<< 04-08-2020 >>
00:00:00*njoseph quit (Quit: http://quassel-irc.org - Chat comfortably. Anywhere.)
00:01:05*njoseph joined #nim
00:04:50*SenasOzys quit (Ping timeout: 256 seconds)
00:17:22ForumUpdaterBotNew thread by Shirleyquirk: Collect and list comprehensions, see https://forum.nim-lang.org/t/6639
00:18:25*FromDiscord quit (Remote host closed the connection)
00:18:26disrupteki don't want the compiler to puke on an import just because i don't have it in a directory named foo-123123123
00:18:29disruptekthat's stupid.
00:18:39*FromDiscord joined #nim
00:28:56*vicfred quit (Quit: Leaving)
00:38:07*audiofile quit (Quit: Default Quit Message)
01:10:37*apahl quit (Ping timeout: 244 seconds)
01:12:34*apahl joined #nim
01:13:20*vicfred joined #nim
01:19:50*maier joined #nim
01:24:45*maier quit (Ping timeout: 240 seconds)
01:25:52*Tongir joined #nim
01:28:01*Tlongir quit (Ping timeout: 246 seconds)
01:45:41*endragor joined #nim
01:58:28*krux02_ quit (Remote host closed the connection)
02:04:30*sepplesWoona quit (Quit: Bye)
02:05:10*endragor quit (Remote host closed the connection)
02:05:18*endragor joined #nim
02:07:22*waleee-cl quit (Quit: Connection closed for inactivity)
02:28:53*sepples joined #nim
02:57:53*muffindrake quit (Ping timeout: 246 seconds)
03:00:25*muffindrake joined #nim
03:20:42*maier joined #nim
03:25:50*maier quit (Ping timeout: 256 seconds)
04:06:02*supakeen quit (Quit: WeeChat 2.8)
04:06:45*supakeen joined #nim
04:10:31*rockcavera quit (Remote host closed the connection)
04:22:35shashlickwhat's @mratsim timezone
04:22:35*FromDiscord quit (Remote host closed the connection)
04:22:41disruptekparis
04:22:48*FromDiscord joined #nim
04:23:14shashlickokay will have to wait then
04:23:24shashlickplaying with weave but isReady doesn't work
04:40:11*narimiran joined #nim
04:56:27*lritter joined #nim
05:10:52disrupteka macro that uses a custom exception fails to compile.. codegen error due to inheritance.
05:14:50disruptek!repo disruptek/testes
05:14:51disbothttps://github.com/disruptek/testes -- 9testes: 11 15 0⭐ 0🍴
05:21:35*maier joined #nim
05:26:32*maier quit (Ping timeout: 256 seconds)
05:34:12FromDiscord<Varriount> @disruptek I think Araqs on vacation
05:37:28Zevvdisruptek: go to bed man
05:53:15FromGitter<alehander92> disruptek
05:53:16FromGitter<alehander92> morning
05:53:20FromGitter<alehander92> i just woke up
05:53:26FromGitter<alehander92> which means it must be late there
05:55:32*solitudesf joined #nim
06:04:34*maier joined #nim
06:07:41FromDiscord<Varriount> Alehander92: Approximately 1 or 2 in the morning
06:08:02FromDiscord<Varriount> Perhaps an hour earlier
06:11:32ZevvI think he's six after nl, so 2 in the night now
06:19:35FromDiscord<vieru> hi quick noob question
06:19:52FromDiscord<vieru> can i compile a lib in visual studio
06:20:02FromDiscord<vieru> amd then include it in nim gcc?
06:20:09FromDiscord<vieru> (edit) 'amd' => 'and'
06:20:48PMunchdisprutek and Zachary Carter, thanks for the wb :)
06:20:57PMunchAnd Zevv, I've noticed that TOML parsing isn't trivial..
06:21:11PMunch(managed to forget my IRC client open at work..)
06:21:28FromDiscord<Elegant Beef> Oh shit pmunch πŸ˜„
06:21:58supakeenWooo toml.
06:22:25PMunchOh and Clyybber as well, yes I'm back :)
06:22:44PMunch(Just looking at the IRC logs, my scrollback wasn't long enough..)
06:22:59PMunchPrestige, is the cane from something?
06:23:48PMunchsupakeen, it's a mess :P
06:25:57supakeenI do like the format, not as much as UCL but I like it.
06:27:09PMunchUCL?
06:27:20supakeenhttps://github.com/vstakhov/libucl
06:27:49PrestigePMunch: the person who made it said it was the cane thing from the cover of Nim in Action
06:28:24ZevvPMunch: well, it's not *hard* but the spec is just flimsy
06:29:22Zevvand some thing are a bit shitty for parsing, like the dates colliding with numbers requiring a certain amount of backtracking
06:29:50PMunchPrestige, aha, that random image of the "in action" series? :P
06:29:55supakeenPMunch: I believe the project came from FreeBSD but I always liked its 'you can include a URL with a signing key'.
06:30:02PMunchZevv, tell me about it..
06:30:34PMunchZevv, here is a diagram I made for parsing "values" in TOML: https://nim-lang.org/docs/critbits.html
06:30:44PMunchWoops: https://uploads.peterme.net/tomlvalues.png
06:31:14FromDiscord<Varriount> PMunch: But everyone says TOML is uber times better than every other format.
06:31:36PMunchI mean the format is pretty nice :P
06:31:49PMunchBasically a standardised INI file with some extra niceties
06:32:10PMunchVery human-readable (and by extension writeable)
06:32:11PrestigeYeah :P what are you working on now that you're back btw?
06:32:28PMunchWell, I just got back, so haven't really spun up anything yet
06:32:45PMunchAlthough I did play around with implementing stacklang v2 while I was gone
06:32:57ZevvPMunch: is that diagram proven & tight
06:33:00FromDiscord<Varriount> PMunch: I just want tags and merge maps for whatever configuration language I'm using
06:33:26PMunchI also started looking up some more stuff for my configuration thing for my VM
06:33:29PMunchWM*
06:33:37ZevvPMunch: http://ix.io/2ta3
06:33:48PMunchZevv, it's what I used for the Nim TOML parser that passes all the tests
06:33:52PMunchSo it should be good
06:34:01Prestigewhich part PMunch, the nimscript?
06:34:09PMunchVarriount, tags and merge maps?
06:34:10Zevvdude you *got* a parser. So why am I even looking at this stuff
06:34:30Zevvrm -rf ./*
06:34:32PMunchPrestige, yes it will be using NimScript :)
06:35:02PMunchBasically the configurator is going to be able to write out configurations for all kinds of programs
06:35:12PMunchA configuration being a program with a set of options
06:35:31PMunchSo you can call a configuration and the configurator will start the correct program with the correct options
06:35:41PrestigeThat's going to be awesome
06:36:01PMunchIt will be able to cross-check all the keyboard shortcuts of your programs, and it will allow you to re-use variables across your configs (like colour codes for example)
06:36:16PrestigeI have a good idea to increase the number of tags/workspaces but I need to iron out bugs first
06:37:13PMunchAnd it will support programs that have callbacks (e.g. i3 which can launch programs on keyboard shortcuts) to run NimScript snippets within the configurator itself
06:37:23PMunchIncrease how?
06:37:24FromDiscord<Varriount> PMunch: Tags essentially allow you to add metadata to a value or key, merge maps let you reference and merge in key/values from somewhere else.
06:37:38ZevvPMunch: do everything xmonad does and Im your man!
06:37:42PMunchZevv, well it's for 0.5.0, not 1.0.0
06:37:49PMunchThat's why I started looking at it again
06:38:12PMunchProblem is just that there is no official test-suite, so it's kinda hard to verify that you're compatible..
06:38:23FromDiscord<Varriount> The former is often used to serialize data types that can't be expressed in the configuration language, but can also be used as a kind of function call syntax
06:38:38PrestigePMunch: like having 81 workspaces per monitor, comfortably. Basically, a 2D array of workspaces, and you select which "row" you are viewing
06:38:43PMunchVarriount, hmm not entirely sure if I get how that works..
06:38:47FromDiscord<Varriount> The latter just helps shorten long, repeated configurations
06:39:00alehander92zevv i read "i do everything xmonad does and im your man"
06:39:05alehander92and it sounded like a dystopian ad
06:39:22PMunchZevv, well the idea is to make it easily extensible, so it would definitely be able to do anything xmonad can do, although you might have to write some code for it :P
06:39:25alehander92a bit, where people need to pretend they are window managers
06:39:41Zevvoh, but that's kind of my philosophy. I have an hourly fee, and as long as I feel my physical integrity is not at stake I'll do about anything for that fee
06:40:15alehander92so flying on rockets is not it
06:40:18ZevvI architect your next gen cloud hype, whip up some compiler extensions, wash your car or manage your windows all day
06:40:21ZevvI don't really caer
06:41:00alehander92maybe this is what i should write in my resume
06:41:03PMunchHaha, a personal window manager :P
06:41:09FromDiscord<Varriount> PMunch: For example, if you need to serialize a set type, how would you do that in a configuration language that only supports lists and maps?
06:41:29alehander92map with T, bool?
06:41:31Zevvvarriount: how can you not represent a set as a list?
06:41:37PMunchJust as a list?
06:41:39FromDiscord<Varriount> You might do it like: `!Set [1,2,3]`
06:41:59FromDiscord<Varriount> So when you deserialize the data, you don't lose the type information
06:42:00PMunchAh, you want to guarantee that the reader knows it's a set
06:42:06PMunchI see
06:42:25PMunchI guess you could do that in TOML with an extension as you have "complex keys"
06:42:41PMunchBut that would require you to put your key in quotes
06:42:55PMunchI guess an inline object could also do it
06:43:00FromDiscord<Varriount> For example, Python's YAML library allows serializing arbitrary objects. It places the object type in the tag, rather than coming up with some odd encoding scheme.
06:43:16PMunch`mykey = {type: set, data: [1, 2, 3]}`
06:43:32PMunchBut now we're getting into data exchange more than configurations..
06:43:32FromDiscord<Varriount> Cloudformation uses tags to represent function calls: https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/intrinsic-function-reference-sub.html
06:45:20FromDiscord<Varriount> (Purely because it's easier to type than `{"Fn::Sub": [...]}`
06:45:33FromDiscord<Elegant Beef> Pmunch, did you see that i rewrote my rofi thing, so it's not a weird tangled mess? πŸ˜„
06:45:46FromDiscord<Elegant Beef> *It's easily expandable now using object variants*
06:45:52PMunchOh, no I missed that!
06:45:59PMunchLink?
06:46:03PMunchWoo
06:46:12FromDiscord<Elegant Beef> https://github.com/beef331/nim_rofi
06:46:28FromDiscord<Elegant Beef> I mean it's super compact, but i added features after talking about it
06:46:52FromDiscord<Elegant Beef> now you can give titles/subtitles so it doesnt say "dmenu"
07:07:04PMunchOh right, this was the documentation thingamajig
07:14:48FromDiscord<Elegant Beef> Well more than that, but yea πŸ˜„
07:16:01*superbia joined #nim
07:17:07*SenasOzys joined #nim
07:18:42*Vladar joined #nim
07:26:27*superbia quit (Quit: WeeChat 2.9)
07:27:28*superbia joined #nim
07:32:54*Guest3540 quit (Ping timeout: 256 seconds)
07:40:20alehander92guys
07:40:24alehander92what is stale bot <3
07:40:42alehander92it's a good idea actually
07:40:50alehander92some kind of triage
07:40:53FromDiscord<Varriount> A bot that has been out in the open too long?
07:42:54FromDiscord<Varriount> @alehander92 or do you mean those "stale issue" bots some projects use?
07:45:08FromDiscord<Varriount> Who added that?
07:49:55alehander92yeah, i just saw it
07:50:21FromDiscord<Varriount> I'll be blunt. I _hate_ those bots.
07:50:59FromDiscord<Varriount> In my mind, they're good for nothing but optics.
07:53:17FromDiscord<Varriount> Too often I've found *valid* bugs in various open-source programs, and they had been reported, but closed because the developers never got around to fixing them.
07:54:45alehander92well .. it makes some kind of sense
07:54:52alehander92if you can still reopen the issue/find it
07:55:07alehander92a open / stale / closed might be a better distinction
07:55:09FromDiscord<Elegant Beef> Closing issues that arent resolved is like closing your blinds when your house is on fire
07:55:12*Vladar quit (Quit: Leaving)
07:55:23FromDiscord<Varriount> Yes, exactly.
07:55:27alehander92i doubt people can solve all issues in today's projects
07:55:44*alehander92 so it's a bit misleading to expect that they all need to be "open+solveable"
07:55:46alehander92ops
07:55:54FromDiscord<Elegant Beef> Issues are to be documented and if possible fixed, not just acknowledge and removed
07:56:07alehander92i agree: but open / stale / closed would be ok
07:56:20skrylar[m]owo
07:56:23FromDiscord<Varriount> Yes, but GitHub doesn't have that distinction
07:56:29alehander92because it's good to have a notion of issues that people actually might work on
07:57:05*alehander92 yeah, priority and stuff: i think there are project "tables"?
07:57:13alehander92o i should stop clicking that
07:57:23FromDiscord<Varriount> > This issue has been automatically marked as stale because it has not had recent activity. If you think it is still a valid issue, write a comment below; *otherwise it will be closed*. Thank you for your contributions.
07:57:56FromDiscord<Varriount> I have no problem with the label, I take issue with closing perfectly valid bug reports
07:57:58alehander92it's also a reminder to review older issues
07:58:20alehander92ok, so what about wontfix
07:58:31alehander92labels
07:58:41FromDiscord<Varriount> Sure, close wontfix.
07:58:42PMunchNice, the playground has updated to 1.2.6 automatically :)
07:58:58PMunchAnd it has installed 1.0.8
07:59:02FromDiscord<Varriount> After all, they won't be fixed, and that's been confirmed.
08:00:03FromDiscord<Varriount> But closing issues that just happen to be old benefits no one. All it does is frustrate people who might otherwise add to an existing bug report.
08:00:19alehander92but they can reopen
08:00:26alehander92i think
08:02:06FromDiscord<Varriount> I don't believe so. I can't reopen issues on repositories I don't have specific privileges on.
08:02:43FromDiscord<Varriount> And who actually looks at updated closed issues?
08:05:23PrestigePMunch: oh I was going to tell you - nimlsp seems to need the lsp client to restart if multiple files are edited (it doesn't see the changes otherwise). Doesn't seem to be the case with other lsp servers
08:05:44Prestigewe were talking about this before you left I believe ^
08:07:49FromDiscord<Clyybber> Ugh, stalebot..
08:08:17FromDiscord<Varriount> @Clyybber Read my rant above.
08:08:24FromDiscord<Clyybber> Fully agree
08:08:55FromDiscord<Varriount> I do not need this at 4 in the morning
08:09:34FromDiscord<Clyybber> @dom96 do you know more about this?
08:10:33FromDiscord<Clyybber> or perhaps narimiran
08:10:56narimiranstalebot is on me
08:11:12FromDiscord<Varriount> Please. Remove it.
08:11:54narimiran"But closing issues that just happen to be old benefits no one." -- not true
08:12:42FromDiscord<Elegant Beef> I really hope there is an explanation written to support that and not just two words in response πŸ˜„
08:13:11FromDiscord<Varriount> Then do the benefits outweigh the downsides?
08:13:59narimiranwe think so
08:14:30FromDiscord<Varriount> And what are the benefits?
08:17:52FromDiscord<Yardanico> Rip my GitHub notifications
08:19:12FromDiscord<Clyybber> this
08:19:26narimiranability to focus on issues that are still relevant in 2020. and that includes the issues that stalebot has left comment on, and anybody replies "oh, wait, this is still a valid issue, don't close it". we have a very limited manpower and we would like (both us and our contributors) to prioritize issues that are bothering people using nim 1.2.x in 2020, and not issues that bother somebody who may have already left nim 2 years ago
08:19:51FromDiscord<Varriount> I think part of that message got cut off.
08:20:33narimiranif we had unlimited manpower, i agree with you that stalebot shouldn't be needed. but in reality, we (unfortunately?) need it
08:20:40FromDiscord<Yardanico> @Varriount do you really not trust my bridge like that?
08:20:45FromDiscord<Yardanico> https://media.discordapp.net/attachments/371759389889003532/740122033500979240/Screenshot_2020-08-04-11-20-26-153_com.iskrembilen.quasseldroid.jpg
08:21:41FromDiscord<Clyybber> The manpower required to keep these from getting closed is bigger
08:22:09FromDiscord<Varriount> And "we" is?
08:22:24FromDiscord<Clyybber> And I really don't think we need it. Its not like old issues jump in our face and force us to fix them now instead of focussing on whatever we think is important
08:22:57FromDiscord<Varriount> You can always bookmark a filter by last updated date.
08:23:20FromDiscord<Clyybber> Its quite annoying (the spam) and the timing is very unfortunate too since we closed 124 issues this month
08:23:22narimiran"You can always bookmark a filter by last updated date." - you can always browse "stale" closed issues :P
08:23:38FromDiscord<Clyybber> Then why close them?
08:23:46FromDiscord<Clyybber> Its just to look fancy on the issue numbers
08:23:51FromDiscord<Varriount> Yes, but a user cannot open stale closed issues
08:24:03alehander92it's a tooling problem
08:24:10superbiaso essentially you are not fixing shithub and introducing new issues
08:24:14alehander92people like having nice backlog/history
08:24:25superbiamove to gitlab, problem solved
08:24:28alehander92but it's not the default behavior
08:24:31narimiran@Clyybber yeah, we want fancy 3-digit number of issues :)
08:24:52FromDiscord<Varriount> So it's about optics then.
08:25:07FromDiscord<Clyybber> optics don't help me fix and search for bugs tbh
08:26:02FromDiscord<Varriount> And this problem was so urgent and clear that it didn't merit an easily viewable public discussion?
08:26:22*Trustable joined #nim
08:26:38narimiran@Varriount let's be honest - would any discussion change your mind?
08:26:49narimiran*opinion
08:27:16FromDiscord<Clyybber> maybe it would change yours :)
08:27:21narimiran:)
08:27:28FromDiscord<Varriount> My mind? Unlikely, but possible, if I'd had time to actually read over the arguments.
08:27:38FromDiscord<Clyybber> in the end it should be about making Nim better
08:27:52superbialess issues === better product
08:27:56FromDiscord<Clyybber> and its easier for us to make nim better if we can look at the issues and not die in spam
08:28:00FromDiscord<Clyybber> (edit) 'die' => 'drown'
08:28:37narimiranthe bot is limited at 30 issues per hour, we can tune it down a bit, if that would help to be perceived as less spammy
08:28:38FromDiscord<Clyybber> also it *feels* better to provide a fix for the issue. And that feeling is eradicated by a bot that will just close it automatically anyways
08:28:57PMunchPrestige, hmm what do you mean?
08:29:03FromDiscord<Clyybber> And after all humans act on feelings
08:29:17PMunchYou need to restart the editor to edit multiple files?
08:30:11Prestigewell just the language client vim is using, yeah
08:30:17FromDiscord<Clyybber> narimiran: Also its not like we don't have people that go through old issues. Yardanico, xflywind and sometimes me go through old issues
08:30:26narimiranme too
08:30:37narimiranand it gets very tiresome, at least for me
08:30:37FromDiscord<Clyybber> Does it bother you that much to do that?
08:30:48FromDiscord<Clyybber> I think Yardanico, flywind and I enjoy it
08:31:01FromDiscord<Clyybber> Lets ask @Yardanico
08:31:06PrestigeFor instance if I make a field on a type public in another file, it needs a restart to see that change PMunch
08:31:13narimiranand make no mistake, 6 year old issue that has had some activity in last year and a half won't receive any bot message
08:31:35narimiranit is not closing *old* issues, just *inactive* ones
08:32:09FromDiscord<Clyybber> Yes but inactive doesn't mean invalid.
08:32:10narimiran"but a user cannot open stale closed issues" -- if the original author still cares about the issue, all they need to do is say something, and it will remain open for at least another year and a half
08:32:22FromDiscord<Clyybber> Its just so much spam
08:32:23narimiraninactive means nobody cares enough
08:32:26FromDiscord<Clyybber> And effort
08:32:43FromDiscord<Clyybber> I have no idea what issues are actually new in my inbox
08:32:52FromDiscord<Varriount> That assumes the author actually looks at their inbox
08:32:57FromDiscord<Clyybber> And then I get the spam twice
08:33:07PMunchPrestige, hmm, that's not good
08:33:09FromDiscord<Clyybber> Once for the staling, and once for the author waking it up
08:33:19FromDiscord<Elegant Beef> What about abusing the github projects, and just adding the issue to a list of "Stale issues"
08:33:28FromDiscord<exelotl> People aren't gonna feel like manually commenting on each issue to stop some bot, so valid issues will get closed
08:33:35FromDiscord<Clyybber> What about simply not caring about the fucking issue number
08:33:40narimiran@Clyybber maybe this way you'll find some interesting issue you know how to fix :)
08:33:42FromDiscord<Clyybber> sorry for the swearing
08:34:01FromDiscord<Elegant Beef> Yea clyybber my solution was more about "If this is needed, why not do something less obtrusive"
08:34:02FromDiscord<Clyybber> narimiran: Fair, but I can find these without the bot
08:34:03PMunchPrestige, keep in mind that it only does the error checking when you save a file though
08:34:17FromDiscord<Clyybber> By simply sorting through least recently commented
08:34:21FromDiscord<Clyybber> (edit) 'through' => 'by'
08:34:37PrestigePMunch: yeah, even if I save it doesn't change behavior
08:34:39PMunchSo you would have to first edit the file with the type, and probably save it. And then save the file that uses it to trigger the checking
08:35:38narimiran@Clyybber trust me, i've done that (sorting through least recently commented) numerous times, found plenty of "already fixed" issues (you know about it, you created PRs with tests for some of those)
08:35:58PrestigePMunch: perhaps I should make a vid to show you?
08:36:13PMunchPlease do :)
08:36:51FromDiscord<Clyybber> narimiran: Yeah exactly. Whats the benefit about spamming me with them?
08:37:02FromDiscord<Clyybber> we can go through them when at our own pace
08:37:32FromDiscord<Elegant Beef> I dont recall does adding a tag spam anyone?
08:37:54FromDiscord<exelotl> No but the bot leaving a comment does
08:38:01narimiranat some point (at this rate, probably next week), the spam should stop
08:38:15FromDiscord<Elegant Beef> So why not just mark it stale then issues are left open, and they can be filtered out
08:38:33solitudesfthis is awful
08:38:34FromDiscord<Clyybber> next fucking week
08:38:39FromDiscord<Clyybber> this is going to take a week?
08:38:45FromDiscord<Ricky Spanish> anyone know if the "nim in action" book is reasonably up-to-date? i purchased it already since prefer to read from a physical book but im worried its potentially out of date for some of the stuff
08:38:46FromDiscord<Clyybber> and every fucking hour I get 30 new issues?
08:38:49narimiranor more if we slow it down :D
08:39:03FromDiscord<Elegant Beef> This removes spam, solves closing issues, and results in the same end goal as far as i can see
08:39:09*dadada joined #nim
08:39:25narimiranok, i can move the limit at 2.5 years, should be much less issues affected
08:39:34*dadada is now known as Guest53595
08:40:12FromDiscord<Clyybber> I think beefs proposal is a fine compromise
08:40:46FromDiscord<exelotl> @Ricky Spanish yeah its still very relevant! The main difference is that seqs and strings can't be `nil` anymore since Nim 0.19
08:40:59FromDiscord<Elegant Beef> Doesnt dom have revisions for his book?
08:41:09alehander92narimiran i like the stale idea
08:41:15alehander92but the mail thing will lead to a revolt
08:41:27narimiranok, we can make stalebot post no message
08:41:29superbiaget Nim in Action: Book errata for Nim v1.0.0 | Deepak's Tech Blog
08:41:37narimirani'll do that
08:41:40FromDiscord<Ricky Spanish> awesome thanks for explaining @exelotl
08:41:48*Vladar joined #nim
08:41:53alehander92if it's possible to do it without a mail i think this would anger less people
08:42:22narimiranbut that way the author won't see that it has been labeled, and then they'll be surprised when after X days it is closed
08:42:25FromDiscord<Elegant Beef> Well i suggested the solution that would be non intrusive
08:42:38narimiran(and yes, the idea is to close stale issues, not just label them)
08:42:51FromDiscord<Clyybber> WHY?
08:42:55FromDiscord<Clyybber> There is not point
08:43:05FromDiscord<Clyybber> aside from bagging about: hurr durr we have so few issues
08:43:10FromDiscord<Clyybber> look what a great lang we are
08:43:13FromDiscord<Elegant Beef> You can filter them out after they're labelled
08:43:13FromDiscord<Clyybber> its cringy
08:43:22FromDiscord<Elegant Beef> So you get the same benefit as if they're closed
08:43:27narimiranyes, in the end it is about marketing and looking good
08:43:51FromDiscord<Clyybber> Well it doesn't look good to have hundreds of wontfix closed issues
08:43:52FromGitter<bung87> oh I see the benefits..
08:44:03FromDiscord<Clyybber> The benefit is looking good
08:44:05PMunchBut what if the issues are still valid?
08:44:08narimiranbut the main reason is limited manpower, as i already said earlier
08:44:11FromDiscord<Clyybber> The downside is driving away developers
08:44:18PMunchA bug doesn't go away simply because no-one has talked about it for a while..
08:44:35FromDiscord<Elegant Beef> I mean that's what we thought happened to your pmunch πŸ˜›
08:44:39FromDiscord<Elegant Beef> (edit) 'your' => 'you'
08:44:43FromDiscord<exelotl> I don't think its a good idea, valid issues will get closed
08:44:53FromDiscord<Clyybber> Yeah
08:44:57narimiranPMunch: if it only affected one person at some point in 2016, then probably isn't as important as something newer affecting our *current* users
08:45:11FromDiscord<Elegant Beef> I think what i suggested hits all the bases that everyone wants, aside from closing issues
08:45:29FromDiscord<Clyybber> closing issues is just marketing
08:45:32FromDiscord<Clyybber> and bad one at that
08:46:34PMunchWell it can still affect current users..
08:46:58PMunchIt's not like people comment on an issue saying "this affects me as well" every time they encounter them
08:47:03PMunchThat would just be super annoying
08:47:13narimiranat what point a stale label is better than searching with issues with `is:issue is:open updated:<=2018-01-01`?
08:47:28*waleee-cl joined #nim
08:47:31superbiadoes github have issue Weight ?
08:47:53FromDiscord<Clyybber> narimiran: On a emotional note, its really discouraging to know that all these issues will get closed anyways and that a fix isn't really going to publicly change anything for Nim
08:48:04FromDiscord<Elegant Beef> I guess no better just a alternative
08:48:08FromDiscord<Clyybber> It really hurts the motivation to work on fixes
08:48:23PMunchTake this for example, definitely still an issue, but hasn't even been addressed.. https://github.com/nim-lang/Nim/issues/10193
08:48:27disbotβž₯ Break in template called from loop creates strange results ; snippet at 12https://play.nim-lang.org/#ix=2taI
08:48:32FromDiscord<Clyybber> Its just feelings, irrational, but perhaps other devs feel so too
08:48:34narimiranno more messages from the bot, keeping the spam down
08:49:11PMunchSo now issues will just be closed without warning?
08:49:16superbiaCorrect
08:49:17PMunchNot sure if that is better..
08:49:22dom96More spam!
08:49:29narimiranwe'll reconsider once 4raq is back from holidays
08:50:13dom96I'm 100% in favour of this, it'll just be spammy at first AFAIK
08:50:14FromDiscord<Clyybber> how long does an issue have to be stale to be closed?
08:50:27narimirannow it is 30 days
08:50:35PMunch30 days?
08:50:42FromDiscord<Varriount> Seriously?
