Opened 12 years ago

Closed 9 years ago

#1913 closed defect (invalid)

reverse geocoder overwrites best street answer with worst street answer

Reported by: robe Owned by: robe
Priority: medium Milestone: PostGIS 2.1.9
Component: tiger geocoder Version: master
Keywords: Cc: woodbri

Description

Case in point: SELECT * from reverse_geocode(ST_Point(-71.22484166666666, 42.303448333333336) )

Was giving a primary answer of Highland Ave even though I- 95 is closer to the point. This is because the primary answer has no name (is a ramp), and the logic is intended to pick the next best street name assuming they are unnamed. Instead it replaces the best answer with the worst that has a street

Attachments (1)

highland.png (24.6 KB ) - added by robe 9 years ago.

Download all attachments as: .zip

Change History (5)

comment:1 by woodbri, 12 years ago

Cc: woodbri added

comment:2 by robe, 12 years ago

Milestone: PostGIS 2.1.0PostGIS 2.2.0

comment:3 by robe, 9 years ago

Milestone: PostGIS 2.2.0PostGIS 2.1.9

by robe, 9 years ago

Attachment: highland.png added

comment:4 by robe, 9 years ago

Resolution: invalid
Status: newclosed

I'm debating if this is an issue. I think I'll just dismiss. With tiger 2015 data

test_tiger=# SELECT pprint_addy(addy[1]), street[1] from reverse_geocode(ST_Point(-71.22484166666666, 42.303448333333336) ) ;
       pprint_addy        |    street
--------------------------+--------------
 I- 95, Needham, MA 02494 | Highland Ave
(1 row)


And OSM shows this:

Given it's at the cross roads of the interstate around a ramp for Highland, I think this answer is fine.

Note: See TracTickets for help on using tickets.