| 00:47:52 | * | q66 quit (Quit: Quit) |
| 03:14:27 | * | fowl_ joined #nimrod |
| 03:16:50 | * | fowl quit (Ping timeout: 260 seconds) |
| 03:44:09 | reactormonk | Araq, ensime looks like a good point to start :-) |
| 03:44:19 | reactormonk | but let's finish this shit first. |
| 07:10:37 | Araq | reactormonk: hu? we are working on 'compiler as a service' already ... |
| 08:19:53 | * | Araq is now known as araq_bnc |
| 08:56:18 | * | araq_bnc is now known as Araq |
| 09:34:18 | * | Trixar_za joined #nimrod |
| 09:38:00 | Araq | wb Trixar_za |
| 09:39:58 | Araq | ping zahary |
| 09:41:17 | Trixar_za | Thank Araq |
| 09:42:06 | Trixar_za | Damn box was hacked and DDoS'd - I love how these script kiddies have to inflate their egos by adding insult to injury |
| 09:42:33 | Araq | for some reasons the number of people in this channel is shrinking :-/ |
| 09:44:23 | Trixar_za | It happens. It'll grow again soon. |
| 09:45:07 | Trixar_za | Everytime it's Spring or Summer somewhere in the world, then IRC shrinks |
| 09:45:08 | Trixar_za | :P |
| 09:46:14 | Araq | I see ;-) |
| 09:52:20 | Araq | wow there is a german version of the slitaz website |
| 09:59:03 | Trixar_za | Probably an ancient translation |
| 10:06:00 | Araq | yeah the news section is not up to date ... |
| 10:12:06 | Trixar_za | We really need more translators :P |
| 10:17:54 | Araq | not really |
| 10:18:06 | Araq | english version suffices IMHO |
| 10:18:19 | Araq | linux users are geeks |
| 10:18:32 | Araq | they all speak a bit of english |
| 10:18:46 | Araq | I have to go, see you later |
| 10:19:13 | Trixar_za | Bye Araq |
| 11:12:23 | * | ekselkiu joined #nimrod |
| 11:21:07 | * | q66 joined #nimrod |
| 11:45:00 | Araq | ping q66 |
| 11:45:09 | q66 | pong Araq |
| 11:45:56 | Araq | I've already discussed this quite a bit with zahary, but like to know a second opinion |
| 11:46:45 | Araq | in Lua there are proper co-routines |
| 11:47:09 | Araq | so you can use them to define an "actor" sequentially |
| 11:47:43 | Araq | something like: "turn left; go 10 meters; start shooting the alien" |
| 11:50:09 | Araq | now if you don't do this in the scripting layer of your engine |
| 11:50:29 | Araq | you have to convert the sequential code to a state machine |
| 11:52:30 | Araq | the question is: how often does this come up? |
| 11:53:31 | Araq | Nimrod recently got support for "crippled" coroutines/ Python-like iterators (no stack capturing) and the question is how useful they are ... ;-) |
| 11:54:16 | q66 | Araq, i use coroutines in a few places but i avoid them in performance heavy things |
| 11:54:41 | q66 | but they're useful indeed |
| 11:54:46 | q66 | they simplify a lot of things |
| 11:55:00 | q66 | they're just a bit slow |
| 11:55:15 | q66 | about 2x slower than if you had a regular state machine |
| 11:56:55 | Araq | but that's Lua specific, right? |
| 11:57:07 | Araq | I mean the "2x slower" part |
| 11:57:34 | q66 | Araq, well, i guess they'll always be somewhat slower |
| 11:58:23 | Araq | but Lua's coroutines support recursion iirc |
| 11:58:37 | Araq | that means they capture the stack / have their own stack |
| 11:59:07 | Araq | this may be the reaons why they're slower |
| 12:03:46 | Araq | so what do they simplify for instance? |
| 12:07:22 | q66 | Araq, well, for example they simplified lexing strings in my language a lot. |
