precedence(ef): fix round 1 (F1) -- band's -14 sentinel inverted the table

entry_14_movable_band = -14 was justified only by "cannot collide
with a real table position (1..28) or unclassified" -- a collision
argument, not an ordering one. band's own value is compared by plain
<, so a negative number beats every I-class entry (1-13), not merely
the fixed Baptism it was built to beat. Not live against the shipped
universal sanctoral data (no I-class universal feast falls 7-13
January, which is why the original 1583-9999 sweep showed only the
intended shape), but entries 12/13 admit PROPER and INDULT I-class
feasts too, and those arrive via adjustments.sexp-style overlays --
this architecture's own advertised extension path. A diocesan
patronal feast landing in Holy Family's window would have made the
inversion live immediately: the I-class feast displaced and
transferred to the next day, reproduced and confirmed before this
fix, reverted after.

Fixed by rescaling every band branch to the real RG 91 entry number
times ten (unclassified untouched at max_int) instead of the bare
1..28, reserving genuine headroom immediately before each entry for
a movable sub-rank -- not an arbitrary rescale: RG 91's own text
records a "primum mobilia, deinde fixa" split at three rows, not
only entry 14 (entry 12's own ninth sub-item, entry 14, and entry
20), so this leaves room for the other two without a second rescale
if either ever gets a live witness. entry_14_movable_band now
derives from entry_14_fixed_band - 1, tied together so they cannot
drift apart.

Added a synthetic I-class end-to-end test reproducing the exact
counter-example that found this: a universal I-class candidate
inside Holy Family's own 7-13 January window is now observed and
does not transfer, with Holy Family itself falling to RG109(a)'s
ordinary privileged-Sunday-commemoration path, the same treatment
any genuine feast outranking an ordinary Sunday gets.

2 files changed, 125 insertions, 65 deletions

diff --git a/lib/rites/rite_ef/precedence_ef.ml b/lib/rites/rite_ef/precedence_ef.ml
index 87cdeeb..39251a7 100644
--- a/lib/rites/rite_ef/precedence_ef.ml
+++ b/lib/rites/rite_ef/precedence_ef.ml
@@ -48,33 +48,72 @@ let universal_layer = "ef-universal"
 let indult_prefix = "indult:"
 let unclassified = max_int
 
-(* Not an RG citation -- an ordering device for RG 91 entry 14's own text
-   (scan-verified, docs/research/rules-register.md §4): "14. Festa Domini II
-   classis, PRIMUM MOBILIA, DEINDE FIXA" -- feasts of the Lord, II class,
-   MOVABLE FIRST, then fixed. Both halves are the SAME numbered table row,
-   but the primary text states a real priority between them, not merely two
-   disjoint categories -- [band]'s own entry-14 branches below must not
-   literally TIE at the shared integer 14, because a tie would fall to
+(* CORRECTED, fix round 1 (coordinator finding 1) -- the ORIGINAL version of
+   this constant was [-14], justified only by "cannot collide with a real
+   table position (1..28) or [unclassified]". That is a COLLISION argument,
+   not an ORDERING one, and [band]'s own return value is compared by
+   {!Precedence.compare_by}/{!compare_precedence} as a plain [<] -- a
+   negative value is not merely "distinct from 1..28", it is LOWER than
+   every one of them, so a movable Class2 Lord feast (Holy Family) would
+   have beaten every I-class day (entries 1-13) outright, not merely the
+   fixed Baptism it was built to beat. Not live against the shipped
+   UNIVERSAL sanctoral data (no I-class universal feast falls 7-13 January),
+   which is why the original 1583-9999 blast radius (§6.2) showed only the
+   one intended shape -- but entries 12/13 admit PROPER and INDULT I-class
+   feasts too (patron/titular/dedication days, {!is_universal}/{!is_indult}
+   below), which arrive via `data/ef/adjustments.sexp`-style overlays, this
+   architecture's own advertised extension point -- a diocesan I-class
+   patronal feast landing 7-13 January would have made the inversion live
+   immediately, transferring the I-class feast as though a II-class Sunday
+   had impeded it. Demonstrated directly (fix round 1, reproduced then
+   reverted): a synthetic Class1 sanctoral candidate on 11 January was
+   OBSERVED as Holy Family (band -14) and the synthetic candidate itself
+   TRANSFERRED to the next day -- backwards on every count.
