On Thu, Apr 16, 2026 at 06:12:24PM +1000, Seth McDonald wrote:
(Why not make B const? Idfk ask the glibc devs.)
because that's most convenient on the user side (_somebody_ has to do
that cast if the object is to be modified), and consistent with the
"normal" memchr(), strchr(), and other similar functions.
c23 aims to resolve this issue using a c11 feature.
there is a pertinent thread on the mutt-dev list right now:
https://marc.info/?l=mutt-dev&m=177631784909117&w=2 and
https://marc.info/?l=mutt-dev&m=177632421112239&w=2 .
regarding the commit message itself, it's way too wordy for my taste
now, but still misses the key point: how does the implementation compare
to what glibc itself actually does?
_______________________________________________
isync-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/isync-devel
