I don’t think the bug I posted (which now appears to be sporadic, btw)
is necessarily related to the bugs reported in ruby-gsl or fxruby.
It could be a bug in string.c. Or, it could be a compiler bug. Or, it
could be some issue with regards to Windows IO stream, since a ‘puts’
will sometimes mysteriously solve (or cause) a segfault.
Yes, that could also be. I’m just struck by the fact that sometimes
identical code causes problems when it is run on different machines,
and the Windows implementation tended to cause significantly more
problems in my personal experience…