08:50:52PMunchThat's just ridiculous
08:50:58FromDiscord<Elegant Beef> I believe they mean when marked stale
08:51:00narimiranwait
08:51:12narimiranit is closed 30 days after marked stalled
08:51:18FromDiscord<Elegant Beef> Based off their previous comment on the label, it seems ~2 years
08:51:23PMunchI can see the argument for a bug affecting one person back in 2016 being marked as stale
08:51:28narimiranit is marked stalled if there was no activity for 550 days, which is ~2.5 years
08:51:47PMunchI have some that have been updated early 2019 at least that's been marked as stale..
08:52:02narimiranoops, ~1.5 years
08:52:03PMunchThis one for example: https://github.com/nim-lang/Nim/issues/10193
08:52:04disbotβž₯ Break in template called from loop creates strange results ; snippet at 12https://play.nim-lang.org/#ix=2taI
08:52:20FromDiscord<Clyybber> The thing is this can't work. No way this will work. Either we get incredible spam which hurts every dev working on Nim or we get issues that are closed as stale because the author didn't even know he was supposed to keep it alive
08:52:20narimiran550/365 = 1.506
08:52:30FromDiscord<Clyybber> Its a net negative either way
08:53:09narimiranto limit spam, but still post a warning message, we can bump the limit to 1000 days, then slowly decrease it
08:53:18PMunchI mean the warning for stale issues is good, that kinda has to be there for this to have any effect at all
08:53:36FromDiscord<Varriount> I will be honest in that those numbers are more sensible than what other projects set
08:53:56dom96Amazed by the outrage on this
08:54:00FromDiscord<Varriount> So am less "against" this.
08:54:01FromDiscord<Clyybber> Can't we just get rid of it entirely. It really is failed automation IMO
08:54:09dom96But maybe I'd be just as outraged if I didn't see this in action working previously
08:54:15narimiranwe have 52 issues with no activity after Jan 1st 2017, 159 (211-52) after Jan 1st 2018, etc.
08:54:20dom96Look, if someone cares about their issue they will keep it active
08:54:23dom96and it won't get closed
08:54:27PMunchThe idea is to remove issues that don't apply any more I hope (closing bugs that are still bugs is just dumb), and marking them as stale and then closing them after a while is a good way to get the original author to have a look and decide if it should still be open or not.
08:54:36dom96beats narimiran going over all issues and asking "Is this still a problem?" manually
08:54:43PMunchProblem is that the original author might be gone from Nim and the issue is still valid but they don't care
08:54:52PMunchBut hopefully that's not too many
08:54:57dom96And there is certainly a label that we can use to mark as "Do Not Stale Close"
08:55:03dom96if not then there should be
08:55:04narimiranthen somebody else can go through "stale" label and see what is still valid and what is not
08:55:07FromDiscord<Varriount> @dom96: And I've felt incredibly frustrated when seeing it in action.
08:55:36FromDiscord<Clyybber> dom96: Its cool that you care all about "people caring about issues" and "number of open issues" but in the end most of these are technical issues, not "I feel like opening an issue just for fun"
08:55:54FromDiscord<Clyybber> And technical issues don't just vanish when the person that opened them disappears
08:55:57FromDiscord<Varriount> @dom96 Part of my "outrage" is that this appears to have been done solely at the discretion of two people, with no community input.
08:56:09FromDiscord<Clyybber> You are applying principles of politics to software development
08:56:30FromDiscord<Clyybber> As in, don't talk about it, then its not a problem.
08:56:35PMunchClyybber, well an issue can be fixed "by accident" or just being not applicable any more because of a major rewrite
08:56:41FromDiscord<Varriount> I do not like to find out about these changes at 4 in the freaking morning
08:56:42PMunchSo sometimes bugs just go away on their own
08:56:55narimiranguys, what about an idea to limit spam by increase stale-limit to some 1000+ days number?
08:57:02dom96Varriount: that's fair.
08:57:03FromDiscord<Clyybber> PMunch: Yeah, but even then its quite enjoyable to close these issues. And we have plenty of people that actually do that
08:57:11PMunchnarimiran, that's certainly better
08:57:16FromDiscord<Clyybber> And they do it out of fun (except for miran :P)
08:57:22narimiran:P :D
08:57:31Yardanicowell, this stale thing is really interesting, on one hand I see how it would help - only keep issues affecting current isuers and stuff, but on the other hand - we'll just lose a lot of valid issues and pretend *they're not there*. Even if an issue was fixed in devel and no one knew - there still needs to be a test for it so it doesn't break again
08:57:41Yardanicothe spam thing is annoying but I understand that it's temporary
08:57:46alehander92but those issues are there
08:57:49alehander92just closed
08:58:07PMunchIsn't this what milestones and sprints are supposed to address?
08:58:18dom96Yeah, if the issues have been closed then they likely weren't important enough
08:58:27dom96We have limited resources, this is how we prioritise them
08:58:29Yardanicowell then we'll just get more duplicates, a lot of people don't search for same issues when opening an issue on github
08:58:43PMunchTake the issues that affects people now, add them to a milestone to signal that they will be fixed in an upcoming release
08:58:47dom96Yardanico, same applies to open issues...
08:58:48FromDiscord<Clyybber> Exactly. we have been surfing well with the current approach. We even now started to close more issues than were opened (two to one ration this month!!)
08:59:14dom96Clyybber: well, apparently narimiran and Araq don't feel that to be the case.
08:59:44PMunchI'd say it's petter to mark some issues as a priority and possibly commit to fixing them rather than just closing issues that could still exist
08:59:51narimiranok, putting 1200 days as a limit, that's only ~70 issues in total.
09:00:01Yardanicowell, the day count isn't an issue for me
09:00:03narimiranso the spam will last for only 2-3 hours
09:00:04FromDiscord<Clyybber> dom96: So we should also close old issues without testcases now?
09:00:09PMunchWhat happens to those already marked as stale?
09:00:11FromDiscord<Clyybber> Get ready for regression fests
09:00:41dom96Clyybber: sure, if those issues only affect one person out of a million
09:00:48FromDiscord<Clyybber> But they don't.
09:00:50dom96Imagine how many issues have never been reported
09:00:57dom96that we don't get test cases for
09:01:25dom96This is just a forcing function which allows us to prioritise what really matters
09:01:27FromDiscord<Clyybber> Most issues go unreported.
09:01:31FromDiscord<Clyybber> If we get one we should value it
09:01:39FromDiscord<Clyybber> (edit) 'one' => 'a report'
09:01:54PrestigePMunch: https://0x0.st/i3Xq.mp4
09:02:01FromDiscord<Clyybber> dom96: No. Its a forcing function that keeps me motivated from doing anything at all
09:02:02dom96we should value the issues which are affecting many users
09:02:06dom96we don't have enough resources for that
09:02:27FromDiscord<Clyybber> So prioritise it. Github has a searching feature
09:02:37FromDiscord<Clyybber> .s/searching/sorting
09:02:53superbiaand in 10 years, github will have weights
09:02:54FromDiscord<Clyybber> But please leave the prioritising to us
09:03:01FromDiscord<Varriount> How do old issues take up resources?
09:03:02FromDiscord<Clyybber> And not to some machine that tells me what to do
09:03:16dom96The bot pinging old issues is also a good distributed way of checking whether those issues still apply
09:03:24narimirannotifications are back on, sorry to all affected; but it should be just a relatively small amount of issues (less than 100)
09:03:32FromDiscord<Clyybber> Contributors are not some cog in the machine that you can dicate what they should fix
09:03:40FromDiscord<Clyybber> They fix what they want
09:04:09narimiranexactly!
09:04:34FromDiscord<Clyybber> exactly. And I feel extremely discouraged from fixing anything rn..
09:04:38FromDiscord<Clyybber> Because it doesn't make a change
09:04:46FromDiscord<Clyybber> The issue will get closed anyways
09:04:54alehander92come on guys
09:04:55narimiranand if somebody new comes and wants to pick some random issue to fix, we would like to steer them to the more relevant issues
09:04:56FromDiscord<Clyybber> If its deemed unimportant by the masses
09:05:00alehander92closed / not closed
09:05:07alehander92it matters if it's fixed
09:05:13FromDiscord<Clyybber> narimiran: Well this steering is just annoying and demotivating
09:05:14narimiran"The issue will get closed anyways" -- now you're being overly dramatic
09:05:36alehander92i am not disagreeing, but on the other hand milestones might be a better
09:05:37alehander92fit
09:05:43narimiranpicking one of 1500 issues instead of 1600 is demotivating?
09:05:44FromDiscord<Varriount> dom96, narimiran: I'm off to bed. I've said my piece. Please don't do this again without getting some community input first, if only so those of us in different parts of the world can get to bed.
09:06:02FromDiscord<Clyybber> narimiran: I'm trying to give my subconcious words, it maybe doesn't make a whole lot of sense :p
09:06:35FromDiscord<Varriount> I believe issues/forum post are better for these discussions anyway.
09:06:35superbiawhy don't you suggest github implements OPEN STALE CLOSED ?
09:06:40dom96We should definitely make a forum thread about this and see what everyone thinks
09:07:01FromDiscord<Clyybber> Make an RFC so we can up/downvote
09:07:11*ForumUpdaterBot quit (Remote host closed the connection)
09:07:19narimirandom96: the problem is that people "on the outside" cannot see the full picture
09:07:20*ForumUpdaterBot joined #nim
09:07:20FromDiscord<Clyybber> (edit) 'Make' => 'Or'
09:07:26narimirani know that first hand
09:08:08narimirani had one set of ideas before employed, and then only afterwards i discovered why they were good only in theory and/or in ideal conditions, but not in practice
09:09:32dom96Sure, and you can make your case. Plus in the end it can be Araq's call. But I like to think that Nim is still a community-influenced project, it's worthwhile to see what the community has to say.
09:10:23dom96Varriount: did the emails wake you up? You should seriously consider muting notifications for those.
09:10:50FromDiscord<Varriount> No, but I found out as I was getting ready for bed
09:10:54PMunchPrestige, hmm that's weird
09:11:24Yardanicosuperbia: we can all argue that github is bad, etc, and gitlab is better - but at the end of the day people also use github because of it's "social" features - discovering projects, etc
09:11:26FromDiscord<Varriount> And because it's apparent this is all being discussed *now*, I have to participate *now*
09:11:32Yardanicogitlab is much less advanced in that regard
09:11:49Yardanicoalso github is miles more popular, believe it or not :)
09:11:52superbiayou really must be full of first hand experience to say stuff like this 2020-08-04 10:43:28 narimiran yes, in the end it is about marketing and looking good
09:12:01Yardanicoso people are more likely to contribute if the project is on github
09:12:13PrestigeI have a C api giving me a pointer to a long, which is supposed to be an array of longs (pointer being at the 0 index). How should I traverse this, cast the variable to an array?
09:12:16PMunchStrange Prestige, that works for me with the exact same files..
09:12:32YardanicoPrestige: yeah, cast it to ptr UncheckedArray[clong]
09:12:32FromDiscord<Varriount> When you're employed, you also have to consider the opinions of others on your team and compromise.
09:12:32PMunch(The LSP issue, not the C API thing)
09:12:56Yardanicolet arr = cast[ptr UncheckedArray[int32]](myptr)
09:13:08superbiathere was also a lot of people like myself who quit github few years ago and transfered to gitlab
09:13:14Yardanicosuperbia: not a lot really
09:13:19Yardanicoit was just "overhyped"
09:13:32Yardanicogithub is really much, much, much more popular than gitlab
09:13:37PrestigeThanks Yardanico
09:13:39Yardanicoit has a lot of it's own issues, yes
09:13:45YardanicoPrestige: the C API should also give you the length
09:13:46PrestigePMunch: neovim or what?
09:14:23superbiaYardanico: so why doesn't GITHUB has it's own internal way of managing stale issues?
09:14:40Yardanicothe bot can be considered as an "internal" way
09:15:07YardanicoJust wait till they make their own bot for that :)
09:15:17Prestigebrb
09:15:20*Prestige quit (Quit: Prestige)
09:15:23Yardanicoas I said, I don't think github is the best out here, but it's really the most popular
09:15:31*Prestige joined #nim
09:15:48dom96Very often the most popular != the best
09:15:55superbiaok, I don't think it's even possible to change mind of someone who already put this bloody thing into production
09:16:14Yardanicosuperbia: if Nim was on gitlab it'd receive less contributions really
09:16:26PMunchPrestige, regular vim
09:16:37Yardanicoespecially the more-or-less simple contributions from not relatively experienced people
09:16:39Prestigeusing Coc?
09:16:40Yardanicolike some doc fixes, etc
09:16:46dom96superbia, that's a really bad attitude to have
09:16:56PMunchPrestige, Coc?
09:17:05Yardanicohttps://github.com/neoclide/coc.nvim maybe?
09:17:58Prestigeyeah - which lsp are you using PMunch ?
09:18:05PMunchPrestige, vimlsp
09:18:16PMunchThis one: https://github.com/prabirshrestha/vim-lsp
09:18:26Prestigeinteresting
09:18:50superbiadom96: I am going to have goulash for lunch, so maybe that will change my mind ... (I never start working on something before I ask someone competent what do they think about it)
09:18:58PMunchI still get a warning on the same line that f is unused though
09:19:08Prestigeprobably a communication issue with coc.nvim again PMunch, maybe we can inspect and see what's going on
09:19:40PMunchYeah if you compile with debug information it outputs a pretty verbose log
09:20:23Yardanico PMunch there's an issue I'm having with vscode-nim-lsp + nimlsp (not sure if caused by nimlsp) - e.g. if I install a nimble package, I have to fully reload/restart nimlsp so I won't get import errors about the module not existing
09:21:11Prestigesounds like a similar issue
09:21:13PMunchHmm, that might be a nimsuggest issue..
09:22:31FromDiscord<Clyybber> dom96: Can you review or merge https://github.com/nim-lang/Nim/pull/15155 ?
09:22:31disbotβž₯ minor change in selectors
09:22:43FromDiscord<Clyybber> I think it just fixes a typo right?
09:23:19PrestigePMunch: I'll try that out soon and create an issue if I find the culprit
09:28:48FromDiscord<flywind> @Clyybber Because it uses templates, so `data` works fine. But I think `pdata` is more clear.
