sys —- 系统相关的参数和函数

sys —- 系统相关的参数和函数

该模块提供了一些变量和函数。这些变量可能被解释器使用，也可能由解释器提供。这些函数会影响解释器。本模块总是可用的。

sys.abiflags
在POSIX系统上，以标准的 configure 脚本构建的 Python 中，这个变量会包含 PEP 3149 中定义的ABI标签。

3.2 新版功能.

sys.argv
一个列表，其中包含了被传递给 Python 脚本的命令行参数。 argv[0] 为脚本的名称（是否是完整的路径名取决于操作系统）。如果是通过 Python 解释器的命令行参数 -c 来执行的， argv[0] 会被设置成字符串 '-c' 。如果没有脚本名被传递给 Python 解释器， argv[0] 为空字符串。

为了遍历标准输入，或者通过命令行传递的文件列表，参照 fileinput 模块

注解

On Unix, command line arguments are passed by bytes from OS. Python decodesthem with filesystem encoding and "surrogateescape" error handler.When you need original bytes, you can get it by[os.fsencode(arg) for arg in sys.argv].

sys.base_exec_prefix
在 site.py 运行之前， Python 启动的时候被设置为跟 exec_prefix 同样的值。如果不是运行在虚拟环境中，两个值会保持相同；如果 site.py 发现处于一个虚拟环境中， prefix 和 exec_prefix 将会指向虚拟环境。然而 base_prefix 和 base_exec_prefix 将仍然会指向基础的 Python 环境（用来创建虚拟环境的 Python 环境）

3.3 新版功能.

sys.base_prefix
在 site.py 运行之前， Python 启动的时候被设置为跟 prefix 同样的值。如果不是运行在虚拟环境中, 两个值会保持相同；如果 site.py 发现处于一个虚拟环境中， prefix 和 exec_prefix 将会指向虚拟环境。然而 base_prefix 和 base_exec_prefix 将仍然会指向基础的 Python 环境（用来创建虚拟环境的 Python 环境）

3.3 新版功能.

sys.byteorder
本地字节顺序的指示符。在大端序（最高有效位优先）操作系统上值为 'big' ，在小端序（最低有效位优先）操作系统上为 'little' 。

sys.builtin_module_names
一个元素为字符串的元组。包含了所有的被编译进 Python 解释器的模块。（这个信息无法通过其他的办法获取， modules.keys() 只包括被导入过的模块。）

sys.calltracing(_func, args)
在启用跟踪时调用 func(*args) 来保存跟踪状态，然后恢复跟踪状态。这将从检查点的调试器调用，以便递归地调试其他的一些代码。

sys.copyright
一个字符串，包含了 Python 解释器有关的版权信息

sys._clear_type_cache()
清除内部的类型缓存。类型缓存是为了加速查找方法和属性的。在调试引用泄漏的时候调用这个函数只会清除不必要的引用。

这个函数应该只在内部为了一些特定的目的使用。

sys._current_frames()
返回一个字典，将每个线程的标识符映射到调用函数时该线程中当前活动的最顶层堆栈帧。注意 traceback 模块中的函数可以在给定帧的情况下构建调用堆栈。

This is most useful for debugging deadlock: this function does not require thedeadlocked threads' cooperation, and such threads' call stacks are frozen for aslong as they remain deadlocked. The frame returned for a non-deadlocked threadmay bear no relationship to that thread's current activity by the time callingcode examines the frame.

这个函数应该只在内部为了一些特定的目的使用。

sys.breakpointhook()
This hook function is called by built-in breakpoint(). By default,it drops you into the pdb debugger, but it can be set to any otherfunction so that you can choose which debugger gets used.

The signature of this function is dependent on what it calls. For example,the default binding (e.g. pdb.set_trace()) expects no arguments, butyou might bind it to a function that expects additional arguments(positional and/or keyword). The built-in breakpoint() function passesits args and *kws straight through. Whateverbreakpointhooks() returns is returned from breakpoint().

The default implementation first consults the environment variablePYTHONBREAKPOINT. If that is set to "0" then this functionreturns immediately; i.e. it is a no-op. If the environment variable isnot set, or is set to the empty string, pdb.set_trace() is called.Otherwise this variable should name a function to run, using Python'sdotted-import nomenclature, e.g. package.subpackage.module.function.In this case, package.subpackage.module would be imported and theresulting module must have a callable named function(). This is run,passing in args and *kws, and whatever function() returns,sys.breakpointhook() returns to the built-in breakpoint()function.

Note that if anything goes wrong while importing the callable named byPYTHONBREAKPOINT, a RuntimeWarning is reported and thebreakpoint is ignored.

Also note that if sys.breakpointhook() is overridden programmatically,PYTHONBREAKPOINT is not consulted.

3.7 新版功能.

sys._debugmallocstats()
Print low-level information to stderr about the state of CPython's memoryallocator.

If Python is configured —with-pydebug, it also performs some expensiveinternal consistency checks.

3.3 新版功能.

CPython implementation detail: This function is specific to CPython. The exact output format is notdefined here, and may change.

sys.dllhandle
Integer specifying the handle of the Python DLL.

可用性: Windows。

sys.displayhook(value)
If value is not None, this function prints repr(value) tosys.stdout, and saves value in builtins._. If repr(value) isnot encodable to sys.stdout.encoding with sys.stdout.errors errorhandler (which is probably 'strict'), encode it tosys.stdout.encoding with 'backslashreplace' error handler.

sys.displayhook is called on the result of evaluating an expressionentered in an interactive Python session. The display of these values can becustomized by assigning another one-argument function to sys.displayhook.

伪代码:

def displayhook(value):
    if value is None:
        return
    # Set '_' to None to avoid recursion
    builtins._ = None
    text = repr(value)
    try:
        sys.stdout.write(text)
    except UnicodeEncodeError:
        bytes = text.encode(sys.stdout.encoding, 'backslashreplace')
        if hasattr(sys.stdout, 'buffer'):
            sys.stdout.buffer.write(bytes)
        else:
            text = bytes.decode(sys.stdout.encoding, 'strict')
            sys.stdout.write(text)
    sys.stdout.write("\n")
    builtins._ = value

在 3.2 版更改: Use 'backslashreplace' error handler on UnicodeEncodeError.

sys.dont_write_bytecode
If this is true, Python won't try to write .pyc files on theimport of source modules. This value is initially set to True orFalse depending on the -B command line option and thePYTHONDONTWRITEBYTECODE environment variable, but you can set ityourself to control bytecode file generation.

sys.excepthook(type, value, traceback)
This function prints out a given traceback and exception to sys.stderr.

When an exception is raised and uncaught, the interpreter callssys.excepthook with three arguments, the exception class, exceptioninstance, and a traceback object. In an interactive session this happens justbefore control is returned to the prompt; in a Python program this happens justbefore the program exits. The handling of such top-level exceptions can becustomized by assigning another three-argument function to sys.excepthook.

sys.breakpointhook
sys.displayhook
sys.excepthook
These objects contain the original values of breakpointhook,displayhook, and excepthook at the start of the program. They aresaved so that breakpointhook, displayhook and excepthook can berestored in case they happen to get replaced with broken or alternativeobjects.

3.7 新版功能: breakpointhook

sys.exc_info()
This function returns a tuple of three values that give information about theexception that is currently being handled. The information returned is specificboth to the current thread and to the current stack frame. If the current stackframe is not handling an exception, the information is taken from the callingstack frame, or its caller, and so on until a stack frame is found that ishandling an exception. Here, "handling an exception" is defined as "executingan except clause." For any stack frame, only information about the exceptionbeing currently handled is accessible.

If no exception is being handled anywhere on the stack, a tuple containingthree None values is returned. Otherwise, the values returned are(type, value, traceback). Their meaning is: type gets the type of theexception being handled (a subclass of BaseException); value getsthe exception instance (an instance of the exception type); traceback getsa traceback object (see the Reference Manual) which encapsulates the callstack at the point where the exception originally occurred.

sys.exec_prefix
A string giving the site-specific directory prefix where the platform-dependentPython files are installed; by default, this is also '/usr/local'. This canbe set at build time with the —exec-prefix argument to theconfigure script. Specifically, all configuration files (e.g. thepyconfig.h header file) are installed in the directoryexec_prefix/lib/pythonX.Y/config, and shared library modules areinstalled in exec_prefix/lib/pythonX.Y/lib-dynload, where _X.Y_is the version number of Python, for example 3.2.

注解

If a virtual environment is in effect, thisvalue will be changed in site.py to point to the virtual environment.The value for the Python installation will still be available, viabase_exec_prefix.

sys.executable
A string giving the absolute path of the executable binary for the Pythoninterpreter, on systems where this makes sense. If Python is unable to retrievethe real path to its executable, sys.executable will be an empty stringor None.

sys.exit([arg])
Exit from Python. This is implemented by raising the SystemExitexception, so cleanup actions specified by finally clauses of trystatements are honored, and it is possible to intercept the exit attempt atan outer level.

