Skip to content

schema-proposal.md

Latest commit

 

History

History
164 lines (139 loc) · 8.46 KB

schema-proposal.md

File metadata and controls

164 lines (139 loc) · 8.46 KB
Raw

CECDR schema proposal (draft for committee review)

Identifier scheme

circ:<iso3166-1-alpha2>-<slug>
  • circ: — namespace prefix (placeholder, like mr: in the CRMEDR, pending a committee decision on prefixes across the registries).
  • <iso3166-1-alpha2> — lowercase ISO country code of the nation in which the circumscription's see is located. Supranational or extraterritorial structures use a reserved segment instead (see below).
  • <slug> — the see name (not the full styled title), ASCII-folded, lowercase, hyphenated: boston, roma, shkodre-pult. Generic type words ("Diocese of", "Diocesi di") are stripped: the type is an attribute, not part of the identity.

Rules

  1. Type is an attribute, not identity. A diocese elevated to an archdiocese, or an apostolic prefecture raised to a vicariate and then to a diocese, keeps its ID — as in the CRMEDR, where a beatus's canonization changes status, not identity.
  2. Homonymous sees within a nation are qualified by their civil region, following official usage where it exists: circ:us-portland-in-oregon, circ:us-lafayette-in-indiana.
  3. Renames and transfers of see keep the ID (with the new name as an attribute and the old among historical names); mergers produce a new ID for the united circumscription, with the predecessors retained as suppressed entries pointing to the successor; suppressed circumscriptions keep their IDs and are flagged, never deleted — historical data must remain referenceable.
  4. Eastern circumscriptions follow the same scheme; the church sui iuris is an attribute, carried as a cross-reference into the CESIDR (church_sui_iuris: esi:latin, esi:ukrainian, esi:maronite, esi:syro-malabar), and where an eparchy shares a city with a Latin see the slug is qualified by the church: circ:us-philadelphia (Latin), circ:us-philadelphia-ukrainian (archeparchy). The slug qualifier stays bare — -ukrainian, not -esi-ukrainian — since it is part of an identifier, not a reference.
  5. Supranational and extraterritorial structures (e.g. personal prelatures, ordinariates for Eastern faithful covering several nations) use the reserved first segment int in place of a country code: circ:int-opus-dei. int is not an ISO 3166-1 alpha-2 code, so no collision is possible; circ:va-<slug> is not used for this (the Holy See is a real territory). The token choice remains open for committee confirmation. Military ordinariates are national by nature: circ:it-ordinariato-militare.

Entry shape

{
  "id": "circ:us-boston",
  "litcal_id": "boston_us",
  "name": "Archdiocese of Boston",
  "nation": "US",
  "province": "Massachusetts",
  "church_sui_iuris": "esi:latin",
  "type": "ctype:archdiocese"
}

Circumscription types

type is not a free string but a cross-reference into a companion registry, data/circumscription_types.json, which mints a canonical ID for each canonical rank or juridic form a circumscription can hold:

ctype:<slug>

ctype:diocese, ctype:archeparchy, ctype:territorial-abbacy, ctype:apostolic-vicariate, ctype:military-ordinariate, ctype:personal-prelature, and so on — 16 in all. Each entry carries the Latin name (name_la, the Annuario's own nomenclature), the church to which the form belongs (latin, eastern, both), whether it is territorial, the title of the one who governs it, and the governing canon or document.

This follows the cross-registry convention already used in the family: COECDR embeds CRPDR rp: IDs, CDOCTDR embeds CRMEDR mr: IDs. An entry therefore reads "type": "ctype:diocese", and scripts/generate_seed.py asserts that every type resolves against the types registry.

The ctype: prefix is a placeholder on the same footing as circ:, pending the namespace coordination in open question 1.

Churches sui iuris

church_sui_iuris is likewise a cross-reference, but into a separate repository — the CESIDR, which mints esi: IDs for the 24 Churches sui iuris: the Latin Church and the 23 Eastern Catholic Churches.

"church_sui_iuris": "esi:latin"

The CESIDR slugs were chosen to match the bare values this field already carried (latin, ukrainian, maronite, syro-malabar), so the migration was a prefix and nothing else — no value was renamed and no entry changed meaning.

Because the target registry is a different repository, scripts/generate_seed.py can only check the shape of these values, not resolve them. Cross-repository resolution is a question for the committee alongside the prefix decision.

Further attributes

Planned beyond the seed: metropolitan (the ID of the metropolitan see), status (active | suppressed | merged_into:), historical_names, erected (date of erection), and external keys (litcal_id now; potentially GCatholic and Catholic-Hierarchy keys as cross-references).

Seed and its limits

The seed (data/circumscriptions.json) covers the 2,935 Latin-rite circumscriptions in the Liturgical Calendar API's world index. Known gaps and flaws, in scope for enrichment:

  • Eastern Catholic circumscriptions are absent (the source index is Latin-rite).
  • type is null throughout: the source names mix bare see names ("Lezhë") with styled ones ("Archdiocese of Boston"), so typing needs an authoritative pass (Annuario Pontificio).
  • Two homonymous Chinese sees (circ:cn-xinjiang-1/-2) carry ordinal qualifiers pending proper disambiguation.
  • The personal prelature of Opus Dei is supranational and therefore seeded as circ:int-opus-dei (rule 5) with no nation, even though the source index lists it under Italy; the source also mislabels the row "Diocesi di Lanusei" — a fix has been submitted upstream (LiturgicalCalendarAPI PR #718).

Open questions for the committee

  1. The namespace prefix (circ:) and its coordination with the other registries' prefixes.
  2. The reserved segment for supranational structures (rule 5).
  3. Whether to encode the church sui iuris in the slug only on collision (rule 4) or always for non-Latin circumscriptions.
  4. The authority order for names: Annuario Pontificio Latin names vs. vernacular official names (the seed currently carries the API's names, which mix both).
  5. Whether Ordinariates for Eastern-rite faithful without their own hierarchy, and pastoral structures like apostolic exarchates in the diaspora, need a dedicated type taxonomy beyond the Annuario's. Both are provisionally minted in data/circumscription_types.json (ctype:ordinariate-for-eastern-faithful, ctype:apostolic-exarchate) so the question can be decided against concrete entries rather than in the abstract.
  6. Whether type IDs should be English (ctype:territorial-abbacy, as minted) or Latin (ctype:abbatia-territorialis), following CDOCTDR's use of the Latin lemma. The Latin name is carried as name_la either way, so this is a question about the identifier alone.
  7. Whether patriarchal and major archiepiscopal sees warrant their own type IDs. They are presently archeparchies, with the dignity of the church sui iuris carried by church_sui_iuris; the Annuario lists them distinctly.
  8. Confirmation of the church field in data/circumscription_types.json against the Annuario Pontificio itself. The field records which churches are known to use each form — an empirical question answerable only by enumerating real circumscriptions, including suppressed ones, since rule 3 keeps those referenceable forever. Two initial classifications have already been corrected on evidence: ctype:territorial-abbacy (Grottaferrata, Byzantine, active) and ctype:apostolic-vicariate (the Bulgarian Byzantine vicariates of Macedonia and Thrace, 1883–1926). The remaining latin values were checked against Catholic-Hierarchy's structured view of the Byzantine tradition and against per-form surveys, which are verification aids and not authorities; no Eastern instance was found for diocese, archdiocese, territorial prelature, apostolic prefecture, mission sui iuris, personal prelature or personal ordinariate. Absence of evidence is weaker than the two positives that forced a change, particularly for suppressed Eastern circumscriptions, whose coverage in those sources is uneven. The Annuario remains the authority and has not been consulted.