<< 07-07-2018 >>
00:00:10Araqpolitics is always #im-offtopic
00:00:58nasusiroAraq: true. I just wanted a quick review of what's going on in Austria, that's all
00:01:24nasusirobecause through news websites and TV, they speak gibberish that I don't understand
00:05:12nasusiroAraq: apart from Nim itself, what is the largest Nim project that people have reported it to you?
00:05:17Araqnasusiro, the law is only about allowing employees to work as much as 12 hours
00:05:30nasusiroAraq: ...are they mad or something?! O.o
00:05:42nasusiro8 hours is already exhausting, not 12!
00:06:12nasusiroeventually they will start paying people the same amount of money they would get with traditional 8 hours
00:06:21nasusiroit's like going back to 1936
00:06:28nasusiroanyway, that's the end of politics
00:06:30nasusiroback to Nim now
00:09:37*dddddd quit (Remote host closed the connection)
00:11:52Araqnasusiro, some computer algebra system with 300K loc, but it's closed source and I may misremember the number
00:12:31nasusiroI guess it gave you an incredible satisfaction when they shared this info with you :D
00:12:40Araqsome games around maybe 100K loc.
00:12:48Araq(closed source, sorry)
00:13:40nasusiroI don't mind whether they are closed source or opened
00:14:06Araqand one I'm eager to announce where they replaced their C++ server with Nim
00:14:27Araqwith a code reduction of a factor of 3 or 4.
00:15:43nasusiroeven though I'm a die-hard C++ lover, I do understand them for doing so
00:16:03nasusiroand let me tell you this: I "know" the basics of the basics
00:16:17nasusiroC++ must be the only language that cannot be taught completely
00:16:44nasusirobut so far with Nim I feel very comfortable learning it.
00:22:07*elrood quit (Quit: Leaving)
00:23:09*fjvallarino quit (Remote host closed the connection)
00:23:35*fjvallar_ joined #nim
00:27:45*fjvallar_ quit (Ping timeout: 248 seconds)
00:46:55FromGitter<kayabaNerve> Is it weird if I group ⏎ Nim, Elm, Elixir, Crystal ⏎ BUT ALSO D?
00:48:09Araqno.
00:48:27ldleworkWell Elm is pretty unique there
00:48:32Araqbut it's weird Rust is missing from your list
00:48:37nasusiroAraq: I wanted to ask. How come you chose Nimrod and then renamed it to Nim? What's up with Nimrod?
00:48:45Araqmarketing.
00:48:50nasusiromeaning?
00:48:54ldleworknimrods didn't understand nimrod
00:49:05FromGitter<kayabaNerve> Eh. Rust too if you want. It's my lift of... new languages I guess
00:49:07nasusiroI have no idea what Nimrod is
00:49:12FromGitter<kayabaNerve> Languages not accepted by the norm
00:49:17ldleworknasusiro: you nimrod
00:49:19ldlework:)
00:49:21FromGitter<kayabaNerve> Nimrod = Nim's original langauge or a fan of Nim
00:49:22FromGitter<kayabaNerve> Or an idiot
00:49:30FromGitter<kayabaNerve> *my group of
00:49:33ldleworkor a great hunter king
00:49:49nasusiroa hunter king?
00:50:15nasusirois it a character from a video game or something?
00:50:27ldleworkhttps://en.wikipedia.org/wiki/Nimrod
00:51:05nasusiroah I see
00:51:18nasusiroso what was the original objective of choosing this guy's name?
00:51:37nasusirothe rebellious concept?
00:51:48*icebattle quit (Quit: leaving)
00:51:55Araqking, godlike, built the tower of Babel, you get the idea
00:52:29AraqBabel a reference for Nim allowing every language concept to be expressed because macros
00:52:45nasusiroah OK
00:54:25FromGitter<rayman22201> No worse than Git as a name for a version control system lol
00:55:28nasusirorayman22201: trust me, there are worse things out there lol
00:55:29nasusirohttps://en.wikipedia.org/wiki/WTFPL
00:56:24FromGitter<rayman22201> lol. Exactly. That's what I mean.
00:57:04FromGitter<rayman22201> Though I agree with the marketing decision to shorten it to Nim
00:57:17nasusiroyes, Nim sounds better than Nimrod
00:57:25nasusiroNimrod sounds aggressive
00:58:42FromGitter<kayabaNerve> NIMROD
00:58:46FromGitter<kayabaNerve> NIMROD NIMROD NIMROD
00:58:53FromGitter<kayabaNerve> NIM ⏎ NIM NIM NIM
00:59:00FromGitter<kayabaNerve> Insult/Pokemon chirp?
00:59:08FromGitter<rayman22201> lol
00:59:58nasusiroit have been named MLP after My Little Pony!
01:01:59FromGitter<kayabaNerve> no
01:02:01FromGitter<kayabaNerve> no
01:02:01FromGitter<kayabaNerve> plz no
01:02:04FromGitter<kayabaNerve> ill do anything
01:02:12FromGitter<kayabaNerve> ill make the compiler 1.0 myself
01:02:13FromGitter<kayabaNerve> plz no
01:02:31*fjvallarino joined #nim
01:02:33nasusirothere you have it Araq, I have found you the ultimate volunteer for v1.0!
01:05:24FromGitter<rayman22201> Who knew it would be so easy? :-P
01:07:19*ashleyk_ joined #nim
01:07:51FromGitter<kayabaNerve> I didn't promise a good 1.0
01:08:35nasusiroit doesn't matter, v1.0 is on its way now ^_^
01:09:43*ashleyk__ quit (Ping timeout: 250 seconds)
01:10:48nasusiroanyway folks, I'm off for today. Keep up the good work everyone!
01:10:51*nasusiro quit (Quit: Quitting for now...)
01:13:44*BitPuffin quit (Remote host closed the connection)
01:17:08*ng0_ joined #nim
01:20:07*ng0 quit (Ping timeout: 250 seconds)
01:20:12*krux02 quit (Read error: Connection reset by peer)
02:48:33*leorize joined #nim
03:00:56*endragor joined #nim
04:03:19*endragor quit (Remote host closed the connection)
04:31:24*fjvallarino quit (Remote host closed the connection)
04:31:31*fjvallarino joined #nim
04:34:30Sargunrayman22201 I'm here.
04:44:48*lompik joined #nim
04:55:47*NimBot joined #nim
04:57:24FromGitter<rayman22201> Hi Sargun! Thanks for responding on your post :-)
05:06:11FromGitter<kayabaNerve> What's the best UI lib for Nim? Must be cross platform for the big 3 OSs. Windows, Linux, and BSD. ⏎ ⏎ Jokes on BSD. Mac OS. If it has BSD compatibility though...
