[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Re: I want to colaborate with debian-med



On Tue, 24 May 2005, Steffen Moeller wrote:

I wrote a lengthier reply earlier today but somehow feel that it did not
make it to the list.
Seems so.

I had a parallel work on Mummer over the weekend
Ahh, just another hint to use the ITP tagging of wnpp ...

that I just brought to a stage that it is close to lintian freeness. It
is only the csh issue remaining that Andreas mentioned and the
sharedlibdeps I commented out,
Upps, how did you do that?  Just not calling dh_shlibdeps?  I would
regard this as quick and dirty hack.

a real weird issue there, annotate
is not detected as an executable by ldd, which it should.
I now found the reason (not yet the solution, but this might be more simple):
I allways suspected a name space polution because I had some program annotate
in mind.  In fact a "locate annotate" has some hits on my system which are
some files from imagemagick and from mzscheme.  Strangely enough none of the
hits look like it would be something which ldd should search for.  Moreover
I'm pretty sure that my first try with ldd was

    $ ldd ./annotate

even if I did not wrote this explicitely in

    http://lists.debian.org/debian-med/2005/05/msg00016.html

So IMHO we have here more a bug in ldd than in the mummer program because
if you try

    $ mv annotate xyz_annotate
    $ ldd xyz_annotate
                libstdc++.so.5 => /usr/lib/libstdc++.so.5 (0x4002e000)
        libm.so.6 => /lib/tls/libm.so.6 (0x400e8000)
        libgcc_s.so.1 => /lib/libgcc_s.so.1 (0x4010b000)
        libc.so.6 => /lib/tls/libc.so.6 (0x40114000)
        /lib/ld-linux.so.2 => /lib/ld-linux.so.2 (0x40000000)

all seems fine.  I'm going to report this on debian-devel and wait for an
advise.  If we really want to push the mummer package we could go with
a renamed annotate for the first time (or even try to do a

     mv <path>/annotate <path>/xyz_annotate
     dh_shlibdeps
     mv <path>/xyz_annotate <path>/annotate

in the debian/rules file but this might have other side effects and would
be so strange that I would really avoid this.

The programs are not tested beyond a call to -h which seems to work
fine.  Adrian, I would be very glad if you could make some use of
my work and hope that it is not too much in redundance to what you have
done. I have addressed many changes to the upstream source that I
would like to see submitted to the upstream developers.
This is always a good idea.  What about these shell scripts.  They should
also be fixed upstream.  Would anybody of you both care for it.  If not
I could volunteer.

Thanks for your work

       Andreas.

--
http://fam-tille.de



Reply to: