Comprehensive Python Cheatsheet

^{Download text file, Buy PDF, Fork me on GitHub or Check out FAQ.}

ToC = {
    '1. Collections': [List, Dictionary, Set, Tuple, Range, Enumerate, Iterator, Generator],
    '2. Types':       [Type, String, Regular_Exp, Format, Numbers, Combinatorics, Datetime],
    '3. Syntax':      [Args, Inline, Closure, Decorator, Class, Duck_Types, Enum, Exceptions],
    '4. System':      [Print, Input, Command_Line_Arguments, Open, Path, Command_Execution],
    '5. Data':        [JSON, Pickle, CSV, SQLite, Bytes, Struct, Array, MemoryView, Deque],
    '6. Advanced':    [Threading, Operator, Introspection, Metaprograming, Eval, Coroutine],
    '7. Libraries':   [Progress_Bar, Plot, Table, Curses, Logging, Scraping, Web, Profile,
                       NumPy, Image, Animation, Audio]
}

#Main

if __name__ == '__main__':     # Runs main() if file wasn't imported.
    main()

#List

<list> = <list>[from_inclusive : to_exclusive : ±step_size]

#Dictionary

<view> = <dict>.keys()                          # Coll. of keys that reflects changes.
<view> = <dict>.values()                        # Coll. of values that reflects changes.
<view> = <dict>.items()                         # Coll. of key-value tuples that reflects chgs.

Counter

>>> from collections import Counter
>>> colors = ['blue', 'red', 'blue', 'red', 'blue']
>>> counter = Counter(colors)
>>> counter['yellow'] += 1
Counter({'blue': 3, 'red': 2, 'yellow': 1})
>>> counter.most_common()[0]
('blue', 3)

#Set

<set> = set()

Frozen Set

• Is immutable and hashable.
• That means it can be used as a key in a dictionary or as an element in a set.

<frozenset> = frozenset(<collection>)

#Tuple

Tuple is an immutable and hashable list.

<tuple> = ()
<tuple> = (<el>, )
<tuple> = (<el_1>, <el_2> [, ...])

Named Tuple

Tuple's subclass with named elements.

>>> from collections import namedtuple
>>> Point = namedtuple('Point', 'x y')
>>> p = Point(1, y=2)
Point(x=1, y=2)
>>> p[0]
1
>>> p.x
1
>>> getattr(p, 'y')
2
>>> p._fields  # Or: Point._fields
('x', 'y')

#Range

<range> = range(to_exclusive)
<range> = range(from_inclusive, to_exclusive)
<range> = range(from_inclusive, to_exclusive, ±step_size)

#Enumerate

for i, el in enumerate(<collection> [, i_start]):
    ...

#Iterator

<iter> = iter(<collection>)                 # `iter(<iter>)` returns unmodified iterator.
<iter> = iter(<function>, to_exclusive)     # A sequence of return values until 'to_exclusive'.
<el>   = next(<iter> [, default])           # Raises StopIteration or returns 'default' on end.
<list> = list(<iter>)                       # Returns a list of iterator's remaining elements.

Itertools

from itertools import count, repeat, cycle, chain, islice

#Generator

• Any function that contains a yield statement returns a generator.
• Generators and iterators are interchangeable.

def count(start, step):
    while True:
        yield start
        start += step

#Type

• Everything is an object.
• Every object has a type.
• Type and class are synonymous.

<type> = type(<el>)                          # Or: <el>.__class__
<bool> = isinstance(<el>, <type>)            # Or: issubclass(type(<el>), <type>)

Some types do not have built-in names, so they must be imported:

from types import FunctionType, MethodType, LambdaType, GeneratorType

ABC

An abstract base class introduces virtual subclasses that don’t inherit from it, but are still recognized by isinstance() and issubclass().

>>> from collections.abc import Sequence, Collection, Iterable
>>> isinstance([1, 2, 3], Iterable)
True

#String

<str>  = <str>.strip()                       # Strips all whitespace characters from both ends.
<str>  = <str>.strip('<chars>')              # Strips all passed characters from both ends.

Property Methods

+---------------+----------+----------+----------+----------+----------+
|               | [ !#$%…] | [a-zA-Z] |  [¼½¾]   |  [²³¹]   |  [0-9]   |
+---------------+----------+----------+----------+----------+----------+
| isprintable() |   yes    |   yes    |   yes    |   yes    |   yes    |
| isalnum()     |          |   yes    |   yes    |   yes    |   yes    |
| isnumeric()   |          |          |   yes    |   yes    |   yes    |
| isdigit()     |          |          |          |   yes    |   yes    |
| isdecimal()   |          |          |          |          |   yes    |
+---------------+----------+----------+----------+----------+----------+

#Regex

import re
<str>   = re.sub(<regex>, new, text, count=0)  # Substitutes all occurrences with 'new'.
<list>  = re.findall(<regex>, text)            # Returns all occurrences as strings.
<list>  = re.split(<regex>, text, maxsplit=0)  # Use brackets in regex to include the matches.
<Match> = re.search(<regex>, text)             # Searches for first occurrence of the pattern.
<Match> = re.match(<regex>, text)              # Searches only at the beginning of the text.
<iter>  = re.finditer(<regex>, text)           # Returns all occurrences as match objects.

Match Object

<str>   = <Match>.group()                      # Returns the whole match. Also group(0).
<str>   = <Match>.group(1)                     # Returns part in the first bracket.
<tuple> = <Match>.groups()                     # Returns all bracketed parts.
<int>   = <Match>.start()                      # Returns start index of the match.
<int>   = <Match>.end()                        # Returns exclusive end index of the match.

Special Sequences

• By default digits, whitespaces and alphanumerics from all alphabets are matched, unless 'flags=re.ASCII' argument is used.
• Use a capital letter for negation.

'\d' == '[0-9]'                                # Matches any digit.
'\w' == '[a-zA-Z0-9_]'                         # Matches any alphanumeric.
'\s' == '[ \t\n\r\f\v]'                        # Matches any whitespace.

#Format

<str> = f'{<el_1>}, {<el_2>}'
<str> = '{}, {}'.format(<el_1>, <el_2>)

Attributes

>>> from collections import namedtuple
>>> Person = namedtuple('Person', 'name height')
>>> person = Person('Jean-Luc', 187)
>>> f'{person.height}'
'187'
>>> '{p.height}'.format(p=person)
'187'

General Options

{<el>:<10}                                     # '<el>      '
{<el>:^10}                                     # '   <el>   '
{<el>:>10}                                     # '      <el>'
{<el>:.<10}                                    # '<el>......'
{<el>:<0}                                      # '<el>'

Strings

'!r' calls object's repr() method, instead of str(), to get a string.

{'abcde'!r:10}                                 # "'abcde'   "
{'abcde':10.3}                                 # 'abc       '
{'abcde':.3}                                   # 'abc'

Numbers

{ 123456:10,}                                  # '   123,456'
{ 123456:10_}                                  # '   123_456'
{ 123456:+10}                                  # '   +123456'
{-123456:=10}                                  # '-   123456'
{ 123456: }                                    # ' 123456'
{-123456: }                                    # '-123456'

Floats

{1.23456:10.3}                                 # '      1.23'
{1.23456:10.3f}                                # '     1.235'
{1.23456:10.3e}                                # ' 1.235e+00'
{1.23456:10.3%}                                # '  123.456%'

Comparison of presentation types:

+---------------+-----------------+-----------------+-----------------+-----------------+
|               |     {<real>}    |    {<real>:f}   |    {<real>:e}   |    {<real>:%}   |
+---------------+-----------------+-----------------+-----------------+-----------------+
|   0.000056789 |    '5.6789e-05' |     '0.000057'  |  '5.678900e-05' |     '0.005679%' |
|   0.00056789  |    '0.00056789' |     '0.000568'  |  '5.678900e-04' |     '0.056789%' |
|   0.0056789   |    '0.0056789'  |     '0.005679'  |  '5.678900e-03' |     '0.567890%' |
|   0.056789    |    '0.056789'   |     '0.056789'  |  '5.678900e-02' |     '5.678900%' |
|   0.56789     |    '0.56789'    |     '0.567890'  |  '5.678900e-01' |    '56.789000%' |
|   5.6789      |    '5.6789'     |     '5.678900'  |  '5.678900e+00' |   '567.890000%' |
|  56.789       |   '56.789'      |    '56.789000'  |  '5.678900e+01' |  '5678.900000%' |
| 567.89        |  '567.89'       |   '567.890000'  |  '5.678900e+02' | '56789.000000%' |
+---------------+-----------------+-----------------+-----------------+-----------------+

Ints

{90:c}                                # 'Z'
{90:b}                                # '1011010'
{90:X}                                # '5A'

#Numbers

<int>      = int(<float/str/bool>)    # Or: math.floor(<float>)
<float>    = float(<int/str/bool>)    # Or: <real>e±<int>
<complex>  = complex(real=0, imag=0)  # Or: <real> ± <real>j
<Fraction> = fractions.Fraction(numerator=0, denominator=1)
<Decimal>  = decimal.Decimal(<str/int/float>)

Basic Functions

<num> = pow(<num>, <num>)             # Or: <num> ** <num>
<num> = abs(<num>)                    # <float> = abs(<complex>)
<num> = round(<num> [, ±ndigits])     # `round(126, -1) == 130`

Math

from math import e, pi, inf, nan
from math import cos, acos, sin, asin, tan, atan, degrees, radians
from math import log, log10, log2

Statistics

from statistics import mean, median, variance, pvariance, pstdev

Random

from random import random, randint, choice, shuffle
<float> = random()
<int>   = randint(from_inclusive, to_inclusive)
<el>    = choice(<list>)
shuffle(<list>)

Bin, Hex

<int>     = 0b<bin>                   # Or: 0x<hex>
<int>     = int('<bin>', 2)           # Or: int('<hex>', 16)
<int>     = int('0b<bin>', 0)         # Or: int('0x<hex>', 0)
'0b<bin>' = bin(<int>)                # Or: hex(<int>)

Bitwise Operators

<int>     = <int> & <int>             # And
<int>     = <int> | <int>             # Or
<int>     = <int> ^ <int>             # Xor (0 if both bits equal)
<int>     = <int> << n_bits           # Shift left
<int>     = <int> >> n_bits           # Shift right
<int>     = ~<int>                    # Compliment (flips bits)

#Combinatorics

• Every function returns an iterator.
• If you want to print the iterator, you need to pass it to the list() function!

from itertools import product, combinations, combinations_with_replacement, permutations

#Datetime

• Module 'datetime' provides 'date' <D>, 'time' <T>, 'datetime' <DT> and 'timedelta' <TD> classes. All are immutable and hashable.
• Time and datetime objects can be 'aware' <a>, meaning they have defined timezone, or 'naive' <n>, meaning they don't.
• If object is naive, it is presumed to be in the system's timezone.

from datetime import date, time, datetime, timedelta
from dateutil.tz import UTC, tzlocal, gettz, resolve_imaginary

Constructors

<D>  = date(year, month, day)
<T>  = time(hour=0, minute=0, second=0, microsecond=0, tzinfo=None, fold=0)
<DT> = datetime(year, month, day, hour=0, minute=0, second=0, ...)
<TD> = timedelta(days=0, seconds=0, microseconds=0, milliseconds=0,
                 minutes=0, hours=0, weeks=0)

Now

<D/DTn>  = D/DT.today()                     # Current local date or naive datetime.
<DTn>    = DT.utcnow()                      # Naive datetime from current UTC time.
<DTa>    = DT.now(<tzinfo>)                 # Aware datetime from current tz time.

Timezone

<tzinfo> = UTC                              # UTC timezone. London without DST.
<tzinfo> = tzlocal()                        # Local timezone. Also gettz().
<tzinfo> = gettz('<Cont.>/<City>')          # 'Continent/City_Name' timezone or None.
<DTa>    = <DT>.astimezone(<tzinfo>)        # Datetime, converted to passed timezone.
<Ta/DTa> = <T/DT>.replace(tzinfo=<tzinfo>)  # Unconverted object with new timezone.

Encode

<D/T/DT> = D/T/DT.fromisoformat('<iso>')    # Object from ISO string. Raises ValueError.
<DT>     = DT.strptime(<str>, '<format>')   # Datetime from str, according to format.
<D/DTn>  = D/DT.fromordinal(<int>)          # D/DTn from days since Christ, at midnight.
<DTn>    = DT.fromtimestamp(<real>)         # Local time DTn from seconds since Epoch.
<DTa>    = DT.fromtimestamp(<real>, <tz.>)  # Aware datetime from seconds since Epoch.

Decode

<str>    = <D/T/DT>.isoformat(sep='T')      # Also timespec='auto/hours/minutes/seconds'.
<str>    = <D/T/DT>.strftime('<format>')    # Custom string representation.
<int>    = <D/DT>.toordinal()               # Days since Christ, ignoring time and tz.
<float>  = <DTn>.timestamp()                # Seconds since Epoch, from DTn in local tz.
<float>  = <DTa>.timestamp()                # Seconds since Epoch, from DTa.

Format

>>> from datetime import datetime
>>> dt = datetime.strptime('2015-05-14 23:39:00.00 +0200', '%Y-%m-%d %H:%M:%S.%f %z')
>>> dt.strftime("%A, %dth of %B '%y, %I:%M%p %Z")
"Thursday, 14th of May '15, 11:39PM UTC+02:00"

Arithmetics

<D/DT>   = <D/DT>   ± <TD>                  # Returned datetime can fall into missing hour.
<TD>     = <D/DTn>  - <D/DTn>               # Returns the difference, ignoring time jumps.
<TD>     = <DTa>    - <DTa>                 # Ignores time jumps if they share tzinfo object.
<TD>     = <DT_UTC> - <DT_UTC>              # Convert DTs to UTC to get the actual delta.

#Arguments

<function>(<positional_args>)                  # f(0, 0)
<function>(<keyword_args>)                     # f(x=0, y=0)
<function>(<positional_args>, <keyword_args>)  # f(0, y=0)

Inside Function Definition

def f(<nondefault_args>):                      # def f(x, y):
def f(<default_args>):                         # def f(x=0, y=0):
def f(<nondefault_args>, <default_args>):      # def f(x, y=0):

#Splat Operator

args   = (1, 2)
kwargs = {'x': 3, 'y': 4, 'z': 5}
func(*args, **kwargs)

Is the same as:

func(1, 2, x=3, y=4, z=5)

Inside Function Definition

Splat combines zero or more positional arguments into a tuple, while splatty-splat combines zero or more keyword arguments into a dictionary.

def add(*a):
    return sum(a)

Legal argument combinations:

def f(x, y, z):                # f(x=1, y=2, z=3) | f(1, y=2, z=3) | f(1, 2, z=3) | f(1, 2, 3)
def f(*, x, y, z):             # f(x=1, y=2, z=3)
def f(x, *, y, z):             # f(x=1, y=2, z=3) | f(1, y=2, z=3)
def f(x, y, *, z):             # f(x=1, y=2, z=3) | f(1, y=2, z=3) | f(1, 2, z=3)

Other Uses

<list>  = [*<collection> [, ...]]
<set>   = {*<collection> [, ...]}
<tuple> = (*<collection>, [...])
<dict>  = {**<dict> [, ...]}

#Inline

<function> = lambda: <return_value>
<function> = lambda <argument_1>, <argument_2>: <return_value>

Comprehension

<list> = [i+1 for i in range(10)]                   # [1, 2, ..., 10]
<set>  = {i for i in range(10) if i > 5}            # {6, 7, 8, 9}
<iter> = (i+5 for i in range(10))                   # (5, 6, ..., 14)
<dict> = {i: i*2 for i in range(10)}                # {0: 0, 1: 2, ..., 9: 18}

Is the same as:

out = []
for i in range(10):
    for j in range(10):
        out.append(i+j)

Map, Filter, Reduce

from functools import reduce
<iter> = map(lambda x: x + 1, range(10))            # (1, 2, ..., 10)
<iter> = filter(lambda x: x > 5, range(10))         # (6, 7, 8, 9)
<obj>  = reduce(lambda out, x: out + x, range(10))  # 45

Any, All

<bool> = any(<collection>)                          # False if empty.
<bool> = all(el[1] for el in <collection>)          # True if empty.

If - Else

<obj> = <expression_if_true> if <condition> else <expression_if_false>

Namedtuple, Enum, Dataclass

from collections import namedtuple
Point     = namedtuple('Point', 'x y')
point     = Point(0, 0)

#Closure

We have a closure in Python when:

• A nested function references a value of its enclosing function and then
• the enclosing function returns the nested function.

def get_multiplier(a):
    def out(b):
        return a * b
    return out

Partial

from functools import partial
<function> = partial(<function> [, <arg_1>, <arg_2>, ...])

Nonlocal

If variable is being assigned to anywhere in the scope, it is regarded as a local variable, unless it is declared as a 'global' or a 'nonlocal'.

def get_counter():
    i = 0
    def out():
        nonlocal i
        i += 1
        return i
    return out

#Decorator

A decorator takes a function, adds some functionality and returns it.

@decorator_name
def function_that_gets_passed_to_decorator():
    ...

Debugger Example

Decorator that prints function's name every time it gets called.

from functools import wraps

def debug(func):
    @wraps(func)
    def out(*args, **kwargs):
        print(func.__name__)
        return func(*args, **kwargs)
    return out

@debug
def add(x, y):
    return x + y

LRU Cache

Decorator that caches function's return values. All function's arguments must be hashable.

from functools import lru_cache

@lru_cache(maxsize=None)
def fib(n):
    return n if n < 2 else fib(n-2) + fib(n-1)

Parametrized Decorator

A decorator that accepts arguments and returns a normal decorator that accepts a function.

from functools import wraps

def debug(print_result=False):
    def decorator(func):
        @wraps(func)
        def out(*args, **kwargs):
            result = func(*args, **kwargs)
            print(func.__name__, result if print_result else '')
            return result
        return out
    return decorator

@debug(print_result=True)
def add(x, y):
    return x + y

#Class

class <name>:
    def __init__(self, a):
        self.a = a
    def __repr__(self):
        class_name = self.__class__.__name__
        return f'{class_name}({self.a!r})'
    def __str__(self):
        return str(self.a)

    @classmethod
    def get_class_name(cls):
        return cls.__name__

Str() use cases:

print(<el>)
print(f'{<el>}')
raise Exception(<el>)
loguru.logger.debug(<el>)
csv.writer(<file>).writerow([<el>])

Repr() use cases:

print([<el>])
print(f'{<el>!r}')
>>> <el>
loguru.logger.exception()
Z = dataclasses.make_dataclass('Z', ['a']); print(Z(<el>))

Constructor Overloading

class <name>:
    def __init__(self, a=None):
        self.a = a

Inheritance

class Person:
    def __init__(self, name, age):
        self.name = name
        self.age  = age

class Employee(Person):
    def __init__(self, name, age, staff_num):
        super().__init__(name, age)
        self.staff_num = staff_num

Multiple Inheritance

class A: pass
class B: pass
class C(A, B): pass

Property

Pythonic way of implementing getters and setters.

class MyClass:
    @property
    def a(self):
        return self._a

    @a.setter
    def a(self, value):
        self._a = value

Dataclass

Decorator that automatically generates init(), repr() and eq() special methods.

from dataclasses import dataclass, field

@dataclass(order=False, frozen=False)
class <class_name>:
    <attr_name_1>: <type>
    <attr_name_2>: <type> = <default_value>
    <attr_name_3>: list/dict/set = field(default_factory=list/dict/set)

Inline:

from dataclasses import make_dataclass
<class> = make_dataclass('<class_name>', <coll_of_attribute_names>)
<class> = make_dataclass('<class_name>', <coll_of_tuples>)
<tuple> = ('<attr_name>', <type> [, <default_value>])

Slots

Mechanism that restricts objects to attributes listed in 'slots' and significantly reduces their memory footprint.

class MyClassWithSlots:
    __slots__ = ['a']
    def __init__(self):
        self.a = 1

Copy

from copy import copy, deepcopy
<object> = copy(<object>)
<object> = deepcopy(<object>)

#Duck Types

A duck type is an implicit type that prescribes a set of special methods. Any object that has those methods defined is considered a member of that duck type.

class MyComparable:
    def __init__(self, a):
        self.a = a
    def __eq__(self, other):
        if isinstance(other, type(self)):
            return self.a == other.a
        return NotImplemented

Hashable

• Hashable object needs both hash() and eq() methods and its hash value should never change.
• Hashable objects that compare equal must have the same hash value, meaning default hash() that returns 'id(self)' will not do.
• That is why Python automatically makes classes unhashable if you only implement eq().

class MyHashable:
    def __init__(self, a):
        self._a = copy.deepcopy(a)
    @property
    def a(self):
        return self._a
    def __eq__(self, other):
        if isinstance(other, type(self)):
            return self.a == other.a
        return NotImplemented
    def __hash__(self):
        return hash(self.a)

Sortable

• With total_ordering decorator you only need to provide eq() and one of lt(), gt(), le() or ge() special methods.

from functools import total_ordering

@total_ordering
class MySortable:
    def __init__(self, a):
        self.a = a
    def __eq__(self, other):
        if isinstance(other, type(self)):
            return self.a == other.a
        return NotImplemented
    def __lt__(self, other):
        if isinstance(other, type(self)):
            return self.a < other.a
        return NotImplemented

Iterator

• Any object that has methods next() and iter() is an iterator.
• Next() should return next item or raise StopIteration.
• Iter() should return 'self'.

class Counter:
    def __init__(self):
        self.i = 0
    def __next__(self):
        self.i += 1
        return self.i
    def __iter__(self):
        return self

Python has many different iterator objects:

• Iterators returned by the iter() function, such as list_iterator and set_iterator.
• Objects returned by the itertools module, such as count, repeat and cycle.
• Generators returned by the generator functions and generator expressions.
• File objects returned by the open() function, etc.

class Counter:
    def __init__(self):
        self.i = 0
    def __call__(self):
        self.i += 1
        return self.i

Context Manager

• Enter() should lock the resources and (optionally) return an object.
• Exit() should release the resources.
• Any exception that happens inside the with block is passed to the exit() method.
• If it wishes to suppress the exception it must return a true value.

class MyOpen():
    def __init__(self, filename):
        self.filename = filename
    def __enter__(self):
        self.file = open(self.filename)
        return self.file
    def __exit__(self, exc_type, exception, traceback):
        self.file.close()

#Iterable Duck Types

class MyIterable:
    def __init__(self, a):
        self.a = a
    def __iter__(self):
        for el in self.a:
            yield el

Collection

• Only required methods are iter() and len().
• This cheatsheet actually means '<iterable>' when it uses '<collection>'.
• I chose not to use the name 'iterable' because it sounds scarier and more vague than 'collection'.

class MyCollection:
    def __init__(self, a):
        self.a = a
    def __iter__(self):
        return iter(self.a)
    def __contains__(self, el):
        return el in self.a
    def __len__(self):
        return len(self.a)

Sequence

• Only required methods are len() and getitem().
• Getitem() should return an item at index or raise IndexError.
• Iter() and contains() automatically work on any object that has getitem() defined.
• Reversed() automatically works on any object that has getitem() and len() defined.

class MySequence:
    def __init__(self, a):
        self.a = a
    def __iter__(self):
        return iter(self.a)
    def __contains__(self, el):
        return el in self.a
    def __len__(self):
        return len(self.a)
    def __getitem__(self, i):
        return self.a[i]
    def __reversed__(self):
        return reversed(self.a)

Collections.abc.Sequence

• It's a richer interface than the basic sequence.
• Extending it generates iter(), contains(), reversed(), index() and count().
• Unlike 'abc.Iterable' and 'abc.Collection', it is not a duck type. That is why 'issubclass(MySequence, abc.Sequence)' would return False even if MySequence had all the methods defined.

from collections import abc

class MyAbcSequence(abc.Sequence):
    def __init__(self, a):
        self.a = a
    def __len__(self):
        return len(self.a)
    def __getitem__(self, i):
        return self.a[i]

Table of required and automatically available special methods:

+------------+------------+------------+------------+--------------+
|            |  Iterable  | Collection |  Sequence  | abc.Sequence |
+------------+------------+------------+------------+--------------+
| iter()     |    REQ     |    REQ     |    Yes     |     Yes      |
| contains() |    Yes     |    Yes     |    Yes     |     Yes      |
| len()      |            |    REQ     |    REQ     |     REQ      |
| getitem()  |            |            |    REQ     |     REQ      |
| reversed() |            |            |    Yes     |     Yes      |
| index()    |            |            |            |     Yes      |
| count()    |            |            |            |     Yes      |
+------------+------------+------------+------------+--------------+

#Enum

from enum import Enum, auto

class <enum_name>(Enum):
    <member_name_1> = <value_1>
    <member_name_2> = <value_2_a>, <value_2_b>
    <member_name_3> = auto()

Inline

Cutlery = Enum('Cutlery', 'fork knife spoon')
Cutlery = Enum('Cutlery', ['fork', 'knife', 'spoon'])
Cutlery = Enum('Cutlery', {'fork': 1, 'knife': 2, 'spoon': 3})

User-defined functions cannot be values, so they must be wrapped:

from functools import partial
LogicOp = Enum('LogicOp', {'AND': partial(lambda l, r: l and r),
                           'OR' : partial(lambda l, r: l or r)})

#Exceptions

try:
    <code>
except <exception>:
    <code>

Complex Example

try:
    <code_1>
except <exception_a>:
    <code_2_a>
except <exception_b>:
    <code_2_b>
else:
    <code_2_c>
finally:
    <code_3>

Catching Exceptions

except <exception>:
except <exception> as <name>:
except (<exception>, ...):
except (<exception>, ...) as <name>:

Raising Exceptions

raise <exception>
raise <exception>()
raise <exception>(<el> [, ...])

Re-raising caught exception:

except <exception> as <name>:
    ...
    raise

Exception Object

arguments   = <name>.args
exc_type    = <name>.__class__
filename    = <name>.__traceback__.tb_frame.f_code.co_filename
func_name   = <name>.__traceback__.tb_frame.f_code.co_name
line        = linecache.getline(filename, <name>.__traceback__.tb_lineno)
error_msg   = traceback.format_exception(exc_type, <name>, <name>.__traceback__)

Built-in Exceptions

BaseException
 +-- SystemExit                   # Raised by the sys.exit() function.
 +-- KeyboardInterrupt            # Raised when the user hits the interrupt key (ctrl-c).
 +-- Exception                    # User-defined exceptions should be derived from this class.
      +-- StopIteration           # Raised by next() when run on an empty iterator.
      +-- ArithmeticError         # Base class for arithmetic errors.
      |    +-- ZeroDivisionError  # Raised when dividing by zero.
      +-- AttributeError          # Raised when an attribute is missing.
      +-- EOFError                # Raised by input() when it hits end-of-file condition.
      +-- LookupError             # Raised when a look-up on a collection fails.
      |    +-- IndexError         # Raised when a sequence index is out of range.
      |    +-- KeyError           # Raised when a dictionary key or set element is not found.
      +-- NameError               # Raised when a variable name is not found.
      +-- OSError                 # Failures such as “file not found” or “disk full”.
      |    +-- FileNotFoundError  # When a file or directory is requested but doesn't exist.
      +-- RuntimeError            # Raised by errors that don't fall in other categories.
      |    +-- RecursionError     # Raised when the maximum recursion depth is exceeded.
      +-- TypeError               # Raised when an argument is of wrong type.
      +-- ValueError              # When an argument is of right type but inappropriate value.
           +-- UnicodeError       # Raised when encoding/decoding strings from/to bytes fails.

Collections and their exceptions:

+-----------+------------+------------+------------+
|           |    list    |    dict    |    set     |
+-----------+------------+------------+------------+
| getitem() | IndexError |  KeyError  |            |
| pop()     | IndexError |  KeyError  |  KeyError  |
| remove()  | ValueError |            |  KeyError  |
| index()   | ValueError |            |            |
+-----------+------------+------------+------------+

Useful built-in exceptions:

raise TypeError('Argument is of wrong type!')
raise ValueError('Argument is of right type but inappropriate value!')
raise RuntimeError('None of above!')

User-defined Exceptions

class MyError(Exception):
    pass

class MyInputError(MyError):
    pass

#Exit

Exits the interpreter by raising SystemExit exception.

import sys
sys.exit()                        # Exits with exit code 0 (success).
sys.exit(<int>)                   # Exits with passed exit code.
sys.exit(<obj>)                   # Prints the object and exits with 1 (failure).

#Print

print(<el_1>, ..., sep=' ', end='\n', file=sys.stdout, flush=False)

Pretty Print

from pprint import pprint
pprint(<collection>, width=80, depth=None, compact=False, sort_dicts=True)

#Input

Reads a line from user input or pipe if present.

<str> = input(prompt=None)

#Command Line Arguments

import sys
script_name = sys.argv[0]
arguments   = sys.argv[1:]

Argparse

from argparse import ArgumentParser, FileType
p = ArgumentParser(description=<str>)
p.add_argument('-<short_name>', '--<name>', action='store_true')  # Flag
p.add_argument('-<short_name>', '--<name>', type=<type>)          # Option
p.add_argument('<name>', type=<type>, nargs=1)                    # First argument
p.add_argument('<name>', type=<type>, nargs='+')                  # Remaining arguments
p.add_argument('<name>', type=<type>, nargs='*')                  # Optional arguments
args  = p.parse_args()                                            # Exits on error.
value = args.<name>

#Open

Opens the file and returns a corresponding file object.

<file> = open('<path>', mode='r', encoding=None, newline=None)

Modes

• 'r' - Read (default).
• 'w' - Write (truncate).
• 'x' - Write or fail if the file already exists.
• 'a' - Append.
• 'w+' - Read and write (truncate).
• 'r+' - Read and write from the start.
• 'a+' - Read and write from the end.
• 't' - Text mode (default).
• 'b' - Binary mode.

<file>.seek(0)                      # Moves to the start of the file.
<file>.seek(offset)                 # Moves 'offset' chars/bytes from the start.
<file>.seek(0, 2)                   # Moves to the end of the file.
<bin_file>.seek(±offset, <anchor>)  # Anchor: 0 start, 1 current pos., 2 end.

Read Text from File

def read_file(filename):
    with open(filename, encoding='utf-8') as file:
        return file.readlines()

Write Text to File

def write_to_file(filename, text):
    with open(filename, 'w', encoding='utf-8') as file:
        file.write(text)

#Path

from os import getcwd, path, listdir
from glob import glob

DirEntry

Using scandir() instead of listdir() can significantly increase the performance of code that also needs file type information.

from os import scandir

Path Object

from pathlib import Path

#OS Commands

import os, shutil

Shell Commands

import os
<str> = os.popen('<shell_command>').read()

Sends '1 + 1' to the calculator and captures its output:

>>> from subprocess import run
>>> run('bc', input='1 + 1\n', capture_output=True, encoding='utf-8')
CompletedProcess(args='bc', returncode=0, stdout='2\n', stderr='')

Sends test.in to the calculator running in standard mode and saves its output to test.out:

>>> from shlex import split
>>> os.popen('echo 1 + 1 > test.in')
>>> run(split('bc -s'), stdin=open('test.in'), stdout=open('test.out', 'w'))
CompletedProcess(args=['bc', '-s'], returncode=0)
>>> open('test.out').read()
'2\n'

#JSON

Text file format for storing collections of strings and numbers.

import json
<str>    = json.dumps(<object>, ensure_ascii=True, indent=None)
<object> = json.loads(<str>)

Read Object from JSON File

def read_json_file(filename):
    with open(filename, encoding='utf-8') as file:
        return json.load(file)

Write Object to JSON File

def write_to_json_file(filename, an_object):
    with open(filename, 'w', encoding='utf-8') as file:
        json.dump(an_object, file, ensure_ascii=False, indent=2)

#Pickle

Binary file format for storing objects.

import pickle
<bytes>  = pickle.dumps(<object>)
<object> = pickle.loads(<bytes>)

Read Object from File

def read_pickle_file(filename):
    with open(filename, 'rb') as file:
        return pickle.load(file)

Write Object to File

def write_to_pickle_file(filename, an_object):
    with open(filename, 'wb') as file:
        pickle.dump(an_object, file)

#CSV

Text file format for storing spreadsheets.

import csv

Read

<reader> = csv.reader(<file>, dialect='excel', delimiter=',')
<list>   = next(<reader>)           # Returns next row as a list of strings.
<list>   = list(<reader>)           # Returns list of remaining rows.

Write

<writer> = csv.writer(<file>, dialect='excel', delimiter=',')
<writer>.writerow(<collection>)     # Encodes objects using `str(<el>)`.
<writer>.writerows(<coll_of_coll>)  # Appends multiple rows.

Parameters

• 'dialect' - Master parameter that sets the default values.
• 'delimiter' - A one-character string used to separate fields.
• 'quotechar' - Character for quoting fields that contain special characters.
• 'doublequote' - Whether quotechars inside fields get doubled or escaped.
• 'skipinitialspace' - Whether whitespace after delimiter gets stripped.
• 'lineterminator' - Specifies how writer terminates rows.
• 'quoting' - Controls the amount of quoting: 0 - as necessary, 1 - all.
• 'escapechar' - Character for escaping 'quotechar' if 'doublequote' is False.

+------------------+--------------+--------------+--------------+
|                  |     excel    |   excel-tab  |     unix     |
+------------------+--------------+--------------+--------------+
| delimiter        |       ','    |      '\t'    |       ','    |
| quotechar        |       '"'    |       '"'    |       '"'    |
| doublequote      |      True    |      True    |      True    |
| skipinitialspace |     False    |     False    |     False    |
| lineterminator   |    '\r\n'    |    '\r\n'    |      '\n'    |
| quoting          |         0    |         0    |         1    |
| escapechar       |      None    |      None    |      None    |
+------------------+--------------+--------------+--------------+

Read Rows from CSV File

def read_csv_file(filename):
    with open(filename, encoding='utf-8', newline='') as file:
        return list(csv.reader(file))

Write Rows to CSV File

def write_to_csv_file(filename, rows):
    with open(filename, 'w', encoding='utf-8', newline='') as file:
        writer = csv.writer(file)
        writer.writerows(rows)

#SQLite

Server-less database engine that stores each database into a separate file.

import sqlite3
db = sqlite3.connect('<path>')                  # Also ':memory:'.
...
db.close()

Read

Returned values can be of type str, int, float, bytes or None.

<cursor> = db.execute('<query>')                # Can raise sqlite3.OperationalError.
<tuple>  = <cursor>.fetchone()                  # Returns next row. Also next(<cursor>).
<list>   = <cursor>.fetchall()                  # Returns remaining rows.

Write

db.execute('<query>')
db.commit()

Or:

with db:
    db.execute('<query>')

Placeholders

• Passed values can be of type str, int, float, bytes, None, bool, datetime.date or datetime.datetme.
• Bools will be stored and returned as ints and dates as ISO formatted strings.

db.execute('<query>', <list/tuple>)             # Replaces '?'s in query with values.
db.execute('<query>', <dict/namedtuple>)        # Replaces ':<key>'s with values.
db.executemany('<query>', <coll_of_above>)      # Runs execute() many times.

Example

In this example values are not actually saved because 'db.commit()' is omitted!

>>> db = sqlite3.connect('test.db')
>>> db.execute('create table t (a, b, c)')
>>> db.execute('insert into t values (1, 2, 3)')
>>> db.execute('select * from t').fetchall()
[(1, 2, 3)]

MySQL

Has a very similar interface, with differences listed below.

# $ pip3 install mysql-connector
from mysql import connector
db = connector.connect(host=<str>, user=<str>, password=<str>, database=<str>)
<cursor> = db.cursor()
<cursor>.execute('<query>')                     # Only cursor has execute method.
<cursor>.execute('<query>', <list/tuple>)       # Replaces '%s's in query with values.
<cursor>.execute('<query>', <dict/namedtuple>)  # Replaces '%(<key>)s's with values.

#Bytes

Bytes object is an immutable sequence of single bytes. Mutable version is called bytearray.

<bytes> = b'<str>'                       # Only accepts ASCII characters and \x00 - \xff.
<int>   = <bytes>[<index>]               # Returns int in range from 0 to 255.
<bytes> = <bytes>[<slice>]               # Returns bytes even if it has only one element.
<bytes> = <bytes>.join(<coll_of_bytes>)  # Joins elements using bytes object as separator.

Encode

<bytes> = bytes(<coll_of_ints>)          # Ints must be in range from 0 to 255.
<bytes> = bytes(<str>, 'utf-8')          # Or: <str>.encode('utf-8')
<bytes> = <int>.to_bytes(n_bytes, byteorder='big/little', signed=False)
<bytes> = bytes.fromhex('<hex>')

Decode

<list>  = list(<bytes>)                  # Returns ints in range from 0 to 255.
<str>   = str(<bytes>, 'utf-8')          # Or: <bytes>.decode('utf-8')
<int>   = int.from_bytes(<bytes>, byteorder='big/little', signed=False)
'<hex>' = <bytes>.hex()

Read Bytes from File

def read_bytes(filename):
    with open(filename, 'rb') as file:
        return file.read()

Write Bytes to File

def write_bytes(filename, bytes_obj):
    with open(filename, 'wb') as file:
        file.write(bytes_obj)

#Struct

• Module that performs conversions between a sequence of numbers and a bytes object.
• Machine’s native type sizes and byte order are used by default.

from struct import pack, unpack, iter_unpack
<bytes>  = pack('<format>', <num_1> [, <num_2>, ...])
<tuple>  = unpack('<format>', <bytes>)
<tuples> = iter_unpack('<format>', <bytes>)

Example

>>> pack('>hhl', 1, 2, 3)
b'\x00\x01\x00\x02\x00\x00\x00\x03'
>>> unpack('>hhl', b'\x00\x01\x00\x02\x00\x00\x00\x03')
(1, 2, 3)

Format

For standard sizes start format string with:

• '=' - native byte order
• '<' - little-endian
• '>' - big-endian (also '!')

Integer types. Use a capital letter for unsigned type. Standard sizes are in brackets:

• 'x' - pad byte
• 'b' - char (1)
• 'h' - short (2)
• 'i' - int (4)
• 'l' - long (4)
• 'q' - long long (8)

Floating point types:

• 'f' - float (4)
• 'd' - double (8)

#Array

List that can only hold numbers of a predefined type. Available types and their sizes in bytes are listed above.

from array import array
<array> = array('<typecode>', <collection>)    # Array from coll. of numbers.
<array> = array('<typecode>', <bytes>)         # Array from bytes object.
<bytes> = bytes(<array>)                       # Or: <array>.tobytes()

#Memory View

• A sequence object that points to the memory of another object.
• Each element can reference a single or multiple consecutive bytes, depending on format.
• Order and number of elements can be changed with slicing.

<mview> = memoryview(<bytes/bytearray/array>)  # Immutable if bytes, else mutable.
<real>  = <mview>[<index>]                     # Returns an int or a float.
<mview> = <mview>[<slice>]                     # Mview with rearranged elements.
<mview> = <mview>.cast('<typecode>')           # Casts memoryview to the new format.
<mview>.release()                              # Releases the object's memory buffer.

Decode

<bin_file>.write(<mview>)                      # Writes mview to the binary file.
<bytes> = bytes(<mview>)                       # Creates a new bytes object.
<bytes> = <bytes>.join(<coll_of_mviews>)       # Joins mviews using bytes object as sep.
<list>  = list(<mview>)                        # Returns list of ints or floats.
<str>   = str(<mview>, 'utf-8')                # Treats mview as a bytes object.
<int>   = int.from_bytes(<mview>, byteorder='big/little', signed=False)
'<hex>' = <mview>.hex()

#Deque

A thread-safe list with efficient appends and pops from either side. Pronounced "deck".

from collections import deque
<deque> = deque(<collection>, maxlen=None)

#Threading

• CPython interpreter can only run a single thread at a time.
• That is why using multiple threads won't result in a faster execution, unless at least one of the threads contains an I/O operation.

from threading import Thread, RLock, Semaphore, Event, Barrier

Thread

thread = Thread(target=<function>, args=(<first_arg>, ))
thread.start()
...
<bool> = thread.is_alive()           # Checks if thread has finished executing.
thread.join()                        # Waits for thread to finish.

Lock

lock = RLock()
lock.acquire()                       # Waits for lock to be available.
...
lock.release()

Or:

lock = RLock()
with lock:
    ...

Semaphore, Event, Barrier

<Semaphore> = Semaphore(value=1)     # Lock that can be acquired 'value' times.
<Event>     = Event()                # Method wait() blocks until set() is called.
<Barrier>   = Barrier(n_times)       # Method wait() blocks until it's called n_times.

Thread Pool Executor

from concurrent.futures import ThreadPoolExecutor
with ThreadPoolExecutor(max_workers=None) as executor:
    <iter>   = executor.map(lambda x: x + 1, range(3))         # (1, 2, 3)
    <iter>   = executor.map(lambda x, y: x + y, 'abc', '123')  # ('a1', 'b2', 'c3')
    <Future> = executor.submit(<function> [, <arg_1>, ...])

Future:

<bool> = <Future>.done()             # Checks if thread has finished executing.
<obj>  = <Future>.result()           # Waits for thread to finish and returns result.

Queue

A thread-safe FIFO queue. For LIFO queue use LifoQueue.

from queue import Queue
<Queue> = Queue(maxsize=0)

#Operator

Module of functions that provide the functionality of operators.

from operator import add, sub, mul, truediv, floordiv, mod, pow, neg, abs
from operator import eq, ne, lt, le, gt, ge
from operator import and_, or_, not_
from operator import itemgetter, attrgetter, methodcaller

#Introspection

Inspecting code at runtime.

<list> = dir()                       # Returns names of local variables (incl. functions).
<dict> = vars()                      # Returns dict of local variables. Also locals().
<dict> = globals()                   # Returns dict of global variables.

Attributes

<list> = dir(<object>)               # Returns names of object's attributes (incl. methods).
<dict> = vars(<object>)              # Returns dict of object's fields. Also <obj>.__dict__.

Parameters

from inspect import signature
<sig>        = signature(<function>)
no_of_params = len(<sig>.parameters)
param_names  = list(<sig>.parameters.keys())
param_kinds  = [a.kind for a in <sig>.parameters.values()]

#Metaprograming

Code that generates code.

<class> = type('<class_name>', <parents_tuple>, <attributes_dict>)

Meta Class

A class that creates classes.

def my_meta_class(name, parents, attrs):
    attrs['a'] = 'abcde'
    return type(name, parents, attrs)

Or:

class MyMetaClass(type):
    def __new__(cls, name, parents, attrs):
        attrs['a'] = 'abcde'
        return type.__new__(cls, name, parents, attrs)

Metaclass Attribute

Right before a class is created it checks if it has a 'metaclass' attribute defined. If not, it recursively checks if any of his parents has it defined and eventually comes to type().

class MyClass(metaclass=MyMetaClass):
    b = 12345

Type Diagram

type(MyClass)     == MyMetaClass     # MyClass is an instance of MyMetaClass.
type(MyMetaClass) == type            # MyMetaClass is an instance of type.

Inheritance Diagram

MyClass.__base__     == object       # MyClass is a subclass of object.
MyMetaClass.__base__ == type         # MyMetaClass is a subclass of type.

#Eval

>>> from ast import literal_eval
>>> literal_eval('1 + 2')
3
>>> literal_eval('[1, 2, 3]')
[1, 2, 3]
>>> literal_eval('abs(1)')
ValueError: malformed node or string

#Coroutines

• Coroutines have a lot in common with threads, but unlike threads, they only give up control when they call another coroutine and they don’t use as much memory.
• Coroutine definition starts with 'async' and its call with 'await'.
• 'asyncio.run(<coroutine>)' is the main entry point for asynchronous programs.
• Functions wait(), gather() and as_completed() can be used when multiple coroutines need to be started at the same time.
• Asyncio module also provides it's own Queue, Event, Lock and Semaphore classes.

import asyncio, collections, curses, enum, random

P = collections.namedtuple('P', 'x y')         # Position
D = enum.Enum('D', 'n e s w')                  # Direction

def main(screen):
    curses.curs_set(0)                         # Makes cursor invisible.
    screen.nodelay(True)                       # Makes getch() non-blocking.
    asyncio.run(main_coroutine(screen))        # Starts running asyncio code.

async def main_coroutine(screen):
    state = {'*': P(0, 0), **{id_: P(30, 10) for id_ in range(10)}}
    moves = asyncio.Queue()
    coros = (*(random_controller(id_, moves) for id_ in range(10)),
             human_controller(screen, moves),
             model(moves, state, *screen.getmaxyx()),
             view(state, screen))
    await asyncio.wait(coros, return_when=asyncio.FIRST_COMPLETED)

async def random_controller(id_, moves):
    while True:
        moves.put_nowait((id_, random.choice(list(D))))
        await asyncio.sleep(random.random() / 2)

async def human_controller(screen, moves):
    while (ch := screen.getch()) != 27:
        key_mappings = {259: D.n, 261: D.e, 258: D.s, 260: D.w}
        if ch in key_mappings:
            moves.put_nowait(('*', key_mappings[ch]))
        await asyncio.sleep(0.01)  

async def model(moves, state, height, width):
    while state['*'] not in {p for id_, p in state.items() if id_ != '*'}:
        id_, d = await moves.get()
        p      = state[id_]
        deltas = {D.n: P(0, -1), D.e: P(1, 0), D.s: P(0, 1), D.w: P(-1, 0)}
        new_p  = P(*[sum(a) for a in zip(p, deltas[d])])
        if 0 <= new_p.x < width-1 and 0 <= new_p.y < height:
            state[id_] = new_p

async def view(state, screen):
    while True:
        screen.clear()
        for id_, p in state.items():
            screen.addstr(p.y, p.x, str(id_))
        await asyncio.sleep(0.01)  

curses.wrapper(main)

Libraries

# $ pip3 install tqdm
from tqdm import tqdm
from time import sleep
for el in tqdm([1, 2, 3]):
    sleep(0.2)

#Plot

# $ pip3 install matplotlib
from matplotlib import pyplot
pyplot.plot(<y_data> [, label=<str>])
pyplot.plot(<x_data>, <y_data>)
pyplot.legend()                                # Adds a legend.
pyplot.savefig(<filename>)                     # Saves the figure.
pyplot.show()                                  # Displays the figure.
pyplot.clf()                                   # Clears the figure.

#Table

# $ pip3 install tabulate
import csv, tabulate
with open('test.csv', encoding='utf-8', newline='') as file:
    rows   = csv.reader(file)
    header = [a.title() for a in next(rows)]
    table  = tabulate.tabulate(rows, header)
    print(table)

#Curses

from curses import wrapper, curs_set, ascii
from curses import KEY_UP, KEY_RIGHT, KEY_DOWN, KEY_LEFT

def main():
    wrapper(draw)

def draw(screen):
    curs_set(0)                                # Makes cursor invisible.
    screen.nodelay(True)                       # Makes getch() non-blocking.
    screen.clear()
    screen.addstr(0, 0, 'Press ESC to quit.')
    while screen.getch() != ascii.ESC:
        pass

def get_border(screen):
    from collections import namedtuple
    P = namedtuple('P', 'x y')
    height, width = screen.getmaxyx()
    return P(width - 1, height - 1)

if __name__ == '__main__':
    main()

#Logging

# $ pip3 install loguru
from loguru import logger

Exceptions

Exception description, stack trace and values of variables are appended automatically.

try:
    ...
except <exception>:
    logger.exception('An error happened.')

Rotation

Argument that sets a condition when a new log file is created.

rotation=<int>|<datetime.timedelta>|<datetime.time>|<str>

Retention

Sets a condition which old log files get deleted.

retention=<int>|<datetime.timedelta>|<str>

#Scraping

# $ pip3 install requests beautifulsoup4
import requests
from bs4 import BeautifulSoup
URL = 'https://en.wikipedia.org/wiki/Python_(programming_language)'
try:
    html  = requests.get(URL).text
    doc   = BeautifulSoup(html, 'html.parser')
    table = doc.find('table', class_='infobox vevent')
    rows  = table.find_all('tr')
    link  = rows[11].find('a')['href']
    ver   = rows[6].find('div').text.split()[0]
    url_i = rows[0].find('img')['src']
    image = requests.get(f'https:{url_i}').content
    with open('test.png', 'wb') as file:
        file.write(image)
    print(link, ver)
except requests.exceptions.ConnectionError:
    print("You've got problems with connection.")

#Web

# $ pip3 install bottle
from bottle import run, route, static_file, template, post, request, response
import json

Run

run(host='localhost', port=8080)        # Runs locally.
run(host='0.0.0.0', port=80)            # Runs globally.

Static Request

@route('/img/<image>')
def send_image(image):
    return static_file(image, 'img_dir/', mimetype='image/png')

Dynamic Request

@route('/<sport>')
def send_page(sport):
    return template('<h1>{{title}}</h1>', title=sport)

REST Request

@post('/odds/<sport>')
def odds_handler(sport):
    team = request.forms.get('team')
    home_odds, away_odds = 2.44, 3.29
    response.headers['Content-Type'] = 'application/json'
    response.headers['Cache-Control'] = 'no-cache'
    return json.dumps([team, home_odds, away_odds])

Test:

# $ pip3 install requests
>>> import requests
>>> url  = 'http://localhost:8080/odds/football'
>>> data = {'team': 'arsenal f.c.'}
>>> response = requests.post(url, data=data)
>>> response.json()
['arsenal f.c.', 2.44, 3.29]

#Profiling

from time import time
start_time = time()                     # Seconds since the Epoch.
...
duration = time() - start_time

High performance:

from time import perf_counter
start_time = perf_counter()             # Seconds since restart.
...
duration = perf_counter() - start_time

Timing a Snippet

>>> from timeit import timeit
>>> timeit('"-".join(str(a) for a in range(100))',
...        number=10000, globals=globals(), setup='pass')
0.34986

Profiling by Line

# $ pip3 install line_profiler memory_profiler
@profile
def main():
    a = [*range(10000)]
    b = {*range(10000)}
main()

Call Graph

# $ pip3 install pycallgraph
from pycallgraph import output, PyCallGraph
from datetime import datetime
time_str = datetime.now().strftime('%Y%m%d%H%M%S')
filename = f'profile-{time_str}.png'
drawer = output.GraphvizOutput(output_file=filename)
with PyCallGraph(drawer):
    <code_to_be_profiled>

#NumPy

Array manipulation mini-language. It can run up to one hundred times faster than the equivalent Python code.

# $ pip3 install numpy
import numpy as np

Indexing

<el>       = <2d_array>[0, 0]        # First element.
<1d_view>  = <2d_array>[0]           # First row.
<1d_view>  = <2d_array>[:, 0]        # First column. Also [..., 0].
<3d_view>  = <2d_array>[None, :, :]  # Expanded by dimension of size 1.

