On Mon, Jun 3, 2019 at 2:59 PM Chapman Flack <chap@anastigmatix.net> wrote:
> 1. (This one seems like a bug.) In the little-endian case, if
> SIZEOF_DATUM is smaller than the type, I'm not convinced by doing
> the DatumBigEndianToNative() after the memcpy(). Seems like that's
> too late to make sure the most-significant bytes got copied.
Uh, when else would you do it? Before the memcpy()?
> 2. (This one seems like an API opportunity.) If it becomes common to
> add abbreviation support for smallish types such that (as here,
> when SIZEOF_DATUM >= 8), an abbreviated "equality" result is in fact
> authoritative, would it be worthwhile to have some way for the sort
> support routine to announce that fact to the caller? That could
> spare the caller the effort of re-checking with the authoritative
> routine.
It's possible that that would make sense, but I don't think that this
patch needs to do that. There is at least one pre-existing cases that
does this -- macaddr.
--
Peter Geoghegan