bool evaluations of generators vs lists

[Chris Rebert]

On Tue, Feb 10, 2009 at 11:15 AM, Josh Dukes <josh.dukes at microvu.com> wrote:
> quite simply...what???
>
> In [108]: bool([ x for x in range(10) if False ])
> Out[108]: False

This evaluates the list comprehension and creates an empty list, which
is considered boolean False by Python.

> In [109]: bool( x for x in range(10) if False )
> Out[109]: True

Whereas this creates a /generator object/, whose inner expression is
*not evaluated until specifically required* (e.g. by for-looping over
the generator object). Generators don't have a specially defined
truthiness critera (it's impossible to define generally -- consider
something like `x for x in all_integers if f(x)`, for a complicated
f(x), which would require a solution to the halting problem to know in
advance if it will have a non-zero length), so they end up using the
default behavior for objects with otherwise undefined boolean truth,
which is to consider them True.

Cheers,
Chris

-- 
Follow the path of the Iguana...
http://rebertia.com