For review: PEP 308 - If-then-else expression

[David Eppstein]

In article <yu994r7fhmnh.fsf at europa.research.att.com>,
 Andrew Koenig <ark at research.att.com> wrote:

> Some people would write something like this:
> 
>         strs = []
>         for i in ints:
>             if i < 0:
>                 strs.append("?")
>             else:
>                 strs.append(str(i))
> 
> but others (including me) would like to be able to write this:
> 
>         strs = [("?" if i < 0 else str(i)) for i in ints]

For both of your versions, I'd reverse the order, since str(i) seems to 
be the primary intended result and "?" a fallback for unusual cases:

strs = [(str(i) if i >= 0 else "?") for i in ints]

I haven't yet thought whether I want to vote for GvR's proposal, but 
one advantage of it to me is that it does emphasize the primary result 
(the if clause), and de-emphasize the logic of how you choose that 
result.

So, when reading such code, you can more quickly see what type of object 
you're going to get: as soon as I see "str(i) if..." I can tell that the 
result will be a stringified int, except for some less-common cases 
where it will likely be something else resembling that.  On the other 
hand, for the C syntax (or similar counterproposals in this thread), the 
first thing you see is the i>=0 comparison, which I don't think is the 
most important feature of the expression -- you have to read more of the 
expression to have any idea what kind of value it will produce.

-- 
David Eppstein       UC Irvine Dept. of Information & Computer Science
eppstein at ics.uci.edu http://www.ics.uci.edu/~eppstein/