winnie regress failures on raster tests

[robe]

I don't think this has anything to do with PostGIS raster. These tests started failing on winnie around the time of changes in interrupt (November 11th). Unfortunately I must have set to not store history that far. I'm suspecting some sort of warnings level got changed:

Note how all the errors are because "DO" is not being stripped.

Running tests

 check_gdal .. ok 
 load_outdb ... ok 
 check_raster_columns .. ok 
 check_raster_overviews .. ok 
 rt_io .. ok 
 rt_bytea .. ok 
 box3d .. ok 
 rt_addband .. ok 
 rt_band .. ok 
 rt_tile .. ok 
 rt_dimensions .. ok 
 rt_scale .. ok 
 rt_pixelsize .. ok 
 rt_upperleft .. ok 
 rt_rotation .. ok 
 rt_georeference .. ok 
 rt_set_properties .. ok 
 rt_isempty .. ok 
 rt_hasnoband .. ok 
 rt_metadata .. ok 
 rt_rastertoworldcoord .. ok 
 rt_worldtorastercoord .. ok 
 rt_convexhull .. ok 
 rt_envelope .. ok 
 rt_band_properties .. ok 
 rt_set_band_properties .. ok 
 rt_pixelaspolygons .. ok 
 rt_pixelaspoints .. ok 
 rt_pixelascentroids .. ok 
 rt_setvalues_array .. ok 
 rt_summarystats .. ok 
 rt_count .. ok 
 rt_histogram .. ok 
 rt_quantile .. ok 
 rt_valuecount .. ok 
 rt_valuepercent .. ok 
 rt_bandmetadata .. ok 
 rt_pixelvalue .. ok 
 rt_neighborhood .. ok 
 rt_nearestvalue .. ok 
 rt_pixelofvalue .. ok 
 rt_polygon .. ok 
 rt_utility .. ok 
 rt_fromgdalraster .. ok 
 rt_asgdalraster .. ok 
 rt_astiff .. ok 
 rt_asjpeg .. ok 
 rt_aspng .. ok 
 rt_reclass .. ok 
 rt_gdalwarp .. ok 
 rt_asraster .. ok 
 rt_dumpvalues .. ok 
 rt_createoverview .. ok 
 rt_mapalgebraexpr .. ok 
 rt_mapalgebrafct .. ok 
 rt_mapalgebraexpr_2raster .. ok 
 rt_mapalgebrafct_2raster .. ok 
 rt_mapalgebrafctngb .. ok 
 rt_mapalgebrafctngb_userfunc .. ok 
 rt_intersection .. ok 
 rt_clip .. ok 
 rt_mapalgebra .. failed (diff expected obtained: /projects/postgis/tmp/2.2.0dev_pg9.4_geos3.5.0dev_gdal1.11.1w64/test_62_diff)
-----------------------------------------------------------------------------
--- rt_mapalgebra_expected	2014-11-11 06:03:25 -0500
+++ /projects/postgis/tmp/2.2.0dev_pg9.4_geos3.5.0dev_gdal1.11.1w64/test_62_out	2014-11-13 15:26:52 -0500
@@ -125,6 +125,7 @@
 NOTICE:  pos = [0:1][1:2]={{2,2},{4,4}}
 NOTICE:  userargs = {1000}
 NOTICE:  record = (17,"(2,-4,2,2,1,-1,0,0,0,1)","(32BUI,0,f,)",1000)
+DO
 NOTICE:  value = {{{1}},{{2}}}
 NOTICE:  pos = [0:2][1:2]={{1,1},{2,2},{1,1}}
 NOTICE:  userargs = <NULL>
-----------------------------------------------------------------------------
 rt_mapalgebra_expr .. ok 
 rt_mapalgebra_mask .. ok 
 rt_union .. ok 
 rt_invdistweight4ma .. ok 
 rt_4ma .. ok 
 rt_setvalues_geomval .. ok 
 rt_elevation_functions .. failed (diff expected obtained: /projects/postgis/tmp/2.2.0dev_pg9.4_geos3.5.0dev_gdal1.11.1w64/test_69_diff)
-----------------------------------------------------------------------------
--- rt_elevation_functions_expected	2014-11-11 06:03:25 -0500
+++ /projects/postgis/tmp/2.2.0dev_pg9.4_geos3.5.0dev_gdal1.11.1w64/test_69_out	2014-11-13 15:26:54 -0500
@@ -1,3 +1,4 @@
+DO
 0|aspect|1|1|315.000000
 0|aspect|2|1|341.565063
 0|aspect|3|1|0.000000
-----------------------------------------------------------------------------
 rt_colormap .. ok 
 rt_gist_relationships .. ok 
 rt_intersects .. ok 
 rt_samealignment .. ok 
 rt_geos_relationships .. ok 
 rt_iscoveragetile .. ok 
 bug_test_car5 .. ok 
 permitted_gdal_drivers .. failed (diff expected obtained: /projects/postgis/tmp/2.2.0dev_pg9.4_geos3.5.0dev_gdal1.11.1w64/test_77_diff)
-----------------------------------------------------------------------------
--- permitted_gdal_drivers_expected	2014-11-11 06:03:25 -0500
+++ /projects/postgis/tmp/2.2.0dev_pg9.4_geos3.5.0dev_gdal1.11.1w64/test_77_out	2014-11-13 15:26:57 -0500
@@ -1,3 +1,4 @@
+DO
 4326
 ERROR:  RASTER_fromGDALRaster: Could not open bytea with GDAL. Check that the bytea is of a GDAL supported format
 ERROR:  rt_band_load_offline_data: Cannot open offline raster: <VRTDataset rasterXSize="1000" rasterYSize="1"><VRTRasterBand band="1" subClass="VRTRawRasterBand"><SourceFilename>/etc/passwd</SourceFilename></VRTRasterBand></VRTDataset>
-----------------------------------------------------------------------------
 tickets .. ok 
 loader/Basic .... ok 
 loader/Projected ..... ok 
 loader/BasicCopy .... ok 
 loader/BasicFilename .... ok 
 loader/BasicOutDB .... ok 
 loader/Tiled10x10 .... ok 
 loader/Tiled10x10Copy .... ok 
 loader/Tiled8x8 .... ok 
 clean .. ok 
 uninstall .  /projects/postgis/branches/2.2/regress/00-regress-install/share/contrib/postgis/uninstall_rtpostgis.sql
  /projects/postgis/branches/2.2/regress/00-regress-install/share/contrib/postgis/uninstall_postgis.sql
. ok (4336
)

