Re: [netcdfgroup] Read NETCDF using Java library (now Python)

• To: "netcdfgroup@xxxxxxxxxxxxxxxx" <netcdfgroup@xxxxxxxxxxxxxxxx>
• Subject: Re: [netcdfgroup] Read NETCDF using Java library (now Python)
• From: Jules A <jules0080@xxxxxxxxxxx>
• Date: Mon, 9 Feb 2015 17:09:18 +1100

Thanks for your comments.  I decided to give python a go (newbie), took a while 
to get setup I'm now closer to a solution.
 
I've tried ncdump and nc2text I need the data formatted to the CSV in a 
specific format (input to a program).
 
My main question is after I have written the coordinates how to find the 
correct data from sfc (variable 'MIN_SFC') for the lat/long and date?
 
I've written the header to the file, looking into parsing and saving the dates 
instead of date/time.
Latitude Longitude Date1 Date2 Date3 etc.  
 
I wrote the first set of coordinates Lat/long to cells A2, B2 of the CSV, when 
I tried to create a function to loop through I'm getting a  ValueError: I/O 
operation on closed file.  The file close for the CSV is at the end of the 
script.
 
Thanks!
 
[Code]
from netCDF4 import Dataset;

from netCDF4 import num2date;

filename = "C:/filename.nc"nc = Dataset(filename, 'r', Format='NETCDF4')

print nc.variables


print 'Variable List'
for i in nc.variables:

   print [i, nc.variables[i].units, nc.variables[i].shape]


# get coordinates variables
lats = nc.variables['latitude'][:]

lons = nc.variables['longitude'][:]


try:
  assert(nc.variables['latitude'].units == 'degrees_north')

except:

  raise AttributeError('latitude units attribute not what was expected')


try:
  assert(nc.variables['longitude'].units == 'degrees_east')

except:

  raise AttributeError('longitude units attribute not what was expected')



 sfc= nc.variables['Min_SFC'][:]
times = nc.variables['time'][:]

# convert time
print "Converting Dates"

units = nc.variables['time'].units

dates = num2date (times[:], units=units, calendar='365_day')


#print [dates.strftime('%Y%m%d%H') for date in dates]

header = ['Latitude', 'Longitude']

# append dates to header stringfor d in dates:

   print d

   header.append(d)


# write to file

import csv


with open('Output.csv', 'wb') as csvFile:
outputwriter = csv.writer(csvFile, delimiter=',')

outputwriter.writerow(header) 
outputwriter.writerow(content)
 
def WriteContent(outputwriter, content):

      outputwriter.writerow(content)   return


for l in lons:

   content =[lats[l], lons[l]]

   WriteContent(outputwriter, content)


    # close the output file
csvFile.close()


# close netcdf 
nc.close()


[/Code]
 
From: chris.barker@xxxxxxxx
Date: Tue, 3 Feb 2015 08:34:20 -0800
Subject: Re: [netcdfgroup] Read NETCDF using Java library
To: jules0080@xxxxxxxxxxx
CC: netcdfgroup@xxxxxxxxxxxxxxxx

On Mon, Feb 2, 2015 at 9:03 PM, Jules A <jules0080@xxxxxxxxxxx> wrote:



I'm having troubles accessing data from a NetCDF file.  
My aim is to access the data and write to a CSV file in the following format:
 
Latitude  Longitude Date1  Date2  Date 3
 <-- MIN_SFC values -->
are you committed to using Java? This would be pretty easy with Python and the 
NetcCDF4 python package (http://unidata.github.io/netcdf4-python/) (or IRIS)
Though this should be pretty straightforward with Java, too.
I have been able to iterate through the MIN_SFC Variable,  I was going to save 
to a dictionary or hash table then realised I don't know which date it belongs 
to or Lat/long and vise versa.
 
Dimensions:
time = 7 // the date field how to convert to a date?
latitude = 292
longitude =341
that means there is dimension for time with 7 items -- so there are 7 time 
steps in the file.
There should be a variable called something like "time", with units like: 
"hours since 2015-02-02T01:43:19Z"
you will  want to convert those units to a regular time stamp -- the python 
package has a utility to do this -- probably netcdfJava does too.
 Variables:
float MIN_SFC (time=7, latitude = 292, longitude = 341)

you'll be helped by the "ncdump" utility -- it will dump a netcdf file in a 
plain text form, making it esy to see what's in there. I suggest:
ncdump -h the_file_name.nc
that will dump only the heady, which is probably what you want.
HTH,
-Chris

 I've looked through the online resources, please post a comment or link.  I 
found the ReadNetCDF example useful however it was for version 1.0 and had 
quite a few deprecated methods.
Thanks!
 
<code>
List<Variable> variableList = ncfile.getVariables();

Array data;
 
                for(int j = 0; j < variableList.size(); j++){                   
Variable r = variableList.get(j);

                        System.out.println(r.getFullName());

                        System.out.println(r.getDatasetLocation());

                        System.out.println(r.getNameAndDimensions());

                        System.out.println(r.getDataType());

                    System.out.println(r.getElementSize());                     
System.out.println("-------------") 
                        data = r.read(); int[] shape = data.getShape(); 
                        Index index = data.getIndex();

                          for (int u=0; u<shape[0]; u++) {

                            for (int x=0; x<shape[1]; x++) {

                              double dval = data.getDouble(index.set(u,x));

}}} 
</code>
 
                                          

_______________________________________________

netcdfgroup mailing list

netcdfgroup@xxxxxxxxxxxxxxxx

For list information or to unsubscribe,  visit: 
http://www.unidata.ucar.edu/mailing_lists/ 


-- 

Christopher Barker, Ph.D.
Oceanographer

Emergency Response Division
NOAA/NOS/OR&R            (206) 526-6959   voice
7600 Sand Point Way NE   (206) 526-6329   fax
Seattle, WA  98115       (206) 526-6317   main reception

Chris.Barker@xxxxxxxx
                                          

• Follow-Ups:
  • Re: [netcdfgroup] Read NETCDF using Java library (now Python)
    • From: Chris Barker

• References:
  • [netcdfgroup] Read NETCDF using Java library
    • From: Jules A
  • Re: [netcdfgroup] Read NETCDF using Java library
    • From: Chris Barker