On Wed, 28 Aug 2024 at 09:52, Tiago Freire <tmiguelf_at_[hidden]> wrote:
>
> It's highly dismissive of other people's field experience other than Gaspar's.

Oh, that's hilarious. First there are comments suggesting that there's
abuse and self-inflicted wounds by people who haven't
even seen what the builds of the people running into #pragma once
problems do and why, and fair amounts of claims that they
are just doing it wrong, and then the same people complain about dismissiveness.

Gasper isn't the only person doing these things. As I said, other
people were reported to be doing that, by implementation vendors
who tried (and failed) to deal with the problems reported to them by
users, and those people weren't Gasper or his colleagues.

> I have countless times faced and fixed this problem production code, I guarantee you Gaspar's is not special. I would be more than happy to fix it for a fee.
> I think I have sufficient reasons to be skeptical.
>
> If you have a problem, you either come up with a concrete example that we can all look at that isn't locked behind closed doors (and then we can agree, yes this is a problem, or maybe not, this isn't a problem you should have done this), or just admit you don't have an example to show.

I'm more than happy to admit I don't have an example to show. I don't
run into these problems myself. Yet they cause me to oppose
standardising #pragma once. I have thus far seen zero reasons to
change that take, in this discussion or a decade ago when I was
listening
to that implementation vendor report of their users running into
exactly the same problems that Gasper would be running into if he'd
use
#pragma once.

And that's the situation we're in. #pragma once as a standardized
facility has been rejected before, and there is no new information
in this proposal or in this discussion.