←back to thread

Reflection for C++26

(isocpp.org)
196 points svlasov | 1 comments | 30 Jun 24 11:26 UTC | HN request time: 0.21s | source
Show context
lallysingh ◴[01 Jul 24 23:00 UTC] No.40851756[source]▶
Wow this got really long. I was one of the coauthors for a reflection proposal (N3340) over a dozen years ago. Implementing compile-time reflection is honestly trivial - you basically transfer data from the symbol table on-demand into template specializations. It was roughly 1500 LOC to modify g++ to do it.

Looking at the examples (https://isocpp.org/files/papers/P2996R4.html#examples) what really stands out is the direct integration of type-syntax into the language. It fits in with a certain token-substitution way that connects back to templates. It also replaces some of the uglier operators (typeof?).

I hope it goes int! During the language's stagnation I left for a while, perhaps it'll be competitive again soon.

replies(2): >>40851990 #>>40855315 #
stiglitz ◴[01 Jul 24 23:36 UTC] No.40851990[source]▶
By ”stagnation” do you mean “not getting new features”?
replies(3): >>40852374 #>>40852457 #>>40852476 #
jacoblambda ◴[02 Jul 24 00:48 UTC] No.40852476[source]▶
C++ has gotten a ton of quality of life features with each update. The issue is less that new features aren't coming and more that new features bake through countless iterations of proposals for close to or often over a decade until everyone in WG21 is happy.

So it's not that we aren't getting features. They are coming quite fast and people regularly complain that new C++ has too many things for them to learn and keep up with. The issue is that those are the same features everyone has been asking for for over a decade so the people that really care found workarounds and eventually move over to the new std way of doing things when they can while everyone else continues waiting for that one feature they really care about.

replies(7): >>40853490 #>>40854139 #>>40854365 #>>40854369 #>>40854878 #>>40855193 #>>40856663 #
lenkite ◴[02 Jul 24 13:50 UTC] No.40856663[source]▶
Wish C++ fixed some of its mistakes in the standard library. std::regex is embarrassing when even Python can beat it and nobody uses std::unordered_map due to its pointer chasing. Basic Maps are something that you shouldn't need to use a third party library for.
replies(1): >>40856821 #
ghosty141 ◴[02 Jul 24 14:04 UTC] No.40856821[source]▶
Could you explain the issue with unordered_map?
replies(1): >>40856883 #
1. lenkite ◴[02 Jul 24 14:11 UTC] No.40856883[source]▶
The standard guarantees that you can get a pointer to an object in the map and this pointer will remain valid after rehashing and insertion/deletion of other elements. That basically forces implementers to use “buckets with linked lists”, also known as separate chaining. This is not cache friendly (and also suffers from excessive allocation). Other hash map implementations like abseil's don't provide this guarantee so they can put everything right there inline, which is much more efficient.