Working with Invalid Data and the SQL Server 2008 Geography Data Type, Part 2

re: Working with Invalid Data and the SQL Server 2008 Geography Data Type, Part 2

Nagarjuna Dilip — Tue, 29 Jan 2013 15:46:50 GMT

Thanks for your work! It worked for me. :)

re: Working with Invalid Data and the SQL Server 2008 Geography Data Type, Part 2

Franck — Tue, 08 Feb 2011 21:57:04 GMT

Great post thanks !!!

Really helpful espacially the Reduce() trick on invalid-geography-polygons-that-comes-from-valid-geometry...

Franck

re: Working with Invalid Data and the SQL Server 2008 Geography Data Type, Part 2

Mike — Mon, 31 May 2010 13:31:49 GMT

I am looking for a way to convert my WGS84 coord that a GPS receiver gives me to a NAD83 coord that another program is expecting. Is there a Stored Porcedure that can do this for me?

Thanks,

re: Working with Invalid Data and the SQL Server 2008 Geography Data Type, Part 2

Wimpy1 — Tue, 24 Nov 2009 17:20:54 GMT

Is there a way to make valid geographies strictly using the FME workbench? What FME components/transformers would handle these steps?

re: Working with Invalid Data and the SQL Server 2008 Geography Data Type, Part 2

vIndEx — Mon, 22 Jun 2009 09:43:20 GMT

Hi Ed,

I think that using .STUnion(@g.STPointN(1)) is the best way to validate these geometries, instead of using Reduce:

1) It will mantain the shape of the geometries

2) Don't have to guess with the value passed to reduce

3) If needed, it changes the rings orientation so you can buid a geography

re: Working with Invalid Data and the SQL Server 2008 Geography Data Type, Part 2

Hugh Saalmans — Mon, 22 Jun 2009 07:57:22 GMT

Hi Ed,

Thanks for the update - the original method (in part 1) didn't work with my data, so good to finally have that sorted...!

What would be great would be if there was a tool to remove the linestrings and points created by using either the Reduce function or a validation function, rather than having to clean them out manually or using a custom application.

re: Working with Invalid Data and the SQL Server 2008 Geography Data Type, Part 2

Spatial Ed — Mon, 08 Jun 2009 06:56:33 GMT

You do raise a good point. However, in this particular case, I would argue that the vertexes removed on ID=3499 were essentially duplicative and remaining problem of spikes was most likely pure data errors.

What about a new MapValidGeography<...> function with Reduce and Densify arguments?

re: Working with Invalid Data and the SQL Server 2008 Geography Data Type, Part 2

Morten Nielsen — Mon, 08 Jun 2009 06:37:52 GMT

When you convert form Geometry to Geography, shouldn't you really be densifying along the line segments?

It might be that the shape file you are importing "claims" to be espg:4269, but my bet is that it is really a projected coordinate system where the line segments are straight in a flat coordinate system, but not in an spherical one. The issues you hit also seem to confirm that.

re: Working with Invalid Data and the SQL Server 2008 Geography Data Type, Part 2

me — Sat, 06 Jun 2009 21:30:03 GMT

thanks for your posts. I am a web dev guy who only knows the basics of sql server and this spatial querying is very interesting. Im trying to go from shape files where the lat/lons are backwards to showing the polygons on google maps so one of these months I think I will put all this together. Anyway thanks for your contributions

national media exposure » Wikipedia:Requests for arbitration/Macedonia 2/Proposed decision

Sat, 06 Jun 2009 14:18:26 GMT

PingBack from http://nationalmediaexposure.blog-giant.com/2009/06/04/wikipediarequests-for-arbitrationmacedonia-2proposed-decision/