deterministic OCaml engine to compute and validate liturgical calendars for multiple rites, template-driven output to year 9999 https://git.labunix.xyz/colitur.git/atom?h=v0.9.0 2026-08-20T15:14:42Z 2026-08-20T15:14:42Z Lukasz Kasprzak lukas@labunix.xyz 2026-08-20T15:14:42Z urn:sha1:d68feb04f71c3dea3b59726e2ffca225d99eaa35 la.ini and en.ini both gain [sigla] (the current Vulgate/Latin punctuation convention, byte-identical to Render.default_style) and [bible] (a .full and .abbr row for every id in Book.all -- 45 cited ids plus the 7 tradition targets, 52 total). Shipping [sigla] changes no output, asserted by the full test run. Shipping [bible] does change rendered book names, which is the point. la.ini's titles are sourced from docs/research/scan1.txt/scan2.txt (the 1962 Missal scans), each row citing the line its incipit pattern was read from. Six pairs (kings_3/4, corinthians_1/2, thessalonians_1/2, timothy_1/2, peter_1/2) share one Missal incipit and differ only in the sourced .abbr, matching what the primary text itself does. Three ids (proverbs, song_of_songs, ecclesiasticus) and the seven tradition targets are marked UNSOURCED and fall back to the data's own spelling, per this project's central rule against inventing a Latin title. Re-running the task's own sourcing note against the scans, counting every hit rather than eyeballing a frequency-sorted list, found nine of its NOT-sourced verdicts were undercounted (a single clean hit, buried under higher-frequency matches): Galatians, Colossians, both Thessalonians, both Peter, Malachi, Numbers, Jonas, Osee and Esdras all have a clean incipit in the scans and are sourced here. The note's other three verdicts stand, confirmed independently. Ecclesiasticus is not simply unfound: this Missal reuses Wisdom's own Lectio libri Sapientiae incipit for Ecclesiasticus readings too (both were anciently classed as one Sapiential group), so using it for Ecclesiasticus would misidentify the book, not merely abbreviate it -- recorded in that row's own comment. en.ini's [bible] is filled in full, not left partial the way [celebration] is: without it, a Vulgate-numbered id would fall through the [meta] fallback chain to la.ini's Latin name, not merely a less complete English one. Traditional Douay-Rheims names for the Vulgate ids (3 Kings, Osee, Ecclesiasticus, Isaias, Apocalypse), modern names for the seven tradition targets, since --sigla-tradition modern is the reader asking for modern numbering. test_lang_coverage.ml gains test_every_book_named, asserting every Book.all id has both forms in la.ini -- the check that catches a forgotten tradition target, since nothing else in the suite ever names them. Shipping real book names changes 11 golden template renders (Render/golden) and several test/cli.t examples that used to demonstrate the pre-Task-10 default-spelling fallback; both are updated to the new, correct output, each line checked against a fresh render before promoting. 2026-08-20T14:43:04Z Lukasz Kasprzak lukas@labunix.xyz 2026-08-20T14:43:04Z urn:sha1:27606c42b7506ab7ffc7f1bd32d4d4a72c6400c8 View.citation_ref and readings' part_ref are the only two places a citation reaches output; every emitter goes through View. --raw passes Sigla.verbatim rather than a style built over Lang.raw: an identity name table would still reformat punctuation and renumber. Loosen reject_sigla_for: --sigla-style/--sigla-book/--sigla-tradition now actually render on readings/table/render/emit/publish (each builds its own Sigla.t via the new load_sigla), and stay refused only on commands that render no citation (day, easter, temporal, ...). names_of degrades a Lang.bible miss to Book.default_spelling rather than printing the lookup key itself ("luke.abbr"); pinned in cli.t against a language file with no [bible] section at all, independent of la.ini/en.ini's own eventual [bible] section. Regenerate the 11 golden templates for 2027: every changed line is one canonical citation replacing a stored variant, verified against the full readings 2027 diff (raw vs default) cell by cell -- 86 distinct (raw, rendered) pairs account for the entire diff across all 11 files, with zero unexplained residue. Two families: the seven duplicate book spellings collapsing onto one canonical form (Isa./Isa, 3 Kgs./3 Kings, ...), and citations reconstructed from parsed structure dropping stray punctuation the parser already treats as noise (a trailing period or semicolon, a comma chapter/verse separator, an elided inherited chapter) -- each of the latter already named and tested in test_citation.ml's parse_suite before this task. 2026-08-20T09:09:34Z Lukasz Kasprzak lukas@labunix.xyz 2026-08-20T09:09:34Z urn:sha1:c30a35e1da315f2a44e5b53c714bdcf4d8992d2c grid.typ used rows: (auto, 1fr) on the assumption that a Typst fractional