Broadcasting

Broadcasting is a set of rules by which NumPy functions operate on arrays of different sizes and/or dimensions.

left  = [[0.1], [0.6], [0.8]]        # Shape: (3, 1)
right = [ 0.1 ,  0.6 ,  0.8 ]        # Shape: (3)

1. If array shapes differ in length, left-pad the shorter shape with ones:

left  = [[0.1], [0.6], [0.8]]        # Shape: (3, 1)
right = [[0.1 ,  0.6 ,  0.8]]        # Shape: (1, 3) <- !

2. If any dimensions differ in size, expand the ones that have size 1 by duplicating their elements:

left  = [[0.1, 0.1, 0.1], [0.6, 0.6, 0.6], [0.8, 0.8, 0.8]]  # Shape: (3, 3) <- !
right = [[0.1, 0.6, 0.8], [0.1, 0.6, 0.8], [0.1, 0.6, 0.8]]  # Shape: (3, 3) <- !

3. If neither non-matching dimension has size 1, raise an error.

>>> points = np.array([0.1, 0.6, 0.8])
 [ 0.1,  0.6,  0.8]
>>> wrapped_points = points.reshape(3, 1)
[[ 0.1],
 [ 0.6],
 [ 0.8]]
>>> distances = wrapped_points - points
[[ 0. , -0.5, -0.7],
 [ 0.5,  0. , -0.2],
 [ 0.7,  0.2,  0. ]]
>>> distances = np.abs(distances)
[[ 0. ,  0.5,  0.7],
 [ 0.5,  0. ,  0.2],
 [ 0.7,  0.2,  0. ]]
>>> i = np.arange(3)
[0, 1, 2]
>>> distances[i, i] = np.inf
[[ inf,  0.5,  0.7],
 [ 0.5,  inf,  0.2],
 [ 0.7,  0.2,  inf]]
>>> distances.argmin(1)
[1, 2, 1]

#Image

# $ pip3 install pillow
from PIL import Image

Modes

• '1' - 1-bit pixels, black and white, stored with one pixel per byte.
• 'L' - 8-bit pixels, greyscale.
• 'RGB' - 3x8-bit pixels, true color.
• 'RGBA' - 4x8-bit pixels, true color with transparency mask.
• 'HSV' - 3x8-bit pixels, Hue, Saturation, Value color space.

WIDTH, HEIGHT = 100, 100
size = WIDTH * HEIGHT
hues = [255 * i/size for i in range(size)]
img = Image.new('HSV', (WIDTH, HEIGHT))
img.putdata([(int(h), 255, 255) for h in hues])
img.convert('RGB').save('test.png')

Adds noise to a PNG image:

from random import randint
add_noise = lambda value: max(0, min(255, value + randint(-20, 20)))
img = Image.open('test.png').convert('HSV')
img.putdata([(add_noise(h), s, v) for h, s, v in img.getdata()])
img.convert('RGB').save('test.png')

Drawing

from PIL import ImageDraw

#Animation

# $ pip3 install pillow imageio
from PIL import Image, ImageDraw
import imageio
WIDTH, R = 126, 10
frames = []
for velocity in range(15):
    y = sum(range(velocity+1))
    frame = Image.new('L', (WIDTH, WIDTH))
    draw  = ImageDraw.Draw(frame)
    draw.ellipse((WIDTH/2-R, y, WIDTH/2+R, y+2*R), fill='white')
    frames.append(frame)
frames += reversed(frames[1:-1])
imageio.mimsave('test.gif', frames, duration=0.03)

#Audio

import wave

Sample Values

+-----------+-------------+------+-------------+
| sampwidth |     min     | zero |     max     |
+-----------+-------------+------+-------------+
|     1     |           0 |  128 |         255 |
|     2     |      -32768 |    0 |       32767 |
|     3     |    -8388608 |    0 |     8388607 |
|     4     | -2147483648 |    0 |  2147483647 |
+-----------+-------------+------+-------------+

Read Float Samples from WAV File

def read_wav_file(filename):
    def get_int(a_bytes):
        an_int = int.from_bytes(a_bytes, 'little', signed=width!=1)
        return an_int - 128 * (width == 1)
    with wave.open(filename, 'rb') as file:
        width  = file.getsampwidth()
        frames = file.readframes(file.getnframes())
    byte_samples = (frames[i: i + width] for i in range(0, len(frames), width))
    return [get_int(b) / pow(2, width * 8 - 1) for b in byte_samples]

Write Float Samples to WAV File

def write_to_wav_file(filename, float_samples, nchannels=1, sampwidth=2, framerate=44100):
    def get_bytes(a_float):
        a_float = max(-1, min(1 - 2e-16, a_float))
        a_float += sampwidth == 1
        a_float *= pow(2, sampwidth * 8 - 1)
        return int(a_float).to_bytes(sampwidth, 'little', signed=sampwidth!=1) 
    with wave.open(filename, 'wb') as file:
        file.setnchannels(nchannels)
        file.setsampwidth(sampwidth)
        file.setframerate(framerate)
        file.writeframes(b''.join(get_bytes(f) for f in float_samples))

Examples

from math import pi, sin
samples_f = (sin(i * 2 * pi * 440 / 44100) for i in range(100000))
write_to_wav_file('test.wav', samples_f)

Adds noise to a mono WAV file:

from random import random
add_noise = lambda value: value + (random() - 0.5) * 0.03
samples_f = (add_noise(f) for f in read_wav_file('test.wav'))
write_to_wav_file('test.wav', samples_f)

Plays a WAV file:

# $ pip3 install simpleaudio
from simpleaudio import play_buffer
with wave.open('test.wav', 'rb') as file:
    p = file.getparams()
    frames = file.readframes(p.nframes)
    play_buffer(frames, p.nchannels, p.sampwidth, p.framerate)

Text to Speech

# $ pip3 install pyttsx3
import pyttsx3
engine = pyttsx3.init()
engine.say('Sally sells seashells by the seashore.')
engine.runAndWait()

#Synthesizer

# $ pip3 install simpleaudio
import simpleaudio, math, struct
from itertools import chain, repeat
F  = 44100
P1 = '71♪,69,,71♪,66,,62♪,66,,59♪,,,'
P2 = '71♪,73,,74♪,73,,74,,71,,73♪,71,,73,,69,,71♪,69,,71,,67,,71♪,,,'
get_pause   = lambda seconds: repeat(0, int(seconds * F))
sin_f       = lambda i, hz: math.sin(i * 2 * math.pi * hz / F)
get_wave    = lambda hz, seconds: (sin_f(i, hz) for i in range(int(seconds * F)))
get_hz      = lambda key: 8.176 * 2 ** (int(key) / 12)
parse_note  = lambda note: (get_hz(note[:2]), 0.25 if '♪' in note else 0.125)
get_samples = lambda note: get_wave(*parse_note(note)) if note else get_pause(0.125)
samples_f   = chain.from_iterable(get_samples(n) for n in f'{P1}{P1}{P2}'.split(','))
samples_b   = b''.join(struct.pack('<h', int(f * 30000)) for f in samples_f)
simpleaudio.play_buffer(samples_b, 1, 2, F)

#Basic Script Template

#!/usr/bin/env python3
#
# Usage: .py
#

from collections import namedtuple
from dataclasses import make_dataclass
from enum import Enum
from sys import argv
import re


def main():
    pass


###
##  UTIL
#

def read_file(filename):
    with open(filename, encoding='utf-8') as file:
        return file.readlines()


if __name__ == '__main__':
    main()