<< 21-12-2018 >>
00:00:10FromGitter<timotheecour> can i make hintlink show linkcmd then?
00:00:56FromGitter<timotheecour> (instead of informationless "Link")
00:10:26*krux02 quit (Remote host closed the connection)
00:14:50*xet7 joined #nim
00:16:30*xet7 quit (Remote host closed the connection)
00:18:21*xet7 joined #nim
00:18:58*xet7 quit (Remote host closed the connection)
00:21:00*xet7 joined #nim
00:21:55*theelous3__ quit (Ping timeout: 246 seconds)
00:22:05*xet7 quit (Remote host closed the connection)
00:22:46*xet7 joined #nim
00:23:48*shpx joined #nim
00:29:21*theelous3 joined #nim
00:35:03*kinkinkijkin quit (Quit: Leaving)
00:54:11*kinkinkijkin joined #nim
00:59:27*xet7 quit (Quit: Leaving)
01:03:07*dddddd quit (Ping timeout: 240 seconds)
01:03:51*kinkinkijkin quit (Remote host closed the connection)
01:07:47*dddddd joined #nim
01:20:09*kinkinkijkin joined #nim
01:28:21*d10n-work quit (Quit: Connection closed for inactivity)
01:33:43*shpx quit (Quit: Textual IRC Client: www.textualapp.com)
01:40:46*theelous3 quit (Ping timeout: 250 seconds)
01:47:58*whaletechno joined #nim
01:49:46*ghost64 quit (Quit: See you!)
01:50:24*whaletechho quit (Ping timeout: 272 seconds)
01:50:48*ghost64 joined #nim
01:55:55*McSoFake quit (Ping timeout: 256 seconds)
01:59:42*zachk quit (Quit: Leaving)
02:03:47*PrimHelios quit (Ping timeout: 250 seconds)
02:37:19*leorize quit (Quit: WeeChat 2.3)
02:37:36*leorize joined #nim
02:48:37*kapil____ joined #nim
02:58:55*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
03:00:01*Tyresc quit (Quit: WeeChat 2.4-dev)
03:00:01*banc quit (Quit: Bye)
03:00:38*theelous3__ joined #nim
03:16:00*theelous3__ quit (Read error: Connection reset by peer)
03:16:00*banc joined #nim
04:12:57*ftsf joined #nim
04:35:29bozaloshtshforward declarations are the work of the devil
04:44:53bozaloshtshis there any way to see more than one compiler error?
04:46:17*dddddd quit (Remote host closed the connection)
04:50:16*nsf joined #nim
04:55:10*endragor joined #nim
05:03:16*stephenw` joined #nim
05:05:18*stephenwithav quit (Ping timeout: 250 seconds)
05:21:04*NimBot joined #nim
05:25:05*plushie quit (Quit: plushie)
05:39:49FromGitter<Varriount> bozaloshtsh: `nim check`
06:00:32FromGitter<timotheecour> i added --errormax which catches more types of errors than nim check
06:23:21*narimiran joined #nim
06:47:31FromGitter<AchalaSB> How to open/ read the any file from directory? and how to display it?
07:10:31*krux02 joined #nim
07:11:22*Vladar joined #nim
07:19:15narimiran`readFile` ?
07:35:30FromGitter<alehander42> @AchalaSB how do you want to display it? if it's a binary file, like a picture, it doesnt make sense to `echo` it, you need to invoke another program which displays it
07:36:18FromGitter<alehander42> (and in this case, probably you can just pass the path to that program: if you dont need to do anything else than show it, you probably dont need to directly read it)
07:38:02FromGitter<alehander42> you can invoke programs with execProcess or execShellCmd (again this is for reading e.g. image files)
07:45:48*narimiran quit (Remote host closed the connection)
07:47:23*narimiran joined #nim
07:52:00*narimiran quit (Remote host closed the connection)
07:53:26*narimiran joined #nim
08:29:14FromGitter<AchalaSB> Its an image or may be a text file
08:29:41FromGitter<AchalaSB> How to invoke it using another program?
