<< 28-05-2019 >>
00:00:23*disruptek joined #nim
00:07:54*cyberjpn joined #nim
00:16:54*disruptek quit (Read error: Connection reset by peer)
00:17:37*disruptek joined #nim
00:33:05FromDiscord_<demotomohiro> @deech
00:33:05FromDiscord_<demotomohiro> !eval import macros;dumptree:echo not 1==1-1
00:33:08NimBotStmtList↵ Infix↵ Ident "=="↵ Infix↵ Ident "not"↵ Ident "echo"↵ IntLit 1↵ Infix↵ Ident "-"↵ IntLit 1↵ IntLit 1
00:34:55FromDiscord_<demotomohiro> https://nim-lang.github.io/Nim/manual.html#syntax-precedence
00:37:04FromDiscord_<demotomohiro> import macros;dumplisp:echo not 1==1-1
00:37:21FromDiscord_<demotomohiro> !eval import macros;dumplisp:echo not 1==1-1
00:37:22NimBotStmtList(Infix(Ident("=="), Infix(Ident("not"), Ident("echo"), IntLit(1)), Infix(Ident("-"), IntLit(1), IntLit(1))))
00:46:09FromDiscord_<demotomohiro> 'echo not 1 == 1 - 1' is 'echo (not 1) == (1 - 1)'
00:46:09FromDiscord_<demotomohiro> And 'not 1' is -2 because it is bitwise not.
00:46:43*kapilp joined #nim
00:57:02*sz0 quit (Quit: Connection closed for inactivity)
00:58:58*cyberjpn quit (Ping timeout: 246 seconds)
01:00:16*elronnd joined #nim
01:07:12*elronnd is now known as Elronnd
02:02:06*banc quit (Quit: Bye)
02:02:49leorizeAraq: devel is failing https://travis-ci.org/nim-lang/Nim/builds/538017333
02:04:43*thomasross joined #nim
02:13:00*ftsf joined #nim
02:17:31*deech quit (Ping timeout: 258 seconds)
02:26:16*couven92 quit (Read error: Connection reset by peer)
02:26:24*banc joined #nim
02:27:48leorizenarimiran: your fixing attempts backfired :p https://travis-ci.org/nim-lang/Nim/jobs/538017334#L4997
02:32:13*rnrwashere joined #nim
02:40:44*rnrwashere quit (Remote host closed the connection)
02:41:20*rnrwashere joined #nim
02:45:34*rnrwashere quit (Ping timeout: 252 seconds)
02:57:09*dddddd quit (Remote host closed the connection)
03:08:50*rnrwashere joined #nim
03:40:59*rnrwashere quit (Remote host closed the connection)
03:41:39*rnrwashere joined #nim
03:45:04*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
03:45:34*laaron joined #nim
04:12:54*theelous3 quit (Ping timeout: 258 seconds)
04:18:09*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
04:34:25*nsf joined #nim
04:56:58*Guest93678 quit (Ping timeout: 246 seconds)
05:02:05*Senketsu quit (Quit: WeeChat 2.4)
05:10:31*rnrwashere quit (Remote host closed the connection)
05:21:08*leorize quit (Quit: WeeChat 2.3)
05:23:03*laaron quit (Remote host closed the connection)
05:24:51FromGitter<rishavs> When I declare a global variable in Nim, all my children modules get to use it automatically or do I need to do something to use them? ⏎ ⏎ My project structure is; ⏎ Main.nim ⏎ ... [https://gitter.im/nim-lang/Nim?at=5cecc62383ae782aeee28074]
05:26:50*laaron joined #nim
05:30:13*narimiran joined #nim
05:34:29FromGitter<Varriount> @rishavs "use" isn't a valid keyword
05:35:02FromGitter<rishavs> i was paraphrasing there. sorry about it. I used the var as `echo x`
05:45:33*kapilp quit (Quit: Connection closed for inactivity)
05:46:48*leorize joined #nim
05:47:50FromGitter<rishavs> If you see the repl.it i have linked you can see a simple bit of code simulating my problem
05:49:01leorizerishavs: no, globals don't work like that
05:49:24leorizetheir scope is the same as for functions
05:49:52leorizeif you wanted to use it outside, you'd have to import the module containing it & have that variable exported
05:49:56FromGitter<rishavs> ah, so it isn't a fully global variable. Do i nee dto export it explicitly?
05:50:08leorizeyep
05:50:17leorizeit's as global as a function would be
05:51:07FromGitter<rishavs> I cant import my main module in my child module as that would be a recursive import. Can you look at the repl here and show me how i should go about it? https://repl.it/repls/QuizzicalFabulousRom
05:51:44leorizemind you, `mod` is a keyword :p
05:53:29leorizerishavs: that example doesn't compile
05:53:31FromGitter<rishavs> learned it the way. Raised a bug on that
05:53:52FromGitter<rishavs> https://repl.it/repls/GrandioseThunderousBoards
05:53:55FromGitter<rishavs> wrong repl!!!
05:54:13leorizehttps://nim-lang.github.io/Nim/manual#modules
05:54:14FromGitter<rishavs> sorry, i linked to one i used to add a bug in github for the mod issue
05:54:16leorize^ see that one
05:55:08leorizehttps://repl.it/repls/SentimentalStrictMaintenance
05:56:33FromGitter<rishavs> thats very interesting. So the placement of my import statement is important
05:56:43FromGitter<rishavs> I am so used to just put it at the top
05:56:43FromGitter<rishavs> :D
05:57:09leorizenarimiran: you know you could just restart the build from travis console, right? :p
05:57:44leorizeI'm not sure if I'd need to rebase the branch for it to test the latest commit
05:59:59narimiranleorize: yeah, but i can't do the same for appveyor, so it is easier for me to just go through various PRS and close-open them
06:01:01narimiranleorize: btw, those stupid errors are one more evidence that i should not try to fix anything when i'm tired
06:01:47leorize:p
06:02:49leorizeif anyone is interested, I could add windows to travis :p
06:02:57leorizethis ci thing is kinda fun
06:07:25Araqrishavs: put your imports at the top and don't use recursive module dependencies.
06:07:48Araqyou cannot fake what you don't have -- modularity, so you might as well use 'include' instead.
06:07:50Zevvleorize: you allright?
06:08:10leorize?
06:08:20Zevv"this ci thing is kinda fun"
06:08:34*solitudesf joined #nim
06:09:00FromGitter<rishavs> Thanks Araq. will give include a try
06:09:08leorizeZevv: writing automated jobs is fun :p
06:09:14leorizelearned a bit of docker on the way
06:09:18Zevvat my job CI is reason #1 for grown men crying like babies
06:10:10leorizeI'm a packager :p writing automated script is kind of a specialty
06:10:52ZevvI bet you're worth a lot of $$$ with that skills and patience :)
06:12:28leorize;)
06:13:40*seerix quit (Remote host closed the connection)
06:14:29narimiranZevv: he is still in school, remember? ;)
06:15:20Zevvoh, right, I forgot
06:16:28narimiranleorize: if you're having fun with CIs, more optimizations like you did recently where you speed them up would be welcome :)
06:17:06leorizeI think I squeezed the most out of them :p
06:17:11leorizevirtualized hardware can only get so far
06:17:32leorizeMS have this azure pipelines thing
06:17:47leorizebut no published specs, so I don't know if their hw is better than travis'
06:20:41narimiranfrom what i heard, travis was sold to some company which fired most of its developers, so now people recommend either circle ci or azure pipelines....
06:21:49leorizeazure seems to be a better deal
06:24:44Araqleorize: can't reproduce these failures
06:25:14leorizeAraq: narimiran pushed some fix recently
06:25:27leorizeyou'd need to wait for the next CI job to finish
06:26:24Araqah!
06:26:34Araqso ... narimiran skips the review process
06:26:36Araq:P
06:26:54Araqand has trouble with closing '''
06:26:59Araqthat explains a lot ;-)
06:28:56narimiranAraq: yeah, narimiran should stop working and go to bed when tired ;)
06:29:20Araqwon't blame you for that, I do the same
06:29:56narimiranbtw, devel is now green, yaaay for morning freshness :)
06:30:33Zevvleorize: have we ran the complete CI job through a `perf record` yet?
06:30:52leorizenope
06:31:19Zevvwell, here we go
06:32:26*PMunch joined #nim
06:32:29leorizeazure only have 2 cpu cores as well
06:32:34leorizeI doubt that it could be any faster
06:43:58Zevvhttp://zevv.nl/div/perf-test.png, for what it's worth
06:44:11*salewski joined #nim
06:45:41salewskinarimiran, the new homepage is really not that great. Look is less friendly compared to former release,
06:45:57salewskiand textual contend is not great also.
06:46:07narimiransalewski: PRs welcome ;)
06:46:18salewskiI suggestest already to hire an expert maybe.
06:46:28salewskiSee my draft on forum.
06:47:20salewskiI would regard that as much better personally, and it is hacked togeter in 30 minutes
06:47:33leorizehmm azure built csources in 32s
06:47:34salewskifrom a non native speaker.
06:47:39salewskiBye.
06:51:08PMunchHmm, my Nimble doesn't appear to run "before" rules
06:53:24Araqsalewski: hi
06:55:36PMunchI kinda agree with salewski, although I'm not entirely sure what he means by unfriendly. And the Julia in a Nutshell page is about the most boring page I've seen in a while. I did think to myself though as I read through the new content that maybe the grouping wasn't the best idea. This way you will start reading about one part of the language, which seems like the most important one, because it's on the top. And as you read the list it gets less and less
06:55:36PMunchinteresting. At least until the next category, which has more cool features in it, before it too drags on a bit.
06:56:54PMunchAnd someone mentioned yesterday that they would like to see more mentions of the type system, which I agree would be nice (plus of course the macro system, but that's a bit harder to show in a short snippet)
06:57:41salewskiHi Araq, I looked at the generatet assembler of the ^ operator. No surprise for me, as it is not inlined :-)
06:58:07FromGitter<alehander42> what about julia in a nutshell :O
06:59:25salewski PMunch: Unfriendly nearly black background, and long text fragments. And not all understandable for every one.
