Язык Zig (канал)
Nuke the C++ implementation of Zig from orbit using WASI #13560 The idea here is to use a small WASI binary as a stage1 kernel that is committed to source control and therefore can be used to build any commit from source. We provide a minimal WASI interpreter…
> We will be temporarily regressing async/await/suspend/resume in master branch.
Компилятор на C++ удалят ДО того, как добавят поддержку вышеуказанного в self–hosted компилятор
> The plan is to merge this into master branch before completing #6025. Users of async functions should stick to 0.10.0 for now.
Компилятор на C++ удалят ДО того, как добавят поддержку вышеуказанного в self–hosted компилятор
> The plan is to merge this into master branch before completing #6025. Users of async functions should stick to 0.10.0 for now.
GitHub
async/await/suspend/resume · Issue #6025 · ziglang/zig
This is a sub-task of #89. Proposal Re-introduce the async/await language feature which existed in the bootstrap compiler, with the same semantics as before. Competes with #23446.
🔥1
Язык Zig (канал)
Nuke the C++ implementation of Zig from orbit using WASI #13560 The idea here is to use a small WASI binary as a stage1 kernel that is committed to source control and therefore can be used to build any commit from source. We provide a minimal WASI interpreter…
Merged два часа назад 🎉🎉🎉
https://github.com/ziglang/zig/commit/e7d28344fa3ee81d6ad7ca5ce1f83d50d8502118
https://github.com/ziglang/zig/commit/e7d28344fa3ee81d6ad7ca5ce1f83d50d8502118
GitHub
Merge pull request #13560 from ziglang/wasi-bootstrap · ziglang/zig@e7d2834
Nuke the C++ implementation of Zig from orbit using WASI
🔥7
In the Zig project, we closed 19% of total open issues during December this year.
This was not the work of a "stale bot", but of humans who painstakingly verified that each and every issue was resolved and had test coverage if applicable.
Finally, our technical debt is being paid off and we can reach for some ambitious goals in 2023.
https://mastodon.social/@andrewrk/109621905821062620
This was not the work of a "stale bot", but of humans who painstakingly verified that each and every issue was resolved and had test coverage if applicable.
Finally, our technical debt is being paid off and we can reach for some ambitious goals in 2023.
https://mastodon.social/@andrewrk/109621905821062620
👍2
In about 30 minutes I'll stream working towards the Zig Package Manager.
This will be a yak-shaving stream - there is much to do before we can work on the package manager logic itself.
https://twitch.tv/andrewrok/
https://mastodon.social/@andrewrk/109627337708582296
This will be a yak-shaving stream - there is much to do before we can work on the package manager logic itself.
https://twitch.tv/andrewrok/
https://mastodon.social/@andrewrk/109627337708582296
Twitch
andrewrok - Twitch
Lead developer & president of Zig Software Foundation.
👍1
Язык Zig (канал)
In about 30 minutes I'll stream working towards the Zig Package Manager. This will be a yak-shaving stream - there is much to do before we can work on the package manager logic itself. https://twitch.tv/andrewrok/ https://mastodon.social/@andrewrk/109627337708582296
Стрим уже начался https://www.twitch.tv/andrewrok
Twitch
andrewrok - Twitch
Lead developer & president of Zig Software Foundation.
For those who haven't followed the discusison in #berlin, March 1st we're going to have a one-off event hosted by Delivery Hero.
We're going to have 3 short talks by Adam (the organizer), Joachim (everyone's favorite german-singaporean, also core team member), and me. The talks are going to be accessible for newcomers so don't be afraid.
Delivery Hero is offering food, drinks and afterwards we're going out to a pub with those who want to spend a bit of extra time together.
https://www.meetup.com/zigberlin-meetup-group/events/290511497
We're going to have 3 short talks by Adam (the organizer), Joachim (everyone's favorite german-singaporean, also core team member), and me. The talks are going to be accessible for newcomers so don't be afraid.
Delivery Hero is offering food, drinks and afterwards we're going out to a pub with those who want to spend a bit of extra time together.
https://www.meetup.com/zigberlin-meetup-group/events/290511497
👍1
Zig tips: v0.11 std.build API / package manager changes
https://devlog.hexops.com/2023/zig-0-11-breaking-build-changes/
#блоги
https://devlog.hexops.com/2023/zig-0-11-breaking-build-changes/
#блоги
Hexops
Zig tips: v0.11 std.build API / package manager changes
With Zig v0.11 will come several changes to the std.build API. We've just updated Mach to the latest nightly Zig version, and wanted to provide some tips on how to update your own Zig code.
