decentraland / events / 35110078221

67%

push

github

fix: reject event submissions with past end dates (#1024) (#1025)

* fix: reject event submissions with past end dates on create and update

Both createEvent and updateEvent accepted events whose computed end date
(finish_at) was already in the past, or whose recurrent_until boundary
was already past. This allowed creators to silently publish events that
would never appear in the listing.

Changes:
- createEvent: add temporal checks after calculateRecurrentProperties.
  finish_at <= now rejects events fully in the past (one-time and
  recurring). recurrent_until <= now catches recurring events whose
  series boundary is past even when start_at is future.
- updateEvent: mirror the same checks, conditioned on date-related
  fields being touched (matching the existing touchesRecurrence pattern)
  so routine metadata edits (name, description, image) to a recently-
  ended event are not blocked.
- Move const now = new Date() in createEvent above the new checks to
  avoid duplicate declaration (the later use was moved up).
- Add createEvent.test.ts with tests for both rejection cases.
- Add updateEvent.test.ts cases: past finish_at blocked, metadata-only
  edit of past event allowed, past recurrent_until blocked.

Closes #1024

* fix: address review feedback — formatting and selector-field validation bypass

- Run Prettier on all changed files to fix P0 CI failures
- Add recurrent_setpos, recurrent_monthday, recurrent_weekday_mask, and
  recurrent_month_mask to touchesDates guard in updateEvent so a PATCH
  changing only selector fields still triggers the past-date validation
- Add selector-only regression test (recurrent_weekday_mask → 0 eliminating
  all future dates) to cover the P1 bypass

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

---------

Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>

777 of 1071 branches covered (72.55%)

Branch coverage included in aggregate %.

66 of 66 new or added lines in 2 files covered. (100.0%)

4290 of 6455 relevant lines covered (66.46%)

7.72 hits per line