Skip to main content

Utility and library to interface with CF-Compliant NetCDF flag variables.

Project description

NetCDF Flag Wrapper (ncflag)

So... you want to inspect CF Compliant NetCDF flag variables?

CF Compliant NetCDF Flag variables are integer flags associated with, or having:

  • flag_values
  • flag_meanings
  • flag_masks (optionally)

Read the CF Conventions on flags for more information.


Install the utility with with pip:

pip install ncflag

On the command line, use ncflag:


  -v, --version                   Show the version and exit.
  --show_flags PATH               Print the flags this tool can inspect.
  --use_time_var TEXT
                                  log level
  --help                          Show this message and exit.


Use --show_flags to discover what flags in a given file can be inspected.

Limitation: can only inspect flags of at most one dimension. See details below for dealing with multidimensional flags.

The nominal output with --use_time_var specified is shown below. Without use_time_var, the index along the dimension will be printed instead of a iso 8601 timestamp.

2017-11-27T21:07:41.543778: [u'data_quality_error']
2017-11-27T21:07:42.543812: [u'good_data']
2017-11-27T21:07:43.543807: [u'good_data']
2017-11-27T21:07:44.543802: [u'good_data']

Multidimensional Flags

Occasionally, by some poor misfortune, you may encounter multidimensional flag variables. These are currently not supported by the Command Line Interface (CLI), however, the FlagWrap class can still be used in code, or through an interactive (IPython) session. The FlagWrap.get_flags_set_at_index can be passed a tuple of indicies to get the flags set in a multidimensional flag variable. Below is an example.

from ncflag import FlagWrap
import netCDF4 as nc

with nc.Dataset("") as nc_in:
    v = nc_in.variables["mutidim_variable"]
    print(v.shape)  # --> (2, 10), is multidim.
    w = FlagWrap.init_from_netcdf(v)
    print(w.get_flags_set_at_index((0, 0)))  # --> ["good_quality_qf"]

API and Documentation

To use the FlagWrap in your own code, see the example above for multidimensional flags.

For documentation, please read It is one file and is documented with comprehensive docstrings. The functions are named descriptively. A following functions are available from a FlagWrap instance.

- get_flag(self, flag_meaning)
- reduce(self, exclude_mask, axis=-1)
- get_flag_at_index(self, flag_meaning, i)
- get_flags_set_at_index(self, i, exit_on_good=False)
- find_flag(self, options)
- set_flag(self, flag_meaning, should_be_set, zero_if_unset=True)
- set_flag_at_index(self, flag_meaning, i)
- get_value_for_meaning(self, flag_meaning)
- get_mask_for_meaning(self, flag_meaning)


There are tests, using both synthetic flags, as well as some more serious tests for some fairly complex "in the wild" flags taken from a sample GOES-16 EXIS-L1b-SFXR product file.

test/ is actually a very thourough read to help anyone really understand what's possible and what's going on with these flags.

Deploy to pip, after testing with python2 and python3:

rm -r dist/
python bdist_wheel --universal
twine upload dist/*

Project details

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distributions

No source distribution files available for this release.See tutorial on generating distribution archives.

Built Distribution

ncflag-0.3.2-py2.py3-none-any.whl (10.1 kB view hashes)

Uploaded py2 py3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page