks |
| 23:39:59 | flaviu | dom96: You have to upload the release tarball manually |
| 23:40:03 | Skrylar | just make sure not to name the tag "latest" |
| 23:40:13 | renesac | dom96, Araq has given the green light for Reddit/HN posts? |
| 23:40:17 | Skrylar | you can't retag an existing tag without it invalidating the tag archives |
| 23:40:38 | dom96 | renesac: yeah, reluctantly, but he did :P |
| 23:40:58 | dom96 | flaviu: Nope. See here: https://github.com/Araq/Nimrod/releases |
| 23:40:59 | renesac | nimrod heart beat release |
| 23:41:01 | dom96 | I didn't upload those. |
| 23:41:56 | renesac | we could start giving funny names to our releases |
| 23:42:13 | renesac | like ubuntu and mmg does |
| 23:42:14 | renesac | XD |
| 23:42:26 | flaviu | dom96: Thats the source. You don't want people to download all 20 historical csources, they just want the latest |
| 23:42:28 | dom96 | We could come up with some fun naming scheme. |
| 23:42:51 | dom96 | flaviu: That's the files in the repo at that tag. |
| 23:43:17 | Skrylar | well here's a headscratch |
| 23:43:21 | xenagi | jturner, where do you maintain the port? |
| 23:43:35 | Skrylar | freetype wants to open a file by default and requires bij to open a system font; GDI wants to open a system font by default and requires bij to open a file |
| 23:44:31 | jturner | xenagi: it's in the offical openbsd ports tree, http://www.openbsd.org/cgi-bin/cvsweb/ports/lang/nimrod/ |
| 23:45:14 | dom96 | I wonder how long it will take the Arch guys to update the Nimrod package. |
| 23:45:21 | xenagi | nice |
| 23:45:28 | flaviu | dom96: It comes directly from git, so instant |
| 23:45:45 | Demos | flaviu: I think he meant the community one.... |
| 23:45:48 | Demos | not the aur one |
| 23:46:44 | dom96 | jturner: I hope this helps: https://github.com/nimrod-code/csources/archive/v0.9.4.tar.gz |
| 23:47:16 | flaviu | dom96: Please upload an archive with just the latest sources |
| 23:47:20 | jturner | dom96: perfect, thanks |
| 23:48:19 | flaviu | Or maybe I'm not understainding how this works |
| 23:48:59 | dom96 | flaviu: Download the archive and see for yourself. |
| 23:49:02 | jturner | flaviu: in my case i'll just download both v0.9.4.tar.gz from the Nimrod and csourses repos and then extracted them in the same folder |
| 23:49:17 | jturner | now i have all I need to build like before |
| 23:50:57 | flaviu | Oh, I thought that the 8_2... were version numbers |
| 23:51:40 | jturner | nope, those related to arch and such |
| 23:58:42 | reactormonk | did we have any interest in advocating using result = X instead of return X? |