Guido van Rossum began working on Python in the late 1980s as a successor to theABC programming language, and he first released it in 1991 as Python 0.9.0.[36] Python 2.0 was released in 2000. Python 3.0, released in 2008, was a major revision not completelybackward-compatible with earlier versions. Python 2.7.18, released in 2020, was the last release of Python 2.[37]
Python consistently ranks as one of the most popular programming languages, and it has gained widespread use in themachine learning community.[38][39][40][41]
Python was conceived in the late 1980s[42] by Guido van Rossum atCentrum Wiskunde & Informatica (CWI) in theNetherlands; it was conceived as a successor to theABC programming language, which was inspired bySETL,[43] capable ofexception handling and interfacing with theAmoeba operating system.[12] Python implementation began in December 1989.[44] Van Rossum assumed sole responsibility for the project, as the lead developer, until 12 July 2018, when he announced his "permanent vacation" from responsibilities as Python's "benevolent dictator for life" (BDFL); this title was bestowed on him by the Python community to reflect his long-term commitment as the project's chief decision-maker.[45] (He has since come out of retirement and is self-titled "BDFL-emeritus".) In January 2019, active Python core developers elected a five-member Steering Council to lead the project.[46][47]
Python 2.0 was released on 16 October 2000, with many major new features such aslist comprehensions,cycle-detecting garbage collection,reference counting, andUnicode support.[49] Python 2.7'send-of-life was initially set for 2015, and then postponed to 2020 out of concern that a large body of existing code could not easily be forward-ported to Python 3.[50][51] It no longer receives security patches or updates.[52][53] While Python 2.7 and older versions are officially unsupported, a different unofficial Python implementation,PyPy, continues to support Python 2, i.e., "2.7.18+" (plus 3.10), with the plus signifying (at least some) "backported security updates".[54]
Python 3.0 was released on 3 December 2008, with some new semantics and changed syntax. At least every Python release since (the now unsupported) 3.5 has added some syntax to the language; a few later releases have removed outdated modules and have changed semantics, at least in a minor way.
As of 8 April 2025[update], Python 3.13.3 is the latest stable release (it's highly recommended to upgrade to it, or upgrade any other older 3.x release). This version currently receives full bug-fix and security updates, while Python 3.12—released in October 2023—had active bug-fix support only until April 2025, and since then only security fixes. Python 3.9[55] is the oldest supported version of Python (albeit in the 'security support' phase), because Python 3.8 has become an end-of-life product.[56][57] Starting with Python 3.13, it and later versions receive two years of full support (which has increased from one and a half years), followed by three years of security support; this is the same total duration of support as previously.
Security updates were expedited in 2021 and again twice in 2022. More issues were fixed in 2023 and in September 2024 (for Python versions 3.8.20 through 3.12.6)—all versions (including 2.7)[58] had been insecure because of issues leading to possibleremote code execution[59] andweb-cache poisoning.[60]
Python 3.10 added the| union type operator[61] and added structuralpattern matching capability to the language, with the newmatch andcase keywords.[62] Python 3.11 expandedexception handling functionality. Python 3.12 added the new keywordtype. Notable changes from version 3.10 to 3.11 include increased program execution speed and improved error reporting.[63] Python 3.11 is claimed to be 10–60% faster than Python 3.10, and Python 3.12 increases by an additional 5%. Python 3.12 also includes improved error messages (again improved in 3.14) and many other changes.
Python 3.13 introduced more syntax for types; a new and improved interactive interpreter (REPL), featuring multi-line editing and color support; an incremental garbage collector, which results in shorter pauses for collection in programs that have many objects, as well as increasing the improved speed in 3.11 and 3.12); anexperimentaljust-in-time (JIT) compiler (such features need to be enabled specifically for the increase in speed);[64] and anexperimental free-threaded build mode, which disables theglobal interpreter lock (GIL), allowing threads to run more concurrently, as enabled inpython3.13t orpython3.13t.exe.
Python Enhancement Proposal (PEP) 711 proposes PyBI—a standard format for distributing Python binaries.[65]
Python 3.14.0 is now in the beta 1 phase (introduces e.g. a new opt-in interpreter, up to 30% faster).
Python 3.15 will "Make UTF-8 mode default";[66] This mode is supported in all current Python versions, but it currently must be opted into.UTF-8 is already used by default on Windows (and other operating systems) for most purposes; an exception is opening files. Enabling UTF-8 also makes code fully cross-platform.
Potentially breaking changes
Python 3.0 introduced very breaking changes, but all breaking changes in 3.x discussed below, are designed to affect few users.
