Merge branch 'ef-bvm-saturday': the Office of Our Lady on Saturday

Builds RG 91 entry 27 under RG 78 -- 'In sabbatis, in quibus occurrit
Officium de feria IV classis, fit de sancta Maria in sabbato' -- a rank
test, not a season list. band had been assigning entry 27 312966 times
across the domain while the office was never constructed, so an
otherwise-unoccupied Saturday got a bare ferial slug and its season's
colour.

75863 days change, in two shapes: colour (green or violet to white) on
74642, and colour plus a dropped commemoration on 1221. The 312966
reconciles to the day across four buckets, 210808 of them impeded by a
sanctoral winner.

The first version introduced a live RG 112(d) violation on every
Saturday falling on 16 July: the Office of Our Lady observed,
commemorating Our Lady. Both photographic scans print a rubric at that
Mass for exactly this collision -- 'Missa dici potest aut de sancta
Maria in sabbato, aut propria' -- absent from the electronic
transcription, and cited nowhere. It was missed because the check was a
grep for subject = Bvm, and Mt Carmel is tagged Saint: the rubric's
criterion is whose intercession the oration implores, not colitur's own
tag.

The I-V Mass numbering is a propers rotation and stays unbuilt;
citations remain empty on all 3073916 days.

Recorded rather than glossed: the new rule has no comparison-layer
witness at all, and lectio's fixture prints a Mt Carmel commemoration on
8 in-window dates where colitur now prints none -- a real divergence
that falls through a gap in what the comparators examine. Time after
Epiphany's 5015 office days have zero oracle witnesses and rest on the
scan alone.

