Public: anyone can read this; the crew can change it.

This week, on the clock and the printed Monday sheet, shows each closing's time and place, like "Closing · 2:00 pm at Lakeside Title", so the week's closings are planned from one page #144

Merged opened this

Check · Review · Land · Live done — Landed and live in production.

Merged

On This week, a closing row now reads like "Closing · 2:00 pm at Bayside Title, 400 Main St", on both the clock's card and the printed /week sheet. The closing, week and server tests pass. I did not run the full bun test.

Notes

shipped production is serving it deploy d2eada7

opened this

On This week, a closing row now reads like "Closing · 2:00 pm at Bayside Title, 400 Main St", on both the clock's card and the printed /week sheet. The closing, week and server tests pass. I did not run the full bun test. I couldn't run bun x tsc --noEmit because the repo has no tsconfig.json and bun x tsc pulled in a different tool.

  • Wording: the text after "Closing ·" comes from closingWhere in src/closing.ts, so the time and place read the same as on the deal page. I did not reuse closingLine, because it also prints the date ("Closing Nov 21 · …"), which the day heading already gives. A closing with neither a time nor a place reads "Closing", as before.
  • Order: within one day, timed closings come first in time order. The other rows follow in their old order, and untimed closings sort with them. thisWeek in src/week.ts still takes plain data and returns plain rows, and order across days is unchanged. closingNote there builds the text for both pages.
  • Escaping: the place is escaped on both pages.
  • JSON/MCP: the this_week MCP tool exists, so closing rows now also carry closing_time and closing_place, and the tool description says so. /openapi.json doesn't describe MCP tools, so it has nothing to update.
  • Phones: I added no CSS. The text goes in the existing label line and wraps like the rest of the row. I didn't check a phone viewport.
  • Tests:

- src/week.test.ts has a unit test that sorts closings on one day by time. - src/week.page.test.ts checks that the sample closing's time and place show on both the sheet and the clock card. - It also checks that a place like <b>Title</b> & Co is escaped on the sheet.

bun scripts/lint-design.ts printed nothing. I ran that file because the design-lint.ts named in the brief doesn't exist.

To try it, open / or /week on the sample data. The deal with a Bayside Title closing this week should show its time and place.

🤖 Generated with Claude Code

Closes #141

1 earlier comment · 1 event
commented

On This week, a closing row now reads like "Closing · 2:00 pm at Bayside Title, 400 Main St", on both the clock's card and the printed /week sheet. The closing, week and server tests pass. I did not run the full bun test. I couldn't run bun x tsc --noEmit because the repo has no tsconfig.json and bun x tsc pulled in a different tool.

  • Wording: the text after "Closing ·" comes from closingWhere in src/closing.ts, so the time and place read the same as on the deal page. I did not reuse closingLine, because it also prints the date ("Closing Nov 21 · …"), which the day heading already gives. A closing with neither a time nor a place reads "Closing", as before.
  • Order: within one day, timed closings come first in time order. The other rows follow in their old order, and untimed closings sort with them. thisWeek in src/week.ts still takes plain data and returns plain rows, and order across days is unchanged. closingNote there builds the text for both pages.
  • Escaping: the place is escaped on both pages.
  • JSON/MCP: the this_week MCP tool exists, so closing rows now also carry closing_time and closing_place, and the tool description says so. /openapi.json doesn't describe MCP tools, so it has nothing to update.
  • Phones: I added no CSS. The text goes in the existing label line and wraps like the rest of the row. I didn't check a phone viewport.
  • Tests:

- src/week.test.ts has a unit test that sorts closings on one day by time. - src/week.page.test.ts checks that the sample closing's time and place show on both the sheet and the clock card. - It also checks that a place like <b>Title</b> & Co is escaped on the sheet.

bun scripts/lint-design.ts printed nothing. I ran that file because the design-lint.ts named in the brief doesn't exist.

To try it, open / or /week on the sample data. The deal with a Bayside Title closing this week should show its time and place.

🤖 Generated with Claude Code

Changes requested — before the latest push

Display, sort and escaping look right. One miss: the new MCP fields have no test. Add to the this_week test in src/server.test.ts (around line 1071) an assertion that a closing row carries closing_time and closing_place and that a non-closing row does not. Without it, removing the spread in src/mcp.ts would leave every test green. Also confirm bun x tsc --noEmit and the full bun test locally if possible; the PR says neither was run.

commented

The this_week MCP test in src/server.test.ts now covers the new fields, which was the one miss in the review.

  • What it asserts: every closing row carries the closing_time and closing_place keys, and no other row carries either. The sample closing has no time or place set, so the keys are present with a null value. Removing the spread in src/mcp.ts:119 now fails this test.
  • What I ran: bun test src/server.test.ts passes (47 tests). I did not run the full bun test, to avoid running everything twice, since the forge runs it on the push.
  • Not run here: bun x tsc --noEmit has nothing to check in this clone. There is no tsconfig.json, so tsc just prints its help text. There is no scripts/gate.ts either.
  • Not done: I did not add a case with a real time and place set through the API. It would draw on the write allowance that the test's own comment says is shared across test files.

Nothing else changed since the earlier push.

🤖 Generated with Claude Code

Approved

The earlier ask (a test for the MCP closing_time/closing_place fields) is met: the this_week test now checks closing rows carry both keys and other rows carry neither. Nothing else changed since my last review. No permit cells loosened, output escaped, ordering tested. Non-blocking: add a case with a real time and place set over MCP.

Sign in to join this conversation.

↑↓ navigate ↵ open ⌘↵ its app esc close

Keyboard shortcuts

Docs