<< 04-02-2017 >>
00:12:00*arnetheduck quit (Ping timeout: 276 seconds)
00:14:25*nsf quit (Quit: WeeChat 1.7)
00:24:43*vitalyx quit (Quit: vitalyx)
00:35:41*couven92 quit (Quit: Client disconnecting)
00:42:49*bjz joined #nim
01:05:47*pregressive quit ()
01:24:41*chemist69 quit (Ping timeout: 255 seconds)
01:25:04*chemist69 joined #nim
01:35:30carterzahttp://imgur.com/a/KTREO
01:35:32carterzarunning in nim
01:35:34carterza:D
01:57:11*Matthias247 quit (Read error: Connection reset by peer)
02:10:55*devted joined #nim
02:12:32*chemist69 quit (Ping timeout: 255 seconds)
02:26:20*chemist69 joined #nim
02:47:48*carterza quit (Quit: carterza)
02:55:54*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
02:59:42*devted quit (Quit: Sleeping.)
03:01:27*bjz joined #nim
03:05:45*devted joined #nim
03:23:17*Kingsquee joined #nim
03:25:21*zachcarter joined #nim
03:25:23*zachcarter is now known as carterza
03:35:41*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
03:40:38ftsfcarterza, what's that?
03:40:54carterzahttps://github.com/memononen/nanovg
03:42:31ftsfsweet
03:47:55carterzaI got some existing bindings to compile
03:48:27carterzahowever when I try to use the wrapper in my own project I start getting forward declaration errors so I must be missing an include somewhere
03:55:07*brson quit (Quit: leaving)
03:58:02*devted quit (Quit: Sleeping.)
04:13:59*chemist69 quit (Ping timeout: 256 seconds)
04:17:22carterzacan anyone explain to me how to include a C header file in my nim code?
04:17:29carterzaI’m trying to avoid a forward declaration error
04:18:45*chemist69 joined #nim
05:06:13*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
05:12:12*bjz joined #nim
05:41:52carterzahttp://imgur.com/a/5rV4l
05:56:38*bjz quit (Ping timeout: 255 seconds)
05:57:15*bjz_ joined #nim
06:11:05*bjz joined #nim
06:12:23*bjz_ quit (Ping timeout: 264 seconds)
06:20:07def-pri-pubcarterza, you will probably need to use the emit pragma
06:20:25carterzadef-pri-pub I got it straightened out
06:20:37def-pri-pubag
06:20:37carterzathank you though
06:20:38def-pri-pubah
06:20:40*def-pri-pub quit (Quit: leaving)
06:21:30*Varriount|Mobile joined #nim
06:21:59Varriount|Mobilecarterza: are you on gitter? The bridge is down again
06:22:05Varriount|MobileI responded though
06:22:09carterzaVarriount I’m not but I’ll hop on
06:23:31*Varriount|Mobile left #nim (#nim)
06:35:57*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
06:49:12*xet7 joined #nim
06:50:41*bjz joined #nim
06:54:04*unlaudable joined #nim
07:18:49*unlaudable quit (Ping timeout: 240 seconds)
07:29:07*zevlg quit (Ping timeout: 240 seconds)
07:30:09*nsf joined #nim
07:46:35*Vladar joined #nim
07:47:16*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
08:03:46*Atomic_Rydbq joined #nim
08:08:14*Atomic_Rydbq quit (Ping timeout: 245 seconds)
08:12:55*bjz joined #nim
08:36:54*yglukhov joined #nim
08:53:59*yglukhov quit (Remote host closed the connection)
08:54:32*yglukhov joined #nim
08:59:11*yglukhov quit (Ping timeout: 255 seconds)
09:00:16*bjz_ joined #nim
09:01:52*bjz quit (Ping timeout: 258 seconds)
09:04:16*carterza quit (Quit: carterza)
09:10:03*yglukhov joined #nim
09:16:20*rokups joined #nim
09:20:52*couven92 joined #nim
09:32:27*chemist69 quit (Ping timeout: 256 seconds)
09:35:17*chemist69 joined #nim
09:47:56*yglukhov quit (Remote host closed the connection)
09:48:32*yglukhov joined #nim
09:51:18*bjz_ quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
09:51:58*bungoman quit (Read error: Connection reset by peer)
09:52:44*yglukhov quit (Ping timeout: 255 seconds)
09:57:45*bjz joined #nim
10:20:52*yglukhov joined #nim
10:22:11*yglukhov quit (Remote host closed the connection)
10:22:47*yglukhov joined #nim
10:27:23*yglukhov quit (Ping timeout: 255 seconds)
10:36:05*Matthias247 joined #nim
10:41:15*yglukhov joined #nim
10:41:51rokupsdom96: is there a way to make nimble perform some tasks on install? my ustring depends on c lib. if it could be downloaded/unpacked automatically along with package that would be great.