| 12:07:42 | q66 | because strings in Vortex are multi-token |
| 12:07:50 | q66 | because of string interpolation |
| 12:08:20 | Araq | I'm asking about examples within a game context :P |
| 12:08:21 | q66 | so read_string is just a coroutine that gets resumed each time until it's dead |
| 12:08:29 | q66 | returning a token and its semantic value |
| 12:08:41 | Araq | yeah yeah yeah |
| 12:09:15 | Araq | no need to explain lexers to me :P |
| 12:09:24 | q66 | Araq, in game context, for example action system, though i don't use coroutines there as i already solved it differently |
| 12:09:29 | q66 | basically, you know, there's a main loop |
| 12:09:33 | q66 | there's an entity storage table |
| 12:09:48 | q66 | each frame, it gets iterated and it "acts" on the entity |
| 12:10:05 | q66 | the entity then contains an action queue - i.e. animation changes etc |
| 12:10:26 | q66 | actions take N seconds, basically they begin, they run for some time and they end |
| 12:10:41 | q66 | so, the entity "runs" on each action in its queue |
| 12:11:15 | q66 | the thing is - without coroutines, you have to store all the data in an object and query it each time it's called, as you have no way to communicate between separate frame calls |
| 12:11:40 | q66 | with a coroutine, you maintain an infinite loop in the action object, store all the data (private ones anyway) as locals, and yield each iteration |
| 12:11:49 | q66 | it simplifies design, but is somewhat slower :) |
| 12:12:18 | q66 | as entities and actions are performance heavy though, i use the non-coroutine approach |
| 12:12:37 | q66 | of course, in specialized cases, custom action objects can handle their own coroutine in the appropriate method |
| 12:14:36 | q66 | speaking of coroutines in this use, they can also be used to create completely table-less objects :) |
| 12:14:49 | q66 | where method calls are not method calls, but instead some sort of "signals" |
| 12:15:36 | q66 | bbl |
| 12:17:00 | Araq | brb too |
| 14:44:40 | zahary | Araq, I was out yesterday and I missed your conversation with dom96. If I was here, I would have corrected you :) |
| 14:44:50 | zahary | Please, read this: https://gist.github.com/4096467 |
| 14:51:07 | * | FreeArtMan joined #nimrod |
| 15:21:59 | * | XAMPP_ quit (Ping timeout: 246 seconds) |
| 16:03:04 | Araq | zahary: alright, I think I figured it out finally |
| 16:03:23 | Araq | but I'm reading your gist now |
| 16:03:46 | Araq | welcome FreeArtMan |
| 16:04:04 | * | ekselkiu quit (Ping timeout: 260 seconds) |
| 16:04:05 | FreeArtMan | hi Araq |
| 16:06:52 | Araq | FreeArtMan: are you new or did you change your nick? :-) |
| 16:07:14 | FreeArtMan | new |
| 16:07:31 | Araq | good |
| 16:09:20 | Araq | just ask if you need any help, nobody here bites |
| 16:11:13 | FreeArtMan | ok ;] |
| 16:11:14 | Araq | zahary: my design looks almost identical; I called TPromise an "event" instead, not sure if that is the better term |
| 16:12:50 | Araq | (and yeah, we are aware 'select' is ancient ... but for now it did the job) |
| 16:15:01 | dom96 | Araq: I think that TPromise is a better term. It rings a bell with me at least. |
| 16:15:39 | dom96 | Also the name has a meaning, you are making a "promise" that the data will be retrieved at some point the future. |
| 16:15:56 | Araq | the problem is that there are quite a few of these concurrency related terms and I'm not sure they have formal definitions |
