Python Tutorial, Release 3.12.

class Dog:

tricks = [] # mistaken use of a class variable

def __init__(self, name):

self.name = name

def add_trick(self, trick):

self.tricks.append(trick)

>>> d = Dog('Fido')
>>> e = Dog('Buddy')
>>> d.add_trick('roll over')
>>> e.add_trick('play dead')
>>> d.tricks # unexpectedly shared by all dogs
['roll over', 'play dead']

Correct design of the class should use an instance variable instead:

class Dog:

def __init__(self, name):

self.name = name
self.tricks = [] # creates a new empty list for each dog

def add_trick(self, trick):

self.tricks.append(trick)

>>> d = Dog('Fido')
>>> e = Dog('Buddy')
>>> d.add_trick('roll over')
>>> e.add_trick('play dead')
>>> d.tricks
['roll over']
>>> e.tricks
['play dead']

9.4 Random Remarks

If the same attribute name occurs in both an instance and in a class, then attribute lookup prioritizes the instance:

>>> class Warehouse:

... purpose = 'storage'
... region = 'west'
...
>>> w1 = Warehouse()
>>> print(w1.purpose, w1.region)
storage west
>>> w2 = Warehouse()
>>> w2.region = 'east'
>>> print(w2.purpose, w2.region)
storage east

Data attributes may be referenced by methods as well as by ordinary users (“clients”) of an object. In other words, classes
are not usable to implement pure abstract data types. In fact, nothing in Python makes it possible to enforce data hiding

9.4. Random Remarks 85

Python Tutorial, Release 3.12.2

— it is all based upon convention. (On the other hand, the Python implementation, written in C, can completely hide
implementation details and control access to an object if necessary; this can be used by extensions to Python written in
C.)
Clients should use data attributes with care — clients may mess up invariants maintained by the methods by stamping
on their data attributes. Note that clients may add data attributes of their own to an instance object without affecting the
validity of the methods, as long as name conflicts are avoided — again, a naming convention can save a lot of headaches
here.
There is no shorthand for referencing data attributes (or other methods!) from within methods. I find that this actually
increases the readability of methods: there is no chance of confusing local variables and instance variables when glancing
through a method.
Often, the first argument of a method is called self. This is nothing more than a convention: the name self has
absolutely no special meaning to Python. Note, however, that by not following the convention your code may be less
readable to other Python programmers, and it is also conceivable that a class browser program might be written that relies
upon such a convention.
Any function object that is a class attribute defines a method for instances of that class. It is not necessary that the function
definition is textually enclosed in the class definition: assigning a function object to a local variable in the class is also ok.
For example:

# Function defined outside the class

def f1(self, x, y):
return min(x, x+y)

class C:
f = f1

def g(self):
return 'hello world'

h = g

Now f, g and h are all attributes of class C that refer to function objects, and consequently they are all methods of
instances of C — h being exactly equivalent to g. Note that this practice usually only serves to confuse the reader of a
program.
Methods may call other methods by using method attributes of the self argument:

class Bag:
def __init__(self):
self.data = []

def add(self, x):

self.data.append(x)

def addtwice(self, x):

self.add(x)
self.add(x)

Methods may reference global names in the same way as ordinary functions. The global scope associated with a method
is the module containing its definition. (A class is never used as a global scope.) While one rarely encounters a good
reason for using global data in a method, there are many legitimate uses of the global scope: for one thing, functions and
modules imported into the global scope can be used by methods, as well as functions and classes defined in it. Usually,
the class containing the method is itself defined in this global scope, and in the next section we’ll find some good reasons
why a method would want to reference its own class.
Each value is an object, and therefore has a class (also called its type). It is stored as object.__class__.

86 Chapter 9. Classes
 Python Tutorial, Release 3.12.2

9.5 Inheritance

Of course, a language feature would not be worthy of the name “class” without supporting inheritance. The syntax for a
derived class definition looks like this:

class DerivedClassName(BaseClassName):
<statement-1>
.
.
.
<statement-N>

The name BaseClassName must be defined in a namespace accessible from the scope containing the derived class
definition. In place of a base class name, other arbitrary expressions are also allowed. This can be useful, for example,
when the base class is defined in another module:

class DerivedClassName(modname.BaseClassName):

Execution of a derived class definition proceeds the same as for a base class. When the class object is constructed, the
base class is remembered. This is used for resolving attribute references: if a requested attribute is not found in the class,
the search proceeds to look in the base class. This rule is applied recursively if the base class itself is derived from some
other class.
There’s nothing special about instantiation of derived classes: DerivedClassName() creates a new instance of the
class. Method references are resolved as follows: the corresponding class attribute is searched, descending down the chain
of base classes if necessary, and the method reference is valid if this yields a function object.
Derived classes may override methods of their base classes. Because methods have no special privileges when calling
other methods of the same object, a method of a base class that calls another method defined in the same base class may
end up calling a method of a derived class that overrides it. (For C++ programmers: all methods in Python are effectively
virtual.)
An overriding method in a derived class may in fact want to extend rather than simply replace the base class method
of the same name. There is a simple way to call the base class method directly: just call BaseClassName.
methodname(self, arguments). This is occasionally useful to clients as well. (Note that this only works if
the base class is accessible as BaseClassName in the global scope.)
Python has two built-in functions that work with inheritance:
• Use isinstance() to check an instance’s type: isinstance(obj, int) will be True only if obj.
__class__ is int or some class derived from int.
• Use issubclass() to check class inheritance: issubclass(bool, int) is True since bool is a sub-
class of int. However, issubclass(float, int) is False since float is not a subclass of int.

9.5.1 Multiple Inheritance

Python supports a form of multiple inheritance as well. A class definition with multiple base classes looks like this:

class DerivedClassName(Base1, Base2, Base3):

<statement-1>
.
.
.
<statement-N>

9.5. Inheritance 87
Python Tutorial, Release 3.12.2

For most purposes, in the simplest cases, you can think of the search for attributes inherited from a parent class as depth-
first, left-to-right, not searching twice in the same class where there is an overlap in the hierarchy. Thus, if an attribute is
not found in DerivedClassName, it is searched for in Base1, then (recursively) in the base classes of Base1, and
if it was not found there, it was searched for in Base2, and so on.
In fact, it is slightly more complex than that; the method resolution order changes dynamically to support cooperative calls
to super(). This approach is known in some other multiple-inheritance languages as call-next-method and is more
powerful than the super call found in single-inheritance languages.
Dynamic ordering is necessary because all cases of multiple inheritance exhibit one or more diamond relationships (where
at least one of the parent classes can be accessed through multiple paths from the bottommost class). For example, all
classes inherit from object, so any case of multiple inheritance provides more than one path to reach object. To
keep the base classes from being accessed more than once, the dynamic algorithm linearizes the search order in a way
that preserves the left-to-right ordering specified in each class, that calls each parent only once, and that is monotonic
(meaning that a class can be subclassed without affecting the precedence order of its parents). Taken together, these
properties make it possible to design reliable and extensible classes with multiple inheritance. For more detail, see https:
//www.python.org/download/releases/2.3/mro/.

9.6 Private Variables

“Private” instance variables that cannot be accessed except from inside an object don’t exist in Python. However, there
is a convention that is followed by most Python code: a name prefixed with an underscore (e.g. _spam) should be
treated as a non-public part of the API (whether it is a function, a method or a data member). It should be considered an
implementation detail and subject to change without notice.
Since there is a valid use-case for class-private members (namely to avoid name clashes of names with names defined by
subclasses), there is limited support for such a mechanism, called name mangling. Any identifier of the form __spam (at
least two leading underscores, at most one trailing underscore) is textually replaced with _classname__spam, where
classname is the current class name with leading underscore(s) stripped. This mangling is done without regard to the
syntactic position of the identifier, as long as it occurs within the definition of a class.
Name mangling is helpful for letting subclasses override methods without breaking intraclass method calls. For example:

class Mapping:
def __init__(self, iterable):
self.items_list = []
self.__update(iterable)

def update(self, iterable):

for item in iterable:
self.items_list.append(item)

__update = update # private copy of original update() method

class MappingSubclass(Mapping):

def update(self, keys, values):

# provides new signature for update()
# but does not break __init__()
for item in zip(keys, values):
self.items_list.append(item)

The above example would work even if MappingSubclass were to introduce a __update identifier since
it is replaced with _Mapping__update in the Mapping class and _MappingSubclass__update in the
MappingSubclass class respectively.

88 Chapter 9. Classes
 Python Tutorial, Release 3.12.2

Note that the mangling rules are designed mostly to avoid accidents; it still is possible to access or modify a variable that
is considered private. This can even be useful in special circumstances, such as in the debugger.
Notice that code passed to exec() or eval() does not consider the classname of the invoking class to be the current
class; this is similar to the effect of the global statement, the effect of which is likewise restricted to code that is
byte-compiled together. The same restriction applies to getattr(), setattr() and delattr(), as well as when
referencing __dict__ directly.

9.7 Odds and Ends

Sometimes it is useful to have a data type similar to the Pascal “record” or C “struct”, bundling together a few named data
items. The idiomatic approach is to use dataclasses for this purpose:
from dataclasses import dataclass

@dataclass
class Employee:
name: str
dept: str
salary: int

>>> john = Employee('john', 'computer lab', 1000)

>>> john.dept
'computer lab'
>>> john.salary
1000

A piece of Python code that expects a particular abstract data type can often be passed a class that emulates the methods
of that data type instead. For instance, if you have a function that formats some data from a file object, you can define a
class with methods read() and readline() that get the data from a string buffer instead, and pass it as an argument.
Instance method objects have attributes, too: m.__self__ is the instance object with the method m(), and m.
__func__ is the function object corresponding to the method.

9.8 Iterators

By now you have probably noticed that most container objects can be looped over using a for statement:
for element in [1, 2, 3]:
print(element)
for element in (1, 2, 3):
print(element)
for key in {'one':1, 'two':2}:
print(key)
for char in "123":
print(char)
for line in open("myfile.txt"):
print(line, end='')

This style of access is clear, concise, and convenient. The use of iterators pervades and unifies Python. Behind the scenes,
the for statement calls iter() on the container object. The function returns an iterator object that defines the method
__next__() which accesses elements in the container one at a time. When there are no more elements, __next__()
raises a StopIteration exception which tells the for loop to terminate. You can call the __next__() method
using the next() built-in function; this example shows how it all works:

9.7. Odds and Ends 89

Python Tutorial, Release 3.12.2

>>> s = 'abc'
>>> it = iter(s)
>>> it
<str_iterator object at 0x10c90e650>
>>> next(it)
'a'
>>> next(it)
'b'
>>> next(it)
'c'
>>> next(it)
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
next(it)
StopIteration

Having seen the mechanics behind the iterator protocol, it is easy to add iterator behavior to your classes. Define an
__iter__() method which returns an object with a __next__() method. If the class defines __next__(), then
__iter__() can just return self:

class Reverse:
"""Iterator for looping over a sequence backwards."""
def __init__(self, data):
self.data = data
self.index = len(data)

def __iter__(self):
return self

def __next__(self):
if self.index == 0:
raise StopIteration
self.index = self.index - 1
return self.data[self.index]

>>> rev = Reverse('spam')

>>> iter(rev)
<__main__.Reverse object at 0x00A1DB50>
>>> for char in rev:
... print(char)
...
m
a
p
s

90 Chapter 9. Classes
 Python Tutorial, Release 3.12.2

9.9 Generators

Generators are a simple and powerful tool for creating iterators. They are written like regular functions but use the yield
statement whenever they want to return data. Each time next() is called on it, the generator resumes where it left off (it
remembers all the data values and which statement was last executed). An example shows that generators can be trivially
easy to create:

def reverse(data):
for index in range(len(data)-1, -1, -1):
yield data[index]

>>> for char in reverse('golf'):

... print(char)
...
f
l
o
g

Anything that can be done with generators can also be done with class-based iterators as described in the previous section.
What makes generators so compact is that the __iter__() and __next__() methods are created automatically.
Another key feature is that the local variables and execution state are automatically saved between calls. This made the
function easier to write and much more clear than an approach using instance variables like self.index and self.
data.
In addition to automatic method creation and saving program state, when generators terminate, they automatically raise
StopIteration. In combination, these features make it easy to create iterators with no more effort than writing a
regular function.

9.10 Generator Expressions

Some simple generators can be coded succinctly as expressions using a syntax similar to list comprehensions but with
parentheses instead of square brackets. These expressions are designed for situations where the generator is used right
away by an enclosing function. Generator expressions are more compact but less versatile than full generator definitions
and tend to be more memory friendly than equivalent list comprehensions.
Examples:

>>> sum(i*i for i in range(10)) # sum of squares

285

>>> xvec = [10, 20, 30]

>>> yvec = [7, 5, 3]
>>> sum(x*y for x,y in zip(xvec, yvec)) # dot product
260

>>> unique_words = set(word for line in page for word in line.split())

>>> valedictorian = max((student.gpa, student.name) for student in graduates)

>>> data = 'golf'

>>> list(data[i] for i in range(len(data)-1, -1, -1))
['f', 'l', 'o', 'g']

9.9. Generators 91
Python Tutorial, Release 3.12.2

92 Chapter 9. Classes
 CHAPTER

TEN

BRIEF TOUR OF THE STANDARD LIBRARY

10.1 Operating System Interface

The os module provides dozens of functions for interacting with the operating system:

>>> import os
>>> os.getcwd() # Return the current working directory
'C:\\Python312'
>>> os.chdir('/server/accesslogs') # Change current working directory
>>> os.system('mkdir today') # Run the command mkdir in the system shell
0

Be sure to use the import os style instead of from os import *. This will keep os.open() from shadowing
the built-in open() function which operates much differently.
The built-in dir() and help() functions are useful as interactive aids for working with large modules like os:

>>> import os
>>> dir(os)
<returns a list of all module functions>
>>> help(os)
<returns an extensive manual page created from the module's docstrings>

For daily file and directory management tasks, the shutil module provides a higher level interface that is easier to use:

>>> import shutil

>>> shutil.copyfile('data.db', 'archive.db')
'archive.db'
>>> shutil.move('/build/executables', 'installdir')
'installdir'

10.2 File Wildcards

The glob module provides a function for making file lists from directory wildcard searches:

>>> import glob

>>> glob.glob('*.py')
['primes.py', 'random.py', 'quote.py']

93
Python Tutorial, Release 3.12.2

10.3 Command Line Arguments

Common utility scripts often need to process command line arguments. These arguments are stored in the sys module’s
argv attribute as a list. For instance, let’s take the following demo.py file:

# File demo.py
import sys
print(sys.argv)

Here is the output from running python demo.py one two three at the command line:

['demo.py', 'one', 'two', 'three']

The argparse module provides a more sophisticated mechanism to process command line arguments. The following
script extracts one or more filenames and an optional number of lines to be displayed:

import argparse

parser = argparse.ArgumentParser(
prog='top',
description='Show top lines from each file')
parser.add_argument('filenames', nargs='+')
parser.add_argument('-l', '--lines', type=int, default=10)
args = parser.parse_args()
print(args)

When run at the command line with python top.py --lines=5 alpha.txt beta.txt, the script sets
args.lines to 5 and args.filenames to ['alpha.txt', 'beta.txt'].

10.4 Error Output Redirection and Program Termination

The sys module also has attributes for stdin, stdout, and stderr. The latter is useful for emitting warnings and error
messages to make them visible even when stdout has been redirected:

>>> sys.stderr.write('Warning, log file not found starting a new one\n')

Warning, log file not found starting a new one

The most direct way to terminate a script is to use sys.exit().

10.5 String Pattern Matching

The re module provides regular expression tools for advanced string processing. For complex matching and manipulation,
regular expressions offer succinct, optimized solutions:

>>> import re
>>> re.findall(r'\bf[a-z]*', 'which foot or hand fell fastest')
['foot', 'fell', 'fastest']
>>> re.sub(r'(\b[a-z]+) \1', r'\1', 'cat in the the hat')
'cat in the hat'

When only simple capabilities are needed, string methods are preferred because they are easier to read and debug:

94 Chapter 10. Brief Tour of the Standard Library

 Python Tutorial, Release 3.12.2

>>> 'tea for too'.replace('too', 'two')

'tea for two'

10.6 Mathematics

The math module gives access to the underlying C library functions for floating point math:

>>> import math

>>> math.cos(math.pi / 4)
0.70710678118654757
>>> math.log(1024, 2)
10.0

The random module provides tools for making random selections:

>>> import random

>>> random.choice(['apple', 'pear', 'banana'])
'apple'
>>> random.sample(range(100), 10) # sampling without replacement
[30, 83, 16, 4, 8, 81, 41, 50, 18, 33]
>>> random.random() # random float
0.17970987693706186
>>> random.randrange(6) # random integer chosen from range(6)
4

The statistics module calculates basic statistical properties (the mean, median, variance, etc.) of numeric data:

>>> import statistics

>>> data = [2.75, 1.75, 1.25, 0.25, 0.5, 1.25, 3.5]
>>> statistics.mean(data)
1.6071428571428572
>>> statistics.median(data)
1.25
>>> statistics.variance(data)
1.3720238095238095

The SciPy project <https://scipy.org> has many other modules for numerical computations.

10.7 Internet Access

There are a number of modules for accessing the internet and processing internet protocols. Two of the simplest are
urllib.request for retrieving data from URLs and smtplib for sending mail:

>>> from urllib.request import urlopen

>>> with urlopen('http://worldtimeapi.org/api/timezone/etc/UTC.txt') as response:
... for line in response:
... line = line.decode() # Convert bytes to a str
... if line.startswith('datetime'):
... print(line.rstrip()) # Remove trailing newline
...
datetime: 2022-01-01T01:36:47.689215+00:00

(continues on next page)

10.6. Mathematics 95
Python Tutorial, Release 3.12.2

(continued from previous page)

>>> import smtplib
>>> server = smtplib.SMTP('localhost')
>>> server.sendmail('[email protected]', '[email protected]',
... """To: [email protected]
... From: [email protected]
...
... Beware the Ides of March.
... """)
>>> server.quit()

(Note that the second example needs a mailserver running on localhost.)

10.8 Dates and Times

The datetime module supplies classes for manipulating dates and times in both simple and complex ways. While date
and time arithmetic is supported, the focus of the implementation is on efficient member extraction for output formatting
and manipulation. The module also supports objects that are timezone aware.

>>> # dates are easily constructed and formatted

>>> from datetime import date
>>> now = date.today()
>>> now
datetime.date(2003, 12, 2)
>>> now.strftime("%m-%d-%y. %d %b %Y is a %A on the %d day of %B.")
'12-02-03. 02 Dec 2003 is a Tuesday on the 02 day of December.'

>>> # dates support calendar arithmetic

>>> birthday = date(1964, 7, 31)
>>> age = now - birthday
>>> age.days
14368

10.9 Data Compression

Common data archiving and compression formats are directly supported by modules including: zlib, gzip, bz2,
lzma, zipfile and tarfile.

>>> import zlib

>>> s = b'witch which has which witches wrist watch'
>>> len(s)
41
>>> t = zlib.compress(s)
>>> len(t)
37
>>> zlib.decompress(t)
b'witch which has which witches wrist watch'
>>> zlib.crc32(s)
226805979

96 Chapter 10. Brief Tour of the Standard Library

 Python Tutorial, Release 3.12.2

10.10 Performance Measurement

Some Python users develop a deep interest in knowing the relative performance of different approaches to the same
problem. Python provides a measurement tool that answers those questions immediately.
For example, it may be tempting to use the tuple packing and unpacking feature instead of the traditional approach to
swapping arguments. The timeit module quickly demonstrates a modest performance advantage:

>>> from timeit import Timer

>>> Timer('t=a; a=b; b=t', 'a=1; b=2').timeit()
0.57535828626024577
>>> Timer('a,b = b,a', 'a=1; b=2').timeit()
0.54962537085770791

In contrast to timeit’s fine level of granularity, the profile and pstats modules provide tools for identifying time
critical sections in larger blocks of code.

10.11 Quality Control

One approach for developing high quality software is to write tests for each function as it is developed and to run those
tests frequently during the development process.
The doctest module provides a tool for scanning a module and validating tests embedded in a program’s docstrings.
Test construction is as simple as cutting-and-pasting a typical call along with its results into the docstring. This improves
the documentation by providing the user with an example and it allows the doctest module to make sure the code remains
true to the documentation:

def average(values):
"""Computes the arithmetic mean of a list of numbers.

>>> print(average([20, 30, 70]))

40.0
"""
return sum(values) / len(values)

import doctest
doctest.testmod() # automatically validate the embedded tests

The unittest module is not as effortless as the doctest module, but it allows a more comprehensive set of tests to
be maintained in a separate file:

import unittest

class TestStatisticalFunctions(unittest.TestCase):

def test_average(self):
self.assertEqual(average([20, 30, 70]), 40.0)
self.assertEqual(round(average([1, 5, 7]), 1), 4.3)
with self.assertRaises(ZeroDivisionError):
average([])
with self.assertRaises(TypeError):
average(20, 30, 70)

unittest.main() # Calling from the command line invokes all tests

10.10. Performance Measurement 97

Python Tutorial, Release 3.12.2

10.12 Batteries Included

Python has a “batteries included” philosophy. This is best seen through the sophisticated and robust capabilities of its
larger packages. For example:
• The xmlrpc.client and xmlrpc.server modules make implementing remote procedure calls into an al-
most trivial task. Despite the modules’ names, no direct knowledge or handling of XML is needed.
• The email package is a library for managing email messages, including MIME and other RFC 2822-based mes-
sage documents. Unlike smtplib and poplib which actually send and receive messages, the email package
has a complete toolset for building or decoding complex message structures (including attachments) and for imple-
menting internet encoding and header protocols.
• The json package provides robust support for parsing this popular data interchange format. The csv module
supports direct reading and writing of files in Comma-Separated Value format, commonly supported by databases
and spreadsheets. XML processing is supported by the xml.etree.ElementTree, xml.dom and xml.sax
packages. Together, these modules and packages greatly simplify data interchange between Python applications and
other tools.
• The sqlite3 module is a wrapper for the SQLite database library, providing a persistent database that can be
updated and accessed using slightly nonstandard SQL syntax.
• Internationalization is supported by a number of modules including gettext, locale, and the codecs pack-
age.

98 Chapter 10. Brief Tour of the Standard Library

 CHAPTER

ELEVEN

BRIEF TOUR OF THE STANDARD LIBRARY — PART II

This second tour covers more advanced modules that support professional programming needs. These modules rarely
occur in small scripts.

11.1 Output Formatting

The reprlib module provides a version of repr() customized for abbreviated displays of large or deeply nested
containers:

>>> import reprlib

>>> reprlib.repr(set('supercalifragilisticexpialidocious'))
"{'a', 'c', 'd', 'e', 'f', 'g', ...}"

The pprint module offers more sophisticated control over printing both built-in and user defined objects in a way that
is readable by the interpreter. When the result is longer than one line, the “pretty printer” adds line breaks and indentation
to more clearly reveal data structure:

>>> import pprint

>>> t = [[[['black', 'cyan'], 'white', ['green', 'red']], [['magenta',
... 'yellow'], 'blue']]]
...
>>> pprint.pprint(t, width=30)
[[[['black', 'cyan'],
'white',
['green', 'red']],
[['magenta', 'yellow'],
'blue']]]

The textwrap module formats paragraphs of text to fit a given screen width:

>>> import textwrap

>>> doc = """The wrap() method is just like fill() except that it returns
... a list of strings instead of one big string with newlines to separate
... the wrapped lines."""
...
>>> print(textwrap.fill(doc, width=40))
The wrap() method is just like fill()
except that it returns a list of strings
instead of one big string with newlines
to separate the wrapped lines.

The locale module accesses a database of culture specific data formats. The grouping attribute of locale’s format
function provides a direct way of formatting numbers with group separators:

99
Python Tutorial, Release 3.12.2

>>> import locale

>>> locale.setlocale(locale.LC_ALL, 'English_United States.1252')
'English_United States.1252'
>>> conv = locale.localeconv() # get a mapping of conventions
>>> x = 1234567.8
>>> locale.format_string("%d", x, grouping=True)
'1,234,567'
>>> locale.format_string("%s%.*f", (conv['currency_symbol'],
... conv['frac_digits'], x), grouping=True)
'$1,234,567.80'

11.2 Templating

The string module includes a versatile Template class with a simplified syntax suitable for editing by end-users.
This allows users to customize their applications without having to alter the application.
The format uses placeholder names formed by $ with valid Python identifiers (alphanumeric characters and underscores).
Surrounding the placeholder with braces allows it to be followed by more alphanumeric letters with no intervening spaces.
Writing $$ creates a single escaped $:

>>> from string import Template

>>> t = Template('${village}folk send $$10 to $cause.')
>>> t.substitute(village='Nottingham', cause='the ditch fund')
'Nottinghamfolk send $10 to the ditch fund.'

The substitute() method raises a KeyError when a placeholder is not supplied in a dictionary or a keyword
argument. For mail-merge style applications, user supplied data may be incomplete and the safe_substitute()
method may be more appropriate — it will leave placeholders unchanged if data is missing:

>>> t = Template('Return the $item to $owner.')

>>> d = dict(item='unladen swallow')
>>> t.substitute(d)
Traceback (most recent call last):
...
KeyError: 'owner'
>>> t.safe_substitute(d)
'Return the unladen swallow to $owner.'

Template subclasses can specify a custom delimiter. For example, a batch renaming utility for a photo browser may elect
to use percent signs for placeholders such as the current date, image sequence number, or file format:

>>> import time, os.path

>>> photofiles = ['img_1074.jpg', 'img_1076.jpg', 'img_1077.jpg']
>>> class BatchRename(Template):
... delimiter = '%'
...
>>> fmt = input('Enter rename style (%d-date %n-seqnum %f-format): ')
Enter rename style (%d-date %n-seqnum %f-format): Ashley_%n%f

>>> t = BatchRename(fmt)
>>> date = time.strftime('%d%b%y')
>>> for i, filename in enumerate(photofiles):
... base, ext = os.path.splitext(filename)
... newname = t.substitute(d=date, n=i, f=ext)
(continues on next page)

100 Chapter 11. Brief Tour of the Standard Library — Part II

 Python Tutorial, Release 3.12.2

(continued from previous page)

... print('{0} --> {1}'.format(filename, newname))

img_1074.jpg --> Ashley_0.jpg

img_1076.jpg --> Ashley_1.jpg
img_1077.jpg --> Ashley_2.jpg

Another application for templating is separating program logic from the details of multiple output formats. This makes it
possible to substitute custom templates for XML files, plain text reports, and HTML web reports.

11.3 Working with Binary Data Record Layouts

The struct module provides pack() and unpack() functions for working with variable length binary record for-
mats. The following example shows how to loop through header information in a ZIP file without using the zipfile
module. Pack codes "H" and "I" represent two and four byte unsigned numbers respectively. The "<" indicates that
they are standard size and in little-endian byte order:

import struct

with open('myfile.zip', 'rb') as f:

data = f.read()

start = 0
for i in range(3): # show the first 3 file headers
start += 14
fields = struct.unpack('<IIIHH', data[start:start+16])
crc32, comp_size, uncomp_size, filenamesize, extra_size = fields

start += 16
filename = data[start:start+filenamesize]
start += filenamesize
extra = data[start:start+extra_size]
print(filename, hex(crc32), comp_size, uncomp_size)

start += extra_size + comp_size # skip to the next header

11.4 Multi-threading

Threading is a technique for decoupling tasks which are not sequentially dependent. Threads can be used to improve
the responsiveness of applications that accept user input while other tasks run in the background. A related use case is
running I/O in parallel with computations in another thread.
The following code shows how the high level threading module can run tasks in background while the main program
continues to run:

import threading, zipfile

class AsyncZip(threading.Thread):
def __init__(self, infile, outfile):
threading.Thread.__init__(self)
self.infile = infile
self.outfile = outfile
(continues on next page)

11.3. Working with Binary Data Record Layouts 101

Python Tutorial, Release 3.12.2

(continued from previous page)

def run(self):
f = zipfile.ZipFile(self.outfile, 'w', zipfile.ZIP_DEFLATED)
f.write(self.infile)
f.close()
print('Finished background zip of:', self.infile)

background = AsyncZip('mydata.txt', 'myarchive.zip')

background.start()
print('The main program continues to run in foreground.')

background.join() # Wait for the background task to finish

print('Main program waited until background was done.')

The principal challenge of multi-threaded applications is coordinating threads that share data or other resources. To that
end, the threading module provides a number of synchronization primitives including locks, events, condition variables,
and semaphores.
While those tools are powerful, minor design errors can result in problems that are difficult to reproduce. So, the preferred
approach to task coordination is to concentrate all access to a resource in a single thread and then use the queue module
to feed that thread with requests from other threads. Applications using Queue objects for inter-thread communication
and coordination are easier to design, more readable, and more reliable.

11.5 Logging

The logging module offers a full featured and flexible logging system. At its simplest, log messages are sent to a file
or to sys.stderr:

import logging
logging.debug('Debugging information')
logging.info('Informational message')
logging.warning('Warning:config file %s not found', 'server.conf')
logging.error('Error occurred')
logging.critical('Critical error -- shutting down')

This produces the following output:

WARNING:root:Warning:config file server.conf not found

ERROR:root:Error occurred
CRITICAL:root:Critical error -- shutting down

By default, informational and debugging messages are suppressed and the output is sent to standard error. Other output
options include routing messages through email, datagrams, sockets, or to an HTTP Server. New filters can select different
routing based on message priority: DEBUG, INFO, WARNING, ERROR, and CRITICAL.
The logging system can be configured directly from Python or can be loaded from a user editable configuration file for
customized logging without altering the application.

102 Chapter 11. Brief Tour of the Standard Library — Part II

 Python Tutorial, Release 3.12.2

11.6 Weak References

Python does automatic memory management (reference counting for most objects and garbage collection to eliminate
cycles). The memory is freed shortly after the last reference to it has been eliminated.
This approach works fine for most applications but occasionally there is a need to track objects only as long as they are
being used by something else. Unfortunately, just tracking them creates a reference that makes them permanent. The
weakref module provides tools for tracking objects without creating a reference. When the object is no longer needed,
it is automatically removed from a weakref table and a callback is triggered for weakref objects. Typical applications
include caching objects that are expensive to create:

>>> import weakref, gc

>>> class A:
... def __init__(self, value):
... self.value = value
... def __repr__(self):
... return str(self.value)
...
>>> a = A(10) # create a reference
>>> d = weakref.WeakValueDictionary()
>>> d['primary'] = a # does not create a reference
>>> d['primary'] # fetch the object if it is still alive
10
>>> del a # remove the one reference
>>> gc.collect() # run garbage collection right away
0
>>> d['primary'] # entry was automatically removed
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
d['primary'] # entry was automatically removed
File "C:/python312/lib/weakref.py", line 46, in __getitem__
o = self.data[key]()
KeyError: 'primary'

11.7 Tools for Working with Lists

Many data structure needs can be met with the built-in list type. However, sometimes there is a need for alternative
implementations with different performance trade-offs.
The array module provides an array() object that is like a list that stores only homogeneous data and stores it more
compactly. The following example shows an array of numbers stored as two byte unsigned binary numbers (typecode
"H") rather than the usual 16 bytes per entry for regular lists of Python int objects:

>>> from array import array

>>> a = array('H', [4000, 10, 700, 22222])
>>> sum(a)
26932
>>> a[1:3]
array('H', [10, 700])

The collections module provides a deque() object that is like a list with faster appends and pops from the left side
but slower lookups in the middle. These objects are well suited for implementing queues and breadth first tree searches:

11.6. Weak References 103

Python Tutorial, Release 3.12.2

>>> from collections import deque

>>> d = deque(["task1", "task2", "task3"])
>>> d.append("task4")
>>> print("Handling", d.popleft())
Handling task1

unsearched = deque([starting_node])
def breadth_first_search(unsearched):
node = unsearched.popleft()
for m in gen_moves(node):
if is_goal(m):
return m
unsearched.append(m)

In addition to alternative list implementations, the library also offers other tools such as the bisect module with functions
for manipulating sorted lists:

>>> import bisect

>>> scores = [(100, 'perl'), (200, 'tcl'), (400, 'lua'), (500, 'python')]
>>> bisect.insort(scores, (300, 'ruby'))
>>> scores
[(100, 'perl'), (200, 'tcl'), (300, 'ruby'), (400, 'lua'), (500, 'python')]

The heapq module provides functions for implementing heaps based on regular lists. The lowest valued entry is always
kept at position zero. This is useful for applications which repeatedly access the smallest element but do not want to run
a full list sort:

>>> from heapq import heapify, heappop, heappush

>>> data = [1, 3, 5, 7, 9, 2, 4, 6, 8, 0]
>>> heapify(data) # rearrange the list into heap order
>>> heappush(data, -5) # add a new entry
>>> [heappop(data) for i in range(3)] # fetch the three smallest entries
[-5, 0, 1]