row always shrinks to share whatever space is left. It does not: an 'fr' track's size is the LARGER of its fair share and its content's own minimum, so it grows PAST 1fr when content is taller than that share rather than compressing to fit. Adding the sigla line (Epistle/Gospel references) grew every cell's natural content past what a 1/6-of-remaining-space row allows, and a 6-week month (three of twelve in 2027) split its own table between rows, stranding the last week alone on a fresh page with no heading -- the user's screenshot showed a row sliced mid-cell, 'I classis' cut off, the next month's own heading printed underneath it. Fixed the same way grid.tex's own \cellh already was: an explicit row height (cellh, computed from the page's text height divided by six, budgeted with 6mm of deliberate slack after a zero-slack first attempt still split a table between rows), with each cell's content wrapped in a height-clipped block sized to match -- content that still does not fit is truncated rather than spilling into the next cell or page. block(breakable: false) around the whole table is a second, independent backstop: even if the budget is ever wrong again, the WHOLE table jumps to a fresh page rather than splitting a row across two, a loud failure (an odd blank page) rather than a silent one (a sliced row). This reverts the file's own prior divergence from grid.tex (Typst stretching every month to fill the page regardless of row count, disclosed as deliberate at the time) -- overflow safety now matters more than that cosmetic fill. A 5-week month still leaves a blank sixth row, the same trade-off grid.tex's own header already accepts. Verified against all twelve months of 2027 (three 6-week: January, May, October; nine 5-week), typeset through typst compile: exactly 12 pages, one month title per page, no clipped or duplicated content, day 31 of a 6-week January fully visible. grid.tex itself re-checked for the same latent risk and found unaffected -- its fixed-height \cellh box already had enough margin. Golden regenerated through the same Test_view.view_of path the other affected templates already use. 2026-08-20T08:07:37Z Lukasz Kasprzak lukas@labunix.xyz 2026-08-20T08:07:15Z urn:sha1:6aebb3ec513c0177a39c2eb8d4b741e90d6f53f4 Adds templates/ef/ordo.typ (A5, one week per page, each day framed with a colour swatch, real names, every fixed string through term.*) and templates/ef/grid.typ (A4 landscape, one month per page, 7 columns, colour-filled cells, Epistle/Gospel sigla), matching what ordo.tex and grid.tex already do. Typst resolves its own table of contents inside a single compile: no pageref/aux-file dance, unlike pdflatex's own two-pass need for ordo.tex. ordo.typ's TOC is one flat heading per week (Month . Week N) rather than ordo.tex's own month-once/weeks-nested layout, since the view model has no first-of-month flag for a template to test and inventing one would be new engine state for a cosmetic grouping. grid.typ deliberately diverges from grid.tex's own partial-fill behaviour for a 5-week month: Typst's row-size list repeats its own last entry for every remaining row, so a two-entry rows spec fills the page completely regardless of week count, where LaTeX's fixed cellh leaves a 5-week month's bottom sixth blank on purpose. Both divergences are documented in the templates' own headers. Two bugs found and fixed while building these against the real typst 0.14.2 binary: two stray hash characters in grid.typ's header comment that were not actually inside a line-comment (would have been parsed as Typst code, not prose); and two lines merging into one wherever they were separated only by a bare newline (still the same soft-wrapped paragraph in Typst, unlike LaTeX's own explicit paragraph break) -- fixed with a trailing backslash to force a real line break without adding inter-paragraph spacing, verified against a full-year render (zero typst warnings across all 365 days of 2027, in both templates). Goldens (test/golden/ordo-2027.typ, test/golden/grid-2027.typ) were generated through the harness's own View.of_days/Template.render_string path, via a throwaway generator, NOT through the CLI: the CLI's default language table differs from the harness's own English-chained-to-Latin table, confirmed live (diffing CLI output against the harness's own render showed exactly the expected language divergence and nothing else) -- the same trap this project has hit before. make check-templates now typesets both templates with the typst binary, using the same SKIPPED-and-exit-0 shape the pdflatex/groff blocks already use when their own tool is absent, verified in both directions: absent (PATH without typst) skips loudly and exits 0, and a deliberately corrupted template fails the target with a non-zero exit and typst's own error text.