OrderedDict 在 Python 3.7 中会变得多余吗? [英] Will OrderedDict become redundant in Python 3.7?

问题描述

来自 Python 3.7 变更日志:

<块引用>

dict 对象的插入顺序保留特性已声明成为 Python 语言规范的官方部分.

这是否意味着 OrderedDict 会变得多余吗?我能想到的唯一用途是保持与旧版 Python 的向后兼容性，旧版 Python 不保留普通词典的插入顺序.

解决方案

不，它不会在 Python 3.7 中变得多余，因为 OrderedDict 不仅仅是一个保留的 dict插入顺序，它还提供了一个依赖顺序的方法，OrderedDict.move_to_end()，并支持reversed() 迭代*.

此外，与 OrderedDict 的相等比较是顺序敏感的，而 Python 3.7 中的 dict 仍然不是这种情况，例如:

<预><代码>>>>OrderedDict([(1,1), (2,2)]) == OrderedDict([(2,2), (1,1)])错误的>>>dict([(1,1), (2,2)]) == dict([(2,2), (1,1)])真的

两个相关问题此处和此处.

* 为 Python 3.8 添加了对常规 Python dict 的 reversed() 迭代的支持，参见 issue33462

From the Python 3.7 changelog:

the insertion-order preservation nature of dict objects has been declared to be an official part of the Python language spec.

Would this mean that OrderedDict will become redundant? The only use I can think of it will be to maintain backwards compatibility with older versions of Python which don't preserve insertion-order for normal dictionaries.

解决方案

No it won't become redundant in Python 3.7 because OrderedDict is not just a dict that retains insertion order, it also offers an order dependent method, OrderedDict.move_to_end(), and supports reversed() iteration*.

Moreover, equality comparisons with OrderedDict are order sensitive and this is still not the case for dict in Python 3.7, for example:

>>> OrderedDict([(1,1), (2,2)]) == OrderedDict([(2,2), (1,1)]) 
False
>>> dict([(1,1), (2,2)]) == dict([(2,2), (1,1)]) 
True

Two relevant questions here and here.

* Support for reversed() iteration of regular Python dict is added for Python 3.8, see issue33462

这篇关于OrderedDict 在 Python 3.7 中会变得多余吗?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！