10:42:46Vladarrokups, check out "requires" key
10:43:33rokupsVladar: but that is for other nimble packages
10:44:08Vladarwell, yes
10:44:12Araqrokups: checkout the new distros.nim module
10:44:16Araqwas designed for that
10:44:40Araqbut for now it only produces output like "run apt-get boo to finish the installation"
10:44:44rokupsAraq: would that work for stuff that is not on os packages?
10:44:59rokupsguess not
10:45:09Araqyes, kind of
10:45:45Araqdistros.foreignCmd(...)
10:45:47yglukhovAraq: SYS_gettid is a new change. doesn't compile on android. have you considered pthread_self for this purpose?
10:46:30Araqwell no. android needs its own implementation then
10:46:57rokupsAraq: how about this: would it be somehow possible to obtain nimble package path at compile time? For {.compile.} consumption for instance.
10:47:37yglukhovwhy not pthread_self? i guess this is a highly available api across platforms
10:49:03yglukhovAraq
10:49:22cheatfateyglukhov, maybe because pthread_self() returns not identifier, but address?
10:50:12yglukhovit returns pthread_t which is a thread id semantically.
10:51:38cheatfateyglukhov, could you please test value `178` for android?
10:51:49cheatfatei mean SYS_gettid = 178
10:52:14cheatfateor 224
10:52:39Araqhmm I thought pthread_self is not available on Linux
10:52:44Araqbut I was wrong.
10:53:39Araqok, so patch it to use pthread_self please
10:53:53Araqcan we cast pthread_t to int?
10:55:46cheatfatetypedef unsigned long int pthread_t;
10:57:02Araqgood enough :-)
10:57:21*stisa joined #nim
11:00:43*nsf quit (Quit: WeeChat 1.7)
11:04:11*derlafff quit (Remote host closed the connection)
11:04:20*derlafff joined #nim
11:10:05yglukhovcheatfate, Araq, now that i've read a bit more about subj, pthread_self is not guaranteed to be unique. and to compare pthread_t one has to use pthread_equal, so thats a bit unwieldy i guess.
11:10:51yglukhovbut i've found that android has pid_t gettid() func, so i'll make a patch for android
11:12:48cheatfateyglukhov, of course it has difference only in syscall number
11:13:05cheatfateand i think we need to define this numbers
11:13:09cheatfatefor all architectures
11:13:33cheatfateso its why i asked you to test two values
11:13:48yglukhovcheatfate: why would we mess with this, if theres a gettid function available?
11:13:54cheatfate178 arm64, 224 for arm32
11:14:19cheatfatebecause gettid is not declared in glibc
11:15:16yglukhovso?
11:15:34yglukhovandroid ndk defines it in unistd.h
11:16:24cheatfategettid uses SYScall right?
11:16:34cheatfateyou have broken android gettid()
11:16:50cheatfateso you can watch your unistd for syscall number
11:16:53yglukhovgettid uses SYScall - i would suspect that
11:17:14cheatfatecould you please grep it for `__NR_gettid`
11:17:41cheatfateso with your patch we can patch whole linuxes at ones
11:17:43cheatfateonce
11:17:54cheatfatenot only android
11:18:02cheatfatewhich i think uses 178 or 224
11:25:25yglukhovcheatfate: 224, 222 and 178
11:26:33cheatfateyglukhov, i was wrong threads.nim don't uses numbers and uses SYS_gettid
11:26:49cheatfatehave you such constant?
