<< 21-09-2018 >>
00:07:37AlexMaxhrm
00:08:07AlexMaxc2nim seems to turn *current = *current*10.0f; into current[] = current * 10.0[]
00:13:28AlexMaxHow do you get the ascii value of a number in Nim?
00:13:32AlexMaxerm
00:13:35AlexMaxof a letter
00:15:16AlexMaxI suppose I could cast it to an integer, but maybe the standard library has something nicer?
00:19:53FromGitter<Quelklef> `ord`
00:19:54FromGitter<Quelklef> probably
00:20:53FromGitter<kayabaNerve> ord() or int()
00:25:46zachcarter gogolxdong: I'm sure I have some code lying around somwhere that does this - let me try to find something
00:26:10*Perkol joined #nim
00:31:42zachcarterwell - I guess not...
00:33:38zachcartergogolxdong: I haven't done any work with nuklear in quite some time - and I'm not an expert on the bindings (I never really was). You may be able to glean some insight from the GUI code in frag / zengine - but no guarantees
00:33:57zachcarterhttps://github.com/vurtun/nuklear/blob/27603d5b97bb010a58dada53105f430b78da0749/nuklear.h#L20928 - I think is the C code you're calling
00:34:02zachcarterso I'd start by looking at that
00:37:39AlexMaxFromGitter: Thanks , not sure why I didn't just try that
00:37:40FromGitterAlexMax, I'm a bot, *bleep, bloop*. I relay messages between here and https://gitter.im/nim-lang/Nim
00:37:52AlexMax@quelklef, what I just said :P
00:38:14AlexMaxman, this sample C program does a lot of raw C string manipulation :P
00:44:41*zachcarter quit (Quit: Lost terminal)
00:46:45AlexMaxman, trying to cast an array to a cstring is giving me a problem, heh
00:50:07*dddddd quit (Remote host closed the connection)
00:58:27*abm quit (Read error: Connection reset by peer)
01:16:23FromGitter<Quelklef> Any way to circumvent the default behavior that getters are ignored in the same module as the declared type?
01:16:26FromGitter<Quelklef> Sorry, setters*
01:36:03AlexMaxhrm, there's no built in way to convert a cstring to a string?
01:36:56FromGitter<kayabaNerve> AlexMax $
01:37:27AlexMax...Oh
01:37:42AlexMaxI should've known that :P
01:47:41FromGitter<gogolxdong> @zacharycarter C code works, not sure whether the binding is right ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba44dbdb9531f2dfa99b2ad]
01:48:32FromGitter<gogolxdong> string implicitly convert to cstring?
01:54:21FromGitter<gogolxdong> or the casting from InputFilter which takes the first parameter as var text_edit to plugin_filter which taks it as ptr text_edit? object cast object to pointer?
01:54:58FromGitter<gogolxdong> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba44f71f4bd1056ac933291]
02:02:35FromGitter<gogolxdong> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba4513b0cfe7f30f1c589ba]
02:03:34FromGitter<gogolxdong> ah, it takes address , the main concern is on string impilictly converted to cstring.
02:05:36FromGitter<gogolxdong> conversion from string to cstring is automatic?
02:06:02FromGitter<kayabaNerve> string -> cstring = automatic ⏎ cstring -> string = `$`
02:08:29FromGitter<gogolxdong> yes, then it's an real issue .
02:08:46*tefter quit (Remote host closed the connection)
02:08:58FromGitter<gogolxdong> a
02:28:37*Perkol quit (Quit: Leaving)
03:03:16FromGitter<gogolxdong> hints Error: undeclared identifier: 'await' .
03:11:37FromGitter<gogolxdong> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba461695df5194734edfbde]
03:16:36Tangergogolxdong: You can't use await outside of a proc with the {.async.} pragma
03:16:48TangerTry waitFor
03:16:50Tangerinstead
03:17:02TangerOr handle the futures directly
03:30:01FromGitter<gogolxdong> thanks for reminding, wouldn't the async module be rewritten?
03:43:46TangerUmmm, I recall one of the async stdlibs going towards being deprecated, but I can't remember which one, sorry
03:55:34*Yardanico quit (Ping timeout: 240 seconds)
03:56:54*gsingh93 quit (Ping timeout: 264 seconds)
03:58:35*Yardanico joined #nim
03:59:43*gsingh93 joined #nim
04:13:00*miran joined #nim
04:50:49FromGitter<iffy> I'm trying to compile a c program using a nim-build statically linked library and I'm getting: : fatal error: 'nimbase.h' file not found
04:51:05FromGitter<iffy> where do I find that file? Or how do I resolve the bug?
04:51:39*jxy quit (Ping timeout: 252 seconds)
04:52:10*jxy joined #nim
04:55:16*erratic quit (Ping timeout: 246 seconds)
05:00:10Tangeriffy: What are you running to comp\ile it?
05:02:07FromGitter<iffy> Tanger: I added `-Ipath/to/Nim/lib` and now I'm getting a different error (probably not anything to do with Nim)? `gcc -o test_cprog -I/Users/matt/lib/Nim//lib/ -Isqlite3 -Inimcache libbuckets.a test.c`
05:02:31FromGitter<iffy> New error ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba47b67913ba7799b1fd99e]
05:03:06FromGitter<iffy> does that mean I compiled libsqlite3.a incorrectly? or libbuckets.a?
05:04:24FromGitter<iffy> oh :) it means I didn't include `libsqlite3.a` in my `gcc` command
05:04:40TangerHaha
05:04:54TangerFrom what I can gather, nimbase.h is in the csources installed by koch when building nim
05:05:57TangerSo it should be in the nim installation directory
05:06:30TangerSo mayeb it needs extra args to link against those headers?
05:08:16FromGitter<iffy> Yeah, `-I/Users/matt/lib/Nim/lib/` did the trick to fix that. So, is there a way to build my statically linked library (from nim code) which includes the statically linked libsqlite3.a?
05:14:07*erratic joined #nim
05:14:12TangerNothing solid except for a couple of SO articles. Maybe try this suggestion iffy: https://stackoverflow.com/questions/27078657/cannot-link-libsqlite3-so-in-gcc
05:23:07*leorize quit (Ping timeout: 250 seconds)
05:31:52*icebattl1 joined #nim
05:34:05*nsf joined #nim
05:34:46*icebattle quit (Ping timeout: 272 seconds)
05:43:03*SenasOzys quit (Remote host closed the connection)
05:43:37*SenasOzys joined #nim
05:44:18*nif joined #nim
05:50:21*leorize joined #nim
05:58:16*miran quit (Ping timeout: 246 seconds)
06:10:36*mhnoyes7 joined #nim
06:12:32*mhnoyes7 quit (Remote host closed the connection)
06:13:31FromGitter<gogolxdong> what to pass for pointer of `proc write*(a1: cint, a2: pointer, a3: int): int {.importc, header: "<unistd.h>".}`
06:19:03*zcu15 joined #nim
06:21:38*zcu15 quit (Remote host closed the connection)
06:28:36*SenasOzys quit (Ping timeout: 272 seconds)
06:30:50*xet7 joined #nim
06:32:33FromGitter<gogolxdong> where goes wrong of this snippet, cannot write buffer to memory ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba49081b4990c30eee818c6]
06:34:04Tangergogolxdong: SIGSEGV error?
06:37:17FromGitter<gogolxdong> not, content should be 'butte' from the 5th char, but it's random hex.
06:41:10FromGitter<gogolxdong> or this ⏎ ⏎ ```code paste, see link``` ⏎ ⏎ be `butterfly` [https://gitter.im/nim-lang/Nim?at=5ba49285e5c2cc56adbf6831]
06:42:03*TheLemonMan joined #nim
06:46:14TangerOooh
06:46:28Tangergogolxdong: I think your problem is `var writeBuff`
06:46:38TangerI reckon that's auto typing to a string
06:46:51TangerIf you get the addr of a string, you get both the length and data parts of the string
06:47:11TangerTry either defining it as a cstring or casting it to cstring
06:47:57TangerI'm pretty sure a cstring is just a ptr[Char] or something similar, so you can probs write it in without `addr`
06:51:24TangerCan also try casting content to a string instead of a char array, as it would be written in as a string
06:52:46*Electrux joined #nim
06:54:07FromGitter<gogolxdong> great, works.
06:55:13FromGitter<gogolxdong> Is this dangerous as writing physical memory directly.
06:57:13FromGitter<gogolxdong> write to memory through driver is more efficient than system call ,right?
06:57:21TangerYeah, more effecient
06:57:26TangerBut yes, less safe'
07:05:18FromGitter<gogolxdong> don't know where it writes the buffer.
07:09:50TangerAs in where in /dev/mem?
07:09:59TangerOr are you talking about the var buffer?
07:10:24FromGitter<gogolxdong> former
07:12:17TheLemonManmratsim, I guess you tried setting minSamples to something like 1000 and were wondering why it never finished, right? :)
07:15:08Tangergogolxdong: As posix.read() returns a cint file descriptor, if you need to track where you're writing to, you might be better off using moveMem or copyMem
07:15:18TangerStill both unsafe, but they return pointers to memory
07:18:39FromGitter<rayman22201> @gogolxdong this is a very dangerous thing you are doing. What are you trying to accomplish? /dev/mem is not just the virtual adress space for your program. It's the full physical ram space of the computer. writing random bits to /dev/mem is a notorious way to crash your system.
07:26:02FromGitter<gogolxdong> thanks for reminding , aware of the potential effect a little, exploring out of curiosity.
07:29:38FromGitter<rayman22201> /dev/mem is typically only used in special cases such as writing a driver for a physical device or something. Some PCI devices map to a specific absolute ram address for example.
07:30:00*navin joined #nim
07:30:57*Electrux quit (Quit: Quit)
07:31:45*Electrux joined #nim
07:33:25FromGitter<rayman22201> The fastest way to write to ram for a normal program is probably mmap
07:42:08*NimBot joined #nim
07:52:35*Electrux quit (Ping timeout: 252 seconds)
07:55:32*Electrux joined #nim
07:57:08*jolo joined #nim
07:58:36*PMunch joined #nim
08:00:32TangerHey guys, can you define your own NimNodeKinds?
08:00:52TangerLike, could you designate something as a distinct type of a node that expectKind can match?
08:01:12TheLemonManyou can pass expectKind a set of nnkNode...
08:02:10TangerAh, and it'll pass if it's in that set?
08:02:26Tanger*return true
08:02:53TheLemonManyep
08:03:20*gmpreussner quit (Quit: kthxbye)
08:03:21TangerAwesome, thanks!
08:03:35*gmpreussner joined #nim
08:08:01*Vladar joined #nim
08:08:39FromGitter<xmonader> @kaushalmodi i think i'll just do it myself in the end
08:10:22TangerHey guys, can anybody explain this example from docs/macros? var res = findChild(n, it.kind == nnkPostfix and it.basename.ident == toNimIdent"foo")
08:10:42Tangerit.kind == nnkPostfix and it.basename.ident == toNimIdent"foo" is a `cond: untyped`, I'm just curious where the `it` is scoped from? Or if it's a separate lib
08:11:31TheLemonMananaphoric macros, sequtils follows the same convention
08:12:15TheLemonManlong story short, the `it` is injected by `findChild` in the user-supplied expression
08:18:55*BenjaminDeuter joined #nim
08:20:44*BenjaminDeuter quit (Remote host closed the connection)
08:24:32*Electrux quit (Read error: Connection reset by peer)
08:25:06*navin quit (Remote host closed the connection)
08:26:55TangerAyy, thanks again TheLemonMan! Killing it this arvo XD
08:27:21*Electrux joined #nim
08:30:18*navin joined #nim
08:55:23*krux02 joined #nim
09:03:48*leorize quit (Ping timeout: 245 seconds)
09:03:51*PMunch quit (Ping timeout: 260 seconds)
09:10:34TheLemonManhmm, isn't openFileStream supposed to be available on stable?
09:17:52FromGitter<mratsim> no
09:17:58FromGitter<mratsim> it has been added after :P
09:18:12FromGitter<mratsim> like 2 days after
09:18:23FromGitter<mratsim> but the documentation was generated 5 days after stable
09:18:35FromGitter<mratsim> so it pops up but isn’t available
09:18:45TheLemonMan_fuckin_ awesome, thanks mratsim
09:19:39FromGitter<mratsim> I think there is a bug or forum post about that already @Araq ^
09:22:00FromGitter<gogolxdong> @Tanger , is it possible to read the memory addressed with the fd opened in posix?
09:22:31FromGitter<gogolxdong> memory address taken by the fd.
09:22:43TheLemonManmratsim, I've just finished adding a way to export the raw benchmark data to json for comparison/post-processing/plotting
09:23:13FromGitter<mratsim> nice =)
09:23:17FromGitter<gogolxdong> it's only cint file descriptor now.
09:23:45FromGitter<mratsim> @gogolxdong I think there is a mmap or memmap module in Nim
09:24:35shashlickhttps://packnback.github.io/blog/programming_languages/
09:24:47FromGitter<gogolxdong> it's the user space memory
09:26:42TheLemonManshashlick, the other day I've finally learned what a shashlick was and now I'm hungry every time I see your nick
09:27:09FromGitter<gogolxdong> also system call context switching is expensive.
09:27:44shashlickThat wasn't the intention but an interesting result nonetheless
09:49:56FromGitter<gogolxdong> wish we can manage memory manually.
09:50:20*navin quit (Remote host closed the connection)
09:51:19TheLemonMan...just use C then?
09:54:18FromGitter<gogolxdong> at this level ,maybe yes.
10:00:08Araqyou can manage memory manually all you want in Nim, there is 'ptr UncheckedArray', alloc, dealloc, cast
10:00:18Araqand the memfiles module for an mmap wrapper
10:01:58*TheLemonMan quit (Quit: "It's now safe to turn off your computer.")
10:03:53FromGitter<gogolxdong> Deeper than that , wonder anyone tried , writing everything via /dev/mem in memory to manipulate machine manually.
10:06:27*djural_18 joined #nim
10:08:11*Electrux quit (Read error: Connection reset by peer)
10:08:51*djural_18 quit (Remote host closed the connection)
10:11:15*navin joined #nim
10:12:11*Electrux joined #nim
10:16:44*PMunch joined #nim
10:19:04FromGitter<gogolxdong> http://highscalability.com/blog/2013/5/13/the-secret-to-10-million-concurrent-connections-the-kernel-i.html
10:20:03*dddddd joined #nim
10:20:57*navin quit (Remote host closed the connection)
10:22:20*navin joined #nim
10:22:26*bytesighs3 joined #nim
10:24:44*Fervor joined #nim
10:26:31*jolo quit (Ping timeout: 256 seconds)
10:27:08*bytesighs3 quit (Remote host closed the connection)
10:27:13*Fervor quit (Killed (Sigyn (Spam is off topic on freenode.)))
10:30:08PMunchHmm, how can I turn on overflow errors in release builds?
10:34:59*refi649 joined #nim
10:39:41*refi649 quit (Remote host closed the connection)
10:41:21*Ven`` joined #nim
10:49:54*Ven`` quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
10:53:28*navin quit (Remote host closed the connection)
10:53:50*navin joined #nim
10:57:10*noonien joined #nim
10:59:46*ErnestG_29 joined #nim
11:00:37*ErnestG_29 quit (Remote host closed the connection)
11:05:14*stefanos82 joined #nim
11:20:41*xet7 quit (Remote host closed the connection)
11:21:06FromGitter<gogolxdong> like how to get the PCI-E Graphic Card address and write something to screen?
11:21:36FromGitter<mratsim> you should use OpenGL or Vulkan or a framebuffer for that @gogolxdong
11:22:06FromGitter<mratsim> `--overflowChecks:on` @PMunch
11:23:28*xet7 joined #nim
11:23:43PMunchAh thanks
11:25:02*navin quit (Remote host closed the connection)
11:25:27*navin joined #nim
11:25:57FromGitter<gogolxdong> It's possible because of nuklear-nim except some issues
11:26:02PMunchHmm, wouldn't it make sense for "try: expect:" to wrap itself in the correct pragmas for the caught exceptions?
11:28:17FromGitter<gogolxdong> I want to know more about memory. I wrote helloworld in memory and it stays at 0x0000000
11:29:45FromGitter<gogolxdong> need to find the address application mapped to.
11:30:19*dgwana joined #nim
11:30:59FromGitter<mratsim> I used to play with directfb or fbdev, you can try: https://www.linuxquestions.org/questions/programming-9/hello-world-or-display-pixels-on-framebuffer-fb-h-4175539854/
11:31:36*dgwana quit (Client Quit)
11:31:39FromGitter<mratsim> the last post. you need access to your framebuffer device like /dev/fb0
11:32:09*dgwana joined #nim
11:36:17*leorize joined #nim
11:37:37*navin quit (Remote host closed the connection)
11:38:40*navin joined #nim
11:56:35FromGitter<gogolxdong> marked.
12:00:26*navin quit (Remote host closed the connection)
12:01:38FromGitter<gogolxdong> the last post doesn't compile.
12:04:28*navin joined #nim
12:16:21*navin quit (Remote host closed the connection)
12:38:41*PMunch quit (Quit: Leaving)
12:41:48*gangstacat quit (Quit: Ĝis!)
12:49:38*pnorman24 joined #nim
12:50:29*pnorman24 quit (Remote host closed the connection)
12:59:06*gangstacat joined #nim
13:00:46*navin joined #nim
13:05:10*navin quit (Ping timeout: 244 seconds)
13:07:50*dgwana quit (Quit: Leaving)
13:17:47FromGitter<kaushalmodi> PMunch: Someday I'd like to write a parser for Org mode. Can you point to a parsing library that serves as the best example if a parser written in Nim?
13:30:26*Diamondcite25 joined #nim
13:31:15FromGitter<mratsim> json from the stdlib or from Araq’s packedJson
13:34:01pigmejyou can also take look into: https://github.com/OpenSystemsLab/jsmn.nim but it's rather hardcore parser (still json though)
13:36:26*SenasOzys joined #nim
13:38:16*ehmry quit (Quit: WeeChat 1.7)
13:38:32*Diamondcite25 quit (Ping timeout: 252 seconds)
13:38:39FromGitter<kaushalmodi> Thanks both, I'll take a look.
13:38:50FromGitter<kaushalmodi> pig
13:39:02FromGitter<kaushalmodi> Sorry, fat fingers on phone
13:39:47FromGitter<kaushalmodi> pigmej: What do you mean by hardcore parser?
13:40:59FromGitter<kaushalmodi> General question to all.. what does a parser ideally return? JSON?
13:40:59pigmejby using it as library :)
13:41:24pigmejstdlib / Araq one return pretty good things, JsonNode
13:41:42pigmejthe one from my link returns tokens
13:42:02pigmejwhich makes it kinda counter-intuitive to use (for me obviously)
13:42:13FromGitter<kaushalmodi> Ok, thanks. Will need to make a mental model and plan before I venture out into writing this parser.
13:44:14FromGitter<mratsim> Ideally a parser returns a syntax tree, that’s the easier to work with.
13:45:01FromGitter<mratsim> if you return tokens you’re a lexer
13:46:02FromGitter<kaushalmodi> But I cannot wrap my head around on how to make a syntax tree "tangible"..
13:46:23FromGitter<kaushalmodi> With JSON output, I can visualize the parsed elements on stdout
13:46:35FromGitter<kaushalmodi> And any other tool can pick up from there
13:46:41FromGitter<mratsim> the first thing you need to write is a pretty-printer then
13:47:08FromGitter<mratsim> or just convert your tree to json and hand it over to a json pretty printer
13:48:21FromGitter<kaushalmodi> Hmm, so the syntax tree is like the primary parsed internal database.
13:48:51FromGitter<kaushalmodi> And JSON output is just one of the outputs
13:50:35FromGitter<mratsim> Nim abstract syntax tree (parsed by macros) can be converted to Lisp or Json for example
13:51:04FromGitter<mratsim> it’s just an easy way to represent structured data (code being structured data)
13:51:19FromGitter<kaushalmodi> Ah, making circle back to macros. I will eventually have to understand those. :)
13:51:29FromGitter<mratsim> but a text editor can also use it to detect indentation or brackets and implement code-folding
13:51:36FromGitter<kaushalmodi> And the macros can output Lisp ?!
13:51:59FromGitter<mratsim> https://nim-lang.org/docs/macros.html#dumpLisp.m,untyped
13:52:30FromGitter<kaushalmodi> Thanks! That might be the key for me to start tinkering with macros :)
13:52:59FromGitter<mratsim> weirdest reason ever :P
13:54:50*brimonk9 joined #nim
13:58:27*brimonk9 quit (Remote host closed the connection)
14:12:42*agk23 joined #nim
14:13:25*agk23 quit (K-Lined)
14:23:04*TheLemonMan joined #nim
14:30:33*navin joined #nim
14:31:41pigmejthese kind of bugs make me sad: https://github.com/nim-lang/Nim/issues/9024 :(
14:34:44*Senketsu quit (Quit: WeeChat 2.2)
14:35:03FromGitter<mratsim> Does https://github.com/status-im/nim-asyncdispatch2 have the same behaviour?
14:35:25FromGitter<mratsim> (it was rewritten from the ground up so hopefully it doesn’t have the same bugs)
14:38:35FromDiscord<Shield> How can I make a proc type that returns itself? example: a callback that returns another callback of the same type
14:41:01*Electrux quit (Ping timeout: 260 seconds)
14:41:19*Electrux joined #nim
14:43:43*miran joined #nim
14:53:00*floppydh quit (Quit: WeeChat 2.2)
14:54:07*vivus joined #nim
14:55:23pigmejso ad2 had similar behaviour
14:55:26pigmejnot sure if not even worse
14:55:29pigmejdeadlocks when socket got closed
14:55:32pigmejnot closing socket when close called etc
14:55:55vivuswhich nim library allows me to execute commands from a file/string? I found reference to nimscript, but im making sure there isnt something else
14:56:11TheLemonManwhat kind of "commands" ?
14:56:51TheLemonManShield, you can probably try with a `auto` return type
14:57:04TheLemonManbut if you can't write down the type then it's probably a bad idea to boot
14:57:50vivusTheLemonMan: like "ls -la"
14:58:14*ctracey11 joined #nim
14:59:10TheLemonMancheck out the osproc module
14:59:26TheLemonManand don't parse `ls`!
14:59:53FromGitter<kaushalmodi> "don't parse ls!" - +1
14:59:53FromGitter<mratsim> @Shield, you can start hacking around that `type Callback[T] = proc (foo: T): Callback[T] {.closure.}`
15:00:49FromGitter<mratsim> don’t forget to kill nimsuggest and nim from time to time :troll: the compiler/suggestions might get into stack overflow while you’re developing
15:01:19pigmejkrux02: may I suggest something? You should split your pr into pieces, because in current form it may endup like never ending story
15:01:32*ctracey11 quit (Remote host closed the connection)
15:01:50FromGitter<kaushalmodi> @krux02: I agree with pigmej
15:02:03FromGitter<kaushalmodi> that last commit fixed that specific issue. So open a new PR for that
15:02:12FromGitter<kaushalmodi> if that PR relies on unmerged PR, mention so in the comments
15:02:29krux02I would agree with that
15:02:42krux02but yuuta yamada isn't maintaining at this point anymore.
15:02:47pigmejbecause you did serious fixes
15:02:57krux02and splitting up the PR into several independent commits is just a lot of pain
15:03:00FromGitter<kaushalmodi> krux02: it's for posterity
15:03:04pigmejkrux02: well
15:03:09pigmejyou didn't mention that you finished your PR
15:03:20FromGitter<kaushalmodi> the PR's and separate commits will document the fixes in progression
15:03:20pigmejyou stated: `This is an on going PR.`
15:03:26pigmejso everyone assumed that it's not yet done
15:03:36FromGitter<kaushalmodi> if you squash that humongous PR, all the info would be lost
15:03:40FromGitter<kaushalmodi> the rationale, etc
15:03:52pigmejI think all but last commit are fine there
15:04:07pigmejjust say that it's ready and create separate for #202
15:04:10FromDiscord<Shield> auto will hide the compilation error but will cause "Error: internal error: mapType"
15:04:24vivusTheLemonMan: what does "don't parse ls" mean?
15:04:45FromGitter<kaushalmodi> https://unix.stackexchange.com/questions/128985/why-not-parse-ls
15:05:07FromGitter<kaushalmodi> vivus: it's fine if you want to just print the ls output
15:05:20FromGitter<kaushalmodi> but don't post-process that ls output to do anything else
15:05:35vivusI just want to take a command from a string and execute it on the commandline
15:06:00krux02pigmej, well I think the last commit is fine in that PR, because it is part of the same refactoring works that fixed an issue on the way.
15:06:13pigmejok, can you then at least rephase PR ?
15:06:18FromGitter<kaushalmodi> vivus: For that, look at osprocs as TheLemonMan said
15:06:19pigmejto say that it's ready?
15:06:32pigmejI'll take short look, and let's ping yuuta
15:06:34vivusalright ty
15:06:35krux02I already had the refactorings done, saw the issue and integrated it. Was around 15 seconds of work
15:06:43pigmejif no response, then I'll merge it
15:07:09pigmejI somehow still have merge rights even after few years..
15:07:15krux02yes it is ready
15:07:34FromDiscord<Shield> is there a way to achieve it? basically I want something like "callback = callback()" where callback holds a proc that can returns the same proc or another one of the same type
15:07:34*Electrux quit (Read error: Connection reset by peer)
15:07:42FromGitter<kaushalmodi> vivus: specific to ls, if you want a list of files the Nim way, look at https://nim-lang.org/docs/os.html#walkFiles.i,string
15:07:49*Electrux joined #nim
15:08:18vivus@kaushalmodi I want to execute different commands from a file. Think of it like ansible
15:08:30FromGitter<kaushalmodi> vivus: And to run a random shell command and return its output as string, look at https://nim-lang.org/docs/osproc.html#execProcess,string,openArray[string],StringTableRef,set[ProcessOption]
15:08:56FromGitter<kaushalmodi> vivus: Well, I don't know what ansible is :) But hopefully the above helps.
15:09:03FromGitter<mratsim> @Shield did you see my message?
15:09:06vivusty i think it will
15:11:58krux02pigmej, I updated the message
15:11:59FromDiscord<Shield> oh sorry i missed it, trying it now
15:13:04krux02pigmej, I am sorry that I forgot to update the PR.
15:14:11*stefanos82_ joined #nim
15:14:28FromDiscord<Shield> wew it just crashes
15:14:37*stefanos82_ quit (Client Quit)
15:16:21*stefanos82 quit (Ping timeout: 252 seconds)
15:16:28FromGitter<mratsim> figures :P
15:17:04FromGitter<mratsim> I can probably makes something work but I don’t have time at the moment. Try opening a forum post. I’ll check when I have more time
15:17:26pigmejkrux02: np :)
15:22:12*stefanos82 joined #nim
15:24:23*Guest42462 joined #nim
15:26:07*Guest42462 quit (Remote host closed the connection)
15:30:21*icebattl1 quit (Ping timeout: 244 seconds)
15:33:10*icebattle joined #nim
15:55:56Araqhttps://github.com/nim-lang/nightlies/releases/tag/t1 testers required
15:56:44*Tariul joined #nim
16:06:21*Senketsu joined #nim
16:16:39TheLemonManAraq, works fine here, noice
16:17:07*Senketsu quit (Quit: WeeChat 2.2)
16:23:57FromGitter<iffy> I'm porting a Node.js program to Nim (it's that awful pseudo-line-parsing HTTP log thing from yesterday) and I'm surprised to find that Nim takes 13 seconds while the node version takes 3 seconds. Same 256MB input file. (And the node script is currently doing more than the Nim, too). Should this be surprising? I could come up with a shareable example case if it's surprising enough to pursue.
16:24:21TheLemonMandid you build it with -d:release ?
16:24:25FromGitter<iffy> yep
16:24:32*vivus quit (Remote host closed the connection)
16:24:51pigmejhow are you parsing that file?
16:25:33FromGitter<iffy> line by line in both cases (Node's `readline` module and a simplified `readLine` in Nim): ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba51b7c33da0f649e329928]
16:26:10FromGitter<iffy> Let me make a shareable direct comparison so that I'm sure I'm comparing apples to apples
16:26:36*tewinget0 joined #nim
16:26:38*tewinget0 quit (Remote host closed the connection)
16:28:06*Tariul quit (Remote host closed the connection)
16:29:41pigmejwhy not readLine from fileStream?
16:33:40FromGitter<iffy> pigmej: this one? https://nim-lang.org/docs/streams.html#readLine,Stream,TaintedString
16:34:37*Electrux quit (Quit: Quit)
16:34:45FromGitter<iffy> it doesn't handle \0 bytes as I was hoping
16:37:42*miran quit (Ping timeout: 264 seconds)
16:39:02*Gaasmann24 joined #nim
16:40:28*navin_ joined #nim
16:41:27*Gaasmann24 quit (Remote host closed the connection)
16:43:53*navin quit (Ping timeout: 252 seconds)
17:02:24FromGitter<rayman22201> I have a feeling that Node is using fgets, which is much faster on Unix than getting one char at a time. that's what Python does anyway: https://hg.python.org/cpython/file/d77db57982a1/Objects/fileobject.c#l1207
17:02:35FromGitter<rayman22201> http://www.cplusplus.com/reference/cstdio/fgets/
17:02:39*wildlander joined #nim
17:03:51FromGitter<rayman22201> I'm curious if that is the issue. If it is, It may be worth doing a pr to the Nim readline implementation to use that optimization on unix as well?
17:03:56*Piraty left #nim ("--")
17:04:53*Trustable joined #nim
17:07:25TheLemonMan"Note: This is not very efficient." yeah
17:08:43*nsf quit (Quit: WeeChat 2.2)
17:09:27*martinda1414 joined #nim
17:09:52federico3iffy: strace the two applications
17:12:45*martinda1414 quit (Remote host closed the connection)
17:14:48FromGitter<iffy> https://gist.github.com/iffy/ede3fb5b637b34566ecd36c447995c75
17:15:11FromGitter<iffy> Let me know if you see the same behavior as I'm seeing ^
17:15:21*dago joined #nim
17:16:33FromGitter<iffy> Node has its own definition of "line" that doesn't agree with Python or `wc`. Nim's built-in `readLine` also doesn't agree with Python or `wc`, nor does it agree with Node. My custom `readLine` agrees with Python and `wc` but is slow.
17:16:52FromGitter<iffy> (although no slower than the builtin `readLine`)
17:17:42FromGitter<iffy> oh, it consumes 200MB to run the test, btw
17:19:17*dago quit (Killed (Sigyn (Spam is off topic on freenode.)))
17:25:21FromGitter<iffy> TheLemonMan: I saw that note on `readLine(s:Stream)`; does it apply to the one I linked, too?
17:27:23FromGitter<iffy> federico3: I'm on a mac and I'm struggling to get `dtruss` to work
17:29:49TheLemonManholy fuck, that jungle of callbacks is nasty
17:31:11FromGitter<iffy> Yeah, not completely straightforward (I'm trying to make each program do roughly the same thing as what I'm actually doing and the same as each other)
17:32:19FromGitter<kaushalmodi> Araq: Tested the t1 release, works great.. extracted, ran sh install.sh and found nim binary installed in bin/. 👍
17:32:53FromGitter<kaushalmodi> only one thing ..
17:33:16FromGitter<kaushalmodi> while extracting (I use `dtrx` for that), I got these warnings (on RHEL 6.8): ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba52b5cd655361f76334291]
17:34:18FromGitter<iffy> Shall I file an issue? Is that the process?
17:34:54TheLemonMankaushalmodi, https://superuser.com/questions/318809/linux-os-x-tar-incompatibility-tarballs-created-on-os-x-give-errors-when-unt
17:35:09*darithorn joined #nim
17:35:19FromGitter<kaushalmodi> yes, functionally there was no issue
17:35:34FromGitter<kaushalmodi> but a user seeing these warnings wouldn't get a comfortable feeling
17:36:12FromGitter<kaushalmodi> so.. this was built on a macOS?
17:36:23FromGitter<kaushalmodi> from that su answer: ⏎ ⏎ > Install GNU tar if you can on Mac OS and use that to create the tar
17:37:40TheLemonManthe tarball is created here: https://github.com/nim-lang/nightlies/blob/master/.travis.yml
17:38:46FromGitter<kaushalmodi> it's not tarred in that travis, it seems
17:38:56FromGitter<kaushalmodi> I see that gnu-tar is installed
17:39:00FromGitter<kaushalmodi> but where is it used?
17:39:46FromGitter<kaushalmodi> ok.. need to look in `koch xz`
17:39:58FromGitter<kaushalmodi> that is probably using plain `tar` and not `gnu-tar`
17:42:04TheLemonManstrace shows the node version uses a lot of pread64 calls to load the whole file into memory
17:43:03TheLemonManscrew that, I misread the pread64 manpage
17:43:09FromGitter<kaushalmodi> the .xz is created in koch.nim here: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba52dad5df5194734f2f417]
17:43:15FromGitter<kaushalmodi> I find that very cryptic
17:43:29TheLemonManit uses a big chunk size (65535 vs 4096)
17:45:04FromGitter<rayman22201> @iffy maybe. I want to see if fgets fixes it first. I'm trying it out in a minute.
17:46:44FromGitter<mratsim> mixin not working properly in generics proc is annoying :/ that prevents defining “rand” proc later :/
17:46:51TheLemonMankaushalmodi, look at niminst.nim
17:47:28FromGitter<kaushalmodi> TheLemonMan: will have a look.. though what is `$#` in that exec?
17:47:32krux02mratsim: where is mixin not working?
17:47:37FromGitter<tim-st> @iffy maybe the python implementation is better. My assumption for python is that it uses a buffered stream then parses on this stream for lines and refills buffer when needed, nim doesnt seem to buffer
17:49:04FromGitter<tim-st> maybe setting `bufSize` makes a difference: https://nim-lang.org/docs/streams.html#newFileStream,string,FileMode,int
17:49:05FromGitter<mratsim> @krux02, here, I’d like to use mixin rand: https://github.com/SimonDanisch/julia-challenge/pull/3/files#diff-7f0523acf6522d8908cac4b96d98dfa5R40
17:50:05FromGitter<mratsim> but you actually already know the issue, it’s this: https://github.com/nim-lang/Nim/issues/6387
17:50:15FromGitter<mratsim> (I think)
17:50:29FromGitter<kaushalmodi> TheLemonMan: from niminst.nim, looks like `tar` is used if `gtar` is not found
17:50:40*Trustable quit (Remote host closed the connection)
17:50:41FromGitter<Vindaar> @kaushalmodi `$#` is one possible placeholder for `strutils.%`, see https://nim-lang.org/docs/strutils.html#%,string,openArray[string]
17:50:44FromGitter<kaushalmodi> Araq: Is `gtar` available on Travis
17:51:00FromGitter<mratsim> The use case is to define `type Point3 = object; x, y, z: float32` later in bench
17:51:14FromGitter<mratsim> and a `rand` proc on those later as well
17:51:28FromGitter<mratsim> so I need to force the rand symbol open so I can define it later
17:51:36krux02honestly I have no idea what I have written there
17:51:52FromGitter<mratsim> but due to generics, Nim tries to resolve the symbol anyway
17:52:40FromGitter<kaushalmodi> @Vindaar Thanks! I am too spoiled by strformat :)
17:52:54FromGitter<kaushalmodi> I have never yet needed to use that % operator
17:53:12FromGitter<kaushalmodi> s/never/not
17:53:53krux02mratsim: why don't you use an explicit mixin
17:55:11FromGitter<Vindaar> @kaushalmodi hehe, almost thought as much :)
17:55:28FromGitter<iffy> @tim-st bufsize=4096 didn't seem to help
17:56:31FromGitter<tim-st> @iffy the pythons default is 8192
17:57:00FromGitter<tim-st> but maybe it's not implemented to work together with FileStream, then it's a bug
17:57:50FromGitter<mratsim> @krux02, doesn’t work, I tried `mixin rand`
17:58:14FromGitter<kaushalmodi> question about the nightly build travis: https://github.com/nim-lang/nightlies/blob/master/.travis.yml
17:58:36FromGitter<kaushalmodi> on which OS is that koch xz run?
17:58:37FromGitter<iffy> @tim-st neither 4096 nor 8192 made any noticeable difference
17:58:54FromGitter<mratsim> full gist here: https://gist.github.com/mratsim/e29de51f73800e5bfce9419abd9ddcfc GTG
17:59:28FromGitter<tim-st> @iffy ok, then it's a bug I think because when `bufsize` is available for a `FileStream` it's expected to make a difference
17:59:48FromGitter<tim-st> I'd like to have buffer support for FileStream :\
17:59:49FromGitter<iffy> @rayman22201 I filed this https://github.com/nim-lang/Nim/issues/9026 (cause I need to go now) so that we can keep track of discussion there
18:01:05*darithorn quit (Quit: WeeChat 2.2)
18:04:49FromGitter<tim-st> are nightly builds planned for windows too?
18:05:07FromGitter<kaushalmodi> they should work on windows too
18:05:27FromGitter<kaushalmodi> download that tar.xz, extract it, and run `sh build.sh` (so needs to run in bash on windows)
18:06:20FromGitter<tim-st> doesnt windows need `build.bat` ?
18:06:25FromGitter<kaushalmodi> actually ignore what I said
18:06:28FromGitter<kaushalmodi> sorry
18:06:41FromGitter<kaushalmodi> but reading through instructions, the same tar.xz should work on windows too
18:06:55FromGitter<kaushalmodi> ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba5333fe5c2cc56adc366af]
18:07:51FromGitter<kaushalmodi> I don't see finish.exe in the tar.xz
18:07:57FromGitter<kaushalmodi> but I do see some *.bat files
18:08:22FromGitter<kaushalmodi> try it out.. and probably open an issue saying that those instructions probably need to be updated
18:08:23FromGitter<kaushalmodi> :)
18:11:36FromGitter<tim-st> ok, thanks, I thought it was a precompiled binary with `nim.exe`
18:13:03*navin_ quit ()
18:13:09enthus1astif you dont have a nim.exe on windows already you can bootstrap from c and build the devel Nim with koch boot
18:13:43FromGitter<kaushalmodi> enthus1ast: This is about testing the new nightly releases
18:13:57FromGitter<tim-st> yes, I have one, but I thought that I read that appveyor can build nightlies with complete binaries like on the official nim download page
18:14:02FromGitter<kaushalmodi> I think anyone should be able to download the release and just build it
18:14:13FromGitter<tim-st> I think ziglang does it
18:14:21enthus1astah ok nice to know
18:15:04FromGitter<kaushalmodi> @tim-st Yes, that would be nice. Have a compiled (and hopefully static) binaries for each platform
18:15:29FromGitter<kaushalmodi> The Go lang distribution can serve as a great example
18:15:54FromGitter<kaushalmodi> I don't code in Go, but I can simply download the release tar, extract it and put the contents in PATH. Done.
