[debhelper-devel] [RFC 0/4] improve verbosity/silentness of debhelper

Bernhard R. Link brlink at debian.org
Sat Jan 3 15:41:38 UTC 2015


* Axel Beckert <abe at debian.org> [150102 09:22]:
> > * Bernhard R. Link <brlink at debian.org> [141229 14:11]:
> > > This series tries to improve the verbosity and silentness of debhelper a
> > > bit.
[...]
> Hadn't time to run debhelper with these patches, but the idea and
> patches look helpful to me.
>
> I though wonder if this is a feature which should depend on the
> compatibility level. (No opinion there, though, just wondering.)

1/4 and 4/4 do not change anything unless DH_QUIET is set, so I hope we
all agree no compatibility level is usefull there at all.

2/4 only prints a bit more, so I hope noone sees any reason here, too.

3/4 will give --disable-silent-rules to autoconf even without DH_VERBOSE
that was given only with DH_VERSOSE before (and --enable-silent-rules
with DH_QUIET, but behaviour with DH_QUIET is totally new).
For actual autoconf configure that option should cause no harm (autoconf
ignores unknown --enable- and --disable, so even very old autoconf will
not be effected. For things not being autoconf but having a configure and
unless configure not ignoring unknown --enable/--disable already have
at least the problem that they will fail with DH_VERBOSE. They also get
--disable-maintainer-mode and --disable-dependency-tracking anyway, so
we a talking about a non-autoconf configure.sh that does ignore
--disable-maintainer-mode and --disable-dependency-tracking but fails
with --disable-silent-rules, which it would already get with DH_VERBOSE.

As that flag also fixes a bug of some autoconf use packages (useless
buildd logs) I'd argue that this nothing needing a compat-guard.
(Especially as it will only happen so early in the stretch release cycle).


	Bernhard R. Link
-- 
F8AC 04D5 0B9B 064B 3383  C3DA AFFC 96D1 151D FFDC



More information about the debhelper-devel mailing list