09:29:32FromDiscord<flywind> sent a code paste, see https://play.nim-lang.org/#ix=2taY
09:29:48FromDiscord<Clyybber> Yeah, agreed
09:30:24*letto quit (Read error: Connection reset by peer)
09:31:04*letto joined #nim
09:41:08PrestigePMunch: Well I pulled, built debug, and now I'm not getting anything it seems: https://0x0.st/i3XS.txt
09:41:09FromDiscord<flywind> hi, @dom96, could you add a license for `httpbeast`. It only mentioned license name in nimble file. Thanks!
09:44:10Prestigemaybe an issue with the nimsuggest source?
09:55:53PrestigeAha, got it. PMunch near the very end it shows where Foo.a is made public/exported (textDocument/didChange), and right after in the other file it says 'a' is an undeclared field: https://0x0.st/i3XQ.txt
09:57:09silvernode[m]Well I'm off work tonight, so that should mean it's a good time to work on Nim projects, but after working 7 days straight, coding is a bit difficult. :P
09:58:51dom96flywind: will do tonight
09:59:25dom96silvernode[m], 7 days straight? wtf
09:59:37FromDiscord<flywind> Thank you!
10:00:38silvernode[m]<dom96 "flywind: will do tonight"> yep, y last night off was the 26th of July.
10:01:00silvernode[m]> <@freenode_dom96:matrix.org> flywind: will do tonight
10:01:00silvernode[m] * yep, my last night off was the 26th of July.
10:03:30silvernode[m]wrong reply reference, oh well
10:05:02dom96Busy period in your workplace or a company that doesn't respect work-life balance? :)
10:07:23silvernode[m]<dom96 "Busy period in your workplace or"> First off, it's a very fast paced night stocking job and management keeps hiring 70 year old people that can't handle the workload. Second, other new hires either stop showing up or not showing up at all, and third, the people who actually show up are getting sick or have to call out because a family member has died from COVID-19.
10:07:58silvernode[m]> <@freenode_dom96:matrix.org> Busy period in your workplace or a company that doesn't respect work-life balance? :)
10:07:58silvernode[m] * First off, it's a very fast paced night stocking job and management keeps hiring 70 year old people that can't handle the workload. Second, other new hires either stop showing up or don't show up at all, and third, the people who actually show up are getting sick or have to call out because a family member has died from COVID-19.
10:08:53dom96oh :/
10:09:11silvernode[m]Then there is me who has been there for 5 years and a guy who has been there for 20+ years along with a couple more core team members carrying the weight of the others.
10:10:05silvernode[m]We keep getting people who can't follow instructions or who can't do things fast enough. I was there for 12 hours the last two days each for the first time since I started there 5 years ago.
10:10:31silvernode[m]Needless to say my next check is going to be interesting.
10:12:37*nc-x joined #nim
10:13:04*NimBot joined #nim
10:15:41*Yardanico left #nim ("https://quassel-irc.org - Chat comfortably. Anywhere.")
10:15:42*Yardanico joined #nim
10:15:43*Yardanico left #nim ("https://quassel-irc.org - Chat comfortably. Anywhere.")
10:15:45*Yardanico joined #nim
10:18:34nc-xi am a bit late to the party but I don't think the stale bot is a good fit for Nim, because the number of devs are very low, so a lot of issues remain open for a long time with no new comments, and now the stale bot is going to close such issues if somebody doesn't keep bumping up the issues
10:19:13*nc-x quit (Remote host closed the connection)
10:21:28FromDiscord<tomck> Can i forward declare types?
10:21:37narimiranhttps://github.com/probot/stale#is-closing-stale-issues-really-a-good-idea
10:21:38FromDiscord<tomck> or have circular types outside of the same `type` block?
10:22:01Yardanicowell, there are "package-level types"
10:22:10Yardanicobut I think no one really uses them so idk if they work ;P
10:22:23Yardanicohttps://nim-lang.org/docs/manual_experimental.html#package-level-objects
10:22:24FromDiscord<tomck> i see, is that the only way to construct circular types?
10:22:31FromDiscord<tomck> ah, 'experimental'
10:22:36Yardanicousually the pattern is to create a types.nim file
10:22:40Yardanicoand put all your types in there :P
10:22:44Yardanicoif they depend on each other
10:24:00FromDiscord<tomck> hmmmmm interesting, is that standard?
10:24:10Yardanicoit's not "standard"
10:24:10FromDiscord<tomck> seems super messy, reminds me of c-land
10:24:31Yardanicoor just rework your code so you don't need circular types :)
10:24:38FromDiscord<Rika> its what we're dealing with for now
10:25:09FromDiscord<tomck> package-level types seem like a solution
10:25:16FromDiscord<tomck> or are they still brand-spanking-new?
10:25:27Yardanicono, they've been there for quite a while
10:25:30Yardanicomore than a year I think
10:25:34Yardanicobut I didn't see anyone use them it :D
10:25:36FromDiscord<Rika> so pretty new
10:27:23FromDiscord<tomck> how do packages even work? what package is my code in by default? AFAIK i haven't specified a package name anywhere
10:28:08Yardanicowell it says that it's a nimble package
10:28:21FromDiscord<tomck> how do i reference 'the current package'
10:28:35FromDiscord<tomck> i'm in the same package, i just want to have circular types across modules
10:28:42YardanicoI've never used that feature myself :)
10:28:42FromDiscord<flywind> When we fix stale closed issue, can the stale label be removed automatically just like `fix issueNumber`?
10:28:51FromDiscord<tomck> fair, ty
10:28:56Yardanicowill check if there are tests for it
10:47:14PMunchPrestige, that is very weird indeed
10:47:35PMunchHmm, let me build a new fresh copy of nimlsp
10:47:36FromDiscord<Zed> hey @dom96 is your opencv wrapper stable? is the wrapper basically the same as the python one?
10:48:11dom96haven't used it in many years, so no idea :)
10:48:14PrestigePMunch: if you need more info or want me to create an issue lmk, I'll get to it after breakfast
10:48:33FromDiscord<Zed> ok, thanks
11:00:20PMunchOh wait..
11:00:52PMunchHmm, I think the Nim plugin I have in Vim is the one giving that error message..
11:01:02PMunchNow I can't see that unused message at all..
11:01:40PMunchZah isn't on is he?
11:06:05TrustableCan someone please close this issue: https://github.com/nim-lang/Nim/issues/5765
11:06:06disbotβž₯ setFilePos and getFilePos do not support files larger than 2GB on Windows
11:08:23FromDiscord<Clyybber> Done, thanks!
11:13:25leorize[m]@flywind nope, it's a known issue: https://github.com/probot/stale/issues/255
11:13:25disbotβž₯ do not mark issues that are being cross referenced
11:15:13*maier quit (Ping timeout: 264 seconds)
11:28:54*maier joined #nim
11:32:41PMunchPrestige, disabled my other Nim plugin and now I get the same issue..
11:36:19PrestigePMunch: ah, at least you were able to reproduce it
11:36:45PMunchIt's really weird though, with the other plug-in the LSP log looked just fine..
11:36:59PrestigeThat is weird
11:37:32PMunchVery..
11:37:50PMunchIt almost seems like the nim.vim plug-in does something that makes Vim behave differently..
11:40:43Oddmongerwhat is the difference between debugging with gdb and nim-gdb ? Some macros for gdb in nim-gdb ?
11:41:19Oddmonger(i've read the entry about debugging , butit didn't mention nim-gdb)
11:41:29leorizeit makes gdb aware of Nim types
11:42:09Oddmongeri've read this: https://nim-lang.org/blog/2017/10/02/documenting-profiling-and-debugging-nim-code.html
11:42:14leorizeyou can see enum values, seqs, sets
11:42:28Yardanicomost of it only works for refc though
11:42:30Yardanicobecause of RTTI
11:44:14Oddmongerah interesting to see native nim values without pointer contortions
11:50:38PMunchPrestige, could you create an issue on GitHub for this?
11:50:47*krux02 joined #nim
11:50:55PrestigeSure thing
11:51:04PMunchHmm, I wonder if this issue also happens with nimsuggest
12:06:02*supakeen quit (Quit: WeeChat 2.8)
12:06:38*supakeen joined #nim
12:07:42FromDiscord<Clyybber> narimiran: Can we maybe increase the limit to infinity for a short period of time? I think its better if it spams anyways that we let it spam once.
12:10:34*audiofile joined #nim
12:13:47*rockcavera joined #nim
12:27:35PMunchHuh, discovered some weird behaviour possibly related to try/except
12:28:11PMunchThis works fine: https://play.nim-lang.org/
12:28:21PMunchWoops: https://play.nim-lang.org/#ix=2tbS
12:29:07PMunch100 gets pushed to the stack, // consumes the 100, 200 gets consumed by //, and then // is finalised by the 20
12:29:45PMunchBut if I move the `var command = new Command` outside the try block: https://play.nim-lang.org/#ix=2tbT
12:30:09PMunchI get the same final result, but the Command object appears to be copied (and the elems list isn't updated correctly)
12:30:22*endragor quit (Remote host closed the connection)
12:30:50*endragor joined #nim
12:31:35FromDiscord<Clyybber> huh
12:31:39FromDiscord<Clyybber> interesting
12:32:53FromDiscord<Clyybber> is this already minimized?
12:34:29PMunchWell it is slightly more minimal than what I have
12:34:39PMunchWhat I have the iterator is chosen based on the parsing of an enum
12:34:46PMunchAnd it's all in a loop with a prompt
12:34:52PMunchSo it works as a calculator
12:36:29FromDiscord<flywind> leorize, thanks.
12:36:35*endragor quit (Ping timeout: 240 seconds)
12:47:41FromDiscord<Clyybber> PMunch: Does it show this behaviour with gc:arc/exceptions:goto too?
12:48:52PMunchHmm, gc:arc just crashes before I get that far
12:49:46FromDiscord<Clyybber> heh :D
12:49:51FromDiscord<Clyybber> latest devel?
12:49:55PMunchLatest stable
12:49:58PMunchJust a sec
12:50:26PMunchYup, latest devel as well
12:51:08Yardanicothe compiler crashes
12:51:13Yardaniconot runtime :P
12:52:23*vsantana joined #nim
12:54:10PMunchHuh, the actual project compiles
12:54:14PMunchBut crashes on runtime
12:54:44PMunchThis thing: http://ix.io/2tbZ
13:00:35narimiran@Clyybber "Can we maybe increase the limit to infinity" - i don't understand
13:01:02FromDiscord<Clyybber> So that we get all the spam at once and not spread out over a few hours
13:01:24FromDiscord<Clyybber> but maybe its already finished
13:02:27narimiranah, that limit. from what i understand, 30 per hour is hard max. ours is now set up to 20/h
13:02:43narimiranand yes, i think the main spam period is now over
13:04:12FromDiscord<Clyybber> ah ok
13:06:34shashlickI somehow feel leaving issues sitting around is less work than using this bot both for maintainers and users
13:06:52*nurse_oda_purse joined #nim
13:10:41FromDiscord<Clyybber> yeah, I feel the same way
13:15:36shashlickSomehow, issues with valid snippets should be given more time to live
13:16:13shashlickI get the sentiment, just that as an issue opener, I don't see what I'm supposed to do
13:16:35narimiranmore than 3 years of inactivity? but ok, we should make another bot which would go through the old/neglected issues and just post a comment "this has a code snippet", and that would do the trick
13:16:56shashlickTissue already can do all that
13:16:59leorizeor we can just not use the bot :P
13:17:41narimiranand there i thought nothing could be hated more than 'fusion' :D
13:18:24shashlickEveryone's new favorite word: bump
13:21:29superbiabut we already did all of the work on the bot... would sux if we dump the bot whilst spending so much time working on it
13:21:38*endragor joined #nim
13:21:40*endragor quit (Read error: Connection reset by peer)
13:21:51*endragor joined #nim
13:22:14FromDiscord<Clyybber> if its for the better than it doesn't matter how much time was already spent on it
13:22:16Prestigesunken cost fallacy
13:23:02PMunchCould be a lesson for next time, don't spend a bunch of time on things no-one wanted in the first place
13:24:55audiofileI can define functions in nim below callee right?
13:24:59audiofilewhat is this 'feature' called
13:25:02Yardanicoforward declaration
13:25:08audiofilethanks!
13:25:21FromDiscord<Rika> you still have to declare the signature above callee
13:25:39Yardanicohence forward declaration :)
13:25:48audiofileah okay, that makes sense
13:26:56FromGitter<iffy> This search might be enough to find "stale" issues, if it's helpful for anyone: https://github.com/nim-lang/Nim/issues?q=is%3Aissue+is%3Aopen+-linked%3Apr+updated%3A%3C%3D2017-08-01+
13:27:27audiofilea bit iffy :P'
13:28:36FromDiscord<Clyybber> narimiran: Look what this leads to https://github.com/nim-lang/Nim/issues/4057 :(
13:28:38disbotβž₯ Windows. Program crashes when control-c if compiled with --threads:on ; snippet at 12https://play.nim-lang.org/#ix=2tcq
13:30:45FromDiscord<Clyybber> Its still an issue though.
13:35:10PMunch@Clybber, more minimal example: https://play.nim-lang.org/#ix=2tct
13:35:25PMunchMove line 19 to inside the except at line 23 and it works as expected
13:37:06narimiran"Nim version: Nim Compiler Version 0.13.0"
13:37:29narimiranyes, it is still an issue, but look at for how many versions that has not been fixed
13:37:41FromDiscord<Clyybber> So what???
13:37:50leorizeyour point?
13:38:31leorizenew contributors are not gonna look around in the "closed" section to find issues to fix
13:38:53*arecacea1 quit (Remote host closed the connection)
13:39:02FromDiscord<Clyybber> exactly
13:39:06narimiranmy point: it was not important enough to block the release any of the previous 9 (!) versions, it is not important enough in the near future either.
13:39:06PMunchJust because it hasn't been fixed yet doesn't mean it should never be fixed..
