Python 3.6 有什么新变化¶

编者:: Elvis Pranskevichus <elvis@magic.io>, Yury Selivanov <yury@magic.io>

本文解释了与3.5相比，Python 3.6中的新功能。 Python 3.6于2016年12月23日发布。请参阅 changelog 以获取完整的更改列表。

参见

PEP 494 - Python 3.6 发布计划

摘要 -- 发布重点¶

新的语法特性：

PEP 498, 格式化字符串字面值
PEP 515, 数字字面值中的下划线。
PEP 526 , 变量标注语法。
PEP 525，异步生成器。
PEP 530: 异步推导式。

新的库模块：

secrets: PEP 506 -- Python 标准库增加一个密码模块。

CPython 实现的改进：

根据 Raymond Hettinger 的提议已将 dict 类型重新实现为使用更紧凑的表示形式并类似于 PyPy dict implementation。这使得字典所使用的内存相对于 Python 3.5 版减少了 20% 到 25%。
类创建的定制过程通过新协议得到了简化。
类属性的定义顺序现在会被保留。
现在 **kwargs 中的元素会与传给函数的关键字参数保持对应顺序。
添加了 DTrace 和 SystemTap 探测支持
新的 PYTHONMALLOC 环境变量可被用来调试解释器的内存分配和访问错误。

标准库中的重大改进：

asyncio 模块获得了许多新特性，显著的可用性和性能提升，以及大量的问题修正。从 Python 3.6 开始 asyncio 模块不再处于待定状态，它已被视为稳定 API。
实现了新的文件系统路径协议以支持路径类对象。所有操作路径的标准库函数已被更新为使用新的协议。
datetime 模块已获得对本地时间消歧义的支持。
typing 模块获得了大量的改进。
tracemalloc 模块已被大幅重写，现在将被用于为 ResourceWarning 提供更好的输出，并为内存分配错误提供更好的诊断。请参阅 PYTHONMALLOC 一节来了解详情。

安全改进：

添加了 secrets 模块以简化适用于密码管理的高加密强度伪随机数的生成，例如账户验证、安全凭据等场景。
在 Linux 上，现在 os.urandom() 会阻塞直到系统的 urandom 熵池被初始化以提升安全性。其理由参见 PEP 524。
hashlib 和 ssl 模块现在支持 OpenSSL 1.1.0。
ssl 模块的默认设置和特性集已得到改进。
hashlib 模块获得了对 BLAKE2, SHA-3 和 SHAKE 哈希算法以及 scrypt() 密钥派生函数的支持。

Windows改进：

PEP 528 和 PEP 529, 将Windows文件系统和控制台的编码更改为UTF-8
在交互式地使用 py.exe 启动器时，当用户未（通过命令行参数或配置文件）指定版本时不再优先选择 Python 2 而是选择 Python 3。对声明行的处理则保持不变 —— 在这种情况下 "python" 还是指 Python 2。
python.exe 和 pythonw.exe 已被标记为支持长路径，这意味着不再有 260 个字符的路径长度限制。详情参见移除 MAX_PATH 限制。
可以添加一个 ._pth 文件来强制使用隔离模式和完整指定所有搜索路径来避免注册表和环境查找。更多信息请参阅相关文档。
现在 python36.zip 文件可以作为推断 PYTHONHOME 的标记物。请参阅相关文档了解详情。

新的特性¶

PEP 498: 格式化字符串字面值¶

PEP 498 引入了一种新型的字符串字面值: f-字符串，或称格式化字符串字面值。

格式化字符串字面值带有 'f' 前缀并且类似于 str.format() 所接受的格式字符串。其中包含由花括号包围的替换字段。替换字段属于表达式，它们会在运行时被求值，然后使用 format() 协议进行格式化:

>>> name = "Fred"
>>> f"He said his name is {name}."
'He said his name is Fred.'
>>> width = 10
>>> precision = 4
>>> value = decimal.Decimal("12.34567")
>>> f"result: {value:{width}.{precision}}"  # nested fields
'result:      12.35'

参见

PEP 498 -- 字符串字面值插值。: PEP 由 Eric V. Smith 撰写并实现

特性文档。

PEP 526: 变量标注的语法¶

PEP 484 引入了函数形参类型标注即类型提示的标准。这个 PEP 为 Python 添加了标注变量类型的语法，包括类变量和实例变量:

primes: List[int] = []

captain: str  # Note: no initial value!

class Starship:
    stats: Dict[str, int] = {}

与函数标注一样，Python 解释器不会为变量标注附加任何特殊含义，仅会将其保存在类或模块的 __annotations__ 属性中。

与静态类型语法的变量声明不同，标注语法的目的是通过抽象语法树和 __annotations__ 属性提供一个简单方式来为第三方工具和库指定结构化类型元数据。

参见

PEP 526 -- 变量标注的语法。: PEP 由 Ryan Gonzalez, Philip House, Ivan Levkivskyi, Lisa Roach, 和 Guido van Rossum 撰写，由 Ivan Levkivskyi 实现。

使用或将要使用此新语法的工具有: mypy, pytype, PyCharm 等等。

PEP 515: 数字字面值中的下划线。¶

PEP 515 增加了在数字字面值中使用下划线的能力以改善可读性。例如:

>>> 1_000_000_000_000_000
1000000000000000
>>> 0x_FF_FF_FF_FF
4294967295

单个下划线允许用在数码之间和任何数制指示符之后。一行内不允许有开头、末尾或多个下划线。

字符串格式化微语言现在也支持以 '_' 选项来表示用下划线作为浮点表示类型和整数表示类型 'd' 的千位分隔符。对于整数表示类型 'b', 'o', 'x' 和 'X'，将每隔 4 个数码插入一个下划线:

>>> '{:_}'.format(1000000)
'1_000_000'
>>> '{:_x}'.format(0xFFFFFFFF)
'ffff_ffff'

参见

PEP 515 -- 数字字面值中的下划线。: PEP 由 Georg Brandl 和 Serhiy Storchaka 撰写

PEP 525: 异步生成器¶

PEP 492 将对原生协程和 async / await 语法的支持引入到 Python 3.5 中。但 Python 3.5 实现的一个明显限制是不可能在同一函数体中同时使用 await 和 yield。在 Python 3.6 中此限制已被解除，这样就就能够定义 异步生成器:

async def ticker(delay, to):
    """Yield numbers from 0 to *to* every *delay* seconds."""
    for i in range(to):
        yield i
        await asyncio.sleep(delay)

这个新语法允许更快速且更精简的代码。

参见

PEP 525 -- 异步生成器: PEP 由 Yury Selivanov 撰写并实现

PEP 530: 异步推导式¶

PEP 530 添加了对在列表、集合与字典推导式和生成器表达式中使用 async for 的支持:

result = [i async for i in aiter() if i % 2]

此外，await 表达式也在所有种类的推导式中得到支持:

result = [await fun() for fun in funcs if await condition()]

参见

PEP 530 -- 异步推导式: PEP 由 Yury Selivanov 撰写并实现

PEP 487: 更简单的自定义类创建¶

现在可以在不使用元类的情况下自定义子类的创建。当一个新的子类被创建时将在基类上调用新的 __init_subclass__ 类方法:

class PluginBase:
    subclasses = []

    def __init_subclass__(cls, **kwargs):
        super().__init_subclass__(**kwargs)
        cls.subclasses.append(cls)

class Plugin1(PluginBase):
    pass

class Plugin2(PluginBase):
    pass

为了让来自 __init_subclass__() 实现的零参数 super() 调用能正确工作，自定义元类必须保证新的 __classcell__ 命名空间入口被传播到 type.__new__ (如创建类对象中所描述)。

参见

PEP 487 -- 更简单的自定义类创建: PEP 由 Martin Teichmann 撰写并实现。

特性文档

PEP 487: 描述器协议的增强¶

PEP 487 扩展了描述器协议以包括新的可选方法 __set_name__()。当创建一个新类时，这个新方法将在定义中包括的所有描述器上被调用，为它们提供对所定义类的引用以及在类命名中间中给予描述器的名称。换句话说，描述器的实例现在能知道描述器在所有者类中的属性名称:

class IntField:
    def __get__(self, instance, owner):
        return instance.__dict__[self.name]

    def __set__(self, instance, value):
        if not isinstance(value, int):
            raise ValueError(f'expecting integer in {self.name}')
        instance.__dict__[self.name] = value

    # this is the new initializer:
    def __set_name__(self, owner, name):
        self.name = name

class Model:
    int_field = IntField()

参见

PEP 487 -- 更简单的自定义类创建: PEP 由 Martin Teichmann 撰写并实现。

特性文档

PEP 519: 添加文件系统路径协议¶

文件系统路径历来被表示为 str 或 bytes 对象。这使得编写对文件系统路径进行操作的代码的人会假定这种对象只能是这两种类型之一（不考虑表示文件描述符的 int 因为它不是文件路径）。不幸的是，这种假定阻止了像 pathlib 这样的文件系统路径的替代对象表示形式与包括 Python 标准库在内的现有代码协同工作。