05:12:37*fjvallarino quit (Remote host closed the connection)
05:13:05*fjvallarino joined #nim
05:16:49*miran joined #nim
05:18:00*fjvallarino quit (Ping timeout: 260 seconds)
05:32:13FromGitter<sclee15> You may for get the U
05:33:23FromGitter<sclee15> UI bindings for NIm
05:34:48FromGitter<citycide> @kayabaNerve nigui & nimx are probably the best ones. I like nigui but it's kind of stagnant
05:50:19*fjvallarino joined #nim
05:51:56*mal`` quit (Quit: Leaving)
06:04:31*mal`` joined #nim
06:25:55*xet7 quit (Quit: Leaving)
06:37:50FromGitter<bevo009> just tried nim secret after seeing earlier comments
06:38:04FromGitter<bevo009> seems completely broken, unless I'm missing a trick
06:38:48FromGitter<kayabaNerve> Is it just a Nim REPL?
06:39:08FromGitter<bevo009> apparently, but it just borks my console
06:39:30FromGitter<bevo009> how do you even exit without a restart?
06:40:30FromGitter<bevo009> maybe that's the secret :)
06:55:13FromGitter<kayabaNerve> Why is it named secret?
06:55:46FromGitter<Quelklef> that's the secret
06:57:47FromGitter<rayman22201> It's named secret because it's really broken and @Araq doesn't want people to use it lol
07:01:37FromGitter<bevo009> It worked like a year ago, mostly, apart from modules that depend on C code
07:01:40FromGitter<bevo009> https://github.com/nim-lang/Nim/issues/5886
07:02:01FromGitter<bevo009> completely broken now though
07:02:40FromGitter<bevo009> there's still iNim :)
07:17:21*lompik quit (Ping timeout: 248 seconds)
07:19:48*nasusiro joined #nim
07:22:38*nkr joined #nim
07:32:10*Trustable joined #nim
07:35:13FromGitter<Varriount> @bevo009 It works fine for the small snippets I throw at it.
07:44:33*nkr quit (Ping timeout: 248 seconds)
07:55:26*Trustable quit (Remote host closed the connection)
08:00:24FromGitter<bevo009> now it works here too
08:00:35FromGitter<bevo009> well that's weird
08:10:08FromGitter<bevo009> I've got another computer that iNim isn't installed on, I'll use nim secret instead...like being back at python, js, ruby etc
08:12:19nasusiroquestion: is this message considered a segfault? "SIGSEGV: Illegal storage access. (Attempt to read from nil?)"
08:16:17FromGitter<mratsim> yes
08:16:33FromGitter<mratsim> it’s when you try to dereference a nil pointer
08:17:27FromGitter<mratsim> If you are using ref objects, check that you properly initialize them with new result, before accessing its fields
08:17:56FromGitter<bevo009> same msg if you run .add on an uninitialized sequence
08:18:02FromGitter<mratsim> or use constructors ObjectType(field1: foo, field2: bar) it will implicitly initialize them for you
08:18:27FromGitter<mratsim> normally on devel seq and string are automatically not nil @bevo009
08:19:07FromGitter<mratsim> @Araq, I think you have to think about a deprecation strategy, seems like this RFC is quite popular: https://github.com/nim-lang/Nim/issues/8208
08:19:11FromGitter<bevo009> I'm still on last release
08:21:44nasusiroI managed to get that aforementioned error message by doing a silly thing inside my local nim.cfg
08:23:56FromGitter<bevo009> @mratsim So this part of the tutorial is changing? ⏎ "Sequence variables are initialized with nil. However, most sequence operations cannot deal with nil (leading to an exception being raised) for performance reasons. Thus one should use empty sequences @[] rather than nil as the empty value. But @[] creates a sequence object on the heap, so there is a trade-off to be made here."
08:23:56nasusiroall I did was to add the following line in it: nimcache=getCurrentDir()
08:24:01FromGitter<mratsim> is it crashing at compile-time or runtime? if compile-time it’s a bug in the compiler, probably related to the not nil change
08:24:20FromGitter<mratsim> @bevo009 yes
08:24:55FromGitter<mratsim> you can use newSeqUninitialized if you want.
08:25:32FromGitter<bevo009> so there wasn't much of a tradeoff really I take it?
08:25:46FromGitter<mratsim> @bevo009 seems like it’s still gated: https://github.com/nim-lang/Nim/blob/devel/changelog.md
08:26:58FromGitter<mratsim> the perf reason if that to support nil, each proc would need “if cpp.isNil: initialize(); doSomething”.
08:27:11FromGitter<bevo009> Thanks, I'll note that newSeqUninitialized down
08:27:33FromGitter<mratsim> in a tight loop, all those if statement would become a bottleneck
08:32:49nasusirocan someone tell me where to find the list of predefined .cfg variable names? All I know is $project_path
08:43:34Araqmratsim: with the fine grained .experimental feature, we have a nice path forward
08:43:59Araqwe can enable your solution with .experimental and then gradually use it everywhere and then make it the default
08:44:39FromGitter<mratsim> sounds good
08:44:59nasusirostill up Araq?
08:50:30FromGitter<bevo009> @mratsim so uninitialized string default will be "", empty string in 0.19.0? ⏎ are typetraits planned to merge into system, or is that rumour
08:53:54*nsf joined #nim
09:01:28*rosshadden quit (Read error: Connection reset by peer)
09:02:00dom96Spammers love to advertise cheap kitchens on our forum
09:03:38*rosshadden joined #nim
09:05:16nasusirodom96: lol. do they look good though?
09:05:42dom96cheapkitchens.uk.com
09:05:47dom96Judge for yourself :P
09:06:08nasusirothey look pretty decent I would say
09:06:33nasusiroand they charge quite cheap I have to admit
09:06:54dom96Woulnd't be surprised if it's a complete scam
09:07:02nasusirodom96: yeah, most probably
09:07:37nasusirodom96: by the way, is there any page where it lists all .cfg variables I could potentially use?
09:07:54dom96nim --help?
09:08:11nasusironope
09:08:19nasusirovariables of $project_path format
09:08:29dom96oh, no idea
09:08:36nasusiroI'm trying to set a different path for nimcache and cannot make it work
09:08:42nasusiroI even managed to crash nim lol
09:10:55nasusiroI have tried to add "nimcache=getCurrentDir()" in my local nim.cfg and crashed nim lol
09:11:13dom96report that as an issue
09:11:25nasusiroright away sir!
09:13:13nasusirodouble-checked it with a vanilla demo and reproduces the same behavior.
09:19:24Araqcan we also kill 'defer' please?
09:20:26dom96lol
09:20:27nasusiroI'm not good with terminology Araq
09:20:46Araqhttps://forum.nim-lang.org/t/4022
09:20:55nasusiroticket #8236 just got opened
09:21:40dom96fix defer?
09:22:25FromGitter<mratsim> yeah, fix defer
09:22:58Araqit's not fixable.
09:23:08Araqthe syntax doesn't have enough information.
09:23:26miranwhat are the advantages of defer, compared to try-finally?
09:23:54FromGitter<mratsim> are there cases when starting the try block at the same line as the defer statement doesn’t work?
