1

我正在尝试编写一个类似于某种数据存储的 Python 类。因此，例如，我不想使用字典，而是想以class.foo的形式访问我的数据，并且仍然能够执行所有很酷的事情，例如像在dataclass中的 x一样对其进行迭代。

以下是我想出的：

class MyStore(object):
    def __init__(self, data):
        self._data = {}
        for d in data:
            # Just for the sake of example
            self._data[d] = d.upper()

    def __iter__(self):
        return self._data.values().__iter__()

    def __len__(self):
        return len(self._data)

    def __contains__(self, name):
        return name in self._data

    def __getitem__(self, name):
        return self._data[name]

    def __getattr__(self, name):
        return self._data[name]


store = MyStore(['foo', 'bar', 'spam', 'eggs'])
print "Store items:", [item for item in store]
print "Number of items:", len(store)
print "Get item:", store['foo']
print "Get attribute:", store.foo
print "'foo' is in store:", 'foo' in store

而且，显然它有效。万岁！但是如何正确实现属性的设置呢？添加以下内容会导致递归限制__getattr__：

def __setattr__(self, name, value):
    self._data[name] = value

阅读文档，我应该调用超类（对象）__setattr__方法以避免递归，但这样我就无法控制我的 self._data 字典。

有人可以指出我正确的方向吗？

4

1 回答 1

8

试试这个：

def __setattr__(self, name, value):
    super(MyStore, self).__setattr__(name, value)
    self._data[name] = value

但是，您可以通过子类化以下内容为自己省去很多麻烦dict：

class MyStore(dict):

    def __init__(self, data):
        for d in data:
            self[d] = d.upper()

    def __getattr__(self, name):
        return self[name]

    def __setattr__(self, name, value):
        self[name] = value

store = MyStore(['foo', 'bar', 'spam', 'eggs'])
print "Store items:", [item for item in store]
print "Number of items:", len(store)
print "Get item:", store['foo']
print "Get attribute:", store.foo
print "'foo' is in store:", 'foo' in store
store.qux = 'QUX'
print "Get qux item:", store['qux']
print "Get qux attribute:", store.qux
print "'qux' is in store:", 'qux' in store

哪个输出...

Store items: ['eggs', 'foo', 'bar', 'spam']
Number of items: 4
Get item: FOO
Get attribute: FOO
'foo' is in store: True
Get qux item: QUX
Get qux attribute: QUX
'qux' is in store: True
于 2012-03-03T22:56:18.127 回答