search

actris-cloudnet / cloudnetpy

Python package for Cloudnet data processing
MIT License
34 stars 26 forks source link

znc-patch-2? #86

Closed spirrobe closed 1 year ago

spirrobe commented 1 year ago

My previous PR for [spirrobe:spirrobe-zncpatch-1](https://github.com/spirrobe/cloudnetpy/tree/spirrobe-zncpatch-1) is working well now on our server with ZNC files. But when running it I realised that the combination of list-input (required as we sometimes have corrupt netCDFs that I need to filter out beforehand) and concat lib makes unnecessary large output files as the radar spectra are contained in the zncs.

As such, I extended keymaps = {"znc": ["DropSize"], "mmclx": None} to keymaps = {"znc": ["DropSize", "doppler", "SPCco", "SPCcx", "SPCcocxRe", "SPCcocxIm"], "mmclx": None}

Additionally, as we run a measurement campaign sometimes unforeseen things happen, such as the change of nfft over one day. This is not great but at the same time should not affect the cloudnet processing per se.

As such I changed L95 in mira.py from 

            concat_lib.concatenate_files(
                valid_files,
                input_filename,
                variables=variables,
                ignore=_miraignorevar(filetypes[0]),
                allow_difference=["nave", "ovl"],
            )

to

            concat_lib.concatenate_files(
                valid_files,
                input_filename,
                variables=variables,
                ignore=_miraignorevar(filetypes[0]),
                allow_difference=["nave", "ovl", "nfft", ],
            )

I'll gladly open another PR/patch for this slight improvement but wanted to get feedback first for part 2 (different nffts) as I do not see that this affects processing in the code but the attribute in the in mira2nc file will not reflect the true input.

siiptuo commented 1 year ago

Changes will be available in the next version.