The optional argument arg can be an integer giving the exit status(defaulting to zero), or another type of object. If it is an integer, zerois considered "successful termination" and any nonzero value is considered"abnormal termination" by shells and the like. Most systems require it to bein the range 0—127, and produce undefined results otherwise. Some systemshave a convention for assigning specific meanings to specific exit codes, butthese are generally underdeveloped; Unix programs generally use 2 for commandline syntax errors and 1 for all other kind of errors. If another type ofobject is passed, None is equivalent to passing zero, and any otherobject is printed to stderr and results in an exit code of 1. Inparticular, sys.exit("some error message") is a quick way to exit aprogram when an error occurs.

Since exit() ultimately "only" raises an exception, it will only exitthe process when called from the main thread, and the exception is notintercepted.

在 3.6 版更改: If an error occurs in the cleanup after the Python interpreterhas caught SystemExit (such as an error flushing buffered datain the standard streams), the exit status is changed to 120.

sys.flags
The named tupleflags exposes the status of command lineflags. The attributes are read only.

attribute — 属性

flag

debug

interactive

isolated

optimize

-O 或 -OO

dont_write_bytecode

no_user_site

no_site

ignore_environment

verbose

bytes_warning

quiet

hash_randomization

dev_mode

-X dev

utf8_mode

-X utf8

在 3.2 版更改: Added quiet attribute for the new -q flag.

3.2.3 新版功能: The hash_randomization attribute.

在 3.3 版更改: Removed obsolete division_warning attribute.

在 3.4 版更改: Added isolated attribute for -I isolated flag.

在 3.7 版更改: Added dev_mode attribute for the new -X dev flagand utf8_mode attribute for the new -X utf8 flag.

sys.float_info
A named tuple holding information about the float type. Itcontains low level information about the precision and internalrepresentation. The values correspond to the various floating-pointconstants defined in the standard header file float.h for the 'C'programming language; see section 5.2.4.2.2 of the 1999 ISO/IEC C standard[C99], 'Characteristics of floating types', for details.

attribute — 属性

float.h macro

explanation

epsilon

DBL_EPSILON

difference between 1 and the least value greaterthan 1 that is representable as a float

dig

DBL_DIG

maximum number of decimal digits that can befaithfully represented in a float; see below

mant_dig

DBL_MANT_DIG

float precision: the number of base-radixdigits in the significand of a float

max

DBL_MAX

maximum representable finite float

max_exp

DBL_MAX_EXP

maximum integer e such that radix**(e-1) isa representable finite float

max_10_exp

DBL_MAX_10_EXP

maximum integer e such that 10**e is in therange of representable finite floats

min

DBL_MIN

minimum positive normalized float

min_exp

DBL_MIN_EXP

minimum integer e such that radix**(e-1) isa normalized float

min_10_exp

DBL_MIN_10_EXP

minimum integer e such that 10**e is anormalized float

radix

FLT_RADIX

radix of exponent representation

rounds

FLT_ROUNDS

integer constant representing the rounding modeused for arithmetic operations. This reflectsthe value of the system FLT_ROUNDS macro atinterpreter startup time. See section 5.2.4.2.2of the C99 standard for an explanation of thepossible values and their meanings.

The attribute sys.float_info.dig needs further explanation. Ifs is any string representing a decimal number with at mostsys.float_info.dig significant digits, then converting s to afloat and back again will recover a string representing the same decimalvalue:

>>> import sys
>>> sys.float_info.dig
15
>>> s = '3.14159265358979'    # decimal string with 15 significant digits
>>> format(float(s), '.15g')  # convert to float and back -> same value
'3.14159265358979'

But for strings with more than sys.float_info.dig significant digits,this isn't always true:

>>> s = '9876543211234567'    # 16 significant digits is too many!
>>> format(float(s), '.16g')  # conversion changes value
'9876543211234568'

sys.float_repr_style
A string indicating how the repr() function behaves forfloats. If the string has value 'short' then for a finitefloat x, repr(x) aims to produce a short string with theproperty that float(repr(x)) == x. This is the usual behaviourin Python 3.1 and later. Otherwise, float_repr_style has value'legacy' and repr(x) behaves in the same way as it did inversions of Python prior to 3.1.

3.1 新版功能.

sys.getallocatedblocks()
Return the number of memory blocks currently allocated by the interpreter,regardless of their size. This function is mainly useful for trackingand debugging memory leaks. Because of the interpreter's internalcaches, the result can vary from call to call; you may have to call_clear_type_cache() and gc.collect() to get morepredictable results.

If a Python build or implementation cannot reasonably compute thisinformation, getallocatedblocks() is allowed to return 0 instead.

3.4 新版功能.

sys.getandroidapilevel()
Return the build time API version of Android as an integer.

Availability: Android.

