Repository navigation
Josep Carod i Rovira reads family Rovira — Catalan and Polish link surnames with i, which is not a conjunction #397
Description
Activity
- added 7 commits that reference this issue
on Aug 18, 2026 10 remaining items
- added a commit that references this issue
on Sep 22, 2026 Closed by #536.
iis in the default conjunctions, and the common shape is fixed, not only the long one:Josep Carod i Roviragives familyCarod i Rovira,Josep Lluis Carod i Roviragives middleLluis, andCarod i Rovira, Josepgives familyCarod i Rovirawhere it used to drop the link and report a suffixi. Polish reads the same way (Jan Kowalski i Nowak).The three questions this issue asked:
- Add
iat all? Yes, marked inconjunctions_ambiguousbesidee, as the comment above anticipated. In a mixed-case name a lower-caseibetween two name words is a link;IandI.stay initials. In a name written wholly in one case it reads as an initial and reportsconjunction-or-initial. A Catalan or Polish caller who wants one-case records joined removes the marking:Lexicon.default().remove(conjunctions_ambiguous={"i"}). Removingifromconjunctionsrestores every earlier reading exactly, and a property test holds the parser to that. - Should a word that is both a conjunction and suffix vocabulary count toward the carve-out total? Yes, where it is placed to join: a single-letter connective counts as a name word whatever else the vocabulary says it is. And one that is also generational vocabulary joins only where a name word stands on each side, so
John Quincy Smith i,Josep Lluis Carod i IIIandJohn Smith Ikeep their suffix. Review caught the first version testing position instead of class (… Carod i IIIread familyCarod i III); the fix and the invariant that now guards it are in the PR. JOSEP LLUIS CAROD I ROVIRAstays an initial, as predicted, and now says so in its report. The report is kept honest: where a one-caseIends the name and is read as the generation (JOHN QUINCY SMITH I,HENRY I), noconjunction-or-initialis raised, so those names report exactly as they did before.
Accepted and recorded under
P3indocs/design/decisions.md: the two-wordCarod ikeeps the generation reading;Henry i of Englandreads as a connective run, likeDuke of Edinburgh; and because a marked letter reads as an initial in one-case names, a few all-lower spellings move to match their all-caps twins (maier, amy i, jr.gives middlei, asMaier, Amy I, Jr.always has).One interaction surfaced on the way and is fixed in the same PR: a maiden clause ends at any suffix word, and a lower-case
iis the roman numeral, soJane Doe nee Puig i Solerhad always been cut to maidenPuig. A link with a name word of the clause on each side now stays in the birth name, maidenPuig i Soler, as theyspelling always read;Jane Doe nee Puig ikeeps suffixi.rules.md#P3now counts seven single-letter connectives, and the recipe is indocs/customize.rst.- Add
- added 10 commits that reference this issue
on Sep 22, 2026
Catalan (and Polish) link two surnames with
i, butiis not inCONJUNCTIONS, so the link is not joined:The family should be the whole link (
Carod i Rovira), which is how Spanishyalready parses today:Why it was left out, and what measurement says
iis the Roman numeral I (suffixes.py:18), and the concern was the collision. Measured by addingito a customLexicon's conjunctions, generation markers do not collide the way expected — every one of these is unchanged:John Smith I,john smith isuffix='I'John I Smithmiddle='I'john i smithmiddle='i'Maier, Amy I, Jr.itokenThe collision is real, but it lands somewhere else
A word can hold both tags. With
iadded it carriesconjunctionandvocab:suffix, and_is_rootname()returns False for a suffix piece — soishrinks thetotalthat P3's carve-out tests (total < 4in _group.py:200). It therefore needs one more name word thanydoes before it joins:given + two surnamesis the commonest Catalan shape, so adding the word alone fixes the longer names and leaves the common one behind.What needs deciding
iat all — it serves Catalan and Polish.conjunctionandvocab:suffixcount toward the carve-out's rootname total? That is a C-ii question (set-vs-set collision,docs/design/decisions.md), and it is what decides whetherJosep Carod i Rovirais fixed.JOSEP LLUIS CAROD I ROVIRAkeepsIan initial under the Latin-capital veto — that is ShouldJose E Maria Santosjoin likeJose e Maria Santosdoes? (the single-letter connective's Latin-capital veto was never adjudicated) #383.Docs affected
rules.md#P3's prose says the vocabulary holds six single-letter conjunctions (y e и і й و); this makes seven.decisions.md#3-0-reevaluationsalready records that which single letters a tradition wants joined is per-locale policy —iis another instance of it.