Hey Christoph et al.,
> > The more I think about it, the more I think this is a misleading
> > and non-intuitive misfeature of file.
[…]
> > Any thoughts before I essentially re-assign this?
>
> Rather somehow deal with it?
Whilst I understand what you mean here, the problem is a little
more subtle. Lintian is deliberately making the distinction between
shared objects and PIE executables at the moment in order to detect
various things in this area.
However, with this change, a whole category of files have been
moved from one category to another due to an (almost!) arbitrary
and certainly orthogonal piece of metadata.
Indeed, as I mentioned in passing there is a specific case in the
testsuite to ensure that lintian correctly warns (or doesn't warn),
despite executable bits on a file in question, so this is unlikely
to be hypothethical pedanticism on my part.
This is also ignoring that dealing with it would be rather ugly
AFIACT.
(Certainly understand what you mean about not wishing to carry a
Debian-specific patch, don't worry.)
Regards,
--
,''`.
: :' : Chris Lamb
`. `'` la...@debian.org / chris-lamb.co.uk
`-
