I would like to create an approach where I can define the structure of a CSV file (obvious extension to excel should follow) where there is a row definition as well has the header. In this approach a simple re-ordering of the definition will move the columns in the output.
My first attempt was to use a namedtuple. Actually handled most of my needs but I can't create an empty row the populate it as needed. I tried to use a recordclass but have much the same problem.
My output file might have > 30 columns so it gets very sloppy to have to create a new instance with a bunch of Nones. I also want to be able to add a column to the structure without having to update the __init__, etc.
My idea pseudo-code (using namedtuples for illustration) would be:
class TableRow(namedtuple(TableRow, "id name password hostip"))
__slots__ = ()
class TableRowHeader:
def __init__(self):
header = TableRow()
header.id = 'ID'
header.name = "Name"
header.password = "Password"
header.hostip = "Host IP"
class OutputTable():
def __init__(self):
self.header = TableRowHeader()
self.rows = list()
def add(self, new_row):
# Example assumes new_row is an instance of TableRow
self.rows.append(new_row)
def to_csv(self, file_name):
with open(file_name, 'w') as csv_file:
# creating a csv writer object
csv_writer = csv.writer(csv_file)
# writing the fields
csv_writer.writerow(self.header)
for row in sorted(self.rows):
csv_writer.writerow(row)
outtable = OutputTable()
row = TableRow()
row.id = 1
row.name = 'Matt'
row.hostip = '10.0.0.1'
row.password = 'obvious'
outtable.add(row)
outtable.to_csv('./example.csv')
I like the pattern but can't figure out a clean way to handle this in Python.
Do you want something like that?
import csv
from collections import namedtuple
TableRowShort = namedtuple('TableRow', "id name password hostip")
TableRowFull = namedtuple('TableRowFull', "id name password hostip description source admin_name")
class TableRowOptional:
def __init__(self, id, name, password=None, hostip=None, description=None, source=None, admin_name=None):
super().__init__()
self.id = id
self.name = name
self.password = password
self.hostip = hostip
self.description = description
self.source = source
self.admin_name = admin_name
class OutputTable():
def __init__(self):
self.headers = []
self.rows = list()
def add(self, row):
if hasattr(row, '_asdict'):
value = row._asdict()
elif hasattr(row, '__dict__'):
value = row.__dict__
elif isinstance(row, dict):
value = row
else:
raise ValueError('Not supported row type: {}'.format(type(row)))
for header in value.keys():
if header not in self.headers:
self.headers.append(header)
self.rows.append(value)
def to_csv(self, file_name):
with open(file_name, 'w') as csv_file:
# creating a csv writer object
csv_writer = csv.writer(csv_file)
# writing the fields
csv_writer.writerow(self.headers)
for row in self.rows:
csv_writer.writerow([row.get(header, None) for header in self.headers])
outtable = OutputTable()
outtable.add(TableRowShort(1, 'Matt', 'obvious', '10.0.0.1'))
outtable.add(TableRowFull(2, 'Maria', 'obvious as usual', '10.1.0.1', 'some description', 'localnet', 'super_admin'))
outtable.add(TableRowOptional(3, 'Maria', hostip='10.1.0.1', description='some description', source='localnet'))
outtable.add({
'id': 1337,
'name': 'hacker',
'hostip': '127.0.0.1',
'extra': "I've hacked you guys lol!",
})
outtable.to_csv('./example.csv')
This solution provides you interface to store some "prepared namedtuples, normal objects (using __dict__ interface) and raw dict objects as rows. It manages CSV headers automatically based on provided rows structures :)
Looks pretty clear & useful to me. What do you think?
Output CSV
# > cat example.csv
id,name,password,hostip,description,source,admin_name,extra
1,Matt,obvious,10.0.0.1,,,,
2,Maria,obvious as usual,10.1.0.1,some description,localnet,super_admin,
3,Maria,,10.1.0.1,some description,localnet,,
1337,hacker,,127.0.0.1,,,,I've hacked you guys lol!