Optional local database

hurahura stores subject metadata on disk under each subject’s META directory ({SubjectID}Tags.json and optional suffixed variants). For local projects that benefit from SQL queries, an optional SQLite index can be enabled. The filesystem remains the source of truth; the database is updated when hurahura writes meta JSON and when existing JSON files are newer than the stored copy.

Enable in miresearch.conf:

[database]
enabled = true
db_path =
sync_on_meta_write = true
study_columns = StudyDate, StudyDescription, PatientID, Modality, ANONYMISED, Age
meta_suffixes =
meta_sync_all = true
extra_fields = ANONYMISED

  • db_path — SQLite file (default: hurahura.db in data_root_dir).

  • sync_on_meta_write — Update the database when updateMetaFile runs (default true).

  • study_columns — Comma-separated top-level keys from the main Tags JSON to mirror as subjects table columns.

  • meta_suffixes — Comma-separated META JSON suffixes to track (empty entry before the first comma means the main {SubjectID}Tags.json). If omitted, -DBSync discovers every *Tags*.json under each subject META folder when meta_sync_all is true (default).

  • meta_sync_all — When true, full sync includes all discovered Tags JSON files.

  • extra_fields — Comma-separated [suffix:]TagName entries promoted to subject_extra for SQL filters.

Rebuild the index from all subjects:

hurahura -config my.conf -sA -DBSync

Tables (conceptual):

  • subjects — Study-level tags from the main Tags JSON (configurable columns).

  • series — One row per DICOM series from the Series list in that JSON.

  • meta_json — Full JSON text per meta suffix (for ad hoc inspection).

  • subject_extra — Key/value pairs from [database] extra_fields.

SQLite is appropriate for single-user, single-machine use. It is not intended for multi-user or networked deployments; use export or a dedicated server database for those.