fabioleitao/data_boar repository overview

⁠Data Boar

Compliance-aware discovery of personal and sensitive data across databases, files, APIs, and more — data soup in, structured findings out. Open-source Python stack with optional ML/DL; aligns with LGPD, GDPR, CCPA, and other frameworks via config.

Current Hub image: latest and 1.7.4.post12 resolve to the same digest (Linux/amd64, measured 2026-07-30):

sha256:ab8f5dad3e33618e5c0dbb6d880ddafa11fc0e39c8372d5f0f1a2316d8597842 (~309 MB compressed on Hub).

Runtime shape: multi-stage build → python:3.14-slim (Debian 13 / trixie, digest-pinned) → gcr.io/distroless/cc-debian13:nonroot (uid 65532, no shell / no apt). ML stack force-reinstalled from the hosted x86-64-v1 wheelhouse (wheelhouse-x86-64-v1-2026-07-29) so popcnt=0 on site-packages .so — runs on any x86-64, including Celeron-class CPUs without AVX. boar_fast_filter (Rust accelerator) is embedded in the image.

Also on Hub (historical, not latest): immutable June GA tag 1.7.4 (sha256:04b4cd7d…) — left untouched when post12 published. Prefer 1.7.4.post12 / latest for new deploys unless you intentionally pin the June digest.

Confirm pyproject.toml and GitHub Releases⁠ for the exact pairing. Free-threaded (no-GIL / cp314t) wheels for lab / Enterprise foresight live in the wheelhouse release⁠ (not inside this container — image uses GIL cp314).

⁠Copyright and maintainer

• Author / copyright: Fabio Leitao — LICENSE⁠ (BSD-3-Clause).
• Professional profile: Add your LinkedIn URL in the Hub Full description editor when publishing (do not embed personal profile URLs in this tracked file).
• Security: Vulnerability reporting — SECURITY.md⁠.

⁠Supported tags

⁠Quick start (web API + dashboard on port 8088)

Prepare a directory with config.yaml and mount it at /data:

docker pull fabioleitao/data_boar:latest
docker run -d -p 8088:8088 -v "$(pwd)/data:/data" -e CONFIG_PATH=/data/config.yaml fabioleitao/data_boar:latest