👍1
Язык Zig (канал)
https://kristoff.it/blog/zig-self-hosted-now-what/ #upstream #блоги
А вот и "new for loop syntax" из поста :)
https://github.com/ziglang/zig/pull/14671
zig fmt пофиксит старый код 😋
#upstream
P.S. тут ответ на более ранний пост в канале, но в чате походу не видно
upd: Смержили 🎉🎉🎉⚡⚡⚡
https://github.com/ziglang/zig/pull/14671
zig fmt пофиксит старый код 😋
#upstream
P.S. тут ответ на более ранний пост в канале, но в чате походу не видно
upd: Смержили 🎉🎉🎉⚡⚡⚡
GitHub
implement multi-object for loops by andrewrk · Pull Request #14671 · ziglang/zig
This exposed a latent bug in the (experimental) x86 backend but I'm not letting it block the merge of this PR, since all the other backends are fine.
Closes #7257.
merge checklist:
implement ...
Closes #7257.
merge checklist:
implement ...
🎉4🤔1
Кстати, совсем забыл: я тут вместе с Gso321 (он начал, я закончил) запилили на Gentoo Wiki статью про Zig:
https://wiki.gentoo.org/wiki/Zig
https://wiki.gentoo.org/wiki/Zig/ru
Там (само собой) в основном генто-специфичная информация, но какая-то информация может быть полезна всем пользователям :)
https://wiki.gentoo.org/wiki/Zig
https://wiki.gentoo.org/wiki/Zig/ru
Там (само собой) в основном генто-специфичная информация, но какая-то информация может быть полезна всем пользователям :)
🔥4❤1👏1
Ещё я тут обвязал libgccjit в Zig, вроде выглядит идиоматично :)
https://git.sr.ht/~bratishkaerik/zig-libgccjit
Там осталось несколько мелочей, но библиотека уже сейчас юзабельна, в папке "examples/" есть 4 из 5 портированных примера с C на Zig, из официального сайта libgccjit (сам примеры выглядят не так идиоматично, потому чтоя пытался воссоздать примеры строка-в-строку)
для сравнения:
https://git.sr.ht/~bratishkaerik/zig-libgccjit/tree/a287291ba38cdf99ebac08b5fef745b547ef177f/item/examples/gcc_tutorial_5.zig#L24
const char_ptr_ptr_type: *GccJit.Type = ctxt.getType(.char).getPointer().getPointer();
https://gcc.gnu.org/git/?p=gcc.git;a=blob;f=gcc/jit/docs/examples/tut05-bf.c;h=f948ede796a48dbae6e903c3b4b99cf2d5d4a103;hb=HEAD#l23
gcc_jit_type *char_ptr_ptr_type =
gcc_jit_type_get_pointer (
gcc_jit_type_get_pointer (
gcc_jit_context_get_type (ctxt, GCC_JIT_TYPE_CHAR)));
#библиотеки
https://git.sr.ht/~bratishkaerik/zig-libgccjit
Там осталось несколько мелочей, но библиотека уже сейчас юзабельна, в папке "examples/" есть 4 из 5 портированных примера с C на Zig, из официального сайта libgccjit (сам примеры выглядят не так идиоматично, потому чтоя пытался воссоздать примеры строка-в-строку)
для сравнения:
https://git.sr.ht/~bratishkaerik/zig-libgccjit/tree/a287291ba38cdf99ebac08b5fef745b547ef177f/item/examples/gcc_tutorial_5.zig#L24
const char_ptr_ptr_type: *GccJit.Type = ctxt.getType(.char).getPointer().getPointer();
https://gcc.gnu.org/git/?p=gcc.git;a=blob;f=gcc/jit/docs/examples/tut05-bf.c;h=f948ede796a48dbae6e903c3b4b99cf2d5d4a103;hb=HEAD#l23
gcc_jit_type *char_ptr_ptr_type =
gcc_jit_type_get_pointer (
gcc_jit_type_get_pointer (
gcc_jit_context_get_type (ctxt, GCC_JIT_TYPE_CHAR)));
#библиотеки
Data Oriented Parallel Value Interner
https://matklad.github.io/2023/04/23/data-oriented-parallel-value-interner.html
#блоги
https://matklad.github.io/2023/04/23/data-oriented-parallel-value-interner.html
#блоги
matklad.github.io
Data Oriented Parallel Value Interner
In this post, I will present a theoretical design for an interner.
It should be fast, but there will be no benchmarks as I haven't implemented the thing.
So it might actually be completely broken or super slow for one reason or another.
Still, I think there…
It should be fast, but there will be no benchmarks as I haven't implemented the thing.
So it might actually be completely broken or super slow for one reason or another.
Still, I think there…
👍3