On Friday, 25 September 2020 at 16:01:32 UTC, Per Nordlöw wrote:
> I just added support for Apple's Swift. It's massively slow on Linux. Check is 61 times slower than dmd and build is 42 times slower than dmd.

The update comments at [1] gives some clues to problems with "Big Agenda Languages" in general och Swift in particular. Jonathan Blow explains the meaning of the term "Big Agenda Languages" at [2].

This should be mentioned when people ask us why we work for Dlang.

[1] https://stackoverflow.com/questions/25537614/why-is-swift-compile-time-so-slow
[2] https://github.com/BSVino/JaiPrimer/blob/master/JaiPrimer.md

On Thursday, 20 August 2020 at 20:50:25 UTC, Per Nordlöw wrote:
> After having evaluated the compilation speed of D compared to other languages at
>
>     https://github.com/nordlow/compiler-benchmark
>
> I wonder; is there any language that compiles to native code anywhere nearly as fast or faster than D, except C?

You should have a look at the self-hosted Zig compiler as well. I'm not sure if it's mature enough to benchmark (it currently only supports a small set of Zig). You would probably need to compile it from source as well.

Have a look at this post [1] I made, that mentions how the Zig compiler uses incremental compilation. Even a full build (non-incremental) seems really fast. Although, I don't know how well it scales.

[1] https://forum.dlang.org/post/ctelroirrkqpkrlupajp@forum.dlang.org

--
/Jacob Carlborg

On Tuesday, 29 September 2020 at 08:03:07 UTC, Jacob Carlborg wrote:
> Have a look at this post [1] I made, that mentions how the Zig compiler uses incremental compilation.
>
> https://forum.dlang.org/post/ctelroirrkqpkrlupajp@forum.dlang.org

Thanks. Have you tried building and using the self-hosted version?

On Tuesday, 29 September 2020 at 08:23:28 UTC, Per Nordlöw wrote:

> Thanks. Have you tried building and using the self-hosted version?

No, I have not. I've looked at a couple of live coding videos about Zig and the self-hosted compiler.

--
/Jacob Carlborg

On Thursday, 27 August 2020 at 10:29:41 UTC, MrSmith wrote:
> void write[Args...](Args... args)

What is the pro of this syntax with `Args... args` compared to D's

void write(Args...)(Args args)

?

On Monday, 5 October 2020 at 14:19:36 UTC, Per Nordlöw wrote:
> What is the pro of this syntax with `Args... args` compared to D's
> void write(Args...)(Args args)
> ?

I think it was due to simpler implementation. This way you know that Args is variadic as early as at parse time.