18:16:20FromGitter<kaushalmodi> useful to routinely build certain Go apps that I love: hugo, peco
18:16:26FromGitter<tim-st> yes, that's nice
18:22:25*Spitfire2 joined #nim
18:24:56TheLemonMantim-st, bufSize works but doesn't influence what fread does
18:26:12*flaviu quit (Remote host closed the connection)
18:27:03*Spitfire2 quit (Ping timeout: 244 seconds)
18:28:10FromGitter<tim-st> ok, then I think it would be better if the bufSize would work on the FileStream level, which had the wished impact I think
18:29:23TheLemonMansure thing
18:35:27*anzuof13 joined #nim
18:36:18*anzuof13 quit (Remote host closed the connection)
18:38:43Araqtim-st: the appveyor nightlies will have the build binaries
18:38:55Araqonce I got it to work... :-)
18:44:08FromGitter<kaushalmodi> Araq: As an aside, does `gtar caf foo.tar.xz foo/` work on OSX?
18:44:17FromGitter<kaushalmodi> you do not need to do tar and xz separately
18:44:29FromGitter<kaushalmodi> `ca` .. `a` stands for automatic
18:44:47FromGitter<kaushalmodi> the .xz extension makes tar figure out by itself what compression to use
18:45:18FromGitter<kaushalmodi> (unless of course that is less useful in niminst.nim if BSD tar does not understand `a`
18:45:20FromGitter<kaushalmodi> )
18:46:00FromGitter<bung87> I remenber I installed xz as required
18:46:34FromGitter<kaushalmodi> @bung87 because right now it is explicitly calling `xz`
18:47:24FromGitter<kaushalmodi> so.. BSD tar supports the `a` switch too: https://www.freebsd.org/cgi/man.cgi?query=bsdtar&sektion=1&manpath=freebsd-release-ports
18:48:07TheLemonManmake sure you check out if OpenBSD and NetBSD and DragonFlyBSD and ... agree :)
18:49:18FromGitter<kaushalmodi> oh well :)
18:54:02FromGitter<bung87> Ok I just awared of the detail,that was a py lib packaged images from docker compose file
18:54:44Araqkaushalmodi: no idea
19:03:31FromGitter<unreadable> Any advatages by choosing cpp backend over c?
19:04:13*ritchie_ quit (Remote host closed the connection)
19:04:13TheLemonMancheaper exceptions & you'll be able to stress-test it :)
19:05:59*Jesin joined #nim
19:08:13*Jesin quit (Remote host closed the connection)
19:09:50AraqTheLemonMan: I did stress test it, the results weren't bad
19:10:06FromGitter<tim-st> Araq: nice, good to hear!
19:10:29Araqbut still made the CIs red and so I haven't merged my patch
19:10:43TheLemonManAraq, the recent codegen problems prove otherwise heh the more people test it the merrier
19:10:49Araqalso ... running all the tests both with C and C++ will produce more timeouts
19:11:32TheLemonManone could shard the tests and run eg. C on travis and C++ on circleci or gitlab or whatever
19:11:57TheLemonManthe alternative is to shell out money for some more time on the boxes
19:13:08FromGitter<unreadable> Overall, what would you guys recommend as a backend
19:13:34*crem quit (Ping timeout: 240 seconds)
19:13:37FromGitter<unreadable> Or when to pick c over cop
19:13:42FromGitter<unreadable> cpp*
19:13:56*Jesin joined #nim
19:15:52*crem joined #nim
19:19:16krux02C is just the default
19:19:24krux02cpp should be as reliable
19:19:52krux02C just has potentionally more backends
19:20:04krux02so maximum compatibility -> C backend
19:20:51AraqTheLemonMan: yeah good idea
19:35:48FromGitter<codenoid> morning all
19:35:49FromGitter<codenoid> <3
19:37:15FromGitter<Clyybber> morning
19:37:30FromGitter<Clyybber> even if its evening here... heh
19:37:43*thebotnet_ joined #nim
19:44:21stefanos82morning...which day is it for you @codenoid?
19:45:28FromGitter<rayman22201> @iffy https://gist.github.com/rayman22201/6b26e1a14b5e4f04776103de976cd8d7
19:45:53FromGitter<rayman22201> using the built in file.readLine, (which uses fgets btw) blows everything else out of the water.
19:47:17FromGitter<rayman22201> also @tim-st and anyone else who cares
19:49:11FromGitter<codenoid> yeah, it's 02:49 AM, technically morning
19:49:33stefanos82ah lol I thought it was more like 8 or 9am for you
19:50:01stefanos82here it's 22:50PM
19:50:11FromGitter<rayman22201> @araq wrote this nice buffered / fgets implementation of readline for Files, but that implementation is not used for generic streams. I think fgets should work for other kinds of streams, so why isn't it used there? https://github.com/nim-lang/Nim/blob/2745111fc333d58f663480639aae5b2e0e0e1636/lib/system/sysio.nim#L144
19:50:34AraqI didn't write this, def- did iirc
19:50:46FromGitter<rayman22201> lol, git blame failed me then :-P
19:51:34*thebotnet_ quit (Quit: Page closed)
19:52:33*noonien quit (Quit: Connection closed for inactivity)
19:54:33FromGitter<rayman22201> Still, my question remains. I know fgets will at least work for File streams and Sockets, the two main uses for Streams, so why not use it in the stream module instead of the naive implementation?
20:20:27FromGitter<tim-st> @rayman22201 intersting, thanks, yes would be good if FileStream has the same performance as File
20:22:45*Jesin quit (Quit: Leaving)
20:24:32*Jesin joined #nim
20:27:39*cooldude13 joined #nim
20:28:05FromGitter<rayman22201> I think a reasonable solution is to do the dynamic dispatch thing and add a readLineImpl to the Stream interface here: https://github.com/nim-lang/Nim/blob/master/lib/pure/streams.nim#L41
20:28:14FromGitter<rayman22201> I may do a PR for this
20:28:49*icebattle quit (Quit: leaving)
20:29:01*Ven`` joined #nim
20:31:18FromGitter<iffy> @rayman22201 Thanks! (and for your comments on GitHub).
