On 19 May 2010 01:12, Will Murnane <will.murn...@gmail.com> wrote: > This involves a lot of string manipulation. Consider using Perl's > Math::BigFloat [1], which may be substantially faster, and gives > arbitrary precision.
Math::BigFloat has a few bugs and does not handle negative infinite in a few cases. We already tried that on Solaris. Solaris perl has been compiled with double floating point resolution and not long double as a normal perl build. That hampers interoperability with Math::BigFloat and makes Solaris /usr/bin/perl useless for our project. We don't want to hack or own perl either, just try to get along with the system resources. Otherwise we have to support the extra build of perl for our customers, too. Nicholas -- Nicholas George MBDA System Integration (home office) _______________________________________________ shell-discuss mailing list shell-discuss@opensolaris.org http://mail.opensolaris.org/mailman/listinfo/shell-discuss
