Fedora Merge Review: slang http://cvs.fedora.redhat.com/viewcvs/devel/slang/ Initial Owner: mlichvar
$ cd slang/devel $ cvs remove -f slang-2.1.2-makefile.patch $ cvs commit -m '' The file is unused and has been superseded. $ rpmlint /home/qa/tmp/rpm/RPMS/slang-2.1.3-1.fc8.i386.rpm slang.i386: W: file-not-utf8 /usr/share/doc/slang-2.1.3/changes.txt => minor -- a few contributor's fornames use various encodings > warning: call to '__warn_memset_zero_len' declared with > attribute warning: memset used with constant zero length > parameter; this could be due to transposed parameters This is a false positive. > checking for the onig library and header files oniguruma.h... no oniguruma-devel is in Fedora, so add --without-onig for reproducible builds if it should not be built with. Else add it as BuildRequires. Else I don't see any packaging problems.
slang-2.1.3-2.fc9 buildrequires oniguruma-devel and the old makefile patch has been removed.
On current version: rpmlint on SRPM is clean. rpmlint on RPMS: slang.i386: W: file-not-utf8 /usr/share/doc/slang-2.1.4/changes.txt The character encoding of this file is not UTF-8. Consider converting it in the specfile's %prep section for example using iconv(1). Easily fixable. slang-static.i386: W: no-documentation The package contains no documentation (README, doc, etc). You have to include documentation files. Fix if static-only docs exist. Source tag should be Source0. Comment in spec on patch upstream status. Otherwise, looks good on full review, no other blockers. Also noticed that this review had been flagged + but never assigned or closed.
The changelog is a mix of UTF8 and ISO-8859-1. Source seems to be a valid rpm tag, or is there a guideline requiring use of Source0? The patches were not send to upstream, they just fix some file permissions, rpath and disable using a private glibc symbol to follow our guidelines.
(In reply to comment #4) > The changelog is a mix of UTF8 and ISO-8859-1. Right. Should be UTF-8. > Source seems to be a valid rpm tag, or is there a guideline requiring use of > Source0? I got confused, I was thinking pf Patch vs. Patch0. Still a good practice, but not a blocker. > The patches were not send to upstream, they just fix some file permissions, > rpath and disable using a private glibc symbol to follow our guidelines. https://fedoraproject.org/wiki/Packaging/Guidelines#All_patches_should_have_an_upstream_bug_link_or_comment So fix the changelog ang comment on the patches in the spec and we're good.
Should be fixed in slang-2.1.4-2.fc11.
Excellent, thank you! APPROVED.