Dataset

The toilets, addresses and location notes come from “toilets with bidets in singapore”, a public spreadsheet maintained by @toiletswithbidetsg. It has three tabs — male toilets, female toilets, and hotel rooms — and the app reads all three.

Coordinates are not in the spreadsheet. They’re looked up from each address via OneMap (Singapore Land Authority), falling back to OpenStreetMap contributors, ODbL. Map tiles are OpenFreeMap.

The spreadsheet records wayfinding as free text — “Male: B1, L4 / Handicap: B1, L5, L7”. The app parses that into one structured entry per toilet, so a venue can carry several, and so you can filter by which toilet you need.

Parsing is best-effort and it does get things wrong, which is why raw always keeps the original sentence. If the structure looks off, the sentence is the truth.

audience

“female”, “male”, “unisex”, “accessible”, or “hotel_room”.

levels

Normalised floors: “L5”, “B1”, or “all” when every floor has one. Empty when the source never said.

cubicle

Which cubicle, when named — “all cubicles”, “elderly cubicle”, “cubicle 3”.

landmark

A nearby fixture to steer by — “near NTUC”, “South Wing”.

squat

True when the source describes a squat pan.

raw

The original sentence, untouched.

sheet

Which tab of the source spreadsheet the entry came from.

The underlying records are the work of @toiletswithbidetsg and are used here with attribution; this app claims no licence over them.

What the app does add — the parsed access points, the geocodes and the schema — is offered under CC BY 4.0. If you use it, credit both: “Structured data from Toilets of SG (CC BY 4.0); source records © @toiletswithbidetsg.”

A wrong floor is worse than no floor. Sign in and submit the correction straight from the map — it’s two taps, and it fixes the listing for everyone who looks it up after you.

Same goes for a toilet that isn’t here yet. If you’ve found one the map is missing, add it.