Possible wrong warning about house numbers in interpolation

[naoliv]

Validating ​https://www.openstreetmap.org/way/259489474 it's possible to see that JOSM gives a warning about "Decreasing house numbers in addresses interpolation"

Reversing the way direction the warning quiets down.

At least reading ​http://wiki.openstreetmap.org/wiki/Addresses#Using_interpolation I can't find anything saying that the way direction is important for interpolations.
Probably JOSM should ignore the way direction.

Also, it could be improved to detect an odd number in an even interpolation (and vice-versa)

JOSM:

Repository Root: http://josm.openstreetmap.de/svn
Build-Date: 2014-07-03 22:03:52
Last Changed Author: Don-vip
Revision: 7287
Repository UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b
URL: http://josm.openstreetmap.de/svn/trunk
Last Changed Date: 2014-07-04 00:00:19 +0200 (Fri, 04 Jul 2014)
Last Changed Rev: 7287

Identification: JOSM/1.5 (7287 pt_BR) Linux Debian GNU/Linux unstable (sid)
Memory Usage: 247 MB / 4029 MB (144 MB allocated, but free)
Java version: 1.7.0_55, Oracle Corporation, OpenJDK 64-Bit Server VM
Java package: openjdk-7-jre:amd64-7u55-2.4.7-2
VM arguments: [-Dawt.useSystemAAFontSettings=on, -Xms256M, -Xmx4g, -XX:+UseConcMarkSweepGC, -XX:+UseCompressedOops]
Dataset consistency test: No problems found

Plugin: AddrInterpolation (30416)
Plugin: Create_grid_of_ways (30416)
Plugin: FixAddresses (30416)
Plugin: ImageryCache (30416)
Plugin: OpeningHoursEditor (30416)
Plugin: PicLayer (30436)
Plugin: SimplifyArea (30416)
Plugin: buildings_tools (30485)
Plugin: editgpx (30416)
Plugin: geochat (30416)
Plugin: geotools (30416)
Plugin: jts (30416)
Plugin: merge-overlap (30416)
Plugin: notes (v0.9.2)
Plugin: opendata (30508)
Plugin: pdfimport (30416)
Plugin: poly (30495)
Plugin: reverter (30436)
Plugin: tagging-preset-tester (30416)
Plugin: todo (29154)
Plugin: turnrestrictions (30454)
Plugin: undelete (30416)
Plugin: utilsplugin2 (30460)

[Klumbumbus]

Yes, this warning is wrong. The way direction doesn't matter. Funny thing is that this test was implemented as result of your ticket #9667 :) It seems there were some missunderstandings 2 years ago. I think we should remove it.

According "detect an odd number in an even interpolation (and vice-versa)" I think in the long run we should ret rid of address interpolations, so such a warning wouldn't have much gain. I think the following rule (at info level) would be better:

way[addr:interpolation] {
        throwOther: tr("Way with {0}. Tag each housenumber separately if possible.", "{0.key}");
}

[Klumbumbus]

In 10084/josm:

fix #10232, see #9667 - remove wrong "Decreasing house numbers in addresses interpolation" warning; add "Even housenumber in odd address interpolation." warning and vice versa; add addr:interpolation info

[Don-vip]

Milestone renamed