Re: [PATCH] Unify human represented timestamp formats.

Mark D. Baushke wrote:
> It is indeed a representation with a different precision. However,
> my recollection is that the +03:00 and +03 are considered
> equivalent. The +03 may be omitted with the :00 assumed if the
> difference is exactly an integral number of hours.

I might be wrong, but I do seem to recall that they are exactly that -
different precision. That the :00 is not assumed - and that the
difference is only precise up to the hour.

> That said, I would favor having the internal stored representation
> in a canonical UTC format. I would further like to see us keep time
> down to at least milliseconds (e.g., 2002-06-23T01:41:13.994Z) or
> smaller fractions of a second if you think that makes sense.

It is. Down to microseconds, with the 'T' and the 'Z' in the end. Over
WebDAV it's required by RFC2518.

-- Naked

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Sun Jun 23 10:02:03 2002