Quick links from the Standard Library
- Operations on Booleans (
True
,False
) - Comparisons
- Operations on numbers (int,floats)
- Operations on sequences (strings, lists, tuples, … indexables), and also for mutable sequences
- Built-in functions
- The range function
- Manipulating strings
Types
Type | Example | Convert to type using |
---|---|---|
None | None | - |
Integer int |
0, 1, 100 | int() |
Float float |
0.0, .5, 10.453 | float() |
String str |
‘hello’ | str() |
List list |
[1,’hi’,None,3.14] | list([iterable]) |
Tuple tuple |
[1,’hi’,None,3.14] | tuple([iterable]) |
Dictionary dict |
{ ‘a’ : 1, ‘b’: 2 } | dict([iterable]) |
Set set |
{ ‘a’, ‘b’ } | dict([iterable]) |
Create your own type using class
Iterating through list elements
for item in aList:
print('List item: ', item)
It is not necessary to iterate list items via their position. You rarely need to do:
for i in range(len(aList)):
print('List item: ', aList[i])
If you are looking for an element in a list and cannot find it, you can use the else
clause.
For example, we are searching for the first integer in aList
that is bigger than 10.
aList = [ 1, 4, 7, 6, 3, 4, 32, 17, 18, 10, 8]
for item in aList:
if item > 10:
return item
else:
return None
Iterating through a dictionary
for key in dictionary:
print('Key item: ', key)
Same as
for key in dictionary.keys():
print('Key item: ', key)
For the values only
for v in dictionary.values():
print('Value item: ', v)
For both keys and values
for k,v in dictionary.items():
print('Key item :', k, ' | Value item: ', v)
Read a file line by line
with open("filename","r") as f: # Add encoding='...' if necessary
for line in f:
print('Line: ', line.rstrip()) # Removing the trailing newline character
# print('Line: ', line, end='') # Or that works too
The with
statement takes care of closing the file in case an error occurs in the for
loop.
Defining a function
def functionName(arg1, arg2, arg3=None, arg4='test.log', arg5=32):
# body of the function
arg1
and arg2
are positional arguments, required when calling the function.
arg3
, arg4
, arg5
are keyword arguments with a default value (respectively None
, the string test.log
, and the interger 32
)
Positional arguments are always first, and in order. Keyword arguments are last, can be omitted and/or given in any desired order.
You can call the above function with, for example, functionName(12,3.14,arg5=10,arg4='Blabla.tex')
.
In the body, at runtime,
arg1
will have the value12
(of typeint
),arg2
will have the value3.14
(of typefloat
),arg3
will have the valueNone
(of typeNoneType
),arg4
will have the value'Blabla.txt'
(of typestr
), andarg5
will have the value10
(of typeint
).
Importing a function from a module
from moduleName import functionName
Or if we want to rename locally the function:
from moduleName import functionName as newName
Testing a condition
We need boolean logic to compare values. We have of course < >, <=, >=, to compare integers, floats or even strings.
If we compare objects, Python will compare their size (by calling __len__()
).
The comparator ==
and !=
checks the values, while is
checks
the id (ie are the same object?). Obviously, when v1 is v2
, it
implies v1 == v2
.
A condition can combine expressions with the above comparators, using
the logical and
and or
operators, and not condition
is its
negation.
if condition:
print('Condition is true')
else:
print('Condition is false')
We can pass an object as the condition. This is very convenient as Python will check if the object is “False”.
False objects are:
- the keywords
None
,False
, - the values
0
,0.0
, - the empty string
''
, - the empty list
[]
, the empty tuple()
- the empty dictionary
{}
,
If you define your own class, you’ll have to add the function
__bool__()
(or __len__()
, since Python defaults to __len__
when
__bool__
is not present)
For example, when calling a function where arguments have default values
def functionName(arg1=None):
if arg1:
print('arg1 was None')
# Body of the function
Catching errors
try:
#
# Do something
#
except ErrorName as e:
print("An error of type ErrorName occured:", repr(e))
print('Continue after the error')