为了修正这种情况，一个由 os.PathLike 表示的新接口被定义出来。通过实现 __fspath__() 方法，对象可以表明它代表一个路径。这样一个对象就能以 str 或 bytes 对象的形式提供文件系统路径的低层级表示。这意味着如果一个对象实现了 os.PathLike 或者是表示文件系统路径的 str 或 bytes 对象它就会被当作是路径型对象。代码可以使用 os.fspath()、os.fsdecode() 或 os.fsencode() 来显式地获取一个路径型对象的 str 和/或 bytes 表示形式。

内置 open() 函数已被更新为接受 os.PathLike 对象，就像 os 和 os.path 模块中的所有相关函数和标准库中的大多数其他函数和类一样。 os.DirEntry 类和 pathlib 中的相关类也已被更新以实现 os.PathLike。

希望对于针对文件系统路径操作的基本函数的更新将会使得第三方代码都能隐式地支持所有路径型对象而无须修改任何代码，或者至少只需极少的修改（例如当对路径型对象进行操作之前在代码的开头调用 os.fspath() 即可）。

下面的这些示例说明了新的接口是如何让 pathlib.Path 被更容易、更透明地用于已有代码的:

>>> import pathlib
>>> with open(pathlib.Path("README")) as f:
...     contents = f.read()
...
>>> import os.path
>>> os.path.splitext(pathlib.Path("some_file.txt"))
('some_file', '.txt')
>>> os.path.join("/a/b", pathlib.Path("c"))
'/a/b/c'
>>> import os
>>> os.fspath(pathlib.Path("some_file.txt"))
'some_file.txt'

（由 Brett Cannon, Ethan Furman, Dusty Phillips 和 Jelle Zijlstra 实现。）

参见

PEP 519 -- 添加文件系统路径协议: PEP 由 Brett Cannon 和 Koos Zevenhoven 撰写。

PEP 495: 消除本地时间的歧义¶

在世界上大多数地方，过去和将来都会存在本地时钟后移的时候。在这种时候，本地时钟会在同一天内两次显示相同的时间。对于这些情况，本地时钟所显示的（或存储在 Python datetime 实例中的）信息将不足以标识某个特定的时间点。

PEP 495 为 datetime.datetime 和 datetime.time 类的实例添加了新的 fold 属性以在本地时间相同的两个时间点之间进行区分:

>>> u0 = datetime(2016, 11, 6, 4, tzinfo=timezone.utc)
>>> for i in range(4):
...     u = u0 + i*HOUR
...     t = u.astimezone(Eastern)
...     print(u.time(), 'UTC =', t.time(), t.tzname(), t.fold)
...
04:00:00 UTC = 00:00:00 EDT 0
05:00:00 UTC = 01:00:00 EDT 0
06:00:00 UTC = 01:00:00 EST 1
07:00:00 UTC = 02:00:00 EST 0

fold 属性的值在大多数实例上均为 0，只有在代表存在歧义的时间的第二个（按发生顺序）实例上例外。

参见

PEP 495 -- 消除本地时间的歧义: PEP 由 Alexander Belopolsky 和 Tim Peters 撰写，由 Alexander Belopolsky 实现。

PEP 529: 将 Windows 文件系统编码格式更改为 UTF-8¶

在表示文件系统路径时最好是使用 str (Unicode) 而不是 bytes。不过，在某些情况下使用 bytes 就是足够而且正确的。

在 Python 3.6 之前，在 Windows 上使用 bytes 路径值可能会导致数据丢失。有了这一更改，现在 Windows 上将支持使用 bytes 来表示路径，前提是这些 bytes 值是用 sys.getfilesystemencoding() 所返回的编码格式来编码的，现在此编码格式默认为 'utf-8'。

不使用 str 来表示路径的应用程序应当使用 os.fsencode() 和 os.fsdecode() 来确保 bytes 被正确地编码。要恢复以前的行为，请设置 PYTHONLEGACYWINDOWSFSENCODING 或调用 sys._enablelegacywindowsfsencoding()。

有关详细信息和可能需要的代码修改的讨论，请参见 PEP 529 。

PEP 528: 将 Windows 控制台编码格式更改为 UTF-8¶

现在，Windows 上的默认控制台将接受所有 Unicode 字符并为 Python 代码提供正确读取的 str 对象。sys.stdin, sys.stdout 和 sys.stderr 现在默认使用 utf-8 编码格式。

此更改仅在使用交互式控制台时适用，而不适用于重定向文件或管道。要恢复以前使用交互式控制台时的行为，请设置 PYTHONLEGACYWINDOWSSTDIO。

参见

PEP 528 -- 将 Windows 控制台编码格式更改为 UTF-8: PEP 由 Steve Dower 撰写并实现。

PEP 520: 保留类属性定义顺序¶

类定义体中的属性具有自然的排序：与名称在源代码中出现的顺序相同。现在此排序会在新类的 __dict__ 属性中被保留。

并且，实际的默认类 execution 命名空间 (从 type.__prepare__() 返回) 现在将是一个保留插入顺序的映射对象。

参见

PEP 520 -- 保留类属性定义顺序: PEP 由 Eric Snow 撰写并实现

PEP 468: 保留关键字参数顺序¶

函数签名中的 **kwargs 现在将保证是一个保留插入顺序的映射对象。

参见

PEP 468 -- 保留关键字参数顺序: PEP 由 Eric Snow 撰写并实现

新的 dict 实现¶

dict 类型现在会使用一种基于 Raymond Hettinger 的提议的“紧凑”表示形式，该表示形式最初由 PyPy 实现。新的The memory usage of the new dict() 的内存占用相比 Python 3.5 减少了 20% 到 25%。

这个新实现对原始顺序的保留被认为是一个实现细节而不应当被依赖（这在将来可能会改变，但我们希望在改变语言规范以强制所有当前和将来的 Python 实现都使用保留顺序的语义之前先在几个发布版的语言内部使用这个新的 dict 实现；这也有助于让仍在使用随机迭代顺序的旧版本语言例如 Python 3.5 保持向下兼容性）。

（由 INADA Naoki 在 bpo-27350 中贡献。该特性最初由 Raymond Hettinger 提议。）

PEP 523: 向 CPython 添加帧求值 API¶

虽然 Python 为自定义代码执行方式提供了广泛的支持，但有一个地方它没有这样做，那就是帧对象的求值。如果您想在 Python 中拦截帧的求值，那么确实没有除了直接操纵自定义函数的函数指针以外的任何办法。

PEP 523 通过提供使帧求值在 C 语言层级上可插入的 API 从而改变了这一状况。这将允许调试器和 JIT 等工具在 Python 代码开始执行之前拦截帧求值。这样就能允许 Python 代码使用替代性求值实现，跟踪帧求值等做法。

这个 API 并不是受限 C API 的组成部分，它被标记为私有以表明该 API 的使用受到限制并且只适用于非常少的、低层级的用例。这个 API 的语义将根据需要随 Python 的一起发生变化。

参见

PEP 523 -- 向 CPython 添加帧求值 API: PEP 由 Brett Cannon 和 Dino Viehland 撰写。

PYTHONMALLOC 环境变量¶

新的 PYTHONMALLOC 环境变量允许设置 Python 内存分配器并安装调试钩子。

现在将可以使用 PYTHONMALLOC=debug 在以发布模式编译的 Python 上为 Python 内存分配器安装调试钩子。调试钩子的效果:

新分配的内存中填充字节 0xCB
释放的内存中填充了字节 0xDB
检测违反 Python 内存分配器 API 的操作。例如，PyObject_Free() 在 PyMem_Malloc() 所分配的内存块上被调用。
在缓冲区开始之前检测写操作（缓冲区下溢）
在缓冲区结束后检测写操作（缓冲区溢出）
检测当调用 PYMEM_DOMAIN_OBJ (如: PyObject_Malloc()) 和 PYMEM_DOMAIN_MEM (如: PyMem_Malloc()) 域的分配器函数时是否持有 GIL。

检查是否保留了 GIL 也是Python 3.6 的新特性。

请参阅 PyMem_SetupDebugHooks() 函数来了解 Python 内存分配器上的调试钩子。

现在还可以使用 PYTHONMALLOC=malloc 为所有的 Python 内存分配强制使用 C 库的 malloc() 分配器。这在以发布模式编译的 Python 上使用外部内存调试器如 Valgrind 时会很有用处。

发生错误时，Python 内存分配器上的调试钩子现在会使用 tracemalloc 模块来获取内存块被分配所在位置上的回溯。

使用 python3.6 -X tracemalloc=5 (在回溯中存储 5 帧) 的缓冲区溢出的致命错误示例:

Debug memory block at address p=0x7fbcd41666f8: API 'o'
    4 bytes originally requested
    The 7 pad bytes at p-7 are FORBIDDENBYTE, as expected.
    The 8 pad bytes at tail=0x7fbcd41666fc are not all FORBIDDENBYTE (0xfb):
        at tail+0: 0x02 *** OUCH
        at tail+1: 0xfb
        at tail+2: 0xfb
        at tail+3: 0xfb
        at tail+4: 0xfb
        at tail+5: 0xfb
        at tail+6: 0xfb
        at tail+7: 0xfb
    The block was made by call #1233329 to debug malloc/realloc.
    Data at p: 1a 2b 30 00

Memory block allocated at (most recent call first):
  File "test/test_bytes.py", line 323
  File "unittest/case.py", line 600
  File "unittest/case.py", line 648
  File "unittest/suite.py", line 122
  File "unittest/suite.py", line 84

Fatal Python error: bad trailing pad byte

Current thread 0x00007fbcdbd32700 (most recent call first):
  File "test/test_bytes.py", line 323 in test_hex
  File "unittest/case.py", line 600 in run
  File "unittest/case.py", line 648 in __call__
  File "unittest/suite.py", line 122 in run
  File "unittest/suite.py", line 84 in __call__
  File "unittest/suite.py", line 122 in run
  File "unittest/suite.py", line 84 in __call__
  ...

