docs(specs,plans): amend #230 — drop 2-letter aliases entirely, week-fix only

3-letter English abbreviations are the supported standard. 2-letter aliases
removed from scope due to collision risk with common English words.

#230

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

Author: pajoma

docs/plans/2026-06-01-fix-230-smart-input-week-and-weekday-aliases.md

@@ -1,4 +1,4 @@
-# Plan — Issue #230: Smart input week-text boundary and 2-letter weekday aliases
+# Plan — Issue #230: Smart input week-text boundary fix
 
 > **Spec:** [docs/specs/2026-06-01-fix-230-smart-input-week-and-weekday-aliases.md](../specs/2026-06-01-fix-230-smart-input-week-and-weekday-aliases.md)
 > **Issue:** [pajoma/vscode-journal#230](https://github.com/pajoma/vscode-journal/issues/230)
@@ -7,21 +7,17 @@
 
 ## Approach
 
-Both bugs are addressed in the #177 tokenizer branch. Bug 1 is already resolved by the tokenizer's architecture — `recognizeWeek` uses `(?=\s|$)` as a zero-width lookahead so no character is consumed from the following text. Bug 2 requires adding `mo`, `tu`, `we`, `th` to `weekdayVocab()` (plus `fr` is already covered by the German section). `getDayOfWeekForString` in `dates.ts` already maps all five 2-letter forms correctly; only the vocab recognizer needs updating.
+Bug 1 (week-text boundary) is already resolved by the #177 tokenizer's architecture — `recognizeWeek` uses `(?=\s|$)` as a zero-width lookahead so no character is consumed from the following text. No code change needed; only regression tests required.
 
-**Trade-off:** `we` and `th` are common English words; adding them as standalone weekday tokens introduces the same ambiguity class as German `di`/`do`. This is accepted per spec: the tokenizer marks them `ambiguous` (≤3 chars), which triggers the yellow border feedback in the QuickPick.
+2-letter English aliases are out of scope — 3-letter forms are the supported standard.
 
 ## Steps
 
 1. **Verify Bug 1 is resolved** — add regression test: `parseInput("task week Finalize Shoppinglist")` → `flags="task"`, `text="Finalize Shoppinglist"`, `hasWeek()=true`. No code change needed.
 
-2. **Add 2-letter English aliases to `weekdayVocab()`** — append `'mo', 'tu', 'fr'` after the 3-letter English block (`mon/tue/wed/thu/fri/sat/sun`). `we` and `th` excluded: both are high-frequency English words that would destructively consume text. `fr` already present in German section; adding it to English section makes intent explicit (same result from `getDayOfWeekForString`). Longest-first ordering preserved.
-
-3. **Invalidate pattern cache** — `_weekdayPatterns` is built lazily from `weekdayVocab()`. Adding entries to the array changes the return value; since the cache is built per `MatchInput` instance, no reset needed.
-
-4. **Add regression tests** to `src/test/suite/match-input-vocab.test.ts`:
+2. **Add regression tests** to `src/test/suite/match-input-vocab.test.ts`:
    - `"task week Finalize Shoppinglist"` → Bug 1 regression
-   - `"mo"`, `"tu"`, `"we"`, `"th"`, `"fr"` each → correct weekday offset (same as 3-letter form)
+   - `"week"` → hasWeek=true, no crash
    - `"w15"` → week 15, no regression from `recognizeWeekNum` priority
 
 ## Test scenarios
@@ -31,10 +27,6 @@ Both bugs are addressed in the #177 tokenizer branch. Bug 1 is already resolved
 | Bug 1 regression — week+text | `"task week Finalize Shoppinglist"` | flag=task, text="Finalize Shoppinglist", hasWeek=true |
 | Bug 1 regression — week alone | `"week"` | hasWeek=true, no crash |
 | Bug 1 regression — w15 still numeric | `"w15"` | week=15 |
-| 2-letter alias mo | `"mo"` | same offset as `"mon"` |
-| 2-letter alias tu | `"tu"` | same offset as `"tue"` |
-| 2-letter alias fr | `"fr"` | same offset as `"fri"` |
-| No alias we/th | `"we"`, `"th"` | text-only (not weekday aliases — collision risk) |
 | No regression — task do | `"task do the shopping"` | flag=task, text="the shopping" (do = German Thursday) |
 
 ## Dependencies
@@ -45,13 +37,11 @@ Both bugs are addressed in the #177 tokenizer branch. Bug 1 is already resolved
 
 | Risk | Mitigation |
 |------|-----------|
-| `we`/`th` collision with English words | Excluded from aliases — not added |
-| `fr` duplication (English + German section) | `getDayOfWeekForString("fr", locale)` returns 5 (Friday) for all locales — result identical |
-| `_weekdayPatterns` cache stale | Cache is per-instance, built lazily on first `recognizeWeekday` call — no stale state possible |
+| `recognizeWeek` lookahead differs from old regex fix | Regression tests cover all three inputs from issue report |
 
 ## Rollback
 
-Part of #177 PR. Revert the `weekdayVocab()` addition (two lines) if 2-letter aliases cause regressions after release.
+Part of #177 PR. Tests-only change — no rollback risk.
 
 ## Reference spec
 

docs/specs/2026-06-01-fix-230-smart-input-week-and-weekday-aliases.md

@@ -1,61 +1,52 @@
-# Spec: Smart Input — week-text boundary and 2-letter English weekday aliases (#230)
+# Spec: Smart Input — week-text boundary fix (#230)
 
 ## Goal
 
-Fix two parsing defects in `MatchInput`: (1) the week-token regex consumes the first character of the following text, and (2) 2-letter English weekday abbreviations (`mo`, `tu`, `we`, `th`, `fr`) are not recognized.
+Fix one parsing defect in `MatchInput`: the week-token regex consumes the first character of the following text. 2-letter English weekday aliases are out of scope — 3-letter forms (`mon`, `tue`, `wed`, `thu`, `fri`) are the supported standard.
 