09:24:07FromGitter<mratsim> @miran, it’s less noisy
09:24:42FromGitter<mratsim> you avoid indentation hell
09:24:44FromGitter<bevo009> @dom96 before the reaper forum got redone, they had a fake general discussion page one level above as a spamtrap...only the spammers posted in there
09:25:09dom96bevo009: nice.
09:25:13miran@mratsim doesn's seem like much of an advantage to me, to be honest
09:25:49nasusiroAraq: how does Nim's defer compare to that of Golang's?
09:26:04AraqI think it's exactly the same
09:26:08FromGitter<mratsim> Some people like indentation: https://github.com/lougehrig10/C-Deep-Reinforcement-Learning/blob/master/src/NeuralNetwork/neuralNet.c
09:27:03FromGitter<mratsim> also you put the defer right next to the acquiring the resource, i.e. you can train that as a muscle/reflex
09:27:05nasusiroAraq: then I guess your example in 4022 is wrong. defer should come first and then setupFoo() should follow
09:27:11FromGitter<mratsim> try/finally has more friction
09:27:46Araqmratsim: usually you would just use a destructor and avoid it altogether
09:28:12FromGitter<mratsim> Where can i find those famed beasts ?;)
09:28:26Araqand if a destructor can't be used, try-finally is good enough and clearer
09:28:58FromGitter<mratsim> I agree destructors are better
09:29:08FromGitter<mratsim> but will they be triggered on exceptions?
09:29:35Araqsure they will.
09:29:40dom96Maybe we should adopt the Python way (TM)?
09:29:44FromGitter<mratsim> great
09:29:59Araqthe Python way is a 'with' template
09:30:03dom96yes
09:30:13Araqwell sure
09:30:14dom96But we need that in the stdlib
09:30:19Araqcan add it to sugar.nim
09:30:30Araqbut I hate 'defer' now
09:30:32FromGitter<mratsim> it’s in the manual iirc.
09:30:52Araqwas hard to implement, it's still a bit buggy and leads to unclear and buggy code
09:31:09dom96I don't really care that much about it personally
09:31:19Araqand it was against Nim's design.
09:31:21dom96AFAIK async still doesn't support it
09:31:27Araqyep. :P
09:31:42FromGitter<mratsim> well just kill it, RFC on the tracker and then deprecate it
09:31:43Araqwhenever we do something against Nim's design, it bites us :P
09:32:07FromGitter<mratsim> “when we do something it bites us” FTFY
09:32:53FromGitter<mratsim> The Python way: https://nim-lang.org/docs/manual.html#templates-passing-a-code-block-to-a-template
09:34:04FromGitter<mratsim> but that supposes we have a concept of “open” and “close” for all resources you can acquire (files, socket, database …)
09:34:51Araqdestructors are that concept.
09:34:54dom96Araq: I'm going to fix c2nim and protect the main branch from pushes
09:34:54FromGitter<mratsim> since open would have different args, you can’t just use a generic template.
09:35:03FromGitter<mratsim> so destructors are just better
09:35:06Araqdom96: I fixed it yesterday
09:35:12Araqwhat's wrong with it?
09:35:19dom96No you didn't. The tests are still failing.
09:35:27Araqthey are green on Windows.
09:35:30dom96You've broken the tests and then left them broken for months
09:35:43Araqprobably some OS specific code in there that I need to fix
09:35:54AraqI updated the tests.
09:35:58Araqyesterday.
09:36:10dom96You've broken installation via Nimble, that's the problem
09:36:49Araqwell travis agrees with you
09:36:56Araqneed to make c2nim not OS specific
09:37:06dom96so from now on it's PRs always, and only when they pass they'll be mergeable
09:38:21Araqdoesn't work for c2nim
09:38:45Araqas the tests were already broken and looking at travis
09:38:55Araq --noBabelPath
09:39:06Araqthis stuff is old...
09:39:49dom96Not sure why that means it doesn't work for c2nim
09:39:58Araqit was red before
09:40:08AraqI create a PR
09:40:13Araqstill red and you reject it
09:40:35dom96yes, that's the point
09:40:58Araqok, but the real problem is that c2nim depends on compiler/renderer
09:41:08Araqand so changes to Nim can change c2nim's output
09:41:25dom96yes, which is why we pin the compiler commit hash
09:41:31dom96and why we use nimble to build it
09:41:41Araqah ok.
09:41:42dom96so we're not constantly after a moving target
09:41:53Araqalright then
09:41:54FromGitter<mratsim> @bevo009 the plan is to make not nil the default, but just monitor the changeling to see if it will be in next rel or not
09:41:57*Vladar joined #nim
09:43:46*noonien joined #nim
09:47:41*xet7 joined #nim
09:49:04Araqdom96: please update the compiler's hash then
09:49:19dom96Yes, I'm doing that
09:49:31Araqok thanks
09:49:56Araqbtw Nimble might have a new regression
09:50:35Araqyeah, a serious one
09:51:01Araq./src/nimblepkg/nimscriptsupport.nim
09:51:01Araq 80: if x.kind == nkPar and x.len == 2 and x[0].isStrLit and x[1].isStrLit:
09:51:21Araqthis needs to be x.kind in {nkTupleConstr, nkPar}
09:51:53dom96what issue does it cause?
09:51:56Araqin fact, I wonder why it worked at all... :-)
09:52:18AraqI think requires can also be used with tuples
09:52:25dom96https://travis-ci.org/nim-lang/c2nim/builds/401178838#L1870
09:52:31dom96Still seeing differences
09:52:39Araq ("packageName", "version")
09:53:16dom96meh, I don't really want that to work anyway
09:53:26dom96What's the point when `requires "packageName version"` works?
09:54:33Araqit's cleaner IMO.
09:54:57dom96The c2nim test-suite is incredibly annoying
09:55:04dom96locally it shows even more errors
09:55:44Araqthe hash computation is based on times.nim and times.nim was changed quite a bit
09:55:56Araqit's just bad, we need to change it
09:56:29dom96It's missing a bunch of empty lines too
09:56:30dom96locally
09:56:46AraqI think locally you just use the wrong compiler version
09:57:19Araqalso, sorry, but c2nim's test suite is not nearly as annoying as Nimble's :P
09:57:33AraqNimble tests take forever on my machine
10:00:48dom96ahh, it's because the tester uses c2nim from my path
10:00:59dom96That's a poor choice :P
10:02:02miransorry to interupt, but a quick question - is it expected that (closure) iterators are 2x slower than procs?
