1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
|
(* This module's name matches the library's own name ("rite_of"), so dune
treats it as the library's top-level module instead of generating one
automatically -- every sibling module this library defines must be
re-exported here explicitly, or external references to e.g.
[Rite_of.Temporal_of] (bin/main.ml, every test/ file that opens this
rite) stop resolving. Same property {!Rite_ef.Rite_ef}'s own top-of-file
comment records for the EF side. *)
module Vocab_of = Vocab_of
module Temporal_of = Temporal_of
module Precedence_of = Precedence_of
module Rubrics_of = Rubrics_of
module Lectionary_of = Lectionary_of
open Colitur_kernel
(* [~lectionary], not a value closed over an internal load: {!Rite_ef.Rite_ef
.context}'s own comment records the exact failure this shape avoids --
fix round 1 (coordinator review) found a prior version that closed over
[Lectionary_ef] loading data/ef/lectionary.sexp as a side effect of being
linked, which made `colitur easter <year>` (a command touching no
lectionary data at all) die at startup the moment that file was absent
from a bare `dune build`. Do not repeat that here: [context] is a
function, and [~lectionary] is caller-supplied, exactly mirroring how the
sanctoral [Layer.t] itself already travels (a loose argument to
{!Colitur_kernel.Calendar.year}/[day], never embedded in this record).
No [~commons] parameter: unlike EF, the OF lectionary chain
({!Lectionary_of.readings}) has no Commons indirection to thread through
-- see that function's own .mli citation for why (OF has no shared-
formulary table the way EF's Commons.sexp is one). *)
let context ~lectionary : (Vocab_of.season, Vocab_of.rank) Rite.t =
{ Rite.id = Temporal_of.id;
vocab = Vocab_of.vocab;
year_start = Temporal_of.year_start;
temporal = Temporal_of.temporal;
anchors = Temporal_of.anchors;
(* The Roman rite -- OF as much as EF -- reckons Easter on the Gregorian
calendar; nothing in the Normae or the IGMR revisits that. Supplied
here rather than assumed by the kernel, so a Julian-reckoning rite
can supply {!Colitur_kernel.Computus.julian_easter} instead -- the
exact reasoning {!Rite_ef.Rite_ef.context}'s own [easter] field
carries, restated here because it is a per-rite decision, not one
this module inherits from EF. Read by [Layer.index] to resolve
{!Date_spec.Easter_offset} (the two movable universal solemnities,
Sacred Heart and the Immaculate Heart of Mary, and any
[--overlay]-supplied movable local feast). *)
easter = Colitur_kernel.Computus.gregorian_easter;
(* Identity: [Some (Date.month d, Date.day d)] for every date, unlike
{!Rite_ef.Temporal_ef.bissextile_fixed_key}. The OF has NO bissextile
(leap-year kalends-doubling) convention for [Layer.on_date]'s fixed
half to correct for -- verified, not merely assumed absent: the only
occurrence of "bissextilis" anywhere in the 2002 Missal (extracted
docs/research/of/missale-romanum-2002.pdf, pdftotext -layout, line
4502) marks leap years in the calendarium's own table of MOVABLE
feasts (an Easter-date lookup table), not a rule that doubles a
kalends position the way the 1962 Missal's February footnote does
(LT.txt:5011-5014, {!Rite_ef.Temporal_ef.bissextile_fixed_key}'s own
citation). A reader coming from the EF side of this codebase should
expect [fixed_key] to carry that mechanism; the OF's own primary
source simply has no such rule to encode, so this field is the
[Rite.t.fixed_key] doc comment's own stated default for "a rite with
no reason to differ". *)
fixed_key = (fun d -> Some (Date.month d, Date.day d));
rules =
{ Precedence.band = Precedence_of.band;
disposition = Precedence_of.disposition;
admit = Precedence_of.admit;
vigil_feast = Precedence_of.vigil_feast };
(* NOT [Vocab_of.seasons] (which lists each season once) -- this is
exactly the case {!Colitur_kernel.Rite.t.season_runs}'s own doc
comment and {!Colitur_kernel.Vocab.t.seasons}'s both name as their
illustration of why [season_runs] is rite-supplied rather than
derived: "a rite may legitimately have one season appear in two
separate runs (the modern form's Ordinary Time does)". Confirmed
directly in {!Temporal_of.season} (Normae n. 43-44, its own "first
run"/"second run" comments): over one Advent-anchored liturgical
year, Ordinary Time is interrupted by Lent+Easter, so the
run-length-compressed sequence {!Validate}'s "seasons" check expects
has SIX runs over five seasons, not five. *)
season_runs = Vocab_of.[ Advent; Christmas; Ordinary_time; Lent; Easter; Ordinary_time ];
transfer_target = Precedence_of.transfer_target;
readings = Lectionary_of.readings ~lectionary ~year_start:Temporal_of.year_start;
creed = Rubrics_of.creed;
gloria = Rubrics_of.gloria;
preface = Rubrics_of.preface }
|