| 16:16:18 | Araq | actually the code doesn't make a 'promise' at all |
| 16:16:33 | dom96 | And changing select to epoll should not be a problem. |
| 16:17:01 | Araq | yield download(url) does not promise anything |
| 16:17:19 | Araq | instead it means: "wait for a downloading-completed event" |
| 16:17:54 | dom96 | It promises that the resource at ``url`` will be downloaded at some point in the future, no? |
| 16:18:13 | Amrykid | :o are you guys talking about asynchronous stuff? |
| 16:18:30 | Araq | Amrykid: indeed and nimrod is getting native support for it |
| 16:18:38 | Amrykid | Araq, thats awesome. |
| 16:18:44 | Araq | in fact, the key feature for it has already been implemented |
| 16:19:26 | Amrykid | ok, so I read up until the TPromise bit. instead, yield seems like 'wait'. |
| 16:19:44 | Araq | that's my point, yes |
| 16:20:16 | Amrykid | can you not like attach a callback (closure) when it returns, that'll execute? |
| 16:20:16 | zahary | I'm just passing by and have to go - it's true that future/promise/task/etc are used a bit inconsistently across the different languages, but still they are relatively established as terms |
| 16:21:05 | Araq | Amrykid: yeah but then you have to wrap the continuation in a callback |
| 16:21:11 | Araq | with 'yield' you don't have to |
| 16:21:18 | zahary | also, I have not covered error handling in my gist - this is also important |
| 16:21:39 | Amrykid | Araq, so what seems to be the issue? |
| 16:21:59 | Amrykid | with 'yield'? |
| 16:22:01 | zahary | see you later |
| 16:22:06 | Araq | no real issues, we're discussing details of the design/API |
| 16:22:13 | Amrykid | ah |
| 16:22:17 | Araq | zahary: alright |
| 16:28:20 | Araq | well for me 'promise' implies that some threading is involved |
| 16:28:44 | Araq | but 'yield' has nothing to do with threads |
| 16:29:05 | Amrykid | yield is for iterators, correct? |
| 16:29:33 | Araq | yes |
| 16:30:17 | Amrykid | best you use a seperate keyword like 'promise' or something to prevent double definitions. |
| 16:34:30 | Araq | well what do you wait for? |
| 16:34:38 | Araq | a 'promise' or an 'event'? |
| 16:36:00 | Amrykid | an event. people don't always keep their promises. events happen unless they are canceled. |
| 16:40:59 | dom96 | Araq: I want to see why nimbuild's CPU usage spikes, can I use your profiler for that? |
| 16:52:09 | Araq | I doubt it ;-) |
| 16:52:49 | Araq | you can try |
| 16:52:55 | Araq | when it spikes you can press ctrl+c and inspect the stack trace :P |
| 17:13:16 | dom96 | hrm, the results are what I suspected |
| 17:13:24 | dom96 | Although i'm not sure if i'm reading it correctly heh |
| 17:18:02 | FreeArtMan | Araq, is in nimrod is possible segfault? |
| 17:18:05 | FreeArtMan | made by user |
| 17:20:04 | FreeArtMan | i more interested in out of array indexing |
| 17:21:41 | * | fowl_ quit (Ping timeout: 245 seconds) |
| 17:24:21 | Araq | FreeArtMan: segfaults are possible but mostly because the compiler doesn't insert null pointer checks |
| 17:24:45 | Araq | array bounds are checked and lots of other checks are performed too in debug builds |
| 17:24:59 | FreeArtMan | that very nice then |
| 17:25:24 | Araq | it's memory safe if you don't misuse some C wrapper and stay away from 'cast' and 'addr' |
| 17:25:47 | Araq | 'cast' and 'addr' are unsafe but easily detected because they are keywords |
