diff options
Diffstat (limited to 'lib/citation/book.ml')
| -rw-r--r-- | lib/citation/book.ml | 104 |
1 files changed, 104 insertions, 0 deletions
diff --git a/lib/citation/book.ml b/lib/citation/book.ml new file mode 100644 index 0000000..2a2b9c0 --- /dev/null +++ b/lib/citation/book.ml @@ -0,0 +1,104 @@ +(* SPDX-License-Identifier: AGPL-3.0-or-later *) + +type id = string + +let to_string t = t + +(* Every book cited across the shipped EF data (lectionary, sanctoral + propers, and commons), with every spelling any of the three files uses. + The dotted/undotted and modern/Vulgate pairs are inherited from lectio -- + see book.mli. Surveyed directly against the data, all three files, not + the lectionary alone -- sanctoral.sexp alone carries more citations than + the lectionary and was the source of every token missed in the first + pass. *) +let table = + [ ("genesis", [ "Gen" ]); + ("exodus", [ "Ex"; "Exod" ]); + ("leviticus", [ "Lev" ]); + ("numbers", [ "Num" ]); + ("kings_3", [ "3 Kings"; "3 Kgs." ]); + ("kings_4", [ "4 Kings" ]); + ("esdras_2", [ "2 Esd." ]); + ("tobit", [ "Tob" ]); + ("judith", [ "Judith" ]); + ("esther", [ "Esther" ]); + ("proverbs", [ "Prov" ]); + ("song_of_songs", [ "Song" ]); + ("wisdom", [ "Wis"; "Wis." ]); + ("ecclesiasticus", [ "Ecclus"; "Sir"; "Eccli" ]); + ("isaiah", [ "Isa"; "Isa." ]); + ("jeremiah", [ "Jer" ]); + ("ezekiel", [ "Ezech"; "Ezek" ]); + ("daniel", [ "Dan" ]); + ("osee", [ "Osee" ]); + ("joel", [ "Joel" ]); + ("jonas", [ "Jonas" ]); + ("malachi", [ "Mal" ]); + ("matthew", [ "Matt"; "Matt." ]); + ("mark", [ "Mark" ]); + ("luke", [ "Luke" ]); + ("john", [ "John" ]); + ("acts", [ "Acts" ]); + ("romans", [ "Rom" ]); + ("corinthians_1", [ "1 Cor"; "1 Cor." ]); + ("corinthians_2", [ "2 Cor"; "2 Cor." ]); + ("galatians", [ "Gal" ]); + ("ephesians", [ "Eph"; "Eph." ]); + ("philippians", [ "Phil" ]); + ("colossians", [ "Col"; "Col." ]); + ("thessalonians_1", [ "1 Thess"; "1 Thess." ]); + ("thessalonians_2", [ "2 Thess" ]); + ("timothy_1", [ "1 Tim." ]); + ("timothy_2", [ "2 Tim"; "2 Tim." ]); + ("titus", [ "Titus" ]); + ("hebrews", [ "Heb" ]); + ("james", [ "Jas"; "James" ]); + ("peter_1", [ "1 Pet"; "1 Pet." ]); + ("peter_2", [ "2 Pet." ]); + ("john_1", [ "1 John" ]); + (* "Apoc" is the Vulgate spelling and "Rev" its modern equivalent, but + BOTH sit inside Vulgate-tradition data, so both resolve to the same + Vulgate id here -- see book.mli's note by [apocalypse] never being an + [of_token] result under that name. Do not add "revelation" as a + spelling: it exists only as a tradition target (below), and giving it + an [of_token] entry would let one book carry two different ids. *) + ("apocalypse", [ "Apoc"; "Rev" ]) ] + +(* Targets a tradition can map ONTO that the Vulgate data never cites + directly. Present so [tradition_of_fields] can validate both sides. + [sirach] and [revelation] exist ONLY here, never as an [of_token] result: + "Sir" and "Rev" already resolve to the Vulgate ids [ecclesiasticus] and + [apocalypse] above, so a modern-numbering tradition maps ONTO these + targets rather than data ever citing them directly. *) +let tradition_targets = + [ "kings_1"; "kings_2"; "nehemiah"; "sirach"; "hosea"; "jonah"; "revelation" ] + +let all = List.map fst table @ tradition_targets + +let tokens = + List.concat_map (fun (id, sp) -> List.map (fun s -> (s, id)) sp) table + +let default_spelling id = + match List.assoc_opt id table with + | Some (first :: _) -> first + | Some [] | None -> id + +let of_token s = + let s = String.trim s in + List.assoc_opt s tokens + +type tradition = (string * string) list + +let vulgate = [] + +let known id = List.mem id all + +let tradition_of_fields fields = + List.filter (fun (a, b) -> known a && known b) fields + +let unknown_fields fields = + List.filter_map + (fun (a, b) -> if known a && known b then None else Some a) + fields + +let map tr id = match List.assoc_opt id tr with Some x -> x | None -> id |