Re: RFS: sqlmap
> There was no ITP for sqlmap package. I filled in one on
> bugs.debian.org. Is it right? Thanks for notifying.
Next time, please file it before start packaging it (so we can avoid
duplicate work on the same tool, etc etc).
>> diff.gz is empty:
>> sqlmap_0.6-1.diff.gz 03-Sep-2008 14:15 20
>> what went wrong?
> Because it's a new package. Should diff.gz have something into it?
Yes, it has to, but I can see your problem: you are both packager and
upstream author. My suggestion is: keep upstream work separate from
debian packaging work.
diff.gz is the difference applied to upstream source code by the
packager (so it usually contains debian/ directory only). You have
"merged" both in orig tarball.
Please split that and only release upstream tarball without debian/
dir, and apply it only in a second moment. I just give a fast look,
and I noticed:
* debian/changelog has to contain changes related to debian package
*only*, so jsut one entry with "Initial package; Closes #<ITP
bugnumber>" is what we expect there
* debian/copyright has Daniele's email wrongly typed; if the source is
GPLv2, than link to /usr/share/common-licenses/GPL-2 (GPL is a link to
the lastest GPL version available, so now is GPL-3)
* it would have been better if you leave debian/rules as dh_make
template proposed to you
* moreover, dpatch integration is done wrong: check
* if you're upstream, why you need a patch system? can't you simply
release a new version? :)
* don't ship pdf (in docs) if they are not generated a build time
* check if some Build-Depends can go to Build-Depends-Indep
* adding a watch file might be intersting
* I still see "Description: an automatic SQL injection tool entirely
developed in Python" and I know why: you uploaded -2 revsion, that
DON'T upload the orig.tar.gz file (see why debianization has to be
That's enough for a 5 minutes check :)
Sandro Tosi (aka morph, Morpheus, matrixhasu)
My website: http://matrixhasu.altervista.org/
Me at Debian: http://wiki.debian.org/SandroTosi