OK, I understand. If you had CC'd me into the discussion at #9914 I could have helped (not all external spkgs' authors are Sage developers too, but some are!).
The only functions which directly call libpari are in my base lib called libjc. all the other code calls functions in libjc. But I always thought that I had to link in libpari as well to get workable binaries. Is that wrong? Incidentally there is a linear dependence lbjc <-- libqcurves <-- librank <-- libg0n. I could reorganise things so that there was a single library, but that would require changing habits of around 15 years.... John On Tue, Oct 26, 2010 at 9:31 PM, leif <not.rea...@online.de> wrote: > On 26 Okt., 21:47, John Cremona <john.crem...@gmail.com> wrote: >> Leif, >> >> Browsing the link led me to >> http://code.google.com/p/sagemath/source/detail?r=bc2913bfc76dd758bc4... >> and some comments of yours about (I think) whether mwrank/eclib/etc >> require the pari library. > > So it appears to be quite useful. :-) > >> They do! They only use it in a small way >> (for factoring integers) but they do use it. > > The question was rather if we have to specify the PARI library in the > extension modules' library lists, as they only use PARI *indirectly* > through the other libs specified (e.g. "curvesntl", "g0nntl", "jcntl", > "rankntl"). > > At least on Linux this is not the case. > > (The dependency on PARI is already recorded in those libraries > directly using it. "readelf -d local/lib/libcurvesntl.so" e.g. shows: > ... > 0x0000000000000001 (NEEDED) Shared library: [libntl.so] > 0x0000000000000001 (NEEDED) Shared library: [libgmp.so.3] > 0x0000000000000001 (NEEDED) Shared library: [libpari- > gmp-2.4.so.3] > 0x0000000000000001 (NEEDED) Shared library: [libjcntl.so] > 0x0000000000000001 (NEEDED) Shared library: [libstdc++.so. > 6] > 0x0000000000000001 (NEEDED) Shared library: [libm.so.6] > 0x0000000000000001 (NEEDED) Shared library: [libgcc_s.so. > 1] > 0x0000000000000001 (NEEDED) Shared library: [libc.so.6] > ...) > > (Though PARI is also unnecessarily linked to /some/ of the libraries > provided by ECLIB, i.e. ECLIB's Makefile(s) could be "simplified" or > improved as well to not link PARI to all of them, cf. #9914 [1].) > > > -Leif > > [1] http://trac.sagemath.org/sage_trac/ticket/9914#comment:5 ff. > > -- > To post to this group, send an email to sage-devel@googlegroups.com > To unsubscribe from this group, send an email to > sage-devel+unsubscr...@googlegroups.com > For more options, visit this group at > http://groups.google.com/group/sage-devel > URL: http://www.sagemath.org > -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
