Use Python to parse records with header delimiters, key/value pairs, and sub records, as a list of dictionaries.

parsetext.py

t = """


---- Header -----
value_a: a
value_b: a
value_c: a
value_d: a

-- Record 1 --
value_a: a
value_b: a
value_c: a
value_d: a
value_d:
  a_a: 1.0
  a_b: 2.0
  a_c: 3.0

-- Record 2 --
value_a: a
value_b: a
value_c: a
value_d: a
value_d:
  a_a: 1.0
  a_b: 2.0
  a_c: 3.0

-- Record 3 --
value_a: a
value_b: a
value_c: a
value_d: a
value_d:
  a_a: 1.0
  a_b: 2.0
  a_c: 3.0

"""


def parse_records(text, record_delimiter="--", key_delimiter=":", sub="  "):

    # convert text to lines, removing empty lines
    lines = (line for line in text.splitlines() if line)

    records = []
    record = {}
    subrecord = {}
    for line in lines:
        if line.startswith(record_delimiter):
            record = {}
            records.append(record)
        else:
            t = line.split(key_delimiter)
            k = t[0].strip()
            v = t[1].strip()
            if line.startswith(sub):
                subrecord[k] = v
            else:
                if not v:
                    # subrecord
                    subrecord = record[k] = {}
                else:
                    record[k] = v

    print(f"records: {len(records)}")
    for r in records:
        print(r)


if __name__ == "__main__":
    parse_records(t)

@jordan-hamilton whenever I do something like this in Python I feel like there is a magic one-liner out there somewhere that can do the job in a more Pythonic way.

Maybe some customization of configparser? I haven’t looked too closely, but I’m doubtful that there’s a better way to handle the sub-records.