Hi, some more updates: * I fixed the sympow build on Solaris/x86 and now I get all those doctests to pass * The Singular pexpect interface hang now has a workaround, that does impact performance to some extend, but it is better than hangs. The fix can be made Solaris specific until we sort things out * A whole bunch of small doctest failures got fixed. * I tracked down one issue that is caused by pow() behaving differently than on Solaris when computing pow of negative values.
But there are at least two more open issues: * symmetrica seems to have some interface issues. I managed to do some of the examples that fail from Sage via symmetrica in C and those seem to work. More work is being done here * numpy/scipy: both of them (for both trunk as well as what we shit) segfault for certain input and this is probably due to some bug in between numpy/scipy, gcc/gfortran 4.2.4 and ATLAS. Not sure what to do here yet, but at least for numpy building without an external BLAS fixes the issues. Scipy seems to be mostly affected by specfun, i.e. its special functions Fortran library. Most of the above fixes as well as some other updates should make it into the next Sage release when I will do Sparc as well as x86 binaries. Cheers, Michael --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
