Huge number of lintian messages

[Holger Levsen]

Hi,

On Freitag, 1. Mai 2009, Thomas Lange wrote:
>     >> W: fai source: out-of-date-standards-version 3.8.0 (current is
>     >> 3.8.1)
>     > yeah...
> ?? What does yeah mean? Did you check if the current version applies
> to 3.8.1?

No, "yeah" just ment this the usual work for upgrading to a new policy version 
has to be done: check the upgrading checklist, do requiered changes if there 
are any, bump the version number.

> Since those commands are never executed by a FAI user (thery are only
> for internal use), I think a man page is not that important. 

Then those commands are located in the wrong directory and should be moved 
somewhere below /usr/share/fai/
/usr/(s)bin is for user commands and those commands are supposed to have a 
manpage.

Policy is there for a reason and one should not usually try to be smarter than 
policy. Or if, please also file bugs against policy.

And IIRC IRC suggestions correctly, faireboot is to be used by the user.

> It you 
> wanna have a man page, feel free to write one. A wishlist bug is not
> needed.

I agree here, but not for the reasons you stated.

>     > btw, I'd recommend to switch to asciidoc and write manpages and the
>     > fai-guide in asciidoc, its way easier to maintain than sgml and
>     > manpage sourcecode.
> I know. This is also my intention.

Cool. Unfortunatly my plate is really too full to reliably say I'll do that 
(and do it then..), but I can say I've done that for piuparts and it was 
straightforward, fun and the resulting documentation is way easier to edit.

>     >> W: fai-client: binary-without-manpage usr/sbin/setup_harddisks
>     > isn't it the perfect time now to remove setup_harddisks completly?!
> I like to do this in lenny+1.

I think you mean squeeze+1? If so, it would be nice to add a note to the guide 
for squeeze :-)

>     >> W: fai-server: maintainer-script-empty postinst
>     > fix in trunk
> No. IMO this is a false positive of lintian. Have a look at the script
> itself.

I did, and I could imagine that lintian is such a smartass to detect that the 
script really does nothing, except being a _unneeded_ placeholder for 
debhelper. On a second look I see those echo commands telling the user how to 
setup fai-server. 

So at first, I thought the correct action would have been to file a bug 
against lintian and I was about to do this. 

After the second look I believe this is actually a bug in FAI and lintian is 
correct, though the output of "lintian -i fai-server_3.2.20_all.deb" could be 
better:

W: fai-server: maintainer-script-empty postinst
[...]
N:   If the package currently doesn't need to do anything in this
N:   maintainer script, it shouldn't be included in the package.

See http://www.debian.org/doc/debian-policy/ch-binary.html#s-maintscripts and 
there the last but one paragraph of 3.9.1:

--quote-----
 If a package has a vitally important piece of information to pass to the user 
(such as "don't run me as I am, you must edit the following configuration 
files first or you risk your system emitting badly-formatted messages"), it 
should display this in the config or postinst script and prompt the user to 
hit return to acknowledge the message. Copyright messages do not count as 
vitally important (they belong in /usr/share/doc/package/copyright); neither 
do instructions on how to use a program (these should be in on-line 
documentation, where all the users can see them).
--quoteend----

Instructions on how to use the programm should not be displayed via maintainer 
scripts. And this is exactly what lintian is complaining about: the 
maintainer script does nothing except informing the user, which should be 
done via README.Debian.


(Policy 3.8.1 is outdated regarding the use of debconf btw, and the is an 
update scheduled regarding this for 3.8.2, see #206684.)


regards,
	Holger
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part.
Url : http://lists.uni-koeln.de/pipermail/linux-fai-devel/attachments/20090504/bc0732c4/attachment.bin