id	summary	reporter	owner	description	type	status	priority	milestone	component	version	resolution	keywords	cc
23836	highway=bus_stop should not be suspicious with addr:*	NeatNit	team	"highway=bus_stop is only meant to be used on nodes, and isn't really a highway. I think it shouldn't be marked as suspicious if it also has an address.

I should also mention, in Israel we used to have imports of bus stops from govt data (GTFS) and the data included addresses for a large number of them. These were imported and now there's a lot of this ""issue"" in Israel.

If you think this combination is still suspicious, I'll be happy to hear your opinion why.

I believe this is the line responsible: https://josm.openstreetmap.de/browser/josm/trunk/resources/data/validator/combinations.mapcss#L461


{{{
/* #9811, #11491, #12865, #14310 */
*[place][place!=farm][place!=plot][/^(addr:housenumber|addr:housename|addr:flats|addr:conscriptionnumber|addr:street|addr:place|addr:city|addr:country|addr:full|addr:hamlet|addr:suburb|addr:subdistrict|addr:district|addr:province|addr:state|addr:interpolation|addr:interpolation|addr:inclusion)$/],
/* addr:postcode is used together with place in some countries */
*[boundary][/^addr:/],
*[highway][/^addr:/][highway!=services][highway!=rest_area][!""addr:postcode""] {
  throwWarning: tr(""{0} together with {1}"", ""{0.key}"", ""addr:*"");
  group: tr(""suspicious tag combination"");
  assertMatch: ""node place=foo addr:housenumber=5"";
  assertMatch: ""node place=foo addr:housenumber=5 addr:postcode=12345"";
  assertNoMatch: ""node place=foo  addr:postcode=12345"";
}
}}}

To fix, I have never seen this language before, but I guess add [!highway=bus_stop] to the end of the highway line?"	defect	closed	normal		Core validator	tested	invalid	addr highway