<list>.sort() # Sorts in ascending order. <list>.reverse() # Reverses the list in-place. <list> = sorted(<collection>) # Returns a new sorted list. <iter> = reversed(<list>) # Returns reversed iterator.
1 2 3 4 5 6 7 8
sum_of_elements = sum(<collection>) elementwise_sum = [sum(pair) for pair in zip(list_a, list_b)] sorted_by_second = sorted(<collection>, key=lambda el: el[1]) sorted_by_both = sorted(<collection>, key=lambda el: (el[1], el[0])) flatter_list = list(itertools.chain.from_iterable(<list>)) product_of_elems = functools.reduce(lambda out, el: out * el, <collection>) list_of_chars = list(<str>)
For details about sorted(), min() and max() see sortable.
Module operator provides functions
itemgetter() and mul() that offer the same functionality as lambda expressions above.
1 2 3 4 5 6 7
<list>.insert(<int>, <el>) # Inserts item at index and moves the rest to the right. <el> = <list>.pop([<int>]) # Removes and returns item at index or from the end. <int> = <list>.count(<el>) # Returns number of occurrences. Also works on strings. <int> = <list>.index(<el>) # Returns index of the first occurrence or raises ValueError. <list>.remove(<el>) # Removes first occurrence of the item or raises ValueError. <list>.clear() # Removes all items. Also works on dictionary and set.
Dictionary
1 2 3 4
<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.
1 2 3 4 5
value = <dict>.get(key, default=None) # Returns default if key is missing. value = <dict>.setdefault(key, default=None) # Returns and writes default if key is missing. <dict> = collections.defaultdict(<type>) # Returns a dict with default value of type. <dict> = collections.defaultdict(lambda: 1) # Returns a dict with default value 1.
1 2 3 4
<dict> = dict(<collection>) # Creates a dict from coll. of key-value pairs. <dict> = dict(zip(keys, values)) # Creates a dict from two collections. <dict> = dict.fromkeys(keys [, value]) # Creates a dict from collection of keys.
1 2 3 4 5
<dict>.update(<dict>) # Adds items. Replaces ones with matching keys. value = <dict>.pop(key) # Removes item or raises KeyError. {k for k, v in <dict>.items() if v == value} # Returns set of keys that point to the value. {k: v for k, v in <dict>.items() if k in keys} # Returns a dictionary, filtered by keys.
for i, el in enumerate(<collection> [, i_start]): ...
Iterator
1 2 3 4 5
<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
1 2
import itertools as it
1 2 3 4
<iter> = it.count(start=0, step=1) # Returns updated value endlessly. Accepts floats. <iter> = it.repeat(<el> [, times]) # Returns element endlessly or 'times' times. <iter> = it.cycle(<collection>) # Repeats the sequence endlessly.
1 2 3
<iter> = it.chain(<coll>, <coll> [, ...]) # Empties collections in order (figuratively). <iter> = it.chain.from_iterable(<coll>) # Empties collections inside a collection in order.
1 2 3
<iter> = it.islice(<coll>, to_exclusive) # Only returns first 'to_exclusive' elements. <iter> = it.islice(<coll>, from_inc, …) # `to_exclusive, +step_size`. Indices can be None.
Generator
Any function that contains a yield statement returns a
generator.
Generators and iterators are interchangeable.
1 2 3 4 5
def count(start, step): while True: yield start start += step
Some
types do not have built-in names, so they must be imported:
1 2
from types import FunctionType, MethodType, LambdaType, GeneratorType, ModuleType
Abstract Base Classes
Each abstract base class specifies a set of virtual
subclasses. These classes are then recognized by isinstance() and
issubclass() as subclasses of the ABC, although they are really not. ABC
can also manually decide whether or not a specific class is its virtual
subclass, usually based on which methods the class has implemented. For
instance, Iterable ABC looks for method iter(), while Collection ABC
looks for iter(), contains() and len().
<str> = <str>.strip() # Strips all whitespace characters from both ends. <str> = <str>.strip('<chars>') # Strips all passed characters from both ends.
1 2 3 4 5
<list> = <str>.split() # Splits on one or more whitespace characters. <list> = <str>.split(sep=None, maxsplit=-1) # Splits on 'sep' str at most 'maxsplit' times. <list> = <str>.splitlines(keepends=False) # On [\n\r\f\v\x1c-\x1e\x85\u2028\u2029] and \r\n. <str> = <str>.join(<coll_of_strings>) # Joins elements using string as a separator.
1 2 3 4 5 6
<bool> = <sub_str> in <str> # Checks if string contains a substring. <bool> = <str>.startswith(<sub_str>) # Pass tuple of strings for multiple options. <bool> = <str>.endswith(<sub_str>) # Pass tuple of strings for multiple options. <int> = <str>.find(<sub_str>) # Returns start index of the first match or -1. <int> = <str>.index(<sub_str>) # Same, but raises ValueError if missing.
1 2 3
<str> = <str>.replace(old, new [, count]) # Replaces 'old' with 'new' at most 'count' times. <str> = <str>.translate(<table>) # Use `str.maketrans(<dict>)` to generate table.
1 2 3
<str> = chr(<int>) # Converts int to Unicode character. <int> = ord(<str>) # Converts Unicode character to int.
Also: 'lstrip()', 'rstrip()' and
'rsplit()'.
Also: 'lower()', 'upper()',
'capitalize()' and 'title()'.
Also: 'isspace()' checks for
'[ \t\n\r\f\v\x1c-\x1f\x85\u2000…]'.
Regex
1 2 3 4 5 6 7 8
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.
Argument 'new' can be a function that accepts a Match object
and returns a string.
Search() and match() return None if they can't find a
match.
Argument 'flags=re.IGNORECASE' can be used with
all functions.
Argument 'flags=re.MULTILINE' makes
'^' and '$' match the start/end of each
line.
Argument 'flags=re.DOTALL' makes dot also
accept the '\n'.
Use r'\1' or '\\1' for
backreference ('\1' returns a character with octal code
1).
Add '?' after '*' and
'+' to make them non-greedy.
Match Object
1 2 3 4 5 6
<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.
By default, decimal characters, alphanumerics and
whitespaces from all alphabets are matched unless
'flags=re.ASCII' argument is used.
As shown above, it restricts all special sequence matches to
the first 128 characters and prevents '\s' from accepting
'[\x1c-\x1f]' (the so-called separator
characters).
Use a capital letter for negation (all non-ASCII characters
will be matched when used in combination with ASCII flag).
Format
1 2 3 4
<str> = f'{<el_1>}, {<el_2>}' # Curly brackets can also contain expressions. <str> = '{}, {}'.format(<el_1>, <el_2>) # Or: '{0}, {a}'.format(<el_1>, a=<el_2>) <str> = '%s, %s' % (<el_1>, <el_2>) # Redundant and inferior C style formatting.
Attributes
1 2 3 4 5 6 7
>>> Person = collections.namedtuple('Person', 'name height') >>> person = Person('Jean-Luc', 187) >>> f'{person.height}' '187' >>> '{p.height}'.format(p=person) '187'
When both rounding up and rounding down are possible, the
one that returns result with even last digit is chosen. That makes
'{6.5:.0f}' a '6' and '{7.5:.0f}'
an '8'.
This rule only effects numbers that can be represented
exactly by a float (.5, .25, …).
from math import e, pi, inf, nan, isinf, isnan # `<el> == nan` is always False. from math import sin, cos, tan, asin, acos, atan # Also: degrees, radians. from math import log, log10, log2 # Log can accept base as second arg.
>>> it.product('abc', 'abc') # a b c [('a', 'a'), ('a', 'b'), ('a', 'c'), # a x x x ('b', 'a'), ('b', 'b'), ('b', 'c'), # b x x x ('c', 'a'), ('c', 'b'), ('c', 'c')] # c x x x
1 2 3 4
>>> it.combinations('abc', 2) # a b c [('a', 'b'), ('a', 'c'), # a . x x ('b', 'c')] # b . . x
1 2 3 4 5
>>> it.combinations_with_replacement('abc', 2) # a b c [('a', 'a'), ('a', 'b'), ('a', 'c'), # a x x x ('b', 'b'), ('b', 'c'), # b . x x ('c', 'c')] # c . . x
1 2 3 4 5
>>> it.permutations('abc', 2) # a b c [('a', 'b'), ('a', 'c'), # a . x x ('b', 'a'), ('b', 'c'), # b x . x ('c', 'a'), ('c', 'b')] # c x x .
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.
1 2 3
from datetime import date, time, datetime, timedelta from dateutil.tz import UTC, tzlocal, gettz, datetime_exists, resolve_imaginary
Use '<D/DT>.weekday()' to get the day of
the week as an int, with Monday being 0.
'fold=1' means the second pass in case of time
jumping back for one hour.
Timedelta normalizes arguments to ±days, seconds
(<86 400) and microseconds (< 1M).
Now
1 2 3 4
<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.
To extract time use '<DTn>.time()',
'<DTa>.time()' or
'<DTa>.timetz()'.
Timezone
1 2 3 4 5 6
<tzinfo> = UTC # UTC timezone. London without DST. <tzinfo> = tzlocal() # Local timezone. Also gettz(). <tzinfo> = gettz('<Continent>/<City>') # 'Continent/City_Name' timezone or None. <DTa> = <DT>.astimezone(<tzinfo>) # Datetime, converted to the passed timezone. <Ta/DTa> = <T/DT>.replace(tzinfo=<tzinfo>) # Unconverted object with a new timezone.
Encode
1 2 3 4 5 6
<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 the Gregorian NYE 1. <DTn> = DT.fromtimestamp(<real>) # Local time DTn from seconds since the Epoch. <DTa> = DT.fromtimestamp(<real>, <tz.>) # Aware datetime from seconds since the Epoch.
ISO strings come in following forms:
'YYYY-MM-DD', 'HH:MM:SS.mmmuuu[±HH:MM]', or
both separated by an arbitrary character. All parts following hours are
optional.
<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 Gregorian NYE 1, ignoring time and tz. <float> = <DTn>.timestamp() # Seconds since the Epoch, from DTn in local tz. <float> = <DTa>.timestamp() # Seconds since the Epoch, from aware datetime.
Format
1 2 3 4
>>> dt = datetime.strptime('2015-05-14 23:39:00.00 +2000', '%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"
'%Z' only accepts 'UTC/GMT' and
local timezone's code. '%z' also accepts
'±HH:MM'.
For abbreviated weekday and month use '%a' and
'%b'.
Arithmetics
1 2 3 4 5 6
<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> = <TD> * <real> # Also: <TD> = abs(<TD>) and <TD> = <TD> ±% <TD>. <float> = <TD> / <TD> # How many weeks/years there are in TD. Also //.
head, *body, tail = <coll.> # Head or tail can be omitted.
Inline
Lambda
1 2 3
<func> = lambda: <return_value> # A single statement function. <func> = lambda <arg_1>, <arg_2>: <return_value> # Also accepts default arguments.
Comprehensions
1 2 3 4 5
<list> = [i+1 for i in range(10)] # Or: [1, 2, ..., 10] <iter> = (i for i in range(10) if i > 5) # Or: iter([6, 7, 8, 9]) <set> = {i+5 for i in range(10)} # Or: {5, 6, ..., 14} <dict> = {i: i*2 for i in range(10)} # Or: {0: 0, 1: 2, ..., 9: 18}
1 2 3
>>> [l+r for l in 'abc' for r in 'abc'] ['aa', 'ab', 'ac', ..., 'cc']
Reduce must be imported from the functools
module.
Any, All
1 2 3
<bool> = any(<collection>) # Is `bool(el)` True for any element. <bool> = all(<collection>) # Is True for all elements or empty.
Conditional Expression
1 2
<obj> = <exp> if <condition> else <exp> # Only one expression gets evaluated.
1 2 3
>>> [a if a else 'zero' for a in (0, 1, 2, 3)] ['zero', 1, 2, 3]
Named Tuple, Enum, Dataclass
1 2 3 4
from collections import namedtuple Point = namedtuple('Point', 'x y') # Creates a tuple's subclass. point = Point(0, 0) # Returns its instance.
1 2 3 4
from enum import Enum Direction = Enum('Direction', 'n e s w') # Creates an enum. direction = Direction.n # Returns its member.
1 2 3 4
from dataclasses import make_dataclass Player = make_dataclass('Player', ['loc', 'dir']) # Creates a class. player = Player(point, direction) # Returns its instance.
Imports
1 2 3 4
import <module> # Imports a built-in or '<module>.py'. import <package> # Imports a built-in or '<package>/__init__.py'. import <package>.<module> # Imports a built-in or '<package>/<module>.py'.
Package is a collection of modules, but it can also define
its own objects.
On a filesystem this corresponds to a directory of Python
files with an optional init script.
Running 'import <package>' does not
automatically provide access to the package's modules unless they are
explicitly imported in its init script.
Closure
We have/get a closure in Python when:
A nested function references a value of its enclosing
function and then
the enclosing function returns the nested
function.
1 2 3 4 5
def get_multiplier(a): def out(b): return a * b return out
Wraps is a helper decorator that copies the metadata of the
passed function (func) to the function it is wrapping
(out).
Without it 'add.__name__' would return
'out'.
LRU Cache
Decorator that caches function's return values. All
function's arguments must be hashable.
1 2 3 4 5 6
from functools import lru_cache
@lru_cache(maxsize=None) def fib(n): return n if n < 2 else fib(n-2) + fib(n-1)
Default size of the cache is 128 values. Passing
'maxsize=None' makes it unbounded.
CPython interpreter limits recursion depth to 1000 by
default. To increase it use
'sys.setrecursionlimit(<depth>)'.
Parametrized Decorator
A decorator that accepts arguments and returns a normal
decorator that accepts a function.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
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
Using only '@debug' to decorate the add()
function would not work here, because debug would then receive the add()
function as a 'print_result' argument. Decorators can however manually
check if the argument they received is a function and act
accordingly.
Class
1 2 3 4 5 6 7 8 9 10 11 12 13
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)
Objects can be made sortable with 'order=True'
and immutable with 'frozen=True'.
For object to be hashable, all attributes must be hashable
and 'frozen' must be True.
Function field() is needed because
'<attr_name>: list = []' would make a list that is
shared among all instances. Its 'default_factory' argument can be any callable.
For attributes of arbitrary type use
'typing.Any'.
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.
Comparable
If eq() method is not overridden, it returns
'id(self) == id(other)', which is the same as
'self is other'.
That means all objects compare not equal by
default.
Only the left side object has eq() method called, unless it
returns NotImplemented, in which case the right object is consulted.
False is returned if both return NotImplemented.
Ne() automatically works on any object that has eq()
defined.
1 2 3 4 5 6 7 8
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().
1 2 3 4 5 6 7 8 9 10 11 12 13
class MyHashable: def __init__(self, a): self._a = 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 and the rest
will be automatically generated.
Functions sorted() and min() only require lt() method, while
max() only requires gt(). However, it is best to define them all so that
confusion doesn't arise in other contexts.
When two lists, strings or dataclasses are compared, their
values get compared in order until a pair of unequal values is found.
The comparison of this two values is then returned. The shorter sequence
is considered smaller in case of all values being equal.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
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.
>>> with open('test.txt', 'w') as file: ... file.write('Hello World!') >>> with MyOpen('test.txt') as file: ... print(file.read()) Hello World!
Iterable
Duck Types
Iterable
Only required method is iter(). It should return an iterator
of object's items.
Contains() automatically works on any object that has iter()
defined.
1 2 3 4 5 6 7 8
class MyIterable: def __init__(self, a): self.a = a def __iter__(self): return iter(self.a) def __contains__(self, el): return el in self.a
1 2 3 4 5 6
>>> obj = MyIterable([1, 2, 3]) >>> [el for el in obj] [1, 2, 3] >>> 1 in obj True
Collection
Only required methods are iter() and len(). Len() should
return the number of items.
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'. The only drawback of this
decision is that a reader could think a certain function doesn't accept
iterators when it does, since iterators are the only built-in objects
that are iterable but are not collections.
1 2 3 4 5 6 7 8 9 10
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 the passed index or raise
IndexError.
Iter() and contains() automatically work on any object that
has getitem() defined.
Reversed() automatically works on any object that has len()
and getitem() defined.
1 2 3 4 5 6 7 8 9 10 11 12 13 14
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)
Discrepancies
between glossary definitions and abstract base classes:
Glossary defines iterable as any object with iter() or
getitem() and sequence as any object with getitem() and len(). It does
not define collection.
Passing ABC Iterable to isinstance() or issubclass() checks
whether object/class has method iter(), while ABC Collection checks for
iter(), contains() and len().
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. It however recognizes
list, tuple, range, str, bytes, bytearray, memoryview and deque, because
they are registered as Sequence's virtual subclasses.
1 2 3 4 5 6 7 8 9 10
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:
If there are no numeric values before auto(), it returns
1.
Otherwise it returns an increment of the last numeric
value.
1 2 3 4 5 6
<member> = <enum>.<member_name> # Returns a member. <member> = <enum>['<member_name>'] # Returns a member or raises KeyError. <member> = <enum>(<value>) # Returns a member or raises ValueError. <str> = <member>.name # Returns member's name. <obj> = <member>.value # Returns member's value.
1 2 3 4 5
list_of_members = list(<enum>) member_names = [a.name for a in <enum>] member_values = [a.value for a in <enum>] random_member = random.choice(list(<enum>))
1 2 3 4 5
def get_next_member(member): members = list(member.__class__) index = (members.index(member) + 1) % len(members) return members[index]
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. +-- ArithmeticError # Base class for arithmetic errors. | +-- ZeroDivisionError # Raised when dividing by zero. +-- AssertionError # Raised by `assert <exp>` if expression returns false value. +-- 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 missing. +-- MemoryError # Out of memory. Could be too late to start deleting vars. +-- NameError # Raised when an object is missing. +-- OSError # Errors such as “file not found” or “disk full” (see Open). | +-- FileNotFoundError # When a file or directory is requested but doesn't exist. +-- RuntimeError # Raised by errors that don't fall into other categories. | +-- RecursionError # Raised when the maximum recursion depth is exceeded. +-- StopIteration # Raised by next() when run on an empty iterator. +-- 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 to/from bytes fails.
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
1 2 3
class MyError(Exception): pass class MyInputError(MyError): pass
Exit
Exits the interpreter by raising SystemExit
exception.
1 2 3 4 5
import sys sys.exit() # Exits with exit code 0 (success). sys.exit(<el>) # Prints to stderr and exits with 1. sys.exit(<int>) # Exits with passed exit code.
'encoding=None' means that the default encoding
is used, which is platform dependent. Best practice is to use
'encoding="utf-8"' whenever possible.
'newline=None' means all different end of line
combinations are converted to '' on read, while on write all ''
characters are converted to system's default line
separator.
'newline=""' means no conversions take place,
but input is still broken into chunks by readline() and readlines() on
every '', ' and ''.
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 ('br',
'bw', 'bx', …).
Exceptions
'FileNotFoundError' can be raised when reading
with 'r' or 'r+'.
'FileExistsError' can be raised when writing
with 'x'.
'IsADirectoryError' and
'PermissionError' can be raised by any.
'OSError' is the parent class of all listed
exceptions.
File Object
1 2 3 4 5
<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 position, 2 end.
1 2 3 4 5
<str/bytes> = <file>.read(size=-1) # Reads 'size' chars/bytes or until EOF. <str/bytes> = <file>.readline() # Returns a line or empty string/bytes on EOF. <list> = <file>.readlines() # Returns a list of remaining lines. <str/bytes> = next(<file>) # Returns a line using buffer. Do not mix.
1 2 3 4
<file>.write(<str/bytes>) # Writes a string or bytes object. <file>.writelines(<collection>) # Writes a coll. of strings or bytes objects. <file>.flush() # Flushes write buffer. Runs every 4096/8192 B.
Methods do not add or strip trailing newlines, even
writelines().
Read
Text from File
1 2 3 4
def read_file(filename): with open(filename, encoding='utf-8') as file: return file.readlines()
Write Text
to File
1 2 3 4
def write_to_file(filename, text): with open(filename, 'w', encoding='utf-8') as file: file.write(text)
Paths
1 2 3
from os import getcwd, path, listdir, scandir from glob import glob
1 2 3 4
<str> = getcwd() # Returns the current working directory. <str> = path.join(<path>, ...) # Joins two or more pathname components. <str> = path.abspath(<path>) # Returns absolute path.
1 2 3 4
<str> = path.basename(<path>) # Returns final component of the path. <str> = path.dirname(<path>) # Returns path without the final component. <tup.> = path.splitext(<path>) # Splits on last period of the final component.
1 2 3
<list> = listdir(path='.') # Returns filenames located at path. <list> = glob('<pattern>') # Returns paths matching the wildcard pattern.
<stat> = os.stat(<path>) # Or: <DirEntry/Path>.stat() <real> = <stat>.st_mtime/st_size/… # Modification time, size in bytes, …
DirEntry
Unlike listdir(), scandir() returns DirEntry objects that
cache isfile, isdir and on Windows also stat information, thus
significantly increasing the performance of code that requires
it.
1 2 3 4 5
<iter> = scandir(path='.') # Returns DirEntry objects located at path. <str> = <DirEntry>.path # Returns whole path as a string. <str> = <DirEntry>.name # Returns final component as a string. <file> = open(<DirEntry>) # Opens the file and returns a file object.
Path Object
1 2
from pathlib import Path
1 2 3
<Path> = Path(<path> [, ...]) # Accepts strings, Paths and DirEntry objects. <Path> = <path> / <path> [/ ...] # First or second path must be a Path object.
1 2 3 4 5
<Path> = Path() # Returns relative cwd. Also Path('.'). <Path> = Path.cwd() # Returns absolute cwd. Also Path().resolve(). <Path> = Path.home() # Returns user's home directory (absolute). <Path> = Path(__file__).resolve() # Returns script's path if cwd wasn't changed.
1 2 3 4 5 6
<Path> = <Path>.parent # Returns Path without the final component. <str> = <Path>.name # Returns final component as a string. <str> = <Path>.stem # Returns final component without extension. <str> = <Path>.suffix # Returns final component's extension. <tup.> = <Path>.parts # Returns all components as strings.
1 2 3
<iter> = <Path>.iterdir() # Returns directory contents as Path objects. <iter> = <Path>.glob('<pattern>') # Returns Paths matching the wildcard pattern.
1 2 3
<str> = str(<Path>) # Returns path as a string. <file> = open(<Path>) # Also <Path>.read/write_text/bytes().
OS Commands
1 2
import os, shutil, subprocess
1 2 3 4
os.chdir(<path>) # Changes the current working directory. os.mkdir(<path>, mode=0o777) # Creates a directory. Permissions are in octal. os.makedirs(<path>, mode=0o777) # Creates all path's dirs. Also: `exist_ok=False`.
1 2 3
shutil.copy(from, to) # Copies the file. 'to' can exist or be a dir. shutil.copytree(from, to) # Copies the directory. 'to' must not exist.
1 2 3
os.rename(from, to) # Renames/moves the file or directory. os.replace(from, to) # Same, but overwrites 'to' if it exists.
1 2 3 4
os.remove(<path>) # Deletes the file. os.rmdir(<path>) # Deletes the empty directory. shutil.rmtree(<path>) # Deletes the directory.
Paths can be either strings, Paths or DirEntry
objects.
Functions report OS related errors by raising either OSError
or one of its subclasses.
Shell
Commands
1 2 3 4
<pipe> = os.popen('<command>') # Executes command in sh/cmd. Returns its stdout pipe. <str> = <pipe>.read(size=-1) # Reads 'size' chars or until EOF. Also readline/s(). <int> = <pipe>.close() # Closes the pipe. Returns None on success.
Sends
'1 + 1' to the basic calculator and captures its output:
Text file format for storing collections of strings and
numbers.
1 2 3 4
import json <str> = json.dumps(<object>) # Converts object to JSON string. <object> = json.loads(<str>) # Converts JSON string to object.
Read Object from JSON File
1 2 3 4
def read_json_file(filename): with open(filename, encoding='utf-8') as file: return json.load(file)
Write Object to JSON File
1 2 3 4
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 Python objects.
1 2 3 4
import pickle <bytes> = pickle.dumps(<object>) # Converts object to bytes object. <object> = pickle.loads(<bytes>) # Converts bytes object to object.
Read
Object from File
1 2 3 4
def read_pickle_file(filename): with open(filename, 'rb') as file: return pickle.load(file)
Write
Object to File
1 2 3 4
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.
1 2
import csv
Read
1 2 3 4
<reader> = csv.reader(<file>) # Also: `dialect='excel', delimiter=','`. <list> = next(<reader>) # Returns next row as a list of strings. <list> = list(<reader>) # Returns a list of remaining rows.
File must be opened with a 'newline=""'
argument, or newlines embedded inside quoted fields will not be
interpreted correctly!
To print the spreadsheet to the console use Tabulate library.
For XML and binary Excel files (xlsx, xlsm and xlsb) use Pandas library.
def read_csv_file(filename, dialect='excel'): with open(filename, encoding='utf-8', newline='') as file: return list(csv.reader(file, dialect))
Write Rows to CSV File
1 2 3 4 5
def write_to_csv_file(filename, rows, dialect='excel'): with open(filename, 'w', encoding='utf-8', newline='') as file: writer = csv.writer(file, dialect) writer.writerows(rows)
SQLite
A server-less database engine that stores each database into
a separate file.
1 2 3 4
import sqlite3 <conn> = sqlite3.connect(<path>) # Opens existing or new file. Also ':memory:'. <conn>.close() # Closes the connection.
Read
1 2 3 4
<cursor> = <conn>.execute('<query>') # Can raise a subclass of sqlite3.Error. <tuple> = <cursor>.fetchone() # Returns next row. Also next(<cursor>). <list> = <cursor>.fetchall() # Returns remaining rows. Also list(<cursor>).
Write
1 2 3 4
<conn>.execute('<query>') # Can raise a subclass of sqlite3.Error. <conn>.commit() # Saves all changes since the last commit. <conn>.rollback() # Discards all changes since the last commit.
Or:
1 2 3
with <conn>: # Exits the block with commit() or rollback(), <conn>.execute('<query>') # depending on whether any exception occurred.
Placeholders
1 2 3 4
<conn>.execute('<query>', <list/tuple>) # Replaces '?'s in query with values. <conn>.execute('<query>', <dict/namedtuple>) # Replaces ':<key>'s with values. <conn>.executemany('<query>', <coll_of_above>) # Runs execute() multiple times.
Passed values can be of type str, int, float, bytes, None,
bool, datetime.date or datetime.datetime.
Bytes object is an immutable sequence of single bytes.
Mutable version is called bytearray.
1 2 3 4 5
<bytes> = b'<str>' # Only accepts ASCII characters and \x00-\xff. <int> = <bytes>[<index>] # Returns an 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 as a separator.
Encode
1 2 3 4 5
<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='little/big', signed=False`. <bytes> = bytes.fromhex('<hex>') # Hex pairs can be separated by whitespaces.
Decode
1 2 3 4 5
<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='little/big', signed=False`. '<hex>' = <bytes>.hex() # Returns hex pairs. Accepts `sep=<str>`.
Read
Bytes from File
1 2 3 4
def read_bytes(filename): with open(filename, 'rb') as file: return file.read()
Write
Bytes to File
1 2 3 4
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.
System’s type sizes, byte order, and alignment rules are
used by default.
1 2 3 4
from struct import pack, unpack <bytes> = pack('<format>', <el_1> [, ...]) # Packages arguments into bytes object. <tuple> = unpack('<format>', <bytes>) # Use iter_unpack() for iterator of tuples.
For
standard type sizes and manual alignment (padding) start format string
with:
'=' - System's byte order (usually
little-endian).
'<' - Little-endian.
'>' - Big-endian (also
'!').
Besides
numbers, pack() and unpack() also support bytes objects as part of the
sequence:
'c' - A bytes object with a single element. For
pad byte use 'x'.
'<n>s' - A bytes object with n
elements.
Integer
types. Use a capital letter for unsigned type. Minimum and standard
sizes are in brackets:
'b' - char (1/1)
'h' - short (2/2)
'i' - int (2/4)
'l' - long (4/4)
'q' - long long (8/8)
Floating point types:
'f' - float (4/4)
'd' - double (8/8)
Array
List that can only hold numbers of a predefined type.
Available types and their minimum sizes in bytes are listed above. Sizes
and byte order are always determined by the system.
1 2 3 4 5 6 7
from array import array <array> = array('<typecode>', <collection>) # Array from collection of numbers. <array> = array('<typecode>', <bytes>) # Array from bytes object. <array> = array('<typecode>', <array>) # Treats array as a sequence of numbers. <bytes> = bytes(<array>) # Or: <array>.tobytes() <file>.write(<array>) # Writes array to the binary file.
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.
Casting only works between char and other types and uses
system's sizes.
Byte order is always determined by the system.
1 2 3 4 5 6
<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
1 2 3 4 5
<bytes> = bytes(<mview>) # Returns a new bytes object. <bytes> = <bytes>.join(<coll_of_mviews>) # Joins mviews using bytes object as sep. <array> = array('<typecode>', <mview>) # Treats mview as a sequence of numbers. <file>.write(<mview>) # Writes mview to the binary file.
1 2 3 4 5
<list> = list(<mview>) # Returns a list of ints or floats. <str> = str(<mview>, 'utf-8') # Treats mview as a bytes object. <int> = int.from_bytes(<mview>, …) # `byteorder='little/big', signed=False`. '<hex>' = <mview>.hex() # Treats mview as a bytes object.
Deque
A thread-safe list with efficient appends and pops from
either side. Pronounced "deck".
1 2 3
from collections import deque <deque> = deque(<collection>, maxlen=None)
1 2 3 4 5
<deque>.appendleft(<el>) # Opposite element is dropped if full. <deque>.extendleft(<collection>) # Collection gets reversed. <el> = <deque>.popleft() # Raises IndexError if empty. <deque>.rotate(n=1) # Rotates elements to the right.
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.
1 2 3
from threading import Thread, RLock, Semaphore, Event, Barrier from concurrent.futures import ThreadPoolExecutor
Thread
1 2 3 4 5
<Thread> = Thread(target=<function>) # Use `args=<collection>` to set the arguments. <Thread>.start() # Starts the thread. <bool> = <Thread>.is_alive() # Checks if the thread has finished executing. <Thread>.join() # Waits for the thread to finish.
Use 'kwargs=<dict>' to pass keyword
arguments to the function.
Use 'daemon=True', or the program will not be
able to exit while the thread is alive.
Lock
1 2 3 4
<lock> = RLock() # Lock that can only be released by acquirer. <lock>.acquire() # Waits for the lock to be available. <lock>.release() # Makes the lock available again.
Or:
1 2 3
with <lock>: # Enters the block by calling acquire(), ... # and exits it with release().
Semaphore, Event, Barrier
1 2 3 4
<Semaphore> = Semaphore(value=1) # Lock that can be acquired by 'value' threads. <Event> = Event() # Method wait() blocks until set() is called. <Barrier> = Barrier(n_times) # Wait() blocks until it's called n_times.
Thread
Pool Executor
Object that manages thread execution.
An object with the same interface called ProcessPoolExecutor
provides true parallelism by running a separate interpreter in each
process. All arguments must be pickable.
1 2 3
<Exec> = ThreadPoolExecutor(max_workers=None) # Or: `with ThreadPoolExecutor() as <name>: …` <Exec>.shutdown(wait=True) # Blocks until all threads finish executing.
1 2 3 4 5
<iter> = <Exec>.map(<func>, <args_1>, ...) # A multithreaded and non-lazy map(). <Futr> = <Exec>.submit(<func>, <arg_1>, ...) # Starts a thread and returns its Future object. <bool> = <Futr>.done() # Checks if the thread has finished executing. <obj> = <Futr>.result() # Waits for thread to finish and returns result.
Queue
A thread-safe FIFO queue. For LIFO queue use
LifoQueue.
1 2 3
from queue import Queue <Queue> = Queue(maxsize=0)
1 2 3 4 5
<Queue>.put(<el>) # Blocks until queue stops being full. <Queue>.put_nowait(<el>) # Raises queue.Full exception if full. <el> = <Queue>.get() # Blocks until queue stops being empty. <el> = <Queue>.get_nowait() # Raises queue.Empty exception if empty.
Operator
Module of functions that provide the functionality of
operators.
<list> = dir() # Names of local variables (incl. functions). <dict> = vars() # Dict of local variables. Also locals(). <dict> = globals() # Dict of global variables.
Attributes
1 2 3 4 5 6 7
<list> = dir(<object>) # Names of object's attributes (incl. methods). <dict> = vars(<object>) # Dict of writable attributes. Also <obj>.__dict__. <bool> = hasattr(<object>, '<attr_name>') # Checks if getattr() raises an AttributeError. value = getattr(<object>, '<attr_name>') # Raises AttributeError if attribute is missing. setattr(<object>, '<attr_name>', value) # Only works on objects with '__dict__' attribute. delattr(<object>, '<attr_name>') # Same. Also `del <object>.<attr_name>`.
Parameters
1 2 3 4 5 6
<Sig> = inspect.signature(<function>) # Function's Signature object. <dict> = <Sig>.parameters # Dict of Parameter objects. <memb> = <Param>.kind # Member of ParameterKind enum. <obj> = <Param>.default # Default value or <Param>.empty. <type> = <Param>.annotation # Type or <Param>.empty.
Metaprogramming
Code that generates code.
Type
Type is the root class. If only passed an object it returns
its type (class). Otherwise it creates a new class.
New() is a class method that gets called before init(). If
it returns an instance of its class, then that instance gets passed to
init() as a 'self' argument.
It receives the same arguments as init(), except for the
first one that specifies the desired type of the returned instance
(MyMetaClass in our case).
Like in our case, new() can also be called directly, usually
from a new() method of a child class
(def __new__(cls): return super().__new__(cls)).
The only difference between the examples above is that
my_meta_class() returns a class of type type, while MyMetaClass()
returns a class of type MyMetaClass.
Metaclass Attribute
Right before a class is created it checks if it has the
'metaclass' attribute defined. If not, it recursively checks if any of
his parents has it defined and eventually comes to type().
1 2 3
class MyClass(metaclass=MyMetaClass): b = 12345
1 2 3
>>> MyClass.a, MyClass.b ('abcde', 12345)
Type Diagram
1 2 3
type(MyClass) == MyMetaClass # MyClass is an instance of MyMetaClass. type(MyMetaClass) == type # MyMetaClass is an instance of type.
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() start multiple
coroutines at the same time.
# $ pip3 install tqdm >>> from tqdm import tqdm >>> from time import sleep >>> for el in tqdm([1, 2, 3], desc='Processing'): ... sleep(1) Processing: 100%|████████████████████| 3/3 [00:03<00:00, 1.00s/it]
Plot
1 2 3 4 5 6 7 8
# $ pip3 install matplotlib import matplotlib.pyplot as plt plt.plot(<x_data>, <y_data> [, label=<str>]) # Or: plt.plot(<y_data>) plt.legend() # Adds a legend. plt.savefig(<path>) # Saves the figure. plt.show() # Displays the figure. plt.clf() # Clears the figure.
import curses, curses.ascii, os from curses import A_REVERSE, KEY_DOWN, KEY_UP, KEY_LEFT, KEY_RIGHT, KEY_ENTER
def main(screen): ch, first, selected, paths = 0, 0, 0, os.listdir() while ch != curses.ascii.ESC: height, _ = screen.getmaxyx() screen.erase() for y, filename in enumerate(paths[first : first+height]): screen.addstr(y, 0, filename, A_REVERSE * (selected == first + y)) ch = screen.getch() selected += (ch == KEY_DOWN) - (ch == KEY_UP) selected = max(0, min(len(paths)-1, selected)) first += (first <= selected - height) - (first > selected) if ch in [KEY_LEFT, KEY_RIGHT, KEY_ENTER, 10, 13]: new_dir = '..' if ch == KEY_LEFT else paths[selected] if os.path.isdir(new_dir): os.chdir(new_dir) first, selected, paths = 0, 0, os.listdir()
if __name__ == '__main__': curses.wrapper(main)
Logging
1 2 3
# $ pip3 install loguru from loguru import logger
1 2 3 4
logger.add('debug_{time}.log', colorize=True) # Connects a log file. logger.add('error_{time}.log', level='ERROR') # Another file for errors or higher. logger.<level>('A logging message.') # Logs to file/s and prints to stderr.
from time import perf_counter start_time = perf_counter() ... duration_in_seconds = perf_counter() - start_time
Timing a
Snippet
1 2 3 4 5
>>> from timeit import timeit >>> timeit("''.join(str(i) for i in range(100))", ... number=10000, globals=globals(), setup='pass') 0.34986
Profiling by
Line
1 2 3 4 5 6 7
# $ pip3 install line_profiler memory_profiler @profile def main(): a = [*range(10000)] b = {*range(10000)} main()
1 2 3 4 5 6 7 8 9
$ kernprof -lv test.py Line # Hits Time Per Hit % Time Line Contents ======================================================= 1 @profile 2 def main(): 3 1 955.0 955.0 43.7 a = [*range(10000)] 4 1 1231.0 1231.0 56.3 b = {*range(10000)}
1 2 3 4 5 6 7 8 9
$ python3 -m memory_profiler test.py Line # Mem usage Increment Line Contents ======================================================= 1 37.668 MiB 37.668 MiB @profile 2 def main(): 3 38.012 MiB 0.344 MiB a = [*range(10000)] 4 38.477 MiB 0.465 MiB b = {*range(10000)}
Call Graph
Generates
a PNG image of the call graph with highlighted bottlenecks:
filename = f'profile-{datetime.datetime.now():%Y%m%d_%H%M%S}.png' drawer = cg.output.GraphvizOutput(output_file=filename) with cg.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. An even faster
alternative that runs on a GPU is called CuPy.
<view> = <array>.reshape(<shape>) # Also `<array>.shape = <shape>`. <array> = <array>.flatten() # Collapses array into one dimension. <view> = <array>.squeeze() # Removes dimensions of length one.
1 2 3 4
<array> = <array>.sum/min/mean/var/std(axis) # Passed dimension gets aggregated. <array> = <array>.argmin(axis) # Returns indexes of smallest elements. <array> = np.apply_along_axis(<func>, axis, <array>) # Func can return a scalar or array.
Shape is a tuple of dimension sizes. A 100x50 RGB image has
shape (50, 100, 3).
Axis is an index of the dimension that gets aggregated.
Leftmost dimension has index 0. Summing the RGB image along axis 2 will
return a greyscale image with shape (50, 100).
Passing a tuple of axes will chain the operations like this:
'<array>.<method>(axis_1, keepdims=True).<method>(axis_2).squeeze()'.
<Image> = Image.new('<mode>', (width, height)) # Also: `color=<int/tuple/str>`. <Image> = Image.open(<path>) # Identifies format based on file contents. <Image> = <Image>.convert('<mode>') # Converts image to the new mode. <Image>.save(<path>) # Selects format based on the path extension. <Image>.show() # Opens image in default preview app.
1 2 3 4 5 6
<int/tuple> = <Image>.getpixel((x, y)) # Returns a pixel. <Image>.putpixel((x, y), <int/tuple>) # Writes a pixel to the image. <ImagingCore> = <Image>.getdata() # Returns a flattened sequence of pixels. <Image>.putdata(<list/ImagingCore>) # Writes a flattened sequence of pixels. <Image>.paste(<Image>, (x, y)) # Writes passed image to the image.
1 2 3 4
<2d_array> = np.array(<Image_L>) # Creates NumPy array from greyscale image. <3d_array> = np.array(<Image_RGB/A>) # Creates NumPy array from color image. <Image> = Image.fromarray(np.uint8(<array>)) # Use <array>.clip(0, 255) to clip the values.
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.
Examples
Creates a PNG
image of a rainbow gradient:
1 2 3 4 5 6 7
WIDTH, HEIGHT = 100, 100 n_pixels = WIDTH * HEIGHT hues = (255 * i/n_pixels for i in range(n_pixels)) 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:
1 2 3 4 5 6
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')
Image Draw
1 2 3
from PIL import ImageDraw <ImageDraw> = ImageDraw.Draw(<Image>)
1 2 3 4 5 6 7
<ImageDraw>.point((x, y)) # Truncates floats into ints. <ImageDraw>.line((x1, y1, x2, y2 [, ...])) # To get anti-aliasing use Image's resize(). <ImageDraw>.arc((x1, y1, x2, y2), deg1, deg2) # Always draws in clockwise direction. <ImageDraw>.rectangle((x1, y1, x2, y2)) # To rotate use Image's rotate() and paste(). <ImageDraw>.polygon((x1, y1, x2, y2, ...)) # Last point gets connected to the first. <ImageDraw>.ellipse((x1, y1, x2, y2)) # To rotate use Image's rotate() and paste().
Use 'fill=<color>' to set the primary
color.
Use 'width=<int>' to set the width of
lines or contours.
Use 'outline=<color>' to set the color of
the contours.
Color can be an int, tuple, '#rrggbb[aa]'
string or a color name.
WIDTH, HEIGHT, R = 126, 126, 10 frames = [] for velocity in range(1, 16): y = sum(range(velocity)) frame = Image.new('L', (WIDTH, HEIGHT)) draw = ImageDraw.Draw(frame) draw.ellipse((WIDTH/2-R, y, WIDTH/2+R, y+R*2), fill='white') frames.append(frame) frames += reversed(frames[1:-1]) imageio.mimsave('test.gif', frames, duration=0.03)
Audio
1 2
import wave
1 2 3 4 5 6 7 8
<Wave_read> = wave.open('<path>', 'rb') # Opens the WAV file. framerate = <Wave_read>.getframerate() # Number of frames per second. nchannels = <Wave_read>.getnchannels() # Number of samples per frame. sampwidth = <Wave_read>.getsampwidth() # Sample size in bytes. nframes = <Wave_read>.getnframes() # Number of frames. <params> = <Wave_read>.getparams() # Immutable collection of above. <bytes> = <Wave_read>.readframes(nframes) # Returns next 'nframes' frames.
1 2 3 4 5 6 7
<Wave_write> = wave.open('<path>', 'wb') # Truncates existing file. <Wave_write>.setframerate(<int>) # 44100 for CD, 48000 for video. <Wave_write>.setnchannels(<int>) # 1 for mono, 2 for stereo. <Wave_write>.setsampwidth(<int>) # 2 for CD quality sound. <Wave_write>.setparams(<params>) # Sets all parameters. <Wave_write>.writeframes(<bytes>) # Appends frames to the file.
Bytes object contains a sequence of frames, each consisting
of one or more samples.
In a stereo signal, the first sample of a frame belongs to
the left channel.
Each sample consists of one or more bytes that, when
converted to an integer, indicate the displacement of a speaker membrane
at a given moment.
If sample width is one byte, then the integer should be
encoded unsigned.
For all other sizes, the integer should be encoded signed
with little-endian byte order.
def read_wav_file(filename): def get_int(bytes_obj): an_int = int.from_bytes(bytes_obj, 'little', signed=(sampwidth != 1)) return an_int - 128 * (sampwidth == 1) with wave.open(filename, 'rb') as file: sampwidth = file.getsampwidth() frames = file.readframes(-1) bytes_samples = (frames[i : i+sampwidth] for i in range(0, len(frames), sampwidth)) return [get_int(b) / pow(2, sampwidth * 8 - 1) for b in bytes_samples]
Write Float Samples to WAV
File
1 2 3 4 5 6 7 8 9 10 11 12
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
Saves a 440 Hz
sine wave to a mono WAV file:
1 2 3 4
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:
1 2 3 4 5
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:
1 2 3 4 5 6 7
# $ pip3 install simpleaudio from simpleaudio import play_buffer with wave.open('test.wav', 'rb') as file: p = file.getparams() frames = file.readframes(-1) play_buffer(frames, p.nchannels, p.sampwidth, p.framerate)
Text to
Speech
1 2 3 4 5 6
# $ pip3 install pyttsx3 import pyttsx3 engine = pyttsx3.init() engine.say('Sally sells seashells by the seashore.') engine.runAndWait()
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: it.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]), 1/4 if '♩' in note else 1/8) get_samples = lambda note: get_wave(*parse_note(note)) if note else get_pause(1/8) samples_f = it.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)
Pygame
1 2 3 4 5 6 7 8 9 10 11 12 13 14
# $ pip3 install pygame import pygame as pg
pg.init() screen = pg.display.set_mode((500, 500)) rect = pg.Rect(240, 240, 20, 20) while all(event.type != pg.QUIT for event in pg.event.get()): deltas = {pg.K_UP: (0, -1), pg.K_RIGHT: (1, 0), pg.K_DOWN: (0, 1), pg.K_LEFT: (-1, 0)} for ch, is_pressed in enumerate(pg.key.get_pressed()): rect = rect.move(deltas[ch]) if ch in deltas and is_pressed else rect screen.fill((0, 0, 0)) pg.draw.rect(screen, (255, 255, 255), rect) pg.display.flip()
Rectangle
Object for storing rectangular coordinates.
