On 4/29/21 6:39 AM, Peter Brett via SG16 wrote:

Hi Corentin,


I think this is a big improvement!


Does anyone know what “physical” means in [lex.phases]:1?  I feel like we could remove both occurrences of “physical” from this paragraph without any normative effect.

I don't have a reference to cite, but I interpret this as follows.  If we think of the standard as an abstract specification, the use of "physical" is what connects it to something that exists in the real world.  This allows the specification to be applied to, for example, code written on a napkin or a blackboard.  I won't lose any sleep over dropping "physical" here, but I can imagine a situation in which the ISO has some rule that specifications may not be completely abstract; that they have to describe something "real".  I have no idea if any such rule exists.



There is an obvious incomplete merge with P2314, in that the current P2295 wording suggests that UCN doesn’t occur for UTF-8 source files.   Just to clarify this, please could we insert a paragraph break before “Any source file character…”?


I don’t have any concerns about forwarding this paper to EWG in its current form.


Best regards,





From: SG16 <sg16-bounces@lists.isocpp.org> On Behalf Of Corentin via SG16
Sent: 29 April 2021 08:35
To: SG16 <sg16@lists.isocpp.org>
Cc: Corentin <corentin.jabot@gmail.com>
Subject: [SG16] P2295R3 Support for UTF-8 as a portable source file encoding



Per request in yesterday's meeting, 

here is P2295R3 Support for UTF-8 as a portable source file encoding.


I am looking forward to your feedback