Using str.translate, this is probably the fastest way to do this:
>>> strs = '000,5\r\n'
>>> from string import ascii_letters, punctuation, whitespace
>>> ignore = ascii_letters + punctuation + whitespace
>>> strs.translate(None, ignore)
'0005'
Using regex:
>>> import re
>>> re.sub(r'[^\d]+','',strs) #or re.sub(r'[^0-9]+','',strs)
'0005'
Using str.join and str.isdigit:
>>> "".join([x for x in strs if x.isdigit()])
'0005'
Use int() to get the integer:
>>> int('0005')
5
Timing comparisons:
>>> strs = strs*10**4
>>> %timeit strs.translate(None, ignore)
1000 loops, best of 3: 441 us per loop
>>> %timeit re.sub(r'[^\d]+','',strs)
10 loops, best of 3: 20.3 ms per loop
>>> %timeit re.sub(r'[^0-9]+','',strs)
100 loops, best of 3: 17.1 ms per loop
>>> %timeit "".join([x for x in strs if x.isdigit()])
10 loops, best of 3: 19.2 ms per loop
0xFFFFFF), or floats (-592.45821543e+04)?