2_corruption

Round 2: Corruption scanner

Problem

You have two equally-sized binary files: a known-good reference.bin and a possibly damaged corrupted.bin.

Find every byte range where they differ and return the ranges as a list of (offset, length) tuples. Adjacent differing bytes must be merged into a single range.

• Input: data/reference.bin and data/corrupted.bin (default 64 MB each; scale up to 1 GB via --size-mb 1024 as a stretch goal).
• Output: list[tuple[int, int]] sorted by offset, with no overlaps.

Files

File	Purpose
baseline.py	Intentionally slow starting point. Don't edit: it is the reference for the comparison.
solution.py	Edit this. Starts out delegating to baseline.py; replace with your faster implementation.
gen_data.py	Generates the reference and corrupted pair alongside a truth.json.
test_corruption.py	Correctness tests and the pytest-codspeed benchmark. Every test is parametrized over both the baseline and your solution.

Generate the data

uv run rounds/2_corruption/gen_data.py                 # default 64 MB.
uv run rounds/2_corruption/gen_data.py --size-mb 1024  # 1 GB stretch goal.

Or run uv run scripts/setup.py to generate every round's data in one go.

Verify correctness

uv run pytest rounds/2_corruption/

Benchmark

Walltime, locally:

uv run pytest --codspeed rounds/2_corruption/

codspeed run --mode walltime -- uv run pytest --codspeed rounds/2_corruption/