11:29:08yglukhovcheatfate, no SYS_gettid is not consistently present in all platforms. __NR_gettid seems to be present everywhere. gettid() is present everywhere.
11:29:45yglukhovi would really go with gettid() solution, because it looks like the most "public" api.
11:30:04yglukhovcheatfate: what advantage do you see in using syscall() ?
11:30:42cheatfatesyscalls are better because they can't be changed :)
11:31:03yglukhovwhat do you mean "changed">
11:31:05yglukhov?
11:32:28cheatfatei mean gettid is just a name which is not part of any standard
11:33:04yglukhovis __NR_gettid a part of any standard?
11:33:34yglukhovor SYS_gettid?
11:34:10cheatfateits system specific constant
11:34:17cheatfatelinux os specific constant
11:34:28cheatfateand i dont think this name will be changed
11:35:30yglukhovjust like gettid() is linux specific function, except SYS_gettid has different values across different linuxes, and gettid() behavior is always the same, so which one is more stable?
11:36:33cheatfateyglukhov, please make `man gettid`
11:36:36cheatfateand read notes
11:36:51cheatfate Glibc does not provide a wrapper for this system call; call it using syscall(2).
11:38:36cheatfateso it looks like this name can be changed
11:39:23*chemist69 quit (Ping timeout: 256 seconds)
11:42:55*chemist69 joined #nim
11:44:30yglukhovcheatfate: let me put it in another way. android ndk doesnt provide SYS_gettid for all platforms, while it does provide gettid(). so other solutions do you see here, if not using gettid?
11:44:50cheatfate__NR_gettid
11:45:16cheatfatewe need to change threads.nim to use `__NR_gettid`
11:45:39yglukhovso you kinda make your bet that SYS_gettid always equals __NR_gettid?
11:46:27cheatfatei don't making any bets here
11:46:44yglukhovwhy? is it documented anywhere?
11:47:04cheatfatejust because android don't use `SYS_` constants for some reason, then we can use `__NR_` constants which are available for both kernels
11:51:25yglukhovok, will change it to __NR_
11:52:11cheatfatei have downloaded NDK and checked headers
11:52:29cheatfateyou also need to change
11:52:38cheatfateheader
11:53:12yglukhovi just renamed SYS_gettid to __NR_gettid and it worked just fine
11:53:20yglukhov* compiled at least =)
11:54:24cheatfatei was wrong everything other is OK
11:54:44cheatfateso just need to change `SYS_gettid` to `__NR_gettid`
11:56:39yglukhovcheatfate: https://github.com/nim-lang/Nim/pull/5338
11:59:12*krux02 joined #nim
12:03:18*Snircle joined #nim
12:03:43*planhths joined #nim
12:05:25demi-is there a method for getting raw keyboard input as part of the standard library?
12:13:34cheatfatedemi-, please explain `raw`
12:14:00dom96rokups: you can do things like this: https://github.com/nim-lang/nimble/blob/master/tests/nimscript/nimscript.nimble#L27
12:14:07dom96but I don't think 'before install' is supported yet
12:14:15dom96it might be, but it's definitely at least buggy
12:14:58demi-right now i am doing a `readline(stdin)` which is fine for reading a whole like, but i want to add keybindings to this, so need to interpret key events that don't belong to a set input (like control-s/arrow keys/tab)
12:16:14cheatfatedemi-, but i think all this control characters goes to stdin as escape sequences, am i right?
12:16:29Amun_Ranope
12:18:07*yglukhov quit (Remote host closed the connection)
12:18:09cheatfatedemi-, you just need to read `stdin` with `read` not `readline`
12:18:31*yglukhov joined #nim
12:18:37*yglukhov quit (Remote host closed the connection)
12:18:51*yglukhov joined #nim
12:20:40*Gilga joined #nim
12:24:09cheatfatedemi-, does it helps?
