[OSM-dev] Broken Nodes

Tom Hughes tom at compton.nu
Thu Sep 20 16:04:34 BST 2007


In message <1190298999.7078.21.camel at localhost>
        Jan Lübbe <jluebbe at lasnet.de> wrote:

> While parsing planet.osm i found some Nodes which seem to have broken
> coordinates:
>
> ignoring node with invalid location (id=27475484 lat=127136 lon=485452)
> ignoring node with invalid location (id=27475485 lat=127216 lon=485548)
> ignoring node with invalid location (id=27475486 lat=127200 lon=485580)
> ignoring node with invalid location (id=27475487 lat=127168 lon=485532)
> ignoring node with invalid location (id=27475488 lat=127096 lon=485548)
> ignoring node with invalid location (id=27475489 lat=127072 lon=485500)
> ignoring node with invalid location (id=27475490 lat=127120 lon=485468)
> ignoring node with invalid location (id=27475492 lat=126104 lon=482980)
> ignoring node with invalid location (id=27475493 lat=126216 lon=483036)
> ignoring node with invalid location (id=27475494 lat=126288 lon=483156)
> ignoring node with invalid location (id=27475508 lat=126304 lon=483204)
> ignoring node with invalid location (id=27475510 lat=126248 lon=483204)
> ignoring node with invalid location (id=27475511 lat=126224 lon=483132)
> ignoring node with invalid location (id=27475513 lat=126184 lon=483052)
> ignoring node with invalid location (id=27475514 lat=126120 lon=482996)
>
> I have not been able to load them in josm and don't know how to fix them
> without it.

I noticed those this morning as well. They all date to a period of
about two minutes on 19th April this year.

I know who they came from, and as a result I think I can guess how
they came about. The person in question has not made their edits
public however so I can't say much more.

The current API will reject any node that is outside the valid range
of latitude and longitude so there shouldn't be any repeat and I will
probably delete those nodes as nobody will be able to find them in any
normal way.

Tom

-- 
Tom Hughes (tom at compton.nu)
http://www.compton.nu/




More information about the dev mailing list