07:00:00*gmpreussner quit (Quit: kthxbye)
07:00:02salewskiThe first page should never give too much details...
07:00:04PMunchOh yeah, I'm not sure I like how big the dark field got either..
07:01:00PMunchHmm, so what about having "cards", ie. smaller text snippets about a feature, that links, or opens up, to more content on that same feature?
07:01:32PMunchPique the interest of the reader with the coolest stuff, then get them clicking as soon as possible to drive engagement?
07:02:08Araqsalewski, it should be specialized in most optimizers worth their money
07:03:23AraqI don't want to have static[T] handwritten specializations everywhere in 2019
07:03:52salewskiYes I understand.
07:04:14Araqand also, the optimization is worthwile to do at runtime too
07:04:25*gmpreussner joined #nim
07:04:39Araqif you don't know the value at compile-time, but it turns out to be 2 or 3 at runtime we also like the faster path
07:04:49Araqthat's why we went with this solution
07:05:18Araqabout the website, I think it's going in the right direction
07:05:25Araqbut your suggestions are good too.
07:05:39salewskiFine.
07:05:48Araqhowever, we need to be careful to not have *too* much text, nobody then reads it
07:06:00*krux02 joined #nim
07:06:23PMunchYeah that's part of our concern now. It looks a bit daunting when you first open it
07:06:49salewskiThat is true. But latest version has already too much text, with too much details.
07:09:07salewskiAnd I like some short whole sentences better than just fragments.
07:10:19salewskiFragments are ok for a teaching lession in a classroom, where teacher explains it.
07:10:56zestyrI agree the new version went more in the "wall of text" direction rather than a clean but still informative landing page like before
07:19:02*nc-x joined #nim
07:19:54nc-xi don't have any particular opinion of whether we need a wall of text or not but i wouldn't use the word "universal" to describe nim.
07:20:07nc-xbecause there is no accepted definition of universal
07:21:27nc-xand if you use the word universal, all the comments on hn/reddit would be picking on it (just like - the "first" native compiled language with hcr - had most comments about lisp and whatever)
07:22:24salewskiUniversal means "there is no real restriction". But ok, maybe only "for systems and application programming"
07:22:53FromGitter<kayabaNerve> I'm more offended about the lack of Oxford commas than the new design
07:25:01salewskiHave to leave now, bye.
07:25:04*salewski quit (Quit: WeeChat 2.3)
07:25:20FromGitter<kayabaNerve> I also highly question that Nimbus description.
07:25:32narimirankayabanerve: "efficiency, expressiveness, and elegance."
07:26:00nc-xCan anybody check the nim website on mobile?
07:26:30nc-xis the top bar with all the links to different pages moving or is it stuck?
07:26:50nc-xfor me it is stuck so half the links are inaccessible
07:27:05leorizenc-x: scroll it sideways
07:27:12nc-xnot working
07:27:20nc-xleorize: ^^^
07:27:36leorizeweird, it works for me here
07:27:53nc-xAlso the examples are too low now on mobile.
07:28:12nc-xpeople might not scroll that down.
07:28:58nc-xleorize: which browser? I am on latest chrome.
07:29:02leorizenarimiran: have fun :) https://travis-ci.org/nim-lang/Nim/jobs/538017335#L7856
07:29:14*dddddd joined #nim
07:29:49leorizenc-x: oh, something changed recently
07:29:53leorizenot working here as well
07:29:58leorizeboth bromite & firefox
07:30:15nc-xnarimiran: ^^^
07:30:16narimirannc-x: shit, it was working yesterday.
07:30:20narimiranwill fix it
07:30:59nc-x👍🏻
07:31:32*cyberjpn joined #nim
07:33:14*nc-x quit (Quit: Page closed)
07:39:25FromGitter<rishavs> as a somewhat new user (i did try nim years ago), i much prefer the new site. ON mobile however, the snippets do not follow the actual feature section so its harder to digest
07:42:52FromGitter<alehander42> i think i like the text but probably
07:43:00FromGitter<alehander42> it can add some of the suggestions
07:43:08*stefanos82 joined #nim
07:43:24FromGitter<alehander42> i also dont get the nimAsgn..v2
07:43:30FromGitter<alehander42> missing ting
07:43:47FromGitter<alehander42> with os:standalone
07:43:55FromGitter<alehander42> do i need to change how strs.nim is included
07:44:40FromGitter<alehander42> is gcDestructors implied?
07:46:26*slugm joined #nim
07:46:39narimiranre website: if y'all can gather all your suggestions in one place (e.g. https://forum.nim-lang.org/t/4880) it will be easier to fix/update the website
07:48:25*slugm left #nim (#nim)
07:49:21*kapilp joined #nim
07:49:48*matt-m joined #nim
07:54:07*hoijui joined #nim
08:12:55PMunch@Vindaar, just updated the persistent vector library. There were some bugs in the slicing operator that should all be fixed now :)
08:18:13*Kaivo quit (Ping timeout: 246 seconds)
08:19:19*Kaivo joined #nim
08:27:30*neceve joined #nim
08:35:19FromGitter<alehander42> hm i get weird behavior in copyString, have to try -useMallc
08:35:24FromGitter<alehander42> malloc*
08:39:24leorizeazure can complete a CI round in 30m
08:39:28leorizethat's rather impressive
08:40:29FromGitter<alehander42> how much for travis?
08:40:46leorize38m43s
08:41:35leorizefor those interested: https://dev.azure.com/alaviss/Nim/_build/results?buildId=8
08:41:52leorizeI only run for linux to compare
08:43:36leorizeit doesn't have the nice output folding function of travis though
08:45:14Zevvmaybe we should make the output of the test run less verbose anyway. What is the use of so much output if all is ok. No news is good news
08:45:52narimiranleorize: since you're having fun with these - how about circle ci?
08:47:49*cyberjpn quit (Ping timeout: 244 seconds)
08:47:57leorizegonna try it out now
08:48:25Zevvis runCI safe to run at home?
08:48:37leorizeyep
08:51:36leorizelooks like parallelism is implemented by these ci services via splitting tests into multiple ci jobs
08:51:36*Vladar joined #nim
08:51:47leorizegonna be quite some work if we want to support that
09:01:32*ftsf quit (Quit: Leaving)
09:05:24FromDiscord_<kodkuce> hmm is there some parser library for jest or asynchttp, for arguments or should i just manual do it ?
09:05:33FromDiscord_<kodkuce> hmm is there some parser library for jester or asynchttp, for arguments or should i just manual do it ?
09:08:33Zevvcgi.decodedata?
09:09:44leorizekodkuce: what kind of parser?
09:10:36*floppydh joined #nim
09:14:12FromDiscord_<kodkuce> uName = fields.Str(required=True, validate=validate.Length(min=3, max=25)) , something like this , tought now am thinking i am gona anyway use json 99% of time so i can just do if try catch and if json dosnet parse , send invalid request , tought i still need manualy to check lenghts and similar
09:14:48FromGitter<alehander42> validation library?
09:15:07FromGitter<alehander42> i am not sure, probably one can make one it would be easy with macro
09:15:35FromGitter<alehander42> <3
09:16:05FromGitter<alehander42> uName = string!(len in 3..25)
09:16:31FromDiscord_<kodkuce> yep, i think validation library is the right word :), duno i still dident even try to write a macro, tought will probably do this manual am just doint reg/auth atm
09:16:52FromDiscord_<kodkuce> "uName = string!(len in 3..25)" i go test this
09:17:06FromGitter<alehander42> or ⏎ validate: ⏎ uName: string!(3..25)
09:17:09FromGitter<alehander42> no no
09:17:12FromGitter<alehander42> this doesnt work sorry
09:17:19FromGitter<alehander42> i was just thinking about possible DSL-s
09:18:14FromGitter<alehander42> otherwise one can still write several simple validation functions and hook them up
09:18:33FromGitter<alehander42> but i really dont know if somebody did it already
09:18:44FromGitter<alehander42> look in nimble
09:19:56FromDiscord_<kodkuce> yep anyway does nim have like terary operator like var x = weight > 20 ? "skiny" : "fat" ?
09:19:58FromDiscord_<kodkuce> yep anyway does nim have like terary operator like var x = weight > 20 ? "skiny" : "fat"
09:20:13FromDiscord_<kodkuce> yep anyway does nim have like terary operator like var x = weight < 20 ? "skiny" : "fat"
09:20:44FromGitter<alehander42> please dont edit your messages in discord #nim
09:21:10FromGitter<alehander42> i know it's usually very good to do so, but the way the irc bridge works, we get a new message for each
09:21:15FromGitter<alehander42> otherwise yes
09:21:18FromGitter<alehander42> if else
09:21:23FromGitter<alehander42> if weight > 20 : .. else: ..
09:21:54FromGitter<alehander42> (you can use `fix word*`)
09:22:18FromDiscord_<kodkuce> ok sorry i newer thinked about editing and IRC
09:22:45FromGitter<alehander42> no problem its a strange issue indeed
09:23:00FromGitter<alehander42> what are you doing? a web app
09:23:02FromGitter<alehander42> ?
09:23:33Zevvkodkuce: alehander42 means that the if yields an expression. So you do `var x = if weight < 20: "skinny" else: "fat"`
09:23:56FromDiscord_<kodkuce> atm want to do a Auth Reg with JSON token atm
09:30:24*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
09:33:25*laaron joined #nim
09:41:21*leorize quit (Remote host closed the connection)
09:47:16*couven92 joined #nim
09:48:32*ShalokShalom joined #nim
09:50:40*leorize joined #nim
09:54:29FromGitter<alehander42> interesting
10:09:09leorizenarimiran: circle ci is a mess
10:09:36leorizeto use it I'd need to spin my own docker image then upload it to docker hub
10:09:39leorizeno thanks
10:09:52Araqleorize, in your PR(s) please remember that I'd like to understand the code too and I don't know what 'RUN' is
10:10:12Araqnor do I want to read code that uses too many Bash ... tricks (?)