+
+   FIXED: every {!band} branch now returns the real RG 91 entry number
+   TIMES TEN (entry 1 -> 10, entry 28 -> 280, {!unclassified} untouched at
+   [max_int]) instead of the bare 1..28 -- not an arbitrary rescale, a
+   deliberate one: RG 91's own text records a "primum mobilia, deinde
+   fixa" (movable-then-fixed) split at THREE rows, not only this one --
+   entry 12's own ninth sub-item ("Festa indulta I classis, primum
+   mobilia, deinde fixa"), entry 14 (this one), and entry 20 ("indult
+   II-class feasts (movable then fixed)", register §4's own summary) --
+   and only entry 14 has a live witness in this codebase's data today. The
+   ×10 scale reserves nine integers of genuine headroom immediately BEFORE
+   each real entry's own value for exactly this kind of sub-rank, without
+   requiring a second rescale if entry 12's or entry 20's own movable half
+   ever needs one too -- a value DERIVABLE from RG 91's own table structure,
+   not an arithmetic convenience chosen to dodge one collision. [band]'s own
+   entry-14 comments (below) restate this scale at each branch it touches. *)
+let entry_14_fixed_band = 140
+
+(* RG 91 entry 14's own text (scan-verified, docs/research/rules-register.md
+   §4): "14. Festa Domini II classis, PRIMUM MOBILIA, DEINDE FIXA" -- feasts
+   of the Lord, II class, MOVABLE FIRST, then fixed. Both halves are the
+   SAME numbered table row, but the primary text states a real priority
+   between them, not merely two disjoint categories -- [band]'s own
+   entry-14 branches below must not literally TIE at
+   {!entry_14_fixed_band}, because a tie would fall to
    {!Precedence.resolve}'s own kernel-level fallback ([compare_by],
    Slug.compare), which is deliberately NOT rubric-authoritative (the exact
    anti-pattern RG 113's own fix already corrected once for commemoration
    ordering, this file's own header) and would in fact pick the WRONG side
    here: "commemoration-of-the-baptism-of-the-lord" sorts before
    "ef-time-after-epiphany-sunday-1" alphabetically, backwards from "primum
-   mobilia". Negative, so it can never collide with any real table position
-   (1..28) or [unclassified] (positive [max_int]) by construction -- not a
-   citation to some entry "13.5" that does not exist in the primary text,
-   just an ordering fact this integer has to carry because RG 91's own two
-   half-rows must compare unequal here. The one real witness today: Holy
-   Family (RG 17(b), temporal_ef.ml's own [holy_family_sunday]) against the
-   fixed Commemoration of the Baptism of the Lord (13 January) -- the only
-   date they can ever coincide, since Holy Family only ever falls 7-13
-   January and no OTHER fixed Class2-Lord sanctoral entry shares that
-   window (register §6.0's own subject audit: Transfiguration 6 Aug,
-   Exaltation of the Cross 14 Sep, Dedication of the Archbasilica 9 Nov,
-   the Purification 2 Feb, the Baptism 13 Jan -- the Baptism alone falls in
-   Holy Family's own window). *)
-let entry_14_movable_band = -14
+   mobilia". Derived from {!entry_14_fixed_band}, one better (lower) than
+   it, using the one integer of headroom the ×10 scale reserves immediately
+   before every real entry -- not a citation to some entry "13.5" that does
+   not exist in the primary text, just the ordering fact RG 91's own two
+   half-rows require, expressed the same way the whole table now is. The
+   one real witness today: Holy Family (RG 17(b), temporal_ef.ml's own
+   [holy_family_sunday]) against the fixed Commemoration of the Baptism of
+   the Lord (13 January) -- the only date they can ever coincide, since
+   Holy Family only ever falls 7-13 January and no OTHER fixed Class2-Lord
+   sanctoral entry shares that window (register §6.0's own subject audit:
+   Transfiguration 6 Aug, Exaltation of the Cross 14 Sep, Dedication of the
+   Archbasilica 9 Nov, the Purification 2 Feb, the Baptism 13 Jan -- the
+   Baptism alone falls in Holy Family's own window). *)
+let entry_14_movable_band = entry_14_fixed_band - 1
 
 let is_indult layer = String.starts_with ~prefix:indult_prefix layer
 let is_universal layer = String.equal layer universal_layer
@@ -155,8 +194,12 @@ let band (ctx : Vocab_ef.season Precedence.context) (c : Vocab_ef.rank Precedenc
   (* Named so entry 8's Sunday exception below can read "one worse than the
      Sunday it must yield to" rather than a bare integer that happens to
      equal entry 15's own value; entry 15's own branch returns this same
-     binding, not a second literal, so the two can never drift apart. *)
-  let entry_15_band = 15 in
+     binding, not a second literal, so the two can never drift apart.
+     CORRECTED, fix round 1 (coordinator finding 1): every branch below now
+     returns the real RG 91 entry number TIMES TEN, not the bare 1..28 --
+     see {!entry_14_fixed_band}'s own comment for why (a genuine ordering
+     bug the bare scale could not express, not a cosmetic rename). *)
+  let entry_15_band = 150 in
   let open Vocab_ef in
   (* CORRECTED (Task B fix round 1, ef-rg16a): a [Commemoration_only]
      celebration has NO row in RG 91's table at all, checked FIRST, ahead of
@@ -199,9 +242,9 @@ let band (ctx : Vocab_ef.season Precedence.context) (c : Vocab_ef.rank Precedenc
      ea22ad2. The guard is right; its former justification was not. *)
   if status = Celebration.Commemoration_only then unclassified
     (* 1: Nativity, Easter Sunday, Pentecost Sunday (I class w/ octave). *)
-  else if is_temporal && rank = Class1 && ((m = 12 && d = 25) || off = 0 || off = 49) then 1
+  else if is_temporal && rank = Class1 && ((m = 12 && d = 25) || off = 0 || off = 49) then 10
     (* 2: Sacred Triduum (Thu-Sat of Holy Week). *)
-  else if is_temporal && rank = Class1 && off >= -3 && off <= -1 then 2
+  else if is_temporal && rank = Class1 && off >= -3 && off <= -1 then 20
     (* 3: Epiphany, Ascension, Holy Trinity, Corpus Christi, Sacred Heart,
        Christ the King. *)
   else if is_temporal && rank = Class1
@@ -209,25 +252,25 @@ let band (ctx : Vocab_ef.season Precedence.context) (c : Vocab_ef.rank Precedenc
              || off = 39 (* Ascension *) || off = 56 (* Trinity *)
              || off = 60 (* Corpus Christi *) || off = 68 (* Sacred Heart *)
              || Date.compare date (Temporal_ef.christ_the_king (Date.year date)) = 0)
-  then 3
+  then 30
     (* 4: Immaculate Conception, Assumption BVM. *)
-  else if (not is_temporal) && rank = Class1 && ((m = 12 && d = 8) || (m = 8 && d = 15)) then 4
+  else if (not is_temporal) && rank = Class1 && ((m = 12 && d = 8) || (m = 8 && d = 15)) then 40
     (* 5: Vigil & Octave day of the Nativity. *)
-  else if is_temporal && rank = Class1 && ((m = 12 && d = 24) || (m = 1 && d = 1)) then 5
+  else if is_temporal && rank = Class1 && ((m = 12 && d = 24) || (m = 1 && d = 1)) then 50
     (* 6: Sundays of Advent, Lent, Passiontide, and Low Sunday. *)
   else if is_temporal && rank = Class1 && is_sunday
           && (season = Advent || season = Lent || season = Passiontide || off = 7)
-  then 6
+  then 60
     (* 7: I-class ferias not above -- Ash Wednesday; Mon/Tue/Wed of Holy Week.
