Using backported debhelper considered tricky
Hi *,
I’d like to break a point in favour of *not* using backported
debhelper when the backport otherwise builds fine in stable
(or oldstable for that matter).
① The recent update of debhelper Breaks qt5-qmake in the version
in stable and there is no backported Qt fulfilling its dependency.
This leads to an irritating error message in pbuilder-satisfy‐
depends-aptitude (depends on debhelper (>= 11~) but cannot be
installed), but is otherwise circumventable.
② Worse, though, it Breaks cmake in the version in stable, but
there *is* a backported cmake available.
This can introduce undesired dependencies on other backports.
I’ve uploaded my package with a debhelper revert now (had to
do this for jessie-backports-sloppy anyway, so no biggie, I
was only using debhelper 11 in sid because lintian started
complaining when one was not using the latest and greatest…).
This is not to say that this should be a general policy, just
describe my perceived issues with it and serve as food for
thought before blindly using the latest and greatest. (Yes,
backreference intended. No slight against the lintian main‐
tainers intended, though.)
bye,
//mirabilos
--
Sometimes they [people] care too much: pretty printers [and syntax highligh-
ting, d.A.] mechanically produce pretty output that accentuates irrelevant
detail in the program, which is as sensible as putting all the prepositions
in English text in bold font. -- Rob Pike in "Notes on Programming in C"
Reply to: