Python inf & nan complexity

[retina]

Write a single line input for Python (2.7 and up) that can output all nine possibilities of ±inf and nan as a complex number.

That is, the output should look like this:

Python 2.7.12 (default, Nov 12 2018, 14:36:49) 
[GCC 5.4.0 20160609] on linux2
Type "help", "copyright", "credits" or "license" for more information.
>>> <your line goes here>
(inf+infj)
(inf-infj)
(inf+nanj)
(-inf+infj)
(-inf-infj)
(-inf+nanj)
(nan+infj)
(nan-infj)
(nan+nanj)
>>>

This puzzle would be too easy if things like functions or loops were permitted, so they aren't allowed. That means no print or for etc.

For those that aren't familiar, complex numbers in python can be entered like this: 3+4j and Python displays them in the same format surrounded by parentheses. But it isn't possible to simply enter inf+infj or nan+nanj those will show an error.

And in case you might be wondering, you can define both inf and nan like this: float('inf') and float('nan'). Hope that helps.

>>> 3+4j
(3+4j)
>>> inf+infj
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
NameError: name 'inf' is not defined
>>> nan+nanj
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
NameError: name 'nan' is not defined
>>> float('inf')
inf
>>> float('-inf'); float('nan')
-inf
nan
>>>

Trying to use strings directly won't help because of the extra '.

>>> '(inf+infj)'
'(inf+infj)'

[uau]

You say that print() is not permitted. Then what is allowed? If the only allowed thing is to return some object and have Python display it in the default manner, that does not seem to enable the desired kind of output. It looks like 9 distinct objects on separate lines (not a list or tuple of objects), and you can't have 9 independent return values from a single line...

[Gimarel]

I don't know much about python. Just from reading your post, is this the intended solution:
complex(float('inf'),float('inf'));complex(float('inf'),float('-inf'));complex(float('inf'),float('nan'));complex(float('-inf'),float('inf'));complex(float('-inf'),float('-inf'));complex(float('-inf'),float('nan'));complex(float('nan'),float('inf'));complex(float('nan'),float('-inf'));complex(float('nan'),float('nan'))

[nordi]

Originally Posted by retina

This puzzle would be too easy if things like functions or loops were permitted, so they aren't allowed. That means no print or for etc.

In Python 2.7, "print" is actually not a function, it is a statement, i.e. a part of the language's grammar. Python 3 changed "print()" to be a function that just happens to be defined by default.

This works with Python 2 and 3:

import itertools as i;x=-1e999,+1e999,float("nan");tuple(i.starmap(complex,i.product(x,x)))

None of these is a function, btw:
>>> type(i.starmap)
<type 'type'>
>>> type(i.product)
<type 'type'>

[retina]

I may have mis-characterised what a function is, but for this purpose I define it as:

A function is something that needs parentheses ( and ). So I stated print explicitly, but this also includes complex and product etc.

Originally Posted by Gimarel

I don't know much about python. Just from reading your post, is this the intended solution:

complex is a function (or statement thingy).

Originally Posted by nordi

In Python 2.7, "print" is actually not a function, it is a statement, i.e. a part of the language's grammar. Python 3 changed "print()" to be a function that just happens to be defined by default.

This works with Python 2 and 3:

starmap and some other things you used are functions.

A solution should work in Python 2.7 and up so even if print is a statement in 2.7, it would be a function in 3.0+ so print is not going to work. That would be just too easy anyway. Easy puzzles are not fun.

[retina]

Originally Posted by uau

You say that print() is not permitted. Then what is allowed?

Anything without ( and ), and not a loop construct.

Originally Posted by uau

If the only allowed thing is to return some object and have Python display it in the default manner, that does not seem to enable the desired kind of output. It looks like 9 distinct objects on separate lines (not a list or tuple of objects), and you can't have 9 independent return values from a single line...

That is not correct.

[nordi]

Originally Posted by retina

A function is something that needs parentheses ( and ).

In your initial post, you suggested that people use float('nan') to get a NaN value, but now you say that's forbidden. Please make up your mind.

[retina]

Originally Posted by nordi

In your initial post, you suggested that people use float('nan') to get a NaN value, but now you say that's forbidden. Please make up your mind.

float is not allowed because of the parenthesis. I said "Hope that helps" which means it won't help.

[Gimarel]

Ok, your definition is still not good enough because you can use print without parenthesis, at least in python2.
But I think you want something like this:
1e400+1e400j;1e400-1e400j;1e400+1e400j-1e400j;-1e400+1e400j;-1e400-1e400j;-1e400+1e400j-1e400j;1e400-1e400+1e400j;1e400-1e400-1e400j;1e400-1e400+1e400j-1e400j

[retina]

Originally Posted by Gimarel

Ok, your definition is still not good enough because you can use print without parenthesis, at least in python2.

But the code must work for all 2.7 and up, so I think that covers it.

Originally Posted by Gimarel

But I think you want something like this:
1e400+1e400j;1e400-1e400j;1e400+1e400j-1e400j;-1e400+1e400j;-1e400-1e400j;-1e400+1e400j-1e400j;1e400-1e400+1e400j;1e400-1e400-1e400j;1e400-1e400+1e400j-1e400j

You are better than any of my minions. Well done.

I guess it wasn't as hard as I anticipated. :(

[retina]

Another possible solution I had was this:
1e666+1e666j;1e666-1e666j;1e666j*-1j;-1e666+1e666j;-1e666-1e666j;1e666j*1j;1e666+1e666*1j;1e666-1e666*1j;1e666-1e666j*-1j

It shows some weird occurrences that Python arithmetic exhibits.