1 2 3 4 5
<Rect> = pg.Rect(x, y, width, height) # Floats get truncated into ints. <int> = <Rect>.x/y/centerx/centery/… # Top, right, bottom, left. Allows assignments. <tup.> = <Rect>.topleft/center/… # Topright, bottomright, bottomleft. Same. <Rect> = <Rect>.move((x, y)) # Use move_ip() to move in-place.
1 2 3 4 5
<bool> = <Rect>.collidepoint((x, y)) # Checks if rectangle contains a point. <bool> = <Rect>.colliderect(<Rect>) # Checks if two rectangles overlap. <int> = <Rect>.collidelist(<list_of_Rect>) # Returns index of first colliding Rect or -1. <list> = <Rect>.collidelistall(<list_of_Rect>) # Returns indexes of all colliding rectangles.
Surface
Object for representing images.
1 2 3 4 5
<Surf> = pg.display.set_mode((width, height)) # Returns a display surface. <Surf> = pg.Surface((width, height)) # New RGB surface. RGBA if `flags=pg.SRCALPHA`. <Surf> = pg.image.load('<path>') # Loads the image. Format depends on source. <Surf> = <Surf>.subsurface(<Rect>) # Returns a subsurface.
1 2 3 4
<Surf>.fill(color) # Tuple, Color('#rrggbb[aa]') or Color(<name>). <Surf>.set_at((x, y), color) # Updates pixel. <Surf>.blit(<Surf>, (x, y)) # Draws passed surface to the surface.
from pygame.draw import line, ... line(<Surf>, color, (x1, y1), (x2, y2), width) # Draws a line to the surface. arc(<Surf>, color, <Rect>, from_rad, to_rad) # Also: ellipse(<Surf>, color, <Rect>, width=0) rect(<Surf>, color, <Rect>, width=0) # Also: polygon(<Surf>, color, points, width=0)
Font
1 2 3 4
<Font> = pg.font.SysFont('<name>', size) # Loads the system font or default if missing. <Font> = pg.font.Font('<path>', size) # Loads the TTF file. Pass None for default. <Surf> = <Font>.render(text, antialias, color) # Background color can be specified at the end.
Sound
1 2 3
<Sound> = pg.mixer.Sound('<path>') # Loads the WAV file. <Sound>.play() # Starts playing the sound.
import collections, dataclasses, enum, io, itertools as it, pygame as pg, urllib.request from random import randint
P = collections.namedtuple('P', 'x y') # Position D = enum.Enum('D', 'n e s w') # Direction W, H, MAX_S = 50, 50, P(5, 10) # Width, Height, Max speed
def main(): def get_screen(): pg.init() return pg.display.set_mode((W*16, H*16)) def get_images(): url = 'https://gto76.github.io/python-cheatsheet/web/mario_bros.png' img = pg.image.load(io.BytesIO(urllib.request.urlopen(url).read())) return [img.subsurface(get_rect(x, 0)) for x in range(img.get_width() // 16)] def get_mario(): Mario = dataclasses.make_dataclass('Mario', 'rect spd facing_left frame_cycle'.split()) return Mario(get_rect(1, 1), P(0, 0), False, it.cycle(range(3))) def get_tiles(): border = [(x, y) for x in range(W) for y in range(H) if x in [0, W-1] or y in [0, H-1]] platforms = [(randint(1, W-2), randint(2, H-2)) for _ in range(W*H // 10)] return [get_rect(x, y) for x, y in border + platforms] def get_rect(x, y): return pg.Rect(x*16, y*16, 16, 16) run(get_screen(), get_images(), get_mario(), get_tiles())
def run(screen, images, mario, tiles): clock = pg.time.Clock() while all(event.type != pg.QUIT for event in pg.event.get()): keys = {pg.K_UP: D.n, pg.K_RIGHT: D.e, pg.K_DOWN: D.s, pg.K_LEFT: D.w} pressed = {keys.get(ch) for ch, is_prsd in enumerate(pg.key.get_pressed()) if is_prsd} update_speed(mario, tiles, pressed) update_position(mario, tiles) draw(screen, images, mario, tiles, pressed) clock.tick(28)
def update_speed(mario, tiles, pressed): x, y = mario.spd x += 2 * ((D.e in pressed) - (D.w in pressed)) x -= (x > 0) - (x < 0) y += 1 if D.s not in get_boundaries(mario.rect, tiles) else (D.n in pressed) * -10 mario.spd = P(x=max(-MAX_S.x, min(MAX_S.x, x)), y=max(-MAX_S.y, min(MAX_S.y, y)))
def update_position(mario, tiles): x, y = mario.rect.topleft n_steps = max(abs(s) for s in mario.spd) for _ in range(n_steps): mario.spd = stop_on_collision(mario.spd, get_boundaries(mario.rect, tiles)) x, y = x + mario.spd.x / n_steps, y + mario.spd.y / n_steps mario.rect.topleft = x, y
def get_boundaries(rect, tiles): deltas = {D.n: P(0, -1), D.e: P(1, 0), D.s: P(0, 1), D.w: P(-1, 0)} return {d for d, delta in deltas.items() if rect.move(delta).collidelist(tiles) != -1}
def stop_on_collision(spd, bounds): return P(x=0 if (D.w in bounds and spd.x < 0) or (D.e in bounds and spd.x > 0) else spd.x, y=0 if (D.n in bounds and spd.y < 0) or (D.s in bounds and spd.y > 0) else spd.y)
def draw(screen, images, mario, tiles, pressed): def get_marios_image_index(): if D.s not in get_boundaries(mario.rect, tiles): return 4 return next(mario.frame_cycle) if {D.w, D.e} & pressed else 6 screen.fill((85, 168, 255)) mario.facing_left = (D.w in pressed) if {D.w, D.e} & pressed else mario.facing_left screen.blit(images[get_marios_image_index() + mario.facing_left * 9], mario.rect) for t in tiles: screen.blit(images[18 if t.x in [0, (W-1)*16] or t.y in [0, (H-1)*16] else 19], t) pg.display.flip()
if __name__ == '__main__': main()
Pandas
1 2 3 4 5
# $ pip3 install pandas matplotlib import pandas as pd from pandas import Series, DataFrame import matplotlib.pyplot as plt
Series
Ordered dictionary with a name.
1 2 3 4 5
>>> Series([1, 2], index=['x', 'y'], name='a') x 1 y 2 Name: a, dtype: int64
1 2 3 4
<Sr> = Series(<list>) # Assigns RangeIndex starting at 0. <Sr> = Series(<dict>) # Takes dictionary's keys for index. <Sr> = Series(<dict/Series>, index=<list>) # Only keeps items with keys specified in index.
<Sr> = <Sr> ><== <el/Sr> # Returns a Series of bools. <Sr> = <Sr> +-*/ <el/Sr> # Items with non-matching keys get value NaN.
1 2 3 4
<Sr> = <Sr>.append(<Sr>) # Or: pd.concat(<coll_of_Sr>) <Sr> = <Sr>.combine_first(<Sr>) # Adds items that are not yet present. <Sr>.update(<Sr>) # Updates items that are already present.
1 2 3
<Sr>.plot.line/area/bar/pie/hist() # Generates a Matplotlib plot. plt.show() # Displays the plot. Also plt.savefig(<path>).
+-----------------+-------------+-------------+---------------+ | | 'rank' | ['rank'] | {'r': 'rank'} | +-----------------+-------------+-------------+---------------+ | sr.apply(…) | | rank | | | sr.agg(…) | x 1 | x 1 | r x 1 | | sr.transform(…) | y 2 | y 2 | y 2 | +-----------------+-------------+-------------+---------------+
Last result has a hierarchical index. Use
'<Sr>[key_1, key_2]' to get its values.