（由 Victor Stinner 在 bpo-26516 和 bpo-26564 中贡献。）

DTrace 和 SystemTap 探测支持¶

Python 现在可以附带 --with-dtrace 来构建以便为解释器中的下列事件启用静态标记:

函数调用/返回
垃圾收集开始/完成
执行的代码行。

这可被用来在生产环境中控制正在运行的解释器，而无需重新编译特定的调试版本或提供应用专属的性能分析/调试代码。

更多信息，请参见使用 DTrace 和 SystemTap 检测CPython 。

当前的实现已在 Linux 和 macOS 上进行了测试。将来可能会添加其他标记。

（由 Łukasz Langa 在 bpo-21590 中贡献，基于 Jesús Cea Avión, David Malcolm 和 Nikhil Benesch 的补丁。）

其他语言特性修改¶

对Python 语言核心进行的小改动：

现在 global 或 nonlocal 语句必须以文本形式出现在同一作用域中首次使用受影响的名称之前。在之前版本中这只是 SyntaxWarning。
现在可以将某个特殊方法设为 None 来表示相应操作不可用。举例来说，如果某个类将 __iter__() 设为 None，则该类就将不可迭代。（由 Andrew Barnert 和 Ivan Levkivskyi 在 bpo-25958 中贡献。）
由重复的回溯行组成的长序列现在将被简化为 "[Previous line repeated {count} more times]" （请参阅回溯获取样例）。（由 Emanuel Barry 在 bpo-26823 中贡献。）
现在导入操作在无法找到模块时将引发新的异常 ModuleNotFoundError (ImportError 的子类)。目前 (在 try-except 中) 检测 ImportError 的代码仍将有效。（由 Eric Snow 在 bpo-15767 中贡献。）
现在依赖于零参数形式 super() 的类方法在类创建期间从元类方法调用时将正确地生效。（由 Martin Teichmann 在 bpo-23722 中贡献。）

新增模块¶

secrets¶

新的 secrets 模块的主要目的是提供一种直观的方式来可靠地生成适用于密码管理的高加密强度的伪随机值，如账户验证、安全凭据等等。

警告

注意 random 模块中的伪随机数发生器不应被用于安全目的。请在 Python 3.6+ 上使用 secrets 而在 Python 3.5 及更早的版本上使用 os.urandom()。

参见

PEP 506 -- Secrets模块被加入Python标准库: PEP 由 Steven D'Aprano 撰写并实现。

改进的模块¶

array¶

现在已被耗尽的输出 array.array 的迭代器将保持耗尽状态，即使迭代后的数组被扩展时也是如此。这将与其他可变序列的行为保持一致。

由 Serhiy Storchaka 在 bpo-26492 中贡献。

ast¶

新增了 ast.Constant AST 节点。它可被外部 AST 优化器用于常量折叠操作。

由 Victor Stinner 在 bpo-26146 中贡献。

asyncio¶

从 Python 3.6 开始 asyncio 模块不再处于暂定状态，其 API 被认为已经稳定。

自 Python 3.5.0 以来 asyncio 模块中值得注意的变化（由于暂定状态所有变化都已反向移植到 3.5.x）:

get_event_loop() 函数已更改为当在例程和回调中被调用时始终返回当前正在运行的循环。（由 Yury Selivanov 在 bpo-28613 中贡献。）
ensure_future() 函数以及所有用到它的函数，比如 loop.run_until_complete()，现在将接受所有种类的可等待对象。（由 Yury Selivanov 贡献。）
新增 run_coroutine_threadsafe() 函数用于从其他线程向事件循环提交协程。（由 Vincent Michel 贡献。）
新增 Transport.is_closing() 方法用于检查传输是否正在关闭或已经关闭。（由 Yury Selivanov 贡献。）
loop.create_server() 方法现在可以接受一个主机列表。（由 Yann Sionneau 贡献。）
新增 loop.create_future() 方法用来创建 Future 对象。这允许替代性的事件循环实现，比如 uvloop，以提供更快速的 asyncio.Future 实现。（由 Yury Selivanov 在 bpo-27041 中贡献。）
新增 loop.get_exception_handler() 方法用于获取当前异常处理句柄。（由 Yury Selivanov 在 bpo-27040 中贡献。）
新增 StreamReader.readuntil() 方法用于从流读取数据直到出现作为分隔符的字节序列。（由 Mark Korenberg 贡献。）
StreamReader.readexactly() 的性能已获得提升。（由 Mark Korenberg 在 bpo-28370 中贡献。）
loop.getaddrinfo() 方法已获得优化已避免当地址已被解析时调用系统 getaddrinfo 函数。（由 A. Jesse Jiryu Davis 贡献。）
loop.stop() 方法已被更改为在当前迭代之后立即停止循环。任何作为上次迭代的结果被加入计划任务的新回调都将被丢弃。（由 Guido van Rossum 在 bpo-25593 中贡献。）
现在 Future.set_exception 在传入一个 StopIteration 异常的实例时将引发 TypeError。（由 Chris Angelico 在 bpo-26221 中贡献。）
新增 loop.connect_accepted_socket() 方法供接受 asyncio 以外的连接，但使用 asyncio 来处理它们的服务器使用。（由 Jim Fulton 在 bpo-27392 中贡献。）
现在 TCP_NODELAY 旗标将默认针对所有 TCP 传输进行设置。（由 Yury Selivanov 在 bpo-27456 中贡献。）
新增 loop.shutdown_asyncgens() 用来在结束循环之前正确地关闭现有的异步生成器。（由 Yury Selivanov 在 bpo-28003 中贡献。）
Future 和 Task 类现在已有经优化过的 C 实现使得 asyncio 代码加速至多 30%。（由 Yury Selivanov 和 INADA Naoki 在 bpo-26081 和 bpo-28544 中贡献。）

binascii¶

b2a_base64() 函数现在接受可选的 newline 关键字参数用来控制是否要在返回值中添加换行符。（由 Victor Stinner 在 bpo-25357 中贡献。）

cmath¶

新增 cmath.tau (τ) 常量。（由 Lisa Roach 在 bpo-12345 中贡献，详情见 PEP 628。）

新增常量: cmath.inf 和 cmath.nan 用于匹配 math.inf 和 math.nan，以及 cmath.infj 和 cmath.nanj 用于匹配由 complex 的 repr 所使用的格式。（由 Mark Dickinson 在 bpo-23229 中贡献。）

collections¶

添加了新的 Collection 抽象基类用于表示有具体大小的可迭代容器类。（由 Ivan Levkivskyi 在 bpo-27598 中贡献并由 Neil Girdhar 撰写文档。）

添加了新的 Reversible 抽象基类用于表示同时提供 __reversed__() 方法的可迭代类。（由 Ivan Levkivskyi 在 bpo-25987 中贡献。）

新增代表异步生成器的 AsyncGenerator 抽象基类。（由 Yury Selivanov 在 bpo-28720 中贡献。）

namedtuple() 函数现在接受可选的关键字参数 module，当指定该参数时，它将被用作所返回的具名元组类的 __module__ 属性。（由 Raymond Hettinger 在 bpo-17941 中贡献。）

namedtuple() 的 verbose 和 rename 参数现在是仅限关键字参数。（由 Raymond Hettinger 在 bpo-25628 中贡献。）

递归的 collections.deque 实例现在可以被 pickle。（由 Serhiy Storchaka 在 bpo-26482 中贡献。）

concurrent.futures¶

ThreadPoolExecutor 类构造器现在接受可选的 thread_name_prefix 参数以便能够自定义由线程池所创建的线程的名称。（由 Gregory P. Smith 在 bpo-27664 中贡献。）

contextlib¶

增加了 contextlib.AbstractContextManager 类用来提供上下文管理器的抽象基类。它为 __enter__() 提供了一个合理的默认实现，该实现将返回 self 并将 __exit__() 设为抽象方法。在 typing 中增加了对应的类 typing.ContextManager。（由 Brett Cannon 在 bpo-25609 中贡献。）

datetime¶

datetime 和 time 类新增了 fold 属性用来在必要时消除本地时间的歧义。在 datetime 中的许多函数已被更新为支持本地时间的消除歧义。请参阅本地时间消歧义一节了解更多信息。（由 Alexander Belopolsky 在 bpo-24773 中贡献。）

现在 datetime.strftime() 和 date.strftime() 方法将支持 ISO 8601 日期指令符 %G, %u 和 %V。（由 Ashley Anderson 在 bpo-12006 中贡献。）.)

datetime.isoformat() 函数现在接受可选的 timespec 参数用来指定时间值要包括的额外组件数量。（由 Alessandro Cucci 和 Alexander Belopolsky 在 bpo-19475 中贡献。）

datetime.combine() 现在接受可选的 tzinfo 参数。（由 Alexander Belopolsky 在 bpo-27661 中贡献。）

decimal¶

新增 Decimal.as_integer_ratio() 方法，它返回一对整数 (n, d) 将给定的 Decimal 实例表示为一个最简形式且分母为正值的分数:

>>> Decimal('-3.14').as_integer_ratio()
(-157, 50)

（由 Stefan Krah 和 Mark Dickinson 在 bpo-25928 中贡献。）

distutils¶

distutils.command.sdist.sdist 的 default_format 属性已被移除且 formats 属性默认为 ['gztar']。虽然不作要求，但任何依赖于 default_format 的存在的代码都可能需要修改。请参阅 bpo-27819 了解更多细节。

email¶

通过多个构造器的 policy 关键字来启用的新 email API 已不再为暂定状态。 email 文档已被重新组织并重新撰写以聚集新 API，同时保留旧式 API 的原有文档。（由 R. David Murray 在 bpo-24277 中贡献。）

email.mime 中的类现在都接受可选的 policy 关键字参数。（由 Berker Peksag 在 bpo-27331 中贡献。）.)

DecodedGenerator 现在支持 policy 关键字。

新增 policy 属性，message_factory 控制当解析器新建消息对象时默认要使用的类。对于 email.policy.compat32 策略来说将为 Message，对于新策略来说将为 EmailMessage。（由 R. David Murray 在 bpo-20476 中贡献。）

encodings¶

在 Windows 上，增加了 'oem' 编码格式用于 CP_OEMCP，以及 'ansi' 别名用于现有的 'mbcs' 编码格式，它使用 CP_ACP 代码页。（由 Steve Dower 在 bpo-27959 中贡献。）

enum¶

在 enum 模块中新增了两个枚举基类: Flag 和 IntFlags。两者均被用于定义可使用按位运算符进行组合的常量。（由 Ethan Furman 在 bpo-23591 中贡献。）

许多标准库模块已被更新以使用 IntFlags 类作为其常量。

新增的 enum.auto 值可被用于自动为枚举成员赋值:

>>> from enum import Enum, auto
>>> class Color(Enum):
...     red = auto()
...     blue = auto()
...     green = auto()
...
>>> list(Color)
[<Color.red: 1>, <Color.blue: 2>, <Color.green: 3>]

faulthandler¶

在 Windows 上，faulthandler 模块现在会为 Windows 异常安装处理句柄：参见 faulthandler.enable()。（由 Victor Stinner 在 bpo-23848 中贡献。）

fileinput¶

hook_encoded() 现在支持 errors 参数。（由 Joseph Hackman 在 bpo-25788 中贡献。）

hashlib¶

hashlib 已支持 OpenSSL 1.1.0。最低的建议版本为 1.0.2。（由 Christian Heimes 在 bpo-26470 中贡献。）

本模块增加了 BLAKE2 哈希函数。 blake2b() 和 blake2s() 将始终可用并支持 BLAKE2 的完整特性集。（由 Christian Heimes 在 bpo-26798 中基于 Dmitry Chestnykh 和 Samuel Neves 的代码贡献。文档由 Dmitry Chestnykh 撰写。）

增加了 SHA-3 哈希函数 sha3_224(), sha3_256(), sha3_384(), sha3_512()，以及 SHAKE 哈希函数 shake_128() 和 shake_256()。（由 Christian Heimes 在 bpo-16113 中贡献。 Keccak 代码包由 Guido Bertoni, Joan Daemen, Michaël Peeters, Gilles Van Assche 和 Ronny Van Keer 编写。）

The password-based key derivation function scrypt() is now available with OpenSSL 1.1.0 and newer. (Contributed by Christian Heimes in bpo-27928.)

http.client¶

HTTPConnection.request() and endheaders() both now support chunked encoding request bodies. (Contributed by Demian Brecht and Rolf Krahl in bpo-12319.)

idlelib 与 IDLE¶

The idlelib package is being modernized and refactored to make IDLE look and work better and to make the code easier to understand, test, and improve. Part of making IDLE look better, especially on Linux and Mac, is using ttk widgets, mostly in the dialogs. As a result, IDLE no longer runs with tcl/tk 8.4. It now requires tcl/tk 8.5 or 8.6. We recommend running the latest release of either.

'Modernizing' includes renaming and consolidation of idlelib modules. The renaming of files with partial uppercase names is similar to the renaming of, for instance, Tkinter and TkFont to tkinter and tkinter.font in 3.0. As a result, imports of idlelib files that worked in 3.5 will usually not work in 3.6. At least a module name change will be needed (see idlelib/README.txt), sometimes more. (Name changes contributed by Al Swiegart and Terry Reedy in bpo-24225. Most idlelib patches since have been and will be part of the process.)

In compensation, the eventual result with be that some idlelib classes will be easier to use, with better APIs and docstrings explaining them. Additional useful information will be added to idlelib when available.

在 3.6.2 中新增:

多个对自动补全的修正。（由 Louie Lu 在 bpo-15786 中贡献。）

在 3.6.3 中新增:

Module Browser (在 File 菜单中，之前称为 Class Browser) 现在会在最高层级函数和类之外显示嵌套的函数和类。（由 Guilherme Polo, Cheryl Sabella 和 Terry Jan Reedy 在 bpo-1612262 中贡献。）

之前以扩展形式实现的 IDLE 特性已作为正常特性重新实现。它们的设置已从 Extensions 选项卡移至其他对话框选项卡。（由 Charles Wohlganger 和 Terry Jan Reedy 在 bpo-27099 中实现。）

Settings 对话框 (Options 中的 Configure IDLE) 已经被部分重写以改进外观和功能。（由 Cheryl Sabella 和 Terry Jan Reedy 在多个问题项中贡献。）

在 3.6.4 中新增:

字体样本现在包括一组非拉丁字符以便用户能更好地查看所选特定字体的效果。（由 Terry Jan Reedy 在 bpo-13802 中贡献。）样本可以被修改以包括其他字符。（由 Serhiy Storchaka 在 bpo-31860 中贡献。）

在 3.6.6 中新增:

编辑器代码上下文选项已经过修改。 Box 会显示所有上下文行直到最大行数。点击一个上下文行会使编辑器跳转到该行。自定义主题的上下文颜色已添加到 Settings 对话框的 Highlights 选项卡。（由 Cheryl Sabella 和 Terry Jan Reedy 在 bpo-33642, bpo-33768 和 bpo-33679 中贡献。）

在 Windows 上，会有新的 API 调用将 tk 对 DPI 的调整告知 Windows。在 Windows 8.1+ 或 10 上，如果 Python 二进制码的 DPI 兼容属性未改变，并且监视器分辨率大于 96 DPI，这应该会令文本和线条更清晰。否则的话它应该不造成影响。（由 Terry Jan Reedy 在 bpo-33656 中贡献。）

在 3.6.7 中新增:

超过 N 行（默认值为 50）的输出将被折叠为一个按钮。 N 可以在 Settings 对话框的 General 页的 PyShell 部分中进行修改。数量较少但是超长的行可以通过在输出上右击来折叠。被折叠的输出可通过双击按钮来展开，或是通过右击按钮来放入剪贴板或是单独的窗口。（由 Tal Einat 在 bpo-1529353 中贡献。）

importlib¶

Import now raises the new exception ModuleNotFoundError (subclass of ImportError) when it cannot find a module. Code that current checks for ImportError (in try-except) will still work. (Contributed by Eric Snow in bpo-15767.)

importlib.util.LazyLoader now calls create_module() on the wrapped loader, removing the restriction that importlib.machinery.BuiltinImporter and importlib.machinery.ExtensionFileLoader couldn't be used with importlib.util.LazyLoader.

importlib.util.cache_from_source(), importlib.util.source_from_cache(), and importlib.util.spec_from_file_location() now accept a path-like object.

inspect¶

The inspect.signature() function now reports the implicit .0 parameters generated by the compiler for comprehension and generator expression scopes as if they were positional-only parameters called implicit0. (Contributed by Jelle Zijlstra in bpo-19611.)

