On Tue, Feb 27, 2001 at 04:14:52PM -0000, [EMAIL PROTECTED] wrote:
> >This distribution has been tested as part of the cpan-testers
> effort to test as many new uploads to CPAN as possible. See
> http://testers.cpan.org/
>
> Please cc any replies to [EMAIL PROTECTED] to keep other
> test volunteers informed and to prevent any duplicate effort.
Thanks. Here are the funny details: The test engine's code to test as
much as possible of the plotting engine even if no output device was
available (no Term::Gnuplot) was confused: integers and floats mean
very different stuff to PARI.
Essentially, I was using the following code:
$w = eval {output_width()} || 1000;
$h = ...
create_rectangle($w - 11, $h - 11);
If Term::Gnuplot is present, then output_width() return a PARI
integer, so the overloaded $w - 11 creates a PARI integer, and
create_rectangle() happily sees what it expects: integer as an
argument. However, if $w is *just* 1000, then $w - 11 is in fact
Perl's 989.0000, which is converted to PARI's floating point type, and
so create_rectangle() fails.
On the first sight, the new (and, in my opinion, busted) logic of
not-propagating to float in 5.7.-current will circumvent this
particular problem. However, since integers and floats are treated
differently by PARI, hiding such problems (by postponing the
conversions until unspecified time in the future) may create more
problems than what is solved.
Shortly: fixed in my copy.
Thanks,
Ilya