10:10:25leorizenot really tricks
10:10:27leorizejust functions
10:10:51leorizethe `run` thing is a function
10:11:11leorizethe default definition for non-docker is to run the command as-is
10:11:18leorizefor docker it runs them in docker
10:11:29Araqthere is RUN vs run and it confuses me :P
10:11:49leorizethe RUN is part of an inline docker file :p
10:12:11*hoijui quit (Ping timeout: 248 seconds)
10:12:24Araqtbh I am not even sure where the YAML ends and the Bash script starts
10:12:43FromDiscord_<kodkuce> did nim-lang.org just got update?
10:12:51AraqI hate this "declarative" yaml shit that doesn't work at all and so needs to fallback to Bash
10:13:12*Vladar quit (Remote host closed the connection)
10:13:23FromGitter<Vindaar> @PMunch: Oh nice, thank you!
10:13:40*Vladar joined #nim
10:16:14*leorize quit (Quit: WeeChat 2.3)
10:17:31narimiran@kodkuce yep, and it is still work in progress as people are giving their suggestions for (further) improvement
10:20:21FromDiscord_<kodkuce> i think it looks nicer now, anyway what hash to use for passwords in db, in python i used that argon2, does nimble have search i go man it
10:23:26FromDiscord_<kodkuce> list and grep worked kinda
10:25:30zestyr`nimble search` is a thing
10:26:00zestyrI recommend the package nimcrypto
10:26:47FromDiscord_<kodkuce> yep i was tarded i searched by name insted by tag
10:28:42Zevvi searched for sha1 the other day. And did not find nimcrypto
10:29:17Zevvso its not always obvious
10:31:19zestyrthere's sha1 in the stdlib though
10:32:43PMunchThere is?
10:32:50narimiranstd/sha1
10:34:26leorize[m]use nim-libsodium
10:34:41leorize[m]there are password hashing functions there
10:34:56leorize[m]@kodkuce ^
10:35:11Zevvyeah but its not ment as ea generic sha1. it only outputs hex and does not support update/finish
10:38:41FromDiscord_<kodkuce> nice libsodium has argon2 thingy 😃
10:41:03*elrood joined #nim
10:41:09FromGitter<kaushalmodi> Araq: Any hint on why parsetoml was commented out in Nim CI? https://github.com/nim-lang/Nim/commit/af6539b0c3b6958251737d5bc140cbcc8652eb8b
10:41:45narimirankaushalmodi see the commit message for a hint :P
10:41:53FromGitter<kaushalmodi> What needs to be fixed?
10:42:05FromGitter<kaushalmodi> What can I run locally to see the issue?
10:43:39narimiranlatest devel and then try to do nimble test parsetoml
10:43:54FromGitter<kaushalmodi> Ok
10:45:52leorize[m]@Araq I will probably change the `run` into `maybe_in_docker`. Is that more clear?
10:48:19*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
10:49:17*laaron joined #nim
10:49:58*laaron quit (Client Quit)
10:50:40*laaron joined #nim
10:54:08leorize[m]if needed I can also move the `install` out of the travis matrix
10:57:36*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
10:57:37*vlad1777d joined #nim
10:58:47Araqleorize[m], it's clearer but not clear
10:59:03Araq"maybe_in_docker"? maybe? when? and when not?
10:59:21*laaron joined #nim
11:04:03*cyberjpn joined #nim
11:04:07FromGitter<Vindaar> @kaushalmodi the problem with `parsetoml` now is that it changes the `kind` of a variant object at runtime. Araq fixed that unintended loophole. here: https://github.com/NimParsers/parsetoml/blob/master/src/parsetoml.nim#L952 and here: https://github.com/NimParsers/parsetoml/blob/master/src/parsetoml.nim#L1046 ⏎ I only looked at it for a couple of minutes yesterday and wasn't sure what the best way to fix the code is
11:04:44*kapilp quit (Quit: Connection closed for inactivity)
11:05:13FromGitter<kaushalmodi> I don't know variant objects, but would the json module be affected the same way?
11:05:35narimiransee here: https://github.com/nim-lang/Nim/commit/383147f5cb3db55cde451902d2c11c5cee9df075
11:05:54*cyberjpn quit (Client Quit)
11:07:40FromGitter<alehander42> Pass a var object and directly assign a value with different kind to it
11:08:00FromGitter<kaushalmodi> Thanks. For future, can a short issue be opened on such 3rd party packages that hints the package dev to such commit that shows how the breaking change on devel could be fixed?
11:08:00FromGitter<alehander42> That seems like a possible solution to me
11:08:50FromGitter<kaushalmodi> In this case, I just happened to notice that commented out commit
11:11:34*rnrwashere joined #nim
11:12:29leorize[m]Araq: http://ix.io/1KgY <-- here's an alternative config
11:14:50leorize[m]I haven't tested it yet though
11:15:33leorize[m]probably won't work as is, but the final result will be similar
11:16:22*rnrwashere quit (Ping timeout: 276 seconds)
11:32:58*deech joined #nim
11:49:01*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
11:49:45*laaron joined #nim
11:56:27*hoijui joined #nim
12:14:02narimiran@alehander42 "i think one can talk a bit more about the type system, it seems underrated to me" --> any suggestion what to add and how to expand it? (for uninitiated: this is for our landing page on the website)
12:17:59leorize[m]narimiran: omg circle ci chew through csources in 8s
12:18:10narimiran:)
12:19:19narimiranso we have travis in 38 min, azure in 30 min, and circle in .....
12:19:40leorize[m]still running :p
12:20:29PMunchnarimiran, well now the landing page barely mentions the type system at all
12:21:14narimiranPMunch: i'm working on the improvements and would like to improve that part too, but i'm out of ideas
12:22:00PMunchI'll try to think of something
12:24:24*Snircle joined #nim
12:25:04leorize[m]I know why nodejs tests fail on 32 bit now :p
12:25:06leorize[m]ubuntu xenial uses nodejs 4.2
12:28:01narimiranbtw, is there a reason why we're not on 18.04? would something else break?
12:29:14leorize[m]because travis don't have anything newer than xenial
12:32:00*solitudesf quit (Ping timeout: 248 seconds)
12:35:13*leorize joined #nim
12:42:58shashlickCI is painful since it is so slow to debug
12:44:21shashlickMostly cannot run locally to verify it things are correct
12:44:40*hoijui quit (Ping timeout: 268 seconds)
12:44:41*matt-m quit (Remote host closed the connection)
12:45:14*matt-m joined #nim
12:45:53*NimBot joined #nim
12:51:03*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
12:51:35*laaron joined #nim
12:57:13Araq# believe it or not, the direct check for ``containsOrIncl(c, a, b)``
12:57:13Araq # increases bootstrapping time from 2.4s to 3.3s on my laptop!
12:57:21Araq^ those were the days...
12:57:30AraqNim bootstraps on 2.4s on my laptop
12:57:41Araqnow it's at 5s ?
12:58:01narimiranrevert all the commits!!11!!!1! :P
12:59:27narimiranor a better way - just remove three of those `containsOrIncl` and boom, instant 2.7 sec improvement! math checks out!
12:59:47narimiranoptimizations are easy!
13:05:37*theelous3 joined #nim
13:12:54FromGitter<alehander42> rewrite nim man
13:12:55FromGitter<alehander42> in rust
13:12:56FromGitter<alehander42> duh
13:12:57FromGitter<alehander42> obvious
13:13:46*Vladar quit (Read error: Connection reset by peer)
13:14:04*apodo joined #nim
13:17:33*nsf quit (Quit: WeeChat 2.4)
13:24:18*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
13:25:30*laaron joined #nim
13:41:03leorize[m]Araq: what do you think about the alt config?
13:43:33Araqplease give me the link once again
13:43:59Zevvhttp://ix.io/1KgY
13:45:20Araqlooks good
13:47:16*lritter joined #nim
13:50:52*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
13:51:59*rnrwashere joined #nim
13:53:02*laaron joined #nim
14:00:50*Senketsu joined #nim
14:01:27*laaron quit (Quit: ZNC 1.7.1 - https://znc.in)
14:01:50*laaron joined #nim
14:04:02*laaron quit (Client Quit)
14:05:14*laaron joined #nim
14:06:54*rnrwashere quit (Remote host closed the connection)
14:07:32*rnrwashere joined #nim
14:11:58*rnrwashere quit (Ping timeout: 252 seconds)
14:13:59*vivus joined #nim
14:14:57FromDiscord_<kodkuce> @leorize[m] hmm ok i just checked sodium_sizes and i dident see any argon2 procs, are there non ? guess i can use scrypt then
14:15:22vivusHello all. so I reformatted my code to use objects now, but I am dealing with an issue in JSON that is causing some of the fields to be displayed as blank, like so: `[{"Field0": 0, "Field1": "", "Field2": {"f0": "", "f1": null}},`
14:16:21narimiranleorize[m]: did circle finish? :P what was its time?
14:16:27vivusI am guessing this is some formatting issue with JSON I need to include in my settings
14:17:35vivusthis is what the JSON output looks like: https://dpaste.de/szcG
14:22:17*PMunch quit (Remote host closed the connection)
14:22:59FromGitter<jrfondren> I think it's more likely that you're providing blank strings, and that the JSON output is correct for what it gets
14:27:35*rnrwashere joined #nim
14:27:35*salewski joined #nim
14:28:34salewskinarimiran, I wonder if we can join/link the examples from the nim homepage with the nim playground.
14:28:53narimiransalewski: that't the idea for the future indeed
14:28:57salewskiHave never used playground, but that would be great for newcommers.
14:29:05narimiranfor now, i'm implementing your suggestions
14:29:06salewskiFine.
14:29:38leorize@kodkuce: the one without any _scrypt or anything like that is the argon
14:29:45leorizebecause that's the default
14:29:59leorizenarimiran: circle failed :p I'm investigating
14:30:39FromDiscord_<kodkuce> hmm i thinki am looking wrong place sodium_sizes only returns cint , i think i have to look in normal sodium just need first find those docs
14:30:40salewskiI will fine tune my text this evening, have alredy added a few forgetten points, and will try to make it a bit shorter.