        Thu-Sat of Holy Week are the Triduum, entry 2 above, not this entry. *)
-  else if is_temporal && rank = Class1 && (off = -46 || (off >= -6 && off <= -4)) then 7
+  else if is_temporal && rank = Class1 && (off = -46 || (off >= -6 && off <= -4)) then 70
     (* 8: All Souls -- RG 91 entry 8's own text (§4) carries a qualifier this
        transcription must honour: "yields to an occurring Sunday". 2 November
        is always Time_after_pentecost (well clear of Advent/Lent/Passiontide
        and of every other entry's own Easter-relative or fixed date), so a
        Sunday landing on it is always an ordinary entry-15 II-class Sunday --
        the one and only rival this exception ever has to lose to. On such a
-       Sunday this returns [entry_15_band + 1]: strictly worse than 15 (an
+       Sunday this returns [entry_15_band + 1]: strictly worse than 150 (an
        exact tie would fall to Precedence.resolve's slug tie-break, which
        for "ef-all-souls" against a "ef-time-after-pentecost-sunday-*" slug
        would make All Souls WIN -- the precise bug this guards against), but
@@ -237,20 +280,20 @@ let band (ctx : Vocab_ef.season Precedence.context) (c : Vocab_ef.rank Precedenc
        feasts transfer) still sees the true I-class candidate it needs to
        move to 3 November. *)
   else if (not is_temporal) && rank = Class1 && m = 11 && d = 2 then
-    if is_sunday then entry_15_band + 1 else 8
+    if is_sunday then entry_15_band + 1 else 80
     (* 9: Vigil of Pentecost. *)
-  else if is_temporal && rank = Class1 && off = 48 then 9
+  else if is_temporal && rank = Class1 && off = 48 then 90
     (* 10: Days within the Octaves of Easter and Pentecost. *)
   else if is_temporal && rank = Class1 && ((off >= 1 && off <= 6) || (off >= 50 && off <= 55))
-  then 10
+  then 100
     (* 11: I-class feasts of the universal Church not above. *)
-  else if (not is_temporal) && (not is_vigil) && rank = Class1 && is_universal layer then 11
+  else if (not is_temporal) && (not is_vigil) && rank = Class1 && is_universal layer then 110
     (* 12: Proper I-class feasts. *)
-  else if (not is_temporal) && (not is_vigil) && rank = Class1 && not (is_indult layer) then 12
+  else if (not is_temporal) && (not is_vigil) && rank = Class1 && not (is_indult layer) then 120
     (* 13: Indult I-class feasts. By elimination once 11 and 12 have failed:
        not the universal layer (11), and marked as an indult overlay (12's
        "not indult" test having just failed). *)
-  else if (not is_temporal) && (not is_vigil) && rank = Class1 then 13
+  else if (not is_temporal) && (not is_vigil) && rank = Class1 then 130
     (* 14, MOVABLE half -- RG 91 entry 14's own "primum mobilia, deinde
        fixa" ({!entry_14_movable_band}'s own comment has the full citation
        and reasoning). Checked before the FIXED half immediately below,
@@ -272,30 +315,32 @@ let band (ctx : Vocab_ef.season Precedence.context) (c : Vocab_ef.rank Precedenc
        occurring II-class Sunday with no universal qualifier either). No
        layer test here, unlike 11/12/13 and 16/19/20: the register does not
        split this entry into universal/proper/indult, so a proper or
-       indult feast of the Lord still bands 14, not 19/20. *)
-  else if (not is_temporal) && (not is_vigil) && rank = Class2 && subject = Subject.Lord then 14
+       indult feast of the Lord still bands {!entry_14_fixed_band}, not
+       19/20. *)
+  else if (not is_temporal) && (not is_vigil) && rank = Class2 && subject = Subject.Lord then
+    entry_14_fixed_band
     (* 15: Sundays, II class (every Sunday not already named at 6). *)
   else if is_temporal && rank = Class2 && is_sunday then entry_15_band
     (* 16: II-class feasts of the universal Church, not of the Lord. *)
-  else if (not is_temporal) && (not is_vigil) && rank = Class2 && is_universal layer then 16
+  else if (not is_temporal) && (not is_vigil) && rank = Class2 && is_universal layer then 160
     (* 17: Days within the Octave of the Nativity (26-28 Dec are Stephen,
        John, the Innocents -- sanctoral, not this entry). *)
-  else if is_temporal && rank = Class2 && m = 12 && (d = 29 || d = 30 || d = 31) then 17
+  else if is_temporal && rank = Class2 && m = 12 && (d = 29 || d = 30 || d = 31) then 170
     (* 18: II-class ferias -- Advent 17-23 Dec; Ember days of Advent, Lent,
        September. *)
   else if is_temporal && rank = Class2
           && ((season = Advent && m = 12 && d >= 17 && d <= 23) || is_ember_18 slug)
-  then 18
+  then 180
     (* 19: Proper II-class feasts. *)
-  else if (not is_temporal) && (not is_vigil) && rank = Class2 && not (is_indult layer) then 19
+  else if (not is_temporal) && (not is_vigil) && rank = Class2 && not (is_indult layer) then 190
     (* 20: Indult II-class feasts. By elimination, as at 13. *)
-  else if (not is_temporal) && (not is_vigil) && rank = Class2 then 20
+  else if (not is_temporal) && (not is_vigil) && rank = Class2 then 200
     (* 21: II-class vigils (Ascension, Assumption, John Baptist, Peter & Paul
        -- can be temporal- or sanctoral-origin, see the file comment above). *)
-  else if rank = Class2 && is_vigil then 21
+  else if rank = Class2 && is_vigil then 210
     (* 22: Ferias of Lent and Passiontide (Thursday after Ash Wednesday to the
        Saturday before Palm Sunday), except the Ember days (18 above). *)
-  else if is_temporal && rank = Class3 && (season = Lent || season = Passiontide) then 22
+  else if is_temporal && rank = Class3 && (season = Lent || season = Passiontide) then 220
     (* 23: III-class feasts in particular calendars. Unlike 11/12 and 14/16
        above, the universal entry (24) is the HIGHER number here -- RG 91's
        own table ranks a particular-calendar III-class feast ahead of a
@@ -303,13 +348,13 @@ let band (ctx : Vocab_ef.season Precedence.context) (c : Vocab_ef.rank Precedenc
        the register states it, not "corrected" into the other classes'
        pattern. RG 91 has no indult sub-rank at III class, so every non-base
        layer lands here, not split further. *)
-  else if (not is_temporal) && (not is_vigil) && rank = Class3 && not (is_universal layer) then 23
+  else if (not is_temporal) && (not is_vigil) && rank = Class3 && not (is_universal layer) then 230
     (* 24: III-class feasts in the universal calendar. *)
-  else if (not is_temporal) && (not is_vigil) && rank = Class3 then 24
+  else if (not is_temporal) && (not is_vigil) && rank = Class3 then 240
     (* 25: Ferias of Advent to 16 Dec, except the Ember days (18 above). *)
-  else if is_temporal && rank = Class3 && season = Advent then 25
+  else if is_temporal && rank = Class3 && season = Advent then 250
     (* 26: III-class vigils (St Lawrence). *)
-  else if rank = Class3 && is_vigil then 26
+  else if rank = Class3 && is_vigil then 260
     (* 27: Office of the BVM on Saturday -- every otherwise-unoccupied IV-class
        Saturday, per the historical default that fills it; ordinary Mass
        propers still make Rogation Mon/Tue/Wed proper without changing the
@@ -318,12 +363,12 @@ let band (ctx : Vocab_ef.season Precedence.context) (c : Vocab_ef.rank Precedenc
        vigils for the same reason 11-13/14/16/19/20/23/24 do: RG 91 has no
        IV-class vigil at all (RG 91's own vigil list, §4 "Vigils",
        stops at III class), so one would be an anomaly, not this entry. *)
-  else if is_temporal && (not is_vigil) && rank = Class4 && weekday = Date.Sat then 27
+  else if is_temporal && (not is_vigil) && rank = Class4 && weekday = Date.Sat then 270
     (* 28: IV-class ferias -- the unqualified catch-all (temporal_ef.ml's own
        comment on [ferial_rank] cites the same primary text, "Feriae IV
        classis"). Excludes vigils for the same reason as 27 above: a IV-class
        "feria" that is also a vigil is not a feria RG 91 describes. *)
-  else if (not is_vigil) && rank = Class4 then 28
+  else if (not is_vigil) && rank = Class4 then 280
   else unclassified
 
 (* Task 8: what happens to the day's LOSING candidate (docs/research/
diff --git a/lib/rites/rite_ef/precedence_ef.mli b/lib/rites/rite_ef/precedence_ef.mli
index 86a0ead..239a254 100644
--- a/lib/rites/rite_ef/precedence_ef.mli
+++ b/lib/rites/rite_ef/precedence_ef.mli
@@ -62,25 +62,40 @@ val ember_prefixes : string list
     new rule, not a silently wrong one. *)
 val unclassified : int
 
+(** {!band}'s value for RG 91 entry 14's FIXED half (a sanctoral, i.e.
