1 files changed, 131 insertions, 1 deletions
diff --git a/data/ef/adjustments.sexp b/data/ef/adjustments.sexp
index a5e7c6f..bd6fe60 100644
--- a/data/ef/adjustments.sexp
+++ b/data/ef/adjustments.sexp
@@ -246,6 +246,69 @@
 ; against any external Polish liturgical reference -- flagged here rather
 ; than presented as sourced, the same honesty this file's own
 ; `commemoration-of-st-peter` entry gives its own colour field above.
+;
+; ef-sanctoral-audit (2026-08-14, branch ef-sanctoral-audit): a full
+; extraction of the CALENDARIUM (both photographic scans, missale-romanum-
+; 1962.pdf and Missale Romanum 1962_text.pdf, pdftoppm-rendered page images
+; read directly, not their pre-existing OCR text layers, which the task
+; brief's own worked examples show are unreliable for exactly this kind of
+; check) against all 327 base entries plus the two prior overlay Adds.
+; Method and full findings: docs/research/rules-register.md's own audit
+; rewrite (bucket recomputation) and .superpowers/sdd/2026-08-12-colitur-
+; rg16a/sanctoral-audit-report.md. Two shapes of finding follow.
+;
+; `Add barbara` -- a genuine DATA GAP, the same shape as the ef-rebootstrap
+; task's five (agnes-secundo/boniface-martyr/eusebius-confessor/evaristus/
+; theodore, above): both scans, 4 December, carry a second line under S.
+; Petri Chrysologi's own entry, "Commemoratio S. Barbarae Virg. et Mart."
+; -- a real, standing commemoration, absent from `data/ef/sanctoral.sexp`
+; AND from lectio's own `tridentine-calendar.ini` (checked directly: its
+; 12-04 block holds only `peter-chrysologus`, then jumps straight to
+; `sabbas`/12-05) -- not a colitur bootstrap miss, the same upstream gap
+; shape the ef-rebootstrap task's own five entries already documented.
+; Rank `Class3`: bootstrapped by the SAME convention every other
+; `Commemoration_only` entry in this file uses (see the rank-inference note
+; below) -- the calendarium gives a bare "Commemoratio", no class numeral,
+; for any commemoration, hers included. Colour Red: RG 124(e), scan-
+; verified word for word (both photographic scans, "XVIII - De coloribus
+; paramentorum", "C) De colore rubro"): "adhibetur color ruber in Officio
+; et Missa de festis: ... e) Sanctorum Martyrum, quorum colitur aut
+; martyrium..." -- red for the feast of a Martyr's own martyrdom, and the
+; calendarium's own line names her "Virg. et Mart." Names: English a plain
+; translation of "Barbarae"; Polish a plain vernacular rendering -- neither
+; independently primary-sourced, the same honesty this file's own
+; `commemoration-of-st-peter` entry already gives its names.
+;
+; Eighteen `Edit ... ((Set_colour ...))` directives -- confirmed lectio/
+; missalemeum bootstrap DATA DEFECTS, the identical shape and root cause as
+; `eusebius-confessor` above (checked individually against
+; `~/git/projects/lectio/internal/caldata/tridentine-calendar.ini`: every
+; one of these 17 slugs' wrong colour is already wrong AT THE SOURCE, not
+; introduced by colitur's own bootstrap step -- e.g. `martha`'s own Polish
+; field there already reads "św. Marty, Dziewicy" (Virgin, no "Męczennicy"
+; suffix), contradicting that SAME ini record's own `colour = red`). Two
+; rules, both scan-verified word for word against RG 124 ("XVIII - De
+; coloribus paramentorum", missale-romanum-1962.pdf, both photographic
+; scans agree): (124.b) "adhibetur color ruber in Officio et Missa de
+; festis: ... Sanctorum Apostolorum et Evangelistarum, in eorum die
+; natalicio, EXCEPTO FESTO S. IOANNIS (27 decembris)" -- red for an
+; Apostle/Evangelist's OWN feast day, John's 27 December the one named
+; exception (already correctly White in the base data, untouched here);
+; (124.e) "Sanctorum Martyrum, quorum colitur aut martyrium" -- red for a
+; Martyr's own martyrdom. Every plain Confessor/Bishop/Virgin entry below
+; (no "Mart."/"Mm." in the calendarium's own line, not an Apostle) is
+; White by RG 117's five-colour residual (not itself independently
+; RG-124-cited the way the red corrections are, since 124 enumerates the
+; RED exceptions, not a white default) -- flagged as inferred-by-category,
+; the same distinction the design spec draws throughout. `peter` (25
+; January, a commemoration of the Apostle nested under `conversion-of-st-
+; paul`) is corrected by CONSISTENCY with this file's own established
+; precedent, not a fresh RG 124 citation of its own: `paul`'s sibling
+; commemoration (22 February, under `chair-of-st-peter`) already carries
+; Red in the base bootstrap, and `commemoration-of-st-peter` above (30
+; June) was deliberately set Red "to match the host day's own Red" by an
+; earlier task -- the identical RG 110 Peter/Paul-pair shape, so `peter`
+; is the one inconsistent survivor, not a newly-discovered rule.
