Clear unavailable selected days after duration changes; disable Continue and remove stale URL selections while preserving valid choices and empty time-picker recovery.
Sort duration choices numerically, independently of valid duration links, configured defaults and booked rescheduling lengths.
Use one compact booking-settings row on day/time/empty states: duration left, timezone right, divider below. Existing dialogs replace the stepper; fixed durations are plain text and final details remain read-only.
Keep selected instants when timezone changes, and clear calendar-day selections that become unavailable.
Each includes a web patch changeset. No new dependencies or role-permission changes.
Validation
Each commit received owner code review, independent correctness/readability reviews (clear), and checks before staging. The final design also received owner visual review.
All web Vitest tests: 1,065 passed after each commit's changes.
Final affected browser journeys: 13 passed, retries disabled — duration/host booking/rescheduling, guest booking, deep links and timezone-crossing absolute slot links.
Earlier commits: existing duration browser suite, 7 passed, retries disabled after each change.
Svelte check: 0 errors / warnings; full web formatting/ESLint, changeset formatting and diff checks passed.
Isolated browser setup built production code without touching developer config/data/build output.
Testing stays focused: one browser regression plus three fast state cases added for the original defect; ordering and shared settings reuse existing tests rather than adding a browser matrix. These are local validation results, not a claim that remote CI has completed.
## Summary
- Clear unavailable selected days after duration changes; disable Continue and remove stale URL selections while preserving valid choices and empty time-picker recovery.
- Sort duration choices numerically, independently of valid duration links, configured defaults and booked rescheduling lengths.
- Use one compact booking-settings row on day/time/empty states: duration left, timezone right, divider below. Existing dialogs replace the stepper; fixed durations are plain text and final details remain read-only.
- Keep selected instants when timezone changes, and clear calendar-day selections that become unavailable.
## Three individually reviewed commits
1. 86dcb56 — unavailable selected-day fix.
2. 5f2e795 — numeric choice ordering, unchanged defaults and permissions.
3. b0ea555 — compact shared duration/timezone settings.
Each includes a web patch changeset. No new dependencies or role-permission changes.
## Validation
Each commit received owner code review, independent correctness/readability reviews (clear), and checks before staging. The final design also received owner visual review.
- All web Vitest tests: **1,065 passed** after each commit's changes.
- Final affected browser journeys: **13 passed, retries disabled** — duration/host booking/rescheduling, guest booking, deep links and timezone-crossing absolute slot links.
- Earlier commits: existing duration browser suite, **7 passed, retries disabled** after each change.
- Svelte check: **0 errors / warnings**; full web formatting/ESLint, changeset formatting and diff checks passed.
- Isolated browser setup built production code without touching developer config/data/build output.
Testing stays focused: one browser regression plus three fast state cases added for the original defect; ordering and shared settings reuse existing tests rather than adding a browser matrix. These are local validation results, not a claim that remote CI has completed.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
Summary
Three individually reviewed commits
86dcb56— unavailable selected-day fix.5f2e795— numeric choice ordering, unchanged defaults and permissions.b0ea555— compact shared duration/timezone settings.Each includes a web patch changeset. No new dependencies or role-permission changes.
Validation
Each commit received owner code review, independent correctness/readability reviews (clear), and checks before staging. The final design also received owner visual review.
Testing stays focused: one browser regression plus three fast state cases added for the original defect; ordering and shared settings reuse existing tests rather than adding a browser matrix. These are local validation results, not a claim that remote CI has completed.