Python 3.12 dropped some outdated modules, and more will be dropped in the future, deprecated as of 3.13; already deprecated array 'u' format code will emitDeprecationWarning since 3.13 and will be removed in Python 3.16. The 'w' format code should be used instead. Part of ctypes is also deprecated andhttp.server.CGIHTTPRequestHandler will emit a DeprecationWarning, and will be removed in 3.15. Using that code already has a high potential for both security and functionality bugs. Parts of the typing module are deprecated, e.g. creating atyping.NamedTuple class using keyword arguments to denote the fields and such (and more) will be disallowed in Python 3.15. Python 3.12 removedwstr meaning Python extensions[67] need to be modified.[68]
Python 3.13 introduces some changes in behavior, i.e., new "well-defined semantics", fixing bugs, and removing many deprecated classes, functions and methods (as well as some of the Python/C API and outdated modules). "The old implementation oflocals() andframe.f_locals was slow, inconsistent and buggy, and it had many corner cases and oddities. Code that works around those may need revising; code that useslocals() for simple templating or print debugging should continue to work correctly."[69]
Python 3.13 introduces the experimental free-threaded build mode, which disables the Global Interpreter Lock (GIL); the GIL is a feature of CPython that previously prevented multiple threads from executing Python bytecode simultaneously. This optional build, introduced through PEP 703, enables better exploitation of multi-core CPUs. By allowing multiple threads to run Python code in parallel, the free-threaded mode addresses long-standing performance bottlenecks associated with the GIL. This change offers a new path for parallelism in Python, without resorting to multiprocessing or external concurrency frameworks.[70]
Regarding annotations in upcoming Python version: "In Python 3.14,from __future__ import annotations will continue to work as it did before, converting annotations into strings."[71]
Python 3.14 drops thePGP digital verification signatures, it had deprecated in version 3.11, when its replacement Sigstore was added for all CPython artifacts; the use of PGP has been criticized by security practitioners.[72]
Some additional standard-library modules will be removed in Python 3.15 or 3.16, as will be many deprecated classes, functions and methods.[73][74]
Python's design offers some support for functional programming in theLisp tradition. It hasfilter,mapandreduce functions;list comprehensions,dictionaries, sets, andgenerator expressions.[82] The standard library has two modules (itertools andfunctools) that implement functional tools borrowed fromHaskell andStandard ML.[83]
Python's core philosophy is summarized in theZen of Python (PEP 20), which includes aphorisms such as these:[84]
Beautiful is better than ugly.
Explicit is better than implicit.
Simple is better than complex.
Complex is better than complicated.
Readability counts.
However, Python features regularly violate these principles and have received criticism for adding unnecessary language bloat.[85] Responses to these criticisms note that the Zen of Python is a guideline rather than a rule.[86] The addition of some new features had been controversial: Guido van Rossum resigned as Benevolent Dictator for Life after conflict about adding the assignment expression operator in Python 3.8.[87][88]
Nevertheless, rather than building all functionality into its core, Python was designed to be highlyextensible via modules. This compact modularity has made it particularly popular as a means of adding programmable interfaces to existing applications. Van Rossum's vision of a small core language with a large standard library and easily extensible interpreter stemmed from his frustrations with ABC, which represented the opposite approach.[42]
Python claims to strive for a simpler, less-cluttered syntax and grammar, while giving developers a choice in their coding methodology. In contrast toPerl's motto "there is more than one way to do it", Python advocates an approach where "there should be one—and preferably only one—obvious way to do it.".[84] In practice, however, Python provides many ways to achieve a given goal. There are, for example, at least three ways to format a string literal, with no certainty as to which one a programmer should use.[89]Alex Martelli is aFellow at thePython Software Foundation and Python book author; he wrote that "To describe something as 'clever' isnot considered a compliment in the Python culture."[90]
Python's developers usually try to avoidpremature optimization; they also reject patches to non-critical parts of theCPython reference implementation that would offer marginal increases in speed at the cost of clarity.[91] Execution speed can be improved by moving speed-critical functions to extension modules written in languages such as C, or by using ajust-in-time compiler likePyPy. It is also possible tocross-compile to other languages; but this approach either fails to achieve the expected speed-up, since Python is a verydynamic language, or only a restricted subset of Python is compiled (with potential minor semantic changes).[92]
Python's developers aim for the language to be fun to use. This goal is reflected in the name—a tribute to the British comedy groupMonty Python[93]—and in playful approaches to some tutorials and reference materials. For instance, some code examples use the terms "spam" and "eggs" (in reference toa Monty Python sketch), rather than the typical terms"foo" and "bar".[94][95] A commonneologism in the Python community ispythonic, which has a wide range of meanings related to program style. Pythonic code may use Pythonidioms well; be natural or show fluency in the language; or conform with Python's minimalist philosophy and emphasis on readability.[96]
An example of Python code and indentationExample ofC# code with curly braces and semicolons
Python is meant to be an easily readable language. Its formatting is visually uncluttered and often uses English keywords where other languages use punctuation. Unlike many other languages, it does not usecurly brackets to delimit blocks, and semicolons after statements are allowed but rarely used. It has fewer syntactic exceptions and special cases thanC orPascal.[97]
Python useswhitespace indentation, rather than curly brackets or keywords, to delimitblocks. An increase in indentation comes after certain statements; a decrease in indentation signifies the end of the current block.[98] Thus, the program's visual structure accurately represents its semantic structure.[99] This feature is sometimes termed theoff-side rule. Some other languages use indentation this way; but in most, indentation has no semantic meaning. The recommended indent size is four spaces.[100]
Theassignment statement, using a single equals sign=
Theif statement, which conditionally executes a block of code, along withelse andelif (a contraction ofelse if)
Thefor statement, which iterates over aniterable object, capturing each element to a local variable for use by the attached block
Thewhile statement, which executes a block of code as long as its condition is true
Thetry statement, which allows exceptions raised in its attached code block to be caught and handled byexcept clauses (or new syntaxexcept* in Python 3.11 for exception groups[101]); thetry statement also ensures that clean-up code in afinally block is always run regardless of how the block exits
Theraise statement, used to raise a specified exception or re-raise a caught exception
Theclass statement, which executes a block of code and attaches its local namespace to aclass, for use in object-oriented programming
Thewith statement, which encloses a code block within a context manager, allowingresource-acquisition-is-initialization (RAII)-like behavior and replacing a common try/finally idiom[102] Examples of a context include acquiring alock before some code is run, and then releasing the lock; or opening and then closing afile
Thecontinue statement, which skips the rest of the current iteration and continues with the next
Thedel statement, which removes a variable—deleting the reference from the name to the value, and producing an error if the variable is referred to before it is redefined[a]
Thepass statement, serving as aNOP (i.e., no operation), which is syntactically needed to create an empty code block
Theassert statement, used in debugging to check for conditions that should apply
Theyield statement, which returns a value from agenerator function (and also an operator); used to implementcoroutines
Thereturn statement, used to return a value from a function
Theimport andfrom statements, used to import modules whose functions or variables can be used in the current program
Thematch andcase statements, analogous to aswitch statement construct, which compares an expression against one or more cases as a control-flow measure
The assignment statement (=) binds a name as areference to a separate, dynamically allocatedobject. Variables may subsequently be rebound at any time to any object. In Python, a variable name is a generic reference holder without a fixeddata type; however, it always refers tosome object with a type. This is calleddynamic typing—in contrast tostatically-typed languages, where each variable may contain only a value of a certain type.
