Bug#969158: expeyes: maybe a false positive generated by mail_autoremovals.pl?
(note: this mail represents my opinions as an ordinary dd, I am not a member of the release team)
due to the fact that it is supposed to (build-)depend on binutils-avr, which
FTBFS.
As I understand it "(build-)depends" should be interpreted as
"depends or build-depends"
The source package expeyes generates one binary package, microhope, which
declares a dependency on avr-libc; should I downgrade this dependency down to
a recommendation? The binary package microhope does not need binutils-avr
as it is mainly an editor for small C or assembly language snippets. It needs
binutils-avr only when the end user will try to compile and link one one the
edited snippets.
The package description for microhope implies that while it may technically
be "mostly an editor" it's reason for existing is avr development. Downgrading
the dependency to a reccomends is arguably correct (reccomends is described as
"The Recommends field should list packages that would be found together with
this one in all but unusual installations.") and would get the autoremovals
tool out of your hair, but I would still question if a package that can't
fulfill it's primary purpose belongs in a Debian release.
IMO what really needs to happen here is that binutils-avr needs to be fixed,
either by changing the actual code or by changing the compiler flags to make
gcc less picky.
Reply to: