libgmpxx4-shlibs _unstable_ port information

Info2: <<
Package: libgmpxx4%type_pkg[-64bit]-shlibs
Version: 4.2.1
Revision: 1009
Distribution: 10.4, 10.5
GCC: 4.0
Type: -64bit (boolean)
Architecture: ( %type_raw[-64bit] = -64bit ) powerpc, ( %type_raw[-64bit] = -64bit ) i386
Source: mirror:gnu:gmp/gmp-%v.tar.gz
Source-MD5: 5603ae75bc0cb1b21ad0187308a23c84
Source2: http://www.math.jmu.edu/~martin/gmp-4.2.1-core2-port.tar.gz
Source2-MD5: 0e0dcd8fc12461a54947c522635287e8
PatchFile: %{Ni}.patch
PatchFile-MD5: 791a47972ef252dffe0d12105c446e4b
PatchScript: <<
patch -p1 < %{PatchFile}
cd mpn/x86; rm *dive_1* */*dive_1* */*/*dive_1*; rm */*mode1o* */*/*mode1o*
<<
Depends: ( %type_raw[-64bit] = -64bit ) 64bit-cpu, gmp%type_pkg[-64bit]-shlibs
BuildDepends: fink (>= 0.26.0), gcc4.0 (>= 4.0.1-1)
Replaces: gmp (<< 4.0.1-1)
NoSetMAKEFLAGS: true
SetMAKEFLAGS: -j1
ConfigureParams: --infodir='${prefix}/share/info' --enable-cxx
CompileScript: <<
#! /bin/sh -ev
if [ "%type_raw[-64bit]" == "." ]; then
ABI=32 ./configure %c --libdir='${prefix}/%lib';
make;
make check ;
else
# Using Jason Martin's core2 port on intel.
# On powerpc, "make check" fails.
case "%m" in
i386) export CFLAGS="-m64 -fast"; ./configure %c --libdir='${prefix}/%lib' --build=amd64-apple-darwin; cp ../gmp-4.2.1-core2-port/amd64call.asm tests; cp ../gmp-4.2.1-core2-port/add_n.asm ../gmp-4.2.1-core2-port/sub_n.asm ../gmp-4.2.1-core2-port/lahf_sahf_test.sh mpn; make; make check ;;
powerpc) ABI=mode64 ./configure %c --libdir='${prefix}/%lib'; make ;;
esac
fi
<<
InstallScript: <<
make install prefix=%i
rm -r %i/include %i/share/info %i/%lib/libgmp.* %i/%lib/libgmpxx.dylib %i/%lib/libgmpxx.a %i/%lib/libgmpxx.la
<<
Shlibs: <<
%p/%lib/libgmpxx.3.dylib 5.0.0 %n (>= 4.2.1-1) %type_num[-64bit]
<<
Description: Shared C++ library for gmp package (obsolete version)
DocFiles: doc/* AUTHORS COPYING COPYING.LIB ChangeLog INSTALL NEWS README ../gmp-4.2.1-core2-port/README:README.core2-port
License: LGPL
DescDetail: <<
GNU MP is a portable library written in C for arbitrary precision
arithmetic on integers, rational numbers, and floating-point numbers.
<<
DescPort: <<
Patch from Ben Hines to build libgmpxx, as of vers. 4.1-2. As of vers.
4.1.4-1, libgmpxx gets its own splitoff.

As of vers. 4.2, mpfr is a separate library, available in fink as the
libmpfr1 package.
<<
Maintainer: Dave Morrison
Homepage: http://www.swox.com/gmp/
<<