diff options
| author | Lukasz Kasprzak <lukas@labunix.xyz> | 2026-08-26 13:30:09 +0200 |
|---|---|---|
| committer | Lukasz Kasprzak <lukas@labunix.xyz> | 2026-08-26 13:30:09 +0200 |
| commit | 9b8821f7b914c94f5b2c8eeda05407ee921b3d5a (patch) | |
| tree | 0aca637921a3792b50e46a2830b869d6b7c4d3ac /lib/kernel/validate.mli | |
| parent | 9cc763c169a6ce4ebbdbed506c5eb2dad1f9f3c1 (diff) | |
| download | colitur-9b8821f7b914c94f5b2c8eeda05407ee921b3d5a.tar.gz colitur-9b8821f7b914c94f5b2c8eeda05407ee921b3d5a.zip | |
feat(of-lectionary): disclose the citation-conversion residual in the header
Fix option (c) from the I2 review finding, done alongside the parser
fix itself rather than deferred: tools/bootstrap_lectionary_of.ml now
runs every emitted reference through Colitur_citation.Parse.parse at
generation time and adds a fourth COVERAGE item plus a named listing
to data/of/lectionary.sexp's own header, so the 41-reference residual
(all chapter-crossing hyphen ranges, see the citation-fix commit) is
disclosed in the artifact a reader actually opens, not only in a task
report or test file.
Regenerated against the pinned lectio snapshot (dune exec
tools/bootstrap_lectionary_of.exe -- ../lectio/internal/caldata/
of-lectionary.ini data/of/lectionary.sexp); diffed against the
pre-change file to confirm the change is header-only, zero lines
removed, the sexp payload byte-identical. Re-pins the file's own
SHA-256 in test_lectionary_of.ml accordingly (re-derived with
sha256sum, not transcribed from generator stdout, matching that
test's own stated discipline).
Diffstat (limited to 'lib/kernel/validate.mli')
0 files changed, 0 insertions, 0 deletions
