05:14 <GitHub164> [zig] tiehuis pushed 1 new commit to master: https://git.io/fNf4U

05:14 <GitHub164> zig/master 82e9190 Marc Tiehuis: Update zig.parser benchmark program

14:45 <GitHub169> [zig] andrewrk pushed 2 new commits to master: https://git.io/fNJUX

14:45 <GitHub169> zig/master 9eb51e2 Andrew Kelley: fix crash on @ptrToInt of a *void...

14:45 <GitHub169> zig/master 42ba061 Andrew Kelley: std.Hashmap - don't use catch unreachable in tests

15:15 <GitHub121> [zig] andrewrk pushed 1 new commit to master: https://git.io/fNJLS

15:15 <GitHub121> zig/master 2ee67b7 Andrew Kelley: langref: docs for invalid error set cast and incorrect pointer alignment...

16:14 <GitHub150> [zig] andrewrk pushed 1 new commit to master: https://git.io/fNJsN

16:14 <GitHub150> zig/master 05f1ea3 Andrew Kelley: ZIG_DEBUG_COLOR=1 overrides tty detection for runtime stack traces

17:25 <jab> Anyone know any other languages that compile to C ABI compatibility?

18:32 <very-mediocre> iirc Nim

18:32 <very-mediocre> and I think Rust but I'm not sure if there are compromises there

18:33 <stratact> Rust just uses libc as a backend for libstd. I don't think it compiles to C per se

18:34 <stratact> although it's debug info tries to emulate C

19:47 <jab> I've heard Nim, not a fan of it. Much prefer Zig's direction

19:52 <very-mediocre> me too, I like c syntax more than ruby/python syntax

19:56 <jab> i also feel nim's development is unweildy and directionless

19:56 <jab> lots of addons

19:57 <very-mediocre> that's true, it got very convoluted over the last few years

19:57 <very-mediocre> i think they're experimenting with it a lot

19:58 <very-mediocre> what i like about Nim is how simple it is to do really basic OOP

19:59 <very-mediocre> this is old but still pretty close to how it is now: https://nim-by-example.github.io/types/objects/

19:59 <very-mediocre> it's really productive

20:01 <very-mediocre> iirc they use tuples as interfaces

20:03 <very-mediocre> i find that fast a OOP feedback loop is a big deal, I like to iterate through different basic architectures when beginning a project

20:05 <very-mediocre> but everything else in Nim got really complex, it's kind of paradoxical

20:05 <Bas_> Ah, that's nice that you can add methods after the fact basically.

20:07 <very-mediocre> Totally subjective, but I feel like Nim OOP does what Python failed to achieve

20:07 <Bas_> That's one of my complaints about doing OO the C++ way. That you in general can't add to the class afterwards. Unless they anticipated it by making the members protected instead of private and giving it a virtual destructor, which you don't actually want to do by default.

20:09 <very-mediocre> zig is probably going to be the best of both worlds in that regard (not productivity, but the thing that you said)

20:09 <very-mediocre> since it'll be static and you'll be able to add stuff "later" at compile time

20:09 <very-mediocre> without going full runtime vtables

20:10 <very-mediocre> or at least I think so, the metaprogramming features are far from complete of course.

20:12 <very-mediocre> I'm in a weird timezone so my brain is melting now, gonna go, bye :]

20:55 <andrewrk> I can't believe how easy it was to do the windows implementation of the std.event.Loop thread pool dispatching system

20:56 <andrewrk> windows > freebsd/macos > linux, for event-based APIs

21:14 <GitHub176> [zig] andrewrk pushed 1 new commit to master: https://git.io/fNJQP

21:14 <GitHub176> zig/master 0ac1b83 Andrew Kelley: fix non-portable format specifier

21:16 <GitHub120> [zig] wilsonk opened pull request #1211: Update throughput_test.zig. (master...fix_throughput_test) https://git.io/fNJQb

21:21 <GitHub146> [zig] andrewrk pushed 1 new commit to master: https://git.io/fNJ7G

21:21 <GitHub146> zig/master a2834d4 wilsonk: Update throughput_test.zig. (#1211)

21:21 <GitHub124> [zig] andrewrk closed pull request #1211: Update throughput_test.zig. (master...fix_throughput_test) https://git.io/fNJQb