# Group Duplicate Files by Content Given a snapshot of files, return groups of paths whose complete contents are identical. The interview report preserves only the label "File Deduplication," so the literal records and ordering rules below are explicit practice assumptions rather than claimed original details. ```python def group_duplicate_files(files: list[list[str]]) -> list[list[str]]: ... ``` Each input record is exactly `[path, content]`. Paths are unique nonempty strings. Contents are arbitrary strings and may be empty. Return only groups containing at least two paths. Sort paths within each group lexicographically, then sort groups lexicographically by their complete path lists. Exact content equality defines a duplicate; path names and metadata do not. ## Example ```text Input: files = [ ["/a/report.txt", "same"], ["/b/empty.txt", ""], ["/c/copy.txt", "same"], ["/d/other.txt", "different"], ["/e/also-empty.txt", ""] ] Output: [ ["/a/report.txt", "/c/copy.txt"], ["/b/empty.txt", "/e/also-empty.txt"] ] ``` ## Constraints and Errors - `0 <= len(files) <= 200_000` - The total number of content characters is at most `20_000_000`. - A record with the wrong arity, non-string field, empty path, or duplicate path raises `ValueError`. - Validate the record shape and path uniqueness before producing output. - Do not mutate the input. ## Hints - A direct map from full content to paths is sufficient for this literal version. - For real files too large to hold in memory, first bucket by size, then by a streaming digest, and verify byte equality before declaring a duplicate. - Digest equality alone is not a proof of content equality when correctness must survive collisions.