10:05:03dom96probably
10:05:09Araqmiran: that seems a good number, in fact
10:06:02miranok, thanks guys
10:08:56FromGitter<mratsim> I did a bench somewhere
10:09:05FromGitter<mratsim> it was worse than than on it
10:12:20mirani had no clue! ok, back to using procs :)
10:13:47*BitPuffin joined #nim
10:23:13FromGitter<mratsim> @retsyo I found something from CERN to process High Energy Physics abstract and automatically tag them: https://github.com/inspirehep/magpie
10:33:24*elrood joined #nim
10:39:33*miran quit (Ping timeout: 265 seconds)
10:43:27*yglukhov[i] joined #nim
10:47:11FromGitter<mratsim> I’m trying to parse/scrap webpages, why are there so many xml parsers? https://nim-lang.org/docs/parsexml.html, https://nim-lang.org/docs/xmltree.html, https://nim-lang.org/docs/xmlparser.html, https://nim-lang.org/docs/xmldomparser.html
10:48:23FromGitter<mratsim> oh and that: https://nim-lang.org/docs/xmldom.html
10:49:17Yardanico@mratsim usually you would use parsexml and xmltree
10:49:27Yardanicoand yeah, you can try to use nimquery for parsing via css selectors
10:50:37FromGitter<mratsim> for now I’m trying to understand with the minimum tool, that’s like a completely new domain for me so I’m approaching like I did for deep learning.
10:51:19FromGitter<mratsim> there is this as well for css selectors it seems: https://github.com/OpenSystemsLab/q.nim
10:51:33FromGitter<mratsim> but AFAIK CSS selectors are less powerful than XPath
10:52:51FromGitter<GULPF> If it's html you're parsing you should use htmlparser, not the XML ones
10:53:14Yardanico@mratsim but css selectors are easier to use than parsehtml :)
10:53:18FromGitter<mratsim> Yeah I’m looking there to try to make sense of this: https://nim-lang.org/docs/lib.html#pure-libraries-xml-processing
10:53:28FromGitter<mratsim> I don’t really care about easy :P
10:54:03FromGitter<GULPF> XPath is probably more powerful, but CSS selectors are more specialized for html
10:55:31FromGitter<mratsim> I see, I’ll check it out, since you’re the author of Nim query @GULPF, what is the state of your lib vs https://github.com/OpenSystemsLab/q.nim? (going AFK)
10:58:26*yglukhov[i] quit (Remote host closed the connection)
10:59:01*yglukhov[i] joined #nim
11:01:48FromGitter<GULPF> @mratsim Nimquery has support for more selectors and is more actively maintained. I subjectively like the API better as well, but that's no surprise since I implemented it :P
11:03:47*yglukhov[i] quit (Ping timeout: 256 seconds)
11:07:20dom96mratsim: You should use htmlparser
11:07:40dom96Alternatively if you want to do it properly: https://github.com/dom96/webdriver
11:08:04dom96You'll end up using a full browser but you'll get 100% compatibility with JS and all weird HTML quirks
11:14:07*BitPuffin quit (Remote host closed the connection)
11:27:39livcdWhat is the most usable ORM pkg out there atm ?
11:35:04*lompik joined #nim
11:39:44*yglukhov[i] joined #nim
11:40:27*BitPuffin joined #nim
11:56:37*yglukhov[i] quit (Remote host closed the connection)
11:56:50*pomdepin joined #nim
11:57:51*dorelix quit (Read error: Connection reset by peer)
11:58:29*dorelix joined #nim
12:00:42*thomasross_ joined #nim
12:01:39*xet7 quit (Quit: Leaving)
12:03:01*pomdepin quit (Quit: Page closed)
12:03:10*thomasross quit (Ping timeout: 265 seconds)
12:04:27*ng0_ is now known as ng0
12:05:38FromGitter<kaushalmodi> dom96: I agree with @mratsim though, are are too many similarly named stdlibs..
12:06:44FromGitter<kaushalmodi> Both parsexml and xmlparsers libs? Shouldn't they be merged? ( I have not looked what both do by the way. So merging them is a dumb idea, then may be name them differently?)
12:07:10FromGitter<kaushalmodi> *xmlparser
12:07:59*aziz joined #nim
12:12:23*nsf quit (Quit: WeeChat 2.1)
12:15:27*aziz quit (Ping timeout: 240 seconds)
12:19:36FromGitter<mratsim> @dom96 webdriver is cool as well
12:20:14FromGitter<Vindaar> @Araq @dom96 is `stmt` and `expr` finally removed from the language?
12:21:36FromGitter<mratsim> basically I’m exploring scraping because it’s the only way to get data at scale and being able to scrap say marketplaces or social networks and apply data science (image and text analysis) in an easy way would be a killer use case and I know several startups that would be interested.
12:26:02elroodjust don't forget to ask for permission if you're scraping at scale
12:34:04FromGitter<mratsim> Isn’t that described in robots.txt?
12:34:35FromGitter<mratsim> I’ve raised https://github.com/nim-lang/Nim/issues/8238 for the html/xml parser issues.
12:38:10FromGitter<Vindaar> ok, I'm really confused. e.g. `SomeReal` is now also throwing `undeclared identifier` errors, instead of deprecated hints. Is that an issue with deprecation or what happened? Getting the same thing on 2 computers
12:38:48FromGitter<mratsim> someone deleted it too early? :p
12:38:58FromGitter<Vindaar> It just broke arraymancer for me :P
12:49:43FromGitter<Vindaar> https://github.com/mratsim/Arraymancer/pull/256
12:49:52ng0I'd simply write a crawler+scraper which respects the codex beyond robots.txt. good user-agent, etc
12:56:49FromGitter<mratsim> @Vindaar Nim is broken here, SomeFloat doesn’t exist on stable so I can’t merge your PR.
12:57:01FromGitter<Vindaar> oh, forgot about that
12:57:31FromGitter<Vindaar> well, it can stay there and merge it later or close it. :)
12:57:49FromGitter<mratsim> I can merge it when 0.19 stable is out.
12:57:55FromGitter<Vindaar> :)
12:58:10FromGitter<Vindaar> but you're seeing the same behavior now?
12:58:17FromGitter<mratsim> but since it changes everything I will have to redo the PR anyway if I add new stuff so we’ll see.
12:58:28FromGitter<Vindaar> yeah
12:58:37FromGitter<mratsim> I didn’t check. I’m trying not too update too often because it breaks work :P
12:58:50FromGitter<Vindaar> haha, good point
13:01:11FromGitter<mratsim> My guess is that there was a plan to remove the {.deprecated: TFooType, Footype.} Pragma and SomeReal was caught in that
13:01:30*miran joined #nim
13:01:47FromGitter<Vindaar> nope, found it: ⏎ https://github.com/nim-lang/Nim/commit/c7298561c199255facc562402700322360408179
13:01:58FromGitter<mratsim> @Araq ^
13:04:05FromGitter<kaushalmodi> @Vindaar You can do this for SomeFloat support on older Nim: https://github.com/kaushalmodi/strfmt/commit/8701480cc9eac5e04389a392bf50bb2962016e7e
13:05:09FromGitter<mratsim> @Vindaar https://github.com/nim-lang/Nim/pull/8239
13:07:41FromGitter<Vindaar> @kaushalmodi thanks, yes, that's a possibility. :) ⏎ @mratsim ok. If Araq decides to not merge your PR for some reason, I'll add your suggestion to my PR
13:12:10*smt quit (Ping timeout: 256 seconds)
13:14:03*sz0 joined #nim
13:14:27dom96Okay, definitely looks like a mistaken removal
13:14:29dom96I'll merge it
13:14:38FromGitter<Vindaar> :)
13:16:51*dddddd joined #nim
13:40:04FromGitter<mratsim> btw @dom96, when you merge a PR with just “merge”, the history is noisier because there is your “Merge foo from branch bar”, while when using squash and merge, you have a single commit. For example my single line change added my commit + your merge: https://github.com/nim-lang/Nim/commits/devel
13:42:13Araqhuh?
13:42:20Araqwhy was that re-added?
13:42:29Araqgot deprecated just in this cycle?
13:42:30*nkr joined #nim
13:42:39FromGitter<mratsim> because 0.18 doesn’t have SomeFloat
13:42:41FromGitter<mratsim> yes
13:44:32FromGitter<kaushalmodi> Araq: That's why I suggest mandating the version to be added with the deprecated pragma. Using changelog to record such stuff is not efficient/scriptable
13:45:09AraqI agree but .deprecated with a list of deprecations is its own beast
13:45:26Araqand needs to be deprecated :-)
13:46:05*fourouge quit (Ping timeout: 245 seconds)
13:50:41Araqwe could have added a comment though
13:53:31*yglukhov[i] joined #nim
13:58:01*yglukhov[i] quit (Ping timeout: 260 seconds)
14:00:17*xet7 joined #nim
14:24:08dom96mratsim: a merge commit provides a link to the PR, that's useful information and I don't consider it "noise"
14:24:30dom96GitHub really needs a "Merge when tests pass" button
14:25:14FromGitter<mratsim> squash and merge also adds the link to the PR in parenthesis
14:30:00*Trustable joined #nim
14:30:26FromGitter<mratsim> squash and merge also adds the link to the PR in parenthesis
14:31:19elrooda programmer could utilize git hooks and not depend on someone else implementing a feature button
14:31:27FromGitter<mratsim> Interesting: https://github.com/nim-lang/Nim/commit/04b3f8f24aa0a39d66fcce174c377e523f382a3a, https://github.com/nim-lang/Nim/commit/9b31f6785947974e89d46bbc0dee11f1c78754dc
14:40:56miranis there a way for nim vscode plugin (nimsuggest?) to notify about unused imports?
14:44:22*xet7 quit (Quit: Leaving)
14:44:43*nkr quit (Quit: Leaving)
15:01:07dom96oh cool. My HN submission made it to the front page.
15:02:54FromGitter<mratsim> congrats
15:03:44nasusirodom96: nice dissertation project mate. Well done +1
15:03:49dom96Thanks :)
15:04:43*fjvallarino quit (Remote host closed the connection)
15:04:50*fjvallarino joined #nim
15:07:03nasusirodom96: how come you haven't the PDF link anywhere on the website? Maybe under an FAQ section?
15:07:12FromGitter<mratsim> this proves that you are only shadow-banned for Nim stuff :P
15:07:33dom96nasusiro: Because I put that website together in 4 hours, a day before the deadline ;)
15:07:48dom96mratsim: hah, totally
15:07:52nasusiroso you just graduated?
15:08:09dom96yep, a week ago to be precise
15:08:18nasusirowelcome to the real world mate
15:09:13dom96Won't be able to pay less for Amazon Prime and Spotify for much longer :(
15:09:58nasusirodon't you worry; NASA will hire you lol
15:11:17mirandom96: chosing simple fizzbuzz and C++ produces an errror
15:11:46dom96That's because it's not C++ code?
15:12:42miranoh, i thought i could choose anything from the drop down menus as an example
15:14:21elroodcongrats, dom
15:20:25dom96miran: oh no, each example is for a specific lang
15:20:42miranthanks for the clarification
15:39:17*crem quit (Ping timeout: 255 seconds)
15:41:02*crem joined #nim
15:48:42federico3nasusiro: plot twist: people don't want to work for NASA
16:03:55*qwargl joined #nim
16:04:35*endragor joined #nim
16:08:57*endragor quit (Ping timeout: 245 seconds)
16:10:30qwargli am trying to create bindings to a library called soundtouch. i followed the "Binding to C Libraries"-guide, created a .nim using c2nim, and it compiles also. But when i try to call a function, i get: Error: internal error: (filename: "ccgcalls.nim", line: 423) ... nim version is 0.18. (more details: https://pastebin.com/G3Rt6JnQ) Any hints? And yes, i am a beginner with this.
16:11:41FromGitter<mratsim> internal error should be reported, they are compiler bugs. Though in that case there is probably an issue in your code as well
16:12:26FromGitter<mratsim> can you remove the @ in soundtouch_getVersionString(@)
16:13:17FromGitter<mratsim> the @ passes all arguments passed to the Nim function to the C++ library but here you have no arguments
16:14:33qwarglsame thing
16:17:18qwargli added information about the SoundTouchDLL.h: https://pastebin.com/cBDGdCRP
16:18:35*sz0 quit (Quit: Connection closed for inactivity)
16:19:02dom96Can you paste your full .nim code?
16:19:15FromGitter<mratsim> use gist for example
16:19:51*gb00s quit (Quit: Ping timeout (120 seconds))
16:20:09*BitPuffin quit (Ping timeout: 264 seconds)
16:20:27*gb00s joined #nim
16:20:33*miran quit (Remote host closed the connection)
16:20:51qwargldom96: https://pastebin.com/ePArDGuz
16:21:04*miran joined #nim
16:21:10FromGitter<mratsim> Does c2nim supports const char return values? https://gitlab.com/soundtouch/soundtouch/blob/master/source/SoundTouchDLL/SoundTouchDLL.h#L76
16:21:51qwarglDo u mean gist instead of soundtouch?
16:22:19FromGitter<mratsim> I mean to post the Nim code, seems like Pastebin doesn’t highlight properly
16:22:57dom96okay, I can reproduce
16:27:04dom96https://gist.github.com/dom96/047accf5fcf024045e6f00760de15401
16:27:07dom96Here is a minimal repro
16:28:01dom96really weird
16:28:08dom96.importc works
16:28:12dom96.importcpp doesn't
16:30:12qwarglhttps://gist.github.com/marcus-wishes/ec3b99592b1a80f7e9f1fc69e3adfe15 (its the SoundTouchDLL.nim as gist)
16:30:24shashlickhaving some trouble with https://github.com/nim-lang/Nim/pull/8232
16:30:47shashlickcan someone try `echo formatBiggestFloat(1234.567, ffDecimal, 0, '.')` and see what output they get on #head?
16:31:23shashlickstrutils had a test which checked output with "1235." which didn't work on my machines which output "1235" (without decimal)
16:31:30shashlickbut appveyor test fails
16:32:44shashlickaah - okay I figured it out - `nim c` outputs "1235." whereas `nim js` outputs "1235" without decimal
16:33:41shashlickwill open another bug
16:37:30dom96qwargl: I created an issue for this: https://github.com/nim-lang/Nim/issues/8241
16:37:43dom96I suppose you can change `importcpp` to `importc` for these as a workaround
16:38:28dom96I'm really puzzled by this though. Araq, are you there?
16:42:00*nsf joined #nim
16:43:03qwargldom96: i am trying. so far i get this error: could not import: soundtouch_getVersionString()
16:43:17qwarglwhen executing the compiled binary
16:48:36dom96Remove the parenthesis
16:52:22qwarglmain.nim(17, 6) Error: type mismatch: got <proc (): cstring{.cdecl.}>
16:53:16FromGitter<mratsim> we don’t have enough info, what is the rest, there should be a line “Expected …"
16:53:30qwarglmain.nim(17, 6) Error: type mismatch: got <proc (): cstring{.cdecl.}>
16:53:32qwarglbut expected one of:
16:53:33qwarglproc `$`(x: string): string
16:53:35qwarglproc `$`(s: WideCString): string
16:53:36qwarglproc `$`[T: tuple |
16:53:38qwargl object](x: T): string
16:53:39qwarglproc `$`(x: uint64): string
16:53:41qwarglproc `$`(x: int64): string
16:53:42qwarglproc `$`[T, IDX](x: array[IDX, T]): string
16:53:44qwarglproc `$`[Enum: enum](x: Enum): string
16:53:45qwarglproc `$`(w: WideCString; estimate: int; replacement: int = 0x0000FFFD): string
16:53:47qwarglproc `$`[T](x: set[T]): string
16:53:48qwarglproc `$`[T](x: seq[T]): string
16:53:50qwarglproc `$`(x: int): string
16:53:51qwarglproc `$`(x: cstring): string
16:53:53qwarglproc `$`(x: bool): string
16:53:54qwarglproc `$`(x: float): string
16:53:56qwarglproc `$`(x: char): string
16:53:57qwarglproc `$`(err: OSErrorCode): string
16:53:59qwarglsorry!
16:54:31FromGitter<mratsim> you called echo soundtouch_getVersionString without parenthesis, it should be echo soundtouch_getVersionString()
16:55:05FromGitter<rayman22201> @dom96 Congrats on your thesis getting HN front paged! I'm assuming it's written in Nim? Just remember to not forget about us when you get your fancy job at FB :-P
16:55:12qwarglthats what i did before when i got this: error: could not import: soundtouch_getVersionString()
16:55:22FromGitter<mratsim> So: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5b40f07a3572e970c17e277e]
16:55:33FromGitter<mratsim> without parenthesis in the importcpp
16:55:39FromGitter<mratsim> with parenthesis for the echo
16:59:10qwarglyay, works
17:20:55FromGitter<tim-st> can I write async downloaded data to a file?
17:20:59FromGitter<tim-st> let html = client.getContent(url) ⏎ ⏎ ```let f = open(filename, fmWrite) ⏎ f.write(html) ⏎ f.close()``` [https://gitter.im/nim-lang/Nim?at=5b40f67b3d8f71623d64ab6b]
17:21:04FromGitter<tim-st> doesnt work
17:23:25FromGitter<tim-st> oh, I forgot `await`, it works now
17:28:49FromGitter<tim-st> This async thing really needs hard fix, I manually need to shutdown pc...
17:30:18dom96hrm?
17:32:37FromGitter<tim-st> when I replaced httpclient by asynchttpclient it didnt write a single html file to my pc, instead it ate all in RAM and the process couldnt be killed by filemanager
17:32:48FromGitter<tim-st> long pressing pc shutdown fixed it
17:33:14dom96what was your code?
17:33:25FromGitter<tim-st> I think the same from httpclient page
17:33:38FromGitter<tim-st> one moment
17:34:14FromGitter<tim-st> My hope was that asynchttpclient can download multiple files instead of one per time, is that the intention?
17:35:00dom96sure, but you need to create multiple instances of AsyncHttpClient
17:35:17FromGitter<tim-st> ok, is there any example in the docs?
17:35:47FromGitter<tim-st> yes, I had a shared client
17:36:36FromGitter<tim-st> will try again with new clients for each call
17:38:10FromGitter<tim-st> at least it wasnt interuptable by ctrl+c when using async, I think this is a bug
17:39:09FromGitter<tim-st> same again, needs shutdown^^
17:42:31FromGitter<tim-st> https://gist.github.com/tim-st/57f9a197366a893feddc7a3b6b6d48bf
17:44:25*miran quit (Ping timeout: 256 seconds)
17:45:37*Hotkeys joined #nim
18:03:19m712is there an up to date LLVM binding for nim?
18:03:22m712llvm 6/7
18:06:26elrood7 is still two months away, there isn't even a release candidate yet
18:09:04elroodyou could have a look at nlvm
18:10:14dom96First time properly trying Docker. Definitely impressed.
18:11:01m712okay
18:11:12m712i always found docker overly convoluted
18:11:26FromGitter<tim-st> I updated my example code for easier testing, does someone know how to actually do async requests?
18:14:26*qwargl quit (Quit: Konversation terminated!)
18:16:53dom96I just set up a container for my obfuscator in like 2 hours.
18:17:41dom96And it was pretty smooth going, I only wish it took up less HD space
18:24:28Araqdom96: well importcpp with '(' is a pattern and patterns require # or @
18:24:57Araqand probably we need to redesign this a bit :-(
18:25:27Araqit should really be 'import' and 'importPattern'
18:25:42Araqand these patterns are not C++ specific
18:25:52Araqand shouldn't trigger C++ compilations
18:26:10Araq(though we also make C++ the default target anyway)
18:30:38shashlickseeing random build failures
18:31:07shashlicktests\method\tmultim2.nim is crashing on #head
18:31:44shashlickpasses on Windows, crashes on Linux
18:33:42FromGitter<data-man> @Araq: https://github.com/nim-lang/c2nim/pull/130 tests are green now
18:33:59shashlicktmultim2.nim(17, 8) Error: internal error: genMagicExpr: mTypeTrait
18:34:05shashlickunable to generate stack trace even with nim_temp
18:34:39Araqshashlick: it's green here :-)
18:35:12shashlickhmm, failing on two PRs
18:35:37shashlick#8240 and #8232
18:38:14shashlicknope, refreshed #head and it still crashes for me on Ubuntu
18:51:57*dorelix quit (Ping timeout: 240 seconds)
18:53:20*noonien quit (Quit: Connection closed for inactivity)
19:00:15FromGitter<tim-st> does the httpclient automatically replace spaces by `%20 `? it seems only these failed
19:07:09dom96Araq: Happens without '()' too
19:09:49*thomasross_ is now known as thomasross
19:24:15*nasusiro quit (Quit: Quitting for now...)
19:42:16*yglukhov[i] joined #nim
19:47:09*yglukhov[i] quit (Ping timeout: 264 seconds)
19:56:31dom96Wow. Somebody just PWND me well.
20:00:09m712dom96: what do you mean?
20:00:29dom96Someone managed to find a security vulnerability in my obfuscator web app :)
20:00:41FromGitter<kayabaNerve> dom96: What app?
20:01:04dom96https://picheta.me/obfuscator
20:02:08shashlickwhat did they do
20:02:35FromGitter<kayabaNerve> ^^
20:02:39FromGitter<kayabaNerve> What's the vuln
20:03:31livcdlol
20:04:00dom96They POSTed: {"code":"...","language":"`bash code here`","rename":true}
20:04:19dom96My app took the `language` and appended it to the filename as the extension
20:04:37dom96But since I didn't escape the ` the bash code was executed
20:04:52dom96Was fun investigating this :D
20:05:03livcdhow did you find out you got pwnd ?
20:06:16elroodas long as your graduation doesn't get revoked for that glaring oversight..
20:07:05dom96hah, like I said: I put this together in 4 hours
20:07:12dom96I should have reviewed the code before posting on HN
20:07:30dom96Anyway, here is how I noticed it: https://i.imgur.com/zL9VXv5.png
20:07:39dom96Basically `systemctl status obfuscator`
20:08:19dom96And I saw the `python ...` and thought "huh, that's odd"
20:08:19livcdi dont get it
20:08:22livcdthe python code ?
20:08:33dom96soon to be followed by "oh shit"
20:08:36dom96yep
20:08:55livcddid you look at the ip ?
20:09:29dom96hrm, let me reup that image without my VPS' IP
20:10:17livcdah :D
20:11:12dom96https://i.imgur.com/AdTjyYt.png
20:11:25FromGitter<kayabaNerve> What did the python do?
20:11:44dom96Set up a reverse SSH connection
20:12:00dom96The attacker then added an SSH key and SSH'd normally
20:12:07FromGitter<kayabaNerve> What did they then do
20:12:28dom96Good question
20:12:37dom96I see nothing in the .bash_history
20:12:41livcdbut they sshd into a docker container right ?
20:12:43dom96So I'm not entirely sure
20:12:55Hotkeysthey responsibly reported the vulnerability and did no damage? :D
20:13:12dom96livcd: Sadly no :)
20:13:34livcddom96: uh how is that possible ?
20:13:54FromGitter<kayabaNerve> dom96: I would've done that... totally... not setup a cryptocurrency miner...
20:14:18dom96livcd: because the web app isn't running in a Docker container
20:14:24Hotkeyscan't imagine you'd mine that much crypto from a vps :P
20:14:40livcddid you rebuild your VPS ?
20:15:11FromGitter<kayabaNerve> livcd: He saved a file to the disc. They mangled the file name to execute bash. This bash allowed a foreign server to send commands to be executed. This is known as a reverse SSH sessions, as it's a shell connection started by the host server.
20:15:31FromGitter<kayabaNerve> Then they enabled themselves to use the standard SSH.
20:15:44FromGitter<kayabaNerve> Or are you asking how something else is possible
20:15:57livcdyeah but i saw docker and thought they would just be able to get into the docker container
20:16:00livcdbut whatever
20:16:01FromGitter<kayabaNerve> Ah
20:16:20FromGitter<kayabaNerve> Hotkeys depends on the service
20:17:20*yglukhov[i] joined #nim
20:17:29dom96hrm, .bashrc was modified
20:17:33dom96but I don't see what changed
20:17:36FromGitter<kayabaNerve> dun dun dunnnnn
20:17:36*mwbrown quit (Ping timeout: 268 seconds)
20:18:00dom96lol
20:18:18*mwbrown joined #nim
20:18:50dom96strangely, .bash_history isn't being populated
20:19:01livcdthat's an old trick
20:19:04FromGitter<kayabaNerve> ps aux?
20:19:05FromGitter<mratsim> Welcome to the world of botnet. 1B computer are celebrating the new joiner
20:19:23FromGitter<mratsim> if you prepend a comment with a space it’s hidden from history
20:19:30FromGitter<mratsim> a command*
20:19:30FromGitter<kayabaNerve> Check all the running processes? Update your secret files sych as an API key?
20:19:36FromGitter<kayabaNerve> Reboot, check it again?
20:19:52FromGitter<kayabaNerve> Check cron and other time based services?
20:20:09dom96I diffed .bashrc with /etc/skel/.bashrc and see no changes
20:20:21FromGitter<mratsim> and .bash_profile?
20:20:41livcdafaik you can do smh like this with no tty allocation ssh -T user@host /bin/bash -i
20:21:06dom96The person had a TTY and I was trying to redirect it somewhere but couldn't :(
20:21:15FromGitter<mratsim> he might be running in a screen or tmux too
20:21:58*yglukhov[i] quit (Ping timeout: 265 seconds)
20:22:17FromGitter<kayabaNerve> What if you ask for a new IP from the service?
20:22:39livcddom96: you should rebuild your VPS
20:22:53dom96yeah, I will
20:23:50FromGitter<kayabaNerve> How important/good is it to use the raises pragma?
20:23:53livcdyou might have just changed shell and unset HISTFILE that usually does the trick
20:24:00shashlickdo we already have scripts in nim-lang/nim to update a cloned devel view?
20:24:02livcdif you change your shell you wont get logged
20:24:08livcdhe*
20:24:14dom96livcd: Sounds like you have experience :P
20:24:27livcdphrack 64, issue 4
20:24:34FromGitter<kayabaNerve> And if I possibly throw an Exception not in raises, the compiler will stop, right?
20:24:47Araqright
20:25:02FromGitter<kayabaNerve> Thanks Araq. So then it's just how important/good is it to use?
20:25:05livcdi just read lots of stuff that i might never need
20:26:39dom96anything else I should check?
20:27:12livcdif he cared enough to cover his traces you should just rebuild immediatelly
20:28:13livcdjust assume the worst case scenario and that everything is compromised. Do you have some sensitive data on that vps ? ssh keys etc ?
20:29:35dom96Yeah, I revoked those
20:29:51livcdso umh you did ??
20:29:52livcd:D
20:30:04FromGitter<kayabaNerve> livcd is the hacker
20:30:07FromGitter<kayabaNerve> That's how he knows
20:30:13dom96lol
20:30:19FromGitter<kayabaNerve> I think my project is a bit stuck. The TODO list is almost out. Before I add in, say, the UI tasks, I'm working on cleaning up the code base. That's why I'm asking about raises. Anyone else have any good suggestions?