 ((id ef-adjustments)
   (directives
     ((Suppress vigil-of-christmas)
@@ -265,4 +328,71 @@
             ((slug major-litanies)
               (names ((en "The Major Litanies") (pl "Litanie Większe")))
               (rank Class4) (status Commemoration_only) (colour Violet)
-              (subject Saint) (citations ()) (layer ef-universal))))))))
+              (subject Saint) (citations ()) (layer ef-universal)))))
+      (Add
+        ((date (Fixed (month 12) (day 4)))
+          (cel
+            ((slug barbara)
+              (names
+                ((en "St. Barbara") (pl "\197\155w. Barbary, Dziewicy i M\196\153czennicy")))
+              (rank Class3) (status Commemoration_only) (colour Red)
+              (subject Saint) (citations ()) (layer ef-universal)))))
+            ; A SECOND genuine gap, found by the same audit and DELIBERATELY NOT
+      ; FILLED -- recorded here, in the tracked tree, because the five prior
+      ; upstream-gap precedents live in this file and this is where the next
+      ; auditor will be standing. Both scans' 25 December row reads
+      ; "IN NATIVITATE DOMINI, I classis cum oct." followed by "In secunda
+      ; Missa: Commemoratio S. Anastasiae Mart." -- she is scoped to the
+      ; SECOND of Christmas's three Masses, and it is the only per-Mass
+      ; scoping notation in the whole twelve-month calendarium.
+      ; Celebration.t attaches a commemoration to a DATE with no Mass scope,
+      ; and temporal_ef builds exactly one office per civil day, so adding
+      ; her the way `barbara` was added would ship a false positive on every
+      ; 25 December. The obvious action on rediscovery is the wrong one.
+      ; Open modelling question, register section 6.
+(Edit prisca ((Set_colour Red)))
+      ; REVERTED by the fix-round review: conversion-of-st-paul (25 Jan),
+      ; peter (25 Jan) and chair-of-st-peter (22 Feb) were set Red here on
+      ; RG 124(b), whose own qualifier "in eorum die natalicio" does not
+      ; reach them -- and RG 120(f), fifteen lines ABOVE 124(b) in the same
+      ; column, NAMES both feasts by name and by date in the enumeration of
+      ; WHITE: "f) S. Ioannis Ap. et Ev. (27 decembris); Cathedrae S. Petri
+      ; (22 februarii); Conversionis S. Pauli (25 ianuarii); Nativitatis S.
+      ; Ioannis Baptistae (24 iunii)". RG 475(e) confirms it a third time:
+      ; "in festis nataliciis Apostolorum et Evangelistarum, NECNON in
+      ; festis Cathedrae S. Petri et S. Barnabae Ap." -- the Missal names
+      ; the Chair separately precisely because "natalician feasts of
+      ; Apostles" does not include it. lectio and missalemeum both said
+      ; white because both were right.
+      ; `peter` (25 Jan) additionally rested on "match the host day's own
+      ; Red", by analogy to commemoration-of-st-peter (30 June). That works
+      ; there because 30 June's host is red BY NAME under RG 124(c); here
+      ; the host is white BY NAME under RG 120(f), so the same precedent
+      ; yields White. Its other anchor, `paul` @ 22 Feb (Red in base data),
+      ; is unverified bootstrap on a day the Missal calls white and is now
+      ; an open question in the register, not a precedent.
+      ; ROOT CAUSE, recorded so it is not repeated: the audit's page-image
+      ; method was applied to the CALENDARIUM and not to the RUBRIC. RG 124
+      ; was read from the OCR text layer, where the two-column interleave
+      ; shreds RG 120(f) across rows belonging to RG 115, renders its clause
+      ; label as "/ )" and hyphen-splits "Ca-"/"thedrae" -- so a text-layer
+      ; grep for "Cathedr" returns nothing, while RG 124's own enumeration
+      ; survives the same interleave cleanly. A text-layer read of that
+      ; chapter finds 124 and STRUCTURALLY CANNOT find 120(f).
+      (Edit vitus ((Set_colour Red)))
+      (Edit margaret ((Set_colour Red)))
+      (Edit apollinaris ((Set_colour Red)))
+      (Edit agapitus ((Set_colour Red)))
+      (Edit josaphat ((Set_colour Red)))
+      (Edit john-of-san-fecundo ((Set_colour White)))
+      (Edit ephrem-of-syria ((Set_colour White)))
+      (Edit julia-of-falconieri ((Set_colour White)))
+      (Edit john-gualbert ((Set_colour White)))
+      (Edit camillus-de-lellis ((Set_colour White)))
+      (Edit jerome-emiliani ((Set_colour White)))
+      (Edit liborii ((Set_colour White)))
+      (Edit martha ((Set_colour White)))
+      (Edit alphonsus-liguori ((Set_colour White)))
+      (Edit augustine ((Set_colour White)))
+      (Edit rose-of-lima ((Set_colour White)))
+      (Edit mark-i ((Set_colour White))))))