Regression failures in 1.5 when building against 9.2 for windows

[robe]

I'm not sure if this is a problem with the regression tests or something amiss with 9.2.

 PostgreSQL 9.2.10, compiled by Visual C++ build 1600, 64-bit
 Postgis 1.5.8 - 2015-02-28 02:02:11
   GEOS: 3.4.2-CAPI-1.8.2 r3924
   PROJ: Rel. 4.8.0, 6 March 2012

Running tests

 loader/Point.............. ok
 loader/PointM.............. ok
 loader/PointZ.............. ok
 loader/MultiPoint.............. ok
 loader/MultiPointM.............. ok
 loader/MultiPointZ.............. ok
 loader/Arc.............. ok
 loader/ArcM.............. ok
 loader/ArcZ.......... ok
 loader/Polygon.............. ok
 loader/PolygonM.............. ok
 loader/PolygonZ.............. ok
 regress. ok
 regress_index. ok
 regress_management. ok
 regress_index_nulls. ok
 lwgeom_regress. ok
 regress_lrs. ok
 removepoint. ok
 setpoint. ok
 simplify. ok
 snaptogrid. ok
 summary. ok
 affine. ok
 wkt. ok
 measures. ok
 long_xact. ok
 ctors. ok
 sql-mm-serialize. ok
 sql-mm-circularstring. ok
 sql-mm-compoundcurve. ok
 sql-mm-curvepoly. ok
 sql-mm-general. ok
 sql-mm-multicurve. ok
 sql-mm-multisurface. ok
 geojson. ok
 gml. ok
 svg. ok
 kml. ok
 in_gml. failed (diff expected obtained: /tmp/pgis_reg_3852/test_40_diff)
 in_kml. failed (diff expected obtained: /tmp/pgis_reg_3852/test_41_diff)
 polygonize. ok
 regress_ogc. failed (diff expected obtained: /tmp/pgis_reg_3852/test_43_diff)
 regress_ogc_cover. ok
 regress_ogc_prep. ok
 regress_bdpoly. ok
 regress_proj. ok
 dumppoints. ok
 wmsservers. ok
 tickets. ok
 regress_buffer_params. ok
 hausdorff. ok

Run tests: 52
Failed: 3
make[1]: Leaving directory `/projects/postgis/branches/postgis-1.5.8/regress'

Diffs attached.

[robe]

failures

[robe]

I should note it also fails in same sports testing against mingw-w64 compiled PostgreSQL

[robe]

Okay I looked at the regres_ogc error and it fails this:

SELECT ST_astext(ST_pointonsurface('POLYGON((0 0, 0 10, 10 10, 10 0, 0 0),(2 2, 2 4, 4 4, 4 2, 2 2))'));

According to the regress expected, I should be getting a POINT(5 5).

I get a POINT(5 7), but I get a POINT(5 7) on my PostgreSQL 9.4 and 9.3 2.2 instances as well. This test looks like it was later changed to be:

WITH inp AS ( SELECT
 'POLYGON((0 0, 0 10, 10 10, 10 0, 0 0),(2 2, 2 4, 4 4, 4 2, 2 2))'
::geometry as g )
SELECT 'pointonsurface', ST_Contains(g, ST_pointonsurface(g)) from inp;

So maybe it's a harmless failure.

Haven't check the in_kml, in_gml errors yet.

[pramsey]

The ogc failure is harmless. The GML/KML ones seem more extensive though. Not so sure I care.

[pramsey]

Nope, dont