aboutsummaryrefslogtreecommitdiff
path: root/CLAUDE.md
diff options
context:
space:
mode:
authorLukasz Kasprzak <lukas@labunix.xyz>2026-08-18 10:31:22 +0200
committerLukasz Kasprzak <lukas@labunix.xyz>2026-08-18 10:31:22 +0200
commit66162b52b1e7c50fa945cab34e0a184b2623967b (patch)
tree1f5b21d5da301d4381070ef1fc1fd96b6c191542 /CLAUDE.md
parente25558fee984837389d0e53cd6e17a3820f8e5ea (diff)
downloadcolitur-66162b52b1e7c50fa945cab34e0a184b2623967b.tar.gz
colitur-66162b52b1e7c50fa945cab34e0a184b2623967b.zip
docs(ef): close C6 and M26; nothing is left adjudicated against colitur
C6 was closed by the Nativity-Octave fix rather than by argument. Its own note had already split the 138 rows into 119 weekday rows, where it recorded colitur's citation as confirmed correct and said verdict colitur would be accurate for that subset alone, and 19 Sunday rows where colitur emitted the weekday formulary against RG 69 and was plainly wrong -- a Temporal_ef defect the entry deliberately left out of its own scope. That defect is fixed, and the count fell 138 to 119: exactly the population the note had enumerated by date. Verdict open becomes colitur on reasoning already written there. M26 closed on a positive argument, not an absence. Both its other shapes are now built -- the BVM Saturday Mass selection and Trinity's week -- leaving Christ the King's week. RG 299's rule has an escape clause and the escape requires an assigned proper Mass; none is assigned for that week, checked at the source, unlike Trinity's week which prints one and unlike Holy Name and Holy Family which each carry a ferial rubric. That absence is not a research gap, it is the condition that triggers RG 299's default -- and by the default the ferias say the preceding Sunday's Mass, which on this week is Christ the King's own. The limit is stated in the entry rather than glossed: this reads a printed silence as the default rather than as an oversight, which the text cannot distinguish. It is the same reading applied everywhere else RG 299 governs, and applying it inconsistently here to match an oracle already adjudicated wrong six times would be the worse error. No entry in either allow-list is now verdict open or verdict missalemeum. The only non-colitur entries are M15 and M18, both unresolvable, and both are comparator limits rather than calendar disagreements: a temporal-origin celebration carries no English name, so its identity cannot be matched even where both streams plainly mean the same thing. Register: section 6.18.
Diffstat (limited to 'CLAUDE.md')
0 files changed, 0 insertions, 0 deletions