I'll reply inline to keep your questions together with the answers...
On Mon, Mar 17, 2014 at 1:21 PM, John Caron <caron@xxxxxxxxxxxxxxxx> wrote:
> Hi Phil:
>
> * 1. does the "+" cause trouble or is it just aesthetics? We wont hand
> correct NCEP/WMO tables, but if it a serious problem we will autocorrect,
> eg replace "+" with "and" anywhere we see it.*
It seems to be the plus causing the problem. We first noticed it in our
NARR catalog with 4.2.x and could fix it with NcML by renaming the
variable. So far we have only seen the issue with this one variable in
NARR that has the "+" symbol and it getting turned into %2b (I guess for
URL encoding?). Here is a snippet of the OPeNDAP form code with the global
attributes containing the long_name for "Volumetric soil moisture frozen +
liquid".
<h3><a href="opendap_form_help.html#global_attr">*Global Attributes:*
</a></h3>
<td><textarea name="global_attr" rows=5 cols=70>
long_name: *Volumetric soil moisture frozen + liquid @ Layer between 2
depths below land surface layer*
units:
missing_value: NaN
grid_mapping: LambertConformal_Projection
Grib_Variable_Id: VAR_7-15-131-144_L112_layer
Grib1_Center: 7
Grib1_Subcenter: 15
Grib1_TableVersion: 131
Grib1_Parameter: 144
Grib1_Parameter_Name: SOILW
Grib1_Level_Type: 112
Grib1_Level_Desc: Layer between 2 depths below land surface
Originating_or_generating_Center: US National Weather Service, National
Centres for Environmental Prediction (NCEP)
Originating_or_generating_Subcenter: North American Regional Reanalysis
Project
Generating_process_or_model: North American Regional Reanalysis (NARR)
Conventions: CF-1.6
history: Read using CDM IOSP Grib1Collection
featureType: GRID
file_format: GRIB-1
_CoordSysBuilder: ucar.nc2.dataset.conv.CF1Convention
</textarea><p>
Later in the form there is a listing for the variable with the name
transformed to %2b....
Volumetric_soil_moisture_frozen_%2b_liquid_layer_between_two_depths_below_surface_layer:
Grid
>
> * 2. assuming "Volumetric_soil_moisture_frozen_+_liquid" really exists
> in the datase, i think*
>
>
>
>
>
> * <variable name="Volumetric_soil_moisture_frozen_and_liquid"
> orgName="Volumetric_soil_moisture_frozen_+_liquid" /> *
> * should have worked. Check you are looking at the variable name, not the
> description.*
>
>
This did work in 4.2.x, but seems to have some issues in 4.3.21, but I've
not tried it in older 4.3.x If I take out the NcML to rename the variable
this is what shows up at the DDS...
*TDS 4.3 w/o NcML*
Grid {
ARRAY:
Float32
Volumetric_soil_moisture_frozen_*%2b*_liquid_layer_between_two_depths_below_surface_layer[time
= 1][layer_between_two_depths_below_surface_layer = 4][y = 277][x =
349];
MAPS:
Int32 time[time = 1];
Float32
layer_between_two_depths_below_surface_layer[layer_between_two_depths_below_surface_layer
= 4];
Float32 y[y = 277];
Float32 x[x = 349];
}
Volumetric_soil_moisture_frozen_%2b_liquid_layer_between_two_depths_below_surface_layer;
*TDS 4.3 with NcML*
Grid {
ARRAY:
Float32
Volumetric_soil_moisture_frozen_%2b_liquid_layer_between_two_depths_below_surface_layer[time
= 1][layer_between_two_depths_below_surface_layer = 4][y = 277][x =
349];
MAPS:
Int32 time[time = 1];
Float32
layer_between_two_depths_below_surface_layer[layer_between_two_depths_below_surface_layer
= 4];
Float32 y[y = 277];
Float32 x[x = 349];
}
Volumetric_soil_moisture_frozen_%2b_liquid_layer_between_two_depths_below_surface_layer;
Grid {
ARRAY:
Float32 Cloud_water_isobaric[time = 1][isobaric = 29][y = 277][x = 349];
MAPS:
Int32 time[time = 1];
Float32 isobaric[isobaric = 29];
Float32 y[y = 277];
Float32 x[x = 349];
}
> 3. Are you sure its in the "Global Attributes", not the "Variable
> Attributes" ??
