Fedora Merge Review: arptables_jf http://cvs.fedora.redhat.com/viewcvs/devel/arptables_jf/
MUSTFIX - Epoch tag can be removed (0 is the default value) - BuildRoot should be %{_tmppath}/%{name}-%{version}-%{release}-root-%(%{__id_u} -n - no URL provided - "BuildPrereq: /usr/bin/perl" can be removed, perl is on the exception list; if you plan to keep it then you should use "BuildRequires(pre)" (and eventually explain why you explicitely want it ) - "Requires(post,postun): chkconfig" should be split in Requires(post), Requires(postun) - You should stick with either %{buildroot} or $RPM_BUILD_ROOT but not both - "rm -rf %{buildroot}" is not needed in %prep - "rm -rf $RPM_BUILD_ROOT" is missing in %install - "service" should be added to Requires - Missing SMP flags. If it doesn't build with it, please add a comment - $RPM_OPT_FLAGS is not used - "%config /etc/rc.d/init.d/arptables_jf" should use a macro, not a fixed path SHOULD FIX - Summary ended with dot - adding "INSTALL="%{__install} -c -p" to the make install line would preserve timestamps rpmlint has some info for us, too: rpmlint of arptables_jf: W: arptables_jf summary-ended-with-dot Userspace control program for the arptables network filter. W: arptables_jf no-url-tag - We know about these two already W: arptables_jf conffile-without-noreplace-flag /etc/rc.d/init.d/arptables_jf E: arptables_jf executable-marked-as-config-file /etc/rc.d/init.d/arptables_jf - /etc/rc.d/init.d/arptables_jf should not be marked as %config; could be left as such till F8T1 according to last week's guidelines W: arptables_jf service-default-enabled /etc/rc.d/init.d/arptables_jf - that's a good point. Do we need/want it enabled at start time? E: arptables_jf incoherent-subsys /etc/rc.d/init.d/arptables_jf arptables E: arptables_jf incoherent-subsys /etc/rc.d/init.d/arptables_jf arptables E: arptables_jf incoherent-subsys /etc/rc.d/init.d/arptables_jf arptables - Harmless, but maybe should be discussed. Why is the package called arptables_jf and not arptables ? W: arptables_jf no-reload-entry /etc/rc.d/init.d/arptables_jf - harmless, although a reload which does just start+stop would be easy to implement
wrt "BuildRequires(pre)" .. should be read "BuildRequires"
any idea what the difference between this package's source tree and arptables-0.0.3-3 from ebtables.sourceforge.net ? Is this package a fork ? Any pointers to the story behind how we ended up with two different ones would be much appreciated...
(In reply to comment #3) > any idea what the difference between this package's source tree > and arptables-0.0.3-3 from ebtables.sourceforge.net ? Is this package a > fork ? Any pointers to the story behind how we ended up with two different > ones would be much appreciated... No idea what should be marked as a fork. These two packages have the same basis.
Some troubles complained by rpmlint are fixed.
arptables_jf.x86_64: W: no-url-tag COMMENT: upstream dead arptables_jf.x86_64: W: missing-lsb-keyword Default-Stop in /etc/rc.d/init.d/arptables_jf COMMENT: intention arptables_jf.x86_64: E: incoherent-subsys /etc/rc.d/init.d/arptables_jf arptables arptables_jf.x86_64: E: incoherent-subsys /etc/rc.d/init.d/arptables_jf arptables arptables_jf.x86_64: E: incoherent-subsys /etc/rc.d/init.d/arptables_jf arptables COMMENT: wan't fix. This makes sense when new package is inserted to Fedora. Trying change it now could generate difficulties to users.
rechecking Version: 0.0.8 / Release: 19%{?dist} Please add a comment in the spec file, explaining that the upstream is dead and that the following two warnings cannot be avoided: arptables_jf.src: W: no-url-tag arptables_jf.src: W: invalid-url Source0: arptables_jf-0.0.8.tbz As of the incoherent-subsys error, I do not agree. I doubt that a sysadmin which requires this package will be puzzled by the fact that the lock file is called arptables or arptables_jf.
(In reply to comment #4) > (In reply to comment #3) > > any idea what the difference between this package's source tree > > and arptables-0.0.3-3 from ebtables.sourceforge.net ? Is this package a > > fork ? Any pointers to the story behind how we ended up with two different > > ones would be much appreciated... > > No idea what should be marked as a fork. These two packages have the same > basis. Looks pretty much like arptables_jf is the fork, since it reuses the arptables name with a _jf suffix, was created by a Redhat employee, and is also only used by Fedora.
