> >> For this proposal to make sense, all deployed autopkgtests must > >> actually test the package involved to some extent. > > > > I fully ACK and understand the issue and background but just to be more > > effective priority-wise are you aware of how many packages, if any, > > are violating this right now? > > codesearch [1] suggests it may be ~170 packages. All packages by the GIS > team seem to be involved. Rather unconstructively, they refuse to fix it > until lintian has the warning [2].
There is also one package that uses "true" not "/bin/true" so I guess that should also be included. Perhaps: Test-Command:\s*(/bin/)?true I've not found any other obvious noops within Test-Command (exit, echo, :, …). It's not possible for lintian to test for all the 'creative' things people can do but since 'true' is the most obvious noop it's probably enough for that. (The policy around the tests obviously needs to broadly forbid noop tests in some way beyond what lintian can actually test for; someone from RT/ci/ Policy(?) needs to document what is acceptable) cheers Stuart -- Stuart Prescott http://www.nanonanonano.net/ stu...@nanonanonano.net Debian Developer http://www.debian.org/ stu...@debian.org GPG fingerprint 90E2 D2C1 AD14 6A1B 7EBB 891D BBC1 7EBB 1396 F2F7