DataFrame
Table with labeled rows and columns.
1 2 3 4 5
>>> DataFrame([[1, 2], [3, 4]], index=['a', 'b'], columns=['x', 'y']) x y a 1 2 b 3 4
1 2 3
<DF> = DataFrame(<list_of_rows>) # Rows can be either lists, dicts or series. <DF> = DataFrame(<dict_of_columns>) # Columns can be either lists, dicts or series.
<Sr/DF> = <DF>[column_key/s] # Or: <DF>.column_key <DF> = <DF>[row_bools] # Keeps rows as specified by bools. <DF> = <DF>[<DF_of_bools>] # Assigns NaN to False values.
1 2 3
<DF> = <DF> ><== <el/Sr/DF> # Returns DF of bools. Sr is treated as a row. <DF> = <DF> +-*/ <el/Sr/DF> # Items with non-matching keys get value NaN.
1 2 3 4 5
<DF> = <DF>.set_index(column_key) # Replaces row keys with values from a column. <DF> = <DF>.reset_index() # Moves row keys to a column named index. <DF> = <DF>.sort_index(ascending=True) # Sorts rows by row keys. <DF> = <DF>.sort_values(column_key/s) # Sorts rows by the passed column/s.
DataFrame — Merge, Join,
Concat:
1 2 3 4 5 6 7 8 9
>>> l = DataFrame([[1, 2], [3, 4]], index=['a', 'b'], columns=['x', 'y']) x y a 1 2 b 3 4 >>> r = DataFrame([[4, 5], [6, 7]], index=['b', 'c'], columns=['y', 'z']) y z b 4 5 c 6 7
+------------------------+---------------+------------+------------+--------------------------+ | | 'outer' | 'inner' | 'left' | Description | +------------------------+---------------+------------+------------+--------------------------+ | l.merge(r, on='y', | x y z | x y z | x y z | Joins/merges on column. | | how=…) | 0 1 2 . | 3 4 5 | 1 2 . | Also accepts left_on and | | | 1 3 4 5 | | 3 4 5 | right_on parameters. | | | 2 . 6 7 | | | Uses 'inner' by default. | +------------------------+---------------+------------+------------+--------------------------+ | l.join(r, lsuffix='l', | x yl yr z | | x yl yr z | Joins/merges on row keys.| | rsuffix='r', | a 1 2 . . | x yl yr z | 1 2 . . | Uses 'left' by default. | | how=…) | b 3 4 4 5 | 3 4 4 5 | 3 4 4 5 | If r is a Series, it is | | | c . . 6 7 | | | treated as a column. | +------------------------+---------------+------------+------------+--------------------------+ | pd.concat([l, r], | x y z | y | | Adds rows at the bottom. | | axis=0, | a 1 2 . | 2 | | Uses 'outer' by default. | | join=…) | b 3 4 . | 4 | | A Series is treated as a | | | b . 4 5 | 4 | | column. Use l.append(sr) | | | c . 6 7 | 6 | | to add a row instead. | +------------------------+---------------+------------+------------+--------------------------+ | pd.concat([l, r], | x y y z | | | Adds columns at the | | axis=1, | a 1 2 . . | x y y z | | right end. Uses 'outer' | | join=…) | b 3 4 4 5 | 3 4 4 5 | | by default. A Series is | | | c . . 6 7 | | | treated as a column. | +------------------------+---------------+------------+------------+--------------------------+ | l.combine_first(r) | x y z | | | Adds missing rows and | | | a 1 2 . | | | columns. Also updates | | | b 3 4 5 | | | items that contain NaN. | | | c . 6 7 | | | R must be a DataFrame. | +------------------------+---------------+------------+------------+--------------------------+
All operations operate on columns by default. Pass
'axis=1' to process the rows instead.
1 2 3 4 5
>>> df = DataFrame([[1, 2], [3, 4]], index=['a', 'b'], columns=['x', 'y']) x y a 1 2 b 3 4
1 2 3 4 5 6 7 8 9
+-----------------+-------------+-------------+---------------+ | | 'sum' | ['sum'] | {'x': 'sum'} | +-----------------+-------------+-------------+---------------+ | df.apply(…) | | x y | | | df.agg(…) | x 4 | sum 4 6 | x 4 | | | y 6 | | | +-----------------+-------------+-------------+---------------+
1 2 3 4 5 6 7 8 9 10
+-----------------+-------------+-------------+---------------+ | | 'rank' | ['rank'] | {'x': 'rank'} | +-----------------+-------------+-------------+---------------+ | df.apply(…) | x y | x y | x | | df.agg(…) | a 1 1 | rank rank | a 1 | | df.transform(…) | b 2 2 | a 1 1 | b 2 | | | | b 2 2 | | +-----------------+-------------+-------------+---------------+
Use '<DF>[col_key_1, col_key_2][row_key]'
to get the fifth result's values.
DataFrame — Plot, Encode,
Decode:
1 2 3
<DF>.plot.line/bar/hist/scatter/box() # Also: `x=column_key, y=column_key/s`. plt.show() # Displays the plot. Also plt.savefig(<path>).
1 2 3 4 5
<DF> = pd.read_json/html('<str/path/url>') # Run `$ pip3 install beautifulsoup4 lxml`. <DF> = pd.read_csv/pickle/excel('<path/url>') # Use `sheet_name=None` to get all Excel sheets. <DF> = pd.read_sql('<table/query>', <conn.>) # Accepts SQLite3 or SQLAlchemy connection. <DF> = pd.read_clipboard() # Reads a copied table from the clipboard.
1 2 3 4 5
<dict> = <DF>.to_dict(['d/l/s/…']) # Returns columns as dicts, lists or series. <str> = <DF>.to_json/html/csv([<path>]) # Also to_markdown/latex([<path>]). <DF>.to_pickle/excel(<path>) # Run `$ pip3 install openpyxl` for xlsx files. <DF>.to_sql('<table_name>', <connection>) # Accepts SQLite3 or SQLAlchemy connection.
GroupBy
Object that groups together rows of a dataframe based on the
value of the passed column.
1 2 3 4 5 6
>>> df = DataFrame([[1, 2, 3], [4, 5, 6], [7, 8, 6]], index=list('abc'), columns=list('xyz')) >>> df.groupby('z').get_group(6) x y b 4 5 c 7 8
1 2 3 4
<GB> = <DF>.groupby(column_key/s) # Splits DF into groups based on passed column. <DF> = <GB>.apply(<func>) # Maps each group. Func can return DF, Sr or el. <GB> = <GB>[column_key] # Single column GB. All operations return a Sr.
>>> gb = df.groupby('z') x y z 3: a 1 2 3 6: b 4 5 6 c 7 8 6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
+-----------------+-------------+-------------+-------------+---------------+ | | 'sum' | 'rank' | ['rank'] | {'x': 'rank'} | +-----------------+-------------+-------------+-------------+---------------+ | gb.agg(…) | x y | x y | x y | x | | | z | a 1 1 | rank rank | a 1 | | | 3 1 2 | b 1 1 | a 1 1 | b 1 | | | 6 11 13 | c 2 2 | b 1 1 | c 2 | | | | | c 2 2 | | +-----------------+-------------+-------------+-------------+---------------+ | gb.transform(…) | x y | x y | | | | | a 1 2 | a 1 1 | | | | | b 11 13 | b 1 1 | | | | | c 11 13 | c 2 2 | | | +-----------------+-------------+-------------+-------------+---------------+
from sys import argv, exit from collections import defaultdict, namedtuple from dataclasses import make_dataclass from enum import Enum import functools as ft, itertools as it, operator as op, re
def main(): pass
###
## UTIL
#
def read_file(filename): with open(filename, encoding='utf-8') as file: return file.readlines()