Callback scoping

[Dan]

So, I think I understand what python's scoping is doing in the
following situation:

>>x = [ lambda: ind for ind in range(10) ]
x

[<function <lambdaat 0x00BEC070>, <function <lambdaat 0x00BEC7F0>,
<function <lambdaat 0x00BECA70>, <function <lambdaat 0x00C1EBF0>,
<function <lambdaat 0x00C1EE30>, <function <lambdaat 0x00C228F0>,
<function <lambdaat 0x00C228B0>, <function <lambdaat 0x00C28730>,
<function <lambdaat 0x00C286F0>, <function <lambdaat 0x00C287F0>]

>>x[0]()

9

>>x[5]()

9

>>x[9]()

9

>>ind

9

>>ind = 2
x[0]()

2

>>>

But, I'm wondering what is the easiest (and/or most pythonic) way to
get the behavior I want? (If you haven't guessed, I want a list of (no
parameter) functions, each of which returns its index in the list.)

-Dan

[Marc 'BlackJack' Rintsch]

On Thu, 05 Jul 2007 19:14:07 +0000, Dan wrote:

So, I think I understand what python's scoping is doing in the
following situation:

>>>x = [ lambda: ind for ind in range(10) ]

[…]

But, I'm wondering what is the easiest (and/or most pythonic) way to
get the behavior I want? (If you haven't guessed, I want a list of (no
parameter) functions, each of which returns its index in the list.)

Default arguments are evaluated when the function is defined:

In [15]: x = [lambda x=i: x for i in xrange(10)]

In [16]: x[0]()
Out[16]: 0

In [17]: x[5]()
Out[17]: 5

Ciao,
Marc 'BlackJack' Rintsch

[Nick Craig-Wood]

Dan <th********@gmail.comwrote:

So, I think I understand what python's scoping is doing in the
following situation:

>x = [ lambda: ind for ind in range(10) ]

But, I'm wondering what is the easiest (and/or most pythonic) way to
get the behavior I want? (If you haven't guessed, I want a list of (no
parameter) functions, each of which returns its index in the list.)

This is the traditional way :-

>>x = [ lambda ind=ind: ind for ind in range(10) ]
x[0]()

0

>>x[2]()

2

>>x[9]()

9

>>>

--
Nick Craig-Wood <ni**@craig-wood.com-- http://www.craig-wood.com/nick