Created
May 1, 2018 04:27
-
-
Save aidanheerdegen/63908c72e8747ea60eb336b65ee99f5b to your computer and use it in GitHub Desktop.
Copy a variable from one netcdf file to another robustly, ignoring HDF read errors
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
#!/usr/bin/env python | |
import netCDF4 as nc | |
import sys | |
import numpy as np | |
import itertools as it | |
name_var = sys.argv[1] | |
orig = sys.argv[2] | |
new = sys.argv[3] | |
orig = nc.Dataset(orig,mode='r') | |
var = orig.variables[name_var] | |
csizes = np.array(var.chunking()) | |
print(csizes) | |
new = nc.Dataset(new,mode='r+') | |
new_var = new.createVariable(name_var, var.datatype, var.dimensions) | |
new_var.setncatts({k: var.getncattr(k) for k in var.ncattrs()}) | |
dimlim = np.asarray(var.shape) | |
steps = (dimlim-1)/csizes + 1 | |
for index in np.ndindex(*steps): | |
index *= csizes | |
origslices = [] | |
newslices = [] | |
# Make up slices of size bufferChunk | |
for start, step, end in it.izip(index, csizes, dimlim): | |
# min checks we don't go beyond the limits of the variable | |
origslices.append(slice(start,min(start+step,end),None)) | |
newslices.append(slice(start,min(start+step,end),None)) | |
while True: | |
try: | |
# Copy the data | |
new_var[newslices] = var[origslices] | |
except: | |
print('Error :',origslices) | |
tslice = origslices[0] | |
# default to using previous time slice | |
increment = -1 | |
if tslice.start == 0: | |
increment = +1 | |
origslices[0] = slice(tslice.start+increment, tslice.stop+increment,None) | |
print('New slice :',origslices) | |
else: | |
break | |
new.sync() # flush data to disk |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
There was an issue with a couple of MOM5 output files with a corrupted chunk. In this case they are super high temporal resolution, there is very little missing data, and the data is designed to be used for an animation, so data integrity isn't an issue.
Move the original file to a backup name, e.g.
Use NCO tool
ncks
to create a copy of the data file without the corrupted variableThen call the program to copy the variable