13:39:22FromDiscord<Clyybber> narimiran: But thats not true. It might be very important in the future
13:39:23PMunchThat doesn't mean it should never be fixed
13:39:33FromDiscord<Clyybber> for example when threads:on will be the default or w/e
13:39:37narimirannear future = still the same amount of manpower
13:39:44leorizeso the rationale is: it doesn't look important now so it doesn't matter?
13:39:47PMunchIt might also prove to be the root cause of some other issue
13:39:48*arecacea1 joined #nim
13:39:58FromDiscord<Clyybber> but having that issue open doesn't reduce the amount of manpower magically
13:40:18leorizeclosing issues doesn't make the compiler/stdlib magically better
13:40:25leorizein fact it does nothing
13:40:27FromDiscord<Clyybber> yeah
13:40:30FromDiscord<Clyybber> its just marketing
13:40:34FromDiscord<Clyybber> and pretty bad one at that
13:40:37PMunchApart from hiding issues..
13:40:38narimiranno, but having some X amount of manpower (including new contributors), it is better spend elsewhere
13:40:45FromDiscord<Clyybber> as can be seen by the reaction from the issue author
13:40:56FromDiscord<Clyybber> narimiran: But its not a good heuristic
13:40:58PMunchSo mark important issues as important
13:41:01leorizenarimiran: isn't that what the "High Priority" tag designed for?
13:41:22narimiranit isn't the first time that some author has closed their issue, and then somebody else opened it, right?
13:41:33PMunchCould introduce new priority tags, and assign issues to the person actually working on it to avoid duplicated efforts
13:41:54leorizewe have a fair share of assigned issues and no progress :P
13:42:13FromDiscord<Clyybber> some things just take time
13:42:21narimiranit is not like the currently opened issues are *all* the issues we have
13:42:22FromDiscord<Clyybber> and its not like we have too many issues
13:42:33FromDiscord<Clyybber> look at the monthly insight on github
13:42:40FromDiscord<Clyybber> we have closed over 120 issues this month
13:42:44narimiranlook at yearly (if there was one)
13:42:49FromDiscord<Clyybber> with only < 50 new opened ones
13:42:56PMunchnarimiran, no, but that shouldn't be necessary. Just look at the person who left the community right before I went on vacation. Everyone told them to keep their issues open, to the point that they were copied by someone else since the original author didn't want to receive notifications
13:42:56narimiranyou would see a bump from 1.2k to 1.7k
13:43:11FromDiscord<Clyybber> Yes, but its not hindering development
13:43:15leorizethe thing with this bot is: it doesn't fix anything
13:43:15FromDiscord<Clyybber> In fact its good for it
13:43:21solitudesfwhat is even the point of the bot, all the issues after 3rd page are stale, why the fuck do we need this bullshit
13:43:47leorizethere weren't any issue with the current system to begin with, and the bot doesn't improve anything
13:44:12leorizeagain, I think this sort of thing warrants some discusion before being introduced
13:44:27narimiranalright alright alright, i'll talk with 4raq about the bot. we had no idea it would be so universally hated
13:44:28solitudesfits literally noise generator for maintainers that think their project is not alive enough
13:45:32narimiranin the mean time - who will be working on fixing all those old bugs? :P :D
13:45:34leorizethe bot is even hated on it's own issue tracker :P I wouldn't take many bets on it won't be hated here
13:45:44FromDiscord<Zachary Carter> fuck robots man, time to start the war vs them
13:45:54leorizewhoever that find those interesting?
13:46:07leorizeor, we can just hire more people to work on those
13:46:17narimiranwhere do we get money from?
13:46:40PMunchHmm, I should really contribute some money to Nim tbh
13:47:07PMunchI've been meaning to do it for the longest time..
13:47:12leorizestatus donated a ton to the project, are we using that money?
13:47:16FromDiscord<Zachary Carter> I used to but then I stopped
13:47:22FromDiscord<Zachary Carter> I used to do like $150 USD / month
13:47:29FromDiscord<Zachary Carter> did that for over a year
13:48:00narimiranleorize: of course
13:48:08PMunchI mean, if I pay for free entertainment through Patreon I can surely spit some money into Nim which I use almost every single day
13:48:21PMunchOh wow, that's quite a lot Zachary
13:48:22FromDiscord<Zachary Carter> well the dontation stuff I feel like got weird
13:48:28PMunchGot weird?
13:48:31FromDiscord<Zachary Carter> like first it was via this one website
13:48:54PMunchMaybe we should do some kind of fundraiser
13:49:00FromDiscord<Zachary Carter> and then it changed I feel like
13:49:06Yardanicomaybe at least some money from https://opencollective.com/nim should be used for bounties?
13:49:33PMunchI mean most big languages have a huge sponsor, since we don't have one we should potentially be more aggressive about raising money
13:50:07narimiranYardanico: so, what do we pay, $1 per closed issue? :D
13:50:19Yardanicono, I mean in general
13:50:25leorizeI think 4raq said zah should be responsible for setting up bounties
13:50:43narimiranok, but let's talk numbers
13:50:58leorizebut his involvement with the project seems to have been overshadowed by his responsibility for status' projects
13:51:25narimiranlet's say we want one person dealing exclusively with issues, doing that 8 hours per day. questions:
13:51:49narimiran1. how much money do you think we need for that?
13:51:57narimiran2. how long until that person burns out?
13:52:03FromDiscord<Clyybber> 3. Do we need that?
13:52:22leorizebounty is a good alternative :P
13:52:29FromDiscord<Clyybber> I think its a pretty self regulating system
13:52:45narimirananswer 3 - if we don't want issues to be fixed, why does it then matter if they're open or closed? :P
13:52:48FromDiscord<Clyybber> Or are you not referring to dealing with issue bureacracy?
13:53:09*endragor quit (Remote host closed the connection)
13:53:12FromDiscord<Clyybber> Oh, I thought you were talking about issue bureaucracy stuff
13:53:24*endragor joined #nim
13:53:27FromDiscord<Clyybber> The thing about keeping them open is
13:53:33FromDiscord<Clyybber> it indicates that it *should* be fixed
13:53:48FromDiscord<Clyybber> no matter if it actually will or will not be in the near future
13:53:52FromDiscord<Clyybber> and thats the point
13:53:59FromDiscord<Clyybber> issues aren't merely a TODO list
13:54:15PMunchExactly
13:54:29leorizewe can start adding bounties to encourage developers to join in :)
13:54:32PMunchI can see the need for a TODO list, but it shouldn't be the issue tracker
13:54:41*endragor quit (Client Quit)
13:55:23supakeenAren't GitHub projects/milestones nice for that?
13:55:25leorizegithub have the "projects" and "milestone" thingy for that
13:55:56PMunchExactly
13:56:51PMunchHmm, this system of having a closure iterator modifying the state of an object seems pretty brittle..
13:57:07FromDiscord<Clyybber> I think your snippet is ripe for an issue :D
13:57:16*endragor joined #nim
13:57:33PMunchI'm trying to boil it down, but I can't really tell what the actual issue is
13:57:52PMunchI'm guessing the ref object gets deepCopied somewhere, but I'm not sure where and why
13:58:00leorizePMunch: probably loop scoping
13:58:14leorizetry to wrap the entire block in `closureScope:`
13:58:55PMunchThat worked
13:59:15PMunchWhat was it closureScope did again?
13:59:24Yardanicowraps the code in a proc
13:59:28Yardanicoto capture variables
13:59:34Yardanicoso yeah, about issues - can't we just label them as "stale" without closing?
13:59:36leorize`proc anon() = body; anon()`
14:00:34PMunchYardanico, then you're back to everything beyond page 3 is stale
14:00:41PMunchOr the same kind of thing
14:00:48PMunchleorize, right..
14:00:51Yardanicothen just don't do anything and keep the current state of things :P
14:01:08PMunchExactly!
14:01:09leorizePMunch: variables in a for-loop is re-used for each iterations
14:01:33leorizeclosureScope() put them in a closure to break that
14:01:43leorizefor more selective scoping: sugar.capture
14:02:33PMunchI see
14:02:39PMunchOh well, got to go
14:02:40*PMunch quit (Quit: Leaving)
14:02:46*endragor quit (Quit: Leaving...)
14:03:59FromDiscord<Clyybber> damn, the spam keeps on giving :p
14:04:47narimiranheh, i thought 20/hour meant 1 every 3 minutes, but no, it does 20 of them then waits an hour, then 20 of them, etc.
14:05:05FromDiscord<Clyybber> better than 1 every 3 minutes tbh :p
14:05:16FromDiscord<Clyybber> because then at least I can mark all of them as read
14:05:24narimiranyay, the bot is only 99% bad
14:05:27FromDiscord<Clyybber> which is not the point of the bot, but thats just how it goes
14:05:37FromDiscord<Clyybber> narimiran: add another .9% :p
14:05:38*endragor joined #nim
14:05:47narimiranneveeerrr
14:06:08leorizeI mean the bot is even hated on it's own issue tracker :P
14:06:10Yardanicoi just honestly don't think it would help with any kind of marketing at all
14:06:20Yardanicoissue count doesn't say anything
14:09:45FromDiscord<Rika> > damn, the spam keeps on giving :p↡@Clyybber what spam?
14:09:50FromDiscord<Rika> what's happening?
14:10:17narimiranjust don't click "watch" on nim's repo and you'll be fine
14:10:29FromDiscord<Clyybber> well yeah, but I wan't to watch nims repo normally
14:10:34FromDiscord<Clyybber> @Rika github notifications
14:10:39leorizewe added a new bot to our repo, and it's uh, not really nice
14:10:42FromDiscord<Rika> i dont see anything weird on the repo
14:10:53Yardanico@Rika because you're not subscribed to notifications :)
14:10:58FromDiscord<Clyybber> @Rika a ton of issues get commented on by a shitty bot
14:11:16Yardanicohttps://github.com/nim-lang/Nim/issues/10204#issuecomment-668435029
14:11:18disbotβž₯ `nkStmtListExpr` with typ == nil for default parameter ; snippet at 12https://play.nim-lang.org/#ix=2tcK
14:11:34FromDiscord<Rika> Yardanico: and you are not (were not) explaining why
14:11:34FromDiscord<Clyybber> Yardanico: Heh, I already looked at that one
14:11:36narimiranwait wait wiat
14:11:40FromDiscord<Clyybber> its not fixed
14:11:44narimiranwhy does the bot comment on that one???
14:11:49Yardanicohahahaha
14:11:56narimiranit is not 3 years old!! wtf, bot?
14:12:01Yardanicoit commented 7 hours ago
14:12:02FromDiscord<Clyybber> because the bot is fucking stupid
14:12:16narimiranYardanico: ah, so it was before the change
14:12:18FromDiscord<Clyybber> probably was the half a year setting
14:12:42Yardanicojust set it to 1 month :)
14:12:56narimiran1 day
14:13:02Yardanicodisable issues in the repo
14:13:10narimirandelete the repo
14:13:16Yardanicodelete the org
14:13:18FromDiscord<Clyybber> open nim issues in rusts issue tracker instead
14:13:26narimiranlol
14:13:34Yardanicothey don't have a stale bot btw :)
14:13:38Yardanicoand they have 5.7k open issues
14:13:38leorize!repo nocode
14:13:41disbothttps://github.com/kelseyhightower/nocode -- 9nocode: 11The best way to write secure and reliable applications. Write nothing; deploy nowhere. 15 42701⭐ 3883🍴 7& 29 more...
14:14:11narimiranYardanico: they have both more manpower and more users
14:14:15Yardanicoso?
14:14:16*nurse_oda_purse quit (Quit: Leaving)
14:14:22Yardanicoa lot of these open issues are also "not fixed" for years
14:14:57leorizedoes the high issue count suddenly cause contributors to experience fatigue? :P
14:15:05Yardanicoyes of course
14:15:11Yardanicothey'll go to the oldest issue and try to solve it first obviously
14:15:16Yardanicowho would look at newest issues, pfft
14:15:17FromDiscord<Clyybber> can *not* confirm. Am experiencing fatigue due to the bot :p
14:16:07Yardanicoa lot of stuff in open source is not because "oh, I will fix a bug which affects a lot of people", but because "oh, I will fix a bug"
14:16:09Yardanicoit's just about interest
14:17:23narimiranand now the bot has sparked all y'all's interest in old issues. big win!
14:18:05FromDiscord<Clyybber> not really :(
14:18:29FromDiscord<Clyybber> since it floods me
14:18:35leorizeeveryone will stop caring the next day and ban the bot due to spam
14:18:58FromDiscord<Clyybber> everybody report the bot :p
14:19:03FromDiscord<Clyybber> rage against the machine!
14:19:29narimirantoday i could delete half of the nim stdlib and no one would notice
14:20:20Yardanicouhh
14:20:38supakeendo it
14:20:40FromDiscord<Clyybber> the tests would
14:20:41Yardanicothat doesn't justifty closing the issues :P
14:20:53Yardanicothey're *just there*
14:20:54narimirani will delete the tests too
14:21:18narimiranno stdlib = no problems = no issues = no need for bot
14:21:27Yardanicono language = no need for bot
14:21:38narimiranno users = no need for bot
14:24:42*arecacea1 quit (Ping timeout: 260 seconds)
14:28:29*arecacea1 joined #nim
14:36:22*vsantana quit (Quit: vsantana)
14:37:18*vsantana joined #nim
14:52:49*Trustable quit (Remote host closed the connection)
14:59:59*nc-x joined #nim
15:02:18nc-x> and there i thought nothing could be hated more than 'fusion' :D
15:02:21nc-xwhats up with fusion anyways?
15:03:00nc-xlast i heard it is now more of an extended stdlib than the distribution of external packages it was supposed to be
15:03:18FromDiscord<Rika> i dont think it was ever supposed to be the latter you said
15:03:22disruptekshhh don't wake it up.
15:03:42disruptekbetter that it stays asleep.
15:04:39nc-xRika: in the fusion rfc > We plan to create a "Nim distribution" which consists of the Nim compiler, Nimble and a "selected/curated set of Nimble packages."
15:04:56nc-xdisruptek: i just want to know what the difference is between stdlib and extended stdlib
15:05:03nc-xapart from them being in different repos
15:06:09disruptekwhat's "extended stdlib"?
