On Mär 03 2023, Robert Elz via tz wrote:
Date: Fri, 03 Mar 2023 09:04:48 +0100 From: Andreas Schwab via tz <tz@iana.org> Message-ID: <87y1oes1jz.fsf@linux-m68k.org>
| On Mr 02 2023, Paul Eggert via tz wrote: | | > If common practice becomes "ET" we couldn't use that, unfortunately, | > as POSIX requires at least three characters. | | When /dst/ is missing, /std/ can be less than 3 bytes.
What gives you that impression?
Where it says
The interpretation of these fields is unspecified if either field is less than three bytes (except for the case when dst is missing),
it is just allowing for a missing dst field being (obviously) 0 bytes, and hence less than 3 - it is the missing dst field that is allowed to be less than 3 bytes in that case, not the std field.
My understanding is that a missing field does not have a length. Though I admit that without explicit delimiters, a missing field is difficult to distinguish from a zero-length field.
But perhaps that could be clearer.
Perhaps it should say: "except that dst can be missing". -- Andreas Schwab, schwab@linux-m68k.org GPG Key fingerprint = 7578 EB47 D4E5 4D69 2510 2552 DF73 E780 A9DA AEC1 "And now for something completely different."