Read an ODF file in Python using rpy2 to interface with R libraries

odf_to_pandas.py

#!/usr/bin/env python3

"""
Use OCE R package from Python3 to read an ODF file

You will need to install R and libraries 'oce' and 'RJSONIO' by running
the following commands in R. I am using R version 3.6.2

install.packages("devtools")
library(devtools)
install_github("dankelley/oce", ref="develop")
install.packages("RJSONIO", dependencies = TRUE)
library(RJSONIO)

Also, install python packages:
pip install pandas xarray rpy2

"""

import json
import pandas as pd
from rpy2.robjects.packages import importr

oce = importr('oce')
rjsonio = importr('RJSONIO')


def read_odf_py(filename):
    """
    interfaces with the the OCE R library by serializing the oce
    object to/from json
    """

    odf = oce.read_odf(
        filename)

    json_odf = rjsonio.toJSON(odf)
    json_odf_unescaped = str(json_odf).encode(
        'utf-8').decode('unicode_escape').replace('[1] "', '')[0:-2]
    odf = json.loads(json_odf_unescaped)

    return odf


odf_dict = read_odf_py(
    '~/dev/odf/odf/CTD_ExampleData_BBMP/2016/CTD_BCD2016667_011_01_DN.ODF')

# load data into Pandas dataframe
df = pd.DataFrame.from_dict(odf_dict['data'])

# metadata is a nested dictionary
metadata = odf_dict['metadata']

print(metadata)
print(df)