15:06:12disruptekdon't ask me about fusion.
15:06:12disrupteki'll just say that the only reason i hate fusion is that it's terrible in nearly every way.
15:09:59nc-xdisruptek: extended stdlib = https://github.com/nim-lang/fusion/issues/1#issuecomment-623552997 (point 3). I may be misunderstanding but that makes it look like once a module makes its way into fusion, then its maintenance burden is on the core nim team and "hopefully" any contributors, which is no different than the stdlib except that fusion is in
15:10:00disbotβž₯ Releases
15:10:00nc-xa separate repo.
15:10:47*vsantana_ joined #nim
15:11:01nc-xand because it is also going to ship with nim releases, it is just stdlib with a different name
15:11:01*vsantana quit (Ping timeout: 264 seconds)
15:11:05FromDiscord<Clyybber> yeah, its a failed idea IMO
15:11:11FromDiscord<Clyybber> it should be bundled packages
15:11:12nc-xso that is what i don't understand
15:11:17nc-xyeah
15:11:25FromDiscord<Clyybber> literally just a .nimble that we ship with the install
15:11:31FromDiscord<Clyybber> but its not :(
15:11:44FromDiscord<Clyybber> disruptek: Oh you are going to love this bot
15:11:49Yardanicoohohoho
15:13:00disruptekwut
15:13:16disruptekit's raining, which means i have like 90% packet-loss.
15:15:04FromDiscord<Clyybber> good for you
15:15:10FromDiscord<Clyybber> don't ever open github notifications
15:15:14FromDiscord<Clyybber> if you watch the nim repo
15:15:15FromDiscord<Clyybber> that is
15:15:38disruptekoh dear.
15:16:25FromDiscord<haxscramper> Is it possible to return type (`typedesc`) from a function? Something like `func test(a: typedesc): typedesc = (if a is float: a else: string)`
15:16:34Yardanicotypedesc is a compile time only value
15:16:38Yardanicoit doesn't exist at runtime
15:17:37FromDiscord<haxscramper> Even in context like `SomeGeneric[test(float)]`?
15:17:45FromDiscord<Rika> yes
15:17:59FromDiscord<Rika> that's a compiletime proc is it not?
15:18:15FromDiscord<Rika> or can compile time procs return typedescs? idk
15:18:47FromDiscord<haxscramper> With generics it just evaluates to `void` - no compilation errors though
15:21:14shashlickinstead of all this talk, let's all fix one so called stale issue each and then narimiran will back off, what say
15:21:35narimirangood idea!
15:21:47narimiranand if i don't back off immediately, fix another one!
15:22:32shashlickfrankly, I don't see how a closed stale issue is any better than an open stale issue but when we reach 100k stale open issues, this topic will come back regardless
15:22:51shashlickso it is not a - nim has few maintainers - issue, it is the case with every project
15:23:21*vicfred quit (Quit: Leaving)
15:23:32Yardanicoshashlick: so do you close stale issues in nimterop? :)
15:23:35shashlickthere's always more issues than can be fixed - some are simply too boring to debug and fix but are legitimate
15:23:54shashlickno, I consciously close issues I won't fix
15:24:06shashlicki don't think time is a valid metric
15:24:26shashlickif you want to close stale issues then consciously mark them as won't fix or something
15:24:44dom96shashlick, can we implement a GitHub API call in Nimble so that the stale bot is automagically enabled for their repos? ;)
15:24:49dom96Stale bot for all Nim projects
15:24:54shashlickor consciously mark them as stale
15:25:03dom96Praise be the Stale bot.
15:26:35*nc-x quit (Remote host closed the connection)
15:26:57shashlickis @mratsim around
15:27:22*waleee-cl quit (Quit: Connection closed for inactivity)
15:41:25*Lord_Nightmare quit (Ping timeout: 240 seconds)
15:57:28FromDiscord<Ricky Spanish> any suggestions on where the best place would be to jump in if i were interested in contributing to either nim or nims tooling?
15:57:48leorizenimsuggest :)
15:58:36FromDiscord<Ricky Spanish> ok thanks will look at the repo @leorize
15:59:07leorizenimsuggest is actually a part of the compiler :P
15:59:24FromDiscord<Ricky Spanish> yeh i seen it was moved back in
16:09:05disrupteknimble needs a ton of work, too.
16:19:45*apahl quit (Remote host closed the connection)
16:20:08*apahl joined #nim
16:28:02*vicfred joined #nim
16:31:27*ehmry quit (Quit: https://quassel-irc.org - Chat comfortably. Anywhere.)
16:31:56*narimiran quit (Quit: leaving)
16:53:23*maier quit (Ping timeout: 240 seconds)
16:59:12*superbia1 joined #nim
16:59:55*lritter quit (Quit: Leaving)
17:01:15Zevvdisruptek: you have a visible-light laser link for your internet?
17:01:29Zevvyou have to time your packets to go _inbetween_ the drops
17:01:30*superbia quit (Ping timeout: 260 seconds)
17:09:07*vsantana_ quit (Quit: leaving)
17:10:45disrupteki'm trying.
17:12:27Zevvdecrease your mtu to 150 bytes
17:12:39Zevvdisable window scaling
17:16:14disruptekdidn't i bump it down to 296 the other day?
17:17:28Zevvwe need to go lower
17:17:33Zevvthroughput is overrated
17:17:50Zevvsay, I don't think you can be bothered to fix more on the current cps implementation, right
17:18:10disruptekwhy not?
17:18:28disruptekwhat doesn't work?
17:18:30ZevvI heard you were kind of sick messing around in that code base
17:18:35Zevvstandalone_tcp_server
17:18:45Zevvit lifts a local that should stay local
17:19:00disruptekthere's a bug that tblock demonstrates that might be pretty hard to fix.
17:19:15disrupteklemme look at tcp.
17:19:46Zevvstandalone_http_server, btw
17:19:50disruptekit's probably the same bug, unfortunately.
17:20:15disruptekwhere is http_server?
17:20:45Zevvno, it is tcp_server, sorry
17:24:58disruptekdo you wanna play with my testes?
17:25:35disruptekthey are all red.
17:25:35ZevvI'm good
17:25:42disrupteklol
17:25:45Yardanicohm
17:25:58disruptekfor some reason, cps really breaks my balls.
17:26:03Zevvall of them
17:26:04Yardanicopag
17:26:07Yardanicosorry wrong chat
17:26:39Zevvdisruptek: we should try to split this up much more strictly somehow. Make isolated stages that share as few as possible
17:26:55disruptekdid you look at the list i made?
17:27:02ZevvI did.
17:27:04disrupteki can't really do much of anything online today.
17:27:07ZevvI was just being a smartass
17:27:25disruptekoh. actually, it's been kinda okay for a few minutes.
17:27:47disrupteki think we will have to run some of those stages more than once, but whatever.
17:29:27*mbuchel joined #nim
17:36:45Zevvi have stage fright
17:36:55disruptekoh yeah?
17:39:26Zevvyeah
17:41:24*audiophile joined #nim
17:43:22disrupteki think they look pretty doable.
17:43:59disrupteki wonder where the gratuitous assignments goes.
17:44:43disrupteki think maybe if they are {.used.} we may have to force them in. they might be there for side-effect reasons.
17:44:52*audiofile quit (Ping timeout: 246 seconds)
17:44:59*audiophile is now known as audiofile
17:49:27Zevvand still we have the elephant in the rome
17:49:29Zevvroom
17:49:44Zevvvar a: int = 42
17:49:50disrupteki know. my testes crashes the compiler.
17:50:00Zevvthat's the rhino
17:50:03disruptekoh, what is that?
17:50:14Zevvthat we still have to explicitly "give a type"
17:50:17disruptek!package disruptek
17:50:17disbotdisruptek hung like rhinocerous🦏
17:50:50disruptekif we fix that, it will probably make testes green.
17:51:10disruptek🟒🟒
17:51:35*endragor quit (Remote host closed the connection)
17:51:39*Lord_Nightmare joined #nim
17:52:09*endragor joined #nim
17:56:58*endragor quit (Ping timeout: 256 seconds)
17:57:18FromDiscord<Clyybber> I'm too stupid to backup a scope..
17:57:31disruptekto what?
17:57:45Zevvmy scope can backup to 1.44" floppy
17:57:47FromDiscord<Clyybber> Trying to fix https://github.com/nim-lang/Nim/issues/5691
17:57:49disbotβž₯ The StmtList processing of template parameters can lead to unexpected errors ; snippet at 12https://play.nim-lang.org/#ix=2tef
17:57:58FromDiscord<Clyybber> since it seems easy enough
18:04:40*endragor joined #nim
18:07:16*zama quit (Ping timeout: 246 seconds)
18:07:49*zama joined #nim
18:08:53FromDiscord<Clyybber> damn hashtags
18:09:50*endragor quit (Ping timeout: 256 seconds)
18:11:36disrupteki want nim to be able to use directories like foobar/ and not merely foobar-1.2.3 or foobar-#special.
18:11:57disruptekit's annoying to keep the directory name up-to-date, and for obvious reasons.
18:12:02ZevvI want my nim files to be able to have hyphens in them
18:12:12disruptekwell, that's just silly.
18:12:17mbucheli want nim to become a subset of lisp
18:12:27disruptekthat's not very silly, honestly.
18:13:01FromDiscord<Clyybber> subset is silly
18:13:05mbuchelbut this means we move to kebab-case
18:13:16disrupteknim used to have that feature.
18:13:17FromDiscord<Clyybber> cursed-1
18:13:36mbuchelwhy did we remove the kebab-case?
18:13:47FromDiscord<Clyybber> isn't it obvious?
18:13:49mbuchelwere you trying to remove kebabs?
18:13:51FromDiscord<Clyybber> it conflicts with minus
18:13:58mbuchelno i know that
18:14:00disruptekto distinguish ourselves from lisp.
18:14:01FromDiscord<Clyybber> heh :D
18:14:11mbuchelBLASPHEMER
18:14:28disruptekdon't ever trim your eyebrows yourself.
18:14:30FromDiscord<Clyybber> kebab belongs in my belly not Nim
18:14:47FromDiscord<Clyybber> why trim your eyebrows
18:15:03mbuchelto prevent autism from spreading
18:15:14FromDiscord<Clyybber> understandable, have a great day
18:15:18disruptekthere's just no stopping it.
18:16:08mbuchelso is there any way to get around "cannot provie it's safe to initialize ... with runtime value for the discrimminator kind"?
18:17:45ForumUpdaterBotNew thread by Jseb: Sets : what are they, see https://forum.nim-lang.org/t/6640
18:18:12disrupteknot really; it was promised but not delivered.
18:18:25disruptekiirc.
18:18:33disruptekclyybber: ^
18:19:03disrupteki mean, "any way"? yes. use a case statement to match constants to runtime values.
18:19:37disruptekbut you can't do what you want to do, which is newFoo(kind: FooKind) ...
18:20:20FromDiscord<Clyybber> disruptek: v
18:20:30FromDiscord<Clyybber> oh
18:27:25*endragor joined #nim
18:32:26*endragor quit (Ping timeout: 256 seconds)
18:49:53*maier joined #nim
18:53:20*Trustable joined #nim
18:55:25*maier quit (Ping timeout: 264 seconds)
19:07:43*opal quit (Ping timeout: 240 seconds)
19:09:25shashlick@leorize - need your help
19:16:42*clemens3 quit (Ping timeout: 256 seconds)
19:22:47*opal joined #nim
19:22:47*clemens3 joined #nim
19:31:52*Kaivo quit (Quit: turning vpn on or off, idk if it's on atm)
19:32:32*Kaivo joined #nim
19:39:20disruptekwho wants to write a markdown renderer for terminal?
19:39:31FromDiscord<Clyybber> neat idea
19:39:43disrupteki would settle for rst, i guess. πŸ™„
19:40:11FromDiscord<Clyybber> damn. Imagine spending an hour to search for why your scope backups dont work
19:40:28FromDiscord<Clyybber> to find out somewhere something uses shallowCopy :p
19:40:45disruptekwhat's a scope backup?
19:41:22disruptekjust a copy of the scope?
19:43:53FromDiscord<Shucks> https://play.nim-lang.org/#ix=2tf8 which data type could I use to access them like colors.White? I don't want to use a seperate nim file and `from colors import nil`
19:45:12disrupteknot sure what you're asking.
19:45:15disruptek!repo chroma
19:45:16disbothttps://github.com/treeform/chroma -- 9chroma: 11Everything you want to do with colors, in nim. 15 41⭐ 7🍴
19:45:36FromDiscord<Elegant Beef> They basically want to have the colour act like an enum but return an array
19:45:43FromDiscord<Shucks> yes
19:45:47*Vladar quit (Quit: Leaving)
19:46:03disruptekuse an enum as an index into an array of consts.
19:46:15disruptekthe consts can be arrays or whatever.
19:46:38FromDiscord<Elegant Beef> Yea that's the closest you'll get with the end result being like `colors[white]`
19:46:51disruptekyou can always template it.
19:49:16FromDiscord<Shucks> Thank you
19:53:59disrupteksure thing, boss.
20:01:33FromDiscord<Technisha (LGBTQIAP+)> If i had some binary data in a json file (please don't ask why <a:HR_PotatoSweat:645117586085773312>), how would I execute it without writing it to a temporary file?
20:02:06FromDiscord<Recruit_main707> i dont think you could
20:02:10FromDiscord<Technisha (LGBTQIAP+)> Hm
20:02:20FromDiscord<Recruit_main707> i do need to ask why though
20:02:29FromDiscord<Technisha (LGBTQIAP+)> <a:HR_PotatoSweat:645117586085773312>
20:02:40FromDiscord<Technisha (LGBTQIAP+)> I'll just send an image to explain ut
20:02:42FromDiscord<Technisha (LGBTQIAP+)> It*
20:04:16FromDiscord<Technisha (LGBTQIAP+)> Two images- https://media.discordapp.net/attachments/371759389889003532/740299077358452896/Screenshot_20200804_210321_com.discord.jpg https://media.discordapp.net/attachments/371759389889003532/740299077874483290/Screenshot_20200804_210346_com.discord.jpg
20:04:29FromDiscord<Technisha (LGBTQIAP+)> So now, because of @codic, I'ma implement this in Nim-
20:05:01FromDiscord<Rika> why?
