IndexError：预期范围为[1，3）的暗淡0索引Python错误

Question

In Python, I am having strange error, where a = [x[1], x[2]] works, but a = x[1:] does not.

>>> out                                                                    
farray([Y[0], Y[1], Y[2]])                                                        
>>> out[1]
Y[1]
>>> remaining_out = [out[1], out[2]]
>>> remaining_out[0]
Y[1]
>>> remaining_out = out[1:]
>>> remaining_out[0]
Traceback (most recent call last):
File "<stdin>", line 1, in <module> 
File "/home/kmshah4/.local/lib/python3.6/site-packages/pyeda/boolalg/bfarray.py", line 485, in __getitem__
nsls = self._norm_slices(fsls)                                                                                                                                        File "/home/kmshah4/.local/lib/python3.6/site-packages/pyeda/boolalg/bfarray.py", line 890, in _norm_slices                             nsls.append(_norm_index(i, fsl, *self.shape[i]))
File "/home/kmshah4/.local/lib/python3.6/site-packages/pyeda/boolalg/bfarray.py", line 1127, in _norm_index
raise IndexError(fstr.format(dim, start, stop))
IndexError: expected dim 0 index in range [1, 3)

Please help.

Answer 1

Objects in Python do not automatically support slice notation just because they support indexing. That has to be programmed in. There may be syntax more convenient than what you're using though.

If it's iterable, you can convert to a list first, and then slice it, like

a = [*x][1:]

For especially large iterables (does not seem to be the case here), this may be inefficient. In that case, islice it before you unpack into the list.

from itertools import islice
a = [*islice(x, 1, None)]

But objects do not automatically support iteration just because they support indexing either. If that's the case, you can still iterate manually with range() and a comprehension,

a = [x[i] for i in range(1, len(x))]

Though this isn't really shorter until you have a few more elements to deal with. If you need this pattern a lot you could abstract it into a function.

And finally, objects do not necessarily support len just because they support indexing. If you don't know the length in advance, about all you can do is loop through and catch the LookupError at that point.