Python does not supporttail call optimization orfirst-class continuations; according to Van Rossum, the language never will.[103][104] However, better support forcoroutine-like functionality is provided by extending Python's generators.[105] Before 2.5, generators werelazyiterators; data was passed unidirectionally out of the generator. From Python 2.5 on, it is possible to pass data back into a generator function; and from version 3.3, data can be passed through multiple stack levels.[106]
The+,-, and* operators for mathematical addition, subtraction, and multiplication are similar to other languages, but the behavior of division differs. There are two types of division in Python:floor division (or integer division)//, and floating-point division/.[107] Python uses the** operator for exponentiation.
Python uses the+ operator for string concatenation. The language uses the* operator for duplicating a string a specified number of times.
The syntax:=, called the "walrus operator", was introduced in Python 3.8. This operator assigns values to variables as part of a larger expression.[110]
In Python,== compares two objects by value. Python'sis operator may be used to compare object identities (i.e., comparison by reference), and comparisons may be chained—for example,a<=b<=c.
Python usesand,or, andnot as Boolean operators.
Python has a type of expression called alist comprehension, and a more general expression called agenerator expression.[82]
Conditional expressions are written asxifcelsey.[111] (This is different in operand order from thec ? x : y operator common to many other languages.)
Python makes a distinction betweenlists andtuples. Lists are written as[1,2,3], are mutable, and cannot be used as the keys of dictionaries (since dictionary keys must beimmutable in Python). Tuples, written as(1,2,3), are immutable and thus can be used as the keys of dictionaries, provided that all of the tuple's elements are immutable. The+ operator can be used to concatenate two tuples, which does not directly modify their contents, but produces a new tuple containing the elements of both. For example, given the variablet initially equal to(1,2,3), executingt=t+(4,5) first evaluatest+(4,5), which yields(1,2,3,4,5); this result is then assigned back tot—thereby effectively "modifying the contents" oft while conforming to the immutable nature of tuple objects. Parentheses are optional for tuples in unambiguous contexts.[112]
Python featuressequence unpacking where multiple expressions, each evaluating to something assignable (e.g., a variable or a writable property) are associated just as in forming tuple literal; as a whole, the results are then put on the left-hand side of the equal sign in an assignment statement. This statement expects aniterable object on the right-hand side of the equal sign to produce the same number of values as the writable expressions on the left-hand side; while iterating, the statement assigns each of the values produced on the right to the corresponding expression on the left.[113]
Python has a "string format" operator% that functions analogously toprintf format strings in the C language—e.g."spam=%s eggs=%d"%("blah",2) evaluates to"spam=blah eggs=2". In Python 2.6+ and 3+, this operator was supplemented by theformat() method of thestr class, e.g.,"spam={0} eggs={1}".format("blah",2). Python 3.6 added "f-strings":spam="blah";eggs=2;f'spam={spam} eggs={eggs}'.[114]
Strings in Python can beconcatenated by "adding" them (using the same operator as for adding integers and floats); e.g.,"spam"+"eggs" returns"spameggs". If strings contain numbers, they are concatenated as strings rather than as integers, e.g."2"+"2" returns"22".
Delimited by single or double quotation marks; single and double quotation marks have equivalent functionality (unlike inUnix shells,Perl, and Perl-influenced languages). Both marks use the backslash (\) as anescape character.String interpolation became available in Python 3.6 as "formatted string literals".[114]
Triple-quoted, i.e., starting and ending with three single or double quotation marks; this may span multiple lines and function likehere documents in shells, Perl, andRuby.
Raw string varieties, denoted by prefixing the string literal withr. Escape sequences are not interpreted; hence raw strings are useful where literal backslashes are common, such as inregular expressions andWindows-style paths. (Compare "@-quoting" inC#.)
Python hasarray index andarray slicing expressions in lists, which are written asa[key],a[start:stop] ora[start:stop:step]. Indexes arezero-based, and negative indexes are relative to the end. Slices take elements from thestart index up to, but not including, thestop index. The (optional) third sliceparameter, calledstep orstride, allows elements to be skipped or reversed. Slice indexes may be omitted—for example,a[:] returns a copy of the entire list. Each element of a slice is ashallow copy.
In Python, a distinction between expressions and statements is rigidly enforced, in contrast to languages such asCommon Lisp,Scheme, orRuby. This distinction leads to duplicating some functionality, for example:
Theeval() vs.exec() built-in functions (in Python 2,exec is a statement); the former function is for expressions, while the latter is for statements
A statement cannot be part of an expression; because of this restriction, expressions such as list anddict comprehensions (and lambda expressions) cannot contain statements. As a particular case, an assignment statement such asa=1 cannot be part of the conditional expression of a conditional statement.
