[THREDDS #YDW-329913]: FMRC time problem again

["Unidata THREDDS Support" <support-thredds@xxxxxxxxxxxxxxxx>]

> John,
> 
> OMG!   The 4.1+ timeUnitsChange combo fixed it!   I'm not even going
> to try to figure out which one!  ;-)
> 
> Thanks!!!!

party time most excellent!

> 
> -Rich
> 
> On Tue, Sep 29, 2009 at 7:45 PM, Rich Signell <rsignell@xxxxxxxx> wrote:
> > John,
> >
> > We will soon go from 24 hour forecasts to 48 hour forecasts, so we'll
> > want the FRMC. Â But it would be nice to have a "eliminate duplicate
> > time values" from the joinExisting, for sure!
> >
> > I don't understand the FRMC behavior, however. There should not be any
> > duplicate coordinate values for the 1st time value. Â(it only occurs
> > once, in the first file in the FRMC). Â So why does it get a zero
> > value? Â Is it because the dateFormatMark pulled from the filename
> > does not match the 1st time value? Â Â

its because there are 12 time steps in the first file and 13 in the second (at 
least with the files i was testing). timeUnitsChange = true forces the fmrc to 
read all the coordinates up front, which correctly deals with the problem.

But I will try upgrading to 4.1
> > and adding the timeUnitsChange. ÂI thought that only applied if the
> > time units changed. Â ÂNow why did I think that? Â;-)

yeah, its a kludge!


Ticket Details
===================
Ticket ID: YDW-329913
Department: Support THREDDS
Priority: Critical
Status: Closed