What does extended slice syntax actually do for ne

问题:

This question already has an answer here:

• Understanding slice notation 31 answers

The extended slice syntax in python has been explained to me as "a[n:m:k] returns every kth element from n to m".

This gives me a good idea what to expect when k is positive. But I'm lost on how to interpret a[n:m:k] for negative k. I know that a[::-1] reverses a, and that a[::-k] takes ever kth element of the reversed a.

But how is this a generalization of the definition for k positive? I'd like to know how a[n:m:k] is actually defined, so that (for example) I can understand why:

"abcd"[-1:0:-1] = "dcb"

Is a[n:m:-k] reversing the sequence a, then taking the elements with original indices starting from n and ending one before m or something? I don't think so, because this pattern doesn't fit other values of n and m I've tried. But I'm at a loss to figure out how this is actually defined, and searching has gotten me nowhere.

回答1:

[-1:0:-1] means: start from the index len(string)-1 and move up to 0(not included) and take a step of -1(reverse).

So, the following indexes are fetched:

le-1, le-1-1, le-1-1-1  .... 1  # le is len(string)

example:

In [24]: strs = 'foobar'

In [25]: le = len(strs)

In [26]: strs[-1:0:-1]  # the first -1 is equivalent to len(strs)-1

Out[26]: 'raboo'

In [27]: strs[le-1:0:-1]   
Out[27]: 'raboo'

回答2:

The Python documentation (here's the technical one; the explanation for range() is a bit easier to understand) is more correct than the simplified "every kth element" explanation. The slicing parameters are aptly named

slice[start:stop:step]

so the slice starts at the location defined by start, stops before the location stop is reached, and moves from one position to the next by step items.