OGC Test 49 - 50

Topics: Developer Forum
Aug 16, 2006 at 4:22 PM
Hy ricardo.
Looking at the expected results for tests 49 and 50, it seems that something is wrong...
Look at the image here ( http://digilander.libero.it/DGuidi/OGC4950.jpg ):
there are the two polygons, and a union of this two polygons necessary generates a multipolygon!
But OGC expected result is a polygon... :(
I think that the results is not due to a NTS bug...
Aug 16, 2006 at 9:09 PM
Sorry, you won't believe what a silly issue was this... OGC test T49 query condition is:
WHERE lakes.name = 'Blue Lake' AND named_places.name = Ashton';
But the narrative enunciation of the test says: "For this test we will determine the union of Blue Lake and Goose Island".

The same with T50. I've already found that the OGC conformance tests are, deliberately or not, plagued of little mistakes but nothing so annoying like this until now. Moral of the day: "Don't believe in conformance tests" :-P

Sorry again... and thanks!
Aug 17, 2006 at 8:56 AM
In any case T51 error is due to a NTS bug now fixed, so thanks to you ;)