Bob,
Glad to be of assistance.
One tiny word of warning when you add role=outer: please make sure you
do it to ALL the ways (without a role) in a relation, because a mix of
empty roles and role=outer in the same relation will definitely cause
more problems than simply having empty roles!
You
Colin
The answer might lie with role=outer. I applied it to Nettlebed in JOSM as a
test, uploaded the change and ran a new query in Overpass turbo. Nettlebed
appears as a polygon now. It seems I should apply it to all my cases for best
practice, as you write, because it does appear other soft
Hi Bob,
They all seem to be made up of multiple ways. I noticed that Bix and
Assendon has role=outer consistently applied, whereas the others have no
explicit role on the ways. This is implicitly equivalent to role=outer
but having explicit roles for the members in a relation may be
considered "b
Colin
I thank you for your prompt reply.
I have been investigating further in the meantime to try to establish what
makes a difference. I wondered if it lay with “type=boundary” or its absence
but my relations I have checked all have type=boundary. I wonder if you have
time to look at South Ox
Hi Bob,
I have been doing a lot of work looking after admin boundaries in the UK
in the last few years, including adding many thousands of Civil Parish
relations.
Admin boundaries are represented by relations with type=boundary.
Syntactically these are similar to multipolygons, whereby the (one
While running an Overpass turbo query recently I noticed a mixture of polygon
and polyline admin_level 10, civil parishes in South Oxfordshire within the
bounding box, but mostly polylines. Its neighbour, South Buckinghamshire,
appears to contain civil parish polygons. I checked other, random
6 matches
Mail list logo
