`geof:buffer` producing unexplainable results

[ldesousa]

The geof:buffer function is producing unexplainable results. A full example is at the end of this message. Whatever the function is returning, it is certainly not a 10 metre buffer around the point. I experimented different distances and the results are always different from the expected. Only with a distance of 0 is it producing a sensible result (empty polygon).

I am using version 07.20.3233. This issue has been discussed in the users list.

PREFIX geo: <http://www.opengis.net/ont/geosparql#>
PREFIX geof: <http://www.opengis.net/def/function/geosparql/>

SELECT geof:buffer("POINT(50 50)"^^geo:wktLiteral, 10, <http://www.opengis.net/def/uom/OGC/1.0/metre>)
WHERE {};
Type the rest of statement, end with a semicolon (;)> Type the rest of statement, end with a semicolon (;)> Type the rest of statement, end with a semicolon (;)> Type the rest of statement, end with a semicolon (;)> Type the rest of statement, end with a semicolon (;)> callret-0
LONG VARCHAR
_______________________________________________________________________________

POLYGON((60 50,59.807852804032 48.049096779839,59.238795325113 46.173165676349,58.314696123025 44.444297669804,57.071067811865 42.928932188135,55.555702330196 41.685303876975,53.826834323651 40.761204674887,51.950903220161 40.192147195968,50 40,48.049096779839 40.192147195968,46.173165676349 40.761204674887,44.444297669804 41.685303876975,42.928932188135 42.928932188135,41.685303876975 44.444297669804,40.761204674887 46.173165676349,40.192147195968 48.049096779839,40 50,40.192147195968 51.950903220161,40.761204674887 53.826834323651,41.685303876975 55.555702330196,42.928932188134 57.071067811865,44.444297669804 58.314696123025,46.173165676349 59.238795325113,48.049096779839 59.807852804032,50 60,51.950903220161 59.807852804032,53.826834323651 59.238795325113,55.555702330196 58.314696123025,57.071067811865 57.071067811865,58.314696123025 55.555702330196,59.238795325113 53.826834323651,59.807852804032 51.950903220161,60 50))

1 Rows. -- 60 msec.

[HughWilliams]

As asked in the source forge email thread, do you have a minimal test case for recreating the issue ?

[ldesousa]

Hi again Hugh,

yes, the test case is in the issue report at GitHub.

Regards.

-- Luís

------- Original Message ------- On Thursday, May 12th, 2022 at 9:50 PM, HughWilliams @.***> wrote:

As asked in the source forge email thread, do you have a minimal test case for recreating the issue ?

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

[HughWilliams]

OK, see what you mean, the reference to the source forge issue confused me as it talk about another related issue with "geof:sfIntersects".

Anyway I have been able to recreate the issue you logged with "geo:buffer" and have reported it to development to look into ...