Using functional tools

[Andrae Muys]

Patrick W <quitelikely at yahoo.com.au> wrote in message news:<82sn58avea.fsf at acropolis.localdomain>...
> Patrick W <quitelikely at yahoo.com.au> writes:
> 
> [...]
> 
> Or, if the OP's religion forbids 'fors' or 'whiles' under any
> circumstances:
> 
> def do_every_nth(seq, func, n):
>     def transform(item, index):
>         if index % n == n - 1: return func(item)
>         else: return item
>     return map(transform, seq, range(len(seq)))

Alternatively you can use the "obsolete" apply() and simplify the code
substantially.

def identity(x):
    return x

def trans(x):
    return x*10

>>> l = range(5)
>>> f = [identity, trans]  # trans() every second item in list
>>>
>>> list(xmap(apply, xextend(f), zip(l)))
[0, 10, 2, 30, 4]

The zip() is required to convert the list of int's into a list of
tuple's for use by apply.  Note you can use multiple argument
functions by simply adding extra sequence arguments to zip().

xextend => seq * infinity
xmap => lazy map, that terminates upon exhaustion of any of it's
arguments.

I implement these as follows:

def xmap(func, *args):
    iterlist = [iter(arg) for arg in args]
    while 1:
        yield func(*[itr.next() for itr in iterlist])

def xextend(seq):
    while 1: 
        for i in seq:
            yield i

Andrae Muys