12:24:28demi-is that posix.read you mean? because i don't see a `read` in system
12:26:28cheatfatedemi-, please look at system.nim (readChar, readAll, readBytes, readChars, readBuffer)
12:26:53demi-yeah i tried readChar, and that isn't doing what i want
12:27:33cheatfatedemi-, maybe because escape sequences is not one byte length?
12:27:59demi-touche
12:29:20*nsf joined #nim
12:29:21demi-ok, i'm giving it one character input and it continues to block until i hit return
12:30:00cheatfateok i see such too
12:30:12cheatfatethen you need to read it via posix.read()
12:30:39cheatfatebecause this functions are aliases for C's FILE functions (which are buffered)
12:30:39*stisa quit (Read error: Connection reset by peer)
12:30:53demi-which will not work on windows, correct?
12:31:04cheatfatedemi-, C's FILE functions?
12:31:10demi-no, the posix.read
12:31:29cheatfateyep, posix.read will not work on windows
12:32:13*derlafff quit (Remote host closed the connection)
12:32:16demi-so i could be better off implementing with readline or some other similar library
12:34:17*derlafff joined #nim
12:34:51cheatfatedemi-, ok you can disable buffering with setvbuf() on `stdin` handle
12:34:57cheatfatedemi-, and _IONBF constant
12:35:30demi-is that going to be reliable?
12:35:35*stisa joined #nim
12:36:21cheatfatenim don't such functions
12:36:32cheatfatebut i think it will be reliable for unix & posix
12:37:33cheatfateand windows too
12:38:19cheatfateof course you can use stdin.getFileHandle() which returns descriptor, but we still don't have native cross-platform api for file reading/writing
12:39:21demi-i see, then for now i might just implement this by handing off to a readline library and then allow users to enable the functionality as an opt-in
12:39:54*Gilga quit (Ping timeout: 245 seconds)
12:41:26*Gilga joined #nim
12:41:30cheatfatebut why you need readline library?
12:42:07cheatfateare you trying to implement some kind of console editor?
12:42:43demi-i was going to expand on the functionality i implmented here: https://github.com/samdmarshall/cmd.nim (interactive command prompt written in nim)
12:46:08*Arrrr joined #nim
12:46:08*Arrrr quit (Changing host)
12:46:08*Arrrr joined #nim
12:46:23cheatfatedemi-, i think it will be much easier to implement windows version of read()/write() :)
12:48:15demi-hmmm, ok, i've got some other things to document and fix up in it first, so i will think on it.
12:49:28demi-also is there an equivalent to python's list comprehension in nim? it seems like map()+do syntax would be it but that seems ugly
12:52:48*Gilga quit (Ping timeout: 252 seconds)
12:54:06demi-ah, figured it out with sequtils.mapIt()
13:09:59subsetparkdemi-: you can write list comprehensions directly using the lc[] macro imported from future
13:10:39subsetparkhttps://nim-lang.org/docs/future.html
13:10:53*yglukhov quit (Remote host closed the connection)
13:11:29*yglukhov joined #nim
13:15:41*yglukhov quit (Ping timeout: 240 seconds)
13:20:38Araqdemi-: rdstdin.nim, terminal.nim modules exist and work
13:26:51rokupsthanks dom96. for sake of simplicity i just bundled relevant source code of utf8rewind in the package.
13:29:30*Gilga joined #nim
13:37:09demi-Araq: yeah i saw rdstdin, but am unsure how to use it since the documentation seems a bit lacking
13:38:01demi-ah, i will definitely give terminal.nim a try, that looks promising. thanks!
