Comprehensions

A comprehension constructs a list, dictionary or set from an existing iterable in a single expression. It is the conventional form in Python for building one sequence from another, and is generally both shorter and faster than the equivalent loop, because the list is built without repeated method lookups.

The expression before for is evaluated once per element and its result becomes the corresponding element of the new list. The original list is unchanged.

Filtering

An if clause at the end restricts which elements are included. Elements for which the condition is false are omitted entirely; the result is therefore shorter than the input.

Transformation and filtering combine in one expression. The if is applied first, and the leading expression is evaluated only for elements that pass.

A conditional expression may also appear in the leading position, which selects between two values rather than filtering. The two forms are distinct: the leading if ... else chooses a value; the trailing if chooses membership.

Dictionary and set comprehensions

The same syntax builds a dictionary when the body is a key: value pair, and a set when it is a single expression in braces. Building a lookup table from two parallel sequences is a frequent use.

Nested iteration

A comprehension may contain more than one for clause. They are evaluated left to right, in the same order as the equivalent nested loops, with the rightmost varying fastest.

Nesting beyond two levels is difficult to read and should be written as an ordinary loop.

Generator expressions

Replacing the brackets with parentheses produces a generator, which yields elements one at a time instead of building the whole sequence in memory. When the result is consumed immediately by sum, max, any or all, a generator avoids constructing an intermediate list.

When not to use a comprehension

A comprehension should perform one transformation and fit on one or two lines. If it requires several conditions, a try block, or side effects such as printing or writing to a file, write an explicit loop instead.

Task

Write passing_genes(records). It receives a list of (gene, depth) tuples and returns the gene names whose depth is 30 or more, uppercased and sorted alphabetically. Use a comprehension; the tuple unpacks directly in the for clause.