 ## Why now
 
-Both defects break user-visible output in the current 1.1.0 milestone.  The week-text bug silently corrupts task and memo content; the missing aliases were discovered in the same report.
+Defect breaks user-visible output in the current 1.1.0 milestone. The week-text bug silently corrupts task and memo content (e.g. `"task week Finalize Shoppinglist"` → text becomes `"inalize Shoppinglist"`).
 
 ## In scope
 
 - Fix `weekPattern` in `MatchInput.getExpression()` so it does not consume any character beyond the week token itself.
-- Add `mo`, `tu`, `fr` to `getWeekdayPattern()` as recognized English aliases (longest-first ordering; after the 3-letter set). `we` and `th` are excluded — both are high-frequency English words that would destructively consume text (e.g. `"task we need to buy milk"` → `we` parsed as Wednesday).
-- Add regression tests in `src/test/suite/input.test.ts` and/or `week-input.test.ts` covering the exact inputs from the issue report.
+- Add regression tests covering the exact inputs from the issue report.
 
 ## Out of scope
 
+- 2-letter English weekday aliases (`mo`, `tu`, `we`, `th`, `fr`) — 3-letter forms (`mon/tue/wed/thu/fri`) are the supported standard. Adding shorter forms introduces unacceptable collision risk with common English words (`we`, `th`).
 - German 2-letter abbreviation conflicts (`di`, `do`, `fr`, `sa`, `so`) — tracked in #177 (parser rewrite).
 - Locale-based weekday filtering — also #177.
-- Any change to `getDayOfWeekForString` in `dates.ts` (already maps `mo/tu/we/th/fr` correctly).
+- Any change to `getDayOfWeekForString` in `dates.ts`.
 
 ## Acceptance criteria
 
 1. `parseInput("task week Finalize Shoppinglist")` → `flags="task"`, `text="Finalize Shoppinglist"`, `week` = current week number.
-2. `parseInput("task do the shopping")` — behaviour unchanged (deferred; `do` still fires as German Thursday, text = "the shopping"). No regression.
-3. `parseInput("mo")` → `weekday` group matched, `offset` resolves to next/prev Monday relative to today (same as `parseInput("mon")`).
-4. Same for `tu`, `fr`. `we` and `th` are NOT added — they collide with common English words.
-5. `parseInput("w15")` → `week=15` (no regression from week-pattern change).
-6. `parseInput("week")` alone (no following text) → `hasWeek()` true, no crash.
-7. Full test suite green.
+2. `parseInput("task do the shopping")` — behaviour unchanged (`do` still fires as German Thursday, text = "the shopping"). No regression.
+3. `parseInput("w15")` → `week=15` (no regression from week-pattern change).
+4. `parseInput("week")` alone (no following text) → `hasWeek()` true, no crash.
+5. Full test suite green.
 
-## Root causes
+## Root cause
 
 ### Bug 1 — week pattern consumes first character of text
 
 Current:
 ```
 const weekPattern = '(?<week>w(?:eek)?(?:\\s\\D|$))';
 ```
-`(?:\\s\\D|$)` alternates between (whitespace + one non-digit character consumed) and end-of-string.  For input `"week Finalize…"` the `\\D` captures `"F"`, leaving `"inalize…"` in the text group.
+`(?:\\s\\D|$)` alternates between (whitespace + one non-digit character consumed) and end-of-string. For input `"week Finalize…"` the `\\D` captures `"F"`, leaving `"inalize…"` in the text group.
 
 Fix: replace the consuming `\\D` with a lookahead so the non-digit character stays available for the trailing `\\s?` and `textPattern`:
 ```
 const weekPattern = '(?<week>w(?:eek)?)(?=\\s(?!\\d)|$)';
 ```
 `(?=\\s(?!\\d)|$)` asserts (without consuming) that the week token is followed by whitespace-then-non-digit or end-of-string, which prevents `"week15"` from matching this branch (it falls through to `weekNumPattern`).
 
-### Bug 2 — 2-letter English aliases absent from weekday pattern
-
-`getDayOfWeekForString` (dates.ts:86-90) already maps `mo → 1`, `tu → 2`, `we → 3`, `th → 4`, `fr → 5`.  `getWeekdayPattern()` only lists 3-letter English forms (`mon`, `tue`, `wed`, `thu`, `fri`), so the 2-letter variants never reach `resolveWeekday`.
-
-Fix: append `'mo', 'tu', 'fr'` after the 3-letter English group in the `alternatives` array (longest-first ordering preserved). `we` and `th` are excluded — `we` is a common English pronoun and would destructively consume words like "we" in `"task we need to buy milk"`; `th` appears at the start of many English words.
-
 ## Entities / interfaces
 
 - `MatchInput.getExpression()` — `weekPattern` constant (one line change).
-- `MatchInput.getWeekdayPattern()` — `alternatives` array (one section addition).
 - No interface or model changes.
 
 ## Constraints
@@ -66,9 +57,9 @@ Fix: append `'mo', 'tu', 'fr'` after the 3-letter English group in the `alternat
 
 ## Open questions
 
-None after clarification.
+None.
 
 ## Related issues
 
 - Blocks: none.
-- Related: #177 (parser rewrite — will supersede the German 2-letter abbreviation side of this report).
+- Related: #177 (parser rewrite — tokenizer's `recognizeWeek` uses zero-width lookahead by design, so Bug 1 is also fixed there).