Re: Bug in isCoord? and prob with addXml.pl

[Petra Udelhofen <petra@xxxxxxxxxxxxxxxxxxxxxx>]

Hi Joe,
could you drop me line on how you fixed the second bug in addXml.pl? I
have a similar problem:
Dimension lon doesn't have a coordinate variable at ./addXml.pl line 76 

Thank you,
Petra
p.s. Here is a sample ncdump from a netcdf file:
netcdf mm_ucla {
dimensions:
        lon = 72 ;
        lat = 44 ;
        level = 10 ;
        time = 12 ;
variables:
        float T(time, level, lat, lon) ;
                T:missing_value = 9.9e+30f ;
                T:long_name = "temperature" ;
                T:units = "K" ;
        float Z(time, level, lat, lon) ;
                Z:missing_value = 9.9e+30f ;
                Z:long_name = "geopotential height" ;
                Z:units = "m" ;
        float lon(lon) ;
                lon:long_name = "longitude" ;
                lon:units = "degrees_east" ;
        float lat(lat) ;
                lat:long_name = "latitude" ;
                lat:units = "degrees_north" ;
        float level(level) ;
                level:long_name = "pressure" ;
                level:units = "millibars" ;
        int time(time) ;
                time:long_name = "time" ;
                time:units = "months since 1900-01-15" ;

// global attributes:
                :title = "UCLA Model data" ;
                :Conventions = "COARDS" ;
                :history = "Converted from GrADS mm_ucla.ctl on Tue Feb 29
20:27:03 2000 GMT -05:00 with conv_ctl_netcdf (SUNY-SB/ITPA/SPARC-DC/PMU)"
;
data:

 lon = -180, -175, -170, -165, -160, -155, -150, -145, -140, -135, -130, 
    -125, -120, -115, -110, -105, -100, -95, -90, -85, -80, -75, -70, -65, 
    -60, -55, -50, -45, -40, -35, -30, -25, -20, -15, -10, -5, 0, 5, 10,
15, 
    20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100,
105, 
    110, 115, 120, 125, 130, 135, 140, 145, 150, 155, 160, 165, 170, 175 ;

 lat = -86, -82, -78, -74, -70, -66, -62, -58, -54, -50, -46, -42, -38,
-34, 
    -30, -26, -22, -18, -14, -10, -6, -2, 2, 6, 10, 14, 18, 22, 26, 30,
34, 
    38, 42, 46, 50, 54, 58, 62, 66, 70, 74, 78, 82, 86 ;

 level = 850, 700, 500, 300, 200, 100, 70, 10, 5, 1 ;

 time = 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11 ;
}
p.p.s. I will drop further info on our LAS site to the list once I have
data online, Thanks.
------------------- 
>Hi Andrew, 
>
>Thanks for the bug fix. The second bug had already been fixed; both bug
>fixes will be in the next release of LAS. 
>
>I created a XML file from your DODS URL using the latest version of
>addXml, plugged into LAS and it works -- kind of. Ferret runs out of
>memory if you
>attempt to visualize the full geographical range; this is easily fixed by
>using a LAS <init_script> (see the documentation). I've attached the XML
>file for you
>to peruse. 
>
>Cheers. 
  

>Andrew Woolf wrote: 
>
>   Two things: 
>
>   (1) netCDF allows scalar variables (having ndims==0). These, of
>course, 
>   are not indexed by any coordinate variable. In LASNetCDF.pm, however, 
>   the function isCoord() tries to get the name of the first dimension,
>unless 
>   the variable has more than one dimension. This seems to be a bug. The
>fix 
>   is to replace: 
>    return 0 if (scalar @dims > 1); 
>   by: 
>    return 0 if (scalar @dims != 1); >

>   (2) I have what I believe to be a COARDS-compliant netCDF but
>addXml.pl 
>   has a problem with it. During the parsing, it reports: 
>   Dimension LONGITUDE_U doesn't have a coordinate variable at
>LASNetCDF.pm line 
>   669 

....