[derek73]

Closes #791.

What changes

si and chi were unambiguous post-nominals from the 2019 bulk import. Both are common surnames (US Census 2010: Si 1,166 family-name bearers, Chi 7,273; measured with tools/name_usage.py on 2026-10-21), so they join the ambiguous set that ba and meng use (may_be_name = true, issue = 791). Their bearer counts sit in family and given readings with as_of.

• Luo Si gives family Si, where 2.0.0 through 2.5.0 gave suffix Si with no family name.
• Wang, Si gives given Si, where those releases gave suffix Si with no given name.
• A Title-case Si after a full name takes the family name, as Ma does (John Smith Si: middle Smith, family Si).
• A capitalized SI after a full name is still a credential: John Smith SI gives suffix SI, flagged.

The ambiguity is reported (suffix-or-name), so the caller can see the fork.

Measured moves

• Corpus: no real corpus name holds si or chi as a word, so the snapshot moves nothing. Four shape-tagged rows (Luo Si, Min Chi, Wang, Si, Smith, Chi) are new entries in corpus_shapes.jsonl and snapshot.jsonl.
• Order grid (144 cells per order; the re-runnable recipe is in the decisions entry): 104 moves under GIVEN_FIRST, 100 under FAMILY_FIRST, 100 under FAMILY_FIRST_GIVEN_LAST. No moved word is dropped.
• Accepted costs, each pinned by a case row or recorded in the entry: John Smith Si and John Smith Si. read family (the ma/Ma shape, S6); John Smith Chi PhD stops at the declined pick; John, Si. reads title, as John, Ma. does; John Smith (Chi) reads nickname.
• Known wording defect: John, Si. reports "read as a credential" while the word lands in the title field. Filed as Reports say "read as a credential" for a period-marked title after a comma #793 (Reports say "read as a credential" for a period-marked title after a comma #793). The field-naming test cannot take a row for it until that is fixed.

Where it is recorded

• nameparser/vocab/en.toml: the two entries, with family and given readings and notes, and the rebuilt _compiled.py.
• tests/v2/cases.py: 17 #791 rows, pinning the fork with the contrasting spellings (John Smith SI, John Smith, SI, John Smith S.I.), the empty and nickname heads, the caps lean, and the family-first roles. test_every_case_name_is_unique guards the table against repeated names.
• docs/design/decisions/suffix-acronym-collisions.md: a dated 2026-10-21 entry with the precedence (S6 before S4), the period behaviour, the accepted costs, and the order-grid recipe.
• docs/release_log.rst: a Behavior Changes bullet under 2.6.0; the Describe each word by its readings: per-word TOML vocabulary compiled to today's sets (phase 1) #737 bullet now says the re-encoding changes no word list and no parse.
• AGENTS.md: "a trailing period-marked word the UNAMBIGUOUS SUFFIX vocabulary knows is still a suffix".

Checks

• uv run pytest: 22,330 passed, 125 skipped, 3 xfailed.
• uv run mypy and uv run ruff check: clean.
• snapshot.py and cldr_agreement.py match their records. The JA view needs namedivider, which is not installed locally; CI's ja-extra job runs it.
• The design-docs reviewer reviewed the entry twice, and the code, test and comment reviewers reviewed the PR. Their findings are applied in b1907247 and b23fb7b1; the remaining test gaps (suffix runs, all-lowercase si, the absence contrast, a title head) are follow-ups.

🤖 Generated with Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.05%. Comparing base (e766bf1) to head (b14e4df).

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #794   +/-   ##
=======================================
  Coverage   99.05%   99.05%           
=======================================
  Files          47       47           
  Lines        4642     4642           
=======================================
  Hits         4598     4598           
  Misses         44       44           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.