Creating nested dictionary from text file in python3 - python-3.x

Question

I have a text file of thousands of blocks like this. For processing I needed to convert it into dictionary.
Text file Pattern
[conn.abc]
domain = abc.com
id = Mike
token = jkjkhjksdhfkjshdfhsd
[conn.def]
domain = efg.com
id = Tom
token = hkjhjksdhfks
[conn.ghe]
domain = ghe.com
id = Jef
token = hkjhadkjhskhfskdj7979
Another sample data
New York
domain = Basiclink.com
token = eyJ0eXAiOiJKV1QiLCJhbGciOiJSUzI1NiIsIng1dCI6Im5PbzNaRHJPRFhFSzFqS1doWHNsSFJfS1hFZyIsImtpZCI6Im5PbzNaRHJPRFhFSzFqS1doWHNsSFJfS1hFZyJ9.eyJhdWQiOiJodHRwczovL21zLmNvbS9zbm93
method = http
username = abc#comp.com
Toronto
domain = hollywoodlink.com
token = eyJ0eXAiOiJKV1QiLCJhbGciOiJSUzI1NiIsIng1dCI6Im5PbzNaRHJPRFhFSzFqS1doWHNsSFJfS1hFZyIsImtpZCI6Im5PbzNaRHJPRFhFSzFqS1doWHNsSFJfS1hFZyJ9.eyJhdWQiOiJodHRwczovL21zLmNvbS9zbm93Zmxha2UvsfdsdcHJvZGJjcy1lYXN0LXVzLTIiLCJpc3MiOiJodHRwczovL3N0cy53aW5kb3dzLm5ldC9lMjliODE
method = http
username = abc#comp.com
Would like to convert it into following.
d1={conn.abc:{'domain':'abc.com','id': 'Mike',token:'jkjkhjksdhfkjshdfhsd'}
conn.def:{'domain':'efg.com', 'id': 'Tom',token:'hkjhjksdhfks'}
conn.ghe:{'domain':'ghe.com', 'id': 'Jef',token:'hkjhadkjhskhfskdj7979'}}
Thanks

Answer 1

Since the input file can have varying # of lines with data, this code should work.
Assumptions:
Each key (eg: conn.abc) will start with open square bracket and end with open square bracket. example [conn.abc]
Each inner dictionary key value will be separated by =
If the key value can either be [key] or key, then use the below line of code instead of the commented line of code.
elif '=' not in line:
#elif line[0] == '[' and line[-1] == ']':
Code for this is:
with open('abc.txt', 'r') as f:
d1 = {}
for i, line in enumerate(f):
line = line.strip()
if line == '': continue
elif line[0] == '[' and line[-1] == ']':
if i !=0: d1[dkey]= dtemp
dkey = line[1:-1]
dtemp = {}
else:
line_key,line_value = line.split('=')
dtemp[line_key.strip()] = line_value.strip()
d1[dkey]=dtemp
print (d1)
If the input file is:
[conn.abc]
domain = abc.com
id = Mike
token = jkjkhjksdhfkjshdfhsd
[conn.def]
domain = efg.com
id = Tom
dummy = Test
token = hkjhjksdhfks
[conn.ghe]
domain = ghe.com
id = Jef
token = hkjhadkjhskhfskdj7979
The output will be as follows:
{'conn.abc': {'domain': 'abc.com', 'id': 'Mike', 'token': 'jkjkhjksdhfkjshdfhsd'},
'conn.def': {'domain': 'efg.com', 'id': 'Tom', 'dummy': 'Test', 'token': 'hkjhjksdhfks'},
'conn.ghe': {'domain': 'ghe.com', 'id': 'Jef', 'token': 'hkjhadkjhskhfskdj7979'}}
Note here that I added dummy = Test as a key value for conn.def. So your output will have that additional key:value in the output.

Answer 2

You can use standard configparser module:
import configparser
config = configparser.ConfigParser()
config.read("name_of_your_file.txt")
Then you can work with config as standard dictionary:
for name_of_section, section in config.items():
for name_of_value, val in section.items():
print(name_of_section, name_of_value, val)
Prints:
conn.abc domain abc.com
conn.abc id Mike
conn.abc token jkjkhjksdhfkjshdfhsd
conn.def domain efg.com
conn.def id Tom
conn.def token hkjhjksdhfks
conn.ghe domain ghe.com
conn.ghe id Jef
conn.ghe token hkjhadkjhskhfskdj7979
Or:
print(config["conn.abc"]["domain"])
Prints:
abc.com