warning street name contains ss (edit in switzerland)

[datendelphin]

Josm shows a wrong warning "street name contains ss" when uploading road names in Switzerland. This check should probably be limited geographically.

[stoecker]

All checks can be turned off individually. We're aware of that issue for years now and think that user can decide themselves. As far as I know this warning has very low priority already.

[datendelphin]

It is new to me. Version history seems to assert me: ​https://github.com/openstreetmap/josm/commit/117671276fe8b8dce92e0ec7f2f344060943f477
I just discussed it on #osm-ch, no-one remembers to have seen this error before. Seems to be new to be on by default.

And I didn't find out how to disable this check.

Please consider to reopen. It changed recently. This might lead people to believe that OSM doesn't use swiss orthography in Switzerland. Or to completely ignore the validator.

[sp8962@…]

It is -always- wrong (the warning) in Switzerland, it is not a matter of judgement. The behaviour furter seems to be new, does not seem to occur with 6502.

[simon04]

Replying to stoecker:

All checks can be turned off individually.

How? Do you mean that errors can be ignored? That doesn't work for those MapCSS validations since an indifferent error code is used.

Replying to datendelphin:

It is new to me. Version history seems to assert me: ​https://github.com/openstreetmap/josm/commit/117671276fe8b8dce92e0ec7f2f344060943f477

That check has been around for a long time. It has only been migrated to a new validation machinery recently (#9414).

A practical way to ignore this error in the CH context could be to evaluate the Java user.country system property, i.e., the country specified in the system's locale.

[anonymous]

Replying to simon04:

That check has been around for a long time. It has only been migrated to a new validation machinery recently (#9414).

Given that you would have noticed immediately (via the outrage) when the test was added, it is safe to assume that it simply didn't work before.

[simon04]

It did work before, but the generated errors didn't got displayed (unless one enabled them via the validator preferences). The same errors again are generated since r6637. So let's keep it this way. :-)

Thanks for reporting!

[stoecker]

We need to reenable the ignore feature for all the new tests and fix it for the old ones. E.g. we could use the English original message instead of an hardcoded ID. But that means some rework, as currently only translated and formatted text is passed.

[datendelphin]

Thank you. With the latest version there is no warning reported anymore for Strasse with ss