| 17:28:28 | FreeArtMan | i have seen nimrod syntax and told to my self that it beautiful =D |
| 17:28:41 | Araq | thanks :-) |
| 17:29:03 | Araq | some people find it "weird" instead ;-) |
| 17:30:38 | FreeArtMan | wierd its maybe not C like |
| 17:38:42 | Araq | yeah, speaking of which ... |
| 17:38:46 | Araq | iterator p(): TEvent = |
| 17:38:54 | Araq | yield download(url) |
| 17:39:17 | Araq | yield download(url2) |
| 17:39:29 | Araq | --> this is not an 'iterator' |
| 17:39:39 | Araq | and it doesn't really 'yield' either |
| 17:40:27 | Araq | it uses the same implementation as an iterator though ... |
| 17:42:01 | Araq | we could introduce new keywords for it: |
| 17:42:23 | Araq | task p(): TEvent = |
| 17:42:30 | Araq | await download(url) |
| 17:42:39 | Araq | await download(url2) |
| 17:43:29 | Amrykid | ^ |
| 17:43:38 | FreeArtMan | is there some standart for nimrod'e? |
| 17:44:53 | dom96 | Araq: Maybe the ': Tevent' should then be implicit for 'task'? |
| 17:45:37 | Amrykid | this all sounds like what C# did, just putting that out there. |
| 17:46:31 | Amrykid | in .NET 4.5, C#/VB have Async/Await keywords that function like this. |
| 17:46:42 | Araq | Amrykid: we know ;-) |
| 17:46:58 | Amrykid | Araq, yeah, I thought this wasn't 'C#' :D |
| 17:47:26 | * | XAMPP joined #nimrod |
| 17:47:50 | Araq | well the new features of C# are all good because they hired people who know what they are doing |
| 17:48:14 | Araq | so it's no shame to copy from C# :P |
| 17:48:25 | Amrykid | heh |
| 17:48:36 | Araq | it's the old Java-like parts that are annoying |
| 17:49:26 | dom96 | hrm, seems Github is failing at webhooks. |
| 17:49:47 | Araq | it's because your dom.co.cc domain ceased to exist |
| 17:50:02 | dom96 | ahh yes. |
| 17:50:06 | dom96 | Of course. |
| 17:50:45 | Araq | FreeArtMan: we have a manual/spec but it's no ANSI/ISO standard ;-) |
| 17:50:54 | Araq | yet :P |
| 17:52:03 | dom96 | meh, that's old |
| 17:52:41 | Araq | yeah github gets webhook switching wrong |
| 17:52:54 | Araq | did the same for the nimrod compiler |
| 17:54:40 | dom96 | nah, test hook just sends the master branch |
| 17:54:46 | dom96 | instead of the latest branch |
| 17:55:14 | Araq | hrm, perhaps |
| 17:55:29 | * | XAMPP quit (Quit: Leaving) |
| 17:56:49 | Araq | both 'await' and 'async' can be implemented as macros |
| 18:01:45 | Amrykid | i see how you would use 'await' but async? |
| 18:01:51 | Amrykid | async proc... ? |
| 18:03:51 | Araq | proc p {.async.} = ... |
| 18:05:28 | * | XAMPP joined #nimrod |
| 18:06:56 | Araq | and yeah, dom96, the ': TEvent' should be implicit then |
| 18:13:25 | * | fowl joined #nimrod |
| 18:14:31 | * | XAMPP quit (Quit: Leaving) |
| 18:14:47 | * | XAMPP joined #nimrod |
| 18:15:35 | FreeArtMan | is some large project with nimrod? |
| 18:15:41 | FreeArtMan | that i can get and compile? |
| 18:16:05 | dom96 | FreeArtMan: Aporia, jester, nimbuild are some examples ;) |
| 18:16:12 | dom96 | and of course the Nimrod compiler itself |
| 18:16:26 | FreeArtMan | ok will check them now |
| 18:23:30 | Araq | hrm C# 5's await does not capture the stack either |
| 18:23:44 | Araq | which means Nimrod's is on par |
| 18:25:23 | FreeArtMan | how was to compile smallest possible code? |
| 18:25:40 | FreeArtMan | i have found it on the site and now cannot find again |