Checked git commit 8b1d1dbab8f89c2301349f176aa57cd049cbbdf5 NO source files match upstream - there is no upstream for arptables_jf package YES package meets naming and versioning guidelines - package name contains underscore but package arptables_jf is on the list of packages which don't have to obey this particular guideline https://fedoraproject.org/wiki/Packaging:NamingGuidelines#Separators YES spec file is properly named, is cleanly written and uses macros consistently YES dist tag is present YES clean section and buildroot present YES licence field matches the actual license NO license text included in package - no text file with license is not included in source tarball, it should be present YES latest version is being packaged NO BuildRequiers are proper (Perl?) - I think perl is not needed to properly build the package YES compiler flags are appropriate YES package builds in mock YES debuginfo package looks complete NO rpmlint is silent - $ rpmlint arptables_jf.spec arptables_jf.spec: W: invalid-url Source0: arptables_jf-0.0.8.tbz 0 packages and 1 specfiles checked; 0 errors, 1 warnings. - warning about invalid-url is ok since there's no upstream, thus there's no url pointing to upstream's remote source code tarball - $ rpmlint arptables_jf-0.0.8-22.fc17.src.rpm arptables_jf.src: W: spelling-error Summary(en_US) Userspace -> User space, User-space, Users pace arptables_jf.src: W: spelling-error Summary(en_US) arptables -> portables, stables, tables arptables_jf.src: W: spelling-error %description -l en_US arptables -> portables, stables, tables arptables_jf.src: W: spelling-error %description -l en_US jf -> hf, jg, j arptables_jf.src: W: spelling-error %description -l en_US arpfilter -> filterer arptables_jf.src: W: spelling-error %description -l en_US firewalling -> fire walling, fire-walling, firewall arptables_jf.src: W: spelling-error %description -l en_US arp -> rap, tarp, carp arptables_jf.src: W: no-url-tag arptables_jf.src: W: invalid-url Source0: arptables_jf-0.0.8.tbz 1 packages and 0 specfiles checked; 0 errors, 9 warnings. - spelling errors could be safely ignored - no url tag and invalid source warnings could be ignored because there is no upstream - $ rpmlint arptables_jf-0.0.8-22.fc17.x86_64.rpm arptables_jf.x86_64: W: spelling-error Summary(en_US) Userspace -> User space, User-space, Users pace arptables_jf.x86_64: W: spelling-error %description -l en_US jf -> hf, jg, j arptables_jf.x86_64: W: spelling-error %description -l en_US arpfilter -> filterer arptables_jf.x86_64: W: spelling-error %description -l en_US firewalling -> fire walling, fire-walling, firewall arptables_jf.x86_64: W: spelling-error %description -l en_US arp -> rap, tarp, carp arptables_jf.x86_64: W: incoherent-version-in-changelog 0:0.0.8-22 ['0.0.8-22.fc17', '0.0.8-22'] arptables_jf.x86_64: W: no-url-tag arptables_jf.x86_64: W: missing-lsb-keyword Default-Stop in /etc/rc.d/init.d/arptables_jf arptables_jf.x86_64: E: incoherent-subsys /etc/rc.d/init.d/arptables_jf arptables arptables_jf.x86_64: E: incoherent-subsys /etc/rc.d/init.d/arptables_jf arptables arptables_jf.x86_64: E: incoherent-subsys /etc/rc.d/init.d/arptables_jf arptables 1 packages and 0 specfiles checked; 3 errors, 8 warnings. - spelling errors could be ignored - incoherent version in changelog warning is caused by use of dist tag and could be safely ignored - incoherent-subsys, changes due to satisfying rpmlint could break user's scripts, thus warning is ignored - spelling errors could be ignored - $ rpmlint arptables_jf-debuginfo-0.0.8-22.fc17.x86_64.rpm arptables_jf-debuginfo.x86_64: W: spelling-error Summary(en_US) arptables -> portables, stables, tables arptables_jf-debuginfo.x86_64: W: spelling-error Summary(en_US) jf -> hf, jg, j arptables_jf-debuginfo.x86_64: W: spelling-error %description -l en_US arptables -> portables, stables, tables arptables_jf-debuginfo.x86_64: W: spelling-error %description -l en_US jf -> hf, jg, j arptables_jf-debuginfo.x86_64: W: no-url-tag 1 packages and 0 specfiles checked; 0 errors, 5 warnings. - spelling errors could be ignored - no-url-tag warning is ok because there's no upstream YES final provides and requires look sane N/A %check is present and all tests pass YES no shared libraries are added to the regular linker search paths N/A owns the directories it creates - package content is copied into already existing directories YES doesn't own any directories it shouldn't YES no duplicates in %files YES scriptlets looks sane YES code, not content N/A large documentation files must go in a -doc subpackage. YES %docs are not necessary for the proper functioning of the package. YES no headers. YES no pkgconfig files. YES no libtool .la droppings. YES not a GUI app.
Ping?
Adding owner.
(In reply to comment #9) > Checked git commit > 8b1d1dbab8f89c2301349f176aa57cd049cbbdf5 > NO source files match upstream > - there is no upstream for arptables_jf package I put a comment to the Source0 tag. > NO license text included in package > - no text file with license is not included in source tarball, it should be > present Licensing guidelines says that: "If (and only if) the source package includes the text of the license(s) in its own file, then that file, containing the text of the license(s) for the package must be included in %doc. If the source package does not include the text of the license(s), the packager should contact upstream and encourage them to correct this mistake." The license text is not included in the tarball and upstream is dead so there's nothing I can do here. > NO BuildRequiers are proper (Perl?) > - I think perl is not needed to properly build the package Fixed in rawhide.
I don't see anything else to fix. Can we close this review ?
I think we can close this one.
This bug just popped out when I searched for packages under review but this looks completed without setting fedora-review+ Setting the flag therefore.