On Tue, Jul 15, 2008 at 10:07 AM, Dirk Stöcker <[EMAIL PROTECTED]> wrote:
> On Tue, 15 Jul 2008, Karl Newman wrote: > > >> No. It is > >> validator:ignore_foo=nodeid > >> Or > >> validator:ignore_bar=nodeid1;wayid2;noideid3 > >> > >> And that's full of problems when moving it to a database. You would need > >> to use relations for it. > > > > No, you're missing the point. It's just another tag. The > > validator_ignore=no_name goes directly on the element that would > otherwise > > be flagged by the validator. It's metadata, which basically means, "yes, > I > > know there is something weird about this entity, but a human says it's > > okay." > > Assume you have a railway station and a village both called "Hubba". You > tag the with validator:ignore_name=true. > > Now a new user adds the same village again. You wont be able to see this > change, as you disabled the whole checking. Thus the only solutions is > for village: validator:ignore_name=id_of_railway_station (or other way > round). > > You see the problem with this? > > Ciao > Well, I don't know how the validator works (I haven't used it much), so I don't know why it's a problem to have both a railway station and village with the same name. But in any case, if you're clever with the implementation, you could have it flag the second instance as a duplicate, even though the first instance says to ignore duplicated names. Basically that means that all items that would be flagged as a duplicate by the validator need to be tagged in order to suppress the warning. Karl _______________________________________________ josm-dev mailing list [email protected] http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/josm-dev