20:33:07*Ven`` quit (Ping timeout: 240 seconds)
20:33:31FromGitter<tim-st> I missed a proc to read a null terminated string from a stream, the readLine breaks on new line or null char, but sometimes you need to parse for null char only :\
20:34:13*cooldude13 quit (Ping timeout: 245 seconds)
20:34:18TheLemonManuse readData plus memchr?
20:35:08FromGitter<rayman22201> @iffy np. It was a small and fun / interesting problem :-P @tim-st This is true. pretty much every implementation of readLine works this way though
20:36:01FromGitter<rayman22201> @TheLemonMan lol, yeah, exactly, or mmap if you can at that point.
20:37:59FromGitter<iffy> wow... my real program is 5x faster than node now. More like what I expected
20:38:34TheLemonManonce again rayman saves the day!
20:39:20FromGitter<rayman22201> lol 😎
20:39:24FromGitter<rayman22201> not really though
20:39:29FromGitter<rayman22201> but thank you
20:39:43*erratic quit (Quit: this server has gone to sleep)
20:44:23Araqrayman: PR is welcome.
20:44:26*Zeno[m] left #nim ("User left")
20:44:37FromGitter<rayman22201> @araq working on it
20:45:46FromGitter<rayman22201> after lunch though. :-)
20:50:03*leru joined #nim
21:09:14*lhavelund joined #nim
21:15:07*lhavelund quit (Ping timeout: 240 seconds)
21:30:20*nsf joined #nim
21:32:11*miran joined #nim
21:32:18*leru quit (Remote host closed the connection)
21:33:16FromGitter<almynic> Hi I just switch from stable to devel and now I got this Error: usage of 'isNil' is a user-defined error ⏎ ⏎ Here is my code: ⏎ ⏎ ```code paste, see link``` [https://gitter.im/nim-lang/Nim?at=5ba5639c33da0f649e346742]
21:33:56FromGitter<almynic> Can anyone help me out?
21:34:08TheLemonManseqs (and strings) cannot be nil anymore
21:34:27FromGitter<kaushalmodi> *There should be a sticky for this somewhere.*
21:34:31TheLemonMan`this.formats.isNil` -> `this.formats.len == 0`
21:35:29FromGitter<almynic> Oh thanks.
21:36:09FromGitter<kaushalmodi> Also, if you want to *temporarily* keep using nil, compile with `--nilseqs:on`
21:37:01*leru joined #nim
21:38:28FromGitter<almynic> I'm not a fan of nil and null I try to avoid that as much as possible ;)
21:39:12TheLemonManhmm, we could slip a `deprecated` pragma with an informative message along with the .error. one
21:40:23*TheLemonMan quit (Quit: "It's now safe to turn off your computer.")
21:40:31FromGitter<almynic> That would help others or at least update the documentation.
21:41:11FromGitter<kaushalmodi> How would documentation help?
21:41:28FromGitter<kaushalmodi> It's not as if people search the documentation on see the error.
21:42:00FromGitter<kaushalmodi> Talking of which, now the documentation from devel branch is published at https://nim-lang.github.io/Nim/theindex.html
21:43:16FromGitter<zacharycarter> nice!
21:43:50FromGitter<kaushalmodi> @almynic OK, I see.. there should have been some sign of `isNil` not working for strings and seqs here: https://nim-lang.github.io/Nim/system.html#isNil,string
21:44:37FromGitter<kaushalmodi> well.. there is a *sign* albeit a hidden one
21:44:48FromGitter<kaushalmodi> clicking on those 3 dots shows: `proc isNil(x: string): bool {.noSideEffect, magic: "IsNil", error.}` :D
21:44:53FromGitter<kaushalmodi> <-- error
21:45:38FromGitter<kaushalmodi> TheLemonMan: Should an issue be opened to track this? addition of deprecated pragma?
21:52:56FromGitter<kaushalmodi> Fun fact of the day: ⏎ ⏎ > 14 representing the 14 original members of the nazi party, and the 88 representing the HH sign ⏎ ⏎ https://github.com/nim-lang/Nim/issues/9030 [https://gitter.im/nim-lang/Nim?at=5ba5683833da0f649e3481e3]
21:56:07FromGitter<rayman22201> Wat
21:57:39miranand so it begins....
21:58:15*WakiMiko3 joined #nim
21:58:17FromGitter<kaushalmodi> It's amusing.. there are a variety of trolls out there
21:59:57mirani really hope so somebody is having fun
22:00:18FromDiscord<Shield> this is how opensource dies
22:00:44miranand this is just a stab at the recent linux kernel events
22:00:46*WakiMiko3 quit (Remote host closed the connection)
22:05:24FromDiscord<Shield> it's sad, they found a way to waste people's time on stupid "issues", what's next? banning words from dictionaries?
22:08:48*Vladar quit (Remote host closed the connection)
22:15:17*nsf quit (Quit: WeeChat 2.2)
22:22:58*leru quit (Ping timeout: 245 seconds)
22:23:47*miran quit (Ping timeout: 252 seconds)
22:25:51FromDiscord<Tomohiro> When I write "const pi = 3.14159", it must be replaced with "const pi = 3.13159"?
22:28:29stefanos82what? -_-
22:34:20FromGitter<kaushalmodi> stefanos82: see above linked issue 😂
22:34:28*problame_ joined #nim
22:36:34stefanos82is this a troll or something?
22:36:49stefanos82"Diversity editor at NimGump Tech"
22:44:01*problame_ quit (Ping timeout: 260 seconds)
23:09:47*ariabuckles13 joined #nim
23:10:40*ariabuckles13 quit (K-Lined)
23:32:20*krux02 quit (Remote host closed the connection)
23:32:34*krux02 joined #nim
23:34:16*druonysus quit (Quit: No Ping reply in 180 seconds.)
23:35:27FromDiscord<exelotl> Is it possible to use a macro to modify a const after it has been defined?
23:35:30*druonysus joined #nim
23:35:30*druonysus quit (Changing host)
23:35:30*druonysus joined #nim
23:44:41FromGitter<rayman22201> Why not just have the macro generate the constant to begin with?
23:54:48*hydraz11 joined #nim
23:56:29*hydraz11 quit (Killed (Sigyn (Spam is off topic on freenode.)))
23:56:57*xet7 quit (Quit: Leaving)
23:57:10FromGitter<kayabaNerve> ^^