sage-6.0: Error installing 4ti2 package

[rickhg12hs]

Trying to install 4ti2 package generates an error.  How can this be fixed?

[from log file]

...

****************************************************

Host system:

Linux steelers.net 3.12.6-200.fc19.i686.PAE #1 SMP Mon Dec 23 16:58:40 UTC 2013 

i686 i686 i386 GNU/Linux

****************************************************

---

In file included from VectorArrayAPI.cpp:26:0:

../../src/groebner/VectorArrayAPI.h: In instantiation of 'static void _4ti2_::VectorArrayAPI::convert(const T1&, T2&) [with T1 = __gmp_expr<__mpz_struct [1], __mpz_struct [1]>; T2 = long long int]':

VectorArrayAPI.cpp:99:30:   required from here

../../src/groebner/VectorArrayAPI.h:67:8: error: cannot convert 'const __gmp_expr<__mpz_struct [1], __mpz_struct [1]>' to 'long long int' in assignment

     v2 = v1;

        ^

make[2]: *** [lib4ti2int64_la-VectorArrayAPI.lo] Error 1

[William Stein]

On Sun, Jan 12, 2014 at 8:36 AM, rickhg12hs <rickh...@gmail.com> wrote:
> Trying to install 4ti2 package generates an error. How can this be fixed?
>

Here's my recipe (for SageMathCloud) for building 4ti2:

# 4ti2 into sage: until the optional spkg gets fixed:


./sage -sh
cd /tmp; wget http://www.4ti2.de/version_1.6/4ti2-1.6.tar.gz &&
tar xf 4ti2-1.6.tar.gz && cd 4ti2-1.6 ; ./configure
--prefix=/usr/local/sage/current/local/; time make -j16
make install # this *must* be a separate step!!
rm -rf /tmp/4ti2*

> --
> You received this message because you are subscribed to the Google Groups
> "sage-support" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to sage-support...@googlegroups.com.
> To post to this group, send email to sage-s...@googlegroups.com.
> Visit this group at http://groups.google.com/group/sage-support.
> For more options, visit https://groups.google.com/groups/opt_out.



--
William Stein
Professor of Mathematics
University of Washington
http://wstein.org

[rickhg12hs]

Unfortunately, the recipe generated the same error.

/bin/sh ../../libtool  --tag=CXX   --mode=compile g++ -DHAVE_CONFIG_H -I. -I../../src -I../../src/4ti2  -D_4ti2_INT64_   -I./..  -D__STDC_LIMIT_MACROS -DNDEBUG -Wall -std=c++0x -MT lib4ti2int64_la-VectorArrayAPI.lo -MD -MP -MF .deps/lib4ti2int64_la-VectorArrayAPI.Tpo -c -o lib4ti2int64_la-VectorArrayAPI.lo `test -f 'VectorArrayAPI.cpp' || echo './'`VectorArrayAPI.cpp

libtool: compile:  g++ -DHAVE_CONFIG_H -I. -I../../src -I../../src/4ti2 -D_4ti2_INT64_ -I./.. -D__STDC_LIMIT_MACROS -DNDEBUG -Wall -std=c++0x -MT lib4ti2int64_la-VectorArrayAPI.lo -MD -MP -MF .deps/lib4ti2int64_la-VectorArrayAPI.Tpo -c VectorArrayAPI.cpp -o lib4ti2int64_la-VectorArrayAPI.o

In file included from VectorArrayAPI.cpp:26:0:

../../src/groebner/VectorArrayAPI.h: In instantiation of 'static void _4ti2_::VectorArrayAPI::convert(const T1&, T2&) [with T1 = __gmp_expr<__mpz_struct [1], __mpz_struct [1]>; T2 = long long int]':

VectorArrayAPI.cpp:99:30:   required from here

../../src/groebner/VectorArrayAPI.h:67:8: error: cannot convert 'const __gmp_expr<__mpz_struct [1], __mpz_struct [1]>' to 'long long int' in assignment

     v2 = v1;

        ^

make[3]: *** [lib4ti2int64_la-VectorArrayAPI.lo] Error 1

make[3]: Leaving directory `/tmp/4ti2-1.6/src/groebner'

[rickhg12hs]

FYI, building 4ti2 independently of Sage on my system completes without error.  In addition, all "make check" tests pass.

The config.log files are largely the same, but there are differences.  Here's a few that may be significant.

[diff -d working/config.log sage/config.log]

...

2228c2064

< CFLAGS='-O3 -fomit-frame-pointer -malign-double -fstrict-aliasing -ffast-math -march=pentiumpro -Wall -march=pentiumpro'

---

> CFLAGS='-O3 -fomit-frame-pointer -malign-double -fstrict-aliasing -ffast-math -march=pentiumpro -Wall'

2234c2070

< CXXFLAGS='-O3 -fomit-frame-pointer -malign-double -fstrict-aliasing -ffast-math -Wall -std=c++0x'

---

> CXXFLAGS=' -Wall -std=c++0x'

...

2600a2437

> #define _4ti2_HAVE_MPZ_INT64_CONVERSION 1

[rickhg12hs]

Was there there something from the original Sage install (compiled from source) that produced the different CXXFLAGS?

Strange that only the Sage config.log had "#define _4ti2_HAVE_MPZ_INT64_CONVERSION 1".