Methods of objects are functions attached to the object's class; the syntax for normal methods and functions,instance.method(argument), issyntactic sugar forClass.method(instance,argument). Python methods have an explicitself parameter to accessinstance data, in contrast to the implicit self (orthis) parameter in some object-oriented programming languages (e.g.,C++,Java,Objective-C,Ruby).[115] Python also provides methods, often calleddunder methods (because their names begin and end with double underscores); these methods allow user-defined classes to modify how they are handled by native operations including length, comparison,arithmetic, and type conversion.[116]
Python usesduck typing, and it has typed objects but untyped variable names. Type constraints are not checked at definition time; rather, operations on an object may fail at usage time, indicating that the object is not of an appropriate type. Despite beingdynamically typed, Python isstrongly typed, forbidding operations that are poorly defined (e.g., adding a number and a string) rather than quietly attempting to interpret them.
Python allows programmers to define their own types usingclasses, most often forobject-oriented programming. Newinstances of classes are constructed by calling the class, for example,SpamClass() orEggsClass()); the classes are instances of themetaclasstype (which is an instance of itself), thereby allowing metaprogramming andreflection.
Before version 3.0, Python had two kinds of classes, both using the same syntax:old-style andnew-style.[117] Current Python versions support the semantics of only the new style.
Python supportsoptional type annotations.[4][118] These annotations are not enforced by the language, but may be used by external tools such asmypy to catch errors.[119][120] Mypy also supports a Python compiler called mypyc, which leverages type annotations for optimization.[121]
Python includes conventional symbols for arithmetic operators (+,-,*,/), the floor-division operator//, and themodulo operator%. (With the module operator, a remainder can be negative, e.g.,4 % -3 == -2.) Python also offers the** symbol forexponentiation, e.g.5**3 == 125 and9**0.5 == 3.0; it also offers the matrix‑multiplication operator@ .[125] These operators work as in traditional mathematics; with the sameprecedence rules, theinfix operators+ and- can also beunary, to represent positive and negative numbers respectively.
Division between integers produces floating-point results. The behavior of division has changed significantly over time:[126]
The current version of Python (i.e., since 3.0) changedthe / operator to always represent floating-point division, e.g.,5/2==2.5.
The floor division// operator was introduced. Thus7//3 == 2,-7//3 == -3,7.5//3 == 2.0, and-7.5//3 == -3.0. For outdated Python 2.7 adding thefrom__future__importdivision statement causes a module in Python 2.7 to use Python 3.0 rules for division instead (see above).
In Python terms, the/ operator representstrue division (or simplydivision), while the// operator representsfloor division. Before version 3.0, the/ operator representsclassic division.[126]
Rounding towards negative infinity, though a different method than in most languages, adds consistency to Python. For instance, this rounding implies that the equation(a+b)//b==a//b+1 is always true. The rounding also implies that the equationb*(a//b)+a%b==a is valid for both positive and negative values ofa. As expected, the result ofa%b lies in thehalf-open interval [0,b), whereb is a positive integer; however, maintaining the validity of the equation requires that the result must lie in the interval (b, 0] whenb is negative.[127]
Python provides around function for rounding a float to the nearest integer. Fortie-breaking, Python 3 uses theround to even method:round(1.5) andround(2.5) both produce2.[128] Python versions before 3 used theround-away-from-zero method:round(0.5) is1.0, andround(-0.5) is−1.0.[129]
Python allows Boolean expressions that contain multiple equality relations to be consistent with general usage in mathematics. For example, the expressiona < b < c tests whethera is less thanb andb is less thanc.[130] C-derived languages interpret this expression differently: in C, the expression would first evaluatea < b, resulting in 0 or 1, and that result would then be compared withc.[131]
Due to Python's extensive mathematics library and the third-party libraryNumPy, the language is frequently used for scientific scripting in tasks such as numerical data processing and manipulation.[134][135]
Functions are created in Python by using thedef keyword. A function is defined similarly to how it is called, by first providing the function name and then the required parameters. Here is an example of a function that prints its inputs:
defprinter(input1,input2="already there"):print(input1)print(input2)printer("hello")# Example output:# hello# already there
To assign a default value to a function parameter in case no actual value is provided at run time, variable-definition syntax can be used inside the function header.
Program to calculate thefactorial of a positive integer:
n=int(input('Type a number, and its factorial will be printed: '))ifn<0:raiseValueError('You must enter a non-negative integer')factorial=1foriinrange(2,n+1):factorial*=iprint(factorial)
Some parts of the standard library are covered by specifications—for example, theWeb Server Gateway Interface (WSGI) implementationwsgiref follows PEP 333[137]—but most parts are specified by their code, internal documentation, andtest suites. However, because most of the standard library is cross-platform Python code, only a few modules must be altered or rewritten for variant implementations.
As of 13 March 2025,[update] thePython Package Index (PyPI), the official repository for third-party Python software, contains over 614,339[138] packages. These have a wide range of functionality, including the following:
Most Python implementations (including CPython) include aread–eval–print loop (REPL); this permits the environment to function as acommand line interpreter, with which users enter statements sequentially and receive results immediately.
CPython is thereference implementation of Python. This implementation is written in C, meeting theC11 standard[141] (since version 3.11, older versions use theC89 standard with several selectC99 features), but third-party extensions are not limited to older C versions—e.g., they can be implemented using C11 or C++.[142][143] CPythoncompiles Python programs into an intermediatebytecode,[144] which is then executed by avirtual machine.[145] CPython is distributed with a large standard library written in a combination of C and native Python.
CPython is available for many platforms, including Windows and most modernUnix-like systems, including macOS (andApple M1 Macs, since Python 3.9.1, using an experimental installer). Starting with Python 3.9, the Python installer intentionally fails to install onWindows 7 and 8;[146][147]Windows XP was supported until Python 3.5, with unofficial support forVMS.[148] Platform portability was one of Python's earliest priorities.[149] During development of Python 1 and 2, evenOS/2 andSolaris were supported;[150] since that time, support has been dropped for many platforms.