13:39:03*Kingsquee quit (Quit: https://i.imgur.com/qicT3GK.gif)
13:39:35*devted joined #nim
13:40:27*cncl joined #nim
13:42:17Araq var line = ""
13:42:17Araq while readLineFromStdin("> ", line):
13:42:17Araq echo „you wrote „, line
13:43:55cncldom96: i'm the guy from jblow's stream chat from a couple of weeks back. i was thinking of updating the game i showed you to work with the newer nim versions. i think i will try nim again but not use msvc this time. is there a guide to installing nim with an already-existing working mingw64 installation? (i have gcc 6.2, non-posix)
13:44:40cncli'll release the code once i get it working and replace its ncurses dependency with pdcurses or something else
13:44:55cncl(i can't remember why on earth i chose ncurses at the time...)
13:45:49dom96cncl: yay! Nice to see you here :)
13:46:13cnclgreetings
13:46:57cnclthe last version i used was like 0.11 or something
13:47:04dom96cncl: I think the best way for you is to use one of the zip files here: https://nim-lang.org/download.html
13:47:34dom96We still have a bit of work ahead of us to streamline Nim's installation process, especially on Windows.
13:47:37*bjz_ joined #nim
13:47:54cncli'm using a home-rolled mingw64 so i'm pretty used to suffering
13:47:54dom96But hopefully you won't run into trouble by using the "zip file" method
13:48:09dom96heh
13:48:18cnclalright, i'll try that first. does it include its own mingw64? or will finish.exe find the one i already have
13:48:26dom96finish.exe will find it
13:48:30cnclnice, ok
13:48:39dom96If it complains it's possible that it's simply wrong.
13:48:54dom96Especially if you have a home-rolled mingw64
13:49:10*yglukhov joined #nim
13:49:36*bjz quit (Ping timeout: 240 seconds)
13:51:13cnclit's one i modified from msys2 for doing some complicated stuff with qt cross compilation. i'm hoping it will work, though
13:53:05cnclhttps://gist.github.com/randrew/f573661fd66f304e2a786b150c03a77b
13:53:29*vlad1777d joined #nim
13:54:08subsetpark...But damn, mapIt is way more convenient than lc :)
13:54:47dom96cncl: yeah... finish.exe is surprisingly unstable.
13:55:19dom96cncl: it doesn't do anything that you can't do yourself though. Just add Nim to your PATH and you should be good to go. I assume you managed to compile it?
13:55:36cnclnim --version works
13:56:24cncli guess i should try compiling something
14:02:01*yglukhov quit (Remote host closed the connection)
14:02:35*yglukhov joined #nim
14:03:34*yglukhov quit (Remote host closed the connection)
14:03:48*yglukhov joined #nim
14:04:47dom96cncl: working?
14:06:56*yglukhov quit (Remote host closed the connection)
14:07:10Jipok[m]wow
14:07:19Jipok[m]it's working
14:07:23Jipok[m]https://riot.im/develop/#/room/#freenode_#nim:matrix.org
14:07:31*yglukhov joined #nim
14:08:09dom96Jipok[m]: matrix bridge?
14:08:27cncldom96: seems to work
14:08:31Jipok[m]+
14:08:48dom96cncl: great
14:09:50Jipok[m]Someone wants to write a client for the matrix?
14:09:51Jipok[m]on nim
14:10:44*Gilga quit (Ping timeout: 255 seconds)
14:11:46cncldom96: the .exe seems kind of big though. i remember them being smaller? 432kb for hello world (release build)
14:11:49*yglukhov quit (Ping timeout: 256 seconds)
14:12:19dom96Jipok[m]: why don't you write one? :)
14:13:11dom96cncl: I'm sure you can get it down. https://hookrace.net/blog/nim-binary-size/
14:13:19dom96But I don't understand why it matters so much
14:13:45cncli often commit compiled binaries into source repos for other people to use as part of a toolchain
14:14:00cnclusually simple command line tools for things that artists run to rebuild assets
14:14:58cnclif there are like 15 - 20 of these tools it can get annoying if the size for each one gets large. it's possible to put them all into a monolithic tool but then it changes frequently and people can get annoyed. not a problem that everyone would have, but it's something i sometimes have to deal with
14:16:06dom96That article should help you get the size down
14:16:13cnclah strip does remove a lot of stuff, i forgot how much of the runtime is unused
14:16:29dom96although I don't know how much it applies to windows
14:21:03Araq"i often commit compiled binaries into source repos for other people to use as part of a toolchain" -- thumbs up! :-)
14:21:18cnclhaha. it's the only way to fly
14:21:30demi-lolsob
14:21:37*byte512 joined #nim
14:21:38Araqwe have lots of usuability issues just because the git nazis are against binaries in git repos
14:21:50Araq*usability
14:22:58dom96sure, it was much better a couple of years ago when a Nim clone was multiple GBs in size.