08:31:39FromGitter<AchalaSB> let consider `var filecontent = readFile("/home/download/cat.png")` is a path. How can display my image??
08:34:12FromGitter<alehander42> I told you: you see which program you have that can display images, how you can pass command args to it and then invoke it with execProcess or execShellCmd: please read their docs and try to do that first
08:39:23FromGitter<alehander42> (modules os.nim for example)
08:39:28FromGitter<alehander42> and osproc
08:42:43FromGitter<alehander42> Araq, I am just reading the nti nodes
08:42:46FromGitter<alehander42> and tm_ etc
08:43:26FromGitter<alehander42> Araq: what I mean is, I want cgen to generate the enum ones not only when reprEnum is used, but always when debug is in options
08:44:09FromGitter<alehander42> but not sure where to put it: in type section generation code(but genTypeSection seems empty, probably happens somehwere else?) , in enum literal generation code or somewhere else?
08:44:22FromGitter<alehander42> (the genTypeInfo invocation)
08:52:46*someone___ joined #nim
08:53:02someone___hey!
08:53:26someone___i'm trying to write a simple program that would compile to js
08:53:40someone___it seems like closures work differently with js backend, is that so?
08:56:47someone___oops, sorry, it was my mistake. all works if you start complaining about that something doesn't.
09:00:59*someone___ quit (Quit: Page closed)
09:05:58*floppydh joined #nim
09:06:38*narimiran quit (Remote host closed the connection)
09:07:45*narimiran joined #nim
09:35:45Araqalehander42: genTypeSection seems appropriate
09:38:46*endragor quit (Remote host closed the connection)
09:45:42*PMunch joined #nim
09:49:45krux02https://www.youtube.com/watch?v=xUsAKstP-AQ&feature=youtu.be
09:53:57FromGitter<timotheecour> Hi @krux02 quick question if you’re here?
09:55:01krux02yes
09:57:38FromGitter<timotheecour> just replying on this 1 point: ⏎ ⏎ > And btw, you did the mistake of importing strutils.cmpIgnoreStyle as well here: https://github.com/nim-lang/Nim/pull/10063/files#diff-dc05830f41a222a0b826e05a04ebd1e0R14 ⏎ Don't do it. [https://gitter.im/nim-lang/Nim?at=5c1cb91223323d58bd01a804]
09:58:04FromGitter<timotheecour> i ‘m not sure i agree, look at $nimc_D/compiler/wordrecg.nim
09:58:30FromGitter<timotheecour> it uses `from strutils import cmpIgnoreStyle` for this: ⏎ ⏎ `````` [https://gitter.im/nim-lang/Nim?at=5c1cb94623323d58bd01a9f5]
09:58:32FromGitter<timotheecour> proc findStr*(a: openArray[string], s: string): int = ⏎ for i in countup(low(a), high(a)): ⏎ ⏎ `````` [https://gitter.im/nim-lang/Nim?at=5c1cb9486649aa1f82ab02a1]
09:59:07FromGitter<timotheecour> so the same `cmpIgnoreStyle` would be used
10:03:45*endragor joined #nim
10:08:15*endragor quit (Ping timeout: 250 seconds)
10:13:04FromGitter<krux02> yes same problem here. strutils.cmpIgnoreStyle is technically incorrect here, but it doesn.t cause a problem, because there are no identifiers used where the case of the first character matters.
10:13:14FromGitter<krux02> it just didn't cause a bug yet.