14:30:53*couven92 quit (Ping timeout: 245 seconds)
14:30:58*pwntus quit (Ping timeout: 246 seconds)
14:31:05narimiransalewski: great, i'll be following the forum, but you can ping me here too if you want
14:31:14vivus@jrfondren I'm not seeing any blank strings in my table here: https://dpaste.de/mbgU
14:31:22salewskiI really like the second page at https://nim-lang.org/features.html
14:31:38salewskiThat page is fine to explain all the details...
14:32:00salewskiBye.
14:32:01narimiransalewski: currently i'm struggling with the 'elegant' part and i'm not sure how to put more emphasis on the type system
14:32:02*salewski quit (Client Quit)
14:36:13FromDiscord_<kodkuce> https://download.libsodium.org/doc/password_hashing says "Sodium's high-level crypto_pwhash_* API leverages the Argon2 function." , and in nim sodium there is no pwhash, there is crypt_auth , meybe that?
14:36:47leorizethere's pwhash
14:36:53leorizeuse the version on git
14:37:31leorizeah, it's not nim-libsodium, it's libsodium
14:37:52leorizeinstall it like this to get the git version: `nimble install 'libsodium@#head'`
14:38:12FromDiscord_<kodkuce> oh ok will try
14:38:24FromDiscord_<kodkuce> nimble remove libsodium
14:39:06*couven92 joined #nim
14:41:12*rnrwashere quit (Remote host closed the connection)
14:41:14laaronvivus, good json fix. I here json.to often `key_not_found` exception, any idea make json.to relax to process non-exist field
14:41:38Zevvnarimiran: new nim-lang page comments here, privmsg or in the forum?
14:41:54narimiranZevv: depending if they are good or bad :P
14:42:06Zevvhehe
14:42:21*rnrwashere joined #nim
14:44:12FromGitter<jrfondren> I think it'd be a lot more approachable, and easier to extend, and just as informative, if you had the topics ('Expressive', etc.) as tabs that selected which bullet points you'd see.
14:44:35vivuslaaron: json.to ?
14:45:03FromGitter<jrfondren> You'd just have "[Efficient]. Elegant. Expressive." with the efficiency bullet-points displayed, and the user can hover over or click 'Elegant' to switch to those.
14:45:46FromGitter<jrfondren> that might change the sample code as well.
14:47:27FromGitter<jrfondren> and one way to expand on that would be to have a separate line of tabs that are questions rather than praise. "GC?", "Types?". same treatment otherwise.
14:48:02laaronvivus, yes, `json.to` unmarshal method, don't know if you familiar with that
14:48:37vivuslaaron: I used marshal previously and it generated the same output. Can you link to me what you are talking about so I can test it out?
14:50:37laaronvivus, when json string have more fields than object. thanks. https://nim-lang.github.io/Nim/json.html#to.m%2CJsonNode%2Ctypedesc
14:52:15vivuslaaron: is this an import or export feature?
14:53:13dom96laaron: use `Option[T]`
14:54:50*rockcavera quit (Remote host closed the connection)
14:57:08*elrood quit (Remote host closed the connection)
14:58:09laaronvivus, I mean usage problem. what's mean import/export feature
14:58:41vivuslaaron: I am trying to export table-data to JSON and it is giving me blank entries. Trying to figure out why, cause my table has no blank entries
14:59:03*rnrwashere quit (Remote host closed the connection)
14:59:14*rockcavera joined #nim
14:59:14leorizenarimiran: almost done
14:59:51*rnrwashere joined #nim
15:00:14laaronvivus, oh, I see. I am trying to decode json string to an object, the reverse way, in case you know that
15:02:48laarondom96, how use `Option[T]`, when there a explict type object
15:03:44dom96not sure what you mean
15:04:44leorizenarimiran: 24mins
15:04:51narimiranoooh, nice
15:05:01*kapilp joined #nim
15:06:48*Trustable joined #nim
15:06:54narimiranleorize: ...and can we have the whole travis matrix in it (linux, osx, cpp)?
15:07:11leorizeosx is paid
15:08:37narimiranleorize: what about azure pipelines?
15:11:06laarondom96, like the `json.to` example in manual, if drop a `name` field, will return `key not found` exception. how can I avoid this exception by use `Option[T]`, put `Option[T]` where
15:15:44narimiranleorize: if i'm reading azure pipelines website correctly, it offers "10 free parallel jobs with unlimited minutes per month" for open source projects, and one of their headlines is "Get cloud-hosted pipelines for Linux, macOS, and Windows"
15:17:17*rnrwashere quit (Remote host closed the connection)
15:18:20dom96laaron: name: Option[string]
15:18:40leorizenarimiran: yea, we can support all of those w azure
15:19:17laarondom96, https://dpaste.de/FRVd
15:19:25narimiranleorize: it sounds good -- what's the catch? ;)
15:19:51dom96laaron: weight: Option[int]?
15:22:42*rnrwashere joined #nim
15:22:48*zestyr quit (Quit: WeeChat 2.4)
15:24:25shashlickDo you all think there's value in a central repo of nimterop wrappers or let the community use the tool to make wrappers and focus on something else
15:24:36leorizenarimiran: it's MS? :p
15:25:48*narimiran DEVELOPERS DEVELOPERS DEVELOPERS
15:25:53laarondom96, I see, thanks. I think and want to just omit the field before.
15:29:12*nanofame joined #nim
15:30:38nanofameHello guys. Just wanted to say Nim is nice but I think it's practically useless unless people have a dedicated team that can keep up with the changes in the compiler and fixing all the libs...
15:31:33shashlickthere's a stable branch of the compiler
15:31:55shashlick0.19.x has had 4 releases already
15:32:55shashlickif you are using nim in production, you wouldn't want to keep updating
15:33:27nanofamethe stable branch is more often broken than the dev branch. With so many changes to dev branch some of the libs are usually broken all the time
15:34:22shashlickdo you have specific examples? i've been using 0.19.x for the last year with no issues
15:34:35shashlickobviously you need to keep the lib also at a fixed tag/commit
15:35:46nanofameI have been using just jester,winim,odbc and a few others
15:35:53Zevvnanofame: stdlibs are all CI tested. The occasional bug slips through of course, but I don't experience problems with that myelf
15:36:30FromGitter<liquid600pgm> I get very strange behavior when I launch VSCodium and write *any* Nim code. I save the code, everything seems to work but then my PC freezes up completely with the disk activity LED on
15:36:45nanofameI am more complaining about 3rd party libs that cannot keep up with devel
15:37:16shashlickwell, the message is to use stable and not devel
15:37:46shashlickstable has been at 0.19.x for the last 8 months and backported bug fixes
15:38:39nanofameI am not competent developer enough to fix the issues myself
15:39:28shashlickdevel is under active development, bleeding edge - only nim developers should be using devel, and lib writers should be porting their libs
15:39:51laarondoes nimble support lock on commit/tag version?
15:39:54nanofameI think I have overestimated my abilities and maturity of Nim's ecosystem
15:40:12shashlicki understand until recently, stable was not maintained but that has changed
15:40:29shashlick@nanofame: what's the challenge with using 0.19.6?
15:40:54nanofameI have had issues with stable before and was told to upgrade to devel
15:41:07nanofameand I was using devel since then
15:41:09shashlick@laaron - nimble doesn't supporting locking but you can install a particular tag or commit if you wanted with the @ operator
15:41:32shashlick@nanofame - was this in the 0.19.x timeframe?
15:41:40*Kaivo quit (Ping timeout: 244 seconds)
15:41:48narimirannanofame: so, you're using devel and you claim "the stable branch is more often broken than the dev branch."!?
15:42:01nanofameI think I am giving up for now on Nim. I need something more mature and with larger community
15:42:28shashlick@nanofame - how about you let us help you get stable on stable
15:42:33nanofamenarimiran: I have had issues with some pks on stable and was just told to install devel...
15:43:10*Kaivo joined #nim
15:43:22laaronok. seems add version locking feature to nimble is good proposal
15:43:32narimirannanofame: ok, have you reported the issues to those packages? have they fixed them in the mean time?
15:43:36shashlickya there have been some discussions aroudn that
15:44:02shashlick@narimiran - we need to change this community mindset - libs should always work on stable
15:44:14shashlickotherwise nim will never be attractive to real world usage
15:45:02narimiranshashlick: yeah, 0.20 should be *the* stable one that everybody should agree on
15:45:16narimiranas it is 1.0rc in disguise
15:47:37vivusnarimiran: do you have any suggestions for the JSON issue I highlighted above?
15:47:44vivusi've run out of ideas to find a solution
15:48:24FromGitter<liquid600pgm> damn Gitter, it glitches out and doesn't show me older messages from today
15:48:29FromGitter<liquid600pgm> their client is terrible
15:50:46nanofameI will try to setup stable tomorrow and run some libs
15:51:15shashlickokay, feel free to post here if you run into any issues
15:52:50narimiranvivus: can you share the link once again, please?
15:52:54nanofameI like Nim but for a mediocre dev like me it's too much
15:53:19vivusJSON output: https://dpaste.de/szcG ... table output: https://dpaste.de/mbgU
15:54:18vivusnanofame: I am forcing myself to use Nim even though I am probably subpar at dev. It has its quirks, but the language is nice overall . the community is ++ though, these folks help A LOT
15:54:26narimiranvivus: so what is the problem, what are you doing, etc.? i haven't been following the discussion
15:55:06nanofameHave a nice eve everyone. GN
15:55:10*nanofame quit (Quit: Page closed)
15:55:15vivusnarimiran: the table output does not contain any empty strings but the JSON output does. All I am doing to convert to JSON is: `var myjs = $$mytable`
15:58:32narimiranvivus: ok, will take a look, just let me grab something to eat first
15:58:41vivusnarimiran: sure, no rush. take your time
15:59:25*lqdev[m] joined #nim
16:00:26FromGitter<alehander42> i love you all
16:06:00*floppydh quit (Quit: WeeChat 2.4)
16:10:38narimiranshashlick: has federico3 contacted you about nightlies, and why they don't ship csources?