14:23:22*arnetheduck joined #nim
14:23:23Araqthat had nothing to do with binaries.
14:24:06dom96yes it did
14:24:16*Arrrr quit (Quit: Leaving.)
14:24:23dom96the C sources were zipped up and placed in the repo
14:24:50dom96then updated periodically, but I guess you will argue that with the binaries you have in mind there will be no updates :)
14:25:19dom96feel free to set up a separate repo for them
14:25:36dom96if uploading them to a server is so hard
14:29:48Araqno, it's not "hard", it's not convenient for anybody. not for me, not for the users of my software.
14:31:41Araqpretty sure not the zip was placed in the repo, but the generated C sources.
14:32:18dom96it was a zip, because the C sources caused the diffs to become too huge.
14:35:54Araqok, that was a bad idea :P
14:36:17Araqbut I'll add DLLs to ui anyway
14:37:04dom96please don't
14:41:57*Gilga joined #nim
14:42:52Araqwhy not?
14:43:08Araqthink 3MB will become "gigabytes" anytime soon?
14:44:09*yglukhov joined #nim
14:46:32dom96You don't need it in Git's history. What is so inconvenient about placing it on a server?
14:51:32AraqI build the library for win32, win64, osx, linux 64bit. I don't have "upload to server" on all of these machines. I do have "git push" though.
14:52:08Araqbesides, Nimble doesn't really support external downloads yet either.
14:52:17yglukhovAraq: can you merge https://github.com/nim-lang/Nim/pull/5338 pls?
14:52:39dom96fine, so at least set up a separate repo for these DLLs
14:53:10dom96on the other hand, downloading filezilla onto these machines shouldn't take that much effort
14:53:41*bjz joined #nim
14:54:07*bjz_ quit (Ping timeout: 240 seconds)
14:55:35Araqyglukhov: tests are red :P
14:55:55yglukhovAraq: thats not me :P
14:57:48*gangstacat quit (Quit: Ĝis)
15:01:11Araqso ... how long does it travis to see exitCode == 0 means "build green"?
15:01:16Araq*it take
15:01:29*Gilga quit (Quit: WeeChat 1.5)
15:01:34cncldom96: let's say i wanted to use nim to parse a c or c++ source file and find invocations of a special macro. it doesn't have to be a perfect parse, just exclude uses of the macro which are in comments or strings
15:01:52cnclis there a parser in a nim library that would be useful for that?
15:02:32yglukhovAraq: an hour per job, as far as im concerned. but the problem is not the timeout, but script error
15:04:09*nsf quit (Quit: WeeChat 1.7)
15:04:14dom96cncl: don't know of any library, but maybe you could reuse c2nim's parser
15:05:04cnclok, thanks
15:07:14Araqc2nim's lexer does the job
15:07:28Araqparser would give you some AST you don't need
15:07:38*gangstacat joined #nim
15:08:56*yglukhov quit (Remote host closed the connection)
15:09:31*yglukhov joined #nim
15:09:33*carterza joined #nim
15:10:04*PMunch joined #nim
15:13:41*yglukhov quit (Ping timeout: 240 seconds)
15:21:29*yglukhov joined #nim
15:36:17*nsf joined #nim
15:41:25*yglukhov quit (Remote host closed the connection)
15:42:01*yglukhov joined #nim
15:46:09*yglukhov quit (Ping timeout: 245 seconds)
15:47:28*yglukhov joined #nim
15:51:17*yglukhov quit (Remote host closed the connection)
15:51:50*yglukhov joined #nim
15:56:39*yglukhov quit (Ping timeout: 256 seconds)
16:03:53cheatfate3 hours ago travis made some PR to issue dom96 posted yesterday, so maybe its time to check commit?
