How to

Conversion

Convert files to netcdf

You can use the odpy command within a terminal to access the different functions of the ocean-data-parser.

Convert all cnv files in subdiretories by using the seabird.cnv parser and save to output directory:

odpy convert --input-path input/**/*.cnv --parser=seabird.cnv --output_path=output

See commmand line section for more detail or use the command: odpy convert --help

Avoid reconverting over again the same files

The ocean-data-parser provides a file retristry which can be used to: - track which files were converted and outputted where - error associated with each files - file modified time - file hash

If activated, a registry file (.csv/.parquet) will be saved and any time a conversion is rerun. odpy will first compare the available files to the already parsed files available within the registry and will only convert the ones which have changes. Those changes are primarily based on the file modified time, each modified file is then rehashed and if that hash is different the file will be reconverted and the output overwritten (default)

Parser handling

Load any compatible files in my own project

To load a compatible file you can use the automated parser detection method:

from ocean_data_parser import read

# Load a file to an xarray object
ds = read.file('Path to file')

# Save to netcdf
ds.to_netcdf('save-path.nc')

The parser detection method relies on the file extension and the first few lines present within the given file. It is preferable to define a specific parser when a tool is used in production.

Load a file with a specific parser in my own project

You can import a specific parser via the ocean_data_parser.parser

from ocean_data_parser.parsers import seabird

# Load a seabird cnv file as an xarray dataset
ds = seabird.cnv('Path to seabird cnv file')

# Save to netcdf
ds.to_netcdf('save-path.nc')

The ocean-data-parser can then be used within either a python package, script or jupyter notebook. See documentation Notebook section for examples on how to use the package within a jupyter notebook.