Exports

Research exports & data dictionary

HealthArchive provides metadata-only exports for research and reproducibility. Exports do not include raw HTML or full diff bodies.

Exports overview

Use the export manifest to discover available formats and limits.

Export manifest

Export manifest API

Dataset releases

Current release posture: existing releases remain available with checksums; scheduled publication is paused, and any future release requires explicit maintainer approval, manual dispatch, and rights/schema review.

Dataset controls verified on 2026-08-12: the published workflows are manual-only, the rights notice resolves, and exact 15-snapshot / 24-change schema guards are active.

HealthArchive dataset releases · Dataset provenance and reuse notice

Download / print

Download the data dictionary (Markdown) or use your browser’s print dialog to save as PDF.

Snapshots export (fields)

  • snapshot_id: numeric snapshot ID.
  • source_code / source_name: source identifiers.
  • captured_url: URL captured at crawl time.
  • normalized_url_group: canonical grouping key.
  • capture_timestamp_utc: UTC timestamp (ISO-8601).
  • language, status_code, mime_type,title: metadata when available.
  • capture_backend / capture_fidelity: capture-system and fidelity labels.
  • job_id / job_name: edition anchor (if available).
  • snapshot_url: stable public URL for citation.

Changes export (fields)

  • change_id: numeric change event ID.
  • source_code / source_name, normalized_url_group.
  • from_snapshot_id / to_snapshot_id and corresponding UTC timestamps.
  • change_type, summary, section/line counts, and change ratio.
  • high_noise and diff_truncated flags.
  • diff_version, normalization_version, computed_at_utc.
  • compare_url: stable public URL for the diff view.

Limitations

  • Exports reflect captured content, not real-time source updates.
  • Coverage is limited to in-scope sources and successful captures.
  • Replay fidelity varies by site and asset type.