10:14:45FromGitter<timotheecour> ok so I’m just propagating an existing maybe-bug then; happy to fix in a separate PR since it’s pre-existing, to avoid conflating 2 things
10:16:47FromGitter<alehander42> @krux02 some of the frontends might target gdb < 7 without pretty printers ? but I doubt it, it is kinda old
10:16:53FromGitter<alehander42> (watching the video blog)
10:16:55FromGitter<timotheecour> at least we both agree that having 5 procs that do slightly different things with the same name is a terrible idea, and shd be fixed
10:18:57AraqI think it's high time you focus on more pressing issues
10:19:12Araqyes, parseEnum is optimized for "stuff comes from command line"
10:19:34Araqyes, it's slightly broken for convenience
10:19:48Araqbut no, this is not a pressing problem
10:20:40Araqand while you're removing cruft others add new cruft like %* vs buildJson
10:24:52FromGitter<Vindaar> now I feel bad :P
10:26:27*stefanos82 joined #nim
10:27:20Araqsorry, wasn't my intention
10:27:25AraqI like buildJson
10:27:48Araqbut now I'm faced with the decision if having both %* and buildJson is worse than just %* or not
10:28:02Araqand I made an old mistake
10:28:17*endragor joined #nim
10:28:26AraqI didn't communicate clear enough that it is a "if we do it, here is how..."
10:29:11FromGitter<timotheecour> i thought buildJson was settled and ready to ship...
10:29:22Araq^ see :-)
10:30:01Araqin this case I say: ensure the docs and tests are really good and we'll merge and move on
10:30:44Araqbtw the Fortran standard arrived https://wg5-fortran.org/f2018.html
10:31:07Araqwhich has a new "C interop" section that also seems to cover how to do matrixes and strides etc
10:32:27*endragor quit (Ping timeout: 240 seconds)
10:40:21FromGitter<Vindaar> heh, no worries, I was half joking ;) happy to make docs + tests really good though, if I'm given some guidance :)
10:55:52*endragor joined #nim
10:56:15FromGitter<mratsim> @araq, oh, I’m jumping ship to Fortran then :P
10:57:34FromGitter<mratsim> I hope the manual/Nim specs don’t become an “ISO-1234” reference fest
10:59:16FromGitter<gogolxdong> @alehander42 How do you use routing in karax? hashData or click?
11:00:03*endragor quit (Ping timeout: 245 seconds)
11:07:07*floppydh quit (Ping timeout: 240 seconds)
11:08:43Araqgogolxdong todoapp shows how routing can be done
11:09:36*floppydh joined #nim
11:10:16FromGitter<gogolxdong> There is an foreign node issue using click routing, echart works fine using hashPart .
11:10:31FromGitter<alehander42> yeah, I haven't used it iirc, I didn't really need it in electron
11:12:08FromGitter<dom96> What even is buildJSON?
11:12:35*narimiran quit (Remote host closed the connection)
11:13:07FromGitter<Vindaar> @dom96 https://github.com/nim-lang/Nim/pull/10037
11:22:07Araqdom96 oh yeah, please take over :-)
11:23:50*PMunch quit (Remote host closed the connection)
11:42:32FromGitter<mratsim> ugh, how many people were pulled in that conversation
12:04:56dom96Sorry, but I closed it
12:08:51Araqfine with me
12:09:44Araqat some point I'll write an araqslib nimble package that uses operators for everything
12:10:42Araqand it will have %*$ and %*$$ that uses marshal.$$ for super-stringifications
12:11:27Araqand $| for integrals
12:12:08Araqand $+ for sum
12:12:18Araqand $* for product
12:12:41krux02∫ΣΠ
12:12:51krux02unicode for the win
12:12:53krux02:P
12:13:40Araqand &* for the concatenative repeat
12:15:17Araqand $- for the derivation
12:16:21krux02so you just want to reinvent APL
12:17:08*Trustable joined #nim
12:17:48FromGitter<mratsim> https://en.wikipedia.org/wiki/K_(programming_language)
12:18:51FromGitter<zacharycarter> most languages get by fine with marshalJSON unmarshalJSON
12:19:51FromGitter<zacharycarter> why not include that "interface" in the stdlib and then allow users to implement their own de/serializlation implementations
12:20:18FromGitter<zacharycarter> and use the json marshaling / unmarshaling facilities that are already present?
