Cannot round-trip UCS-2 with unpaired surrogates using UTF-16 and surrogateescape

Bug report

The following Python code throws a UnicodeEncodeError:

b'\xdc\xff'.decode('UTF-16BE', 'surrogateescape').encode('UTF-16', 'surrogateescape')

I would expect that it would return b'\xdc\xff' instead.

Your environment

• CPython versions tested on: 3.10.4 from Fedora 35 repository
• Operating system and architecture: Fedora 35 on x86-64

首先在指定的 CPython 版本上运行所报告的 UTF-16BE 解码和 UTF-16 编码代码片段，然后跟踪相关的编码和 surrogateescape 行为。完成的标准是字节 b'\xdc\xff' 可以往返转换而不会出现 UnicodeEncodeError，并根据所报告的预期检查其行为。