16:15:00rokupsAraq: turns out align pragma is not even implemented. want to guide me through the process? i already enabled it in pragmas.nim for types and vars. not sure which part of codegen to poke now.
16:19:13*yglukhov joined #nim
16:24:55*stisa quit ()
16:27:47*yglukhov quit (Remote host closed the connection)
16:28:21*yglukhov joined #nim
16:29:29*stisa[m] joined #nim
16:33:12Araqrokups: ccgtypes.nim I guess
16:33:21*yglukhov quit (Ping timeout: 252 seconds)
16:35:11rokupsAraq: yup, found the spot. thing is computeSizeAux() in types.nim overwrites alignment set by pragma. as temp measure i added check against default alignment of 2 but its wrong. besides compiler already uses that typ.align field though i dont quite grasp for what purpose exactly. maybe its different thing and that field is not a thing align pragma should
16:35:11rokupseven set?
16:37:11*nsf quit (Quit: WeeChat 1.7)
16:38:07*planhths quit (Quit: Konversation terminated!)
16:45:06Araqwell align should default to 0 and computeSizeAux touches it if it's 0
16:50:06*vlad1777d quit (Remote host closed the connection)
16:52:05rokupsAraq: ill try to implement that then. but when im done compiler would automatically insert custom alignment for structs all over the place because computeSizeAux() would set alignment. is this what should be happening?
16:52:09*vlad1777d joined #nim
16:52:40*irrequietus quit (Remote host closed the connection)
16:56:06*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
16:56:49Araqyeah, won't cause problems
16:56:57Araqwe don't use align for code generation
16:57:03*yglukhov joined #nim
16:57:07rokupswhat is it used for?
16:57:17Araqnothing
16:57:38Araqfor "correct" sizeof computations
16:57:54Araqthat I never got right and so 'sizeof' is not available at compiletime for complex types
16:58:16rokupsi see
16:59:57*irrequietus joined #nim
17:11:24rokupsactually as i understand currently setting packing on some struct will set it for all following structs as well. scary. luckily we can fix that with using __declspec(align(1)) instead of #pragma pack(1) ;)
17:15:09*Varriount|Mobile quit (Ping timeout: 252 seconds)
17:17:46*yglukhov quit (Remote host closed the connection)
17:18:21*yglukhov joined #nim
17:19:33*Jesin quit (Ping timeout: 252 seconds)
17:19:56*yglukhov quit (Remote host closed the connection)
17:20:11*yglukhov joined #nim
17:22:57*Jesin joined #nim
17:31:31*carterza quit (Quit: carterza)
17:54:52*Arrrr joined #nim
17:54:52*Arrrr quit (Changing host)
17:54:52*Arrrr joined #nim
18:32:15*arnetheduck quit (Remote host closed the connection)
18:33:49*yglukhov quit (Remote host closed the connection)
18:34:23*yglukhov joined #nim
18:38:16*yglukhov quit (Remote host closed the connection)
18:38:33*yglukhov joined #nim
18:40:10*yglukhov quit (Remote host closed the connection)
18:40:44*yglukhov joined #nim
18:44:49*yglukhov quit (Ping timeout: 240 seconds)
18:45:05cheatfaterokups, this is bug
18:45:15cheatfateit needs to be fixed
18:46:10*couven92 quit (Quit: Client disconnecting)
18:46:21cheatfaterokups, and this packing mostly used to crafting network packets
18:46:28cheatfatebecause they don't need to be aligned in memory
18:53:49cheatfatehttps://github.com/nim-lang/Nim/issues/4763
19:00:45*themagician joined #nim
19:05:56*yglukhov joined #nim
19:08:37*byte512 quit (Ping timeout: 240 seconds)
19:11:39*couven92 joined #nim
19:35:50*carterza joined #nim
19:42:59*Jesin quit (Quit: Leaving)
19:44:53*Jesin joined #nim
19:57:24*vitalyx joined #nim
20:02:50*unlaudable joined #nim
20:11:36*unlaudable quit (Ping timeout: 240 seconds)
20:17:56*Arrrr quit (Quit: Leaving.)
