[Bug c++/96570] Warnings desired for time_t to int coversions

["terra at gnome dot org" <gcc-bugzilla@gcc.gnu.org>]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=96570

--- Comment #4 from M Welinder <terra at gnome dot org> ---
> Explicit casts don't, and that's what I was questioning.

They most certainly do.  That's an empirical statement from having gone over a
fairly large code base.  It is not a statement that they should occur there and
there is likely nothing "inadvertent[ly]" about their presence.  "Mistaken" and
"ill-advised" are probably better descriptions, but the reasons have long since
been forgotten.  I.e., the code is buggy.

Look, you are being a bit defensive here which is strange as no-one is
attacking you.  Please try looking at the goal:

*** The goal here is to find time handling bugs.

Is that a worthy goal for gcc?

I will assert that there is a lot of buggy time handling code out there and
that fixing it will receive more and more attention over the next 15 years.

The compiler may or may not be the right tool to help find and fix these, but
gcc has in the past taken it upon itself to warn about other classes of
likely-wrong code and it is fairly well positioned to do so given its access to
a parse tree and type information.