20:05:03FromDiscord<Technisha (LGBTQIAP+)> @Recruit_main707 is this a good explanation?-
20:05:15FromDiscord<Technisha (LGBTQIAP+)> > why?↡Don't ask me, ask him-
20:05:23FromDiscord<Technisha (LGBTQIAP+)> For once it wasn't my idea-
20:05:25FromDiscord<Rika> why do you need it to not write into a file
20:05:37FromDiscord<Rika> > Don't ask me, ask him-↡@Technisha (LGBTQIAP+) yeah but you want to implement it in nim, why?
20:06:14FromDiscord<Recruit_main707> Prove nim can do the same as rust
20:06:20FromDiscord<Recruit_main707> I guess...
20:06:30FromDiscord<Recruit_main707> Or curiosity
20:07:05leorizeshashlick: o/
20:08:20FromDiscord<Varriount> @Elegant Beef Converter?
20:08:50FromDiscord<Technisha (LGBTQIAP+)> > Or curiosity↡Winner winner chicken dinner! You want a prize?
20:09:13FromDiscord<Elegant Beef> @Varriount Diverter?
20:09:58FromDiscord<Rika> > why do you need it to not write into a file↡@Technisha (LGBTQIAP+) next question, is this important?
20:10:40FromDiscord<Technisha (LGBTQIAP+)> Just in case there's no space to write it anywhere
20:10:52FromDiscord<Elegant Beef> But i do think i see what you mean varriount, use a pure enum and then use a convert to sample the underlying arrayt
20:10:54FromDiscord<Technisha (LGBTQIAP+)> And again, curiosity
20:10:56FromDiscord<Elegant Beef> (edit) 'convert' => 'converter' | 'arrayt' => 'array'
20:11:23FromDiscord<Rika> write it in the tmpfs
20:11:26FromDiscord<Elegant Beef> > there's no space to write it anywhere↡Ah yea fuck forget we're running c64s!
20:11:32FromDiscord<Rika> it'll be written to memory only
20:11:54FromDiscord<Rika> if you're out of space *and* memory you're fucked anyway so
20:12:05FromDiscord<Elegant Beef> Well that's when you go use vram
20:12:48FromDiscord<Rika> if you dont have tmpfs (windows or smth) there's https://nim-lang.org/docs/memfiles.html
20:13:22FromDiscord<Rika> im not actually sure if you can make files with the memfiles module
20:13:28FromDiscord<Rika> because i only skimmed through it
20:13:35FromDiscord<Rika> but you can read the docs right?
20:13:48*waleee-cl joined #nim
20:14:28disruptekrika: what happened to you?
20:16:00FromDiscord<Rika> i dont know, perhaps ive been this idiotic all along
20:16:03FromDiscord<Rika> dont ask really
20:16:05FromDiscord<Rika> i wouldnt know
20:16:13FromDiscord<Technisha (LGBTQIAP+)> Hmm...
20:16:20FromDiscord<Technisha (LGBTQIAP+)> Thanks Rika!
20:16:39FromDiscord<Technisha (LGBTQIAP+)> https://nim-lang.org/docs/streams.html
20:16:42FromDiscord<Technisha (LGBTQIAP+)> Hmm-
20:17:06FromDiscord<Technisha (LGBTQIAP+)> How does that work?
20:17:25FromDiscord<Rika> streams still needs a real file (unless, again, you can use tmpfs)
20:17:50disruptekyou could use a stringstream.
20:17:54disruptekwhat are you trying to do?
20:18:16FromDiscord<Rika> trying to execute a binary blob without writing a file onto a regular file system
20:19:35disrupteki guess you could use a pipe or something.
20:20:40disruptekman, what a hideous day.
20:20:49FromDiscord<Technisha (LGBTQIAP+)> So basically memfiles work like normal files?-
20:21:32FromDiscord<Elegant Beef> They're files that point to ram instead of a drive
20:21:43FromDiscord<Technisha (LGBTQIAP+)> Coolio
20:21:52FromDiscord<Rika> okay honestly i think you should just go with writing a regular file to a `tmpfs` filesystem because its easier
20:21:56FromDiscord<Rika> but not portable
20:22:04FromDiscord<Rika> i dont know how the rust version pulls it off
20:22:05FromDiscord<Varriount> @Technisha (LGBTQIAP+) You just have to mark the memory as executable
20:22:15disruptekyou can't get much more portable than their computer.
20:22:20FromDiscord<Technisha (LGBTQIAP+)> That can be done with the os module, correct?
20:22:33FromDiscord<Technisha (LGBTQIAP+)> > you can't get much more portable than their computer.↡True xD
20:22:42FromDiscord<Technisha (LGBTQIAP+)> But I am a distrohopper
20:22:43FromDiscord<Technisha (LGBTQIAP+)> So
20:22:46FromDiscord<Technisha (LGBTQIAP+)> Β―\\_(ツ)\_/Β―
20:22:52FromDiscord<Elegant Beef> You code on your phone was the joke
20:23:03FromDiscord<Rika> @Technisha (LGBTQIAP+) maybe read https://stackoverflow.com/questions/13690454/how-to-compile-and-execute-from-memory-directly
20:23:07FromDiscord<Technisha (LGBTQIAP+)> I do code on my phone mainly
20:23:12FromDiscord<Technisha (LGBTQIAP+)> But i do have a PC
20:23:15FromDiscord<Technisha (LGBTQIAP+)> Thanks Rika!
20:23:25FromDiscord<Technisha (LGBTQIAP+)> (edit) 'But i do have a PC ... ' => 'But i do have a PCthat i rarely code on-'
20:23:40FromDiscord<Varriount> @Technisha (LGBTQIAP+) So, just to make sure my assumptions are correct, you want your program to retrieve executable data from somewhere, and execute it, without writing the data to a file?
20:23:55FromDiscord<Varriount> (on the disk)
20:23:55FromDiscord<Technisha (LGBTQIAP+)> Yup
20:23:57disrupteki find this so strange, but maybe there are some good things about the phone as a dev box.
20:24:03Yardanicothat's the area of JITs and similar
20:24:05FromDiscord<Technisha (LGBTQIAP+)> It'll read the dsta from a json file
20:24:16Yardanicoit's far from simple to actually *execute*
20:24:16disruptekwhat's good about dev on a phone?
20:24:19FromDiscord<Technisha (LGBTQIAP+)> > i find this so strange, but maybe there are some good things about the phone as a dev box.↡It's portable :)
20:24:19Yardanicoif it's just instructions
20:24:21Yardanicowithout ELF etc
20:24:36FromDiscord<Elegant Beef> You can code and shit disruptek, which is not much difference to what i normal do as i code shit
20:24:46FromDiscord<Varriount> @Technisha (LGBTQIAP+) Perhaps Linux allows exec() from memory. If it doesn't, then you'll have to write an executable loader
20:25:00FromDiscord<Technisha (LGBTQIAP+)> > it's far from simple to actually *execute*↡> if it's just instructions↡> without ELF etc↡It's data from a binary
20:25:04disruptekwhatever blows your skirt up, i guess.
20:25:12FromDiscord<Varriount> Or compile your retrieved blob as a position-independent blob
20:25:12FromDiscord<Rika> @Varriount maybe the link i sent technisha above would also inform you πŸ˜›
20:25:27FromDiscord<Technisha (LGBTQIAP+)> > whatever blows your skirt up, i guess.↡No thanks-
20:25:28FromDiscord<Rika> @Technisha (LGBTQIAP+) that means it has ELF
20:25:30FromDiscord<Technisha (LGBTQIAP+)> Lmao
20:25:32FromDiscord<Recruit_main707> Code an emulator,boot windows and then run it there
20:25:33FromDiscord<Technisha (LGBTQIAP+)> Mhm
20:25:59disrupteki mean whatever peels your banana, i guess.
20:26:10FromDiscord<Recruit_main707> Lol
20:26:36FromDiscord<Varriount> @Rika Ah, memexec makes it much easier
20:26:58FromDiscord<codic> @Technisha (LGBTQIAP+) please don't
20:27:20FromDiscord<Rika> i really think its hard to implement the "load from memory w/o writing a file" idea because programmers made it hard intentionally, to combat arbitrary code execution i assume
20:27:34FromDiscord<Varriount> @Rika No having to do function fixups and the like
20:27:55FromDiscord<Rika> the memexec function still writes a file
20:28:00FromDiscord<Rika> just not to a traditional file system
20:28:04FromDiscord<Technisha (LGBTQIAP+)> > @Technisha (LGBTQIAP+) please don't↡@codic why not?-
20:28:14FromDiscord<codic> because it's a duplication of efforts
20:28:22FromDiscord<Technisha (LGBTQIAP+)> Eh-
20:28:24FromDiscord<Technisha (LGBTQIAP+)> Idc-
20:28:27FromDiscord<codic> and yeah, I'm going to be using a file stream similar to memexec
20:28:27FromDiscord<Rika> duplication of effort isnt always bad
20:28:33FromDiscord<Technisha (LGBTQIAP+)> You gave me the stupid idea-
20:28:41FromDiscord<Varriount> Why do you not want to write to a file?
20:28:42FromDiscord<Rika> especially when the second effortmaker just wants to learn or try it out
20:29:06FromDiscord<codic> that's true
20:29:32FromDiscord<Technisha (LGBTQIAP+)> > Why do you not want to write to a file?↡A) Curiosity↡B) If there's no space on the disk
20:29:40FromDiscord<codic> ...
20:29:45FromDiscord<codic> There will always be space on the disk
20:29:45FromDiscord<Technisha (LGBTQIAP+)> :3
20:29:53FromDiscord<Varriount> "A" is good enough for me
20:29:55FromDiscord<Technisha (LGBTQIAP+)> Not in my case-
20:29:58FromDiscord<codic> Because they will have space to store the original file
20:29:58FromDiscord<Technisha (LGBTQIAP+)> Lmao
20:30:00FromDiscord<codic> What is your "case"?
20:30:08*vicfred quit (Quit: Leaving)
20:30:23FromDiscord<Technisha (LGBTQIAP+)> I don't have alot of storage on my PC because i run out from all the shit i install
20:30:31FromDiscord<codic> How much exactly do you have
20:30:41disruptekplease, be exact.
20:30:43FromDiscord<Rika> binaries arent that large, you know
20:30:53disrupteklet's talk numbers.
20:30:53FromDiscord<Rika> unless theyre go binaries but w/e
20:31:00FromDiscord<codic> yeah, especially if you strip them
20:31:06disruptekdo you strip them?
20:31:06FromDiscord<codic> also, the file you write to would be temporary
20:31:09FromDiscord<Technisha (LGBTQIAP+)> I have a 250gb SSD which gets bloated quickly-
20:31:13FromDiscord<Varriount> It's usually secondary data (images, etc) that take up space
20:31:17FromDiscord<codic> yeah, i do↡i have a 100gb ssd
20:31:32FromDiscord<Technisha (LGBTQIAP+)> You don't install as much shit as me-
20:31:36FromDiscord<Varriount> @Technisha (LGBTQIAP+) Have you turned on transparent compression?
20:31:42FromDiscord<Technisha (LGBTQIAP+)> Wdym?
20:31:45FromDiscord<Varriount> And what OS?
20:31:56FromDiscord<Technisha (LGBTQIAP+)> Rn I'm using Solus
20:32:00FromDiscord<codic> I install a *lot* of stuff
20:32:02Yardanicoand how you not have free space?
20:32:08FromDiscord<Rika> > Have you turned on transparent compression?↡@Varriount not all filesystems support transparent compression
20:32:09Yardanicodid you incorrectly size your root partition?
20:32:14Yardanicois it separate from the home one?
20:32:23FromDiscord<codic> i very highly doubt that you do not have enough space to extract, at most, a mb or two of data
20:32:28FromDiscord<Rika> and theyre kinda "not beginner friendly" because it usually needs editing the fstab
20:32:31FromDiscord<codic> and then delete it afterwards anyways
20:33:12FromDiscord<Technisha (LGBTQIAP+)> I'm not a beginner at using Linux so it should be alright with me :P
20:33:14FromDiscord<Varriount> @Rika The supposed "beauty" of Linux is that it places emphasis on multiple filesystem mounts.;D
20:33:43Yardanico@Technisha I doubt your Linux ability a bit if you say you don't have enough space :)
20:33:48FromDiscord<Rika> i agree, but theyre usually unnecessary for single user PCs (like most desktops are)
20:33:48FromDiscord<codic> &
20:33:51FromDiscord<codic> (edit) '&' => '^'
20:33:51FromDiscord<codic> @Technisha (LGBTQIAP+) They are saying to resize the partition using your fstab *if* you don't have enough space in your parititon, because you sized it incorrectly
20:34:03FromDiscord<Technisha (LGBTQIAP+)> O
20:34:04FromDiscord<Rika> no
20:34:05FromDiscord<Rika> what i mean
20:34:07FromDiscord<codic> oh
20:34:13FromDiscord<codic> then what?
20:34:14FromDiscord<Technisha (LGBTQIAP+)> O again-
20:34:31FromDiscord<Rika> we're saying to add a setting to turn transparent compression on in the fstab if you really need the space on your drive
20:34:42FromDiscord<Rika> and only if you're using something like zfs or btrfs
20:34:48FromDiscord<Technisha (LGBTQIAP+)> Oh
20:34:50FromDiscord<Rika> i dont remember ext4 supporting transparent compression
20:34:57FromDiscord<Technisha (LGBTQIAP+)> I think I'm using ext4 or something
20:35:41FromDiscord<Technisha (LGBTQIAP+)> I haven't checked in god knows how long-
20:35:41FromDiscord<Rika> well you cant do much about that then
20:35:47FromDiscord<codic> you are using ext4
20:35:53FromDiscord<codic> solus is ext4
20:35:57FromDiscord<Technisha (LGBTQIAP+)> O
20:35:58FromDiscord<Rika> if you dont know what you're using you're using ext4
20:36:00FromDiscord<Technisha (LGBTQIAP+)> :P
20:36:06FromDiscord<Technisha (LGBTQIAP+)> :P
20:36:06FromDiscord<codic> i played around with zfs a bit
20:36:07FromDiscord<codic> it was nice
20:36:16FromDiscord<codic> bit too unstable for me, but that's offtopic
20:36:18FromDiscord<Technisha (LGBTQIAP+)> It's faster then ext4, right?