12:21:16Araqthat's a different topic
12:21:46FromGitter<zacharycarter> okay
12:25:34*fthe joined #nim
12:27:28*floppydh quit (Ping timeout: 246 seconds)
12:27:52FromGitter<alehander42> I almost used Σ in my lang
12:28:20FromGitter<alehander42> actually its easy to type with a good editor plugin
12:28:29FromGitter<alehander42> but still wouldn't bet on it :D
12:28:49FromGitter<mratsim> debugging that in vim remote would be annoying ...
12:29:07FromDiscord_<smal> I love using flame emojis in my code
12:29:13FromDiscord_<smal> Makes it run faster
12:29:52*floppydh joined #nim
12:30:16FromGitter<alehander42> hm Araq, smal has a point, the compiler should handle a racecar emoji variable name as automatic `-opt:speed`
12:30:25FromGitter<mratsim> @smal, now you just need to upgrade your comments: https://github.com/pjreddie/darknet
12:31:07FromGitter<alehander42> ew
12:31:34*floppydh quit (Client Quit)
12:32:46*Tyresc joined #nim
12:38:00FromGitter<survivorm> Is this a sane world? ^D
12:45:19*fthe quit (Ping timeout: 246 seconds)
12:57:37*kapil____ quit (Quit: Connection closed for inactivity)
13:02:32Araqany ideas why tasynctry fails with 'nim cpp'?
13:05:23*kapil____ joined #nim
13:20:59*abm joined #nim
13:22:51*Snircle joined #nim
13:32:06*legogh joined #nim
13:49:35*legogh is now known as littleli
13:54:01*nsf quit (Quit: WeeChat 2.3)
14:15:29littlelihello everyone (waving)
14:15:57FromGitter<brentp> Hi am getting `SIGSEGV: Illegal storage access. (Attempt to read from nil?)` in a newSeq call from at least nim-0.19.0 and devel. Here is the gdb traceback: ⏎ ⏎ ```code paste, see link``` ⏎ ⏎ how would I go about debugging this? I am wrapping a C library and maybe something odd is happening there, but can't find anything suspicious. [https://gitter.im/nim-lang/Nim?at=5c1cf59dbabbc178b2b823ef]
14:17:00*rockcavera quit (Remote host closed the connection)
14:18:41*littleli quit (Quit: leaving)
14:20:38FromGitter<brentp> the error from nim is: ⏎ ⏎ ``` g.afs = newSeqUninitialized[float32](L) ⏎ echo g.afs.len``` ⏎ ⏎ so immediately after the newSeq, printing the len causes a problem. It's a fairly large seq. [https://gitter.im/nim-lang/Nim?at=5c1cf6b63c4093612c3ea89e]
14:22:46leorizebrentp: if you're using gdb, you can build Nim with `-d:noSignalHandler` to disable the default SIGSEGV handler
14:23:09leorizeoh, the crash looks like it come from the GC
14:25:16leorizeif it worked before 0.19.0 then it must be a regression. Can you create a small reproducible sample?
14:27:00Araq"must be a regression" -- that's not how memory unsafe code works...
14:27:34FromGitter<brentp> I don't know if it worked before 0.19.0, I just mean it's not specific to a recent version. I haven't been able to make a reproducer. I think it's related to the c-lib I am using.
14:27:43FromGitter<brentp> but don't know how to find the problem.
14:27:55Araqcdecl vs stdcall problems
14:28:01Araqptr T vs ptr ptr T problems
14:28:46Araq"I wrapped a nano-threads C library" problems
14:29:09Araq(= C library that mucks around with the stacks)
14:29:17FromGitter<brentp> ok. I'll look into those. the lib is here fwiw : https://github.com/brentp/nim-miniz/tree/Zip
14:30:01Araq"C's int is not Nim's int" problems
14:30:08Araq"C's float is not Nim's float" problems
14:30:34AraqC func of an unknown prototype is called
14:30:48Araqthe wrappers doesn't use the .bycopy pragma
14:31:00Araqthe wrapper wasn't created by c2nim
14:32:03FromGitter<brentp> @Araq, are you describing actual issues with this nim-miniz now, or enumerating possible problems?
