Problem
13 records in sat_centre.json (all UK schools, all added_by: Shauryagangrade, same batch as #188/#171) have gmaps_link still set to the literal template placeholder instead of a real link:
"gmaps_link": "https://maps.google.com/?q=lat,lng"
That's the raw <lat>,<lng> placeholder text from the record-shape template, never filled in with the record's own coordinates. lat/lng themselves are correct on these records - it's specifically the gmaps_link field that got left as the template string. This passes npm run validate today because the schema only checks the URL's prefix shape, not that q= is followed by real numbers.
Affected ids: 57137, 57159, 57231, 57295, 57210, 57272, 57630, 57296, 57183, 58351, 60245, 59501, 57216.
Fix
For each of the 13, set gmaps_link to https://maps.google.com/?q=<lat>,<lng> using that record's own already-correct lat/lng (or a real Google Maps place link if you have one handy for that school). While you're in there, scripts/validate-places.mjs could also gain a check that gmaps_link doesn't contain the literal string `lat,lng" - happy to see that added in the same PR so this class of bug can't ship silently again.
Acceptance criteria
Problem
13 records in
sat_centre.json(all UK schools, alladded_by: Shauryagangrade, same batch as #188/#171) havegmaps_linkstill set to the literal template placeholder instead of a real link:That's the raw
<lat>,<lng>placeholder text from the record-shape template, never filled in with the record's own coordinates.lat/lngthemselves are correct on these records - it's specifically thegmaps_linkfield that got left as the template string. This passesnpm run validatetoday because the schema only checks the URL's prefix shape, not thatq=is followed by real numbers.Affected ids:
57137,57159,57231,57295,57210,57272,57630,57296,57183,58351,60245,59501,57216.Fix
For each of the 13, set
gmaps_linktohttps://maps.google.com/?q=<lat>,<lng>using that record's own already-correctlat/lng(or a real Google Maps place link if you have one handy for that school). While you're in there,scripts/validate-places.mjscould also gain a check thatgmaps_linkdoesn't contain the literal string `lat,lng" - happy to see that added in the same PR so this class of bug can't ship silently again.Acceptance criteria
gmaps_linkthat actually points at their coordinateslat,lngplaceholder going forward