[postgis-users] Failing regression tests

Imre Samu pella.samu at gmail.com
Sun Dec 3 05:16:38 PST 2023


Paul Förster via postgis-users <postgis-users at lists.osgeo.org> :

> Especially protobuf-c is a pain. I have to use SLES 15.5 (and for testing
> openSUSE Leap 15.5). With the SUSE flavor distribution(s), the package
> versions are a mess and I have to stay current. Hence the downloaded source
> packages above. Also, this happens no matter what PostgreSQL versions I use
> (also compiled from source).
>

Hello Paul,
I would like to clarify a few details:
- For openSUSE, which architecture are you using? Is it amd64, arm64,
powerpc, or ibmz/linuxone?
- Could you please provide the complete installation scripts (are they in
bash?) and the full logs?
- When you execute your installation script, can you reproduce the process
in Docker? (For instance, using "docker pull opensuse/leap:15.5")

 regress/core/in_gml .. failed (diff expected obtained:
> /home/postgres/postgis-3.4.1/postgis_reg/test_24_diff)
>  regress/core/in_kml .. failed (diff expected obtained:
> /home/postgres/postgis-3.4.1/postgis_reg/test_25_diff)
>

Based on the Debian docke-postgis CI found at
https://github.com/postgis/docker-postgis/blob/master/16-master/Dockerfile ,
we need to look for specific lines in the GDAL build logs.
Please search for lines like:
- "Scanning dependencies of target ogr_KML"
- "Scanning dependencies of target ogr_GML"
- "Built target ogr_KML"
- "Built target ogr_GML"

Could you please check if there is a " -DGDAL_USE_LIBKML=ON" setting in the
GDAL cmake parts?
Please examine the output of the command "ogr2ogr --formats". Are the
following lines present in the output?
  GML -vector- (rw+v): Geography Markup Language (GML)
  LIBKML -vector- (rw+v): Keyhole Markup Language (LIBKML)
  KML -vector- (rw+v): Keyhole Markup Language (KML)

Best regards,
   Imre
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/postgis-users/attachments/20231203/d82de0d7/attachment.htm>


More information about the postgis-users mailing list