On Mon, 2020-05-18 at 14:35 +0200, Pierre-Elliott Bécue wrote:
> Le dimanche 17 mai 2020 à 22:27:52+0100, Adam D. Barratt a écrit :
> > Control: tags -1 + moreinfo
> >
> > On Sat, 2020-05-16 at 22:39 +0200, Pierre-Elliott Bécue wrote:
> > > +policyd-rate-limit (1.0.1.1-1+deb10u1) buster; urgency=medium
> >
> > The version needs to be lower than the package in unstable, so
> > 1.0.1.1-
> > 1~deb10u1
>
> Arf, my bad, I forgot that again.
:-)
> > -Build-Depends: debhelper (>= 11~),
> > +Build-Depends: debhelper-compat (= 11),
> >
> > As a general note, that's not particularly great for a stable
> > update,
> > even though it's effectively a no-op (because it's not part of
> > resolving the issues). As part of a backport I wouldn't request not
> > including it though.
>
> Dropped.
OK. As I said, in this instance it'd be OK.
> > - raise ValueError("connection closed")
> > + raise PolicydConnectionClosed()
> > [...]
> > - except Exception as error:
> > + except PolicydConnectionClosed:
> > + if config.debug:
> > + sys.stderr.write("Connection closed\n")
> >
> > Does anything rely on the specific strings being output here?
>
> It's debug output. Although anyone could decide to use this output to
> handle some things, it's not supposed to be.
Fair enough.
> Here is a new diff.
Thanks.
That appears to have dropped a number of entries from the changelog for
the unstable upload it's based on, which is a little confusing to say
the least.
If you're dropping changes for the backport, then the unstable
changelog stanza should be exactly as it was for the upload to
unstable, and any changes that weren't included should be itemised in
the changelog for the stable upload, if they're sufficiently relevant.
Regards,
Adam