16:12:05*thomasross quit (Remote host closed the connection)
16:12:24federico3no he hasn't :)
16:12:31federico3shashlick: do you see my queries on IRC?
16:12:41shashlicki don't get DMs, nope 😞
16:13:04*natrys joined #nim
16:14:27narimiranvivus: it is hard for me to work with these giant files - do you maybe have some minimal example where you see these differences? also, can you share your code?
16:15:02federico3shashlick: is there anything publishing source tarballs for the nightlies similar to the releases?
16:15:51shashlickcsources are already in the releases posted online
16:16:08shashlickwhat are you looking for in specific?
16:17:34shashlicklinux x32 and x64 xz files posted online have binaries included but they are just like any other xz file so you can recompile if you prefer
16:17:44shashlicksame with all others
16:22:07federico3shashlick: I'm looking for a packages with .nim sources, csources, build.sh and no binaries
16:23:09shashlickwell, the only one without any binaries right now is the osx package
16:23:38shashlickbut all you really need to do is remove the bin directory from one of the binary releases and it is the same as a source release
16:24:07shashlickthere's 6 binaries in the bin file, other than that, everything else is identical to source release
16:24:26shashlickoh and koch
16:24:32federico3ok thanks
16:31:04narimiranvivus: my manual inspection tells me that for `"67": (f0: "...", f2: @["We don\'t see things as they are, but as we are."]),"` you introduce `f2` field, and after that for every entry in the table you get one good entry + one `{"Field0": 0, "Field1": "", "Field2": {"f0": "", "f1": null}},`
16:31:51narimiranso something is going on in your source around that place
16:33:22vivusnarimiran: just got back, sorry for not replying sooner
16:33:42FromDiscord_<kodkuce>
16:33:42FromDiscord_<kodkuce> https://cdn.discordapp.com/attachments/371759389889003532/582969750469083140/2019-05-28_183129.png
16:34:25*nsf joined #nim
16:35:33narimiranvivus: heh, that `f2` might be because of my edit, ignore that :D but there problems start :)
16:36:55narimiranvivus: look at the `f0` string in "70", "68", and so on - the last character might be the one causing problems
16:37:47narimiran"了。"
16:38:16narimiranbut this is just guessing on my part, based on "unicode is hard"
16:39:00*matt-m quit (Ping timeout: 268 seconds)
16:39:31FromGitter<zacharycarter> love you too @alehander42
16:40:03vivusnarimiran: my next idea is to try a different foreign language to see what happens there. I will try a Slavic or some other language with its own unique writing
16:42:40*matt-m joined #nim
16:43:40*ng0 joined #nim
16:44:09FromGitter<zacharycarter> could try Finnish :P I have no idea what you're doing but I'm all about everything Finland since I moved here haha
16:44:09FromDiscord_<kodkuce> do droped images create link on irc or not show at all?
16:44:34narimiranwe see a link, but we don't know what to do with that image
16:45:46FromDiscord_<kodkuce> frame it and put it in your house?
16:46:08FromDiscord_<kodkuce> btw, kodkuce==poopBot if you havent notice xD
16:46:42FromGitter<zacharycarter> I think it belongs in a museum personally
16:48:21FromDiscord_<kodkuce> duno but i was able to call this function until i added async pragma and switched to pg thingy
16:48:28narimirankodkuce i did, don't worry ;)
16:48:32FromGitter<alehander42> @zacharycarter <3
16:48:36FromGitter<alehander42> how is helsinki
16:48:43FromGitter<Varriount> Hey @zacharycarter ! :D
16:49:03FromGitter<zacharycarter> Hi guys :)
16:49:09FromGitter<alehander42> i only remember suomi of my nokia
16:49:12FromGitter<alehander42> language menu
16:49:29*elrood joined #nim
16:49:38FromGitter<zacharycarter> Haha - my Finnish is still awful - I can say thank you and hi and that's about it :P
16:50:20FromGitter<zacharycarter> Helisnki is pretty awesome - moving into my first apartment here on Saturday (I've been staying in a temporary one work got for me) - but I really like the city
16:51:33FromGitter<zacharycarter> Also my first main task at work is building a web application lmao - at least it's integrating with an actual game server and the work schedule is more rigorous, plus I work solo on my project which is nice
16:53:04FromGitter<Varriount> @zacharycarter Neat, you moved across the pond?
16:53:48*vivus quit (Remote host closed the connection)
16:54:13*neceve quit (Remote host closed the connection)
16:54:15FromGitter<zacharycarter> Yup! I managed to land a job with Wargaming's (World of Tanks) mobile division as a tools programmer
16:55:03FromGitter<zacharycarter> the hitch was I had to move to Helsinki, but having never lived outside of the US before, it's an opportunity I'm glad I came across
16:56:14FromDiscord_<kodkuce> World of Tanks sux, but nice place to go only sux its kinda cold
16:56:38*rockcavera quit (Remote host closed the connection)
16:56:39*FromGitter * Varriount dooesn't play many moble games
16:56:41FromGitter<zacharycarter> it's a pretty fun studio though - we have a music room with a drum set / guitars / electric piano and people play Warhammer 40k, MTG, etc... after hours
16:57:11FromGitter<alehander42> yes come to balkans we dont need global warming, we already have the local one
16:57:20FromGitter<zacharycarter> I played a lot of World of Warships on PC and liked it - I wasn't as into World of Tanks (played on xbox 360)
16:58:17FromGitter<zacharycarter> the weather's been pretty mild - it was cold when I got here and snowed lightly a few times, but since then it's been 10 - 20 Celsius
16:58:44FromGitter<alehander42> people should play Bach on piano, not warhammer
16:58:48FromGitter<alehander42> :P
16:59:05FromGitter<alehander42> what about your dog :O
16:59:16FromGitter<zacharycarter> he's with me :D he came along for the journey
17:00:25FromGitter<zacharycarter> I think he likes it over here more than back home haha
17:01:41FromGitter<zacharycarter> I do want to visit the Balkans btw - maybe next summer
17:02:40FromGitter<zacharycarter> oh and the head of the studio offered to buy me (with company money) warhammer 40k units this morning so I'm not complaining :P although I think I'll pass on his offer as I'd like to actually own my army
17:03:29FromDiscord_<kodkuce> my problem number 1 was i writed future insted Future
17:07:40FromGitter<zacharycarter> need to remember to proper case those type names...
17:13:46FromGitter<alehander42> Wow that's a big change
17:15:36FromGitter<alehander42> First
17:15:47FromGitter<alehander42> Is it a huskey dog
17:17:09FromGitter<zacharycarter> (https://files.gitter.im/nim-lang/Nim/KwVW/cooper.jpg)
17:17:17FromGitter<zacharycarter> Nope! He's a lab mix
17:17:50*solitudesf joined #nim
17:18:13FromGitter<zacharycarter> There's a lot more to explore though here (parks and nature everywhere), and I think he he likes being in the city where there are a lot of other people and dogs around
17:21:19FromGitter<kayabaNerve> narimiran: "Windows, Linux, BSD and Mac OS X" "C, C++ or JavaScript"
17:22:14narimirankayabaNerve heh, we discussed oxford comma and the conclusion was that we should be consistent and not use it anywhere, sorry (so my example with it will be gone)
17:22:45*salewski joined #nim
17:23:29salewskiI have posted second draft to the forum. Have added REPL and hot code reloading as suggested by pablo,
17:23:45*kapilp quit (Quit: Connection closed for inactivity)
17:23:59FromGitter<kayabaNerve> So instead of fixing oxford commas, you deleted them
17:24:00FromGitter<kayabaNerve> Why
17:24:01FromGitter<kayabaNerve> Why
17:24:02salewskibut I dont know if it already works, maybe you can remove it again. Maybe text is a bit long still,
17:24:09narimiran...delilah
17:24:13salewskibut it is easy and fast to read.
17:24:20narimiransalewski: thanks, will take a look
17:24:30salewskiBye.
17:28:08*salewski quit (Quit: WeeChat 2.3)
17:29:05FromGitter<Varriount> @zacharycarter Nice to hear that you're working in game development. :)
17:30:59FromGitter<data-man> @salewski optimisations -> optimizations ?
17:44:45FromGitter<zacharycarter> @Varriount thanks! Took me a while but I'm happy to finally have entered the industry :)
17:49:24*zyklon joined #nim
17:49:32*uvegbot quit (Quit: Konversation terminated!)
17:52:52*marcazar joined #nim
17:59:15*rockcavera joined #nim
18:02:16lqdev[m]so the VS Code Nim extension gets stuck on `Nim: check project` and crashes my PC because of abnormally high disk usage
18:02:17lqdev[m]hooray
18:04:42lqdev[m]are there any good VS Code alternatives with decent Nim support that aren't Vim?
18:05:39Mister_MagisterHello. i want to run execShellCmd to generate some image using imagemagic but i want to protect myself from nasty users strings so i would like to escape it somehow. Any ideas?
18:07:05lqdev[m]oh shit, it seems there's a terrible memory leak with Nim and nimsuggest, I can't explain why but it consumes all of my memory in seconds
18:07:27FromGitter<zacharycarter> Nimsuggest definitely leaks
18:07:33leorizeMister_Magister: there's quoteShell in the os module
18:07:39Mister_Magisterthanks!
18:07:55FromGitter<zacharycarter> it's a known issue with no fix pending currently
18:08:11lqdev[m]but it leaks so fast it takes like 30 seconds to crash my whole PC because of a lack of free memory
18:08:22leorizeexcept if you use my neovim plugin :p
18:08:33leorizethen it won't leak as much :p
18:08:39narimiranlqdev[m]: does it happen when you have some macro with illegal synax?
18:08:42*Vladar joined #nim
18:09:07lqdev[m]perhaps? no idea. it only happens with a project that uses NPeg
18:09:07narimirani've seen that take some cpu and consume memory, but never as fast as you say
18:09:13FromGitter<kayabaNerve> lqdev[m]: How much mem do you have?
