[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: GRASS 8.0.0



On 12/30/21 18:07, Sebastiaan Couwenberg wrote:
The first release candidate for GRASS 8.0.0 has been released.

This new major version is an opportunity for major changes in the packaging as well.

Should we drop the grass<XY> naming and just use grass since we don't support installing multiple versions simultaneously?

This would affect executables like usr/bin/grass80 & usr/bin/x-grass80 and subdirectories like usr/lib/grass80 & usr/share/grass80.

QGIS likely required changes to support GRASS 8 any way, so having to adapt its packaging too won't be much of an issue.

The Fedora packaging also uses the grass<XY> naming for their 7.8 packages, so we're not the only one doing this. We may want to keep it for 8.x.

Thoughts?

The silence is deafening.

Some tests with libgdal-grass built with GRASS 8.0.1 revealed some issues, but those turned out to be present in 7.x too:

 https://lists.osgeo.org/pipermail/grass-dev/2022-February/095610.html

It also revealed that libgdal-grass is broken in bullseye:

 https://bugs.debian.org/1006446

QGIS explicitly doesn't support GRASS 8 yet, and patching it to build with it is too much work:

 https://github.com/qgis/QGIS/blob/final-3_22_4/CMakeLists.txt#L122

There seems to be some support for GRASS 8 in the 3.22 branch, so we might be able to upgrade to GRASS 8 after the 3.22.5 release.

Since I don't use GRASS, I'm in no hurry to update. At the time of writing there has been response to my question (on the grass-dev list) whether any work has been done to test gdal-grass and qgis with GRASS 8, that would help build confidence whether we can update or not.

Kind Regards,

Bas

--
 GPG Key ID: 4096R/6750F10AE88D4AF1
Fingerprint: 8182 DE41 7056 408D 6146  50D1 6750 F10A E88D 4AF1


Reply to: