[gdal-dev] MVT fields/geometry alignment

[Michael Sumner]

Oh I see, thanks for explaining!

Cheers, Mike


On Wed., 7 Aug. 2019, 18:19 Even Rouault, <even.rouault at spatialys.com>
wrote:
>
> Michael,
>
> >
> > In this MVT source poLayer->GetFeatureCount() reports that there are 85
> > features, as does ogrinfo summary:
> >
> > But, iterating features finds 81 valid features, as reported by
>
> The difference is due to clipping geometries to the tile extent.
> GetFeatureCount() do a super fast scan, counting all features, whereas by
> default GetNextFeature() also checks if the geometry intersects the tile
> extent. And in that case there are a few features that are actually
outside of
> the tile extent. If you add -oo CLIP=NO then GetNextFeature() will report
85
> features as well.
>
> As a workaround for GetFeatureCount() to return 81 as well, you can
specify a
> dummy attribute filter that will force it to go to the slow scan mode:
>
> ogrinfo 14/4823/6160.mvt landuse -al -so -where 1=1
>
> Even
>
> --
> Spatialys - Geospatial professional services
> http://www.spatialys.com
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/gdal-dev/attachments/20190807/6cd31077/attachment.html>