14:32:14Araqenumerating possible problems :-)
14:32:39Araqmz_alloc_func* = proc (opaque: pointer; items: csize; size: csize): pointer <-- missing calling convention
14:33:26Araqall your callback types need to be annotated with {.cdecl.}
14:34:24FromGitter<brentp> ok. I jumped into someone else's lib assuming it was c2nim'ed. It did not have bycopy pragma, but adding that doesn't resolve. I'll look at the calling convention stuff.
14:36:12Araqdiscard pZip.mz_zip_writer_finalize_archive() # lines like these also make debugging harder, don't ignore errors, turn them into exceptions
14:36:45Araqvar pZip: ptr mz_zip_archive = cast[ptr mz_zip_archive](alloc0(sizeof(mz_zip_archive))) # seems easier to do
14:36:55Araqvar pZip: mz_zip_archive
14:37:07Araqand then the 'addr' operator
14:39:00Araqand things like MINIZ_USE_UNALIGNED_LOADS_AND_STORES are implementation hacks that have nothing to do in a Nim wrapper
14:39:47FromGitter<brentp> my code is the `Zip` type at the end where it `doAssert`s those instead of discarding. but maybe starting from a c2nim version would be simpler than tracking down the cause in this.
14:40:07*xet7 joined #nim
14:41:49Araqin general, wrap what you need, don't wrap the billion's typedef unsigned int zlib_uint; anachronism
14:44:39FromGitter<brentp> understood. thanks for taking a look.
14:59:49*dddddd joined #nim
15:02:36*d10n-work joined #nim
15:02:48*Ven`` joined #nim
15:06:37*endragor joined #nim
15:10:47*endragor quit (Ping timeout: 240 seconds)
15:27:19dom96Araq, krux02: Nice video guys!
15:27:26dom96For those that missed it: https://forum.nim-lang.org/t/4491
15:29:02FromGitter<mratsim> reminds me @shashlick, you really should change the calling convention of nimgen to cdecl by default, maybe with a `when defined(vcc): {.push callingConvention: stdcall.} else: {.push callingConvention: cdecl.}`
15:29:20FromGitter<mratsim> only microsoft uses stdcall
15:30:07krux02dom96: thanks
15:38:27*krux02 quit (Remote host closed the connection)
15:40:25*PrimHelios joined #nim
15:40:51*Ven`` quit (Quit: My MacBook has gone to sleep. ZZZzzz…)
15:43:50Araqdom96, oh thanks
15:59:20*whaletechno quit (Quit: ha det bra)
16:16:57*Vladar quit (Remote host closed the connection)
16:26:16*ftsf quit (Read error: Connection reset by peer)
16:29:05*PMunch joined #nim
16:33:42*OrganicAnywhere joined #nim
16:38:25*skelett1 quit (Quit: WeeChat 2.1)
16:41:02*thomasross quit (Remote host closed the connection)
16:44:42*PMunch quit (Ping timeout: 250 seconds)
16:50:28*PMunch joined #nim
16:56:18*Cthalupa quit (Ping timeout: 245 seconds)
16:59:13*Cthalupa joined #nim
17:04:03*endragor joined #nim
17:08:35*endragor quit (Ping timeout: 250 seconds)
17:35:55*plushie joined #nim
17:44:33*OrganicAnywhere quit (Ping timeout: 250 seconds)
17:53:34*endragor joined #nim
17:58:20*endragor_ joined #nim
18:01:43*endragor quit (Ping timeout: 246 seconds)
18:17:10*Vladar joined #nim
18:18:14*endragor_ quit (Remote host closed the connection)
18:24:06*endragor joined #nim
18:26:13*steshaw quit (Quit: Connection closed for inactivity)
18:28:19*endragor quit (Ping timeout: 246 seconds)
18:28:24*nsf joined #nim
18:37:20*abm quit (Quit: Leaving)
18:57:35*whaletechno joined #nim
18:59:47*MyMind joined #nim
19:00:23*Sembei quit (Ping timeout: 250 seconds)
19:15:36FromGitter<brentp> I started from scratch, used c2nim on the ~10 functions I needed, used cdecl, and get the exact same error. i already switched my code to use the nim/zipfiles/zip instead. ⏎ with -d:useLibzipSrc
19:20:42*OrganicAnywhere joined #nim
19:24:26*PMunch quit (Remote host closed the connection)
19:32:07*OrganicAnywhere left #nim ("Opi out.")
