[kevin9327]

Follow-up to #4208, as discussed there: past and future should handle a leap-day refDate the same way birthdate does.

faker.date.past() and faker.date.future() build their bounds by moving refDate by whole years with setUTCFullYear. When refDate is February 29 and the target year is a common year, the bound rolls over to March 1:

const refDate = new Date(Date.UTC(2024, 1, 29, 12));
faker.date.past({ years: { min: 1, max: 2 }, refDate });
// latest possible result: 2023-03-01T11:59:59Z, which is less than one year before refDate
faker.date.future({ years: { min: 1, max: 2 }, refDate });
// latest possible result: 2026-03-01T12:00:00Z, which is more than two years after refDate

This moves sameDayInYear (added for birthdate in #4208, which clamps February 29 to February 28 in a common year, like date-fns addYears/subYears) to src/modules/date/_utils.ts and uses it for both bounds in past and future. Other dates are unchanged.

Tests: four fixed-randomizer tests (like the existing past/future range tests) pin each end of the range for a 2024-02-29 refDate; all four fail on next and pass with the fix. test/modules/date.spec.ts and test/internal pass locally (775 tests), and no snapshots change.

AI-assisted: I directed the investigation and reviewed and verified the change myself.

[netlify]

✅ Deploy Preview for fakerjs ready!

Built without sensitive environment variables

Name	Link
🔨 Latest commit	f6b12ab
🔍 Latest deploy log	https://app.netlify.com/projects/fakerjs/deploys/6ae31017bc630e76bd936d83
😎 Deploy Preview	https://deploy-preview-4215.fakerjs.dev
📱 Preview on mobile	Toggle QR Code... Use your smartphone camera to open QR code link.
🤖 Make changes	Run an agent on this branch

---

To edit notification comments on pull requests, go to your Netlify project configuration.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.18%. Comparing base (a9d2058) to head (f6b12ab).
⚠️ Report is 1 commits behind head on next.

Additional details and impacted files

@@           Coverage Diff           @@
##             next    #4215   +/-   ##
=======================================
  Coverage   99.18%   99.18%           
=======================================
  Files        1179     1180    +1     
  Lines        3947     3947           
  Branches      713      713           
=======================================
  Hits         3915     3915           
  Misses         28       28           
  Partials        4        4           

Files with missing lines	Coverage Δ
src/modules/date/_utils.ts	100.00% <100.00%> (ø)
src/modules/date/birthdate.ts	100.00% <ø> (ø)
src/modules/date/future.ts	100.00% <100.00%> (ø)
src/modules/date/past.ts	100.00% <100.00%> (ø)

... and 1 file with indirect coverage changes

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[ST-DDT]

I'm not sure about the lower bounds each, as it now is technically one day before the refDate.

[kevin9327]

Good point, thanks. With refDate 2024-02-29, February 28 is the right answer only for the latest bound of each range; for the earliest bound it puts the result one day outside the range (2022-02-28 is more than two years before, 2025-02-28 is less than one year after). Fixed in 4e43327: the earliest bound of past and future keeps the March 1 rollover, the latest bound clamps to February 28. That matches how birthdate treats the max age (1943-03-01). The two tests for the earliest bound now expect March 1. The two tests for the latest bound still fail on next. test/modules/date.spec.ts and test/internal pass (775).

[kevin9327]

Thanks for the suggestions, applied all four in f6b12ab: both test comments now use the "would be less/more than" wording, and in past and future each comment sits right above the line it describes (from → March 1, to → February 28). Comment-only change, no code changes.