1 files changed, 174 insertions, 4 deletions

diff --git a/test/test_precedence_ef.ml b/test/test_precedence_ef.ml
index ed9a933..6f8b5b5 100644
--- a/test/test_precedence_ef.ml
+++ b/test/test_precedence_ef.ml
@@ -291,10 +291,21 @@ let cases =
       cand ~origin:P.Sanctoral ~rank:V.Class3 ~layer:PE.universal_layer "vigil-of-st-lawrence",
       260 );
     (* Entry 27 -- RG 91 entry 27 (§4): an otherwise-unoccupied IV-class
-       Saturday. *)
-    ( "27 Office of the BVM on Saturday", off 62,
-      cand ~rank:V.Class4 "ef-time-after-pentecost-1-sat",
-      270 );
+       Saturday. ef-bvm-saturday task: sourced from [Temporal_ef.temporal]'s
+       own real output (see [of_temporal], the same review-finding-3
+       discipline entry 18's/21's own rows above use), not a hand-typed
+       slug/rank pair -- this row used to be the one place in this whole
+       suite where [band]'s entry-27 branch was exercised by a value nothing
+       in [Temporal_ef.temporal] could actually produce (the office itself
+       was unbuilt): [band] read [rank = Class4 && weekday = Sat]
+       correctly, but nothing proved [temporal] ever handed it a candidate
+       shaped that way for the RIGHT reason. Now it does: [of_temporal (off
+       62)] is 6 June 2026, a real otherwise-contested Time-after-Pentecost
+       Saturday (Norbert, Class3, wins the DAY outright in the real
+       pipeline -- irrelevant here, since this row tests [band] on the
+       TEMPORAL CANDIDATE alone, which [Temporal_ef.temporal] builds as the
+       BVM Saturday office regardless of what wins). *)
+    ( "27 Office of the BVM on Saturday (from Temporal_ef.temporal)", off 62, of_temporal (off 62), 270 );
     (* Entry 28 -- RG 91 entry 28 (§4): the unqualified IV-class catch-all. *)
     ("28 IV-class feria", off 65, cand ~rank:V.Class4 "ef-time-after-pentecost-1-tue", 280);
     (* Not an RG 91 row at all: a I-class candidate marked as a vigil, which
@@ -794,6 +805,52 @@ let disposition_cases =
         "conversion-of-st-paul",
       cand ~origin:P.Sanctoral ~rank:V.Class2 ~subject:Sub.Lord ~layer:PE.universal_layer
         "commemoration-of-the-baptism-of-the-lord",
+      "Commemorate(Ordinary)" );
+    (* RG112(d), fix round 1 (coordinator finding F1) -- PE.marian_slugs's
+       own citation has the full argument (both scans, including the 16
+       July "Missa dici potest aut ... aut ..." rubric). [of_temporal] on a
+       REAL 16-July-Saturday (2033) sources the winner from
+       Temporal_ef.temporal's own real output, the same review-finding-3
+       discipline this file's Ember/vigil rows already use -- not a
+       hand-typed slug/subject pair that could silently drift from what the
+       office actually builds. The loser is the real data/ef/sanctoral.sexp
+       entry, [Commemoration_only], the exact shape that made this branch
+       need to be checked BEFORE the Commemoration_only catch-all. *)
+    ( "RG112(d): the BVM Saturday Office excludes a commemoration invoking \
+       the SAME BVM's intercession -- Omit, not Commemorate (Mt Carmel, \
+       16 July 2033, a real otherwise-unoccupied Saturday)",
+      of_temporal (mk 2033 7 16),
+      cand ~origin:P.Sanctoral ~rank:V.Class3 ~status:Cel.Commemoration_only ~subject:Sub.Saint
+        ~layer:PE.universal_layer "our-lady-of-mt-carmel",
+      "Omit" );
+    (* Conjunct 1/2, LOSER identity: an ORDINARY Commemoration_only saint
+       NOT on PE.marian_slugs must still reach the ordinary commemorate
+       flow even when the winner is the BVM Saturday Office -- proves
+       RG112(d) excludes a specific same-BVM collision, not every
+       commemoration the office happens to outrank (the real live shape,
+       1 August, Holy Machabees, PE.marian_slugs's own comment). *)
+    ( "RG112(d) does NOT fire when the LOSER is not a Marian commemoration \
+       -- an ordinary saint stays Commemorate(Ordinary) even under the BVM \
+       Saturday Office (1 August 2026, Holy Machabees)",
+      of_temporal (mk 2026 8 1),
+      cand ~origin:P.Sanctoral ~rank:V.Class3 ~status:Cel.Commemoration_only ~subject:Sub.Saint
+        ~layer:PE.universal_layer "holy-machabees",
+      "Commemorate(Ordinary)" );
+    (* Conjunct 2/2, WINNER identity: an ordinary (non-Bvm, non-Marian-
+       titled) winner must not exclude Mt Carmel either -- RG112(d) is
+       keyed on the WINNER also being "de B. Maria Virg.", not merely on
+       the loser's own identity. SYNTHETIC: no live date has an ordinary
+       Class3 saint winning outright while Mt Carmel also loses that same
+       day (Mt Carmel's own fixed date, 16 July, has no other real
+       sanctoral entry to construct this from) -- proves the conjunct is
+       real, not merely untested, the same "SYNTHETIC" discipline the
+       RG112(a) rows above already use for their own second conjunct. *)
+    ( "SYNTHETIC: RG112(d) does NOT fire when only the LOSER is Marian -- \
+       an ordinary winner leaves Mt Carmel Commemorate(Ordinary)",
+      cand ~origin:P.Sanctoral ~rank:V.Class3 ~subject:Sub.Saint ~layer:PE.universal_layer
+        "ef-synthetic-ordinary-winner",
+      cand ~origin:P.Sanctoral ~rank:V.Class3 ~status:Cel.Commemoration_only ~subject:Sub.Saint
+        ~layer:PE.universal_layer "our-lady-of-mt-carmel",
       "Commemorate(Ordinary)" )
   ]
 
@@ -1421,6 +1478,107 @@ let test_ii_class_sunday_privileged_witness_admitted_end_to_end () =
     [ ("ef-some-saint", "omitted: admission limit reached") ]
     (List.map (fun (c, reason) -> (S.to_string c.P.cel.Cel.slug, reason)) resolution.P.omitted)
 
