Question

Suppose I have a text file with numbers separated by colons and spaces as the following.

0:-83 1: -51 2: -69 3: -82 4: -85 8: -90 9: -69 QUAD
0:-88 1: -88 2: -98 3: -75 4: -42 5: -71 6: -89 7: -28 8: -83 9: -78 STADIUM

A pair is defined as two numbers separated by a colon. Spaces interrupt pairs of numbers arbitrarily.

Currently, I have the following.

with open('data.txt') as file:
  lines = file.read().splitlines()
for line in lines: 
  line = line[:-1]  
  # What is the regex I should be using? 
  # data = re.split(r'[:\s]',line) includes the space after the colon if it exists

What is the best way to parse the text file so that each line is stored as a list of tuples where each tuple is a pair?

Answer 1

The following regex will give you your pairs of numbers (including minus signs) in groups:

r'\b(-?\d+)\s*:\s*(-?\d+)\b'

This matches a word-boundary (\b), then a set of digits (with an optional - minus sign before it), followed by a colon surrounded by optional whitespace, followed by another set of digits with optional minus sign, followed by a word boundary.

Demo:

>>> import re
>>> numpairs = re.compile(r'\b(-?\d+)\s*:\s*(-?\d+)\b')
>>> example = '0:-83 1: -51 2: -69 3: -82 4: -85 8: -90 9: -69 QUAD'
>>> numpairs.findall(example)
[('0', '-83'), ('1', '-51'), ('2', '-69'), ('3', '-82'), ('4', '-85'), ('8', '-90'), ('9', '-69')]

You probably don’t want to reach all lines into memory at once; just loop over your file line by line:

import re
numpairs = re.compile(r'\b(-?\d+)\s*:\s*(-?\d+)\b')
with open('data.txt') as file:
    for line in file:
        for pair in numpairs.findall(line):
            print pair

You don’t need to remove the newline for the above code to work, but if you do, use line.strip() instead of line[:-1].