Source code for pycif.plugins.obsparsers.obspack.parse_txt

import pandas as pd




def parse_txt(obs_file):

    # Get header information
    with open(obs_file, "r") as f:
        nheader = int(f.readline().split(":")[1])
        header = []
        for k in range(nheader - 1):
            header.append(f.readline())

    # Fetch information in the header
    for ln in header:
        if "site_code" in ln:
            site_code = ln.split(":")[1].strip()
        elif "site_elevation_unit" in ln:
            site_elevation_unit = ln.split(":")[1].strip()
        elif "dataset_parameter" in ln:
            dataset_parameter = ln.split(":")[1].strip()

    # Read data
    col2use = [0, 1, 2, 3, 4, 5, 10, 11, 12, 13, 14, 18, 20]
    colname = [
        "year", "month", "day", "hour", "minute", "second",
        "duration", "obs", "lat", "lon", "alt",
        "flag", "obspack_id"
    ]
    data = pd.read_csv(
        obs_file, skiprows=nheader, header=None,
        sep=r"\s+",
        usecols=col2use, names=colname,
        index_col="obspack_id",
        parse_dates={
            "date": ["year", "month", "day", "hour", "minute", "second"]
        }
    )
    data.loc[:, "date"] = pd.to_datetime(
        data["date"], format="%Y %m %d %H %M %S"
    )

    # Replace fixed parameters
    data.loc[:, "station"] = site_code
    data.loc[:, "parameter"] = dataset_parameter.upper()
    data.loc[:, "network"] = "ObsPack"
    data.loc[:, "obserror"] = 0.

    return data