Re: Bug#731466: RFS - policyd-weight/0.1.15.2-6
+++ Werner Detter [2013-12-10 10:57 +0100]:
> Am 10.12.13 09:04, schrieb gregor herrmann:
> > On Tue, 10 Dec 2013 07:14:02 +0100, Werner Detter wrote:
> >
> >>>>> -. /lib/init/vars.sh
> >>>>> . /lib/lsb/init-functions
> >>>
> >>> You'll have to explain to me why this is right? /lib/init/vars.sh still
> >>> exists and appears to contain default values. Although it does seem the
> >>> the policyd-weight init script doesn't use of them - maybe that's the point?
> >>
> >> Right, but I had the lintian warning: init.d-script-call-internal-API
> >> before with ". /lib/init/vars.sh" in the init-script. I've googled
> >> around and found this bugreport:
> >>
> >> http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=710559
> >>
> >> So, I took it out.
> >
> > ... where Niels says that the lintian tag was premature and he
> > deactivated it in lintian git for the time being.
>
> Yes .. but .. Josh states:
>
> > /etc/init.d/skeleton is wrong; only sysvinit's own internal scripts
> > should use vars.sh directly. Other scripts should source
> > /lib/lsb/init-functions and use the log_* functions.
>
> Yes tag was premature and deactivated BUT will it be reenabled if
> /etc/init.d/skeleton is fixed? I can take it back in - no problem
> - but then i get the mentioned lintian-warning
> init.d-script-call-internal-API" .... so what to do here now?
Given that the init script doesn't use any of the variables from vars.sh
there is no reason to use an interface that is now deprecated to do
nothing, so I agree that taking this line out is the right thing to do.
The 'waiting for a fix' part is only relvaant if your script actually _is_ using those vars, IMHO.
So I'll upload as-is.
Wookey
--
Principal hats: Linaro, Emdebian, Wookware, Balloonboard, ARM
http://wookware.org/
Reply to: