Groupes de capture et remplacements

A capturing group wraps part of a pattern in parentheses so you can pull out just that piece, not the whole match.

After a match, group(1) returns the first parenthesized part, group(2) the second, and group(0) the entire matched string.

When you use findall with groups, each result becomes a tuple of the captured parts, making structured data easy to collect.

You can give a group a name for clearer code. Reach it later with group by that name instead of remembering a number.

The re.sub function finds every match and replaces it with text you choose. It is search and replace, powered by patterns.

Inside a replacement you can reference a captured group with \1. This lets you reorder or rewrite matched text on the fly.

A common use of re.sub is redaction. Match phone numbers or emails, then replace each one with a placeholder like a row of stars.

For tricky edits, pass a function to re.sub. It receives each match and returns the replacement, so you can compute results per match.

Sometimes you group only to apply a quantifier, not to capture. A non-capturing group starts with ?: so it never clutters your results.

Use re.subn instead of re.sub to get back both the new string and a count of how many replacements it actually made.

Capturing groups turn raw text into structured fields. A date string becomes year, month, and day you can store or compare directly.

You matched a full date and want only the year part you wrapped in parentheses. How do you read it?

You learned capturing groups, named groups, and re.sub. Together they let you extract precise fields and rewrite text in one clean pass. ✨