All current Python versions (since 3.7) support only operating systems that feature multithreading, by now supporting not nearly as many operating systems (dropping many outdated) than in the past.
All alternative implementations have at least slightly different semantic. For example, an alternative may include unordered dictionaries, in contrast to other current Python versions. As another example in the larger Python ecosystem, PyPy does not support the full C Python API. Alternative implementations include the following:
PyPy is a fast, compliant interpreter of Python 2.7 and 3.10.[151][152] PyPy'sjust-in-time compiler often improves speed significantly relative to CPython, but PyPy does not support some libraries written in C.[153] PyPy offers support for theRISC-V instruction-set architecture, for example.
Codon is an implentation with anahead-of-time (AOT) compiler, which compiles a statically-typed Python-like language whose "syntax and semantics are nearly identical to Python's, there are some notable differences"[154] For example, Codon uses 64-bit machine integers for speed, not arbitrarily as with Python; Codon developers claim that speedups over CPython are usually on the order of ten to a hundred times. Codon compiles to machine code (viaLLVM) and supports native multithreading.[155] Codon can also compile to Python extension modules that can be imported and used from Python.
Pyston is a variant of the Python runtime that uses just-in-time compilation to speed up execution of Python programs.[157]
Cinder is a performance-oriented fork of CPython 3.8 that features a number of optimizations, including bytecode inline caching, eager evaluation of coroutines, a method-at-a-timeJIT, and an experimental bytecode compiler.[158]
The Snek[159][160][161] embedded computing language "is Python-inspired, but it is not Python. It is possible to write Snek programs that run under a full Python system, but most Python programs will not run under Snek."[162] Snek is compatible with 8-bitAVR microcontrollers such asATmega 328P-based Arduino, as well as larger microcontrollers that are compatible withMicroPython. Snek is an imperative language that (unlike Python) omitsobject-oriented programming. Snek supports only one numeric data type, which features 32-bitsingle precision (resemblingJavaScript numbers, though smaller).
Stackless Python is a significant fork of CPython that implementsmicrothreads. This implementation uses thecall stack differently, thus allowing massively concurrent programs. PyPy also offers a stackless version.[163]
Just-in-time Python compilers have been developed, but are now unsupported:
Google began a project namedUnladen Swallow in 2009: this project aimed to speed up the Python interpreter five-fold by usingLLVM, and improvemultithreading capability for scaling to thousands of cores,[164] while typical implementations are limited by theglobal interpreter lock.
Psyco is a discontinuedjust-in-timespecializing compiler, which integrates with CPython and transforms bytecode to machine code at runtime. The emitted code is specialized for certaindata types and is faster than standard Python code. Psyco does not support Python 2.7 or later.
PyS60 was a Python 2 interpreter forSeries 60 mobile phones, which was released byNokia in 2005. The interpreter implemented many modules from Python's standard library, as well as additional modules for integration with theSymbian operating system. The NokiaN900 also supports Python through theGTK widget library, allowing programs to be written and run on the target device.[165]
There are several compilers/transpilers to high-level object languages; the source language is unrestricted Python, a subset of Python, or a language similar to Python:
Cython compiles a superset of Python to C. The resulting code can be used with Python via direct C-level API calls into the Python interpreter.
PyJL compiles/transpiles a subset of Python to "human-readable, maintainable, and high-performance Julia source code".[92] Despite the developers' performance claims, this is not possible forarbitrary Python code; that is, compiling to a faster language or machine code is known to be impossible in the general case. The semantics of Python might potentially be changed, but in many cases speedup is possible with few or no changes in the Python code. The faster Julia source code can then be used from Python or compiled to machine code.
Nuitka compiles Python into C.[169] This compiler works with Python 3.4 to 3.12 (and 2.6 and 2.7) for Python's main supported platforms (and Windows 7 or even Windows XP) and for Android. The compiler developers claim full support for Python 3.10, partial support for Python 3.11 and 3.12, and experimental support for Python 3.13. Nuitka supports macOS including Apple Silicon-based versions. The compiler is free of cost, though it has commercial add-ons (e.g., for hiding source code).
Numba is a JIT compiler that is used from Python; the compiler translates a subset of Python and NumPy code into fast machine code. This tool is enabled by adding a decorator to the relevant Python code.
Pythran compiles a subset of Python 3 to C++ (C++11).[170]
RPython can be compiled to C, and it is used to build the PyPy interpreter for Python.
The Python → 11l → C++ transpiler[171] compiles a subset of Python 3 to C++ (C++17).
IronPython allows running Python 2.7 programs with the .NETCommon Language Runtime.[175] Analpha version (released in 2021), is available for "Python 3.4, although features and behaviors from later versions may be included."[176]
Jython compiles Python 2.7 to Java bytecode, allowing the use of Java libraries from a Python program.[177]
Pyrex (last released in 2010) andShed Skin (last released in 2013) compile to C and C++ respectively.
A perforance comparison among various Python implementations, using a non-numerical (combinatorial) workload, was presented at EuroSciPy '13.[178] In addition, Python's performance relative to other programming languages is benchmarked byThe Computer Language Benchmarks Game.[179]
There are several approaches to optimizing Python performance, given the inherent slowness of aninterpreted language. These approaches include the following strategies or tools:
Just-in-time compilation: Dynamically compiling Python code just before it is executed. This technique is used in libraries such asNumba andPyPy.
Static compilation: Python code is compiled into machine code sometime before execution. An example of this approach is Cython, which compiles Python into C.
