* David Megginson -- Wednesday 12 January 2005 17:02: > There's one gotcha -- the file for the current cycle is usually > incomplete (the file grows as reports are collected), so you always > want the previous one as a backup.
I noticed already. That's no problem, though. You just dump as much METAR data into the cache as you like (downloaded cycle files, or particular station data) and the proxy picks up the most appropriate set. If you are out of data, then you get invalid data (and fgfs will stop querying) or everything jumps back one day, or ... Works quite nicely already, btw. The only change to fgfs was that the GET request sends now an "X-Time" header containing the simulated GMT as UNIX epoch. noaa ignores this header, but the proxy considers this information. (The whole proxy is programmed in a few lines of Perl. :-) m. _______________________________________________ Flightgear-devel mailing list Flightgear-devel@flightgear.org http://mail.flightgear.org/mailman/listinfo/flightgear-devel 2f585eeea02e2c79d7b1d8c4963bae2d