pangram: Implemented new exercise

[behrtam]

Example solution for new proposed exercise exercism/problem-specifications#133
Will be updated later based on possible feedback there.

[Dog]

Under what case is your set going to have a length over 26? (assuming ascii_letters)

Isn't this exercise region specific due to the use of isalpha? I don't know much about region differences, so I'm going to use German characters in my example only because I'm more familiar with the language.

My understanding of isalpha is it uses the constant string.letters to determine how it should return. If we switched to a "German" region, I'm worried ä, ö, ü, ß will also be included in the set because string.letters is region dependent. Since the size of string.letters changed, this means we could potentially substitute the letter z for the letter ä and still pass.

This raises a few questions:

• Should a German pangram include ä, ö, ü, ß?
• Should we consider limiting to ascii_letters?

[behrtam]

Good point! Yes we should limit it to ascii. The string method isalpha() is a bad choice here. 'abcdefghijklmnopqrstuvwΣΩδ' as input would return true while it clearly is missing 'xyz'.
We should use ? in ascii_lowercase instead of isalpha().

>>> from string import ascii_lowercase
>>> pan = lambda s: len(set(char for char in s.lower() if char in ascii_lowercase)) == 26
>>> pan('abcdefghijklmnopqrstuvwxyz')
True
>>> pan('abcdefghijklmnopqrstuvwΣΩδ')
False
>>> pan('abcdefghijklmnopqrstuvwΣΩδxyz')
True
>>>

[behrtam]

We could even get rid of the len/set combination. This is fare more readable.

>>> pan = lambda s: all(char in s.lower() for char in ascii_lowercase)
>>> pan('abcdefghijklmnopqrstuvwΣΩδxyz')
True
>>> pan('abcdefghijklmnopqrstuvwΣΩδ')
False
>>>

[Dog]

I really like the solution with all. I think its ready for merging otherwise.