--- /dev/null
+%include /usr/lib/rpm/macros.perl
+%define pdir AI
+%define pnam NeuralNet-BackProp
+Summary: A simple back-prop neural net that uses Delta's and Hebbs' rule.
+Name: perl-%{pdir}-%{pnam}
+Version: 0.89
+Release: 1
+License: GPL/Artistic
+Group: Development/Languages/Perl
+Source0: ftp://ftp.cpan.org/pub/CPAN/modules/by-module/%{pdir}/%{pdir}-%{pnam}-%{version}.zip
+BuildRequires: perl >= 5.6
+BuildRequires: rpm-perlprov >= 3.0.3-26
+BuildArch: noarch
+BuildRoot: %{tmpdir}/%{name}-%{version}-root-%(id -u -n)
+
+%description
+AI::NeuralNet::BackProp is a simply back-propagation, feed-foward neural
+network designed to learn using a generalization of the Delta rule and
+a bit of Hopefield theory.
+
+# %description -l pl
+# TODO
+
+%prep
+%setup -q -n %{name}-%{version} -c
+
+%build
+perl Makefile.PL
+%{__make}
+%{!?_without_tests:%{__make} test}
+
+%install
+rm -rf $RPM_BUILD_ROOT
+
+%{__make} install DESTDIR=$RPM_BUILD_ROOT
+
+install -d $RPM_BUILD_ROOT%{_examplesdir}/%{name}-%{version}
+install examples/* $RPM_BUILD_ROOT%{_examplesdir}/%{name}-%{version}
+
+%clean
+rm -rf $RPM_BUILD_ROOT
+
+%files
+%defattr(644,root,root,755)
+%{perl_sitelib}/AI/NeuralNet/*.pm
+%doc Changes *.htm README
+%attr(755,root,root) %{_examplesdir}/%{name}-%{version}/*.pl
+%{_examplesdir}/%{name}-%{version}/*.dat
+%{_mandir}/man3/*