#9836 closed defect (fixed)
crossing barrier and highway is not reported
Reported by: | mkoniecz | Owned by: | team |
---|---|---|---|
Priority: | normal | Milestone: | 20.01 |
Component: | Core validator | Version: | |
Keywords: | barrier highway crossing regression | Cc: |
Description (last modified by )
According to Barriers "If a linear barrier crosses a highway, they must connect with a node at intersection. Always tag this node with the appropriate node-barrier tag" but this is currently not reported. I attach osm testcase.
Attachments (4)
Change History (20)
by , 11 years ago
Attachment: | testcase.osm added |
---|
comment:1 by , 11 years ago
comment:2 by , 11 years ago
Component: | Core → Core validator |
---|---|
Summary: | crossing barrier and highway is not reported by validator → crossing barrier and highway is not reported |
comment:3 by , 5 years ago
Description: | modified (diff) |
---|---|
Keywords: | barrier highway crossing added |
Version: | → tested |
comment:4 by , 5 years ago
Version: | tested |
---|
Still reproducible
Relative:URL: ^/trunk Repository:UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b Last:Changed Date: 2020-01-13 00:44:06 +0100 (Mon, 13 Jan 2020) Revision:15697 Build-Date:2020-01-13 02:30:54 URL:https://josm.openstreetmap.de/svn/trunk
comment:5 by , 5 years ago
I think there is actually no code to detect this case. We report crossing barriers and crossing highways but not a mix of both.
@Dirk: Maybe it was reported before r6581. Do you remember why you added a special test for barriers?
follow-up: 7 comment:6 by , 5 years ago
#8220. ;-)
Highway and barrier was not asked for yet I'd say :-)
comment:7 by , 5 years ago
follow-up: 11 comment:8 by , 5 years ago
I think it was implemented before r6581, but probably the old implementation caused too many false positives?
by , 5 years ago
Attachment: | 9836.patch added |
---|
detect crossings barrier/building, barrier/highway, barrier/railway, barrier/waterway, and barrier/way
comment:9 by , 5 years ago
Hmm, not sure about barrier/way. Probably never happens because of the filter implemented with ignoreWaySegmentCombination()
comment:10 by , 5 years ago
Milestone: | → 20.01 |
---|
comment:11 by , 5 years ago
Should the changes on src/org/openstreetmap/josm/gui/NavigatableComponent.java be part of the patch ?
Replying to GerdP:
I think it was implemented before r6581, but probably the old implementation caused too many false positives?
Yes, was implemented years ago.
comment:12 by , 5 years ago
Hmm. Isn't the idea of unit tests that you expand them when changing code instead of reducing them?
follow-up: 14 comment:13 by , 5 years ago
Not sure if this is a false positive? way 617592269 & way 617380123 give Crossing barrier/highway
barrier=retaining_wall / highway=steps
I see a lot of new crossing warnings in Genua.
comment:14 by , 5 years ago
Replying to GerdP:
Not sure if this is a false positive? way 617592269 & way 617380123 give Crossing barrier/highway
barrier=retaining_wall / highway=steps
I would say no. Either a common node with barrier=entrance
or a spit of the barrier=*
in two pieces is needed.
Alternativly, layer=*
could be used.
I see a lot of new crossing warnings in Genua.
The test was silent for years!
comment:16 by , 5 years ago
Keywords: | regression added |
---|
Of course, they need to have the same layer value.