[openstreetmap/openstreetmap-website] "Query feature" should show relations in a more intuitive way (Issue #6622)
Minh Nguyễn
notifications at github.com
Tue Dec 16 20:14:43 UTC 2025
1ec5 left a comment (openstreetmap/openstreetmap-website#6622)
> the result sometimes gives the impression there's a duplicate object on the map
This is exacerbated by the identical feature type prefixes, which we’re trying to keep consistent with the presentation of Nominatim search results. This is more of a problem with some relation types than others. After all, most boundaries don’t have other boundaries as members, but a `type=waterway` relation by definition has `waterway=*` ways as members. Even in plain language, I’d struggle to succinctly explain the difference without falling back to raw element types.
It looks like the Overpass API response for [this particular query](https://github.com/openstreetmap/openstreetmap-website/blob/a30c8d037e23b2ee48e0aabb9a0e555f27ef8be3/app/assets/javascripts/index/query.js#L275) includes the full list of members of each relation. That’s overkill for the current display, but it does enable the hierarchical display you’re proposing.
If we do choose a hierarchical display, I’d suggest inverting it. That way, multiple members of the same relation get collated under a single relation, in case the user queries for features near multiple members:
* Stream: Houghton Brook
* Stream: Houghton Brook
* Stream: Houghton Brook
I think the sidebar currently only has the ability to display these items in a flat list, so we’d need to style these items slightly differently.
--
Reply to this email directly or view it on GitHub:
https://github.com/openstreetmap/openstreetmap-website/issues/6622#issuecomment-3662211669
You are receiving this because you are subscribed to this thread.
Message ID: <openstreetmap/openstreetmap-website/issues/6622/3662211669 at github.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstreetmap.org/pipermail/rails-dev/attachments/20251216/9248522c/attachment-0001.htm>
More information about the rails-dev
mailing list