20:36:47FromDiscord<codic> Depends
20:37:00FromDiscord<codic> Here are some benchmarks: https://www.phoronix.com/scan.php?page=article&item=ubuntu1910-ext4-zfs&num=1
20:37:15FromDiscord<Rika> i'm using btrfs on my backup drive (because i need to store a lot of backups in a small amount of space, deduplication and compression come in really handy here) but i dont trust it enough for daily usage so i still use ext4 on my main
20:37:29Yardanicobtrfs is stable enough
20:37:32Yardanicounless you have raid5/6
20:37:37FromDiscord<Rika> really?
20:37:37FromDiscord<codic> never really tried BTRFS
20:37:50FromDiscord<codic> but the point is, you should definitely have enough space
20:37:59FromDiscord<Rika> ive heard horror stories but theyre a few years old
20:38:02FromDiscord<Rika> anyway yeah
20:38:07FromDiscord<codic> if you don't, you have bigger problems to worry about lol
20:38:08FromDiscord<Varriount> Does ext3 support transparent compression?
20:38:18FromDiscord<Rika> you have bigger problems to worry about other than storage space
20:38:24Yardanicoext doesn't support compression at all
20:38:32FromDiscord<Rika> > Does ext3 support transparent compression?↡@Varriount i dont remember it having such
20:38:42FromDiscord<codic> I believe that ext3 has a patch for transparent compression, but not official
20:38:49Yardanicobtrfs/zfs/f2fs do support it
20:38:54FromDiscord<codic> Ah yeah, e3compr
20:38:54YardanicoI mean compression
20:38:55Yardanicomaybe more
20:39:04FromDiscord<codic> https://sourceforge.net/projects/e3compr/
20:39:08FromDiscord<Varriount> @Technisha (LGBTQIAP+) Could you just... Buy a bigger drive?
20:39:20FromDiscord<codic> I think that they *do* have enough space
20:39:24FromDiscord<codic> just are a bit confused
20:39:46disruptekbtrfs is the least stable fs i've used since ext/2 days.
20:40:39FromDiscord<Rika> i'd love to believe you but your sarcasm is really hard to detect so i can't
20:40:55FromDiscord<codic> πŸ‘€
20:41:11disruptekis it?
20:41:27Yardanicodisruptek: idk when you used it
20:41:30Yardanicobtrfs nowadays is pretty stable
20:41:40disrupteki use it still.
20:42:01disruptekthe last time i got annoyed enough to try to switch, i though i'd try bcachefs.
20:42:18FromDiscord<Varriount> I use NTFS
20:42:21disruptekbut there was something i needed in a later kernel and i didn't feel like porting it.
20:42:22Yardanicomeh :)
20:42:23Yardanicontfs
20:42:24FromDiscord<codic> Ew
20:42:40FromDiscord<Varriount> The only thing bad about it is the small file performance
20:42:42FromDiscord<codic> zfs gang
20:42:51FromDiscord<codic> hmm if you search up "zfs gang" turns out zfs has a thing called "gang blocks"
20:42:56FromDiscord<codic> https://utcc.utoronto.ca/~cks/space/blog/solaris/ZFSGangBlocks
20:43:03Yardanico"gang" cringe stuff
20:43:06Yardanico:)
20:43:08disrupteki saw that in a movie once.
20:43:16FromDiscord<codic> gang blocks?
20:43:17disruptekokay, i've seen it in lots of movies.
20:43:19leorize@Rika you don't trust btrfs enough to daily it but you trust it enough to store your backups?
20:43:23*supakeen promotes Yardanico to nim gang member
20:43:27Yardanicono, "<insert something here> gang"
20:43:29Yardanicoit's cringe really
20:43:35FromDiscord<codic> lol
20:43:41disruptekhey, if you want an @nim.fan email address, just ask.
20:43:52supakeendisruptek: do you offer @nim.gang email addresses?
20:43:57disrupteknope.
20:44:33disrupteki use btrfs over btrfs and it's a disaster.
20:44:46FromDiscord<codic> @nimmer.fan please
20:44:50FromDiscord<Rika> leorize: i only just started making these backups (in fact, i only have one right now) so i trust it NOW but before i didnt
20:45:01FromDiscord<codic> uh oh, you know we just pinged someone named "nimmer" πŸ‘€
20:45:07FromDiscord<Rika> i dont trust a CONVERSION from ext4 to btrfs, though
20:45:34disruptek@nimmer isn't a thing because i don't have @nimmer or @nimmer.gang or @nimmer, uh, @nimmer.fan.
20:45:43leorizethat makes more sense :P
20:45:44disruptek@nimmer isn't a thing because i don't have @nimmer or @nimmer.gang or @nimmer, uh, @nimmer.fan. but i do have nim.fan.
20:45:57FromDiscord<codic> Lmao
20:45:57Yardanicoyes
20:46:02Yardanicoconversion from ext4 to btrfs is fully broken
20:46:03Yardaniconever do that
20:46:08FromDiscord<codic> https://media.discordapp.net/attachments/371759389889003532/740309612149997578/unknown.png
20:46:12Yardanicospeaking from experienc e
20:46:19Yardanicobtrfs itself is fine
20:46:24leorizeI'm running btrfs daily here
20:46:38leorizeperks over zfs: you don't have to compile zfs
20:46:42disruptekterribad.
20:46:46Yardanicoleorize: true
20:46:49Yardanicocompatible license :)
20:46:50FromDiscord<Rika> i'll use it when i distrohop on my laptop (to nixos)
20:46:56disrupteki did switch my cache from btrfs and ext4 and now it purrs.
20:46:59disruptekproblem solved.
20:47:06disruptekso, btrfs over btrfs. terrible.
20:47:39FromDiscord<codic> πŸ€”
20:47:45disruptekZevv: you know what this bug is?
20:47:49disrupteki finally looked at it.
20:49:21disruptekwe go through the nodes and create something locally, `s`. now it's in the env. now we figure out what the type is; oh, it's env0 we're returning. what goes into env0? s. let's make sure it's locallized... argh.
20:50:39FromDiscord<Varriount> Anyone know if there's a way to switch off functions that read/write/exec things at compile time?
20:50:42*maier joined #nim
20:51:15FromDiscord<Varriount> I'm running the compiler through a fuzzer, and don't want a randomly generated program to delete the entire system.
20:51:24Yardanicohehe
20:51:26FromDiscord<Rika> ~~jail them~~
20:51:31Yardanicobut really
20:51:35Yardanicorun it in a container lol
20:51:41FromDiscord<Rika> or that
20:51:51FromDiscord<Varriount> Can't, not with the fuzzer
20:52:09Yardanicoare you sure?
20:52:11Yardaniconot even a VM?
20:52:40*NimBot joined #nim
20:52:49FromDiscord<Varriount> If I run the fuzzer, which stores its state locally, in a VM or container, I end up with the same problem.
20:53:08Zevvdisruptek: told you that!
20:53:19Yardanicobut you can send the state once in a while?
20:53:24FromDiscord<Varriount> If the fuzzer is told to run "docker run", or whatever, then it can't analyze the binary
20:53:25Yardanicowhy the fuzzer is so stupid you can't save the state?
20:53:39disruptekit's weird, i didn't think i was sharing the locals that much.
20:53:57FromDiscord<Varriount> Yardanico: Where am I going to save the state?
20:54:03Yardanicooutside of the container/VM
20:54:07Yardanicoin the *cloud*
20:54:10Zevvtake me to your locals
20:54:29disruptekgotta compose them on the way in; i guess i'm not doing that.
20:55:00FromDiscord<Varriount> Yardanico: If the fuzzer is running in the VM, and can access the external files, then the fuzzer compiler can too
20:55:22FromDiscord<Varriount> *fuzzed Nim compiler
20:55:37*maier quit (Ping timeout: 246 seconds)
20:56:51FromDiscord<Rika> cant you pass into the container while disallowing outwards access?
20:56:57FromDiscord<Rika> pass data into*
20:57:14FromDiscord<Rika> (btw i have not much of an idea how containers work)
20:57:23disrupteki'm worried his ai will steal my identity.
20:57:44FromDiscord<Varriount> I guess I could run a job outside the container to periodically copy the fuzzer state
20:57:52disruptekmaybe it's trying to login as me, from the future, in order to warn us.
20:58:16FromDiscord<Varriount> But then, when restoring the state, it will just run the bad program input again
20:58:56disruptekyou have that risk in any event, so to speak.
20:59:11disruptekbtw, the cloud is a place with lots of casualities.
20:59:21disruptekplan accordingly.
21:02:31disruptekZevv: you're blowing my mind with this installLocal().
21:03:06disruptekyou define it in cps, after importing env, which is the only place it is used. via a name reference.
21:03:29disruptekis it me or are oreos way sweeter than they used to be?
21:05:11Zevvdisruptek: if it works, it aint stupid
21:05:25disrupteki know, it's just a little more than i can handle right now.
21:05:41disruptekdid i mention it's raining?
21:05:47Zevvno?
21:06:00Yardanicohe did
21:06:02Yardanicoa few hours ago
21:06:09disruptekget with the programm.
21:08:48FromDiscord<Clyybber> @Varriount Theres a mechanism for that
21:08:59FromDiscord<Clyybber> not sure if its possible to enable it via a flag though
21:09:01FromDiscord<Clyybber> Lemme check
21:11:25*vicfred joined #nim
21:11:37*dulsi_ joined #nim
21:13:15FromDiscord<Clyybber> theres vmopsDanger
21:14:30*dulsi quit (Ping timeout: 256 seconds)
21:18:51FromDiscord<exelotl> hey, I was wondering what's Nim's exception model like currently? I'm wrapping a DLL where every single function returns a possible error code. If I make it so that each function in my wrapper checks the return value and possibly throws an exception, should I expect a performance hit?
21:19:31disrupteknot under cpp or arc.
21:20:05disrupteki mean, it's hard to beat the speed of checking an int against a constant. πŸ˜‰
21:20:38disruptekit's basically free.
21:20:43Yardanico@exelotl with default GC and C backend it's setjmp
21:20:44FromDiscord<exelotl> yeah :(
21:20:51disruptekprobably most of the time it's truly, 100% free.
21:20:53YardanicoI think --exceptions:goto works for refc as well if you enable it
21:21:00Yardanicoand C++ backend uses native C++ exceptions
21:21:49ZevvI just roll my own exceptions, who needs language support
21:22:05disruptekheavy are the testes.
21:23:33Zevvmy tcp server is broken
21:24:49*dulsi_ is now known as dulsi
21:29:12Zevvdisruptek: I defined it in cps because it is actually used by the transformed code, which is the 'user' code
21:29:21ZevvI did similar tricks with npeg
21:29:47Zevvin this case a template is no different then a proc. You insert a call in the macro, so in the end the call is in the user code.
21:29:53Zevvwhich includes cps, and not encv
21:29:53Zevvenv
21:30:09disruptekyeah, i know, i'm just a little distracted.
21:30:17Zevvaren't we all
21:41:44*Trustable quit (Remote host closed the connection)
21:59:11*darrion_oakenbow joined #nim
22:03:07*wowi42 quit (Ping timeout: 240 seconds)
22:03:08*wowi42_ joined #nim
22:03:25disruptekdo you want a version that might work without checks?
22:08:30FromDiscord<reilly> Compiling a thing with the C++ backend, and running throws a "System Error" popup stating that `libgcc_s_dw2-1.dll` was not found. I found a StackOverflow thread, but at a glance it doesn't really appear to be helpful since I'm not using C++ directly.
22:09:41*solitudesf quit (Ping timeout: 256 seconds)
22:16:54FromDiscord<reilly> Thread in question: https://stackoverflow.com/questions/4702732/the-program-cant-start-because-libgcc-s-dw2-1-dll-is-missing
22:21:32*SenasOzys quit (Read error: Connection reset by peer)
22:51:36*maier joined #nim
22:55:31*rockcavera quit (*.net *.split)
22:55:31*MightyJoe quit (*.net *.split)
22:55:31*pulux quit (*.net *.split)
22:55:31*junland quit (*.net *.split)
22:55:32*antranigv quit (*.net *.split)
22:55:32*GaveUp quit (*.net *.split)
22:55:32*disbot quit (*.net *.split)
22:55:32*federico3 quit (*.net *.split)
22:55:32*shashlick quit (*.net *.split)
22:56:43*maier quit (Ping timeout: 256 seconds)
22:57:11*rockcavera joined #nim
22:57:11*pulux joined #nim
22:57:11*MightyJoe joined #nim
22:57:11*junland joined #nim
22:57:11*antranigv joined #nim
22:57:11*GaveUp joined #nim
22:57:11*disbot joined #nim
22:57:11*federico3 joined #nim
22:57:11*shashlick joined #nim
23:15:16*Guest53595 quit (Ping timeout: 256 seconds)
23:16:15*vicfred quit (Quit: Leaving)
23:16:35*vicfred joined #nim
23:17:17*dadada joined #nim
23:17:40*dadada is now known as Guest39052
23:29:57FromDiscord<flywind> I use `jmeter` to benchmark `asynchttpserver`(using `-d:danger`), I set 400 threads and 1s and the throughput is 13388/s. Is it right?
23:30:56FromDiscord<flywind> windows
23:34:02*einichi joined #nim
23:34:08disrupteki dunno, but it's not a very large number.
23:38:51FromDiscord<flywind> Nerver mind, at least my modification of `httpbeast`(based on `wepoll`) in windows is faster than `asynchttpserver`(600 threads + 1s) though I may be wrong. 😜
23:38:55FromDiscord<flywind> https://media.discordapp.net/attachments/371759389889003532/740353099692245023/unknown.png
23:54:30*krux02_ joined #nim
23:57:27*krux02 quit (Ping timeout: 260 seconds)