Question

I have a string like this:

str = "something move 11 something move 12 something 13 copy 14 15"

where the “something” means some text, or no text at all.
and as a result I want to have a list like:

[('move', 11, ''), ('move', 12, 13), ('copy', 14, 15)]

I tried using this:

re.findall('(move|copy).+?([0-9]+).+?([0-9]+)*', str)

but it gives my the output:

[('move', 11, ''), ('move', 12, ''), ('copy', 14, '')]

I understand that is because the last number is optional, but I just have no idea how I could it get working.

How can I do this?

Answer 1

Based on @Ashwini Chaudhary’s answer:

#!/usr/bin/env python
import re

commands = "copy move".split()
input_string  = "something move 11 something move 12 something 13 copy 14 15"
tokens = iter(re.split("(%s)" % "|".join(map(re.escape, commands)), input_string))
result = []
for tok in tokens:
     if tok in commands:
        args = re.findall(r"\d+", next(tokens, ""))
        result.append((tok,) + tuple(args) + ("",)*(2 - len(args)))
print(result)

Output

[('move', '11', ''), ('move', '12', '13'), ('copy', '14', '15')]

To limit each command to two arguments, just use slicing: tuple(arg[:2]).