3.7 新版功能.

sys.getcheckinterval()
Return the interpreter's "check interval"; see setcheckinterval().

3.2 版后已移除: Use getswitchinterval() instead.

sys.getdefaultencoding()
Return the name of the current default string encoding used by the Unicodeimplementation.

sys.getdlopenflags()
Return the current value of the flags that are used fordlopen() calls. Symbolic names for the flag values can befound in the os module (RTLD_xxx constants, e.g.os.RTLD_LAZY).

可用性: Unix。

sys.getfilesystemencoding()
Return the name of the encoding used to convert between Unicodefilenames and bytes filenames. For best compatibility, str should beused for filenames in all cases, although representing filenames as bytesis also supported. Functions accepting or returning filenames should supporteither str or bytes and internally convert to the system's preferredrepresentation.

This encoding is always ASCII-compatible.

os.fsencode() and os.fsdecode() should be used to ensure thatthe correct encoding and errors mode are used.

In the UTF-8 mode, the encoding is utf-8 on any platform.
On Mac OS X, the encoding is 'utf-8'.
On Unix, the encoding is the locale encoding.
On Windows, the encoding may be 'utf-8' or 'mbcs', dependingon user configuration.

在 3.2 版更改: getfilesystemencoding() result cannot be None anymore.

在 3.6 版更改: Windows is no longer guaranteed to return 'mbcs'. See PEP 529and _enablelegacywindowsfsencoding() for more information.

在 3.7 版更改: Return 'utf-8' in the UTF-8 mode.

sys.getfilesystemencodeerrors()
Return the name of the error mode used to convert between Unicode filenamesand bytes filenames. The encoding name is returned fromgetfilesystemencoding().

os.fsencode() and os.fsdecode() should be used to ensure thatthe correct encoding and errors mode are used.

3.6 新版功能.

sys.getrefcount(object)
Return the reference count of the object. The count returned is generally onehigher than you might expect, because it includes the (temporary) reference asan argument to getrefcount().

sys.getrecursionlimit()
Return the current value of the recursion limit, the maximum depth of the Pythoninterpreter stack. This limit prevents infinite recursion from causing anoverflow of the C stack and crashing Python. It can be set bysetrecursionlimit().

sys.getsizeof(object[, default])
Return the size of an object in bytes. The object can be any type ofobject. All built-in objects will return correct results, but thisdoes not have to hold true for third-party extensions as it is implementationspecific.

Only the memory consumption directly attributed to the object isaccounted for, not the memory consumption of objects it refers to.

If given, default will be returned if the object does not provide means toretrieve the size. Otherwise a TypeError will be raised.

getsizeof() calls the object's sizeof method and adds anadditional garbage collector overhead if the object is managed by the garbagecollector.

See recursive sizeof recipefor an example of using getsizeof() recursively to find the size ofcontainers and all their contents.

sys.getswitchinterval()
Return the interpreter's "thread switch interval"; seesetswitchinterval().

3.2 新版功能.

sys.getframe([_depth])
Return a frame object from the call stack. If optional integer depth isgiven, return the frame object that many calls below the top of the stack. Ifthat is deeper than the call stack, ValueError is raised. The defaultfor depth is zero, returning the frame at the top of the call stack.

CPython implementation detail: This function should be used for internal and specialized purposes only.It is not guaranteed to exist in all implementations of Python.

sys.getprofile()
Get the profiler function as set by setprofile().

sys.gettrace()
Get the trace function as set by settrace().

CPython implementation detail: The gettrace() function is intended only for implementing debuggers,profilers, coverage tools and the like. Its behavior is part of theimplementation platform, rather than part of the language definition, andthus may not be available in all Python implementations.

sys.getwindowsversion()
Return a named tuple describing the Windows versioncurrently running. The named elements are major, minor,build, platform, service_pack, service_pack_minor,service_pack_major, suite_mask, product_type andplatform_version. service_pack contains a string,platform_version a 3-tuple and all other values areintegers. The components can also be accessed by name, sosys.getwindowsversion()[0] is equivalent tosys.getwindowsversion().major. For compatibility with priorversions, only the first 5 elements are retrievable by indexing.

platform will be 2 (VER_PLATFORM_WIN32_NT).

product_type may be one of the following values:

常数

含义

1 (VER_NT_WORKSTATION)

The system is a workstation.

2 (VER_NT_DOMAIN_CONTROLLER)

The system is a domaincontroller.

3 (VER_NT_SERVER)

The system is a server, but nota domain controller.

This function wraps the Win32 GetVersionEx() function; see theMicrosoft documentation on OSVERSIONINFOEX() for more informationabout these fields.

platform_version returns the accurate major version, minor version andbuild number of the current operating system, rather than the version thatis being emulated for the process. It is intended for use in logging ratherthan for feature detection.

可用性: Windows。

在 3.2 版更改: Changed to a named tuple and added service_pack_minor,service_pack_major, suite_mask, and product_type.

在 3.6 版更改: Added platform_version

sys.get_asyncgen_hooks()
Returns an asyncgen_hooks object, which is similar to anamedtuple of the form (firstiter, finalizer),where firstiter and finalizer are expected to be either None orfunctions which take an asynchronous generator iterator as anargument, and are used to schedule finalization of an asynchronousgenerator by an event loop.

3.6 新版功能: See PEP 525 for more details.

注解

This function has been added on a provisional basis (see PEP 411for details.)

sys.get_coroutine_origin_tracking_depth()
Get the current coroutine origin tracking depth, as set byset_coroutine_origin_tracking_depth().

3.7 新版功能.

注解

This function has been added on a provisional basis (see PEP 411for details.) Use it only for debugging purposes.

sys.get_coroutine_wrapper()
Returns None, or a wrapper set by set_coroutine_wrapper().

3.5 新版功能: See PEP 492 for more details.

注解

This function has been added on a provisional basis (see PEP 411for details.) Use it only for debugging purposes.

3.7 版后已移除: The coroutine wrapper functionality has been deprecated, andwill be removed in 3.8. See bpo-32591 for details.

sys.hash_info
A named tuple giving parameters of the numeric hashimplementation. For more details about hashing of numeric types, see数字类型的哈希运算.

attribute — 属性

explanation

width

width in bits used for hash values

modulus

prime modulus P used for numeric hash scheme

inf

hash value returned for a positive infinity

nan

hash value returned for a nan

imag

multiplier used for the imaginary part of acomplex number

algorithm

name of the algorithm for hashing of str, bytes,and memoryview

hash_bits

internal output size of the hash algorithm

seed_bits

size of the seed key of the hash algorithm

3.2 新版功能.

在 3.4 版更改: Added algorithm, hash_bits and seed_bits

sys.hexversion
The version number encoded as a single integer. This is guaranteed to increasewith each version, including proper support for non-production releases. Forexample, to test that the Python interpreter is at least version 1.5.2, use:

if sys.hexversion >= 0x010502F0:
    # use some advanced feature
    ...
else:
    # use an alternative implementation or warn the user
    ...

This is called hexversion since it only really looks meaningful when viewedas the result of passing it to the built-in hex() function. Thenamed tuple sys.version_info may be used for a morehuman-friendly encoding of the same information.

More details of hexversion can be found at API 和 ABI 版本管理.

sys.implementation
An object containing information about the implementation of thecurrently running Python interpreter. The following attributes arerequired to exist in all Python implementations.

name is the implementation's identifier, e.g. 'cpython'. The actualstring is defined by the Python implementation, but it is guaranteed to belower case.

version is a named tuple, in the same format assys.version_info. It represents the version of the Pythonimplementation. This has a distinct meaning from the specificversion of the Python language to which the currently runninginterpreter conforms, which sys.version_info represents. Forexample, for PyPy 1.8 sys.implementation.version might besys.version_info(1, 8, 0, 'final', 0), whereas sys.version_infowould be sys.version_info(2, 7, 2, 'final', 0). For CPython theyare the same value, since it is the reference implementation.

hexversion is the implementation version in hexadecimal format, likesys.hexversion.

cache_tag is the tag used by the import machinery in the filenames ofcached modules. By convention, it would be a composite of theimplementation's name and version, like 'cpython-33'. However, aPython implementation may use some other value if appropriate. Ifcache_tag is set to None, it indicates that module caching shouldbe disabled.

sys.implementation may contain additional attributes specific tothe Python implementation. These non-standard attributes must start withan underscore, and are not described here. Regardless of its contents,sys.implementation will not change during a run of the interpreter,nor between implementation versions. (It may change between Pythonlanguage versions, however.) See PEP 421 for more information.

3.3 新版功能.

注解

The addition of new required attributes must go through the normal PEPprocess. See PEP 421 for more information.

sys.int_info
A named tuple that holds information about Python's internalrepresentation of integers. The attributes are read only.

属性

解释

bits_per_digit

number of bits held in each digit. Pythonintegers are stored internally in base2**int_info.bits_per_digit

sizeof_digit

size in bytes of the C type used torepresent a digit

3.1 新版功能.

sys.interactivehook
When this attribute exists, its value is automatically called (with noarguments) when the interpreter is launched in interactive mode. This is done after the PYTHONSTARTUP file isread, so that you can set this hook there. The site modulesets this.

