Re: [ferret_users] 30-FEB with 360_day calendar

[Ansley Manke <ansley.b.manke@xxxxxxxx>]

Hi Patrick,
This was a bug, which has been fixed in the Ferret v6.2 release.  It's 
listed in the release notes.
http://ferret.pmel.noaa.gov/Ferret/documentation/version-6-2-release-notes#d360

I think if you could work around this, if you don't want to upgrade to 
v6.2 at this time, with something like this:

 yes? def 
axis/t="01-JAN-1860:00:00:00":"1-MAR-1890:00:00:00":1/edges/units="Months"/cal=360_day 
simutaxis2

Since the units are months, this definition results in monthly time 
steps ending with 16-feb-1890, just as the definition you used would 
have done.  In fact specifying the hours, minutes, seconds as I did 
above is not necessary in this case.

Ansley


Brockmann Patrick wrote:
> Sorry. My question was confused and had kept traces of tries I have 
> made to identify the bug.
>
> To resume and of course with a correct syntax for february month (ie 
> FEB) and specification
> for calendar (ie /cal=360_day), I get an error with ferret v6.08.
> Unfortunately this is the ferret release used in our production mode.
>
> This is now ok with the last release aka ferret v6.20.
> But I confirmed that it is a bug in release v6.08
>
> ######################################
> FERRET v6.08
> yes? def 
> axis/t="01-JAN-1860":"30-FEB-1890":1/edges/units="Months"/cal=360_day 
> simutaxis2
> **ERROR: command syntax: 30-FEB-1890
>
>
> ######################################
> FERRET v6.2
> yes? def 
> axis/t="01-JAN-1860":"30-FEB-1890":1/edges/units="Months"/cal=360_day 
> simutaxis2
> *** NOTE: /UNIT=MONTHS  ... using 1/12 of 360.00 days
>
> Have a nice week-end.
> Patrick