19:37:50*zachk joined #nim
19:41:19*zachk quit (Changing host)
19:41:19*zachk joined #nim
19:55:46*nsf quit (Quit: WeeChat 2.3)
20:07:37*kapil____ quit (Quit: Connection closed for inactivity)
20:13:19*Vladar quit (Remote host closed the connection)
20:23:55*druonysus quit (Ping timeout: 244 seconds)
20:41:08*elrood joined #nim
20:42:06Araqbrentp interesting
20:49:09FromGitter<brentp> /shrug I got it to work with boehm, but figured it best to move on.
20:51:31FromGitter<Clyybber> Araq: Is ditching the GC an eventual goal?
20:52:43*Trustable quit (Remote host closed the connection)
20:53:24Araqyes and no
20:55:14Araqbrentp: can you show me the code please?
20:56:30Araqclyyber: it's still a bit unclear what to do with 'ref' but mostly because we have so many different ideas about it :P
20:59:19FromGitter<brentp> @Araq : https://github.com/brentp/nim-minizip
21:00:02FromGitter<brentp> the test pass fine, but get weird stuff elsewhere.
21:01:52AraqGC is seen as suboptimal because (a) worse interop with C(++) and (b) O(heap size) tracing algorithm. That doesn't mean we have to throw away the idea of automatic MM though.
21:02:07*wildlander joined #nim
21:02:20Araqbrentp: mz_alloc_func lacks the .cdecl afaict
21:02:36Araqmz_realloc_func too
21:04:35FromGitter<Clyybber> Araq: Ok, so destructors + GC will probably work together? Or do they already?
21:04:50ldleworkwould be interesting to see Nim become the second serious memory-safe language
21:05:15FromGitter<Clyybber> ldlework I would like *optionally* memory-safe
21:05:21ldleworksee if it can make all that lifetime tracking stuff "friendly" like pythonize/delphize it
21:05:45FromGitter<Clyybber> Var parameters are already some kind of borrow checking
21:06:27AraqClyybber: Yes they will work together and that's not the hard part in designing the system, ymmv
21:06:52FromGitter<Clyybber> ldlework And with features like no nil, or non nil strings that evade allocation the future looks great :D
21:06:59FromGitter<Clyybber> Araq: Ok, nice
21:07:08Araqthe hard part is deciding on what 'ref' actually means :-)
21:07:22FromGitter<Clyybber> A safe ptr?
21:07:32Araqit's obviously an owning pointer, but does it allow cycles
21:08:27FromGitter<Clyybber> Why would cycles be a problem?
21:08:41FromGitter<Clyybber> Or would they?
21:08:53FromGitter<brentp> @Araq, if I add cdecl to those, then the tests won't even pass.
21:08:53Araqbecause then we need some kind of tracing mechanism
21:09:16Araqbrentp: also compile with -d:useSysAssert -d:useGcAssert for more error checking
21:09:23FromGitter<Clyybber> Ah, yeah, without a gc that seems kinda tricky
21:09:44Araqthere is a world of hardly explored semi-automatic schemes
21:11:43FromGitter<Clyybber> Araq: Do you think it's a good idea to optionally convert C "enums" to distinct cints, and then generate borrowed `+`, `-`, `=` procs?
