# The read function below reads SWAN HSR K0 txt files. This data can be found
# at https://data.phys.ucalgary.ca/sort_by_project/SWAN/hsr/l0/multi_freq/. The
# function takes in the path of a K0 txt file (must be downloaded prior) and
# returns a Pandas dataframe.
#
# Changelog:
#   2023-07-13: Filobateer Ghaly
#     - initial creation
#   2023-09-05: Filobateer Ghaly
#     - bugfix for timestamp and indexing

import pandas as pd


def read(filename: str) -> dict:
    """ 
    Reads k0 data files for a given site, date and file version.

    Args:
        filename: the name of the HSR K0 file to read

    Raises:
        FileNotFoundError: in case the file does not exist

    Returns:
        dict: A dictionary containing the metadata and the data
              frame. The metadata includes the band ids, central
              frequencies, passbands, and more.
    """
    # init
    metadata_dict = {}

    # read data
    try:
        fp = open(filename, 'r')
        for line in fp:
            line = line.strip()
            if (len(line) == 0):
                continue
            elif (line[0] == '#' and ':' in line):
                key, val = line[2:].strip().split(": ")
                metadata_dict[key] = val
            else:
                break
        fp.close()
    except FileNotFoundError:
        raise FileNotFoundError(f"The file {filename} does not exist.")

    # create dataframe
    columns = ["Date", "Time"] + metadata_dict["Band Unique IDs"].split(', ')
    df = pd.read_csv(filename, comment='#', delim_whitespace=True, names=columns)

    # convert time column to datetime objects
    df.insert(0, "timestamp", pd.to_datetime(df["Date"] + " " + df["Time"], format="%Y-%m-%d %H:%M:%S"))
    df.drop(["Date", "Time"], axis=1, inplace=True)

    # organize data return object
    ret_data = {
        "data": df,
        "metadata": metadata_dict,
    }

    # return
    return ret_data