18:09:26narimiranZevv: NPeg crashes vscode!? ^
18:09:29lqdev[m]kayabaNerve: 16GB
18:09:33FromGitter<kayabaNerve> Dang
18:10:04lqdev[m]leorize: I don't want to use Vim, I'm used to the GUI experience too much
18:10:33leorizethere are neovim gui's around
18:10:42leorizesome of them are on par with vscode
18:10:52narimiranlqdev[m]: i switched to neovim + leorize's plugin from vscode and GUI experience, and i like it very much
18:10:53leorizeI'm kinda allergic to electron so I don't use them :p
18:11:36narimiranbut i had some prior experience with vim and i used vim keybindings in vscode, so that part wasn't a problem for me
18:11:48FromGitter<kayabaNerve> I just use Atom
18:11:56FromGitter<zacharycarter> I just don't use nimsuggest
18:12:13lqdev[m]my experience with Atom was sub-par, I didn't have any autocomplete with Nim
18:12:17FromGitter<zacharycarter> I know Arne and a few others use Emacs
18:12:22*lqdev[m] uploaded an image: image.png (4KB) < https://matrix.org/_matrix/media/v1/download/matrix.org/YuNDCyatLsbwadeqccydiPCH >
18:12:25lqdev[m]would you look at this monster
18:12:34lqdev[m]and it got even worse
18:12:40FromGitter<kayabaNerve> lqdev[m]: The Nim plugin does almost nothing for me.
18:12:41narimirancongrats, you win!
18:12:43lqdev[m]I killed it when it reached over 8GB
18:13:10leorizehmm, isn't nimsuggest programmed to die after 4gb?
18:13:11narimiranlqdev[m]: did you use some over-complicated npeg stuff?
18:13:19lqdev[m]nope
18:13:21FromGitter<kayabaNerve> It doesn't have autocomplete. It doesn't have jump to X. It sometimes shows errors/hints and sometimes doesn't.
18:13:23lqdev[m]@kayabaNerve: it doesn't do much for me as well, it's a problem with this specific project
18:13:26narimiranleorize: yep, but this is not nimsuggest, i think
18:13:47narimiranit is probably some compile-time monster happening
18:13:50FromGitter<kayabaNerve> I miss Aporia
18:13:55FromGitter<zacharycarter> @kayabaNerve - I think you have to set VSCode up so that the Nim plugin knows where there nimsuggest binary is
18:13:58FromGitter<kayabaNerve> Not even because I liked the GUI that much
18:14:00FromGitter<kayabaNerve> Because it works
18:14:13FromGitter<kayabaNerve> @zacharycarter I refuse to use a MS IDE
18:14:39Mister_Magisterleorize: but that won't do. i want to put text in parameter that is inside "" and want to escape that text so it won't escape my ""
18:14:54FromGitter<kayabaNerve> I like text editors + terminals. The comment here is the Atom Nim plugin doesn't work. :P
18:15:22FromGitter<zacharycarter> oh
18:15:29leorizeMister_Magister: quote shell can produce an entire shell command
18:15:34FromGitter<zacharycarter> Atom is electron too no?
18:15:40Mister_Magisteri could replace it with regex and remove everything that is not a char
18:15:48Mister_Magisterleorize: hmm
18:15:51FromGitter<kayabaNerve> I just prefer Atom over every other text editor I've ever used so... it balances out.
18:15:56FromGitter<zacharycarter> Also VSCode is supposed to be an editor - not an IDE
18:15:57FromGitter<Varriount> @zacharycarter I'm *finally* finishing up University - I'm working on https://github.com/Varriount/commandant for my final course this summer (an independent project course).
18:16:00lqdev[m]this is the parser: https://github.com/liquid600pgm/rod/blob/master/src/rod/parser.nim
18:16:01FromGitter<kayabaNerve> Electron was built as Atom
18:16:09leorizeMister_Magister: oh, it's quoteShellCommand
18:16:26Mister_Magisterleorize: i used it but it changed text to '"; RM -RF --NO-PRESERVE-ROOT /' which won't do
18:16:29Mister_Magisterit will escape ""
18:16:51leorizestop quoting :p
18:16:53FromGitter<kayabaNerve> "Initially developed for GitHub's Atom editor, Electron has since been adopted by companies all over the world."
18:16:58Mister_Magisterleorize: i need quote
18:17:00leorizethe point of that proc is so that you don't have to quote
18:17:16FromGitter<zacharycarter> @Varriount - awesome!!! The project sounds neat!
18:17:16Mister_Magisterthe parameter is
18:17:18Mister_Magister-draw "text 0,0 'SADASSAJDKJLJASKLDJASLKDJASLD'"
18:17:26Mister_Magisterso ' or " will escape
18:17:54Mister_Magisterleorize: let's go different approch. can i replace text using regex?
18:18:40leorizeyes, but that's just overkill
18:18:48leorizeso what's your full command that you need to run?
18:18:58leorizethat you afraid would escape?
18:20:16Mister_Magisterhow did you use that bot to test nim
18:20:20Mister_Magister!eval ?
18:20:21NimBotCompile failed: in.nim(1, 2) Error: invalid indentation
18:20:25leorizeyep
18:20:47Mister_Magister!eval echo escape(
18:20:48NimBotCompile failed: in.nim(1, 13) Error: invalid indentation
18:20:51Mister_Magistermissclick
18:21:01Mister_Magister!eval echo escape("\"")
18:21:03NimBotCompile failed: in.nim(1, 6) Error: undeclared identifier: 'escape'
18:21:16Mister_Magister!eval import strutils; echo escape("\"")
18:21:19NimBot"\""
18:21:46Mister_Magisteri think that would be proper function to do the job
18:21:53Mister_Magisterbut i'm getting multiple definition of `nsuEscape'; /home/foidbgen/.cache/nim/bot_d/stdlib_strutils.c.o:stdlib_strutils.c:(.text+0x5e33): first defined here collect2: error: ld returned 1 exit status
18:22:08leorizethat's not how you should do it...
18:22:15leorizeshell have weird escaping rules
18:22:39leorizetry adding the `-f` param to Nim
18:22:50Mister_Magistermy command is in short convert file.png -draw "text 0,0 'SADASSAJDKJLJASKLDJASLKDJASLD'" img1b.jpg
18:23:03Mister_Magisterand i want to allow users to put text there
18:23:23leorize!eval import os; echo quoteShellCommand ["convert", "file.png", "text 0,0 'SADASSAJDKJLJASKLDJASLKDJASLD'", "img1b.jpg"]
18:23:25NimBotCompile failed: in.nim(1, 17) Error: type mismatch: got <proc (args: openarray[string]): string{.noSideEffect, gcsafe, locks: 0.}>
18:23:44leorize!eval import os; echo quoteShellCommand(["convert", "file.png", "text 0,0 'SADASSAJDKJLJASKLDJASLKDJASLD'", "img1b.jpg"])
18:23:47NimBotconvert file.png 'text 0,0 '"'"'SADASSAJDKJLJASKLDJASLKDJASLD'"'"'' img1b.jpg
18:24:00Mister_Magisterwell that won't do
18:24:08leorizewhy?
18:24:13Mister_Magisterto many """?
18:24:15Mister_Magisteralso
18:24:29leorizetry running that exact command on your shell
18:24:34leorizeit'll behave just as you expected
18:24:40Mister_Magister!eval import os; echo quoteShellCommand(["convert", "file.png", "text 0,0 ''; rm -rf /; ''", "img1b.jpg"])
18:24:43NimBotconvert file.png 'text 0,0 '"'"''"'"'; rm -rf /; '"'"''"'"'' img1b.jpg
18:24:52Mister_Magisteru think that won't escape?
18:25:01Mister_Magisteru can run it
18:25:23leorizejust did
18:25:27*clyybber joined #nim
18:25:43Mister_Magisterand it escaped
18:26:00leorizenope
18:26:01Mister_Magistercommand failed
18:26:30Mister_Magisterhttps://paste.opensuse.org/view//27287799
18:26:50leorize!eval import os; discard execShellCmd quoteShellCommand(["echo", "text 0,0 ''; rm -rf /; ''"])
18:26:53NimBottext 0,0 ''; rm -rf /; ''
18:27:19leorizeMister_Magister: ^ there, no weird injections possible
18:27:21Mister_Magisterno that's wrong parameter
18:28:02leorizewait, so you wanted the user to be able to inject more than one param?
18:28:09Mister_Magisterno no only one param
18:28:15Mister_Magisterin place of SADASSAJDKJLJASKLDJASLKDJASLD
18:28:18Mister_Magisteronly text
18:28:30Mister_Magisterremoving other chars will be simple enough tbh
18:29:03leorize!eval import os; discard execShellCmd quoteShellCommand(["echo", "text 0,0 " & "; rm -rf /"])
18:29:04FromGitter<jrfondren> `for c in suspicious_user_content: if c notin {'a'..'z', 'A'..'Z', '0'..'9', '-', '.'}: complain()`
18:29:06NimBottext 0,0 ; rm -rf /
18:29:34Mister_Magisterkiiind of
18:29:42leorize!eval import os; discard execShellCmd quoteShellCommand(["echo", "text 0,0 " & quoteShell "; rm -rf /"])
18:29:45NimBottext 0,0 '; rm -rf /'
18:29:48Mister_Magisteryou can't replace with regex?
18:29:55FromGitter<jrfondren> you can.
18:29:57leorizeyou can, but it's slow
18:30:19FromGitter<jrfondren> anyway, if your only concern is that the shell not interpret your command in any way, then don't use execShellCmd
18:30:28Mister_Magister i would need to write some function like you did leorize and clear bad characters
18:30:35FromGitter<jrfondren> os osproc.startProcess
18:30:52FromGitter<jrfondren> then the user can give binary data for all you care. as long as command you're running is OK with it.
18:31:48Mister_Magisternot sure as i have to connect text they give me with parameter strign
18:33:19Mister_Magisterif the text was standalone parameter that would be easy but it isn't
18:33:52FromGitter<jrfondren> that's not a problem at all.
