https://github.com/Endilll closed
https://github.com/llvm/llvm-project/pull/94876
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/cor3ntin approved this pull request.
I think as a follow up we should get rid of the "exported templates are
unsupported" warning which makes no sense whatsoever and either use the c++20
error in all language modes or just a generic "unexpected 'export' keyword
here" diagno
https://github.com/Endilll updated
https://github.com/llvm/llvm-project/pull/94876
>From e4028ec6e70f6d86325393a8d03e407404643bc0 Mon Sep 17 00:00:00 2001
From: Vlad Serebrennikov
Date: Sun, 9 Jun 2024 00:59:09 +0300
Subject: [PATCH 1/3] [clang] Cover CWG issues about `export template`
---
cl
https://github.com/shafik commented:
LGTM after addressing coloring issues.
https://github.com/llvm/llvm-project/pull/94876
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Endilll wrote:
Thank you for chiming in!
> It might be fine to give a different color to superseded, but I would guess
> we'd want that to be a lighter form version of whatever the superseding issue
> is colored. e.g., given that conforming = green and non-conforming = red, if
> we have a sup
AaronBallman wrote:
The goal of the status page is to convey implementation status to our users,
and so from that perspective I think N/A provides the least information to
users because it basically says "this entry doesn't apply to us". So in these
kinds of cases, `sup` conveys more informati
Endilll wrote:
Me and Corentin discussed this offline. Two points emerged there:
1) He's more concerned with the fact that 204 and other Core issues are
highlighted green, inheriting the status of the issue that superseded them,
than with the fact that they are marked as superseded. Styles we u
Endilll wrote:
> We never implemented export template - and no one did, therefore we are not
> affected by changes to it (no one did except for EDG, i don't even know if
> they shipped it)
When I said "wording changes that do not affect implementations", I meant a
hypothetical conforming impl
cor3ntin wrote:
We never implemented `export template` - and no one did, therefore we are not
affected by changes to it (no one did except for EDG, i don't even know if they
shipped it)
It is perfectly reasonable not to have tests whatsoever for that either
(although I find the pre-c++20 diag
Endilll wrote:
Using `N/A` for 204, 323, and 335 would be novel, because at the moment it's
used for wording changes that do not affect implementations. Those Core issues
clearly affected the implementations back when they were resolved.
https://github.com/llvm/llvm-project/pull/94876
cor3ntin wrote:
I would prefer we say `N/A` rather than pretend we support something that was
never implemented (to the extend i think it might be better to have an error
about modules rather than exported templates in older language modes)
https://github.com/llvm/llvm-project/pull/94876
_
llvmbot wrote:
@llvm/pr-subscribers-clang
Author: Vlad Serebrennikov (Endilll)
Changes
This PR covers the following Core issues:
[CWG204](https://cplusplus.github.io/CWG/issues/204.html) "Exported class
templates"
[CWG323](https://cplusplus.github.io/CWG/issues/323.html) "Where must `expo
https://github.com/Endilll created
https://github.com/llvm/llvm-project/pull/94876
This PR covers the following Core issues:
[CWG204](https://cplusplus.github.io/CWG/issues/204.html) "Exported class
templates"
[CWG323](https://cplusplus.github.io/CWG/issues/323.html) "Where must `export`
appea
13 matches
Mail list logo