21:11:55FromGitter<Clyybber> In c2nim
21:11:57Araqthat's always the best idea
21:12:57FromGitter<Clyybber> I have patched c2nim a time ago to do that, but I remember my code being very hacky
21:13:08FromGitter<Clyybber> So I might clean that up and do a PR
21:14:10Araqthat would be really nice, thank you
21:16:36FromGitter<Clyybber> Araq: Do you happen to know what to do with enums that contain code like ⏎ ⏎ ``` SOMETHING = SOMETHING``` ⏎ ⏎ It's in the vulkan headers, and I really don't know what the value is supposed to be then. [https://gitter.im/nim-lang/Nim?at=5c1d58336649aa1f82af1171]
21:17:04FromGitter<Clyybber> I currently just skip those
21:18:48Araqno idea either, sorry
21:27:09*druonysus joined #nim
21:27:09*druonysus quit (Changing host)
21:27:09*druonysus joined #nim
21:33:30FromGitter<Clyybber> Araq: How does c2nim currently deal with C identifiers resulting in the same Nim identifier?
21:35:54Araqnot at all you use #mangle to fix it
21:37:33FromGitter<Clyybber> Ah ok thank you
21:37:50FromGitter<brentp> @Araq with those flags, I see: `[GCASSERT] markS: foreign heap root detected A!`
21:39:15FromGitter<Clyybber> Araq: Are objects `{.bycopy.}` by default?
21:39:41FromGitter<Clyybber> Or is that only for C interop?
21:40:25FromGitter<Clyybber> Oh, nvrmind, found it in the docs :P
21:40:59FromGitter<Clyybber> Still, what are objects by default? Are they a smart combination of both?
21:44:49Araqbrentp: seems you have mapped 'ref' to something that isn't
21:46:07Araqor you assumed the old "it's immutable so I am free to share it" myth
21:48:13FromGitter<brentp> what would "mapping to 'ref' to something that isn't" look like?
21:48:39FromGitter<brentp> I think I understand the other one.
21:51:48Araqcasting to 'ref'
21:52:01Araqor pretending C code returns a 'ref'
21:52:27*kinkinkijkin quit (Ping timeout: 240 seconds)
21:57:00FromGitter<brentp> I think the most skechcy thing I am doing is using an object with {.bitsize.} pragmas (which you previously told me to avoid) and then casting that to it's component type.
21:58:30FromGitter<brentp> ```code paste, see link``` ⏎ ⏎ and then casting that to uint64. [https://gitter.im/nim-lang/Nim?at=5c1d6206d945b96882d6f2ac]
22:03:07Araqthat doesn't matter, we fixed bitsize for sizeof
22:03:17Araqand it doesn't contain GC'ed memory either
22:10:27*kinkinkijkin joined #nim
22:18:13FromGitter<mratsim> So Rust is adding macros (beyond string mixing): https://blog.rust-lang.org/2018/12/21/Procedural-Macros-in-Rust-2018.html
22:18:20FromGitter<mratsim> mixin*
22:20:40Araqthat's old news as far as I'm concerned
22:24:44Araqdoes anybody know good offline speech-to-text software?
22:26:57kinkinkijkinsam for the commodore 64
22:27:07kinkinkijkinjust kidding, that's a little old
22:27:15kinkinkijkinand requires a commodore 64
22:34:10*fthe joined #nim
22:35:25FromGitter<arnetheduck> I used dragon naturally speaking around 98', maybe that's still around :)
22:37:08FromGitter<Clyybber> Araq: https://github.com/kaldi-asr/kaldi is great
22:45:40*stefanos82 quit (Remote host closed the connection)
22:45:58*elrood quit (Remote host closed the connection)
23:30:44*McSoFake joined #nim
23:30:53*fthe quit (Ping timeout: 245 seconds)
23:44:01*fthe joined #nim