11.8 Decimal Floating Point Arithmetic

The decimal module offers a Decimal datatype for decimal floating point arithmetic. Compared to the built-in
float implementation of binary floating point, the class is especially helpful for
• financial applications and other uses which require exact decimal representation,
• control over precision,
• control over rounding to meet legal or regulatory requirements,
• tracking of significant decimal places, or
• applications where the user expects the results to match calculations done by hand.
For example, calculating a 5% tax on a 70 cent phone charge gives different results in decimal floating point and binary
floating point. The difference becomes significant if the results are rounded to the nearest cent:

>>> from decimal import *

>>> round(Decimal('0.70') * Decimal('1.05'), 2)
Decimal('0.74')
(continues on next page)

104 Chapter 11. Brief Tour of the Standard Library — Part II

 Python Tutorial, Release 3.12.2

(continued from previous page)

>>> round(.70 * 1.05, 2)
0.73

The Decimal result keeps a trailing zero, automatically inferring four place significance from multiplicands with two
place significance. Decimal reproduces mathematics as done by hand and avoids issues that can arise when binary floating
point cannot exactly represent decimal quantities.
Exact representation enables the Decimal class to perform modulo calculations and equality tests that are unsuitable for
binary floating point:

>>> Decimal('1.00') % Decimal('.10')

Decimal('0.00')
>>> 1.00 % 0.10
0.09999999999999995

>>> sum([Decimal('0.1')]*10) == Decimal('1.0')

True
>>> 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 == 1.0
False

The decimal module provides arithmetic with as much precision as needed:

>>> getcontext().prec = 36
>>> Decimal(1) / Decimal(7)
Decimal('0.142857142857142857142857142857142857')

11.8. Decimal Floating Point Arithmetic 105

Python Tutorial, Release 3.12.2

106 Chapter 11. Brief Tour of the Standard Library — Part II

 CHAPTER

TWELVE

VIRTUAL ENVIRONMENTS AND PACKAGES

12.1 Introduction

Python applications will often use packages and modules that don’t come as part of the standard library. Applications will
sometimes need a specific version of a library, because the application may require that a particular bug has been fixed or
the application may be written using an obsolete version of the library’s interface.
This means it may not be possible for one Python installation to meet the requirements of every application. If application
A needs version 1.0 of a particular module but application B needs version 2.0, then the requirements are in conflict and
installing either version 1.0 or 2.0 will leave one application unable to run.
The solution for this problem is to create a virtual environment, a self-contained directory tree that contains a Python
installation for a particular version of Python, plus a number of additional packages.
Different applications can then use different virtual environments. To resolve the earlier example of conflicting require-
ments, application A can have its own virtual environment with version 1.0 installed while application B has another
virtual environment with version 2.0. If application B requires a library be upgraded to version 3.0, this will not affect
application A’s environment.

12.2 Creating Virtual Environments

The module used to create and manage virtual environments is called venv. venv will usually install the most recent
version of Python that you have available. If you have multiple versions of Python on your system, you can select a specific
Python version by running python3 or whichever version you want.
To create a virtual environment, decide upon a directory where you want to place it, and run the venv module as a script
with the directory path:

python -m venv tutorial-env

This will create the tutorial-env directory if it doesn’t exist, and also create directories inside it containing a copy
of the Python interpreter and various supporting files.
A common directory location for a virtual environment is .venv. This name keeps the directory typically hidden in your
shell and thus out of the way while giving it a name that explains why the directory exists. It also prevents clashing with
.env environment variable definition files that some tooling supports.
Once you’ve created a virtual environment, you may activate it.
On Windows, run:

tutorial-env\Scripts\activate

107
Python Tutorial, Release 3.12.2

On Unix or MacOS, run:

source tutorial-env/bin/activate

(This script is written for the bash shell. If you use the csh or fish shells, there are alternate activate.csh and
activate.fish scripts you should use instead.)
Activating the virtual environment will change your shell’s prompt to show what virtual environment you’re using, and
modify the environment so that running python will get you that particular version and installation of Python. For
example:

$ source ~/envs/tutorial-env/bin/activate
(tutorial-env) $ python
Python 3.5.1 (default, May 6 2016, 10:59:36)
...
>>> import sys
>>> sys.path
['', '/usr/local/lib/python35.zip', ...,
'~/envs/tutorial-env/lib/python3.5/site-packages']
>>>

To deactivate a virtual environment, type:

deactivate

into the terminal.

12.3 Managing Packages with pip

You can install, upgrade, and remove packages using a program called pip. By default pip will install packages from
the Python Package Index. You can browse the Python Package Index by going to it in your web browser.
pip has a number of subcommands: “install”, “uninstall”, “freeze”, etc. (Consult the installing-index guide for complete
documentation for pip.)
You can install the latest version of a package by specifying a package’s name:

(tutorial-env) $ python -m pip install novas

Collecting novas
Downloading novas-3.1.1.3.tar.gz (136kB)
Installing collected packages: novas
Running setup.py install for novas
Successfully installed novas-3.1.1.3

You can also install a specific version of a package by giving the package name followed by == and the version number:

(tutorial-env) $ python -m pip install requests==2.6.0

Collecting requests==2.6.0
Using cached requests-2.6.0-py2.py3-none-any.whl
Installing collected packages: requests
Successfully installed requests-2.6.0

If you re-run this command, pip will notice that the requested version is already installed and do nothing. You can supply
a different version number to get that version, or you can run python -m pip install --upgrade to upgrade
the package to the latest version:

108 Chapter 12. Virtual Environments and Packages

 Python Tutorial, Release 3.12.2

(tutorial-env) $ python -m pip install --upgrade requests

Collecting requests
Installing collected packages: requests
Found existing installation: requests 2.6.0
Uninstalling requests-2.6.0:
Successfully uninstalled requests-2.6.0
Successfully installed requests-2.7.0

python -m pip uninstall followed by one or more package names will remove the packages from the virtual
environment.
python -m pip show will display information about a particular package:

(tutorial-env) $ python -m pip show requests

---
Metadata-Version: 2.0
Name: requests
Version: 2.7.0
Summary: Python HTTP for Humans.
Home-page: http://python-requests.org
Author: Kenneth Reitz
Author-email: [email protected]
License: Apache 2.0
Location: /Users/akuchling/envs/tutorial-env/lib/python3.4/site-packages
Requires:

python -m pip list will display all of the packages installed in the virtual environment:

(tutorial-env) $ python -m pip list

novas (3.1.1.3)
numpy (1.9.2)
pip (7.0.3)
requests (2.7.0)
setuptools (16.0)

python -m pip freeze will produce a similar list of the installed packages, but the output uses the format that
python -m pip install expects. A common convention is to put this list in a requirements.txt file:

(tutorial-env) $ python -m pip freeze > requirements.txt

(tutorial-env) $ cat requirements.txt
novas==3.1.1.3
numpy==1.9.2
requests==2.7.0

The requirements.txt can then be committed to version control and shipped as part of an application. Users can
then install all the necessary packages with install -r:

(tutorial-env) $ python -m pip install -r requirements.txt

Collecting novas==3.1.1.3 (from -r requirements.txt (line 1))
...
Collecting numpy==1.9.2 (from -r requirements.txt (line 2))
...
Collecting requests==2.7.0 (from -r requirements.txt (line 3))
...
Installing collected packages: novas, numpy, requests
Running setup.py install for novas
Successfully installed novas-3.1.1.3 numpy-1.9.2 requests-2.7.0

12.3. Managing Packages with pip 109

Python Tutorial, Release 3.12.2

pip has many more options. Consult the installing-index guide for complete documentation for pip. When you’ve
written a package and want to make it available on the Python Package Index, consult the Python packaging user guide.

110 Chapter 12. Virtual Environments and Packages

 CHAPTER

THIRTEEN

WHAT NOW?

Reading this tutorial has probably reinforced your interest in using Python — you should be eager to apply Python to
solving your real-world problems. Where should you go to learn more?
This tutorial is part of Python’s documentation set. Some other documents in the set are:
• library-index:
You should browse through this manual, which gives complete (though terse) reference material about types,
functions, and the modules in the standard library. The standard Python distribution includes a lot of additional
code. There are modules to read Unix mailboxes, retrieve documents via HTTP, generate random numbers, parse
command-line options, compress data, and many other tasks. Skimming through the Library Reference will give
you an idea of what’s available.
• installing-index explains how to install additional modules written by other Python users.
• reference-index: A detailed explanation of Python’s syntax and semantics. It’s heavy reading, but is useful as a
complete guide to the language itself.
More Python resources:
• https://www.python.org: The major Python web site. It contains code, documentation, and pointers to Python-
related pages around the web.
• https://docs.python.org: Fast access to Python’s documentation.
• https://pypi.org: The Python Package Index, previously also nicknamed the Cheese Shop1 , is an index of user-
created Python modules that are available for download. Once you begin releasing code, you can register it here so
that others can find it.
• https://code.activestate.com/recipes/langs/python/: The Python Cookbook is a sizable collection of code examples,
larger modules, and useful scripts. Particularly notable contributions are collected in a book also titled Python
Cookbook (O’Reilly & Associates, ISBN 0-596-00797-3.)
• https://pyvideo.org collects links to Python-related videos from conferences and user-group meetings.
• https://scipy.org: The Scientific Python project includes modules for fast array computations and manipulations
plus a host of packages for such things as linear algebra, Fourier transforms, non-linear solvers, random number
distributions, statistical analysis and the like.
For Python-related questions and problem reports, you can post to the newsgroup comp.lang.python, or send them
to the mailing list at [email protected]. The newsgroup and mailing list are gatewayed, so messages posted to one
will automatically be forwarded to the other. There are hundreds of postings a day, asking (and answering) questions,
suggesting new features, and announcing new modules. Mailing list archives are available at https://mail.python.org/
pipermail/.
Before posting, be sure to check the list of Frequently Asked Questions (also called the FAQ). The FAQ answers many
of the questions that come up again and again, and may already contain the solution for your problem.
1 “Cheese Shop” is a Monty Python’s sketch: a customer enters a cheese shop, but whatever cheese he asks for, the clerk says it’s missing.

111
Python Tutorial, Release 3.12.2

112 Chapter 13. What Now?

 CHAPTER

FOURTEEN

INTERACTIVE INPUT EDITING AND HISTORY SUBSTITUTION

Some versions of the Python interpreter support editing of the current input line and history substitution, similar to
facilities found in the Korn shell and the GNU Bash shell. This is implemented using the GNU Readline library, which
supports various styles of editing. This library has its own documentation which we won’t duplicate here.

14.1 Tab Completion and History Editing

Completion of variable and module names is automatically enabled at interpreter startup so that the Tab key invokes
the completion function; it looks at Python statement names, the current local variables, and the available module names.
For dotted expressions such as string.a, it will evaluate the expression up to the final '.' and then suggest comple-
tions from the attributes of the resulting object. Note that this may execute application-defined code if an object with a
__getattr__() method is part of the expression. The default configuration also saves your history into a file named
.python_history in your user directory. The history will be available again during the next interactive interpreter
session.

14.2 Alternatives to the Interactive Interpreter

This facility is an enormous step forward compared to earlier versions of the interpreter; however, some wishes are left:
It would be nice if the proper indentation were suggested on continuation lines (the parser knows if an indent token is
required next). The completion mechanism might use the interpreter’s symbol table. A command to check (or even
suggest) matching parentheses, quotes, etc., would also be useful.
One alternative enhanced interactive interpreter that has been around for quite some time is IPython, which features tab
completion, object exploration and advanced history management. It can also be thoroughly customized and embedded
into other applications. Another similar enhanced interactive environment is bpython.

113
Python Tutorial, Release 3.12.2

114 Chapter 14. Interactive Input Editing and History Substitution

 CHAPTER

FIFTEEN

FLOATING POINT ARITHMETIC: ISSUES AND LIMITATIONS

Floating-point numbers are represented in computer hardware as base 2 (binary) fractions. For example, the decimal
fraction 0.625 has value 6/10 + 2/100 + 5/1000, and in the same way the binary fraction 0.101 has value 1/2 + 0/4 +
1/8. These two fractions have identical values, the only real difference being that the first is written in base 10 fractional
notation, and the second in base 2.
Unfortunately, most decimal fractions cannot be represented exactly as binary fractions. A consequence is that, in general,
the decimal floating-point numbers you enter are only approximated by the binary floating-point numbers actually stored
in the machine.
The problem is easier to understand at first in base 10. Consider the fraction 1/3. You can approximate that as a base 10
fraction:

0.3

or, better,

0.33

or, better,

0.333

and so on. No matter how many digits you’re willing to write down, the result will never be exactly 1/3, but will be an
increasingly better approximation of 1/3.
In the same way, no matter how many base 2 digits you’re willing to use, the decimal value 0.1 cannot be represented
exactly as a base 2 fraction. In base 2, 1/10 is the infinitely repeating fraction

0.0001100110011001100110011001100110011001100110011...

Stop at any finite number of bits, and you get an approximation. On most machines today, floats are approximated using
a binary fraction with the numerator using the first 53 bits starting with the most significant bit and with the denominator
as a power of two. In the case of 1/10, the binary fraction is 3602879701896397 / 2 ** 55 which is close to
but not exactly equal to the true value of 1/10.
Many users are not aware of the approximation because of the way values are displayed. Python only prints a decimal
approximation to the true decimal value of the binary approximation stored by the machine. On most machines, if Python
were to print the true decimal value of the binary approximation stored for 0.1, it would have to display:

>>> 0.1
0.1000000000000000055511151231257827021181583404541015625

That is more digits than most people find useful, so Python keeps the number of digits manageable by displaying a rounded
value instead:

115
Python Tutorial, Release 3.12.2

>>> 1 / 10
0.1

Just remember, even though the printed result looks like the exact value of 1/10, the actual stored value is the nearest
representable binary fraction.
Interestingly, there are many different decimal numbers that share the same nearest approxi-
mate binary fraction. For example, the numbers 0.1 and 0.10000000000000001 and 0.
1000000000000000055511151231257827021181583404541015625 are all approximated by
3602879701896397 / 2 ** 55. Since all of these decimal values share the same approximation, any one
of them could be displayed while still preserving the invariant eval(repr(x)) == x.
Historically, the Python prompt and built-in repr() function would choose the one with 17 significant digits, 0.
10000000000000001. Starting with Python 3.1, Python (on most systems) is now able to choose the shortest of
these and simply display 0.1.
Note that this is in the very nature of binary floating-point: this is not a bug in Python, and it is not a bug in your code
either. You’ll see the same kind of thing in all languages that support your hardware’s floating-point arithmetic (although
some languages may not display the difference by default, or in all output modes).
For more pleasant output, you may wish to use string formatting to produce a limited number of significant digits:
>>> format(math.pi, '.12g') # give 12 significant digits
'3.14159265359'

>>> format(math.pi, '.2f') # give 2 digits after the point

'3.14'

>>> repr(math.pi)
'3.141592653589793'

It’s important to realize that this is, in a real sense, an illusion: you’re simply rounding the display of the true machine
value.
One illusion may beget another. For example, since 0.1 is not exactly 1/10, summing three values of 0.1 may not yield
exactly 0.3, either:
>>> 0.1 + 0.1 + 0.1 == 0.3
False

Also, since the 0.1 cannot get any closer to the exact value of 1/10 and 0.3 cannot get any closer to the exact value of
3/10, then pre-rounding with round() function cannot help:
>>> round(0.1, 1) + round(0.1, 1) + round(0.1, 1) == round(0.3, 1)
False

Though the numbers cannot be made closer to their intended exact values, the math.isclose() function can be useful
for comparing inexact values:
>>> math.isclose(0.1 + 0.1 + 0.1, 0.3)
True

Alternatively, the round() function can be used to compare rough approximations:

>>> round(math.pi, ndigits=2) == round(22 / 7, ndigits=2)
True

Binary floating-point arithmetic holds many surprises like this. The problem with “0.1” is explained in precise detail
below, in the “Representation Error” section. See Examples of Floating Point Problems for a pleasant summary of how

116 Chapter 15. Floating Point Arithmetic: Issues and Limitations

 Python Tutorial, Release 3.12.2

binary floating-point works and the kinds of problems commonly encountered in practice. Also see The Perils of Floating
Point for a more complete account of other common surprises.
As that says near the end, “there are no easy answers.” Still, don’t be unduly wary of floating-point! The errors in Python
float operations are inherited from the floating-point hardware, and on most machines are on the order of no more than
1 part in 2**53 per operation. That’s more than adequate for most tasks, but you do need to keep in mind that it’s not
decimal arithmetic and that every float operation can suffer a new rounding error.
While pathological cases do exist, for most casual use of floating-point arithmetic you’ll see the result you expect in the
end if you simply round the display of your final results to the number of decimal digits you expect. str() usually
suffices, and for finer control see the str.format() method’s format specifiers in formatstrings.
For use cases which require exact decimal representation, try using the decimal module which implements decimal
arithmetic suitable for accounting applications and high-precision applications.
Another form of exact arithmetic is supported by the fractions module which implements arithmetic based on rational
numbers (so the numbers like 1/3 can be represented exactly).
If you are a heavy user of floating-point operations you should take a look at the NumPy package and many other packages
for mathematical and statistical operations supplied by the SciPy project. See <https://scipy.org>.
Python provides tools that may help on those rare occasions when you really do want to know the exact value of a float.
The float.as_integer_ratio() method expresses the value of a float as a fraction:

>>> x = 3.14159
>>> x.as_integer_ratio()
(3537115888337719, 1125899906842624)

Since the ratio is exact, it can be used to losslessly recreate the original value:

>>> x == 3537115888337719 / 1125899906842624

True

The float.hex() method expresses a float in hexadecimal (base 16), again giving the exact value stored by your
computer:

>>> x.hex()
'0x1.921f9f01b866ep+1'

This precise hexadecimal representation can be used to reconstruct the float value exactly:

>>> x == float.fromhex('0x1.921f9f01b866ep+1')
True

Since the representation is exact, it is useful for reliably porting values across different versions of Python (platform
independence) and exchanging data with other languages that support the same format (such as Java and C99).
Another helpful tool is the sum() function which helps mitigate loss-of-precision during summation. It uses extended
precision for intermediate rounding steps as values are added onto a running total. That can make a difference in overall
accuracy so that the errors do not accumulate to the point where they affect the final total:

>>> 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 + 0.1 == 1.0
False
>>> sum([0.1] * 10) == 1.0
True

The math.fsum() goes further and tracks all of the “lost digits” as values are added onto a running total so that the
result has only a single rounding. This is slower than sum() but will be more accurate in uncommon cases where large
magnitude inputs mostly cancel each other out leaving a final sum near zero:

117
Python Tutorial, Release 3.12.2

>>> arr = [-0.10430216751806065, -266310978.67179024, 143401161448607.16,

... -143401161400469.7, 266262841.31058735, -0.003244936839808227]
>>> float(sum(map(Fraction, arr))) # Exact summation with single rounding
8.042173697819788e-13
>>> math.fsum(arr) # Single rounding
8.042173697819788e-13
>>> sum(arr) # Multiple roundings in extended precision
8.042178034628478e-13
>>> total = 0.0
>>> for x in arr:
... total += x # Multiple roundings in standard precision
...
>>> total # Straight addition has no correct digits!
-0.0051575902860057365

15.1 Representation Error

This section explains the “0.1” example in detail, and shows how you can perform an exact analysis of cases like this
yourself. Basic familiarity with binary floating-point representation is assumed.
Representation error refers to the fact that some (most, actually) decimal fractions cannot be represented exactly as binary
(base 2) fractions. This is the chief reason why Python (or Perl, C, C++, Java, Fortran, and many others) often won’t
display the exact decimal number you expect.
Why is that? 1/10 is not exactly representable as a binary fraction. Since at least 2000, almost all machines use IEEE 754
binary floating-point arithmetic, and almost all platforms map Python floats to IEEE 754 binary64 “double precision”
values. IEEE 754 binary64 values contain 53 bits of precision, so on input the computer strives to convert 0.1 to the
closest fraction it can of the form J/2**N where J is an integer containing exactly 53 bits. Rewriting

1 / 10 ~= J / (2**N)

as

J ~= 2**N / 10

and recalling that J has exactly 53 bits (is >= 2**52 but < 2**53), the best value for N is 56:

>>> 2**52 <= 2**56 // 10 < 2**53

True

That is, 56 is the only value for N that leaves J with exactly 53 bits. The best possible value for J is then that quotient
rounded:

>>> q, r = divmod(2**56, 10)

>>> r
6

Since the remainder is more than half of 10, the best approximation is obtained by rounding up:

>>> q+1
7205759403792794

Therefore the best possible approximation to 1/10 in IEEE 754 double precision is:

118 Chapter 15. Floating Point Arithmetic: Issues and Limitations

 Python Tutorial, Release 3.12.2

7205759403792794 / 2 ** 56

Dividing both the numerator and denominator by two reduces the fraction to:

3602879701896397 / 2 ** 55

Note that since we rounded up, this is actually a little bit larger than 1/10; if we had not rounded up, the quotient would
have been a little bit smaller than 1/10. But in no case can it be exactly 1/10!
So the computer never “sees” 1/10: what it sees is the exact fraction given above, the best IEEE 754 double approximation
it can get:

>>> 0.1 * 2 ** 55
3602879701896397.0

If we multiply that fraction by 10**55, we can see the value out to 55 decimal digits:

>>> 3602879701896397 * 10 ** 55 // 2 ** 55
1000000000000000055511151231257827021181583404541015625

meaning that the exact number stored in the computer is equal to the decimal value
0.1000000000000000055511151231257827021181583404541015625. Instead of displaying the full decimal
value, many languages (including older versions of Python), round the result to 17 significant digits:

>>> format(0.1, '.17f')

'0.10000000000000001'

The fractions and decimal modules make these calculations easy:

>>> from decimal import Decimal

>>> from fractions import Fraction

>>> Fraction.from_float(0.1)
Fraction(3602879701896397, 36028797018963968)

>>> (0.1).as_integer_ratio()
(3602879701896397, 36028797018963968)

>>> Decimal.from_float(0.1)
Decimal('0.1000000000000000055511151231257827021181583404541015625')

>>> format(Decimal.from_float(0.1), '.17')

'0.10000000000000001'

15.1. Representation Error 119

Python Tutorial, Release 3.12.2

120 Chapter 15. Floating Point Arithmetic: Issues and Limitations

 CHAPTER

SIXTEEN

APPENDIX

16.1 Interactive Mode

16.1.1 Error Handling

When an error occurs, the interpreter prints an error message and a stack trace. In interactive mode, it then returns to the
primary prompt; when input came from a file, it exits with a nonzero exit status after printing the stack trace. (Exceptions
handled by an except clause in a try statement are not errors in this context.) Some errors are unconditionally fatal
and cause an exit with a nonzero exit status; this applies to internal inconsistencies and some cases of running out of
memory. All error messages are written to the standard error stream; normal output from executed commands is written
to standard output.
Typing the interrupt character (usually Control-C or Delete) to the primary or secondary prompt cancels
the input and returns to the primary prompt.1 Typing an interrupt while a command is executing raises the
KeyboardInterrupt exception, which may be handled by a try statement.

16.1.2 Executable Python Scripts

On BSD’ish Unix systems, Python scripts can be made directly executable, like shell scripts, by putting the line

#!/usr/bin/env python3.5

(assuming that the interpreter is on the user’s PATH) at the beginning of the script and giving the file an executable mode.
The #! must be the first two characters of the file. On some platforms, this first line must end with a Unix-style line
ending ('\n'), not a Windows ('\r\n') line ending. Note that the hash, or pound, character, '#', is used to start a
comment in Python.
The script can be given an executable mode, or permission, using the chmod command.

$ chmod +x myscript.py

On Windows systems, there is no notion of an “executable mode”. The Python installer automatically associates .py files
with python.exe so that a double-click on a Python file will run it as a script. The extension can also be .pyw, in that
case, the console window that normally appears is suppressed.
1 A problem with the GNU Readline package may prevent this.

121
Python Tutorial, Release 3.12.2

16.1.3 The Interactive Startup File

When you use Python interactively, it is frequently handy to have some standard commands executed every time the
interpreter is started. You can do this by setting an environment variable named PYTHONSTARTUP to the name of a file
containing your start-up commands. This is similar to the .profile feature of the Unix shells.
This file is only read in interactive sessions, not when Python reads commands from a script, and not when /dev/tty
is given as the explicit source of commands (which otherwise behaves like an interactive session). It is executed in the
same namespace where interactive commands are executed, so that objects that it defines or imports can be used without
qualification in the interactive session. You can also change the prompts sys.ps1 and sys.ps2 in this file.
If you want to read an additional start-up file from the current directory, you can program this in the global start-up file
using code like if os.path.isfile('.pythonrc.py'): exec(open('.pythonrc.py').read()).
If you want to use the startup file in a script, you must do this explicitly in the script:

import os
filename = os.environ.get('PYTHONSTARTUP')
if filename and os.path.isfile(filename):
with open(filename) as fobj:
startup_file = fobj.read()
exec(startup_file)

16.1.4 The Customization Modules

Python provides two hooks to let you customize it: sitecustomize and usercustomize. To see how it works, you need first
to find the location of your user site-packages directory. Start Python and run this code:

>>> import site

>>> site.getusersitepackages()
'/home/user/.local/lib/python3.5/site-packages'

Now you can create a file named usercustomize.py in that directory and put anything you want in it. It will affect
every invocation of Python, unless it is started with the -s option to disable the automatic import.
sitecustomize works in the same way, but is typically created by an administrator of the computer in the global site-
packages directory, and is imported before usercustomize. See the documentation of the site module for more details.

122 Chapter 16. Appendix

 APPENDIX

GLOSSARY

>>>
The default Python prompt of the interactive shell. Often seen for code examples which can be executed interac-
tively in the interpreter.
...
Can refer to:
• The default Python prompt of the interactive shell when entering the code for an indented code block, when
within a pair of matching left and right delimiters (parentheses, square brackets, curly braces or triple quotes),
or after specifying a decorator.
• The Ellipsis built-in constant.
2to3
A tool that tries to convert Python 2.x code to Python 3.x code by handling most of the incompatibilities which can
be detected by parsing the source and traversing the parse tree.
2to3 is available in the standard library as lib2to3; a standalone entry point is provided as Tools/scripts/
2to3. See 2to3-reference.
abstract base class
Abstract base classes complement duck-typing by providing a way to define interfaces when other techniques
like hasattr() would be clumsy or subtly wrong (for example with magic methods). ABCs introduce vir-
tual subclasses, which are classes that don’t inherit from a class but are still recognized by isinstance() and
issubclass(); see the abc module documentation. Python comes with many built-in ABCs for data structures
(in the collections.abc module), numbers (in the numbers module), streams (in the io module), import
finders and loaders (in the importlib.abc module). You can create your own ABCs with the abc module.
annotation
A label associated with a variable, a class attribute or a function parameter or return value, used by convention as
a type hint.
Annotations of local variables cannot be accessed at runtime, but annotations of global variables, class attributes,
and functions are stored in the __annotations__ special attribute of modules, classes, and functions, respec-
tively.
See variable annotation, function annotation, PEP 484 and PEP 526, which describe this functionality. Also see
annotations-howto for best practices on working with annotations.
argument
A value passed to a function (or method) when calling the function. There are two kinds of argument:
• keyword argument: an argument preceded by an identifier (e.g. name=) in a function call or passed as a value
in a dictionary preceded by **. For example, 3 and 5 are both keyword arguments in the following calls to
complex():

123
Python Tutorial, Release 3.12.2

complex(real=3, imag=5)
complex(**{'real': 3, 'imag': 5})

• positional argument: an argument that is not a keyword argument. Positional arguments can appear at the
beginning of an argument list and/or be passed as elements of an iterable preceded by *. For example, 3 and
5 are both positional arguments in the following calls:

complex(3, 5)
complex(*(3, 5))

Arguments are assigned to the named local variables in a function body. See the calls section for the rules governing
this assignment. Syntactically, any expression can be used to represent an argument; the evaluated value is assigned
to the local variable.
See also the parameter glossary entry, the FAQ question on the difference between arguments and parameters, and
PEP 362.
asynchronous context manager
An object which controls the environment seen in an async with statement by defining __aenter__() and
__aexit__() methods. Introduced by PEP 492.
asynchronous generator
A function which returns an asynchronous generator iterator. It looks like a coroutine function defined with async
def except that it contains yield expressions for producing a series of values usable in an async for loop.
Usually refers to an asynchronous generator function, but may refer to an asynchronous generator iterator in some
contexts. In cases where the intended meaning isn’t clear, using the full terms avoids ambiguity.
An asynchronous generator function may contain await expressions as well as async for, and async with
statements.
asynchronous generator iterator
An object created by a asynchronous generator function.
This is an asynchronous iterator which when called using the __anext__() method returns an awaitable object
which will execute the body of the asynchronous generator function until the next yield expression.
Each yield temporarily suspends processing, remembering the location execution state (including local variables
and pending try-statements). When the asynchronous generator iterator effectively resumes with another awaitable
returned by __anext__(), it picks up where it left off. See PEP 492 and PEP 525.
asynchronous iterable
An object, that can be used in an async for statement. Must return an asynchronous iterator from its
__aiter__() method. Introduced by PEP 492.
asynchronous iterator
An object that implements the __aiter__() and __anext__() methods. __anext__() must return an
awaitable object. async for resolves the awaitables returned by an asynchronous iterator’s __anext__()
method until it raises a StopAsyncIteration exception. Introduced by PEP 492.
attribute
A value associated with an object which is usually referenced by name using dotted expressions. For example, if
an object o has an attribute a it would be referenced as o.a.
It is possible to give an object an attribute whose name is not an identifier as defined by identifiers, for example
using setattr(), if the object allows it. Such an attribute will not be accessible using a dotted expression, and
would instead need to be retrieved with getattr().
awaitable
An object that can be used in an await expression. Can be a coroutine or an object with an __await__()

124 Appendix A. Glossary

 Python Tutorial, Release 3.12.2

method. See also PEP 492.

BDFL
Benevolent Dictator For Life, a.k.a. Guido van Rossum, Python’s creator.
binary file
A file object able to read and write bytes-like objects. Examples of binary files are files opened in binary mode
('rb', 'wb' or 'rb+'), sys.stdin.buffer, sys.stdout.buffer, and instances of io.BytesIO
and gzip.GzipFile.
See also text file for a file object able to read and write str objects.
borrowed reference
In Python’s C API, a borrowed reference is a reference to an object, where the code using the object does not
own the reference. It becomes a dangling pointer if the object is destroyed. For example, a garbage collection can
remove the last strong reference to the object and so destroy it.
Calling Py_INCREF() on the borrowed reference is recommended to convert it to a strong reference in-place,
except when the object cannot be destroyed before the last usage of the borrowed reference. The Py_NewRef()
function can be used to create a new strong reference.
bytes-like object
An object that supports the bufferobjects and can export a C-contiguous buffer. This includes all bytes,
bytearray, and array.array objects, as well as many common memoryview objects. Bytes-like ob-
jects can be used for various operations that work with binary data; these include compression, saving to a binary
file, and sending over a socket.
Some operations need the binary data to be mutable. The documentation often refers to these as “read-write bytes-
like objects”. Example mutable buffer objects include bytearray and a memoryview of a bytearray. Other
operations require the binary data to be stored in immutable objects (“read-only bytes-like objects”); examples of
these include bytes and a memoryview of a bytes object.
bytecode
Python source code is compiled into bytecode, the internal representation of a Python program in the CPython
interpreter. The bytecode is also cached in .pyc files so that executing the same file is faster the second time
(recompilation from source to bytecode can be avoided). This “intermediate language” is said to run on a virtual
machine that executes the machine code corresponding to each bytecode. Do note that bytecodes are not expected
to work between different Python virtual machines, nor to be stable between Python releases.
A list of bytecode instructions can be found in the documentation for the dis module.
callable
A callable is an object that can be called, possibly with a set of arguments (see argument), with the following syntax:

callable(argument1, argument2, argumentN)

A function, and by extension a method, is a callable. An instance of a class that implements the __call__()
method is also a callable.
callback
A subroutine function which is passed as an argument to be executed at some point in the future.
class
A template for creating user-defined objects. Class definitions normally contain method definitions which operate
on instances of the class.
class variable
A variable defined in a class and intended to be modified only at class level (i.e., not in an instance of the class).
complex number
An extension of the familiar real number system in which all numbers are expressed as a sum of a real part and an

125
Python Tutorial, Release 3.12.2

imaginary part. Imaginary numbers are real multiples of the imaginary unit (the square root of -1), often written
i in mathematics or j in engineering. Python has built-in support for complex numbers, which are written with
this latter notation; the imaginary part is written with a j suffix, e.g., 3+1j. To get access to complex equivalents
of the math module, use cmath. Use of complex numbers is a fairly advanced mathematical feature. If you’re
not aware of a need for them, it’s almost certain you can safely ignore them.
context manager
An object which controls the environment seen in a with statement by defining __enter__() and
__exit__() methods. See PEP 343.
context variable
A variable which can have different values depending on its context. This is similar to Thread-Local Storage in
which each execution thread may have a different value for a variable. However, with context variables, there may
be several contexts in one execution thread and the main usage for context variables is to keep track of variables in
concurrent asynchronous tasks. See contextvars.
contiguous
A buffer is considered contiguous exactly if it is either C-contiguous or Fortran contiguous. Zero-dimensional buffers
are C and Fortran contiguous. In one-dimensional arrays, the items must be laid out in memory next to each other,
in order of increasing indexes starting from zero. In multidimensional C-contiguous arrays, the last index varies
the fastest when visiting items in order of memory address. However, in Fortran contiguous arrays, the first index
varies the fastest.
coroutine
Coroutines are a more generalized form of subroutines. Subroutines are entered at one point and exited at another
point. Coroutines can be entered, exited, and resumed at many different points. They can be implemented with the
async def statement. See also PEP 492.
coroutine function
A function which returns a coroutine object. A coroutine function may be defined with the async def statement,
and may contain await, async for, and async with keywords. These were introduced by PEP 492.
CPython
The canonical implementation of the Python programming language, as distributed on python.org. The term
“CPython” is used when necessary to distinguish this implementation from others such as Jython or IronPython.
decorator
A function returning another function, usually applied as a function transformation using the @wrapper syntax.
Common examples for decorators are classmethod() and staticmethod().
The decorator syntax is merely syntactic sugar, the following two function definitions are semantically equivalent:

def f(arg):
...
f = staticmethod(f)

@staticmethod
def f(arg):
...

The same concept exists for classes, but is less commonly used there. See the documentation for function definitions
and class definitions for more about decorators.
descriptor
Any object which defines the methods __get__(), __set__(), or __delete__(). When a class attribute
is a descriptor, its special binding behavior is triggered upon attribute lookup. Normally, using a.b to get, set or
delete an attribute looks up the object named b in the class dictionary for a, but if b is a descriptor, the respective
descriptor method gets called. Understanding descriptors is a key to a deep understanding of Python because

126 Appendix A. Glossary

 Python Tutorial, Release 3.12.2

they are the basis for many features including functions, methods, properties, class methods, static methods, and
reference to super classes.
For more information about descriptors’ methods, see descriptors or the Descriptor How To Guide.
dictionary
An associative array, where arbitrary keys are mapped to values. The keys can be any object with __hash__()
and __eq__() methods. Called a hash in Perl.
dictionary comprehension
A compact way to process all or part of the elements in an iterable and return a dictionary with the results. results
= {n: n ** 2 for n in range(10)} generates a dictionary containing key n mapped to value n **
2. See comprehensions.
dictionary view
The objects returned from dict.keys(), dict.values(), and dict.items() are called dictionary
views. They provide a dynamic view on the dictionary’s entries, which means that when the dictionary changes,
the view reflects these changes. To force the dictionary view to become a full list use list(dictview). See
dict-views.
docstring
A string literal which appears as the first expression in a class, function or module. While ignored when the suite is
executed, it is recognized by the compiler and put into the __doc__ attribute of the enclosing class, function or
module. Since it is available via introspection, it is the canonical place for documentation of the object.
duck-typing
A programming style which does not look at an object’s type to determine if it has the right interface; instead,
the method or attribute is simply called or used (“If it looks like a duck and quacks like a duck, it must be a
duck.”) By emphasizing interfaces rather than specific types, well-designed code improves its flexibility by allowing
polymorphic substitution. Duck-typing avoids tests using type() or isinstance(). (Note, however, that
duck-typing can be complemented with abstract base classes.) Instead, it typically employs hasattr() tests or
EAFP programming.
EAFP
Easier to ask for forgiveness than permission. This common Python coding style assumes the existence of valid
keys or attributes and catches exceptions if the assumption proves false. This clean and fast style is characterized
by the presence of many try and except statements. The technique contrasts with the LBYL style common to
many other languages such as C.
expression
A piece of syntax which can be evaluated to some value. In other words, an expression is an accumulation of
expression elements like literals, names, attribute access, operators or function calls which all return a value. In
contrast to many other languages, not all language constructs are expressions. There are also statements which
cannot be used as expressions, such as while. Assignments are also statements, not expressions.
extension module
A module written in C or C++, using Python’s C API to interact with the core and with user code.
f-string
String literals prefixed with 'f' or 'F' are commonly called “f-strings” which is short for formatted string literals.
See also PEP 498.
file object
An object exposing a file-oriented API (with methods such as read() or write()) to an underlying resource.
Depending on the way it was created, a file object can mediate access to a real on-disk file or to another type of
storage or communication device (for example standard input/output, in-memory buffers, sockets, pipes, etc.). File
objects are also called file-like objects or streams.
There are actually three categories of file objects: raw binary files, buffered binary files and text files. Their interfaces
are defined in the io module. The canonical way to create a file object is by using the open() function.

127
Python Tutorial, Release 3.12.2

file-like object
A synonym for file object.
filesystem encoding and error handler
Encoding and error handler used by Python to decode bytes from the operating system and encode Unicode to the
operating system.
The filesystem encoding must guarantee to successfully decode all bytes below 128. If the file system encoding
fails to provide this guarantee, API functions can raise UnicodeError.
The sys.getfilesystemencoding() and sys.getfilesystemencodeerrors() functions can
be used to get the filesystem encoding and error handler.
The filesystem encoding and error handler are configured at Python startup by the PyConfig_Read() function:
see filesystem_encoding and filesystem_errors members of PyConfig.
See also the locale encoding.
finder
An object that tries to find the loader for a module that is being imported.
Since Python 3.3, there are two types of finder: meta path finders for use with sys.meta_path, and path entry
finders for use with sys.path_hooks.
See PEP 302, PEP 420 and PEP 451 for much more detail.
floor division
Mathematical division that rounds down to nearest integer. The floor division operator is //. For example, the
expression 11 // 4 evaluates to 2 in contrast to the 2.75 returned by float true division. Note that (-11) //
4 is -3 because that is -2.75 rounded downward. See PEP 238.
function
A series of statements which returns some value to a caller. It can also be passed zero or more arguments which
may be used in the execution of the body. See also parameter, method, and the function section.
function annotation
An annotation of a function parameter or return value.
Function annotations are usually used for type hints: for example, this function is expected to take two int argu-
ments and is also expected to have an int return value:

def sum_two_numbers(a: int, b: int) -> int:

return a + b

Function annotation syntax is explained in section function.

See variable annotation and PEP 484, which describe this functionality. Also see annotations-howto for best
practices on working with annotations.
__future__
A future statement, from __future__ import <feature>, directs the compiler to compile the current
module using syntax or semantics that will become standard in a future release of Python. The __future__
module documents the possible values of feature. By importing this module and evaluating its variables, you can
see when a new feature was first added to the language and when it will (or did) become the default:

>>> import __future__

>>> __future__.division
_Feature((2, 2, 0, 'alpha', 2), (3, 0, 0, 'alpha', 0), 8192)

garbage collection
The process of freeing memory when it is not used anymore. Python performs garbage collection via reference

128 Appendix A. Glossary

 Python Tutorial, Release 3.12.2

counting and a cyclic garbage collector that is able to detect and break reference cycles. The garbage collector can
be controlled using the gc module.
generator
A function which returns a generator iterator. It looks like a normal function except that it contains yield expres-
sions for producing a series of values usable in a for-loop or that can be retrieved one at a time with the next()
function.
Usually refers to a generator function, but may refer to a generator iterator in some contexts. In cases where the
intended meaning isn’t clear, using the full terms avoids ambiguity.
generator iterator
An object created by a generator function.
Each yield temporarily suspends processing, remembering the location execution state (including local variables
and pending try-statements). When the generator iterator resumes, it picks up where it left off (in contrast to
functions which start fresh on every invocation).
generator expression
An expression that returns an iterator. It looks like a normal expression followed by a for clause defining a loop
variable, range, and an optional if clause. The combined expression generates values for an enclosing function:

>>> sum(i*i for i in range(10)) # sum of squares 0, 1, 4, ... 81

285

generic function
A function composed of multiple functions implementing the same operation for different types. Which imple-
mentation should be used during a call is determined by the dispatch algorithm.
See also the single dispatch glossary entry, the functools.singledispatch() decorator, and PEP 443.
generic type
A type that can be parameterized; typically a container class such as list or dict. Used for type hints and
annotations.
For more details, see generic alias types, PEP 483, PEP 484, PEP 585, and the typing module.
GIL
See global interpreter lock.
global interpreter lock
The mechanism used by the CPython interpreter to assure that only one thread executes Python bytecode at a time.
This simplifies the CPython implementation by making the object model (including critical built-in types such as
dict) implicitly safe against concurrent access. Locking the entire interpreter makes it easier for the interpreter
to be multi-threaded, at the expense of much of the parallelism afforded by multi-processor machines.
However, some extension modules, either standard or third-party, are designed so as to release the GIL when doing
computationally intensive tasks such as compression or hashing. Also, the GIL is always released when doing I/O.
Past efforts to create a “free-threaded” interpreter (one which locks shared data at a much finer granularity) have not
been successful because performance suffered in the common single-processor case. It is believed that overcoming
this performance issue would make the implementation much more complicated and therefore costlier to maintain.
hash-based pyc
A bytecode cache file that uses the hash rather than the last-modified time of the corresponding source file to
determine its validity. See pyc-invalidation.
hashable
An object is hashable if it has a hash value which never changes during its lifetime (it needs a __hash__()
method), and can be compared to other objects (it needs an __eq__() method). Hashable objects which compare
equal must have the same hash value.

