I'm trying to write a regex for the following situation. I have a file with hundreds of dictionaries as string.
EG:
{'a':1'}
{{'a':1, 'b':2}{'c':3}}
{'a':4, 'b':6}
I read the file and removed the newlines. Now I'm trying split them based on a regex.
{'a':1'}{{'a':1, 'b':2}{'c':3}}{'a':4, 'b':6}
re.split("({.*?})", str). This wouldn't work because the whole second dict wouldn't match. How can I write a regex that would match all the lines return a list of dictionaries.
You could simply do:
(\{[^{}]+\})
# look for an opening {
# and anything that is not { or }
# as well as an ending }
In Python this would be:
import re
rx = r'(\{[^{}]+\})'
string = "{'a':1'}{{'a':1, 'b':2}{'c':3}}{'a':4, 'b':6}"
matches = re.findall(rx, string)
print matches
# ["{'a':1'}", "{'a':1, 'b':2}", "{'c':3}", "{'a':4, 'b':6}"]
Python regular expressions are not able to handle nested structures by themselves. you would have to do some looping or recursion separately.
However, you commented above that each line is a json response. Why not use json.loads() on each line.
import json
with open('path_to_file', 'r') as f:
data = [json.loads(line) for line in f]
data is now a list of dictionaries.
1intentional? Thanks.{{'a':1, 'b':2}{'c':3}}is not valid Python syntax. If it's a single dictionary with nested dictionaries then it's missing keys and a comma, and if you treat it as two separate dictionaries then you have extra braces.json?