Re: [PATCH] support ptrdiff_t if it has the same storage size as int, but does not alias it

"'Robin Haberkorn' via scintilla-interest" <[email protected]> Thu, 09 Oct 2025 22:21:36 +0000
Newsgroups gmane.comp.lib.scintilla.devel
Message-ID <[email protected]>
On Fri Oct 10, 2025 at 00:08:54 GMT +03, Neil wrote:
> Still looks to be too much of a change when a simple -D__HAIKU__ can avoid 
> most of it. It increases the proportion of code in headers leading to worse 
> build times and is a bad direction to move in.

Even if we could assume that ptrdiff_t is always long on 32-bit Haiku
and therefore doesn't alias int that would be a lot trickier on NetBSD where
apparently some ports use long and others int.
Also, do we really know that on Haiku? They support armv7 as well.

Am I supposed to write some kind of custom Autoconf compile test to check
for the aliasability of ptrdiff_t with int if sizeof(ptrdiff_t) == sizeof(int)
and pass in -D__HAIKU__ on all platforms that do?
Ugly, but possible. Then we should at the very least give it a more
neutral name like RUNSTYLES_NEEDS_PTRDIFF or something.

Robin

-- 
You received this message because you are subscribed to the Google Groups "scintilla-interest" group.
To unsubscribe from this group and stop receiving emails from it, send an email to [email protected].
To view this discussion visit https://groups.google.com/d/msgid/scintilla-interest/DDE54CVV6UM0.28ABM2AGR9BWO%40googlemail.com.
signature.asc (application/pgp-signature, 488 B)
-----BEGIN PGP SIGNATURE-----

iQEzBAABCAAdFiEEB/6AD3js1eGB4raKarMDYY4q6voFAmjoNXIACgkQarMDYY4q
6voVMAf+Ikic1X0ldAjrSIvmowtmnudaAuuKPJ56h1iwT+Q7U3aBEaeaAxb3k+BL
m1/OQ7iix26z5xk1KMq6dFNI0UGa4pWZKRjLDKkiw8CRkdhb1s8NpiyDHqnBq7YA
b5o14UEB4ypsUTDYjXXbo73KtSyu5gouP3OAR8HV9meOIeSenQ4HEIoR+YPIdDku
MCxq6mtPyiZiGJk8YbSvSMTpVddKo/Fn/vE3nMlNSLcJfhNIuiExzddwDgbZexUw
EUP777E0Dyl/IA5hL9VymnOpLfHOk1UtZooXFWYqZxKSfcfMzn42ICqfJDdXsu+g
y6y/b7mgsnWY8Ifn1v169ZAisdVkCQ==
=FLSo
-----END PGP SIGNATURE-----