Lookaround in practice

Let's put all four forms of lookaround together in real-world scenarios. Lookarounds shine when you need to:

1. Extract a value without the context that identifies it.
2. Validate a string against multiple independent conditions.
3. Filter matches that satisfy some conditions but not others.

Password validation

Pattern: ^(?=.*[A-Z])(?=.*\d)(?=.*[^a-zA-Z\d]).{8,}$
Checks:
  - at least one uppercase     (?=.*[A-Z])
  - at least one digit         (?=.*\d)
  - at least one symbol        (?=.*[^a-zA-Z\d])
  - minimum length 8           .{8,}

Each lookahead checks one condition, starting from the beginning (^). They are all zero-width: the engine stays at position 0 and then consumes with .{8,}$.

Extraction between delimiters

Pattern: (?<=\().+?(?=\))
Sample:  Funzione foo(bar) e baz(qux, qix)
Match: "bar", "quux, qix" (without parentheses)

Lookbehind + lookahead + lazy quantifier: extracts the content inside the parentheses without including them.

Combining lookahead and lookbehind

Using lookahead and lookbehind together allows isolating strings that lie within specific tags or formats (e.g. extracting text between two tags without including the tags in the final match). This avoids subsequent string cleanup operations.

Try it

Review exercise

Additional challenge