{"approve": "license and version number unchanged: 1.8.4"}

Please review sources

Check script succeeded

Please review build success

Builds for repo openSUSE_Factory

Hi, would you mind to explain why you have changed keyring? It is quite unusual to do it when tarball is not changed.

Oh I see - you simply use the keyring from upstream. I'd not say this is correct as especially Apache tends to have KEY files with all past release managers in it, so gpg-offline is happy if one of those keys matches. Would you mind to create a subversion.keyring which just one ID (those which has been used to sign the tarball)?

BTW: if disagree, please discuss it on opensuse-packaging as we definitely have to make it clear ...

Hi, would you mind to explain why you have changed keyring? It is quite unusual to do it when tarball is not changed.

Oh I see - you simply use the keyring from upstream. I'd not say this is correct as especially Apache tends to have KEY files with all past release managers in it, so gpg-offline is happy if one of those keys matches. Would you mind to create a subversion.keyring which just one ID (those which has been used to sign the tarball)?

BTW: if disagree, please discuss it on opensuse-packaging as we definitely have to make it clear ...

- print error logs on on regression test failures
- always build with and statically link to in-tree SQLite to
prevent performance problems and test failures with SQLite 3.8
- fix regression tests for ppc/ppc64 architectures, found in
openSUSE package build and fixed with upstream developers, adding
subversion-1.8.x-fix-ppc-tests.patch
- improve sqlite and python-ctypes conditionals, the latter not
building for any current targets
- require python-sqlite when running regression tests for all
targets, no longer pulled in implicitly
- update keyring, use Subversion Project Management Committee
keyring rather than all committers