Concurrency and parallelism: Multiple tasks can be run simultaneously. Python contains modules such as `multiprocessing` to support this form of parallelism. Moreover, this approach helps to overcome limitations of theGlobal Interpreter Lock (GIL) in CPU tasks.
Efficient data structures: Performance can also be improved by using data types such asSet for membership tests, ordeque fromcollections forqueue operations.
Python's development is conducted largely through thePython Enhancement Proposal (PEP) process; this process is the primary mechanism for proposing major new features, collecting community input on issues, and documenting Python design decisions.[180] Python coding style is covered in PEP 8.[181] Outstanding PEPs are reviewed and commented on by the Python community and the steering council.[180]
Enhancement of the language corresponds with development of the CPython reference implementation. The mailing list python-dev is the primary forum for the language's development. Specific issues were originally discussed in theRoundupbug tracker hosted by the foundation.[182] In 2022, all issues and discussions were migrated toGitHub.[183] Development originally took place on aself-hosted source-code repository runningMercurial, until Python moved toGitHub in January 2017.[184]
CPython's public releases have three types, distinguished by which part of the version number is incremented:
Backward-incompatible versions, where code is expected to break and must be manuallyported. The first part of the version number is incremented. These releases happen infrequently—version 3.0 was released 8 years after 2.0. According to Guido van Rossum, a version 4.0 will probably never exist.[185]
Major or "feature" releases are largely compatible with the previous version but introduce new features. The second part of the version number is incremented. Starting with Python 3.9, these releases are expected to occur annually.[186][187] Each major version is supported by bug fixes for several years after its release.[188]
Bug fix releases,[189] which introduce no new features, occur approximately every three months; these releases are made when a sufficient number of bugs have been fixedupstream since the last release. Security vulnerabilities are also patched in these releases. The third and final part of the version number is incremented.[189]
Manyalpha, beta, and release-candidates are also released as previews and for testing before final releases. Although there is a rough schedule for releases, they are often delayed if the code is not ready yet. Python's development team monitors the state of the code by running a largeunit test suite during development.[190]
Tools that can generate documentation for PythonAPI includepydoc (available as part of the standard library);Sphinx; andPdoc and its forks,Doxygen andGraphviz.[191]
Python's name is inspired by the British comedy groupMonty Python, whom Python creator Guido van Rossum enjoyed while developing the language. Monty Python references appear frequently in Python code and culture;[192] for example, themetasyntactic variables often used in Python literature arespam andeggs, rather than the traditionalfoo andbar.[192][193] The official Python documentation also contains various references to Monty Python routines.[194][195] Python users are sometimes referred to as "Pythonistas".[196]
TheaffixPy is often used when naming Python applications or libraries. Some examples include the following:
Since 2003, Python has consistently ranked in the top ten of the most popular programming languages in theTIOBE Programming Community Index; as of December 2022[update], Python was the most popular language.[40] Python was selected as Programming Language of the Year (for "the highest rise in ratings in a year") in 2007, 2010, 2018, and 2020—the only language to have done so four times as of 2020[update][197]). In the TIOBE Index, monthly rankings are based on the volume of searches for programming languages on Google, Amazon, Wikipedia, Bing, and 20 other platforms. According to the accompanying graph, Python has shown a marked upward trend since the early 2000s, eventually passing more established languages such as C, C++, and Java. This trend can be attributed to Python's readable syntax, comprehensive standard library, and application in data science and machine learning fields.[198]
TIOBE Index Chart showing Python's popularity compared to other programming languages
The combination of Python andProlog has proven useful for AI applications, with Prolog providing knowledge representation and reasoning capabilities. The Janus system, in particular, exploits similarities between these two languages, in part because of their dynamic typing and their simple, recursive data structures. This combination is typically applied natural language processing, visual query answering, geospatial reasoning, and handling semantic web data.[222][223]The Natlog system, implemented in Python, usesDefinite Clause Grammars (DCGs) to create prompts for two types of generators: text-to-text generators such as GPT3, and text-to-image generators such as DALL-E or Stable Diffusion.[224]
Python can be used for graphical user interfaces (GUIs), by using libraries such asTkinter.[225] Similarly, for theOne Laptop per Child XO computer, most of theSugar desktop environment is written in Python (as of 2008).[226]
Python is embedded in many software products (and some hardware products) as a scripting language. These products include the following:
Similarly,GNU Debugger uses Python as apretty printer to show complex structures such as C++ containers.Esri promotes Python as the best choice for writing scripts inArcGIS.[229] Python has also been used in several video games,[230][231] and it has been adopted as first of the threeprogramming languages available inGoogle App Engine (the other two beingJava andGo).[232]LibreOffice includes Python, and its developers plan to replace Java with Python; LibreOffice's Python Scripting Provider is a core feature[233] since version 4.0 (from 7 February 2013).
Ruby's creator,Yukihiro Matsumoto, said that "I wanted a scripting language that was more powerful than Perl, and more object-oriented than Python. That's why I decided to design my own language."[247]
Swift, a programming language developed by Apple, has some Python-inspired syntax.[248]
Kotlin blends Python and Java features, which minimizes boilerplate code and enhances developer efficiency.[249]
Python's development practices have also been emulated by other languages. For example, Python requires a document that describes the rationale and context for any language change; this document is known as aPython Enhancement Proposal or PEP. This practice is also used by the developers ofTcl,[250]Erlang,[251] and Swift.[252]
^del in Python does not behave the same waydelete in languages such asC++ does, where such a word is used to call thedestructor and deallocate heap memory.
^ab"Why was Python created in the first place?".General Python FAQ. Python Software Foundation.Archived from the original on 24 October 2012. Retrieved22 March 2007.I had extensive experience with implementing an interpreted language in the ABC group at CWI, and from working with this group I had learned a lot about language design. This is the origin of many Python features, including the use of indentation for statement grouping and the inclusion of very high-level data types (although the details are all different in Python).
