Python inf & nan complexity

[uau]

Originally Posted by retina

That is not correct.

Guess I'm too used to using the Python shell only through IPython. In IPython, a line with semicolons will only show the return value from the last statement (and the semicolon-using solutions here do not work), whereas plain Python shell will handle semicolon-separated statements as basically independent input lines - each will produce independent output, but '_' for example will refer to the return value of the last only.

BTW if anyone is using Python through its plain shell, I recommend trying IPython. I think things like multi-line block support are especially useful - if you write a loop with multiple lines, it's in the command history as one object, and you can easily repeat it with one part edited (instead of recalling each single line individually as the default shell would require).

[retina]

Originally Posted by uau

... plain Python shell will handle semicolon-separated statements as basically independent input lines ...

I gave the hint in the first post with the output from float('-inf'); float('nan').

I did give some real hints, along with the fake float hint.

[nordi]

OK, so you were looking for something like

1e666+1e666j;_*_;_*_;1e666-1e666j;_+1e666j;-1e666+1e666j;_-1e666j;-1e666-1e666j;_+1e666


Or, to make it even shorter and less readable:

j=1e666;e=3E333J;e+j;_*_;_*_;j-e;_+e;e-j;_-e;-j-e;_+j

[SmartMersenne]

Originally Posted by nordi

OK, so you were looking for something like

1e666+1e666j;_*_;_*_;1e666-1e666j;_+1e666j;-1e666+1e666j;_-1e666j;-1e666-1e666j;_+1e666


Or, to make it even shorter and less readable:

j=1e666;e=3E333J;e+j;_*_;_*_;j-e;_+e;e-j;_-e;-j-e;_+j

These don't produce the output as requested. Here is what shows up on my computer:

(inf+infj)
(nan+infj)
(nan+nanj)
(inf-infj)
(inf+nanj)
(-inf+infj)
(-inf+nanj)
(-inf-infj)
(nan-infj)

[chalsall]

To share something fellow geeks might enjoy... The Worst Programming Language Ever - Mark Rendle - NDC Oslo 2021

[retina]

A shorter version:
i=1e666;k=1e666j;i+k;i-k;k*-1j;-i+k;-i-k;k*1j;i+i*1j;i-i*1j;i-k*-1j

[nordi]

Originally Posted by SmartMersenne

These don't produce the output as requested. Here is what shows up on my computer:

(inf+infj)
(nan+infj)
(nan+nanj)
(inf-infj)
(inf+nanj)
(-inf+infj)
(-inf+nanj)
(-inf-infj)
(nan-infj)

And that's exactly what was requested:

output all nine possibilities of ±inf and nan as a complex number.

[Gimarel]

If the order doesn't matter this is shorter:
a=666e666;b=666e666j;a+b;-_;_+a;-_;a-b;-_;_-b;-_;_-a
And it should be obvious how to strip 4 more charakters.

[retina]

The order doesn't matter.

"... should look like this." isn't the same as "... shall be this." IMO.

IIUC "like" means similar, not exact. At least in this context. I know there are other meanings of "like", but most of those would just make the sentence look weird.

[SmartMersenne]

Originally Posted by retina

The order doesn't matter.

"... should look like this." isn't the same as "... shall be this." IMO.

IIUC "like" means similar, not exact. At least in this context. I know there are other meanings of "like", but most of those would just make the sentence look weird.

Sorry, my bad. I took it literally including the order of the output.

[axeltherabbit]

my solution:
i=0+9e+999j;r=9e+999;r+i;r-i;r+i-i;-r+i;-r-i;-r+i-i;r-r+i;r-r-i;r-r+i-i

it's a cool problem for a codegolf, I wonder what is the shortest possible solution