Splitting strings into lists and splitting again

Question

I want to split the string

"   510     -9999999  9             99         12             5             [3, 0]        []            [6]       "

(which contains more or less random numbers of whitespaces between the entries) into it's component parts, including the lists within the string. I can get to this

['510', '-9999999', '9', '99', '12', '5', '[3,', '0]', '[]', '[6]']

through using split and replace. However, I then want to reconstitute the lists within the original string so that I can get to

['510', '-9999999', '9', '99', '12', '5', '[3,0]', '[]', '[6]'].

The real problem is that this string is one of many and the lists may contain many, or no components so I have to deal with this is a general way.

I could potentially search for '[' , then search for ']' to close up the list but, as I don't know the length of any of the lists going in, this seems an inefficient way of doing things.

Any help greatly appreciated!

Answer 1

If lists can't be nested then I think it is possible to preprocess string with:

s = "   510     -9999999  9             99         12             5             [3, 0]        []            [6]       "

opened = False
s_new = ""
for i in s:
    if i == "[":
        opened = True
    if i == "]":
        opened = False
    if not opened or (opened and i != " "):
        s_new += i

And then split it into list:

l = s_new.split()

Answer 2

If lists aren't nested, you can try this:

def mysplit (a):
    return re.split(' +', re.sub('\\[(.*?)\\]', lambda m: '[{}]'.format(m.groups()[0].replace(' ', '')), a))

Answer 3

There is always regex, but you can do it on the cheap like this

>>> import shlex
>>> shlex.split(s.replace('[','"[').replace(']',']"'))
['510', '-9999999', '9', '99', '12', '5', '[3, 0]', '[]', '[6]']

The proper solution would be to use pyparsing module, or even better to control the input source to give you something more sensible like json.