2003-01-09T19:32:09 Mikael Olsson: > I thought it was "z", when in fact it was "%z". Sorry. Your code > is right. Well, it's right where you're fortunate enough to have a strftime that handles %z, as I do, and apparently you do, but as Andy Bach pointed out to me, that's not universal. In fact, I don't know of a simple, portable way to get this info. So my test implementation is falling back and punting, simply using gmtime and a literal "Z", as you first suggested. > Anyhow, this gives me the opportunity to point out that it'd be > real nice if your test client would also send fractional seconds, > since we want that to be (an optional) part of this new protocol. Good idea, just done, used Time::HiRes, I'll repost the client and server pair shortly, using gmtime w/ 'Z', and with the client emitting fractional seconds. -Bennett
This archive was generated by hypermail 2b30 : Fri Jan 10 2003 - 12:01:34 PST