To reduce code churn when upgrading from Python 2.7 and the legacy inspect.getargspec() API, the previously documented deprecation of inspect.getfullargspec() has been reversed. While this function is convenient for single/source Python 2/3 code bases, the richer inspect.signature() interface remains the recommended approach for new code. (Contributed by Nick Coghlan in bpo-27172)

json¶

json.load() and json.loads() now support binary input. Encoded JSON should be represented using either UTF-8, UTF-16, or UTF-32. (Contributed by Serhiy Storchaka in bpo-17909.)

logging¶

The new WatchedFileHandler.reopenIfNeeded() method has been added to add the ability to check if the log file needs to be reopened. (Contributed by Marian Horban in bpo-24884.)

math¶

The tau (τ) constant has been added to the math and cmath modules. (Contributed by Lisa Roach in bpo-12345, see PEP 628 for details.)

multiprocessing¶

Proxy Objects returned by multiprocessing.Manager() can now be nested. (Contributed by Davin Potts in bpo-6766.)

os¶

See the summary of PEP 519 for details on how the os and os.path modules now support path-like objects.

scandir() now supports bytes paths on Windows.

A new close() method allows explicitly closing a scandir() iterator. The scandir() iterator now supports the context manager protocol. If a scandir() iterator is neither exhausted nor explicitly closed a ResourceWarning will be emitted in its destructor. (Contributed by Serhiy Storchaka in bpo-25994.)

在 Linux 上，现在 os.urandom() 会阻塞直到系统的 urandom 熵池被初始化以提升安全性。其理由参见 PEP 524。

The Linux getrandom() syscall (get random bytes) is now exposed as the new os.getrandom() function. (Contributed by Victor Stinner, part of the PEP 524)

pathlib¶

pathlib now supports path-like objects. (Contributed by Brett Cannon in bpo-27186.)

See the summary of PEP 519 for details.

pdb¶

The Pdb class constructor has a new optional readrc argument to control whether .pdbrc files should be read.

pickle¶

Objects that need __new__ called with keyword arguments can now be pickled using pickle protocols older than protocol version 4. Protocol version 4 already supports this case. (Contributed by Serhiy Storchaka in bpo-24164.)

pickletools¶

pickletools.dis() now outputs the implicit memo index for the MEMOIZE opcode. (Contributed by Serhiy Storchaka in bpo-25382.)

pydoc¶

The pydoc module has learned to respect the MANPAGER environment variable. (Contributed by Matthias Klose in bpo-8637.)

help() and pydoc can now list named tuple fields in the order they were defined rather than alphabetically. (Contributed by Raymond Hettinger in bpo-24879.)

random¶

The new choices() function returns a list of elements of specified size from the given population with optional weights. (Contributed by Raymond Hettinger in bpo-18844.)

re¶

Added support of modifier spans in regular expressions. Examples: '(?i:p)ython' matches 'python' and 'Python', but not 'PYTHON'; '(?i)g(?-i:v)r' matches 'GvR' and 'gvr', but not 'GVR'. (Contributed by Serhiy Storchaka in bpo-433028.)

Match object groups can be accessed by __getitem__, which is equivalent to group(). So mo['name'] is now equivalent to mo.group('name'). (Contributed by Eric Smith in bpo-24454.)

Match objects now support index-like objects as group indices. (Contributed by Jeroen Demeyer and Xiang Zhang in bpo-27177.)

readline¶

Added set_auto_history() to enable or disable automatic addition of input to the history list. (Contributed by Tyler Crompton in bpo-26870.)

rlcompleter¶

Private and special attribute names now are omitted unless the prefix starts with underscores. A space or a colon is added after some completed keywords. (Contributed by Serhiy Storchaka in bpo-25011 and bpo-25209.)

shlex¶

The shlex has much improved shell compatibility through the new punctuation_chars argument to control which characters are treated as punctuation. (Contributed by Vinay Sajip in bpo-1521950.)

site¶

When specifying paths to add to sys.path in a .pth file, you may now specify file paths on top of directories (e.g. zip files). (Contributed by Wolfgang Langner in bpo-26587).

sqlite3¶

sqlite3.Cursor.lastrowid now supports the REPLACE statement. (Contributed by Alex LordThorsen in bpo-16864.)

socket¶

The ioctl() function now supports the SIO_LOOPBACK_FAST_PATH control code. (Contributed by Daniel Stokes in bpo-26536.)

The getsockopt() constants SO_DOMAIN, SO_PROTOCOL, SO_PEERSEC, and SO_PASSSEC are now supported. (Contributed by Christian Heimes in bpo-26907.)

The setsockopt() now supports the setsockopt(level, optname, None, optlen: int) form. (Contributed by Christian Heimes in bpo-27744.)

The socket module now supports the address family AF_ALG to interface with Linux Kernel crypto API. ALG_*, SOL_ALG and sendmsg_afalg() were added. (Contributed by Christian Heimes in bpo-27744 with support from Victor Stinner.)

New Linux constants TCP_USER_TIMEOUT and TCP_CONGESTION were added. (Contributed by Omar Sandoval, bpo-26273).

socketserver¶

Servers based on the socketserver module, including those defined in http.server, xmlrpc.server and wsgiref.simple_server, now support the context manager protocol. (Contributed by Aviv Palivoda in bpo-26404.)

The wfile attribute of StreamRequestHandler classes now implements the io.BufferedIOBase writable interface. In particular, calling write() is now guaranteed to send the data in full. (Contributed by Martin Panter in bpo-26721.)

ssl¶

ssl supports OpenSSL 1.1.0. The minimum recommend version is 1.0.2. (Contributed by Christian Heimes in bpo-26470.)

3DES has been removed from the default cipher suites and ChaCha20 Poly1305 cipher suites have been added. (Contributed by Christian Heimes in bpo-27850 and bpo-27766.)

SSLContext has better default configuration for options and ciphers. (Contributed by Christian Heimes in bpo-28043.)

SSL session can be copied from one client-side connection to another with the new SSLSession class. TLS session resumption can speed up the initial handshake, reduce latency and improve performance (Contributed by Christian Heimes in bpo-19500 based on a draft by Alex Warhawk.)

The new get_ciphers() method can be used to get a list of enabled ciphers in order of cipher priority.

All constants and flags have been converted to IntEnum and IntFlags. (Contributed by Christian Heimes in bpo-28025.)

Server and client-side specific TLS protocols for SSLContext were added. (Contributed by Christian Heimes in bpo-28085.)

statistics¶

A new harmonic_mean() function has been added. (Contributed by Steven D'Aprano in bpo-27181.)

struct¶

struct now supports IEEE 754 half-precision floats via the 'e' format specifier. (Contributed by Eli Stevens, Mark Dickinson in bpo-11734.)

subprocess¶

subprocess.Popen destructor now emits a ResourceWarning warning if the child process is still running. Use the context manager protocol (with proc: ...) or explicitly call the wait() method to read the exit status of the child process. (Contributed by Victor Stinner in bpo-26741.)

The subprocess.Popen constructor and all functions that pass arguments through to it now accept encoding and errors arguments. Specifying either of these will enable text mode for the stdin, stdout and stderr streams. (Contributed by Steve Dower in bpo-6135.)

sys¶

The new getfilesystemencodeerrors() function returns the name of the error mode used to convert between Unicode filenames and bytes filenames. (Contributed by Steve Dower in bpo-27781.)

On Windows the return value of the getwindowsversion() function now includes the platform_version field which contains the accurate major version, minor version and build number of the current operating system, rather than the version that is being emulated for the process (Contributed by Steve Dower in bpo-27932.)

telnetlib¶

Telnet is now a context manager (contributed by Stéphane Wirtel in bpo-25485).

time¶

The struct_time attributes tm_gmtoff and tm_zone are now available on all platforms.

timeit¶

The new Timer.autorange() convenience method has been added to call Timer.timeit() repeatedly so that the total run time is greater or equal to 200 milliseconds. (Contributed by Steven D'Aprano in bpo-6422.)

timeit now warns when there is substantial (4x) variance between best and worst times. (Contributed by Serhiy Storchaka in bpo-23552.)

tkinter¶

Added methods trace_add(), trace_remove() and trace_info() in the tkinter.Variable class. They replace old methods trace_variable(), trace(), trace_vdelete() and trace_vinfo() that use obsolete Tcl commands and might not work in future versions of Tcl. (Contributed by Serhiy Storchaka in bpo-22115).

回溯¶

Both the traceback module and the interpreter's builtin exception display now abbreviate long sequences of repeated lines in tracebacks as shown in the following example:

>>> def f(): f()
...
>>> f()
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "<stdin>", line 1, in f
  File "<stdin>", line 1, in f
  File "<stdin>", line 1, in f
  [Previous line repeated 995 more times]
RecursionError: maximum recursion depth exceeded

（由 Emanuel Barry在 bpo-26823 中贡献。）

tracemalloc¶

The tracemalloc module now supports tracing memory allocations in multiple different address spaces.

The new DomainFilter filter class has been added to filter block traces by their address space (domain).

（由 Victor Stinner 在 bpo-26588 中贡献。）

typing¶

Since the typing module is provisional, all changes introduced in Python 3.6 have also been backported to Python 3.5.x.

The typing module has a much improved support for generic type aliases. For example Dict[str, Tuple[S, T]] is now a valid type annotation. (Contributed by Guido van Rossum in Github #195.)

