Re: C and C++, promotion, stabilization, migrationFor embedded

On 8/25/2024 12:27 AM, Lawrence D'Oliveiro wrote:

On Sat, 24 Aug 2024 22:49:24 -0400, Arne Vajhøj wrote:

On 8/24/2024 10:47 PM, Lawrence D'Oliveiro wrote:

On Sat, 24 Aug 2024 22:39:44 -0400, Arne Vajhøj wrote:

I fully get the need for UTF-8 support in anything data related.

>
I wonder what you would consider to *not* be “data related” ...

>
Why wonder when I provided plenty of examples of what was
in and what was out?

 I didn’t see the connection. You don’t count “file names, in usernames, in
logicals, in identifiers” as “data”?

No. Or at least not in the business relevant sense.
The users care that the XML file get delivered in an encoding that
the recipient system can parse correctly. They do not care what
identifiers are used in the program creating it and they probably
prefer the filename to be something not causing problems anywhere
(all ASCII, no spaces etc.).
Seen from their perspective what is in that XML file is data - all
the surrounding stuff is not. Obviously the source code and
file system are data from the compiler and OS perspective. But
the OS and compiler got purchased to deliver that XML file to
the users.
It may likely be a requirement to sell the system that it can produce
a file with <data>Blåbærsyltetøj</data>.
It is highly unlikely to impact the sale whether the file can
be named Blåbærsyltetøj.txt and whether the file handle in the
program can be called Blåbærsyltetøjfile.
Arne