Some (I believe) spurious warnings in Lintian for e2fsprogs

[Theodore Tso <tytso@mit.edu>]

Per the Lintain's user's manual, I'd like to ask your opinion on these
three Lintian warnings before I override them:

W: libuuid1: maintainer-script-needs-depends-on-adduser postinst
N:
N:   This script calls adduser, but the package does not depend or
N:   pre-depend on the adduser package.
N:

I'm not actually calling the adduser program.  Instead the script is
checking for the adduser configuration file and respecting it, per a
user request that e2fsprogs respect the adduser configuration, but
because e2fsprogs is an essential package, I didn't want to actually
depend on the adduser package.  The shell script fragment in question
which seems to be trigger lintian is:

ADDUSERCONF='/etc/adduser.conf'

if test -f $ADDUSERCONF; then
    FIRST_SYSTEM_UID=$(sed -n "s/^[[:space:]]*FIRST_SYSTEM_UID[[:space:]]*=[[:space:]]*[\"']\?\([^\"']*\)[\"']\?/\1/p" $ADDUSERCONF)
    LAST_SYSTEM_UID=$(sed -n "s/^[[:space:]]*LAST_SYSTEM_UID[[:space:]]*=[[:space:]]*[\"']\?\([^\"']*\)[\"']\?/\1/p" $ADDUSERCONF)
    FIRST_SYSTEM_GID=$(sed -n "s/^[[:space:]]*FIRST_SYSTEM_GID[[:space:]]*=[[:space:]]*[\"']\?\([^\"']*\)[\"']\?/\1/p" $ADDUSERCONF)
    LAST_SYSTEM_GID=$(sed -n "s/^[[:space:]]*LAST_SYSTEM_GID[[:space:]]*=[[:space:]]*[\"']\?\([^\"']*\)[\"']\?/\1/p" $ADDUSERCONF)
fi

I suspect lintian is simply searching for the string "adduser" to
trigger the lintian warning.

W: uuid-runtime: command-with-path-in-maintainer-script prerm:5 /usr/sbin/uuidd
N:
N:   The indicated program run in a maintainer script has a prepended path.
N:   Programs called from maintainer scripts normally should not have a
N:   path prepended. dpkg ensures that the PATH is set to a reasonable
N:   value, and prepending a path may prevent the local administrator from
N:   using a replacement version of a command for some local reason.
N:   
N:   Refer to Debian Policy Manual section 6.1 (Introduction to package
N:   maintainer scripts) for details.
N:

Here, I want to make sure the uuidd daemon is shutdown before removing
the package.  Since the uuidd daemon is provided by the uuid-runtime
package, and I *know* exactly where it is, and I ***don't*** want the
script to be confused by some other pathname being used by the local
administrator, I specify an explicit pathname.  In addition, given the
explicit test to make sure /usr/sbin/uuidd exists, it seems
non-sensical to remove the prepended path here:

if [ -x /usr/sbin/uuidd ]
then
	/usr/sbin/uuidd -k || true
fi

W: e2fsprogs source: debian-rules-calls-debhelper-in-odd-order dh_gencontrol (line 519)
N:
N:   One of the targets in the debian/rules file for this package calls
N:   debhelper programs in an odd order. Normally, dh_makeshlibs should be
N:   called before dh_shlibdeps or dh_installdeb, dh_shlibdeps should be
N:   called before dh_gencontrol, and all should be called before
N:   dh_builddeb. Calling them in the wrong order may cause incorrect or
N:   missing package files and metadata.
N:

Things are getting called out of order here because of the need to
create udeb files.  And the rules file is complicated enough that
keeping the udeb files close together seems to much easier to
understand and in the long run make the rules file as maintainable as
possible given the complexity of the e2fsprogs package.

I'm not sure there's much that can be done here, other than my just
overriding the lintian warning for this package.

Any comments or suggestions are greatly appreciated.

Regards,


							- Ted