>
>
I'm positive that it's the Global Attributes that are showing the variable
data in them. In the HTML source above there is the Global Attributes text
box. Please let me know if you need any more information or if I can send
you something else.
Thanks for looking into this for us! :-D
--Phil
> John
>
>
> On 3/17/2014 11:08 AM, Phil Cogbill (NOAA Affiliate) wrote:
>
>> I am working TDS 4.3.21 and NcML to to fix an issue with "Volumetric
>> soil moisture frozen + liquid" showing up in the "Global Attributes" in
>> the OPeNDAP Dataset Access Form.
>>
>> Inline image 1
>>
>> Let me start from the beginning. With TDS 4.2.9 we saw this same
>> behaviour and renamed the variable this way...
>>
>> <variable name="Volumetric_soil_moisture_frozen_and_liquid"
>> orgName="Volumetric_soil_moisture_frozen_+_liquid" />
>>
>> With 4.2.9 this fixes the issue with the garbled metadata in the Global
>> Attributes. When this is added to the 4.3.21 catalog this happens...
>>
>> Error {
>> code = 500;
>> message = "New variable (Volumetric_soil_moisture_frozen_and_liquid)
>> must have datatype attribute";
>> };
>>
>>
>> After finding this I dug around in the code to add a type and added
>> this...
>>
>> <variable name="Volumetric_soil_moisture_frozen_and_liquid"
>> orgName="Volumetric_soil_moisture_frozen_+_liquid"
>> type="float" />
>>
>> Then I get two variables, the original and the new one with the _and_ in
>> the name, plus the Global Attributes metadata is still not correct. I have
>> even tried to remove the variable in multiple ways and it still shows up in
>> the metadata as it is in the image above...
>>
>> <remove name="Volumetric_soil_moisture_frozen_+_liquid"
>> type="variable" />
>>
>> <remove name="Volumetric_soil_moisture_frozen_%2b_liquid"
>> type="variable" />
>>
>>
>>
>> Any help to fix this would be super! :-D
>>
>>
>>
>>
>> Thanks!
>>
>> Phil Cogbill (Government Contractor)
>>
>>
>> Senior Architect, STG, Inc.
>> NOAA's National Climatic Data Center
>> Climate Services and Monitoring Division
>> Data Access and Applications Branch
>>
>> National Climatic Data Center
>> Veach-Baley Federal Building
>> 151 Patton Ave.
>> Room 355
>> Asheville, NC 28801-5001
>> Phil.Cogbill@xxxxxxxx <mailto:Phil.Cogbill@xxxxxxxx>
>>
>> Desk: (828) 271-4474
>> Cell: (828) 419-0303
>>
>> This electronic transmission contains information that may be internal
>> use only,confidential, or proprietary. If you are not the intended
>> recipient, be aware that any disclosure, copying, distribution or use
>> of the contents hereof is strictly prohibited. If you have received
>> this transmission in error, please notify Phil.Cogbill@xxxxxxxx
>> <mailto:Phil.Cogbill@xxxxxxxx>
>>
>>
>> _______________________________________________
>> thredds mailing list
>> thredds@xxxxxxxxxxxxxxxx
>> For list information or to unsubscribe, visit:
>> http://www.unidata.ucar.edu/mailing_lists/
>>
>>
> _______________________________________________
> thredds mailing list
> thredds@xxxxxxxxxxxxxxxx
> For list information or to unsubscribe, visit:
> http://www.unidata.ucar.edu/mailing_lists/
