Michele Petrazzo wrote:
> Just a question about that "different algorithm", because it force the
> developer to do other work for make the "split" result more "logically
> compatible":
>
> S = "" # this can become from an external source like config parser
>
> for s n S.split():
> do the work... # don't go inside
>
> that isn't "compatible", so python split it into two different methods
> the string, with:
>
> for s n S.split(','):
> do the work... # run one time into this
No question.
>>> def split_any(s, seps):
... for sep in seps:
... parts = s.split(sep)
... if len(parts) > 1:
... return parts
... raise ValueError
...
>>> split_any(" alpha beta ", [",", None])
['alpha', 'beta']
>>> split_any("alpha,beta", [",", None])
['alpha', 'beta']
>>> split_any("alpha_beta", [",", None])
Traceback (most recent call last):
File "<stdin>", line 1, in ?
File "<stdin>", line 6, in split_any
ValueError
The answer :-)
Seriously, I think you have to be a bit more explicit on what you want to
know.
Peter
--
http://mail.python.org/mailman/listinfo/python-list
