Hello everybody
I am currently working on a project that fetches information about an address inside the confines of Switzerland. Retrieving the values as JSON, I have noticed that under the “address” key, different naming schemes are used for the neighborhood/suburb/quarter with respect to the city’s address. For example Basel has the schema “house_number”, “road”,“suburb”,“borough”,“city” while the city of Zürich has the schema “house_number”, “road”, “suburb”, “city”. Looking at the structure of how administrative boundaries are defined in openStreetMap Switzerland follows an ambiguous structure that makes each city have a different ranking and also a different key name. This discrepancy of districts can also be confirmed by examining the boundaries map interface. Here, the lowest neighborhood inside Zürich has the ranking of 10, while the one on Basel has the ranking of 9.
Is there a way to correctly retrieve the lowest administrative boundary information of an address without relying on the “address” schema or the admin_level ranking? If so, how could a API call with the corresponding request parameter look like?
None of that is correct, Swiss addresses are always of the format
street_level_localisation number
post_code postal_city
country (optional if using the address outside of Switzerland)
Further, subdivisions of municipalities (admin level 8) only exist formally in a handful of locations, but these would only be address components if they just happen to use the same name as the postal city, just as the municipality name only turns up if it just happens to be the same as the postal city. However in both cases the postal city extent can differ substantially from that of the admin entity with the same name.
Where from?
IIRC, Nominatim “enriches” the returned data with suburb and borough, even though they are not part is Swiss addresses.
I suspect that the confusion might be between postal address and the many other localisations an object might have. For example the many different parishes it might be in, non-admin toponyms for the enclosing area and so on.