Storing file input into dictionary

Question

I have written small program in Python which reads the following file and stores the result into a dictionary. I am getting expected output but I think it can be done in a better way. I am using Python 3.

Input

check_name : xa25
not run:
del_l6w_
dl_l4w_
dl_l22w_
de_l3w_
ckt_pw_
ckt_pw_
run:
inv_w_
buf_w_
End

Code

import collections
def main():
 chk_cell = "input.txt"
 chk_cell_data = collections.defaultdict(dict)
 with open(chk_cell, "r") as fchk_cell:
 check_name = ""
 not_run_flag = False
 run_flag = False
 for data in fchk_cell.readlines():
 data = data.strip()
 if not data:
 continue
 if data.startswith('check_name'):
 data = data.split(":")
 check_name = data[1].strip()
 continue
 elif data.startswith('End') or '-' in data:
 check_name = ""
 not_run_flag = False
 run_flag = False
 continue
 elif data.startswith('not run'):
 chk_cell_data[check_name]['not_run'] = []
 not_run_flag = True
 continue
 elif data.startswith('run'):
 chk_cell_data[check_name]['run'] = []
 not_run_flag = False
 run_flag = True
 continue
 
 if not_run_flag:
 chk_cell_data[check_name]['not_run'].append(data)
 
 elif run_flag:
 chk_cell_data[check_name]['run'].append(data)
 print(chk_cell_data)
 
 
if __name__ == "__main__":
 main()

• Reinderien: Can you show an example of the output?

Answer 1

Type strength

Unless you have a really (really) good reason to use a dict - such as planning to immediately serialize to JSON for some network operation - dict is a poor choice for an internal representation. Use a @dataclass.

Parsing strictness

You should be a little more strict about your parsing; your current implementation considers

check_name_for_something_that_makes_no_sense

to be a check_name heading, and a data entry under the run: heading called

check_name

to be a heading instead of a data entry. Instead, just compare the whole line.

Your example data do not show any reason for your - check to be there, so I see no reason for it to exist.

Parse state

Instead of hanging onto flags to remember parse state, you can simply assign a reference to the list currently being populated.

Explicit iteration

No need to call readlines. Just iterate over the file object itself.

Example implementation

from dataclasses import dataclass
from typing import TextIO, List
@dataclass
class RunData:
 name: str
 run: List[str]
 not_run: List[str]
 @classmethod
 def from_file(cls, f: TextIO) -> 'RunData':
 run = []
 not_run = []
 current_list = None
 name = None
 for line in f:
 line = line.rstrip()
 if line == 'run:':
 current_list = run
 elif line == 'not run:':
 current_list = not_run
 elif line.startswith('check_name :'):
 name = line.split(': ', 1)[1]
 elif line == 'End':
 return cls(name, run, not_run)
 else:
 current_list.append(line)
def main():
 chk_cell = "input.txt"
 with open(chk_cell, "r") as fchk_cell:
 chk_cell_data = RunData.from_file(fchk_cell)
 print(chk_cell_data)
if __name__ == "__main__":
 main()

prints

RunData(name='xa25', run=['inv_w_', 'buf_w_'], not_run=['del_l6w_', 'dl_l4w_', 'dl_l22w_', 'de_l3w_', 'ckt_pw_', 'ckt_pw_'])