129
Python Tutorial, Release 3.12.2

Hashability makes an object usable as a dictionary key and a set member, because these data structures use the hash
value internally.
Most of Python’s immutable built-in objects are hashable; mutable containers (such as lists or dictionaries) are not;
immutable containers (such as tuples and frozensets) are only hashable if their elements are hashable. Objects which
are instances of user-defined classes are hashable by default. They all compare unequal (except with themselves),
and their hash value is derived from their id().
IDLE
An Integrated Development and Learning Environment for Python. idle is a basic editor and interpreter environ-
ment which ships with the standard distribution of Python.
immutable
An object with a fixed value. Immutable objects include numbers, strings and tuples. Such an object cannot be
altered. A new object has to be created if a different value has to be stored. They play an important role in places
where a constant hash value is needed, for example as a key in a dictionary.
import path
A list of locations (or path entries) that are searched by the path based finder for modules to import. During
import, this list of locations usually comes from sys.path, but for subpackages it may also come from the parent
package’s __path__ attribute.
importing
The process by which Python code in one module is made available to Python code in another module.
importer
An object that both finds and loads a module; both a finder and loader object.
interactive
Python has an interactive interpreter which means you can enter statements and expressions at the interpreter
prompt, immediately execute them and see their results. Just launch python with no arguments (possibly by
selecting it from your computer’s main menu). It is a very powerful way to test out new ideas or inspect modules
and packages (remember help(x)).
interpreted
Python is an interpreted language, as opposed to a compiled one, though the distinction can be blurry because of the
presence of the bytecode compiler. This means that source files can be run directly without explicitly creating an
executable which is then run. Interpreted languages typically have a shorter development/debug cycle than compiled
ones, though their programs generally also run more slowly. See also interactive.
interpreter shutdown
When asked to shut down, the Python interpreter enters a special phase where it gradually releases all allocated
resources, such as modules and various critical internal structures. It also makes several calls to the garbage collector.
This can trigger the execution of code in user-defined destructors or weakref callbacks. Code executed during the
shutdown phase can encounter various exceptions as the resources it relies on may not function anymore (common
examples are library modules or the warnings machinery).
The main reason for interpreter shutdown is that the __main__ module or the script being run has finished
executing.
iterable
An object capable of returning its members one at a time. Examples of iterables include all sequence types (such
as list, str, and tuple) and some non-sequence types like dict, file objects, and objects of any classes you
define with an __iter__() method or with a __getitem__() method that implements sequence semantics.
Iterables can be used in a for loop and in many other places where a sequence is needed (zip(), map(), …).
When an iterable object is passed as an argument to the built-in function iter(), it returns an iterator for the
object. This iterator is good for one pass over the set of values. When using iterables, it is usually not necessary to
call iter() or deal with iterator objects yourself. The for statement does that automatically for you, creating

130 Appendix A. Glossary

 Python Tutorial, Release 3.12.2

a temporary unnamed variable to hold the iterator for the duration of the loop. See also iterator, sequence, and
generator.
iterator
An object representing a stream of data. Repeated calls to the iterator’s __next__() method (or passing it
to the built-in function next()) return successive items in the stream. When no more data are available a
StopIteration exception is raised instead. At this point, the iterator object is exhausted and any further calls
to its __next__() method just raise StopIteration again. Iterators are required to have an __iter__()
method that returns the iterator object itself so every iterator is also iterable and may be used in most places where
other iterables are accepted. One notable exception is code which attempts multiple iteration passes. A container
object (such as a list) produces a fresh new iterator each time you pass it to the iter() function or use it in a
for loop. Attempting this with an iterator will just return the same exhausted iterator object used in the previous
iteration pass, making it appear like an empty container.
More information can be found in typeiter.
CPython implementation detail: CPython does not consistently apply the requirement that an iterator define
__iter__().
key function
A key function or collation function is a callable that returns a value used for sorting or ordering. For example,
locale.strxfrm() is used to produce a sort key that is aware of locale specific sort conventions.
A number of tools in Python accept key functions to control how elements are ordered or grouped. They in-
clude min(), max(), sorted(), list.sort(), heapq.merge(), heapq.nsmallest(), heapq.
nlargest(), and itertools.groupby().
There are several ways to create a key function. For example. the str.lower() method can serve as a key
function for case insensitive sorts. Alternatively, a key function can be built from a lambda expression such as
lambda r: (r[0], r[2]). Also, operator.attrgetter(), operator.itemgetter(), and
operator.methodcaller() are three key function constructors. See the Sorting HOW TO for examples of
how to create and use key functions.
keyword argument
See argument.
lambda
An anonymous inline function consisting of a single expression which is evaluated when the function is called. The
syntax to create a lambda function is lambda [parameters]: expression
LBYL
Look before you leap. This coding style explicitly tests for pre-conditions before making calls or lookups. This
style contrasts with the EAFP approach and is characterized by the presence of many if statements.
In a multi-threaded environment, the LBYL approach can risk introducing a race condition between “the looking”
and “the leaping”. For example, the code, if key in mapping: return mapping[key] can fail if
another thread removes key from mapping after the test, but before the lookup. This issue can be solved with locks
or by using the EAFP approach.
locale encoding
On Unix, it is the encoding of the LC_CTYPE locale. It can be set with locale.setlocale(locale.
LC_CTYPE, new_locale).
On Windows, it is the ANSI code page (ex: "cp1252").
On Android and VxWorks, Python uses "utf-8" as the locale encoding.
locale.getencoding() can be used to get the locale encoding.
See also the filesystem encoding and error handler.

131
Python Tutorial, Release 3.12.2

list
A built-in Python sequence. Despite its name it is more akin to an array in other languages than to a linked list
since access to elements is O(1).
list comprehension
A compact way to process all or part of the elements in a sequence and return a list with the results. result
= ['{:#04x}'.format(x) for x in range(256) if x % 2 == 0] generates a list of strings
containing even hex numbers (0x..) in the range from 0 to 255. The if clause is optional. If omitted, all elements
in range(256) are processed.
loader
An object that loads a module. It must define a method named load_module(). A loader is typically returned
by a finder. See PEP 302 for details and importlib.abc.Loader for an abstract base class.
magic method
An informal synonym for special method.
mapping
A container object that supports arbitrary key lookups and implements the methods specified in the
collections.abc.Mapping or collections.abc.MutableMapping abstract base classes.
Examples include dict, collections.defaultdict, collections.OrderedDict and
collections.Counter.
meta path finder
A finder returned by a search of sys.meta_path. Meta path finders are related to, but different from path entry
finders.
See importlib.abc.MetaPathFinder for the methods that meta path finders implement.
metaclass
The class of a class. Class definitions create a class name, a class dictionary, and a list of base classes. The metaclass
is responsible for taking those three arguments and creating the class. Most object oriented programming languages
provide a default implementation. What makes Python special is that it is possible to create custom metaclasses.
Most users never need this tool, but when the need arises, metaclasses can provide powerful, elegant solutions.
They have been used for logging attribute access, adding thread-safety, tracking object creation, implementing
singletons, and many other tasks.
More information can be found in metaclasses.
method
A function which is defined inside a class body. If called as an attribute of an instance of that class, the method
will get the instance object as its first argument (which is usually called self). See function and nested scope.
method resolution order
Method Resolution Order is the order in which base classes are searched for a member during lookup. See The
Python 2.3 Method Resolution Order for details of the algorithm used by the Python interpreter since the 2.3
release.
module
An object that serves as an organizational unit of Python code. Modules have a namespace containing arbitrary
Python objects. Modules are loaded into Python by the process of importing.
See also package.
module spec
A namespace containing the import-related information used to load a module. An instance of importlib.
machinery.ModuleSpec.
MRO
See method resolution order.

132 Appendix A. Glossary

 Python Tutorial, Release 3.12.2

mutable
Mutable objects can change their value but keep their id(). See also immutable.
named tuple
The term “named tuple” applies to any type or class that inherits from tuple and whose indexable elements are also
accessible using named attributes. The type or class may have other features as well.
Several built-in types are named tuples, including the values returned by time.localtime() and os.
stat(). Another example is sys.float_info:

>>> sys.float_info[1] # indexed access

1024
>>> sys.float_info.max_exp # named field access
1024
>>> isinstance(sys.float_info, tuple) # kind of tuple
True

Some named tuples are built-in types (such as the above examples). Alternatively, a named tuple can be created
from a regular class definition that inherits from tuple and that defines named fields. Such a class can be written
by hand or it can be created with the factory function collections.namedtuple(). The latter technique
also adds some extra methods that may not be found in hand-written or built-in named tuples.
namespace
The place where a variable is stored. Namespaces are implemented as dictionaries. There are the local, global
and built-in namespaces as well as nested namespaces in objects (in methods). Namespaces support modularity by
preventing naming conflicts. For instance, the functions builtins.open and os.open() are distinguished
by their namespaces. Namespaces also aid readability and maintainability by making it clear which module im-
plements a function. For instance, writing random.seed() or itertools.islice() makes it clear that
those functions are implemented by the random and itertools modules, respectively.
namespace package
A PEP 420 package which serves only as a container for subpackages. Namespace packages may have no physical
representation, and specifically are not like a regular package because they have no __init__.py file.
See also module.
nested scope
The ability to refer to a variable in an enclosing definition. For instance, a function defined inside another function
can refer to variables in the outer function. Note that nested scopes by default work only for reference and not for
assignment. Local variables both read and write in the innermost scope. Likewise, global variables read and write
to the global namespace. The nonlocal allows writing to outer scopes.
new-style class
Old name for the flavor of classes now used for all class objects. In earlier Python versions, only new-style classes
could use Python’s newer, versatile features like __slots__, descriptors, properties, __getattribute__(),
class methods, and static methods.
object
Any data with state (attributes or value) and defined behavior (methods). Also the ultimate base class of any
new-style class.
package
A Python module which can contain submodules or recursively, subpackages. Technically, a package is a Python
module with a __path__ attribute.
See also regular package and namespace package.
parameter
A named entity in a function (or method) definition that specifies an argument (or in some cases, arguments) that
the function can accept. There are five kinds of parameter:

133
Python Tutorial, Release 3.12.2

• positional-or-keyword: specifies an argument that can be passed either positionally or as a keyword argument.
This is the default kind of parameter, for example foo and bar in the following:

def func(foo, bar=None): ...

• positional-only: specifies an argument that can be supplied only by position. Positional-only parameters can
be defined by including a / character in the parameter list of the function definition after them, for example
posonly1 and posonly2 in the following:

def func(posonly1, posonly2, /, positional_or_keyword): ...

• keyword-only: specifies an argument that can be supplied only by keyword. Keyword-only parameters can be
defined by including a single var-positional parameter or bare * in the parameter list of the function definition
before them, for example kw_only1 and kw_only2 in the following:

def func(arg, *, kw_only1, kw_only2): ...

• var-positional: specifies that an arbitrary sequence of positional arguments can be provided (in addition to any
positional arguments already accepted by other parameters). Such a parameter can be defined by prepending
the parameter name with *, for example args in the following:

def func(*args, **kwargs): ...

• var-keyword: specifies that arbitrarily many keyword arguments can be provided (in addition to any key-
word arguments already accepted by other parameters). Such a parameter can be defined by prepending the
parameter name with **, for example kwargs in the example above.
Parameters can specify both optional and required arguments, as well as default values for some optional arguments.
See also the argument glossary entry, the FAQ question on the difference between arguments and parameters, the
inspect.Parameter class, the function section, and PEP 362.
path entry
A single location on the import path which the path based finder consults to find modules for importing.
path entry finder
A finder returned by a callable on sys.path_hooks (i.e. a path entry hook) which knows how to locate modules
given a path entry.
See importlib.abc.PathEntryFinder for the methods that path entry finders implement.
path entry hook
A callable on the sys.path_hooks list which returns a path entry finder if it knows how to find modules on a
specific path entry.
path based finder
One of the default meta path finders which searches an import path for modules.
path-like object
An object representing a file system path. A path-like object is either a str or bytes object representing a path,
or an object implementing the os.PathLike protocol. An object that supports the os.PathLike protocol
can be converted to a str or bytes file system path by calling the os.fspath() function; os.fsdecode()
and os.fsencode() can be used to guarantee a str or bytes result instead, respectively. Introduced by PEP
519.
PEP
Python Enhancement Proposal. A PEP is a design document providing information to the Python community,
or describing a new feature for Python or its processes or environment. PEPs should provide a concise technical
specification and a rationale for proposed features.

134 Appendix A. Glossary