Prefer Generator Expressions Has Fix

• Playground Link

Description

List comprehensions like [x for x in range(10)] are a concise way to create lists in Python. However, we can achieve better memory efficiency by using generator expressions like (x for x in range(10)) instead. List comprehensions create the entire list in memory, while generator expressions generate each element one at a time. We can make the change by replacing the square brackets with parentheses.

YAML

id: prefer-generator-expressions
language: python
rule:
  pattern: $LIST
  kind: list_comprehension
transform:
  INNER:
    substring: {source: $LIST, startChar: 1, endChar: -1 }
fix: ($INNER)

This rule converts every list comprehension to a generator expression. However, not every list comprehension can be replaced with a generator expression. If the list is used multiple times, is modified, is sliced, or is indexed, a generator is not a suitable replacement.

Some common functions like any, all, and sum take an iterable as an argument. A generator function counts as an iterable, so it is safe to change a list comprehension to a generator expression in this context.

id: prefer-generator-expressions
language: python
rule:
  pattern: $FUNC($LIST)
constraints:
  LIST: { kind: list_comprehension }
  FUNC:
    any:
      - pattern: any
      - pattern: all
      - pattern: sum
      # ...
transform:
  INNER:
    substring: {source: $LIST, startChar: 1, endChar: -1 }
fix: $FUNC($INNER)

Example

any([x for x in range(10)])

Diff

any([x for x in range(10)]) 
any(x for x in range(10)) 

Contributed by

Steven Love