StatsHub Docs

csv-deduplicator

A standalone Flask utility for de-duplicating CSV exports. Not part of the Bun workspace.

apps/csv-deduplicator is a small Flask tool, independent of everything else in this repository: upload a CSV, see which rows repeat, download those rows.

cd apps/csv-deduplicator
python3 -m venv .venv
. .venv/bin/activate
pip install -r requirements.txt
flask --app app run
Python 3Flask 3four routespandasthe actual workGunicornin the imageDocker

What it does

Upload a CSV

One form, one file, 16 MB ceiling. Anything pandas can read.

See the duplicates

Rows duplicated across every column, with keep=False so each copy is shown and not just the extras. The original table renders beside them.

Download them

duplicates.csv, from the session — the upload itself is never written to disk.

How it is built

Not a Bun workspace

No package.json, so bun install and bun run dev do not touch it and Turbo does not see it. It is in apps/ because it deploys, not because it shares anything.

State lives in the session

The deduplicated CSV is held as a string in the Flask session between the upload and the download, so there is no upload directory to clean up and no file to leak between users.

/health for the container

The probe the HEALTHCHECK and Coolify both call. Same contract as every other deployable here.

It ships its own Dockerfile and a docker-compose.yml service, which is how it actually gets deployed.

On this page