+(* RG 91 entry 27 / RG 78-79 (Caput IX, "De sancta Maria in sabbato") reached
+   end-to-end through the REAL pipeline -- this is THE hazard this task's own
+   brief names explicitly: {!PE.band}'s entry-27 branch has been reachable
+   only in isolation (the table-driven row above) since it was written, never
+   through a real [Temporal_ef.temporal] candidate winning an actual day, so
+   this test is checked for whether it exercises entry 27 for the RIGHT
+   reason -- proved by the mutation test recorded in the task report
+   (reverting temporal_ef.ml's own [is_bvm_saturday] branches back to the
+   generic ferial fallback reddens THIS test, not merely the [band]-table
+   row, which would still pass unchanged since it is sourced from real
+   [T.temporal] output either way).
+
+   1 August 2026, matching the task brief's own worked example against the
+   oracle exactly: a Time-after-Pentecost Saturday, otherwise unoccupied but
+   for Holy Machabees (Commemoration_only, Class3, real data/ef/sanctoral.sexp
+   entry, 1 August) -- the office wins outright (band 270, ahead of Holy
+   Machabees's own [unclassified]) and the Commemoration_only loser still
+   rides along as an ordinary commemoration, the SAME shape the oracle's own
+   row for this date shows ("V Mass of the B. V. M. -- Salve, Sancta Parens",
+   commemorating Holy Machabees). *)
+let test_bvm_saturday_wins_and_commemorates_end_to_end () =
+  let date = mk 2026 8 1 in
+  let day_ctx = ctx date in
+  let bvm_saturday = of_temporal date in
+  let holy_machabees =
+    cand ~origin:P.Sanctoral ~rank:V.Class3 ~status:Cel.Commemoration_only ~layer:PE.universal_layer
+      "holy-machabees"
+  in
+  let resolution = P.resolve real_rules day_ctx ~temporal:bvm_saturday ~sanctoral:[ holy_machabees ] in
+  Alcotest.(check string) "the BVM Saturday office is observed, not Holy Machabees"
+    "ef-time-after-pentecost-9-saturday" (S.to_string resolution.P.observed.P.cel.Cel.slug);
+  Alcotest.(check string) "observed rank is class-4, unconditional" "class-4"
+    (V.rank_to_string resolution.P.observed.P.cel.Cel.rank);
+  Alcotest.(check string) "observed colour is white, overriding the season's own green" "white"
+    (Col.to_string resolution.P.observed.P.cel.Cel.colour);
+  Alcotest.(check bool) "observed subject is Bvm" true (resolution.P.observed.P.cel.Cel.subject = Sub.Bvm);
+  Alcotest.(check (list (pair string string))) "Holy Machabees still rides along as an ordinary commemoration"
+    [ ("holy-machabees", "Ordinary") ]
+    (List.map
+       (fun (c, p) -> (S.to_string c.P.cel.Cel.slug, match p with P.Privileged -> "Privileged" | P.Ordinary -> "Ordinary"))
+       resolution.P.commemorations);
+  Alcotest.(check int) "nothing omitted" 0 (List.length resolution.P.omitted)
+
+(* The other half of the SAME hazard: a real Saturday where RG 78's own
+   protasis is FALSE (a genuine sanctoral feast wins the day outright), so
+   the BVM-Saturday-shaped temporal candidate must lose, and be OMITTED under
+   RG 26 (a temporal Class4 non-vigil loser is never commemorated) exactly
+   like the plain ferial candidate it replaces always was -- proving [band]/
+   [disposition] need no special case for this office at all, the design
+   claim the task report makes explicitly. 12 September 2026 is a Saturday
+   where `most-holy-name-of-mary` (data/ef/adjustments.sexp's own [Set_subject
+   Bvm] entry, Class3 Feast) wins outright -- ALSO tagged [subject = Bvm],
+   the one live data witness that could in principle confuse a
+   subject-keyed disposition rule with RG 26's own rank-keyed one: this test
+   proves they do NOT collide (RG 26 fires first, unconditionally, on the
+   loser's own [rank = Class4], never reaching a subject comparison at all). *)
+let test_bvm_saturday_loses_to_a_real_feast_end_to_end () =
+  let date = mk 2026 9 12 in
+  let day_ctx = ctx date in
+  let bvm_saturday = of_temporal date in
+  let name_of_mary =
+    cand ~origin:P.Sanctoral ~rank:V.Class3 ~subject:Sub.Bvm ~layer:PE.universal_layer
+      "most-holy-name-of-mary"
+  in
+  let resolution = P.resolve real_rules day_ctx ~temporal:bvm_saturday ~sanctoral:[ name_of_mary ] in
+  Alcotest.(check string) "Most Holy Name of Mary is observed, not the BVM Saturday office"
+    "most-holy-name-of-mary" (S.to_string resolution.P.observed.P.cel.Cel.slug);
+  Alcotest.(check (list string)) "nothing is commemorated" []
+    (List.map (fun (c, _) -> S.to_string c.P.cel.Cel.slug) resolution.P.commemorations);
+  Alcotest.(check (list (pair string string)))
+    "the BVM-Saturday-shaped temporal candidate is omitted under RG26, not commemorated -- no Bvm-vs-Bvm \
+     confusion with RG112(a)-style subject logic"
+    [ ("ef-time-after-pentecost-15-saturday", "omitted: yielded to a higher day") ]
+    (List.map (fun (c, reason) -> (S.to_string c.P.cel.Cel.slug, reason)) resolution.P.omitted)
+
+(* RG112(d), fix round 1 (coordinator finding F1) -- PE.marian_slugs's own
+   citation has the full argument. End-to-end through the REAL pipeline
+   (PE.band, PE.disposition AND PE.admit together): 16 July 2033, a real
+   otherwise-unoccupied Saturday, with the real data/ef/sanctoral.sexp
+   [our-lady-of-mt-carmel] entry as the day's only competing candidate.
+   Before this fix, [resolution.P.commemorations] held
+   [our-lady-of-mt-carmel:Ordinary] (confirmed live against real 2005/2033
+   `colitur day` output before this fix existed); after, it is empty and
+   Mt Carmel is reported omitted, not silently dropped. *)
+let test_bvm_saturday_excludes_mt_carmel_end_to_end () =
+  let date = mk 2033 7 16 in
+  let day_ctx = ctx date in
+  let bvm_saturday = of_temporal date in
+  let mt_carmel =
+    cand ~origin:P.Sanctoral ~rank:V.Class3 ~status:Cel.Commemoration_only ~subject:Sub.Saint
+      ~layer:PE.universal_layer "our-lady-of-mt-carmel"
+  in
+  let resolution = P.resolve real_rules day_ctx ~temporal:bvm_saturday ~sanctoral:[ mt_carmel ] in
+  Alcotest.(check string) "the BVM Saturday office is observed"
+    "ef-time-after-pentecost-5-saturday" (S.to_string resolution.P.observed.P.cel.Cel.slug);
+  Alcotest.(check (list string)) "Mt Carmel is NOT commemorated (RG112(d))" []
+    (List.map (fun (c, _) -> S.to_string c.P.cel.Cel.slug) resolution.P.commemorations);
+  Alcotest.(check (list (pair string string))) "Mt Carmel is reported omitted, not silently dropped"
+    [ ("our-lady-of-mt-carmel", "omitted: yielded to a higher day") ]
+    (List.map (fun (c, reason) -> (S.to_string c.P.cel.Cel.slug, reason)) resolution.P.omitted)
+
 (* RG 16(a) (register §6.0) reached end-to-end through the REAL pipeline
    ([PE.band], [PE.disposition] AND [PE.admit] together, not any one of
    them in isolation as this file's own hand-tagged rows test them
@@ -1895,6 +2053,18 @@ let suite =
           "II-class Sunday override: RG109(b) witness admitted over an ordinary saint, end-to-end"
           `Quick test_ii_class_sunday_privileged_witness_admitted_end_to_end;
         Alcotest.test_case
+          "RG91 e27/RG78: the BVM Saturday office wins and commemorates Holy Machabees, end-to-end \
+           (2026-08-01)"
+          `Quick test_bvm_saturday_wins_and_commemorates_end_to_end;
+        Alcotest.test_case
+          "RG91 e27/RG78/RG26: the BVM Saturday office loses to a real feast and is omitted, end-to-end \
+           (2026-09-12)"
+          `Quick test_bvm_saturday_loses_to_a_real_feast_end_to_end;
+        Alcotest.test_case
+          "RG112(d), fix round 1 (F1): the BVM Saturday office excludes Mt Carmel, end-to-end \
+           (2033-07-16)"
+          `Quick test_bvm_saturday_excludes_mt_carmel_end_to_end;
+        Alcotest.test_case
           "RG16(a): a Feast of the Lord suppresses the Sunday's own commemoration entirely, \
            end-to-end (2028-08-06)"
           `Quick test_rg16a_lord_feast_suppresses_sunday_end_to_end;