[Tilesathome] rendering bug!!!??

[Dirk-Lüder Kreie]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Robert Hart schrieb:
> On 28/09/2007, Thomas Krauss <mail at der-eichi.de> wrote:
>> Hello, me again,
>>
>> I'm mapping at the moment some areas in germany,
>> you can see the result here:
>> http://informationfreeway.org/?lat=48.17112823915294&lon=9.191224345557755&zoom=12&layers=00B0F000
>>
>> on the bottom, there are 2 straight lines. They never exist in the osm data,
>> never seen them with josm on the data layer.
>>
>> This must be a confused rendering bug,
>> does someone know this kind of problem?
>>
>> Is there a way to fix this?
> 
> I'd hazard a guess the lines2curves script choked on something. You

I'd think so, too.

> had a number of quite long ways that were unordered. Maybe frollo
> gives up and then lines2curves chokes? I fixed all the problems with
> unordered ways, segments in the same way multiple times (there should
> be a validator check for that), and so forth, and have requested a
> rerender. Lets see what happens....

Yes frollo chokes on too long ways, that's why it stops reordering them
after a certain length.

Please don't (not only for the renderer, but OSM in general) create
a) too long ways (use lots of resources on api bbox requests)
b) unordered ways (all kinds of crazy stuff happens, with routing or
rendering
c) non-contiguous ways (branched or segmented)


- --

Dirk-Lüder "Deelkar" Kreie
Bremen - 53.0952°N 8.8652°E

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFG/n8iFUbODdpRVDwRAvFbAKCjsmqx3NVJJ8NpsjFhtIJ9zD/zxACgkTaH
2byciQIy0NY2G3ep0Fv+av0=
=35VR
-----END PGP SIGNATURE-----