3.4 新版功能.

sys.intern(string)
Enter string in the table of "interned" strings and return the interned string— which is string itself or a copy. Interning strings is useful to gain alittle performance on dictionary lookup — if the keys in a dictionary areinterned, and the lookup key is interned, the key comparisons (after hashing)can be done by a pointer compare instead of a string compare. Normally, thenames used in Python programs are automatically interned, and the dictionariesused to hold module, class or instance attributes have interned keys.

Interned strings are not immortal; you must keep a reference to the returnvalue of intern() around to benefit from it.

sys.is_finalizing()
Return True if the Python interpreter isshutting down, False otherwise.

3.5 新版功能.

sys.last_type
sys.last_value
sys.last_traceback
These three variables are not always defined; they are set when an exception isnot handled and the interpreter prints an error message and a stack traceback.Their intended use is to allow an interactive user to import a debugger moduleand engage in post-mortem debugging without having to re-execute the commandthat caused the error. (Typical use is import pdb; pdb.pm() to enter thepost-mortem debugger; see pdb module formore information.)

The meaning of the variables is the same as that of the return values fromexc_info() above.

sys.maxsize
An integer giving the maximum value a variable of type Py_ssize_t cantake. It's usually 231 - 1 on a 32-bit platform and 263 - 1 on a64-bit platform.

sys.maxunicode
An integer giving the value of the largest Unicode code point,i.e. 1114111 (0x10FFFF in hexadecimal).

在 3.3 版更改: Before PEP 393, sys.maxunicode used to be either 0xFFFFor 0x10FFFF, depending on the configuration option that specifiedwhether Unicode characters were stored as UCS-2 or UCS-4.

sys.meta_path
A list of meta path finder objects that have theirfind_spec() methods called to see if oneof the objects can find the module to be imported. Thefind_spec() method is called with atleast the absolute name of the module being imported. If the module to beimported is contained in a package, then the parent package's pathattribute is passed in as a second argument. The method returns amodule spec, or None if the module cannot be found.

参见

importlib.abc.MetaPathFinder
The abstract base class defining the interface of finder objects onmeta_path.
importlib.machinery.ModuleSpec
The concrete class whichfind_spec() should returninstances of.

在 3.4 版更改: Module specs were introduced in Python 3.4, byPEP 451. Earlier versions of Python looked for a method calledfind_module().This is still called as a fallback if a meta_path entry doesn'thave a find_spec() method.

sys.modules
This is a dictionary that maps module names to modules which have already beenloaded. This can be manipulated to force reloading of modules and other tricks.However, replacing the dictionary will not necessarily work as expected anddeleting essential items from the dictionary may cause Python to fail.

sys.path
A list of strings that specifies the search path for modules. Initialized fromthe environment variable PYTHONPATH, plus an installation-dependentdefault.

As initialized upon program startup, the first item of this list, path[0],is the directory containing the script that was used to invoke the Pythoninterpreter. If the script directory is not available (e.g. if the interpreteris invoked interactively or if the script is read from standard input),path[0] is the empty string, which directs Python to search modules in thecurrent directory first. Notice that the script directory is inserted _before_the entries inserted as a result of PYTHONPATH.

A program is free to modify this list for its own purposes. Only stringsand bytes should be added to sys.path; all other data types areignored during import.

参见

Module site This describes how to use .pth files to extendsys.path.

sys.path_hooks
A list of callables that take a path argument to try to create afinder for the path. If a finder can be created, it is to bereturned by the callable, else raise ImportError.

Originally specified in PEP 302.

sys.path_importer_cache
A dictionary acting as a cache for finder objects. The keys arepaths that have been passed to sys.path_hooks and the values arethe finders that are found. If a path is a valid file system path but nofinder is found on sys.path_hooks then None isstored.

Originally specified in PEP 302.

在 3.3 版更改: None is stored instead of imp.NullImporter when no finderis found.

sys.platform
This string contains a platform identifier that can be used to appendplatform-specific components to sys.path, for instance.

For Unix systems, except on Linux, this is the lowercased OS name asreturned by uname -s with the first part of the version as returned byuname -r appended, e.g. 'sunos5' or 'freebsd8', at the timewhen Python was built. Unless you want to test for a specific systemversion, it is therefore recommended to use the following idiom:

if sys.platform.startswith('freebsd'):
    # FreeBSD-specific code here...
elif sys.platform.startswith('linux'):
    # Linux-specific code here...

For other systems, the values are:

System

platform value

Linux

'linux'

Windows

'win32'

Windows/Cygwin

'cygwin'

Mac OS X

'darwin'

在 3.3 版更改: On Linux, sys.platform doesn't contain the major version anymore.It is always 'linux', instead of 'linux2' or 'linux3'. Sinceolder Python versions include the version number, it is recommended toalways use the startswith idiom presented above.

参见

os.name has a coarser granularity. os.uname() givessystem-dependent version information.

platform 模块对系统的标识有更详细的检查。

sys.prefix
A string giving the site-specific directory prefix where the platformindependent Python files are installed; by default, this is the string'/usr/local'. This can be set at build time with the —prefixargument to the configure script. The main collection of Pythonlibrary modules is installed in the directory prefix/lib/pythonX.Ywhile the platform independent header files (all except pyconfig.h) arestored in prefix/include/pythonX.Y, where X.Y is the versionnumber of Python, for example 3.2.

注解

If a virtual environment is in effect, thisvalue will be changed in site.py to point to the virtualenvironment. The value for the Python installation will still beavailable, via base_prefix.

sys.ps1
sys.ps2
Strings specifying the primary and secondary prompt of the interpreter. Theseare only defined if the interpreter is in interactive mode. Their initialvalues in this case are '>>> ' and '… '. If a non-string object isassigned to either variable, its str() is re-evaluated each time theinterpreter prepares to read a new interactive command; this can be used toimplement a dynamic prompt.

sys.setcheckinterval(interval)
Set the interpreter's "check interval". This integer value determines how oftenthe interpreter checks for periodic things such as thread switches and signalhandlers. The default is 100, meaning the check is performed every 100Python virtual instructions. Setting it to a larger value may increaseperformance for programs using threads. Setting it to a value <= 0 checksevery virtual instruction, maximizing responsiveness as well as overhead.

3.2 版后已移除: This function doesn't have an effect anymore, as the internal logic forthread switching and asynchronous tasks has been rewritten. Usesetswitchinterval() instead.

sys.setdlopenflags(n)
Set the flags used by the interpreter for dlopen() calls, such as whenthe interpreter loads extension modules. Among other things, this will enable alazy resolving of symbols when importing a module, if called assys.setdlopenflags(0). To share symbols across extension modules, call assys.setdlopenflags(os.RTLD_GLOBAL). Symbolic names for the flag valuescan be found in the os module (RTLD_xxx constants, e.g.os.RTLD_LAZY).

可用性: Unix。

sys.setprofile(profilefunc)
Set the system's profile function, which allows you to implement a Python sourcecode profiler in Python. See chapter Python 分析器 for more information on thePython profiler. The system's profile function is called similarly to thesystem's trace function (see settrace()), but it is called with different events,for example it isn't called for each executed line of code (only on call and return,but the return event is reported even when an exception has been set). The function isthread-specific, but there is no way for the profiler to know about context switches betweenthreads, so it does not make sense to use this in the presence of multiple threads. Also,its return value is not used, so it can simply return None. Error in the profilefunction will cause itself unset.

Profile functions should have three arguments: frame, event, andarg. frame is the current stack frame. event is a string: 'call','return', 'ccall', 'c_return', or 'c_exception'. _arg dependson the event type.

The events have the following meaning:

'call'
A function is called (or some other code block entered). Theprofile function is called; arg is None.
'return'
A function (or other code block) is about to return. The profilefunction is called; arg is the value that will be returned, or Noneif the event is caused by an exception being raised.
'c_call'
A C function is about to be called. This may be an extension function ora built-in. arg is the C function object.
'c_return'
A C function has returned. arg is the C function object.
'c_exception'
A C function has raised an exception. arg is the C function object.

sys.setrecursionlimit(limit)
Set the maximum depth of the Python interpreter stack to limit. This limitprevents infinite recursion from causing an overflow of the C stack and crashingPython.

The highest possible limit is platform-dependent. A user may need to set thelimit higher when they have a program that requires deep recursion and a platformthat supports a higher limit. This should be done with care, because a too-highlimit can lead to a crash.

If the new limit is too low at the current recursion depth, aRecursionError exception is raised.

在 3.5.1 版更改: A RecursionError exception is now raised if the new limit is toolow at the current recursion depth.

sys.setswitchinterval(interval)
Set the interpreter's thread switch interval (in seconds). This floating-pointvalue determines the ideal duration of the "timeslices" allocated toconcurrently running Python threads. Please note that the actual valuecan be higher, especially if long-running internal functions or methodsare used. Also, which thread becomes scheduled at the end of the intervalis the operating system's decision. The interpreter doesn't have itsown scheduler.

