[Bug 564520] Review Request: frama-c - Framework for source code analysis of C software

bugzilla at redhat.com bugzilla at redhat.com
Tue Feb 16 21:09:42 UTC 2010


Please do not reply directly to this email. All additional
comments should be made in the comments box of this bug.


https://bugzilla.redhat.com/show_bug.cgi?id=564520

--- Comment #9 from David A. Wheeler <dwheeler at dwheeler.com> 2010-02-16 16:09:31 EST ---
I've started to walk through the Fedora Guidelines and comparing with this
draft package.  Here are some more comments.

The %files list isn't right.  It ends with:
 %exclude %{_datadir}/frama-c
which makes these lines pointless:
 %{_datadir}/frama-c/why
 %{_datadir}/frama-c/manuals
Basically, %{_datadir}/frama-c/why and ../manuals don't get packaged at all.
The file list in -devel don't look right at all to me; they look like examples
but NOT code necessary for developers depending on frama-c.
(See http://fedoraproject.org/wiki/Packaging:OCaml for more on OCaml -devel
packages.)
I suggest re-examining the %files list, so that they get split more cleanly
*AND* so that there's a -doc subpackage.

Strictly speaking, what you're packaging is "Frama-C Beryllium 2" not
"Beryllium".

This package contains a GUI, so there should be a .desktop file.

The Makefile uses "$(CP)" everywhere, but its definition (in
share/Makefile.common) doesn't preserve timestamps (this impacts the 'make
install' in the -devel package in particular).  You need to try to preserve
timestamps.  One way would be to modify share/Makefile.common so that:
 CP      = cp -f
becomes:
 CP      = cp -f --preserve=timestamps


Have you tried passing the SMP flags, e.g.:
  make %{?_smp_mflags}
if that FAILS, then that should be documented, otherwise you should try to
build using SMP.


Thanks for working on this package, I really appreciate it.

-- 
Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.



More information about the package-review mailing list