| 18:28:49 | Araq | http://nimrod-code.org/question.html |
| 18:28:57 | Araq | For the standard configuration file, -d:quick --opt:size does the trick. |
| 18:29:17 | FreeArtMan | thx |
| 18:29:28 | dom96 | Araq: What limitation does not capturing the stack introduce? |
| 18:30:19 | Araq | actually I may be wrong for C#, still trying to figure it out |
| 18:30:49 | Araq | the limitation means that the async "workflow" needs all to be within a single iterator |
| 18:31:01 | Araq | you can't do: |
| 18:31:17 | Araq | proc foo() = yield |
| 18:31:26 | Araq | iterator bar() = |
| 18:31:28 | Araq | foo() |
| 18:32:13 | Araq | I don't think how often this will come up in practice |
| 18:32:57 | Araq | I tried to ask q66 about it but he instead explained efficiency tradeoffs in Lua :P |
| 18:34:13 | dom96 | will that mean then that I will not be able to call an async proc from within an async proc? |
| 18:34:34 | FreeArtMan | cannot compile Aporia |
| 18:34:38 | FreeArtMan | some error |
| 18:34:54 | dom96 | FreeArtMan: You need the latest compiler from git. |
| 18:35:04 | FreeArtMan | ok |
| 18:39:27 | Araq | dom96: you can do that ... mostly ... |
| 18:39:33 | Araq | it's complicated :P |
| 18:39:47 | dom96 | lol. ok |
| 18:40:38 | Araq | 0.9.0 is not good enough to compile aporia? |
| 18:40:53 | dom96 | I think I added some stuff to the gtk wrapper |
| 18:40:54 | dom96 | so no |
| 18:41:04 | Araq | I see |
| 18:41:20 | Araq | you should provide stable aporia releases :P |
| 18:41:48 | dom96 | I do. |
| 18:42:09 | Araq | really? where to get those? |
| 18:42:34 | Araq | FreeArtMan: nimrod's forum is also written in Nimrod |
| 18:43:03 | Araq | and so is NimBot |
| 18:43:21 | FreeArtMan | i am not in nimrod!!! |
| 18:44:02 | Amrykid | Im still sad that nimbuild doesn't provide builds of aporia, which i thought was the offical nimrod ide |
| 18:44:02 | Araq | hu? read it again |
| 18:44:08 | dom96 | Araq: https://github.com/nimrod-code/Aporia/downloads |
| 18:45:51 | Araq | dom96: maybe we should put these on nimrod's website ... |
| 18:46:08 | dom96 | Araq: Sure, if you want. |
| 18:46:12 | Araq | I don't expect github to host binaries |
| 18:46:25 | dom96 | Might as well wait for the new release though |
| 18:46:39 | FreeArtMan | host base64 binaries =D |
| 18:46:43 | Araq | and Amrykid is right, nimbuild should provide them |
| 18:47:10 | dom96 | Well pull requests are always welcome for Nimbuild ;) |
| 18:47:27 | dom96 | I'm having a hard enough time finishing branch support |
| 18:48:21 | FreeArtMan | it possible to compile kernel module with nimrod? =D |
| 18:49:49 | Araq | FreeArtMan: people wrote toy OS kernels in nimrod |
| 18:50:32 | FreeArtMan | prove in the studio |
| 18:50:49 | FreeArtMan | Aporia compiles fine with newest nimrad |
| 18:51:00 | FreeArtMan | and works |
| 18:51:26 | Araq | I know, I recompiled it today ;-) |
| 18:51:45 | Araq | I'm using it all the time |
| 18:51:49 | FreeArtMan | on 64bit system? =] |
| 18:51:54 | Araq | yes |
| 18:52:10 | FreeArtMan | ;] |
| 18:52:47 | Araq | on linux |
| 18:53:28 | FreeArtMan | ;] |
| 18:56:00 | Araq | can't see a reason why Linux kernel modules can't be written in nimrod but it requires some command line options to strip away the stdlib |
| 18:56:30 | FreeArtMan | module dont have standart headers |
| 18:57:55 | Araq | you need to patch nimbase.h then :P |
| 19:19:04 | Araq | ugh annotating the stdlib with tags is quite some work :-/ |
| 19:31:25 | Araq | I have the effect 'FDb' and then have to annotate almost every operation in db_sqlite |
| 19:32:21 | Araq | however this is redudant as every operation having a 'db: TDbConn' parameter should be annotated with FDb ... |
| 19:33:30 | dom96 | Maybe being able to push {.effect: FDb.} or whatever it is so that all procedures get it would be useful? |
| 19:34:03 | Araq | it's not all procs, but most |
| 19:34:18 | dom96 | yeah, then you could pop around them? |
| 19:34:34 | Araq | that would work |
| 19:34:59 | Araq | but I need to distinguish between FReadDB and FWriteDB anyway |
| 19:35:32 | dom96 | ahh |
| 19:35:39 | Araq | and then it may even catch some real bugs |
| 19:36:06 | Araq | instead of being a "feeling good" feature ;-) |
| 19:45:14 | Araq | hrm, is 'open' and 'close' an effect? |
| 19:45:35 | Araq | yes. |
| 19:54:59 | * | FreeArtMan quit (Quit: Out of this 3D) |
| 20:02:59 | reactormonk | Araq, yeah, but the emacs API |
| 20:03:34 | Araq | adhereing to an API that I didn't invent is no fun :P |
| 20:08:25 | reactormonk | Araq, go implement SWANK :-P |
| 20:12:15 | Araq | swig support sounds more useful :P |
| 20:24:20 | * | XAMPP_ joined #nimrod |
| 20:27:47 | * | XAMPP quit (Ping timeout: 246 seconds) |
| 20:30:59 | * | fowl quit (Changing host) |
| 20:30:59 | * | fowl joined #nimrod |
| 20:31:42 | Araq | deploy this new awesome NimBot asap :D |
| 20:31:54 | Araq | does it pass the Turing test now? |
| 20:32:20 | dom96 | Yep. It has a female personality too. |
| 20:32:51 | Araq | with a sexy voice? |
| 20:33:08 | Araq | !lag |
| 20:33:08 | NimBot | 7ms between me and the server. |
| 20:33:09 | dom96 | of course. |
| 20:33:32 | Araq | !strip |
| 20:33:41 | dom96 | lol |
| 20:33:45 | Araq | not implemented, hu? |
| 20:33:52 | dom96 | no :P |
| 20:34:54 | Araq | more realistic that way ;-) |
| 20:35:24 | Araq | most girls don't obey this command either |
| 20:35:32 | dom96 | hah |
| 20:37:08 | fowl | dom96: you should use something like this for the commands: https://gist.github.com/4013528 |
| 20:38:46 | dom96 | perhaps, but NimBot doesn't really need anything complicated. At least for now. |
| 20:39:31 | dom96 | Wish I had time for fancy macros :| |
| 20:47:05 | dom96 | grr. I think I just found a bug in the IRC module. |
| 20:47:26 | Araq | can't be a serious bug, can it? |
| 20:49:04 | dom96 | Well, it's not super bad. |
| 20:49:07 | dom96 | But not trivial either |
| 20:49:28 | dom96 | I need to see how the IRC protocol works in this situation to know. |
| 20:53:09 | Araq | that sounds like an "undocumented feature" then |
| 20:53:32 | dom96 | It shouldn't cause any issues unless you are changing your nick. |
| 20:53:39 | dom96 | Fixing now. |
| 21:25:42 | Araq | damn writing to a PSurface can be an IO operation but often is not ... :-/ |
| 21:25:57 | Araq | so graphics.nim can't annotate anything |
| 21:26:48 | fowl | when is writing to a surface IO |
| 21:27:19 | fowl | also i thought about it and why not just have PSurface = ref sdl.PSurface |
| 21:27:57 | fowl | sdl.psurface has w/h fields already |
| 21:28:36 | Araq | yeah but it's more convenient this way |
| 21:29:31 | Araq | once destructors work properly we can even do: PSurface = sdl.PSurface |
| 21:30:30 | Araq | the indirection is only necessary for the finalizer |