The typing.ContextManager class has been added for representing contextlib.AbstractContextManager. (Contributed by Brett Cannon in bpo-25609.)

The typing.Collection class has been added for representing collections.abc.Collection. (Contributed by Ivan Levkivskyi in bpo-27598.)

The typing.ClassVar type construct has been added to mark class variables. As introduced in PEP 526, a variable annotation wrapped in ClassVar indicates that a given attribute is intended to be used as a class variable and should not be set on instances of that class. (Contributed by Ivan Levkivskyi in Github #280.)

A new TYPE_CHECKING constant that is assumed to be True by the static type checkers, but is False at runtime. (Contributed by Guido van Rossum in Github #230.)

A new NewType() helper function has been added to create lightweight distinct types for annotations:

from typing import NewType

UserId = NewType('UserId', int)
some_id = UserId(524313)

The static type checker will treat the new type as if it were a subclass of the original type. (Contributed by Ivan Levkivskyi in Github #189.)

unicodedata¶

The unicodedata module now uses data from Unicode 9.0.0. (Contributed by Benjamin Peterson.)

unittest.mock¶

The Mock class has the following improvements:

Two new methods, Mock.assert_called() and Mock.assert_called_once() to check if the mock object was called. (Contributed by Amit Saha in bpo-26323.)
The Mock.reset_mock() method now has two optional keyword only arguments: return_value and side_effect. (Contributed by Kushal Das in bpo-21271.)

urllib.request¶

If a HTTP request has a file or iterable body (other than a bytes object) but no Content-Length header, rather than throwing an error, AbstractHTTPHandler now falls back to use chunked transfer encoding. (Contributed by Demian Brecht and Rolf Krahl in bpo-12319.)

urllib.robotparser¶

RobotFileParser now supports the Crawl-delay and Request-rate extensions. (Contributed by Nikolay Bogoychev in bpo-16099.)

venv¶

venv accepts a new parameter --prompt. This parameter provides an alternative prefix for the virtual environment. (Proposed by Łukasz Balcerzak and ported to 3.6 by Stéphane Wirtel in bpo-22829.)

warnings¶

A new optional source parameter has been added to the warnings.warn_explicit() function: the destroyed object which emitted a ResourceWarning. A source attribute has also been added to warnings.WarningMessage (contributed by Victor Stinner in bpo-26568 and bpo-26567).

When a ResourceWarning warning is logged, the tracemalloc module is now used to try to retrieve the traceback where the destroyed object was allocated.

Example with the script example.py:

import warnings

def func():
    return open(__file__)

f = func()
f = None

Output of the command python3.6 -Wd -X tracemalloc=5 example.py:

example.py:7: ResourceWarning: unclosed file <_io.TextIOWrapper name='example.py' mode='r' encoding='UTF-8'>
  f = None
Object allocated at (most recent call first):
  File "example.py", lineno 4
    return open(__file__)
  File "example.py", lineno 6
    f = func()

The "Object allocated at" traceback is new and is only displayed if tracemalloc is tracing Python memory allocations and if the warnings module was already imported.

winreg¶

Added the 64-bit integer type REG_QWORD. (Contributed by Clement Rouault in bpo-23026.)

winsound¶

Allowed keyword arguments to be passed to Beep, MessageBeep, and PlaySound (bpo-27982).

xmlrpc.client¶

The xmlrpc.client module now supports unmarshalling additional data types used by the Apache XML-RPC implementation for numerics and None. (Contributed by Serhiy Storchaka in bpo-26885.)

zipfile¶

A new ZipInfo.from_file() class method allows making a ZipInfo instance from a filesystem file. A new ZipInfo.is_dir() method can be used to check if the ZipInfo instance represents a directory. (Contributed by Thomas Kluyver in bpo-26039.)

The ZipFile.open() method can now be used to write data into a ZIP file, as well as for extracting data. (Contributed by Thomas Kluyver in bpo-26039.)

zlib¶

The compress() and decompress() functions now accept keyword arguments. (Contributed by Aviv Palivoda in bpo-26243 and Xiang Zhang in bpo-16764 respectively.)

性能优化¶

The Python interpreter now uses a 16-bit wordcode instead of bytecode which made a number of opcode optimizations possible. (Contributed by Demur Rumed with input and reviews from Serhiy Storchaka and Victor Stinner in bpo-26647 and bpo-28050.)
The asyncio.Future class now has an optimized C implementation. (Contributed by Yury Selivanov and INADA Naoki in bpo-26081.)
The asyncio.Task class now has an optimized C implementation. (Contributed by Yury Selivanov in bpo-28544.)
Various implementation improvements in the typing module (such as caching of generic types) allow up to 30 times performance improvements and reduced memory footprint.
The ASCII decoder is now up to 60 times as fast for error handlers surrogateescape, ignore and replace (Contributed by Victor Stinner in bpo-24870).
The ASCII and the Latin1 encoders are now up to 3 times as fast for the error handler surrogateescape (Contributed by Victor Stinner in bpo-25227).
The UTF-8 encoder is now up to 75 times as fast for error handlers ignore, replace, surrogateescape, surrogatepass (Contributed by Victor Stinner in bpo-25267).
The UTF-8 decoder is now up to 15 times as fast for error handlers ignore, replace and surrogateescape (Contributed by Victor Stinner in bpo-25301).
bytes % args is now up to 2 times faster. (Contributed by Victor Stinner in bpo-25349).
bytearray % args is now between 2.5 and 5 times faster. (Contributed by Victor Stinner in bpo-25399).
Optimize bytes.fromhex() and bytearray.fromhex(): they are now between 2x and 3.5x faster. (Contributed by Victor Stinner in bpo-25401).
Optimize bytes.replace(b'', b'.') and bytearray.replace(b'', b'.'): up to 80% faster. (Contributed by Josh Snider in bpo-26574).
Allocator functions of the PyMem_Malloc() domain (PYMEM_DOMAIN_MEM) now use the pymalloc memory allocator instead of malloc() function of the C library. The pymalloc allocator is optimized for objects smaller or equal to 512 bytes with a short lifetime, and use malloc() for larger memory blocks. (Contributed by Victor Stinner in bpo-26249).
pickle.load() and pickle.loads() are now up to 10% faster when deserializing many small objects (Contributed by Victor Stinner in bpo-27056).
Passing keyword arguments to a function has an overhead in comparison with passing positional arguments. Now in extension functions implemented with using Argument Clinic this overhead is significantly decreased. (Contributed by Serhiy Storchaka in bpo-27574).
Optimized glob() and iglob() functions in the glob module; they are now about 3--6 times faster. (Contributed by Serhiy Storchaka in bpo-25596).
Optimized globbing in pathlib by using os.scandir(); it is now about 1.5--4 times faster. (Contributed by Serhiy Storchaka in bpo-26032).
xml.etree.ElementTree parsing, iteration and deepcopy performance has been significantly improved. (Contributed by Serhiy Storchaka in bpo-25638, bpo-25873, and bpo-25869.)
Creation of fractions.Fraction instances from floats and decimals is now 2 to 3 times faster. (Contributed by Serhiy Storchaka in bpo-25971.)

构建和 C API 的改变¶

Python now requires some C99 support in the toolchain to build. Most notably, Python now uses standard integer types and macros in place of custom macros like PY_LONG_LONG. For more information, see PEP 7 and bpo-17884.
Cross-compiling CPython with the Android NDK and the Android API level set to 21 (Android 5.0 Lollipop) or greater runs successfully. While Android is not yet a supported platform, the Python test suite runs on the Android emulator with only about 16 tests failures. See the Android meta-issue bpo-26865.
The --enable-optimizations configure flag has been added. Turning it on will activate expensive optimizations like PGO. (Original patch by Alecsandru Patrascu of Intel in bpo-26359.)
The GIL must now be held when allocator functions of PYMEM_DOMAIN_OBJ (ex: PyObject_Malloc()) and PYMEM_DOMAIN_MEM (ex: PyMem_Malloc()) domains are called.
New Py_FinalizeEx() API which indicates if flushing buffered data failed. (Contributed by Martin Panter in bpo-5319.)
PyArg_ParseTupleAndKeywords() now supports positional-only parameters. Positional-only parameters are defined by empty names. (Contributed by Serhiy Storchaka in bpo-26282).
PyTraceback_Print method now abbreviates long sequences of repeated lines as "[Previous line repeated {count} more times]". (Contributed by Emanuel Barry in bpo-26823.)
The new PyErr_SetImportErrorSubclass() function allows for specifying a subclass of ImportError to raise. (Contributed by Eric Snow in bpo-15767.)
The new PyErr_ResourceWarning() function can be used to generate a ResourceWarning providing the source of the resource allocation. (Contributed by Victor Stinner in bpo-26567.)
The new PyOS_FSPath() function returns the file system representation of a path-like object. (Contributed by Brett Cannon in bpo-27186.)
现在 PyUnicode_FSConverter() 和 PyUnicode_FSDecoder() 函数将接受路径类对象。

其他改进¶

当 --version (简短形式: -V) 提供了两次时，Python 将针对细节信息打印 sys.version。

$ ./python -VV
Python 3.6.0b4+ (3.6:223967b49e49+, Nov 21 2016, 20:55:04)
[GCC 4.2.1 Compatible Apple LLVM 8.0.0 (clang-800.0.42.1)]

弃用¶

新关键字¶

async and await are not recommended to be used as variable, class, function or module names. Introduced by PEP 492 in Python 3.5, they will become proper keywords in Python 3.7. Starting in Python 3.6, the use of async or await as names will generate a DeprecationWarning.

