控制嵌套列表/字符串上的递归(不检查类型) [英] Control recursion on nested lists / strings (without checking for the type)

问题描述

假设我有以下输入内容:

Assume I have the following input:

items = [1, 2, [3, 4], (5, 6), 'ciao', range(3), (i for i in range(3, 6))]

，我想对items执行一些递归操作.

and I want to perform some recursive operation on items.

为简单起见，假设我要展平项目(但可以是其他项目)，一种方法是:

For the sake of simplicity, let's say I want to flatten items (but could be anything else), one way of doing this would be:

def flatten(items, max_depth=-1, shallow=(str, bytes, bytearray)):
    for item in items:
        if shallow and isinstance(item, shallow) or max_depth == 0:
            yield item
        else:
            try:
                for subitem in flatten(item, max_depth - 1, shallow):
                    yield subitem
            except TypeError:
                yield item

这将产生:

print(list(flatten(items)))
[1, 2, 3, 4, 5, 6, 'ciao', 0, 1, 2, 3, 4, 5]

现在如何修改flatten()，以便产生以下内容(对于任意嵌套级别)?

Now how could I modify flatten() so that I could produce the following (for arbitrary nesting levels)?

print(list(flatten(items, shallow=None)))
[1, 2, 3, 4, 5, 6, 'c', 'i', 'a', 'o', 0, 1, 2, 3, 4, 5]

---

要支持的其他输入:

---

Other inputs to be supported:

items = [['best', 'func'], 'ever']
print(list(flatten(items, shallow=None)))
# ['b', 'e', 's', 't', 'f', 'u', 'n', 'c', 'e', 'v', 'e', 'r']

---

注意:我正在寻找一种不依赖显式检查str，bytes或bytearray的方法，因为它很脆弱，如果它出现具有迭代功能的另一种类型，则该类型会产生项目本身.

---

Note: I am looking for an approach that does not rely on explicitly checking for str, bytes or bytearray as this is quite brittle, should it appear another type which has the property that iterating through it yields the item itself.

这与以下内容有关:对嵌套列表/字符串的控制递归

推荐答案

如果您更改

        if shallow and isinstance(item, shallow) or max_depth == 0:

到

        if shallow and isinstance(item, shallow) or max_depth == 0 or item == items:

然后正确地处理了所有三个示例.

then it gets all three examples right.

这篇关于控制嵌套列表/字符串上的递归(不检查类型)的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！