| 21:31:04 | Araq | "writing to a surface" is no IO unless it's the screen surface |
| 21:31:27 | fowl | ahh |
| 21:33:14 | Araq | however for a game server any graphics operation is likely a bug ... hrm |
| 21:41:12 | fowl | game server dont need to run any graphics |
| 21:45:12 | Araq | yes that's the point |
| 21:45:49 | Araq | so you do: proc serve() {.tags: [].} and the compiler validates that 'serve' has no tag 'FGraphics' |
| 21:46:27 | dom96 | why would anyone put graphics code in server code? 0_o |
| 21:46:47 | Araq | because of weird dependencies? |
| 21:46:58 | Araq | these things tend to creep in ... |
| 21:47:10 | Araq | and it's nice you can shield against it |
| 22:01:28 | Araq | http://dendory.net/blog.php?id=509ec629 |
| 22:13:44 | * | NimBot_ joined #nimrod |
| 22:13:44 | * | NimBot_ quit (Remote host closed the connection) |
| 23:22:18 | * | NimBot_ joined #nimrod |
| 23:22:49 | dom96 | !trusted |
| 23:22:50 | NimBot_ | Trusted users: dom96@unaffiliated/dom96 |
| 23:22:52 | dom96 | yay |
| 23:23:37 | Araq | !strip |
| 23:23:51 | Araq | !untrusted |
| 23:23:55 | dom96 | She doesn't like ya |
| 23:24:02 | Araq | !trusted |
| 23:24:02 | NimBot_ | Trusted users: dom96@unaffiliated/dom96 |
| 23:24:11 | Araq | yeah, she doesn't trust me |
| 23:24:21 | dom96 | !addtrust |
| 23:24:22 | NimBot_ | Syntax: !addtrust <nick> <host> |
| 23:24:27 | dom96 | Araq: Try using that command |
| 23:25:54 | Araq | !addtrust Araq host |
| 23:25:54 | NimBot_ | Access denied. |
| 23:25:57 | dom96 | :D |
| 23:26:09 | Araq | !addtrust trust me |
| 23:26:10 | NimBot_ | Access denied. |
| 23:26:34 | dom96 | !addtrust Araq 2a01:7e00::f03c:91ff:fe93:822b |
| 23:26:35 | NimBot_ | Done. |
| 23:26:38 | dom96 | !trusted |
| 23:26:38 | NimBot_ | Trusted users: dom96@unaffiliated/dom96, Araq@2a01:7e00::f03c:91ff:fe93:822b |
| 23:26:55 | Araq | ipv6? |
| 23:27:05 | dom96 | yeah. IP of the BNC |
| 23:27:21 | dom96 | Freenode picks that up for some reason |
| 23:27:24 | Araq | use my MAC instead :P |
| 23:27:52 | dom96 | lol nah |
| 23:28:00 | dom96 | Try using !addtrust again |
| 23:28:10 | Araq | no |
| 23:28:15 | Araq | I'm busy |
| 23:28:28 | dom96 | Fine :( |
| 23:30:25 | * | NimBot_ quit (Remote host closed the connection) |
| 23:31:35 | dom96 | Araq: Marshal creates quite odd json files |
| 23:32:45 | Araq | how so? |
| 23:32:59 | dom96 | it doesn't seem valid |
| 23:33:10 | dom96 | yeah... |
| 23:33:18 | dom96 | Seems like a bug |
| 23:33:23 | dom96 | :( |
| 23:34:58 | Araq | I screwed up json generation? |
| 23:35:14 | * | Araq doesn't believe it :P |
| 23:35:24 | dom96 | https://gist.github.com/7519dfbd52ff4ebf211b |
| 23:35:32 | dom96 | Might be something else |
| 23:36:04 | Araq | yes |
| 23:36:39 | * | XAMPP_ quit (Read error: Connection reset by peer) |
| 23:37:07 | * | XAMPP_ joined #nimrod |
| 23:43:44 | dom96 | Araq: here is the object I serialize: https://gist.github.com/f8e1861e89c35f2138ba |
| 23:46:21 | * | XAMPP_ is now known as XAMPP |
| 23:46:26 | * | XAMPP quit (Changing host) |
| 23:46:26 | * | XAMPP joined #nimrod |
| 23:55:24 | Araq | dom96: proper bug report please |
| 23:55:31 | dom96 | will do |
| 23:55:32 | Araq | looks difficult |
| 23:58:52 | Araq | please check if the unit tests in marshal.nim work for you |
| 23:59:12 | Araq | (nimrod c -r lib/pure/marshal.nim) |
| 23:59:48 | dom96 | seems to work |