IMHO it doesn't make sense to restrict the rangeoperator to signed integers when floats can still represent bigger integers

I can sort of sympathise with that.

But for counterpoint, does it make sense to restrict the iteration of a float iterator to unitary integer increments?

I don't know if I understand you right, you mean to allow rational stepwidth?

Something like BASIC's FOR 3.3 TO 7.8 STEP 0.5?

IMHO the C-Style for is already meant to handle this.

Cheers Rolf

Nit: C-Style for doesn't handle that very well. STEP 0.5 is ok since it's a sum of powers of 2 (2^-1), but let's take STEP 0.1 instead:

$ perl -E'for (my $_=0; $_<=7; $_+=0.1) { say }'
...
5.6
5.7
5.8
5.9
5.99999999999999
6.09999999999999
6.19999999999999
6.29999999999999
6.39999999999999
6.49999999999999
6.59999999999999
6.69999999999999
6.79999999999999
6.89999999999999
6.99999999999999
[download]

Generate the floats from integers to avoid compounding errors:

$ perl -E'for (0..70) { say $_/10 }'
...
5.6
5.7
5.8
5.9
6
6.1
6.2
6.3
6.4
6.5
6.6
6.7
6.8
6.9
7
[download]

the C-Style for is already meant to handle this.

Indeed. Which is (probably) why the range operator only handles integers.