[Bug 1114737] Review Request: openms - Open-source software for LC/MS data management and analyses

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1114737



--- Comment #20 from Jerry James <loganjerry at gmail.com> ---
There is now a symlink %{python2_sitearch}/pyopenms/OpenMS to the files in
openms-data, but those files still exist in
%{python2_sitearch}/pyopenms/share/OpenMS, too.

The python shared object is being stripped, which prevents good debuginfo
generation.  Look for "-Wl,-s" in pyOpenMS/setup.py.

Also, there are still a few rpmlint complaints that should be addressed, namely
these:

openms.x86_64: W: unused-direct-shlib-dependency
/usr/lib64/libOpenSwathAlgo.so.1.11.1 /lib64/libQtCore.so.4
openms.x86_64: W: unused-direct-shlib-dependency
/usr/lib64/libOpenSwathAlgo.so.1.11.1 /lib64/libQtGui.so.4

These happen because of the order of the command line arguments when linking,
with -Wl,--as-needed coming after -lQtCore -lQtGui.  The LDFLAGS setting in the
spec file should instead be this:

export LDFLAGS="$LDFLAGS -Wl,--as-needed -lQtCore -lQtGui"

openms-data.noarch: E: zero-length /usr/share/OpenMS/IDPool/IDPool.txt

Not necessarily a problem, just want to check that that is correct.

openms-devel.x86_64: W: no-dependency-on openms/openms-libs/libopenms

Since the libraries are in the main package, -devel should explicitly depend on
the main package.

openms-devel.x86_64: E: zero-length
/usr/include/OpenMS/VISUAL/moc_EnhancedTabBarWidgetInterface.cxx

As above, not necessarily a problem, but I want to check.

openms-doc.noarch: W: spurious-executable-perm
/usr/share/doc/openms-doc/html/style_ini.css
openms-doc.noarch: E: wrong-script-end-of-line-encoding
/usr/share/doc/openms-doc/html/style_ini.css
python-openms.x86_64: E: non-standard-executable-perm
/usr/lib64/python2.7/site-packages/pyopenms/pyopenms.so 0775L

I think these have obvious fixes.

Finally, can you update the title of this bug to match the new Summary line? 
Past experience shows that git creation for this package won't happen unless
those two match.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are always notified about changes to this product and component