Thanks for the fast response.  I updated the l3kernel bundle and all is now well.  Unfortunately, I'm not doing anything interesting with it: testing for division by zero in a context that shouldn't give an error.  It's quite nice to have the option to do so though!

On Mon, Aug 27, 2012 at 2:35 PM, Bruno Le Floch <[log in to unmask]> wrote:
Hello Scott,

I had mistakenly defined the internal \__fp_trap:nn and documented \fp_trap:nn.

This should be fixed in a more recent version of the l3kernel bundle
(and I updated the "added" date).  Can you update to a version later
than 2012/07/19?

I'm very interested to have feedback on that function: this is very
much experimental, and I am not yet sure how much I should implement
of IEEE's trap mechanism.

Sorry for the mistake.


On 8/27/12, scott heard <[log in to unmask]> wrote:
> Hello, I'm getting an undefined control sequence when trying to use this in
> a
> minimal document as "\fp_trap:nn {invalid_operation}{none}".
> The documentation I have says that it was added on 2012-05-28 and the
> version of
> expl3 that I have is v3990, dated 2012/07/16.  Is this an issue on my end?
> scott