Please do not reply directly to this email. All additional comments should be made in the comments box of this bug report. Summary: Review Request cppad - A Package for Differentiation of C++ Algorithms https://bugzilla.redhat.com/show_bug.cgi?id=416761 ------- Additional Comments From bradbell@xxxxxxxxxx 2007-12-21 13:24 EST ------- (In reply to comment #6) A new version of http://www.seanet.com/~bradbell/cppad/cppad.spec has been uploaded. The corresponding rpm source file is http://www.seanet.com/~bradbell/cppad/cppad-20071221-1.fc7.src.rpm > * Source tarball > - Well, I tried to download the source from the written SourceURL, > however only I could find CppAD-2.0.1.tgz?> (source tarball check is needed for license check) I am working on how to best solve this problem (see comment number 7 for more details). > > * Build option > - build.log says: > --------------------------------------------------------- > 52 checking --with-Documentation... yes > 53 checking --with-Introduction... no > 54 checking --with-Example... yes > 55 checking --with-TestMore... yes > 56 checking --with-Speed... no > 57 checking --with-PrintFor... no > 58 checking --with-stdvector... no > --------------------------------------------------------- > You don't set some conditional configure option. Would you > please explain why? The Introduction and Speed options have been added to the %configure and %check sections of cppad.spec. The PrintFor option builds a test that is not automated (the users actually looks at the output). The stdvector option replaces the template vector class that has the most extensive testing. The standard vector template class is just one of three choices (and the default is being used by the current spec file). > > * Timestamps > - As this is noarch and the installed files are only texts, > keeping timestamps on installed files is highly preferable. > > For this package, the following keeps timestamps. > ---------------------------------------------------------- > %prep > %setup -q > > sed -i.stamp -e 's|cp -r|cp -a|' makefile > > %build > ..... > %install > rm -rf $RPM_BUILD_ROOT > export CPPROG="cp -p" > make install DESTDIR=$RPM_BUILD_ROOT > ....... This has been fixed in the corresponding upstream makefile.am by using the cp -a command (this made it necessary to increase the version number of cppad). > > ---------------------------------------------------------- > i.e. macros in %changelog are expanded (you can gain the warning > about these by rpmlint). > To avoid this, use %% in %changelog to stop macros expanding. > ---------------------------------------------------------- > * Wed Dec 20 2007 Brad Bell < bradell at seanet dot com > 20071203-2 > - Use the commands %%configure, %%check > ----------------------------------------------------------- Done. > BTW I recommend to use <>, not () for mail address. Done. > > * Argument list too long > - By the way, when I try normal rpmbuild: > ------------------------------------------------------------ > [tasaka1@localhost SPECS]$ LANG=C rpmbuild -bi --short-circuit cppad.spec > Executing(%install): /bin/sh -e /home/tasaka1/rpmbuild/INSTROOT/rpm-tmp.58659 > + umask 022 > + cd /home/tasaka1/rpmbuild/BUILD > + cd cppad-20071208 > ....... > cp -r ./doc/* > This has also been fixed in the upstream makefile.am file by copying the directory instead of files; i.e. cp -r ./doc/* should now be cp -a ./doc This error did not occur on my test system, so please see if the fix works for you. > ---------------------------------------------------------- > > * Some rpmlint > $ rpmlint <your srpm> shows: > ------------------------------------------------------------ > cppad.src: W: mixed-use-of-spaces-and-tabs (spaces: line 8, tab: line 5) > cppad.src: W: summary-ended-with-dot cppad base package (not installed). > ------------------------------------------------------------ > Please fix these (you can check what these mean by > $ rpmlint -I mixed-use-of-spaces-and-tabs , for example) Fixed. -- 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, or are watching someone who is. _______________________________________________ Fedora-package-review mailing list Fedora-package-review@xxxxxxxxxx http://www.redhat.com/mailman/listinfo/fedora-package-review