Skip to content

DataFog v4.5.0

Choose a tag to compare

View all tags
@sidmohan0 sidmohan0 released this 02 Jul 17:34
v4.5.0
2fbce0a

What's New

Released: 2026-07-02

4.5 Release Focus

DataFog 4.5.0 is a focused release for lightweight text PII screening. The core install remains dependency-light while the text APIs, CLI, guardrail helpers, German structured PII coverage, optional-profile docs, and Python 3.13 compatibility story become clearer and easier to verify.

German Structured PII

German VAT IDs and German IBANs are detected by default in the regex engine. Broader German identifiers such as tax IDs, postal codes, passport numbers, residence permit numbers, and pension insurance numbers require locales=["de"] or explicit entity selection.

Python 3.13 Optional Profiles

Python 3.13 is certified for the core SDK, CLI, nlp, nlp-advanced, and ocr install profiles. Donut OCR still requires a model already available locally. distributed and all are not newly certified on Python 3.13 in 4.5.0.

Optional OCR And Spark Surfaces

OCR and Spark remain supported optional surfaces. They are not deprecated, but their broader overhaul is deferred beyond 4.5.0 so the core package can stay tight and text-first.

Telemetry Defaults

Telemetry remains disabled unless DATAFOG_TELEMETRY=1 is set. DATAFOG_NO_TELEMETRY=1 and DO_NOT_TRACK=1 continue to force telemetry off.

🐛 Bug Fixes

  • Merge pull request #149 from DataFog/fix/4.5-release-review
  • fix: restore v4.4.0 detection parity for SSN and detect() output keys

🔧 Other Changes

  • Merge pull request #150 from DataFog/release/4.5.0
  • Prepare v4.5.0 stable release
  • ci: exclude historical audit artifacts from prettier
  • chore: finalize 4.5.0 changelog and document guardrail engine defaults
  • chore: bump version to 4.4.0 [skip ci]
  • Prepare v4.4.0 stable release (#133)

📥 Installation

# Core package (lightweight)
pip install datafog

# With all features
pip install datafog[all]

📊 Metrics

  • Package size: ~2MB (core)
  • Install time: ~10 seconds
  • Tests passing: ✅
  • Commits this week: 8
Assets 4
Loading