osmose-backend
osmose-backend copied to clipboard
type=disused:route should not be treated as high priority alert
While keeping historic data about suspended routes forever is invalid, this check reports all of them.
And when route is suspended for say 6 months it makes sense to tag it as say type=disused:route
rather than later recreate it (unless report should be complaining that proper tagging is disused:type=route
? Both are in use https://taginfo.openstreetmap.org//search?q=disused%3Atype#keys https://taginfo.openstreetmap.org/search?q=type%3Ddisused%3Aroute#tags (
When report an issue on existing analyser, in addition on the name of the issue, report:
item 3040 class 3040
https://osmose.openstreetmap.fr/en/issue/bf46ce0a-be63-76ff-43c8-55a5eeb1b03d
https://osmose.openstreetmap.fr/en/map/#zoom=18&lat=50.079719&lon=20.064033&item=xxxx&level=1&issue_uuid=ef2482aa-8ef0-eb5d-5273-0563ea0e7d78
It comes from here
https://github.com/osm-fr/osmose-backend/blob/master/plugins/TagFix_BadValue.py#L47
A generic rule allowing some values by regex.
Note there is already some exceptions, but we cannot list all exceptions like this https://github.com/osm-fr/osmose-backend/blob/master/plugins/TagFix_BadValue.py#L75
Not sure what to do with that.
I thought the life cycle prefixes were only for the keys, not for the values? (If not, let's hope there won't ever be a shop called disused:computers or we wouldn't be able to name it, for instance)
It seems to be pretty rare also, mostly restricted to type=disused:route(_master)
: taginfo for *=disused*
. Maybe this is due to people trying to avoid this validation rule in JOSM? (I mean, if you in JOSM want to mark a bus route as disused, and you use disused:route
, you'll get the warning route relation without route tag
. To avoid that you could say disused:type
, but then you'll get the warning relation without type
)
I thought the life cycle prefixes were only for the keys, not for the values?
On the other hand relations require type
....
No idea which is better - but if we have clear preference then it is qualifying for an obvious bot edit (not manual drudgery) and if there is no obvious preference then it is not a high priority issue.
If not, let's hope there won't ever be a shop called disused:computers or we wouldn't be able to name it, for instance
There are some like that - but I am fixing them/gathering more popular for bot edit. See https://codeberg.org/matkoniecz/OpenStreetMap_cleanup_scripts/src/branch/master/script_assisted_cleanup/shops_migrate_complex_cases.py and https://codeberg.org/matkoniecz/OpenStreetMap_cleanup_scripts/src/branch/master/recurrent_bot_edits/NEXT_shops_migrate_complex_cases.py (https://codeberg.org/matkoniecz/OpenStreetMap_cleanup_scripts/src/branch/master/recurrent_bot_edits/shops_migrate_complex_cases.py does not exist yet, approved bot edit will be there once I have enough time for that - BTW, if someone wants to contribute/use that code feel free to do so, just remember about following OSM bot edit rules and how manual cleanup should be done)