ChanServ changed the topic of #zig to: zig programming language | ziglang.org | be excellent to each other | channel logs: https://irclog.whitequark.org/zig/
thither has left #zig ["WeeChat 1.6"]
<GitHub31>
[zig] bnoordhuis opened pull request #1094: support `--target-arch wasm32` (master...wasm) https://git.io/vh2y6
<bheads___>
I have always hated that as well about c and c++
<andrewrk>
the only macros that are intended to be used by users are the aliases for native and foreign endian. and even then I probably should have made it enums or something
<andrewrk>
ok and SOUNDIO_MAX_CHANNELS because C doesn't let you use static consts as array lengths
<bheads___>
I also hate how macros are how you handle cross platform, and build logic. But ehhh the 70s
<donlzx>
hi andrewrk, I'm new to Zig. Upon reading the doc the first time, I couldn't tell what is the difference between strict and optimized float modes. It would be better that these concepts were briefly explained in the langref.html#Floats section before the example codes.
<andrewrk>
donlzx, I see, maybe the docs that are under @setFloatMode can be more prominently advertised in the main float section
<bheads___>
I bet you cant wait to port that lib to zig! But looks like a good library to me.
<andrewrk>
it's tempting, but there's a big hurdle to overcome: on linux, alsa-lib kind of *is* the kernel API to the sound system
<andrewrk>
linux sound essentially depends on alsa-lib
<andrewrk>
and alsa-lib does things like dump log lines to stdout, and it has a lisp interpreter in it
<andrewrk>
I'm sure it has some panics on out of memory in there
<bheads___>
Ugh... What about pulse?
<andrewrk>
one of my projects will be to look into avoiding alsa lib
<andrewrk>
pulse should be reasonable to have a pure zig lib for it, assuming the protocol is reasonably stable
<andrewrk>
the protocol between libpulseaudio and pulseaudio-server
<bheads___>
I would think so, its been around for a long time now
<bheads___>
Linux legacy is such a crux, even wayland has failed to replace x11.
<GitHub84>
zig/master dc8bda7 Marc Tiehuis: Add arbitrary-precision integer to std...
dbandstra has quit [Quit: Leaving]
jjido has joined #zig
IntoxicatedHippo has quit [Read error: Connection reset by peer]
IntoxicatedHippo has joined #zig
quc has joined #zig
atk has quit [Ping timeout: 240 seconds]
atk has joined #zig
davr0s has joined #zig
isd has quit [Quit: Leaving.]
donlzx has quit [Remote host closed the connection]
ysengrimm has quit [Quit: Ping timeout (120 seconds)]
ysengrimm has joined #zig
jjido has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
tiehuis has joined #zig
davr0s has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
<tiehuis>
bb010g: thanks for those aur patches, i've updated the zig/zig-git packages to omit the stage2 artifacts
<tiehuis>
namcap is also handy, wasn't aware of it, i've not changed the dependencies for the moment though since there is no harm in them and i haven't checked if the xml2 was fixed in the system llvm yet
tiehuis has quit [Quit: WeeChat 2.1]
mahmudov has joined #zig
<bb010g>
tiehus: Since you didn't make git a makedepend, this still fails when building without git installed
tiehuis has joined #zig
<tiehuis>
thanks, fixed
tiehuis has quit [Quit: WeeChat 2.1]
davr0s has joined #zig
<bb010g>
tiehuis: It still fails when building in a chroot because you didn't properly place your patches in the source array, as recommended in https://wiki.archlinux.org/index.php/Patching_packages . My PKGBUILD does fully work in a chroot, if you just want to use that instead of continually modifying the old one.
<bb010g>
Also, the libxml2 change makes no difference besides that namcap lint because the llvm package depends on libxml2.
Ichorio has joined #zig
jjido has joined #zig
davr0s has quit [Quit: My MacBook Pro has gone to sleep. ZZZzzz…]
jjido has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
very-mediocre has quit [Ping timeout: 260 seconds]
hoppetosse has joined #zig
bheads has joined #zig
davr0s has joined #zig
_whitelogger has joined #zig
mahmudov has quit [Remote host closed the connection]
<GitHub115>
zig/master 854f90a tgschultz: Added C string support to fmt by using "{s}". The format string requirement is for saftey. (#1092)
<GitHub13>
[zig] andrewrk closed pull request #1092: Add C string support to std.fmt w/ "{s}" (master...zig-std.fmt-c-string) https://git.io/vh2rC
isaachier has joined #zig
Tanner_ has joined #zig
<Tanner_>
I love the idea of using Zig because it seems like a blatantly better C, but my use case requires cross compilation. Can I do that without the build system for now?
<Tanner_>
I have compiled Zig from source.
<isaachier>
Tanner_ zig is inherently a cross compiler
<Tanner_>
Yes, I know, but there isn't documentation on using that feature.
<clownpriest>
benchmark code is in the test folder. maybe i set thing up improperly? i think for the most part i'm comparing apples to apples
hoppetosse has quit [Ping timeout: 240 seconds]
<byronh__>
Hummm only big difference i see (and i am on a phone) is that it looks like your not including the init call in the benchmark. Is the reference c or c++?
<byronh__>
Other thing to make sure is maybe use clang6 to compile the reference implementation, maybe llvm6 is giving a performance boost
<byronh__>
But yeah awsome you got xxhash lib. I have been thinking about a zlib zig lib
<dbandstra>
i'm playing with zlib atm
<byronh__>
Ooo... I will definitely be interested in what you come up with
<clownpriest>
byronh__, the init call is definitely outside of the benchmark, you think i should include it? i think the bare XXH64 function in C is also stateless
<clownpriest>
whereas the streaming API would have a stateful init as well
<clownpriest>
but yeah, i should probably test an llvm6 compile C version. tho i doubt it could make up that large of a gap, but could definitely be wrong
<byronh__>
I would messure the init, yeah i doubt it will make a difference but better to be sure
<clownpriest>
just tried, still faster than C on all load sizes
<clownpriest>
tho slightly slower than current numbers
<clownpriest>
(with init included)
<byronh__>
Nice!!
<dbandstra>
does anyone know how to link against a c library (like zlib) in the `zig test` or `zig build-exe` command? i only have it working with a build.zig file
<clownpriest>
either way, the checksum() function in the zig impl doesn't actually need an init() call
<clownpriest>
all state is in the function
<dbandstra>
i use --library but then i get `lld: error: unable to find library -lz`