[Tutor] What does "TypeError: 'int' object is not iterable" mean?

[Richard D. Moores]

It's great to have you chime in, Steven. I do wish you would stop
pulling your punches, however. ;)

On Fri, Oct 22, 2010 at 17:23, Steven D'Aprano <steve at pearwood.info> wrote:
> On Sat, 23 Oct 2010 12:42:50 am Richard D. Moores wrote:
>
>> So I wrote a function:
>>
>> def float2n_decimals(floatt, n):
>>     """
>>     Given a float (floatt), return floatt to n decimal places.
>>
>>     E.g., with n = 2, 81.34567 -> 81.35
>>     """
>>     return ("%%.%sf" % n) % floatt
>>
>> which works fine,
>
>
> float2n_decimals(x, 3)
>
> is better written in place as:
>
> "%.*f" % (3, x)
>
> There's no need for a function for something so simple.

Yes, but I needed one for ("%%.%sf" % n) % floatt .

>> but a question remains: n is an integer. Why the
>> 's' in '%sf'?
>
> Your function might be more clear if you split the formatting into two
> lines:
>
> template = "%%.%sf" % n
> return template % floatt
>
> The first line builds a template string from n. Now it becomes clear
> what %s is for: it's to insert the n as a string into the template. %d
> would express the intention of the function better.
>
OK, I'll do that.
>
> By the way, the function docstring is seriously misleading. It describes
> a completely different function:
>
>>     Given a float (floatt), return floatt to n decimal places.
>>
>>     E.g., with n = 2, 81.34567 -> 81.35
>
> It does nothing of the sort! What you are describing is the built-in
> function round:
>
>>>> print round(81.34567, 2)
> 81.35
>
> What your function does is return a string, not a float.

OK, point well-taken.

> Besides, given
> a float, where does n come from? A global variable?

Yes. See line 5 of <http://tutoree7.pastebin.com/M2wnPzLr>. This
version incorporates all your suggestions.

> There is a convention for giving examples that you should follow.
> Putting the three of these together, your docstring should say
> something like:
>
>    Given a float floatt and an integer n, returns floatt formatted
>    as a string to n decimal places.
>
>    >>> float2n_decimals(2, 81.34567)
>    '81.35'

I've never seen that convention, but I'll try to follow it.

>
> (BTW, I really hate the name "floatt". It makes me think you're
> stuttering.)

I'll use "x" instead. Anything you'd like to say about the rest of the script?

Thanks, Steven.

Dick