+    fixed-date, Class2 feast of the Lord) -- see {!entry_14_movable_band}'s
+    own comment for why entry 14 needs two distinct values at all. *)
+val entry_14_fixed_band : int
+
 (** RG 91 entry 14's own text (§4): "Festa Domini II classis, primum
     mobilia, deinde fixa" -- feasts of the Lord, II class, MOVABLE first,
     then fixed. {!band}'s value for the movable half (Holy Family, RG
-    17(b)) -- strictly better than the fixed half's own literal [14],
-    negative so it can never collide with a real table position or
-    {!unclassified}. See precedence_ef.ml's own comment for the full
-    citation and why the two halves must not literally tie. *)
+    17(b)) -- [entry_14_fixed_band - 1], strictly better than the fixed
+    half without colliding with entry 13 or any other real table position.
+    CORRECTED, fix round 1 (coordinator finding 1): this used to be a
+    negative sentinel, justified only by "cannot collide with a real table
+    position" -- a COLLISION argument, not an ORDERING one; since {!band}
+    values are compared by plain [<], a negative value beats every I-class
+    entry (1-13) as well as the fixed half, not merely the fixed half it
+    was built to beat. {!band}'s own entries now use the real RG 91 number
+    TIMES TEN throughout, reserving genuine headroom before every entry --
+    see precedence_ef.ml's own comment for the full citation, the counter-
+    example that found this, and why entries 12 and 20 may need the same
+    treatment if their own "movable then fixed" halves ever get a witness. *)
 val entry_14_movable_band : int
 
 (** [band ctx c]: RG 91's Table of Precedence. Returns the table's own entry
-    number -- I class 1-13, II class 14-21, III class 22-26, IV class 27-28;
-    lower wins (see {!Precedence.rules.band}) -- EXCEPT where the table's own
-    text states an exception: entry 8 (All Souls) reads "yields to an
-    occurring Sunday", so on a Sunday this returns a value that
-    loses to entry 15 rather than the literal integer 8 (see the comment on
-    entry 8 in precedence_ef.ml for the exact value and why). ALSO
-    {!unclassified} for any [Cel.Commemoration_only] candidate, checked
-    first, ahead of every rank-keyed branch -- CORRECTED, Task B fix round
-    1 (coordinator finding 1): RG 91's own table enumerates only "dies
+    number TIMES TEN -- I class 10-130, II class 140-210, III class 220-260,
+    IV class 270-280; lower wins (see {!Precedence.rules.band}) -- EXCEPT
+    where the table's own text states an exception: entry 8 (All Souls)
+    reads "yields to an occurring Sunday", so on a Sunday this returns a
+    value that loses to entry 15 (150) rather than 80 (see the comment on
+    entry 8 in precedence_ef.ml for the exact value and why); entry 14
+    (Feasts of the Lord, II class) splits into {!entry_14_movable_band} and
+    {!entry_14_fixed_band} per the table's own "primum mobilia, deinde
+    fixa". ALSO {!unclassified} for any [Cel.Commemoration_only] candidate,
+    checked first, ahead of every rank-keyed branch -- CORRECTED, Task B fix
+    round 1 (coordinator finding 1): RG 91's own table enumerates only "dies
     liturgici" (entry 24's own text, "Festa III classis..." -- FEASTS), so
     a celebration reduced to a bare commemoration was never a row in the
     table to begin with, confirmed by the calendarium's own notation (a