已弃用的 Python 行为¶

Raising the StopIteration exception inside a generator will now generate a DeprecationWarning, and will trigger a RuntimeError in Python 3.7. See PEP 479: Change StopIteration handling inside generators for details.

The __aiter__() method is now expected to return an asynchronous iterator directly instead of returning an awaitable as previously. Doing the former will trigger a DeprecationWarning. Backward compatibility will be removed in Python 3.7. (Contributed by Yury Selivanov in bpo-27243.)

A backslash-character pair that is not a valid escape sequence now generates a DeprecationWarning. Although this will eventually become a SyntaxError, that will not be for several Python releases. (Contributed by Emanuel Barry in bpo-27364.)

When performing a relative import, falling back on __name__ and __path__ from the calling module when __spec__ or __package__ are not defined now raises an ImportWarning. (Contributed by Rose Ames in bpo-25791.)

已弃用的 Python 模块、函数和方法¶

asynchat¶

The asynchat has been deprecated in favor of asyncio. (Contributed by Mariatta in bpo-25002.)

asyncore¶

The asyncore has been deprecated in favor of asyncio. (Contributed by Mariatta in bpo-25002.)

dbm¶

Unlike other dbm implementations, the dbm.dumb module creates databases with the 'rw' mode and allows modifying the database opened with the 'r' mode. This behavior is now deprecated and will be removed in 3.8. (Contributed by Serhiy Storchaka in bpo-21708.)

distutils¶

The undocumented extra_path argument to the distutils.Distribution constructor is now considered deprecated and will raise a warning if set. Support for this parameter will be removed in a future Python release. See bpo-27919 for details.

grp¶

The support of non-integer arguments in getgrgid() has been deprecated. (Contributed by Serhiy Storchaka in bpo-26129.)

importlib¶

The importlib.machinery.SourceFileLoader.load_module() and importlib.machinery.SourcelessFileLoader.load_module() methods are now deprecated. They were the only remaining implementations of importlib.abc.Loader.load_module() in importlib that had not been deprecated in previous versions of Python in favour of importlib.abc.Loader.exec_module().

The importlib.machinery.WindowsRegistryFinder class is now deprecated. As of 3.6.0, it is still added to sys.meta_path by default (on Windows), but this may change in future releases.

os¶

Undocumented support of general bytes-like objects as paths in os functions, compile() and similar functions is now deprecated. (Contributed by Serhiy Storchaka in bpo-25791 and bpo-26754.)

re¶

Support for inline flags (?letters) in the middle of the regular expression has been deprecated and will be removed in a future Python version. Flags at the start of a regular expression are still allowed. (Contributed by Serhiy Storchaka in bpo-22493.)

ssl¶

OpenSSL 0.9.8, 1.0.0 and 1.0.1 are deprecated and no longer supported. In the future the ssl module will require at least OpenSSL 1.0.2 or 1.1.0.

SSL-related arguments like certfile, keyfile and check_hostname in ftplib, http.client, imaplib, poplib, and smtplib have been deprecated in favor of context. (Contributed by Christian Heimes in bpo-28022.)

A couple of protocols and functions of the ssl module are now deprecated. Some features will no longer be available in future versions of OpenSSL. Other features are deprecated in favor of a different API. (Contributed by Christian Heimes in bpo-28022 and bpo-26470.)

tkinter¶

The tkinter.tix module is now deprecated. tkinter users should use tkinter.ttk instead.

venv¶

The pyvenv script has been deprecated in favour of python3 -m venv. This prevents confusion as to what Python interpreter pyvenv is connected to and thus what Python interpreter will be used by the virtual environment. (Contributed by Brett Cannon in bpo-25154.)

已弃用的 C API 函数和类型¶

Undocumented functions PyUnicode_AsEncodedObject(), PyUnicode_AsDecodedObject(), PyUnicode_AsEncodedUnicode() and PyUnicode_AsDecodedUnicode() are deprecated now. Use the generic codec based API instead.

弃用的构建选项¶

The --with-system-ffi configure flag is now on by default on non-macOS UNIX platforms. It may be disabled by using --without-system-ffi, but using the flag is deprecated and will not be accepted in Python 3.7. macOS is unaffected by this change. Note that many OS distributors already use the --with-system-ffi flag when building their system Python.

移除¶

API 与特性的移除¶

Unknown escapes consisting of '\' and an ASCII letter in regular expressions will now cause an error. In replacement templates for re.sub() they are still allowed, but deprecated. The re.LOCALE flag can now only be used with binary patterns.
inspect.getmoduleinfo() was removed (was deprecated since CPython 3.3). inspect.getmodulename() should be used for obtaining the module name for a given path. (Contributed by Yury Selivanov in bpo-13248.)
traceback.Ignore class and traceback.usage, traceback.modname, traceback.fullmodname, traceback.find_lines_from_code, traceback.find_lines, traceback.find_strings, traceback.find_executable_lines methods were removed from the traceback module. They were undocumented methods deprecated since Python 3.2 and equivalent functionality is available from private methods.
The tk_menuBar() and tk_bindForTraversal() dummy methods in tkinter widget classes were removed (corresponding Tk commands were obsolete since Tk 4.0).
The open() method of the zipfile.ZipFile class no longer supports the 'U' mode (was deprecated since Python 3.4). Use io.TextIOWrapper for reading compressed text files in universal newlines mode.
The undocumented IN, CDROM, DLFCN, TYPES, CDIO, and STROPTS modules have been removed. They had been available in the platform specific Lib/plat-*/ directories, but were chronically out of date, inconsistently available across platforms, and unmaintained. The script that created these modules is still available in the source distribution at Tools/scripts/h2py.py.
The deprecated asynchat.fifo class has been removed.

移植到Python 3.6¶

本节列出了先前描述的更改以及可能需要更改代码的其他错误修正.

'python' 命令行为的变化¶

The output of a special Python build with defined COUNT_ALLOCS, SHOW_ALLOC_COUNT or SHOW_TRACK_COUNT macros is now off by default. It can be re-enabled using the -X showalloccount option. It now outputs to stderr instead of stdout. (Contributed by Serhiy Storchaka in bpo-23034.)