18:34:12FromGitter<jrfondren> `"connected text " & userstuff` can just be one of the parameters.
18:34:22leorizeyou can see my example just handled it well
18:34:32FromGitter<jrfondren> unix processes are passed arguments as an array, not a string.
18:35:02Mister_Magisterso startProcess has no way of injecting second command?
18:35:20FromGitter<jrfondren> if you don't tell it to process your command with the shell, the shell is not involved.
18:35:20leorizeas long as you don't pass poEvalCommand, yes
18:35:44FromGitter<jrfondren> and shell metacharacters have no meaning, and are just passed to the program as part of its arguments, as if you'd gone through this whole quoting rigamarole.
18:35:51Mister_Magisterlet's try it
18:36:00Mister_Magisterthe startProcess
18:37:36FromGitter<jrfondren> you'll probably want poUsePath and poParentStreams
18:37:48Mister_Magisterwhat do you mean?
18:38:06leorizeread the docs :p
18:39:33Mister_Magisteri mean what you mean
18:39:34Mister_Magisterxd
18:41:02Mister_Magisterno such command huh
18:42:46FromGitter<jrfondren> you'll probably want poUsePath and poParentStreams
18:43:14Mister_Magisterxdd
18:43:27Calinouhow do I pass an openarray[string] to a function?
18:43:38CalinouI used to be able to pass just a string, but it seems the Nim compiler now wants me to pass that typer
18:43:40Calinou-r
18:43:44FromGitter<jrfondren> pass it an array[##, string] or a seq[string]
18:44:05FromGitter<jrfondren> sure it wasn't a varargs before?
18:44:15leorize["string"] :p
18:44:21Calinouit's execProcess() from osproc
18:45:38Calinouhttps://0x0.st/zaUI.txt
18:45:41Calinouthat's what I get
18:45:47Calinou(the 2nd argument is the culprit)
18:45:58FromGitter<jrfondren> where's your code?
18:45:59Calinouthat is, if I try to use ["string"] as the 2nd argumen,t
18:46:13Calinouhttps://github.com/Calinou/godot-size-benchmarks/blob/master/src/godot_size_benchmarks.nim#L108-L114
18:46:14FromGitter<jrfondren> oh, last line
18:46:50CalinouI'm using execProcess 3 times and they're all throwing errors
18:46:55Calinouit worked back in March :P
18:47:03FromGitter<jrfondren> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ced8227faac64393400aaa5]
18:47:04Calinou(I'm on 0.19.6 now)
18:47:11FromGitter<jrfondren> the `[godot]` is fine
18:50:28*natrys quit (Ping timeout: 245 seconds)
18:51:32Calinouthe compiler doesn't want me to use either an array or a sequence to match an openArray[string]
18:51:42FromGitter<jrfondren> that's not the problem
18:52:02FromGitter<jrfondren> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ced835282c2dc79a511622a]
18:53:32Calinouoh, right
18:54:41Calinouoh, so I need to use key-value pairs for parameters now, I guess
18:55:02FromGitter<jrfondren> no.
18:55:07FromGitter<jrfondren> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ced840b7c1dab468c617a1e]
18:55:13FromGitter<jrfondren> env:, that's the one with StringTableRef
18:55:56FromGitter<jrfondren> ah but I can imagine why you think it's a problem with parameters
18:56:04FromGitter<jrfondren> sconsFlags are your params to `godot`?
18:56:31FromGitter<jrfondren> or rather, `godot` is the first parameter and you have a list of flags to pass as well
18:56:39Calinouyeah
18:56:42FromGitter<jrfondren> pass them in a single openArray[string], not two of them.
18:57:04Calinouoh, so the cwd argument was removed?
18:57:14Mister_Magisterhow to get output from startProcess because it's not working
18:57:36FromGitter<jrfondren> Mister, where do you want your output?
18:57:48Mister_Magisterto shell
18:57:49leorizenarimiran: should I proceed with adding azure pipelines to Nim?
18:57:58Mister_Magisterlike echo
18:58:25FromGitter<jrfondren> jrfondren @jrfondren 13:37 ⏎ you'll probably want poUsePath and poParentStreams
18:58:57Calinounow, what would be the best way to run each command in the right directory in a cross-platform way?
18:59:15FromGitter<jrfondren> @Calinou, `execProcess(scons, args=@[godot] & sconsFlags, options={poUsePath, poStdErrToStdOut})`
18:59:28Calinougodot isn't an arg, it's the CWD
18:59:32narimiranleorize: if you can setup it to completely replace our current travis+appveyor, that would be great. btw, how does the output look like? e.g. i like appveyor more than travis - it is much easier to see which tests failed
18:59:46Mister_Magister@jrfondren am i stupid? lol
19:00:52FromGitter<jrfondren> @Calinou, don't think there's any way but to repeatedly os.setCurrentDir
19:01:17*natrys joined #nim
19:01:26leorizenarimiran: https://dev.azure.com/alaviss/Nim/_build/results?buildId=8
19:02:06FromGitter<jrfondren> @Calinou, the old code probably did that for you, so doing it yourself should at least be marginally more efficient
19:02:26FromGitter<jrfondren> since you can restore your old working directory just once instead of N times.
19:03:05*rnrwashere quit (Remote host closed the connection)
19:03:51*rnrwashere joined #nim
19:03:55*rnrwashere quit (Remote host closed the connection)
19:04:07*rnrwashere joined #nim
19:04:21Mister_Magister@jrfondren sadly command using startProcess is not working
19:04:57FromGitter<jrfondren> I'm sure it's not a moral failing. what's the problem?
19:06:08Mister_Magisterwait i just fixed it xd sorry for disturbing
19:06:13Calinouthanks for the help :) trying to run the script now
19:07:08FromGitter<jrfondren> np. I'm a little out of it so it took me a while to see what you were doing.
19:07:34Calinouthis is still easier than doing it all in Bash :P
19:07:52Calinouplus, it's much easier to make it work natively on Windows (if I ever need that)
19:08:44Mister_Magister@jrfondren you said it won't escape the command but it did! one parameter is "text 190,410 '" & toUpper(command.params) & "'" and if i put ' inside command.params it will escape parameter
19:09:40FromGitter<jrfondren> what do you mean by 'escape'?
19:09:41*couven92 quit (Quit: Client disconnecting)
19:10:21Mister_Magisterin command it looks like convert -text "text 190,410 ' text ' "
19:10:30FromGitter<jrfondren> what's escaped about that?
19:10:36Mister_Magisterif you put ' instead of text it will interpret the text as parameter
19:10:53Mister_Magisterand that's what's happening
19:12:07Mister_Magisterand i can't use escape function from strutils for some reason
19:12:53Mister_Magistermultiple definition of `nsuEscape'; /home/foidbgen/.cache/nim/bot_d/stdlib_strutils.c.o:stdlib_strutils.c:(.text+0x5e33): first defined here collect2: error: ld returned 1 exit status
19:13:20FromGitter<jrfondren> well that's a serious problem on its own, but it's unrelated
19:13:37FromGitter<jrfondren> how are you invoking startProcess?
19:13:53Mister_Magisterdiscard startProcess("convert", "", ["PiS.png", "-font", "font.ttf", "-pointsize", "81", "-fill", "white", "-draw", "text 190,410 '" & toUpper(command.params) & "'", $command.message.messageId], nil, {poUsePath, poParentStreams})
19:14:21Araqdon't 'include' stdlib modules
19:14:46Mister_Magistero dpnm
19:14:48Mister_Magisteri don't
19:14:56FromGitter<jrfondren> yeah, that's not going to pass anything to the shell. convert's going to get "text 190,410 'blah''" (sans quotes) as a param, supposing that command.params == "blah'"
19:15:32FromGitter<jrfondren> I don't know convert though. if convert itself has an expression language there that requires some kind of escaping, obviously you'll have to do that still
19:15:40Mister_Magisterand if the command.params is like ' -color black it will change color
19:15:48Mister_Magisterbecause it will interpret -color as paramter
19:16:06Mister_Magisterbecaue ' in string will escape parameter ' '
19:16:14FromGitter<jrfondren> that's convert being weird then, because startProcess absolutely isn't passing -color as a separate param based on command.params
19:16:30FromGitter<jrfondren> convert might be a bash script :/
19:16:31Mister_Magister@jrfondren the problem is ending parameter
19:16:38Mister_Magisternot the convert
19:16:50Mister_Magisterif user put ' blah command wil lfail
19:17:05FromGitter<jrfondren> that's absolutely something convert is doing
19:17:18FromGitter<jrfondren> strace your Nim script and look at the exec() syscall.
19:17:22Mister_Magisterfrom the beginning i want to escape ' and " tho
19:17:37Mister_Magisterthere is no command to do that?
19:17:51FromGitter<jrfondren> aye, so the conclusion is that startProcess won't help you and you'll need to do this yourself.
19:18:10FromGitter<jrfondren> well you said strutils had a command you wanted to use.
19:18:14Mister_Magisterand we are back to my first question
19:18:19Mister_Magisterhow do i escape characters
19:18:29Mister_Magister@jrfondren but can't compile
19:18:38FromGitter<jrfondren> then ask a question about that. it works fine for me.
19:19:05Mister_Magisteri have problem when i'm using escape from strutils multiple definition of `nsuEscape'; /home/foidbgen/.cache/nim/bot_d/stdlib_strutils.c.o:stdlib_strutils.c:(.text+0x5e33): first defined here collect2: error: ld returned 1 exit status
19:19:19FromGitter<jrfondren> escapes still may not work though. convert is *already* completely out in the woods with its current behavior. I wouldn't trust it at all.
19:20:41FromGitter<jrfondren> !eval import strutils; echo "hello '".escape
19:20:44NimBot"hello \'"
19:21:53FromGitter<rishavs> Vladar, are you there on IRC?