Run tests: 88
Failed: 3

I'm getting this in the diff files for example:

--- permitted_gdal_drivers_expected	2014-11-11 06:03:25 -0500
+++ /projects/postgis/tmp/2.2.0dev_pg9.4_geos3.5.0dev_gdal1.11.1w64/test_77_out	2014-11-13 15:26:57 -0500
@@ -1,3 +1,4 @@
+DO
 4326
 ERROR:  RASTER_fromGDALRaster: Could not open bytea with GDAL. Check that the bytea is of a GDAL supported format
 ERROR:  rt_band_load_offline_data: Cannot open offline raster: <VRTDataset rasterXSize="1000" rasterYSize="1"><VRTRasterBand band="1" subClass="VRTRawRasterBand"><SourceFilename>/etc/passwd</SourceFilename></VRTRasterBand></VRTDataset>

[robe]

aha it must have been change in r13123

 Revision 13123 by strk:
Disreguard DO responses during testing
The file was modified	/trunk/regress/run_test.pl

[robe]

I was wrong it was this one that broke it: r13123

 Revision 13125 by strk:
Do not expect DO responses in raster regress

Fixes testsuite broke by r13123
The file was modified	/trunk/raster/test/regress/permitted_gdal_drivers_expected
The file was modified	/trunk/raster/test/regress/rt_elevation_functions_expected
The file was modified	/trunk/raster/test/regress/rt_mapalgebra_expected

so seems the run_test.pl change perhaps winnie is ignoring. I'll try to do a fresh svn pull

[robe]

nope she is already doing an svn clean. I'm going to change your regex script after I have done some testing.

I'm guessing your

  @lines = grep(!/^(DO)$/, @lines);

Doesn't work on windows presumably because of the line break difference or something so as far as windows is concerned (or at least running under mingw), DO is not the only thing on the line.

[robe]

committed a fix at r13128 fixed issue on my windows 7 desktop so assume it will fix Winnie's issue as well. Will close once confirmed winnie is happy again and doesn't cause other bots to complain

[robe]

and r13129 (got rid of the redundant DO replace)

[strk]

Sounds good as long as nothing broke (any output starting with "DO" will be hidden).

[robe]

hmm this was really odd. On debbie I got a weird error after this change:

 /var/lib/jenkins/workspace/postgis/branches/2.2/regress/00-regress-install/share/contrib/postgis/postgis.sql
 failed (Error encountered loading /var/lib/jenkins/workspace/postgis/branches/2.2/regress/00-regress-install/share/contrib/postgis/postgis.sql: /var/lib/jenkins/workspace/postgis/tmp/2_2_pg9.2w64/regress_log)
-----------------------------------------------------------------------------
createlang: language "plpgsql" is already installed in database "postgis_reg"
SET
BEGIN
DO
psql:/var/lib/jenkins/workspace/postgis/branches/2.2/regress/00-regress-install/share/contrib/postgis/postgis.sql:78: ERROR:  duplicate key value violates unique constraint "pg_proc_proname_args_nsp_index"
DETAIL:  Key (proname, proargtypes, pronamespace)=(_postgis_deprecate, 25 25 25, 2200) already exists.

but forcing a rerun, the error doesn't show the second time and seems fine. Might be because I did two commits very close to one another and there is some chaining issue with how I have things setup.