Python API 的变化¶

open() will no longer allow combining the 'U' mode flag with '+'. (Contributed by Jeff Balogh and John O'Connor in bpo-2091.)
sqlite3 no longer implicitly commits an open transaction before DDL statements.
On Linux, os.urandom() now blocks until the system urandom entropy pool is initialized to increase the security.
When importlib.abc.Loader.exec_module() is defined, importlib.abc.Loader.create_module() must also be defined.
PyErr_SetImportError() now sets TypeError when its msg argument is not set. Previously only NULL was returned.
The format of the co_lnotab attribute of code objects changed to support a negative line number delta. By default, Python does not emit bytecode with a negative line number delta. Functions using frame.f_lineno, PyFrame_GetLineNumber() or PyCode_Addr2Line() are not affected. Functions directly decoding co_lnotab should be updated to use a signed 8-bit integer type for the line number delta, but this is only required to support applications using a negative line number delta. See Objects/lnotab_notes.txt for the co_lnotab format and how to decode it, and see the PEP 511 for the rationale.
The functions in the compileall module now return booleans instead of 1 or 0 to represent success or failure, respectively. Thanks to booleans being a subclass of integers, this should only be an issue if you were doing identity checks for 1 or 0. See bpo-25768.
Reading the port attribute of urllib.parse.urlsplit() and urlparse() results now raises ValueError for out-of-range values, rather than returning None. See bpo-20059.
The imp module now raises a DeprecationWarning instead of PendingDeprecationWarning.
The following modules have had missing APIs added to their __all__ attributes to match the documented APIs: calendar, cgi, csv, ElementTree, enum, fileinput, ftplib, logging, mailbox, mimetypes, optparse, plistlib, smtpd, subprocess, tarfile, threading and wave. This means they will export new symbols when import * is used. (Contributed by Joel Taddei and Jacek Kołodziej in bpo-23883.)
When performing a relative import, if __package__ does not compare equal to __spec__.parent then ImportWarning is raised. (Contributed by Brett Cannon in bpo-25791.)
When a relative import is performed and no parent package is known, then ImportError will be raised. Previously, SystemError could be raised. (Contributed by Brett Cannon in bpo-18018.)
Servers based on the socketserver module, including those defined in http.server, xmlrpc.server and wsgiref.simple_server, now only catch exceptions derived from Exception. Therefore if a request handler raises an exception like SystemExit or KeyboardInterrupt, handle_error() is no longer called, and the exception will stop a single-threaded server. (Contributed by Martin Panter in bpo-23430.)
如果用户没有权限， spwd.getspnam() 现在会抛出 PermissionError 而非之前的 KeyError 。
The socket.socket.close() method now raises an exception if an error (e.g. EBADF) was reported by the underlying system call. (Contributed by Martin Panter in bpo-26685.)
The decode_data argument for the smtpd.SMTPChannel and smtpd.SMTPServer constructors is now False by default. This means that the argument passed to process_message() is now a bytes object by default, and process_message() will be passed keyword arguments. Code that has already been updated in accordance with the deprecation warning generated by 3.5 will not be affected.
All optional arguments of the dump(), dumps(), load() and loads() functions and JSONEncoder and JSONDecoder class constructors in the json module are now keyword-only. (Contributed by Serhiy Storchaka in bpo-18726.)
type 的子类如果未重载 type.__new__，将不再能使用一个参数的形式来获取对象的类型。
As part of PEP 487, the handling of keyword arguments passed to type (other than the metaclass hint, metaclass) is now consistently delegated to object.__init_subclass__(). This means that type.__new__() and type.__init__() both now accept arbitrary keyword arguments, but object.__init_subclass__() (which is called from type.__new__()) will reject them by default. Custom metaclasses accepting additional keyword arguments will need to adjust their calls to type.__new__() (whether direct or via super) accordingly.
In distutils.command.sdist.sdist, the default_format attribute has been removed and is no longer honored. Instead, the gzipped tarfile format is the default on all platforms and no platform-specific selection is made. In environments where distributions are built on Windows and zip distributions are required, configure the project with a setup.cfg file containing the following:
```
[sdist]
formats=zip
```
This behavior has also been backported to earlier Python versions by Setuptools 26.0.0.
In the urllib.request module and the http.client.HTTPConnection.request() method, if no Content-Length header field has been specified and the request body is a file object, it is now sent with HTTP 1.1 chunked encoding. If a file object has to be sent to a HTTP 1.0 server, the Content-Length value now has to be specified by the caller. (Contributed by Demian Brecht and Rolf Krahl with tweaks from Martin Panter in bpo-12319.)
The DictReader now returns rows of type OrderedDict. (Contributed by Steve Holden in bpo-27842.)
The crypt.METHOD_CRYPT will no longer be added to crypt.methods if unsupported by the platform. (Contributed by Victor Stinner in bpo-25287.)
namedtuple() 的 verbose 和 rename 参数现在是仅限关键字参数。（由 Raymond Hettinger 在 bpo-25628 中贡献。）
On Linux, ctypes.util.find_library() now looks in LD_LIBRARY_PATH for shared libraries. (Contributed by Vinay Sajip in bpo-9998.)
The imaplib.IMAP4 class now handles flags containing the ']' character in messages sent from the server to improve real-world compatibility. (Contributed by Lita Cho in bpo-21815.)
The mmap.write() function now returns the number of bytes written like other write methods. (Contributed by Jakub Stasiak in bpo-26335.)
The pkgutil.iter_modules() and pkgutil.walk_packages() functions now return ModuleInfo named tuples. (Contributed by Ramchandra Apte in bpo-17211.)
re.sub() now raises an error for invalid numerical group references in replacement templates even if the pattern is not found in the string. The error message for invalid group references now includes the group index and the position of the reference. (Contributed by SilentGhost, Serhiy Storchaka in bpo-25953.)
zipfile.ZipFile will now raise NotImplementedError for unrecognized compression values. Previously a plain RuntimeError was raised. Additionally, calling ZipFile methods on a closed ZipFile or calling the write() method on a ZipFile created with mode 'r' will raise a ValueError. Previously, a RuntimeError was raised in those scenarios.
when custom metaclasses are combined with zero-argument super() or direct references from methods to the implicit __class__ closure variable, the implicit __classcell__ namespace entry must now be passed up to type.__new__ for initialisation. Failing to do so will result in a DeprecationWarning in Python 3.6 and a RuntimeError in Python 3.8.
With the introduction of ModuleNotFoundError, import system consumers may start expecting import system replacements to raise that more specific exception when appropriate, rather than the less-specific ImportError. To provide future compatibility with such consumers, implementors of alternative import systems that completely replace __import__() will need to update their implementations to raise the new subclass when a module can't be found at all. Implementors of compliant plugins to the default import system shouldn't need to make any changes, as the default import system will raise the new subclass when appropriate.

C API 的变化¶

The PyMem_Malloc() allocator family now uses the pymalloc allocator rather than the system malloc(). Applications calling PyMem_Malloc() without holding the GIL can now crash. Set the PYTHONMALLOC environment variable to debug to validate the usage of memory allocators in your application. See bpo-26249.
Py_Exit() (and the main interpreter) now override the exit status with 120 if flushing buffered data failed. See bpo-5319.

CPython 字节码的改变¶

There have been several major changes to the bytecode in Python 3.6.

The Python interpreter now uses a 16-bit wordcode instead of bytecode. (Contributed by Demur Rumed with input and reviews from Serhiy Storchaka and Victor Stinner in bpo-26647 and bpo-28050.)
The new FORMAT_VALUE and BUILD_STRING opcodes as part of the formatted string literal implementation. (Contributed by Eric Smith in bpo-25483 and Serhiy Storchaka in bpo-27078.)
The new BUILD_CONST_KEY_MAP opcode to optimize the creation of dictionaries with constant keys. (Contributed by Serhiy Storchaka in bpo-27140.)
The function call opcodes have been heavily reworked for better performance and simpler implementation. The MAKE_FUNCTION, CALL_FUNCTION, CALL_FUNCTION_KW and BUILD_MAP_UNPACK_WITH_CALL opcodes have been modified, the new CALL_FUNCTION_EX and BUILD_TUPLE_UNPACK_WITH_CALL have been added, and CALL_FUNCTION_VAR, CALL_FUNCTION_VAR_KW and MAKE_CLOSURE opcodes have been removed. (Contributed by Demur Rumed in bpo-27095, and Serhiy Storchaka in bpo-27213, bpo-28257.)
The new SETUP_ANNOTATIONS and STORE_ANNOTATION opcodes have been added to support the new variable annotation syntax. (Contributed by Ivan Levkivskyi in bpo-27985.)

Python 3.6.2 中的重要变化¶

New `make regen-all` build target¶

To simplify cross-compilation, and to ensure that CPython can reliably be compiled without requiring an existing version of Python to already be available, the autotools-based build system no longer attempts to implicitly recompile generated files based on file modification times.

Instead, a new make regen-all command has been added to force regeneration of these files when desired (e.g. after an initial version of Python has already been built based on the pregenerated versions).

More selective regeneration targets are also defined - see Makefile.pre.in for details.

（由 Victor Stinner 在 bpo-23404 中贡献。）

在 3.6.2 版本加入.

Removal of `make touch` build target¶

The make touch build target previously used to request implicit regeneration of generated files by updating their modification times has been removed.

它已被新的 make regen-all 目标所替代。

（由 Victor Stinner 在 bpo-23404 中贡献。）

在 3.6.2 版本发生变更.

Python 3.6.4 中的重要变化¶

曾经作为 API 一部分的 PyExc_RecursionErrorInst 单例已被移除，因为它的成员永远不会被清理，可能在解释器的最终化过程中导致段错误。（由 Xavier de Gaye 在 bpo-22898 和 bpo-30697 中贡献。）

Python 3.6.5 中的重要变化¶

在某些情况下 locale.localeconv() 函数现在会临时将 LC_CTYPE 语言区域设为 LC_NUMERIC 语言区域。（由 Victor Stinner 在 bpo-31900 中贡献。）

Python 3.6.7 中的重要变化¶

在 3.6.7 中当提供不带末尾换行符的输入时 tokenize 模块现在会隐式地发出 NEWLINE 形符。此行为现在已与 C 分词器的内部行为相匹配。（由 Ammar Askar 在 bpo-33899 中贡献。）

Python 3.6.10 中的重要变化¶

出于重要的安全性考量，asyncio.loop.create_datagram_endpoint() 的 reuse_address 形参不再被支持。这是由 UDP 中的套接字选项 SO_REUSEADDR 的行为导致的。更多细节请参阅 loop.create_datagram_endpoint() 的文档。（由 Kyle Stanley, Antoine Pitrou 和 Yury Selivanov 在 bpo-37228 中贡献。。）

Python 3.6.13 中的重要变化¶

早先的 Python 版本允许使用 ; 和 & 作为 urllib.parse.parse_qs() 和 urllib.parse.parse_qsl() 中 query 形参的分隔键。出于安全考虑，也为了遵循更新的 W3C 推荐设置，这已被改为只允许单个分隔键，默认为 &。这一改变还会影响 cgi.parse() 和 cgi.parse_multipart() 因为它们在内部使用了受影响的函数。要了解更多细节，请查看它们各自的文档。（由 Adam Goldschmidt, Senthil Kumaran 和 Ken Jin 在 bpo-42967 中贡献。）