Re: Problem with time specification

[Matt Howard <mkhoward@xxxxxxxx>]

Dear John,

No I guess I still have problems with the time specification.

Ferret 5.53 Las 6.2.1.0 on a SGI FUEL Irix 6.5.22

This is a 100 by 100 gridded data set at 4 different times.
The times are made up.  If I use hours LAS will not plot
anything but the first raster.  If I simply change "hours"
to "days" in the units specification it plots fine.

I note that when hours is used the pull down menu for time
shows - note the difference in representation of hours is
different after the first

01-Jan-2004 01:00:00
01-Jan-2004 2:0:0
01-Jan-2004 3:0:0
01-Jan-2004 4:0:0

Both the .nc and .cdl files are available through my website.

http://seawater.tamu.edu/cgi-bin/nph-nc/noppdodsgom/data/junk.nc.info
http://seawater.tamu.edu/~howie/junk.cdl
-- 

I also note that the UI regeneration has a few complaints

Serializing file:/int2/las/server/las.xml to database
Building indexes...
Resolving links...
Generating dataset COADS Climatology
Generating dataset junk_nc
Generating dataset Levitus Climatology
Use of uninitialized value in addition (+) at ../xml/perl/genLas.pl line 218,
<STDIN> line 3.
Use of uninitialized value in numeric eq (==) at ../xml/perl/genLas.pl line 218,
<STDIN> line 3.
Use of uninitialized value in addition (+) at ../xml/perl/genLas.pl line 229,
<STDIN> line 3.

+----------------------------------------------------------------------+
| Dr. Matthew K. Howard                   Voice: (979)-862-4169        |
| Department of Oceanography                FAX: (979)-847-8879        |
| Texas A&M University              Internet: mkhoward@tamu.edu        |
| College Station, TX 77843-3146      webcam: http://seawater.tamu.edu |
+----------------------------------------------------------------------+