Address validation should take multipolygon squares into account

[MinerMax555]

Item 2060 Class 19

Hey, so while resolving issues in a city near me I discovered that some issues persist near what is pretty much the main square of the city, even after I changed the addr:street tags of the addresses around to match those of the square. I suspect this has to do with the fact that this square is mapped as a multipolygon. Have a look at this issue and it's neighbors. Square this address belongs to: https://www.openstreetmap.org/relation/4003691

would be really nice if you could either fix this behaviour or tell me if I am overlooking something. Thanks in advance

[Bibi56]

If it's a square, it should be described with place=square and then the addr field to use is addr:place, not adddr:street

[frodrigo]

The relation have not place tag.

Yes. It is linked to the missing relation support in this analysis.

highway is not build using relation https://github.com/osm-fr/osmose-backend/blob/master/analysers/Analyser_Osmosis.py#L36

[Bibi56]

Note: each address, like https://www.openstreetmap.org/node/2749571952 should also use addr:place instead of addr:street. Yes for multilingual places it would be good associatedStreet would also support addr:place to reduce duplication.

[frodrigo]

@Bibi56 non, no addr:place here since the square (plazza) is not a place=* but rather a highway=pedestrian.

[frodrigo]

@MinerMax555 Did you mark the issues as false-positives?

[frodrigo]

https://github.com/osm-fr/osmose-backend/blob/master/analysers/analyser_osmosis_relation_associatedStreet.py#L570

[Bibi56]

@Bibi56 non, no addr:place here since the square (plazza) is not a place=* but rather a highway=pedestrian.

But as I said, a plazza should be tagged place=square, as you mention it's a square,