Getting Started¶

Install¶

uv add nc-check
# or
pip install nc-check

Install optional CF checker dependencies:

uv add "nc-check[cf]"
# or
pip install "nc-check[cf]"

First Python Run¶

import xarray as xr
import nc_check  # registers ds.check accessor

ds = xr.Dataset(
    data_vars={"temp": (("time", "lat", "lon"), [[[280.0]]])},
    coords={"time": [0], "lat": [10.0], "lon": [20.0]},
)

report = ds.check.compliance(report_format="python")
print(report["counts"])

First CLI Run¶

# shorthand for: nc-check compliance input.nc
nc-check input.nc

# run all checks
nc-check all input.nc

# apply safe fixes and write a new file
nc-comply input.nc output.nc

Output Formats¶

All check methods support report_format:

auto: notebooks -> HTML, CLI -> rich tables, scripts/tests -> Python dict
python: return a dictionary
tables: print rich tables to stdout
html: return HTML string (and optionally save with report_html_file)

Common Workflow¶

Run ds.check.compliance() (or nc-check compliance) to find metadata issues.
Apply safe fixes with ds.check.make_cf_compliant() (or nc-comply for files).
Run ds.check.all() (or nc-check all) and save an HTML report for review.