Re: Emacs pretest 28.0.90 is out

[Po Lu]

Eli Zaretskii <eliz@gnu.org> writes:

> There's at-func.c in lib/, which implements the *at functions.  But
> msdos.c was emulating them on its own, see readlinkat, faccessat, and
> other similar functions there.  Their code is quite boilerplate, so it
> shouldn't be hard to add a couple more of them.  You can use the code
> in w32.c as inspiration (but without the conversion of file names from
> UTF-8).  That should take care of openat and fchmodat.

> I wouldn't recommend implementing copy_file_range, unless you mean the
> trivial implementation that returns an error indication, like the
> Gnulib version does.

Thanks, I made the part in fileio.c that makes use of it conditional on
!MSDOS, so on MS-DOS it will fall back to using read/write instead,

> As for the rest, my recommendation is to implement futimens based on
> DJGPP's setftime.

Thanks, but I'm confused with the DJGPP `struct ftime': do I have to
take leap years and the different lengths of each month into account
when setting the various members of a `struct ftime' from a time_t?

The comment here makes that uncertain:

    struct ftime {
      unsigned ft_tsec:5;       /* 0-29, double to get real seconds */
      unsigned ft_min:6;        /* 0-59 */
      unsigned ft_hour:5;       /* 0-23 */
      unsigned ft_day:5;        /* 1-31 */
                                     ^^
      unsigned ft_month:4;      /* 1-12 */
      unsigned ft_year:7;       /* since 1980 */
    }

Alternatively, is there some function I overlooked that populates
`struct ftime' from time_t?