20:21:11*unlaudable joined #nim
20:21:12demi-also used to map memory without having to build out constructors :P
20:22:07*rokups quit (Quit: Connection closed for inactivity)
20:36:41*unlaudable quit (Ping timeout: 255 seconds)
20:42:50*vitalyx quit (Quit: vitalyx)
20:45:12cheatfateomg, travis still has this bug
20:55:37*bjz joined #nim
20:55:53Araqyup, maybe we should make our build less verbose
20:56:11Araqmaybe it's just the fact that's longer than 1000 lines
21:00:02cheatfateI think we need to remove this `Copying` lines
21:00:20cheatfatebut i dont hink it helps us with travis bug
21:00:25cheatfatethink*
21:18:47*nsf joined #nim
21:19:17*Arrrr joined #nim
21:21:23*yglukhov quit (Remote host closed the connection)
21:21:58*yglukhov joined #nim
21:22:08*yglukhov quit (Remote host closed the connection)
21:22:23*yglukhov joined #nim
21:24:30*Ven joined #nim
21:41:50cheatfateheh, i'm disappointed `micro` don't show me any colors for some reason
21:43:04*Ven quit (Read error: Connection reset by peer)
21:43:27demi-cheatfate: ? you may need a theme set?
21:44:25cheatfateset colorscheme xxx (monokai, solarized) doesn't help
21:44:38cheatfatelooks like problem with my putty client and some terminal strings
21:48:15cheatfatedemi-, is it possible to run micro in some kind of debug mode?
21:48:38cheatfatebecause everything is works but in pure white color
21:49:24demi-not that i'm aware of -- yeah it is highly dependent on terminal settings and env
21:54:36cheatfateannoying `set colorscheme monokai` shows only colored comments
21:54:47cheatfate`set colorscheme simple` everything is white
21:54:52demi-what is your TERM set to?
21:58:24*chemist69 quit (Ping timeout: 252 seconds)
22:11:21*def-pri-pub joined #nim
22:24:32*libmanCloud joined #nim
22:25:21*chemist69 joined #nim
22:30:43*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
22:42:13*bjz joined #nim
22:43:13Araqcheatfate: I'm considering moving to a different CI host
22:43:29cheatfateAraq, how?
22:46:20Araqwell travis is not the only website that can run a yml file with github integration
22:49:58Araqbut good night
22:53:58*Vladar quit (Quit: Leaving)
23:10:24*Snircle quit (Quit: Textual IRC Client: www.textualapp.com)
23:10:55carterzagitlab CI is nice
23:13:55*yglukhov quit (Remote host closed the connection)
23:23:46*Arrrr quit (Quit: Leaving.)
23:27:46*nsf quit (Quit: WeeChat 1.7)
23:28:02carterzacould someone help me understand how to call this proc
23:28:13carterzaproc nvgTextBounds*(ctx: NVGcontextPtr; x: cfloat; y: cfloat; string: cstring;
23:28:13carterza `end`: cstring; bounds: ptr cfloat): cfloat {.nvg.}
23:28:47carterzawhat do I provide for the bounds argument
23:29:08carterzait wants an array of four elements
23:29:15carterzaor that is what the method is supposed to fill
23:42:52*bjz quit (Quit: My MacBook Pro has gone to sleep. ZZZzzz…)
23:49:37*pleiosaur quit (Ping timeout: 240 seconds)
23:50:24def-pri-pubcartersa: can you give me a link to the original NanoVG docs?
23:50:25*pleiosaur joined #nim
23:54:02*couven92 quit (Read error: Connection reset by peer)