Python provides a sophisticated mechanism to build a list called a list comphrehension or list maker. A list comprehension is a single expression that combines an expression, for statement and an optional if statement. This allows a simple, clear expression of the processing that will build up the values of a list.

The basic syntax follows the syntax for a list literal. It encloses the processing in []'s. Here's the simplest form, omitting the optional if clause.

[ expr for-clause ]

The for-clause mirrors the for statement:

for v in sequence

Here are some examples.

even = [ 2*x for x in range(18) ]
hardways = [ (x,x) for x in (2,3,4,5) ]
samples = [ random.random() for x in range(10) ]

A list comprehension behaves like the following loop:

r= []
for 
v
 in 
sequence
:
    r.append( 
expr
 )

The basic process, then, is to iterate through the sequence in the for-clause , evaluating the expression, expr . The values that result are assembled into the list. If the expression depends on the for-clause , each value in the list can be different. If the expression doesn't depend on the for-clause , each value will be the same.

Here's an example where the expression depends on the for-clause.

>>> 
a= [ v*2+1 for v in range(10) ]

>>> 
a

[1, 3, 5, 7, 9, 11, 13, 15, 17, 19]
            

This creates the first 10 odd numbers. It starts with the sequence created by range( 10 ). The for-clause at assigns each value in this sequence to the local variable v. The expression, v*2+1, is evaluated for each distinct value of v. The expression values are assembled into the resulting list.

Typically, the expression depends on the variable set in the for-clause . Here's an example, however, where the expression doesn't depend on the for-clause .

b= [ 0 for i in range(10) ]

This creates a list of 10 zeroes. Because the expression doesn't depend on the for-clause , this could also be done as

b= 10*[0]

A comprehension can also have an if-clause .

The basic syntax is as follows:

[ expr for-clause if-clause ]

The for-clause mirrors the for statement:

for v in sequence

The if-clause mirrors the if statement:

if filter

Here is an example.

hardways = [ (x,x) for x in range(1,7) if x+x not in (2, 12) ]

This more complex list comprehension behaves like the following loop:

r= []
for 
v
 in 
sequence
:
    if 
filter
:
        r.append( 
expr
 )

The basic process, then, is to iterate through the sequence in the for-clause , evaluating the if-clause . When the the if-clause is true, evaluate the expression, expr . The values that result are assembled into the list.

>>> 
v = [ (x,2*x+1) for x in range(10) if x%3==0 ]

>>> 
v

[(0, 1), (3, 7), (6, 13), (9, 19)]
            

This works as follows:

A list comprehension can have any number of for-clauses and if-clauses , in any order. A for-clause must be first. The clauses are evaluated from left to right.