20:31:12livcddom96: yeah anyway just get rid of that vps, any git repos you cloned there, any mirrors etc
20:31:15livcdnuke it
20:31:22dom96But ... downtime :(
20:31:49dom96well, actually, I can rebuild it without shutting this one down
20:32:03livcdwell at least you learned a lesson :)
20:32:05FromGitter<kayabaNerve> Rebuild then transfer the DNS
20:32:08FromGitter<kayabaNerve> Then update it
20:32:12livcdyou are famous enough to get hacked just by posting to HN
20:32:14FromGitter<kayabaNerve> *delete it
20:33:21dom96Yeah, I've got cloudflare too so it should be painless
20:33:31dom96Need my pizza to arrive first though :P
20:33:35livcdi would check for the ips of the attacker of the nim forum registered users if you can etc
20:34:06dom96Pretty sure the IP is a droplet
20:34:26dom96The attacker was smart to do it through a droplet, not from their own PC
20:34:37dom96I feel like I'm a character from the movie Hackers
20:35:20*brainproxy quit (Quit: WeeChat 2.1)
20:35:37livcdah DO droplet ?
20:35:49livcdthen just contact DO maybe they will give you an identity of the attacker
20:36:08livcdmight have been just dude that tried to just poke around
20:36:10livcdstill unpleasant
20:40:34dom96Educational experience though
20:42:36livcdit would scare me tbh
20:43:29livcdwe dont have a torrent lib right ?
20:44:21*[ZED] joined #nim
20:44:53*[ZED] left #nim ("Leaving")
20:44:54FromGitter<Varriount> dom96: You didn't put the app in a docker container?
20:45:04dom96The whole app? no
20:45:31shashlicklivcd: could wrap one of the c libs
20:45:48livcdshashlick: i have never done it but i guess it's possible right?
20:45:57FromGitter<Varriount> Hrm, why doesn't Github use the "recently updated" sorting of issues/pull requests by default?
20:45:58livcdDo nim apps behave correctly in docker containers ?
20:46:08dom96sure, why wouldn't they?
20:46:57livcdCrystal apps were suffering from huge performance hits
20:48:20FromGitter<Varriount> Why?
20:48:44livcdI do not know the reason and if it has been fixed already
20:48:55shashlicklivcd: I could wrap one if you have some time to wait
20:49:34FromGitter<Varriount> livcd: https://github.com/crystal-lang/crystal/issues/5725
20:49:39shashlickam currently working on ensuring nimgen wrappers work from 0.17.0 upto #head
20:50:03FromGitter<Varriount> shashlick: <3 Nimgen. Makes maintaining wrappers so much easier
20:50:25livcdshashlick: that's cool!
20:53:16shashlickvarriount: do you have any wrappers you've made? i want to add them to the nimgen test matrix so that i can ensure something doesn't break as the tool evolves
20:53:53FromGitter<Varriount> shashlick: I have a toy one I made for linenoise, however I think it's broken at the moment.
20:54:05FromGitter<Varriount> (Not due to Nimgen, due to me messing around with it)
20:54:50shashlickcool, let me know when its on github and i'll add it
20:57:39*lompik quit (Ping timeout: 256 seconds)
21:03:57*yglukhov[i] joined #nim
21:06:12FromGitter<tim-st> I have a download running since two hours using httpclient and ssl and the ram grows slowly for unknown reason (nothing is hold in memory for a long time), it's now at 700mb, is this a "memory leak"?
21:06:59FromGitter<tim-st> or is this memory growth "normal"
21:08:49*yglukhov[i] quit (Ping timeout: 260 seconds)
21:10:14FromGitter<tim-st> hm, maybe I should call `client.close` the examples in the docs arent good
21:13:20*PMunch joined #nim
21:14:16*fjvallarino quit (Remote host closed the connection)
21:14:43*fjvallarino joined #nim
21:19:33*fjvallarino quit (Ping timeout: 264 seconds)
21:23:59FromGitter<tim-st> yes, calling close makes it stay constant at 14mb
21:26:17*Vladar quit (Quit: Leaving)
21:28:27*Trustable quit (Remote host closed the connection)
21:36:32*Jesin joined #nim
21:41:54*rockcavera quit (Remote host closed the connection)
21:45:35*crem quit (Ping timeout: 255 seconds)
21:46:36*crem joined #nim
21:53:45shashlickhow do you test your app/project against multiple nim versions? do you use any tool to automate?
21:56:35dom96choosenim :P
21:59:01shashlickya i'm already using choosenim, but something to automate picking a version and doing nimble install && nimble test
21:59:12shashlicki have a bat script but would like something cross-platform
22:22:02shashlickcool so now all nimgen wrappers are tested and working from 0.17.0 onwards
22:26:25PMunchshashlick, a nim script?
22:27:26shashlicklooks like that's the best option
22:27:47*bevo quit (Quit: Page closed)
22:30:00PMunchBut how do you test the script for different version?!
22:31:16shashlickwell, i find all versions installed by choosenim, set each version and run specified tests
22:31:21shashlickcould do the same with nims
22:31:26shashlickinstead of a batch file
22:33:45FromGitter<Varriount> Looks like we need a tox-like application for Nim
22:33:45PMunchI was just joking :P
22:34:25shashlick🙂
22:34:34shashlickvarriount: would be nice, yes
22:34:48PMunchIf you test your Nim code with Nim, how do you test the testing script? :P
22:35:37shashlickyou could use the testing script to test the testing script
22:35:38*elrood quit (Quit: Leaving)
22:37:53*mwbrown quit (Ping timeout: 248 seconds)
22:53:40FromDiscord<exelotl> is it planned to have myNumbers.sort() work without having to pass system.cmp[int] ?
22:56:32PMunchHmm, is there something special I need to do in order to get the gc working in a dll?
22:56:47PMunchI get this: http://ix.io/1go5
22:58:24PMunchpimolayout is compiled with "nim c --app:lib --noMain --out:dynlibs/liblayout.so -d:use_pkg_config_static --debugger:native pimolayout.nim" and pimo.nim is compiled with: "nim c --passL:-rdynamic --debugger:native pimo.nim"
23:04:57PMunchAh wait, this is because I'm not using nimrtl isn't it..
23:11:21*fjvallarino joined #nim
23:16:52*rockcavera joined #nim
23:19:46*nsf quit (Quit: WeeChat 2.1)
23:20:07FromDiscord<exelotl> I'm loving `/` as a syntactic sugar to join file paths in a cross platform way
23:20:37FromDiscord<exelotl> I wasn't really digging `&` for string concatenation but this makes up for it xD
23:23:02*crem quit (Ping timeout: 256 seconds)
23:24:59*crem joined #nim
23:36:15PMunchHmm, Nim really doesn't like to load libraries..
23:36:17*rockcavera quit (Remote host closed the connection)
23:45:52*find0x90 joined #nim
23:49:05FromGitter<kayabaNerve> @Varriount I have one planned
23:49:07FromGitter<kayabaNerve> Not over Nim
23:49:14FromGitter<kayabaNerve> Just a secure decentralized chat
23:49:33FromGitter<kayabaNerve> I can't remember my issue with Tox but Keybase is centralized. It's secure, but a single point of failure.
23:50:25*ng0 quit (Quit: Alexa, when is the end of world?)
23:53:25FromGitter<Varriount> @kayabaNerve I'm referring to https://pypi.org/project/tox/
23:59:18PMunchI keep getting illegal storage access from initTable..
23:59:20PMunchLine 340