19:22:01FromGitter<jrfondren> sure you're not using `include` elsewhere? it may not be enough for this one part you're looking at to have `import`
19:22:20Mister_Magistermaybe my lib somehow but doubt
19:23:13FromGitter<jrfondren> rather than use escapes and hope that convert doesn't have bugs in addition to the completely buggy argument handling already, I suggest using a regex or set[char] to require only valid characters in the text you're including
19:23:33Mister_Magisteri will just replace(s, "'", "")
19:23:34*rnrwashere quit (Read error: Connection reset by peer)
19:23:37Mister_Magisterand go to sleep xd
19:24:01FromGitter<jrfondren> see if \x27 doesn't break it still
19:25:15FromGitter<jrfondren> `\x27 -color black` and `\047 -color black`
19:25:16Zevvsomeone called?
19:25:29*rnrwashere joined #nim
19:25:34Vladarrishavs: what's up?
19:26:41Zevvnarimiran: wow, I crashed vscode?
19:26:44Zevvhow?
19:27:07FromGitter<rishavs> Hey, i was going through the bounce tut for ng2. In the part where I add the fonts to the title scree, i am getting this error ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ced8b8bb76eac527a654f5e]
19:27:37narimiranZevv: "<lqdev[m]> perhaps? no idea. it only happens with a project that uses NPeg"
19:27:47Zevvcool
19:28:11narimiranzevv: "<lqdev[m]> but it leaks so fast it takes like 30 seconds to crash my whole PC because of a lack of free memory"
19:28:18Zevvwoohoo \o/
19:28:29Zevvfix your frigging editor then!
19:28:43ZevvMy vim works just fine
19:28:45Mister_Magister@jrfondren nope it doesn't
19:29:12FromGitter<jrfondren> that's good at least
19:29:12Zevvso, how do I reproduce this? I just install vscode .deb
19:29:18Zevvand now it does not know what Nim is
19:29:39Mister_Magister@jrfondren actually replace(toUpper(command.params), "\'", "\\'") does the job
19:29:42Mister_Magisteraka escaping it
19:29:55Mister_Magisterso all i had to do is to escape characters
19:29:56FromGitter<jrfondren> ................ what
19:29:59narimiranZevv: this is the code: https://github.com/liquid600pgm/rod/blob/master/src/rod/parser.nim
19:30:06Mister_Magisternothing
19:30:11Mister_Magisterworks :) thanks :)
19:30:53FromGitter<jrfondren> isn't that strutils.replace that you're using there?
19:30:54narimiranZevv: there is nim plugin for vscode, but maybe this can be reproduced without vscode?
19:31:10FromGitter<rishavs> Vladar, i was going through the bounce tut for ng2. In the part where I add the fonts to the title scree, i am getting this error ⏎ ⏎ ```code paste, see link``` ⏎ ⏎ What library am i to initialize there? I do have the sdl_ttf dll in my bin folder [https://gitter.im/nim-lang/Nim?at=5ced8c7e5de053468bd47997]
19:31:50Zevvnarimiran: Ah, yes, nim c parser.nim already blows stuff up
19:32:06narimiranas i expected :)
19:32:15Zevvlemmesee
19:32:23narimiranlqdev[m]: are you here, btw?
19:32:41Vladarrishavs: hmm, and the path to the font exist, I assume? Might be some missed init proc. Can you drop your code into pastebin or something, so I could look at it?
19:33:55FromGitter<rishavs> path is correct, because if I try to mess that up the error becomes `CRITICAL: Can't load font data/fnt/FSEsX300.ttf: Couldn't open data/fnt/FSEsX300.ttf` ⏎ I will copy paste the code into pastebin and get back to you
19:37:12Zevvliquid600pgm: you here?
19:37:28FromGitter<rishavs> Vladar, I have added the code to a repl https://repl.it/repls/ElderlyBountifulDevices
19:38:07*matt-m quit (Ping timeout: 248 seconds)
19:40:49Zevvnarimiran: the grammar explodes exponentally in its implementation and it generates a zillion lines of Nim
19:41:00*matt-m joined #nim
19:41:03narimiranas i expected :)
19:41:29ZevvI'll have to put some safeguard in I guess
19:41:49narimiran"<narimiran> lqdev[m]: did you use some over-complicated npeg stuff?", one and a half hour ago
19:44:13Vladarrishavs: if you didn't rename the font, it looks like you have a typo in the font file name (data.nim:20 and data.nim:24). Could you check it before I go deeper?
19:45:00FromGitter<rishavs> My bad. I just added that typo to get the normal path error
19:45:17FromGitter<rishavs> bounce.zip (https://files.gitter.im/nim-lang/Nim/MkKw/bounce.zip)
19:45:51FromGitter<rishavs> Added a zip of my project folder in case you want to compile it. Not sure if you can see it in IRC
19:45:52*Jesin quit (Quit: Leaving)
19:45:54Vladarrishavs: also, loadData() call should be _after_ game init
19:46:50Vladarso you are calling ttf procs before the initialization (that happens in game.init())
19:48:37Vladarlet me know if this helps
19:49:09FromGitter<rishavs> That fixed it! After placing the loaddata func call correctly, the error disappeared
19:49:25Vladarawesome )
19:49:40FromGitter<rishavs> Thanks! the error made me think that I was missing some dependency which is not there in the tut
19:51:14Zevvnarimiran: how did you hear of this memory-eating problem?
19:51:50Vladarrishavs: unmet lib deps usually crash the program before it gets to the actual game logic, in my experience.
19:52:43FromGitter<rishavs> 👍 got it!
19:53:17narimirani think either you mentioned it some time ago about npeg specifically or just the general well-known "macros can explode at CT if you're not careful"
19:53:58*elrood quit (Remote host closed the connection)
19:56:35Mister_Magisterhow can i make round return int? without .0 at the end
19:56:40*matt-m_ joined #nim
19:56:49*nsf quit (Quit: WeeChat 2.4)
19:56:51narimiranMister_Magister: you convert the result to int? :)
19:56:57Mister_Magisterbut how
19:57:00Zevvno, I mean, who reported it just now
19:57:05narimiranint round(3.2)
19:57:13Zevvoh wait, lqdev == liquid600pgm
19:57:17Zevvlqdev[m]: you here?
19:57:39Zevvdon't change your nick dude, I have enough trouble to keep people apart in real life already
19:58:00narimiranyeah, lqdev[m] reported it two hours ago, see the logs
19:58:21narimiranoff i go, see y'all tomorrow
19:58:27Zevvciao!
19:58:55Mister_Magisterint(somethin_
19:58:57Mister_Magistercool
19:59:25*matt-m quit (Ping timeout: 268 seconds)
20:00:12*Trustable quit (Remote host closed the connection)
20:01:50*Jesin joined #nim
20:02:46*narimiran quit (Ping timeout: 246 seconds)
20:07:47*matt-m_ quit (Ping timeout: 258 seconds)
20:10:45*absolutejam joined #nim
20:11:37*absolutejam left #nim (#nim)
20:12:22*absolutejam joined #nim
20:12:53*absolutejam quit (Client Quit)
20:15:22*couven92 joined #nim
20:15:35*ShalokShalom quit (Remote host closed the connection)
20:16:28Zevvnarimiran[m] / lqdev[m]: I now added a safeguard that prevents exponential grammar explosions, and added a note in the docs on how to prevent that.
20:19:00*cybj quit (Ping timeout: 252 seconds)
20:20:33*cybj joined #nim
20:22:41*sealmove joined #nim
20:25:49*shadowbane quit (Quit: Konversation terminated!)
20:35:01*coffeeturtle joined #nim
20:40:35*coffeeturtle quit (Remote host closed the connection)
20:49:59FromDiscord_<DeltaPHC> When you have a ref object type, is using the `new` proc required, or does it do that implicitly when you do `MyObjectRef(field: foo, ...)`?
20:51:56FromGitter<alehander42> the last thing is enough
20:52:13*rnrwashere quit (Remote host closed the connection)
21:13:24*Vladar quit (Remote host closed the connection)
21:22:46*sealmove quit (Quit: WeeChat 2.4)
21:27:21*marcazar quit (Quit: Leaving)
21:32:44dom96marcazar: not sure if you'll read IRC logs, but you /just/ missed me :)
21:35:41*deech_ joined #nim
21:40:19*laaron quit (Remote host closed the connection)
21:40:43*laaron joined #nim
21:41:38*rnrwashere joined #nim
21:42:28dom96So now that Rust will use the `future.await` syntax, should we also support it?
21:45:25FromDiscord_<DeltaPHC> If it already works in prefix notation, then doesn't UFCS already allow .await?
21:46:18FromDiscord_<DeltaPHC> (I haven't used async stuff in Nim)
21:49:58*solitudesf- joined #nim
21:49:59*solitudesf quit (Remote host closed the connection)
21:51:58*vlad1777d quit (Ping timeout: 246 seconds)
21:58:20FromDiscord_<DeltaPHC> I'm still learning, but Nim seems pretty flexible in both its syntax and metaprogramming features. I wouldn't be surprised if you can use existing mechanisms to do postfix await
21:59:11*rnrwashere quit (Remote host closed the connection)
22:01:28*apodo quit (Ping timeout: 272 seconds)
22:04:30*Jesin quit (Quit: Leaving)
22:05:06*deech_ quit (Remote host closed the connection)
22:06:14*Jesin joined #nim
22:07:27FromDiscord_<DeltaPHC> That being said, there's definitely been a lot of controversy about Rust's decision. I think it's well reasoned, even if opinions aren't unanimous
22:08:04*solitudesf- quit (Ping timeout: 246 seconds)
22:14:59*sealmove joined #nim
22:18:30*ng0 quit (Quit: Alexa, when is the end of world?)
22:30:24*kapilp joined #nim
22:52:13*natrys quit (Quit: natrys)
23:05:57*rnrwashere joined #nim
23:08:56*krux02 quit (Remote host closed the connection)
23:16:53*Jesin quit (Quit: Leaving)
23:17:58*Jesin joined #nim
23:30:31*smitop joined #nim
23:47:51*clyybber quit (Quit: WeeChat 2.4)
23:58:24Calinouheh, nice to see DeltaPHC in here :P
23:58:48*couven92 quit (Quit: Client Disconnecting)