^abKuchling, Andrew M. (22 December 2006)."Interview with Guido van Rossum (July 1998)".amk.ca. Archived fromthe original on 1 May 2007. Retrieved12 March 2012.I'd spent a summer at DEC's Systems Research Center, which introduced me to Modula-2+; the Modula-3 final report was being written there at about the same time. What I learned there later showed up in Python's exception handling, modules, and the fact that methods explicitly contain 'self' in their parameter list. String slicing came from Algol-68 and Icon.
^van Rossum, Guido (1993). "An Introduction to Python for UNIX/C Programmers".Proceedings of the NLUUG Najaarsconferentie (Dutch UNIX Users Group).CiteSeerX10.1.1.38.2023.even though the design of C is far from ideal, its influence on Python is considerable.
^ab"Classes".The Python Tutorial. Python Software Foundation.Archived from the original on 23 October 2012. Retrieved20 February 2012.It is a mixture of the class mechanisms found in C++ and Modula-3
^Lundh, Fredrik."Call By Object".effbot.org.Archived from the original on 23 November 2019. Retrieved21 November 2017.replace "CLU" with "Python", "record" with "instance", and "procedure" with "function or method", and you get a pretty accurate description of Python's object model.
^Simionato, Michele."The Python 2.3 Method Resolution Order". Python Software Foundation.Archived from the original on 20 August 2020. Retrieved29 July 2014.The C3 method itself has nothing to do with Python, since it was invented by people working on Dylan and it is described in a paper intended for lispers
^Kuchling, A. M."Functional Programming HOWTO".Python v2.7.2 documentation. Python Software Foundation.Archived from the original on 24 October 2012. Retrieved9 February 2012.List comprehensions and generator expressions [...] are a concise notation for such operations, borrowed from the functional programming language Haskell.
^Schemenauer, Neil; Peters, Tim; Hetland, Magnus Lie (18 May 2001)."PEP 255 – Simple Generators".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 5 June 2020. Retrieved9 February 2012.
^"More Control Flow Tools".Python 3 documentation. Python Software Foundation.Archived from the original on 4 June 2016. Retrieved24 July 2015.By popular demand, a few features commonly found in functional programming languages like Lisp have been added to Python. With the lambda keyword, small anonymous functions can be created.
^ab"Why We Created Julia".Julia website. February 2012.Archived from the original on 2 May 2020. Retrieved5 June 2014.We want something as usable for general programming as Python [...]
^Lattner, Chris (3 June 2014)."Chris Lattner's Homepage". Chris Lattner.Archived from the original on 25 December 2018. Retrieved3 June 2014.The Swift language is the product of tireless effort from a team of language experts, documentation gurus, compiler optimization ninjas, and an incredibly important internal dogfooding group who provided feedback to help refine and battle-test ideas. Of course, it also greatly benefited from the experiences hard-won by many other languages in the field, drawing ideas from Objective-C, Rust, Haskell, Ruby, Python, C#, CLU, and far too many others to list.
^"About Python". Python Software Foundation.Archived from the original on 20 April 2012. Retrieved24 April 2012., second section "Fans of Python use the phrase "batteries included" to describe the standard library, which covers everything from asynchronous processing to zip files."
^ab"TIOBE Index". TIOBE.Archived from the original on 25 February 2018. Retrieved3 January 2023.The TIOBE Programming Community index is an indicator of the popularity of programming languages Updated as required.
^abVenners, Bill (13 January 2003)."The Making of Python".Artima Developer. Artima.Archived from the original on 1 September 2016. Retrieved22 March 2007.
^van Rossum, Guido (20 January 2009)."A Brief Timeline of Python".The History of Python.Archived from the original on 5 June 2020. Retrieved20 January 2009.
^Briggs, Jason R.; Lipovača, Miran (2013).Python for kids: a playful introduction to programming. San Francisco, Calif: No Starch Press.ISBN978-1-59327-407-8.
^Kuchling, A. M.; Zadka, Moshe (16 October 2000)."What's New in Python 2.0". Python Software Foundation.Archived from the original on 23 October 2012. Retrieved11 February 2012.
^"3.3. Special method names".The Python Language Reference. Python Software Foundation.Archived from the original on 15 December 2018. Retrieved27 June 2009.
^"Extending and Embedding the Python Interpreter: Reference Counts". Docs.python.org.Archived from the original on 18 October 2012. Retrieved5 June 2020.Since Python makes heavy use ofmalloc() andfree(), it needs a strategy to avoid memory leaks as well as the use of freed memory. The chosen method is calledreference counting.
^abHettinger, Raymond (30 January 2002)."PEP 289 – Generator Expressions".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 14 June 2020. Retrieved19 February 2012.
^abPeters, Tim (19 August 2004)."PEP 20 – The Zen of Python".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 26 December 2018. Retrieved24 November 2008.
^Lutz, Mark (January 2022)."Python Changes 2014+".Learning Python.Archived from the original on 15 March 2024. Retrieved25 February 2024.
^ab"Transpiling Python to Julia using PyJL"(PDF).Archived(PDF) from the original on 19 November 2023. Retrieved20 September 2023.After manually modifying one line of code by specifying the necessary type information, we obtained a speedup of 52.6×, making the translated Julia code 19.5× faster than the original Python code.
^Guttag, John V. (12 August 2016).Introduction to Computation and Programming Using Python: With Application to Understanding Data. MIT Press.ISBN978-0-262-52962-4.
