RFC: xinetd autoreconf

2012-07-27 Thread Salvo Tomaselli
Hello,

i am taking over the orphaned xinetd, and i am working here:
https://github.com/ltworf/xinetd-debian

the current package has a patch that basically is a replacement of the 
configure script. The patch is extremely complicated and i guess not really 
meant for human eyes.

I must premise that on my architecture xinetd compiles fine without that 
patch, but someone on #debian-it suggested that it might be necessary for 
other architectures.

What i was trying to do is to use dh_autoreconf and have the configure script 
regenerated automatically instead of including a patch i can't check.

But the autoreconf command fails. I've tried multiple versions and it always 
fails giving a long list of warnings about missing template and then
 autoreconf: /usr/bin/autoheader failed with exit status: 1

The original maintainer of xinetd replied after a long delay to my request and 
pointed me here, so i have absolutely no clue of what he did to re-generate 
the configure script.

At this point i would think of making an experimental upload just to see if it 
builds on every arch without that patch.
If someone is more expert on autotools and can see why autoreconf actually 
fails would be the ideal solution.


Suggestions anyone?

Bye

-- 
Salvo Tomaselli


signature.asc
Description: This is a digitally signed message part.


Re: RFC: xinetd autoreconf

2012-07-27 Thread Russ Allbery
Salvo Tomaselli tipos...@tiscali.it writes:

 But the autoreconf command fails. I've tried multiple versions and it
 always fails giving a long list of warnings about missing template and
 then
 autoreconf: /usr/bin/autoheader failed with exit status: 1

xinetd doesn't use autoheader or config.h.in templates.  You can't run
autoheader on that package without rewriting the configure.in script
(which I think should be done, but upstream seems to be using a
hand-rolled config.h.in, so this may be intentional on their part).  You
have to only run autoconf.

It looks like you can do that by overriding dh_autoreconf and running:

dh_autoreconf -- autoconf

instead of its default action.  That should be safe, since the package
doesn't use automake or libtool either, so the other things that
autoreconf runs are not relevant to it.

-- 
Russ Allbery (r...@debian.org)   http://www.eyrie.org/~eagle/


-- 
To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/87394d8v1a@windlord.stanford.edu



Re: RFC: xinetd autoreconf

2012-07-27 Thread Adam Borowski
On Fri, Jul 27, 2012 at 09:28:49AM +0200, Salvo Tomaselli wrote:
 the current package has a patch that basically is a replacement of the 
 configure script. The patch is extremely complicated and i guess not really 
 meant for human eyes.
 
 What i was trying to do is to use dh_autoreconf and have the configure script 
 regenerated automatically instead of including a patch i can't check.
 
 But the autoreconf command fails. I've tried multiple versions and it always 
 fails giving a long list of warnings about missing template and then
  autoreconf: /usr/bin/autoheader failed with exit status: 1

Ie, the package is shipping code without source.  A configure script is as
far away from something readable/editable as you can be while still using
shell.  I don't think anyone can possibly call it a preferred form for
modification with a straight face.

Yet another example why --disable-maintainer-mode is bad.


-- 
Copyright and patents were never about promoting culture and innovations;
from the very start they were legalized bribes to give the king some income
and to let businesses get rid of competition.  For some history, please read
https://en.wikipedia.org/wiki/Statute_of_Monopolies_1623


-- 
To UNSUBSCRIBE, email to debian-mentors-requ...@lists.debian.org
with a subject of unsubscribe. Trouble? Contact listmas...@lists.debian.org
Archive: http://lists.debian.org/20120727150302.ga13...@angband.pl