[Bug 866325] New: Review Request: ugene - genome analysis suite
by Red Hat Bugzilla
https://bugzilla.redhat.com/show_bug.cgi?id=866325
Bug ID: 866325
QA Contact: extras-qa(a)fedoraproject.org
Severity: medium
Version: rawhide
Priority: unspecified
CC: notting(a)redhat.com,
package-review(a)lists.fedoraproject.org
Assignee: nobody(a)fedoraproject.org
Summary: Review Request: ugene - genome analysis suite
Regression: ---
Story Points: ---
Classification: Fedora
OS: Linux
Reporter: yalgaer(a)unipro.ru
Type: ---
Documentation: ---
Hardware: All
Mount Type: ---
Status: NEW
Component: Package Review
Product: Fedora
Spec URL: http://ugene.unipro.ru/downloads/ugene.spec
SRPM URL: http://ugene.unipro.ru/downloads/ugene-1.11.2-1.fc17.src.rpm
Description:
ReReview is needed for UGENE - complex bioinformatics toolkit based on Qt.
Unipro UGENE is a cross-platform visual environment for DNA and protein
sequence analysis. UGENE integrates the most important bioinformatics
computational algorithms and provides an easy-to-use GUI for performing
complex analysis of the genomic data. One of the main features of UGENE
is a designer for custom bioinformatics workflows.
Fedora Account System Username: yalgaer
See also https://bugzilla.redhat.com/show_bug.cgi?id=481717
--
You are receiving this mail because:
You are on the CC list for the bug.
10 years, 9 months
[Bug 658754] Review Request: cubrid - a very fast and reliable open source SQL database server
by Red Hat Bugzilla
https://bugzilla.redhat.com/show_bug.cgi?id=658754
--- Comment #59 from Michael Schwendt <bugs.michael(a)gmx.net> ---
> We tried to strip everything and yet the warning still occurs.
You are not supposed to strip anything, since that is done during -debuginfo
extraction by rpmbuild.
Stripping manually seems to work, though (just for testing):
$ file cm_admin
cm_admin: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically
linked (uses shared libs), for GNU/Linux 2.6.32,
BuildID[sha1]=0x526ce58d29207bb44fda19c7ad91e43471e1afbe, not stripped
$ strip cm_admin
$ file cm_admin
cm_admin: ELF 64-bit LSB executable, x86-64, version 1 (SYSV), dynamically
linked (uses shared libs), for GNU/Linux 2.6.32,
BuildID[sha1]=0x526ce58d29207bb44fda19c7ad91e43471e1afbe, stripped
$
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=4Cu5iTdlpD&a=cc_unsubscribe
10 years, 9 months
[Bug 658754] Review Request: cubrid - a very fast and reliable open source SQL database server
by Red Hat Bugzilla
https://bugzilla.redhat.com/show_bug.cgi?id=658754
--- Comment #56 from Esen Sagynov <kadishmal(a)gmail.com> ---
Hi,
> * First of all you must use %configure instead of ./configure. It sets all necessary variables (libdir, etc). if you'e insisting on using ./configure then please explain this.
Explained in the spec file.
# We cannot use (%)configure instead of (./)configure because
# (%)configure sets everything including `libdir`, `bindir`, etc.
# If `libdir` is set, CUBRID installation will fail because at this
# moment CUBRID does not support this option. In fact, there are
# other options which are also not supported. CUBRID requires
# most of its libraries to exist under the same directory where
# it is installed. Thus, because we insist on using `./configure`,
# `rpmlint` generates the following warning mesage:
# `W: configure-without-libdir-spec`. This is expected.
> * rpmling throws a lot of warnings about unstripped-binary-or-object. I advise you to fix ./configure usage as mentioned above and repeat building in Koji.
Yes, we know about these warnings, but for some reason they don't get resolved
even if we strictly check all binaries and make them executable. We tried to
strip everything and yet the warning still occurs.
> * Requires(post): systemd, Requires(preun): systemd, Requires(postun): systemd - these should be dropped. The only viable target w/o systemd is EPEL 6 (I personally don't recommend adding new packages into EPEL 5).
All dropped.
> * "Requires: %{name}%{?_isa} = %{cubrid_version}.%{build_version}" is not enough. You should require %{release} as well.
Replaced with "Requires: %{name}%{?_isa} = %{version}-%{release}".
> * demodb has "%defattr(-,cubrid,cubrid,-)". Is it really necessary - to lock out demo (e.g. samples, publicly available in Internet) from a user which has shell access to the computer? I suppose this is a leftover and should be simply removed. Otherwise please explain this.
Added comments to explain this.
# `demodb` demo database is not a sample database publicly
# available on the Internet. It is just a demo database
# distributed together with CUBRID binaries. `demodb`
# database can be installed only by cubrid user, and
# since the related files are stored in
# `/usr/share/cubrid/demo`, it is logical that they are
# owned by the same cubrid user.
> * Regarding "%defattr(-,cubrid,cubrid,-)" in the main %files section. I suppose it should be "%defattr(-,%{cubrid_user},%{cubrid_user},-)". Just for better consistency with other places where you setting permissions.
Fixed.
Latest spec is at
http://sourceforge.net/projects/cubrid/files/CUBRID-9.1.0/Linux/Fedora-RP....
F19 koji build: http://koji.fedoraproject.org/koji/taskinfo?taskID=5788595
RPM:
http://kojipkgs.fedoraproject.org//work/tasks/8596/5788596/cubrid-9.1.0.0...
F18 koji build: http://koji.fedoraproject.org/koji/taskinfo?taskID=5788647
http://kojipkgs.fedoraproject.org//work/tasks/8648/5788648/cubrid-9.1.0.0...
rawhide build fails at this moment
http://koji.fedoraproject.org/koji/taskinfo?taskID=5788704 because, as I've
noticed, armv7hl architecture was added which is not supported by CUBRID.
--
You are receiving this mail because:
You are on the CC list for the bug.
Unsubscribe from this bug https://bugzilla.redhat.com/token.cgi?t=ZTYjn2YJYC&a=cc_unsubscribe
10 years, 9 months