[Tutor] "if clause" in list comprehensions.

[Eduardo Vieira]

On Mon, Oct 19, 2009 at 1:20 PM, Alan Gauld <alan.gauld at btinternet.com> wrote:
>
> "Sander Sweers" <sander.sweers at gmail.com> wrote
>
>>> mylist = ['John', 'Canada', 25, 32, 'right']
>>> a = [item.upper() for item in mylist if type(item) == type('good')]
>>
>> Usually it is recommended to use hasattr() instead of type()
>>   hasattr(s, 'upper')
>
> Nope, they do  completely different things
> I think you might be thinking of isinstance() which can be used instead of
> type(). I see you use hasattr as a means of testing for a method but that is
> still different from testing type - the method names might be the same but
> the functions be completely different in effect!
>
>>> returned this: ['JOHN', 'CANADA', 'RIGHT']
>>> I was expecting this: ['JOHN', 'CANADA', 25, 32, 'RIGHT']
>>> So, actually the "if" acted like a filter.
>
> It is intended to be used as a filter.
>
>>> In order to use a list comprehension I created this function instead.
>>> def upperfy(item)
>>>   try:
>>>       item = item.upper()
>>>   except AttributeError:
>>>       pass
>>>   return item
>
>> I would move return item under the except and remove the pass, other
>> might disagree on this.
>
> I would :-)
> Doing that would result in None being returned for each successful
> conversion. The OPs code is correct (even if unnecessary)
>
>>> a = [upperfy(item) for item in mylist]
>
> a = [item.upper() if type(item) == str else item for item in mylist]
>
> should do it I think.
>
> HTH,
>
>
> --
> Alan Gauld
> Author of the Learn to Program web site
> http://www.alan-g.me.uk/
>
> _______________________________________________
> Tutor maillist  -  Tutor at python.org
> To unsubscribe or change subscription options:
> http://mail.python.org/mailman/listinfo/tutor
>
Thanks for all the guidance. I'm glad I learned I could use "else" in
the list comp. after all. I thought I could, but when I tried it, I
got it wrong and forgot how I did it. Yeah, I had an impression that a
function was unnecessary in this case, but didn't know how to get
around it.
To: Emmanuel: Thanks for your remark in the case of unicode. In my
code that wasn't a problem, but it's something to consider.

Regards,

Eduardo