On 12 February 2016 at 09:51, Stuart Reynolds <[email protected]> wrote: > As a side issue, does your tool currently ignore disused:amenity=* tags? If > not, can it please. Thanks.
The tool only fetches amenity=school and amenity=college objects, so disused:amenity=* (and other similar variants) will always be ignored. On 12 February 2016 at 10:07, Stuart Reynolds <[email protected]> wrote: > …oh, and if the name matching could be tolerant to “&” and “and", that would > be brilliant, too. It’s not a biggie, because I’m getting matched against the > Edubase reference, but the names are otherwise the same and I’m using & for > brevity. The name matching (which currently just checks the two strings for an exact match) isn't actually used for much. The only place it's used is when looking for a geographically close match between an OSM object and an official location, a pair with an exact name match will be preferred over other closer matches. Matches based on ref:* tags always take priority, and if there's no exact name match, then the nearest unmatched object with 1km is chosen. That said, it's on my To Do list to look at doing a fuzzy name comparison -- probably by deleting common words / phrases (School, Academy, CofE, Voluntary Controlled, and, &, etc) from both strings and then seeing if what's left is equal. This would probably be used to flag up possible errors in matched pairs (either in the OSM name or the matching), rather than as a way to get better matching in the first place. The latter can always be solved by adding the ref:* values to OSM. Best wishes, Robert. -- Robert Whittaker _______________________________________________ Talk-GB mailing list [email protected] https://lists.openstreetmap.org/listinfo/talk-gb