^van Rossum, Guido (22 April 2009)."Tail Recursion Elimination". Neopythonic.blogspot.be.Archived from the original on 19 May 2018. Retrieved3 December 2012.
^van Rossum, Guido; Eby, Phillip J. (10 May 2005)."PEP 342 – Coroutines via Enhanced Generators".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 29 May 2020. Retrieved19 February 2012.
^"PEP 380". Python.org.Archived from the original on 4 June 2020. Retrieved3 December 2012.
^"division".python.org.Archived from the original on 20 July 2006. Retrieved30 July 2014.
^van Rossum, Guido; Hettinger, Raymond (7 February 2003)."PEP 308 – Conditional Expressions".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 13 March 2016. Retrieved13 July 2011.
^Zadka, Moshe; van Rossum, Guido (11 March 2001)."PEP 237 – Unifying Long Integers and Integers".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 28 May 2020. Retrieved24 September 2011.
^abZadka, Moshe; van Rossum, Guido (11 March 2001)."PEP 238 – Changing the Division Operator".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 28 May 2020. Retrieved23 October 2013.
^abBatista, Facundo (17 October 2003)."PEP 327 – Decimal Data Type".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 4 June 2020. Retrieved24 November 2008.
^"What's New in Python 2.6".Python v2.6.9 documentation. 29 October 2013.Archived from the original on 23 December 2019. Retrieved26 September 2015.
^van Rossum, Guido (5 June 2001)."PEP 7 – Style Guide for C Code".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 1 June 2020. Retrieved24 November 2008.
^"Jython FAQ".www.jython.org.Archived from the original on 22 April 2021. Retrieved22 April 2021.
^Murri, Riccardo (2013).Performance of Python runtimes on a non-numeric scientific code. European Conference on Python in Science (EuroSciPy).arXiv:1404.6388.Bibcode:2014arXiv1404.6388M.
^abWarsaw, Barry; Hylton, Jeremy; Goodger, David (13 June 2000)."PEP 1 – PEP Purpose and Guidelines".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 6 June 2020. Retrieved19 April 2011.
^abAahz; Baxter, Anthony (15 March 2001)."PEP 6 – Bug Fix Releases".Python Enhancement Proposals. Python Software Foundation.Archived from the original on 5 June 2020. Retrieved27 June 2009.
^"Python Buildbot".Python Developer's Guide. Python Software Foundation.Archived from the original on 5 June 2020. Retrieved24 September 2011.
^Blake, Troy (18 January 2021)."TIOBE Index for January 2021".Technology News and Information by SeniorDBA.Archived from the original on 21 March 2021. Retrieved26 February 2021.
^Taft, Darryl K. (5 March 2007)."Python Slithers into Systems".eWeek.com. Ziff Davis Holdings.Archived from the original on 13 August 2021. Retrieved24 September 2011.
^Andersen, C. and Swift, T., 2023. The Janus System: a bridge to new prolog applications. In Prolog: The Next 50 Years (pp. 93–104). Cham: Springer Nature Switzerland.
^Tarau, P., 2023. Reflections on automation, learnability and expressiveness in logic-based programming languages. In Prolog: The Next 50 Years (pp. 359–371). Cham: Springer Nature Switzerland.
^CCP porkbelly (24 August 2010)."Stackless Python 2.7".EVE Community Dev Blogs.CCP Games.Archived from the original on 11 January 2014. Retrieved11 January 2014.As you may know, EVE has at its core the programming language known as Stackless Python.
^Caudill, Barry (20 September 2005)."Modding Sid Meier's Civilization IV".Sid Meier's Civilization IV Developer Blog.Firaxis Games. Archived fromthe original on 2 December 2010.we created three levels of tools ... The next level offers Python and XML support, letting modders with more experience manipulate the game world and everything in it.
^"Modular Docs – Why Mojo".docs.modular.com.Archived from the original on 5 May 2023. Retrieved5 May 2023.Mojo as a member of the Python family [..] Embracing Python massively simplifies our design efforts, because most of the syntax is already specified. [..] we decided that the right long-term goal for Mojo is to provide a superset of Python (i.e. be compatible with existing programs) and to embrace the CPython immediately for long-tail ecosystem enablement. To a Python programmer, we expect and hope that Mojo will be immediately familiar, while also providing new tools for developing systems-level code that enable you to do things that Python falls back to C and C++ for.
^Yegulalp, Serdar (16 January 2017)."Nim language draws from best of Python, Rust, Go, and Lisp".InfoWorld.Archived from the original on 13 October 2018. Retrieved7 June 2020.Nim's syntax is strongly reminiscent of Python's, as it uses indented code blocks and some of the same syntax (such as the way if/elif/then/else blocks are constructed).
^Lattner, Chris (3 June 2014)."Chris Lattner's Homepage". Chris Lattner.Archived from the original on 22 December 2015. Retrieved3 June 2014.I started work on the Swift Programming Language in July of 2010. I implemented much of the basic language structure, with only a few people knowing of its existence. A few other (amazing) people started contributing in earnest late in 2011, and it became a major focus for the Apple Developer Tools group in July 2013 [...] drawing ideas from Objective-C, Rust, Haskell, Ruby, Python, C#, CLU, and far too many others to list.
^Jalan, Nishant Aanjaney (10 November 2022)."Programming in Kotlin".CodeX. Retrieved29 April 2024.
^Kupries, Andreas; Fellows, Donal K. (14 September 2000)."TIP #3: TIP Format".tcl.tk. Tcl Developer Xchange.Archived from the original on 13 July 2017. Retrieved24 November 2008.
^"Swift Evolution Process".Swift Programming Language Evolution repository on GitHub. 18 February 2020.Archived from the original on 27 April 2020. Retrieved27 April 2020.
.svg)