3.2 新版功能.

sys.settrace(tracefunc)
Set the system's trace function, which allows you to implement a Pythonsource code debugger in Python. The function is thread-specific; for adebugger to support multiple threads, it must register a trace function usingsettrace() for each thread being debugged or use threading.settrace().

Trace functions should have three arguments: frame, event, andarg. frame is the current stack frame. event is a string: 'call','line', 'return', 'exception' or 'opcode'. arg depends onthe event type.

The trace function is invoked (with event set to 'call') whenever a newlocal scope is entered; it should return a reference to a local tracefunction to be used for the new scope, or None if the scope shouldn't betraced.

The local trace function should return a reference to itself (or to anotherfunction for further tracing in that scope), or None to turn off tracingin that scope.

If there is any error occurred in the trace function, it will be unset, justlike settrace(None) is called.

The events have the following meaning:

'call'
A function is called (or some other code block entered). Theglobal trace function is called; arg is None; the return valuespecifies the local trace function.
'line'
The interpreter is about to execute a new line of code or re-execute thecondition of a loop. The local trace function is called; arg isNone; the return value specifies the new local trace function. SeeObjects/lnotab_notes.txt for a detailed explanation of how thisworks.Per-line events may be disabled for a frame by settingf_trace_lines to False on that frame.
'return'
A function (or other code block) is about to return. The local tracefunction is called; arg is the value that will be returned, or Noneif the event is caused by an exception being raised. The trace function'sreturn value is ignored.
'exception'
An exception has occurred. The local trace function is called; arg is atuple (exception, value, traceback); the return value specifies thenew local trace function.
'opcode'
The interpreter is about to execute a new opcode (see dis foropcode details). The local trace function is called; arg isNone; the return value specifies the new local trace function.Per-opcode events are not emitted by default: they must be explicitlyrequested by setting f_trace_opcodes to True on theframe.

Note that as an exception is propagated down the chain of callers, an'exception' event is generated at each level.

For more fine-grained usage, it's possible to set a trace function byassigning frame.f_trace = tracefunc explicitly, rather than relying onit being set indirectly via the return value from an already installedtrace function. This is also required for activating the trace function onthe current frame, which settrace() doesn't do. Note that in orderfor this to work, a global tracing function must have been installedwith settrace() in order to enable the runtime tracing machinery,but it doesn't need to be the same tracing function (e.g. it could be alow overhead tracing function that simply returns None to disableitself immediately on each frame).

For more information on code and frame objects, refer to 标准类型层级结构.

CPython implementation detail: The settrace() function is intended only for implementing debuggers,profilers, coverage tools and the like. Its behavior is part of theimplementation platform, rather than part of the language definition, andthus may not be available in all Python implementations.

在 3.7 版更改: 'opcode' event type added; f_trace_lines andf_trace_opcodes attributes added to frames

sys.setasyncgen_hooks(_firstiter, finalizer)
Accepts two optional keyword arguments which are callables that accept anasynchronous generator iterator as an argument. The firstiter_callable will be called when an asynchronous generator is iterated for thefirst time. The _finalizer will be called when an asynchronous generatoris about to be garbage collected.

3.6 新版功能: See PEP 525 for more details, and for a reference example of afinalizer method see the implementation ofasyncio.Loop.shutdown_asyncgens inLib/asyncio/base_events.py

注解

This function has been added on a provisional basis (see PEP 411for details.)

sys.setcoroutine_origin_tracking_depth(_depth)
Allows enabling or disabling coroutine origin tracking. Whenenabled, the cr_origin attribute on coroutine objects willcontain a tuple of (filename, line number, function name) tuplesdescribing the traceback where the coroutine object was created,with the most recent call first. When disabled, cr_origin willbe None.

To enable, pass a depth value greater than zero; this sets thenumber of frames whose information will be captured. To disable,pass set depth to zero.

This setting is thread-specific.

3.7 新版功能.

注解

This function has been added on a provisional basis (see PEP 411for details.) Use it only for debugging purposes.

sys.setcoroutine_wrapper(_wrapper)
Allows intercepting creation of coroutine objects (only ones thatare created by an async def function; generators decorated withtypes.coroutine() or asyncio.coroutine() will not beintercepted).

The wrapper argument must be either:

a callable that accepts one argument (a coroutine object);
None, to reset the wrapper.

If called twice, the new wrapper replaces the previous one. The functionis thread-specific.

The wrapper callable cannot define new coroutines directly or indirectly:

def wrapper(coro):
    async def wrap(coro):
        return await coro
    return wrap(coro)
sys.set_coroutine_wrapper(wrapper)
 
async def foo():
    pass
 
# The following line will fail with a RuntimeError, because
# ``wrapper`` creates a ``wrap(coro)`` coroutine:
foo()