Replace Ebola Zaire with Ebola Sudan to have a small, fast loading, real organism (Zaire is 3k sequences, Sudan 150, we already have large ones with WNV and CCHF). This speeds up time to useful preview.
Change processing of dates to make sampleCollectionDate be of type string and of form YYYY or YYYY-MM or YYYY-MM-DD (instead of type date with YYYY-MM-DD and MM and DD forced to 01 if missing)
Introduce new processed data fields of type date sampleCollectionDateRangeLower and sampleCollectionDateRangeUpper
Upper bound of date range is constrained by submission date (sequences cannot be from samples collected after they were submitted)
Currently no lower bound is chosen, though doing so might be useful for ranged searches - we might decide to introduce an artificial min date for search purposes only, otherwise this might require a LAPIS/SILO feature, see https://github.com/GenSpectrum/LAPIS/issues/1012 for thoughts
Screenshot
PR Checklist
[ ] All necessary documentation has been adapted.
[ ] The implemented feature is covered by an appropriate test.
Discussed on Slack: https://loculus.slack.com/archives/C05G172HL6L/p1732110859357449
preview URL: https://date-ranges.loculus.org
Summary
sampleCollectionDatebe of type string and of formYYYYorYYYY-MMorYYYY-MM-DD(instead of type date withYYYY-MM-DDand MM and DD forced to 01 if missing)sampleCollectionDateRangeLowerandsampleCollectionDateRangeUpperScreenshot
PR Checklist