I've used j to parse web sites. The program runs daily in batch
somewhere on the Amazon Web Service virtual computer.
Python seamlessly converts machine integers to an "infinite" size
representation. Which is nice. There must be a performance penalty for
this, and super size integers are primarily useful for puzzles such as
those in project Euler. Long integers can also be used for fixed point
scientific programs where the orders of magnitude have a large span.
The upshot is that I wish I knew the rules j uses to preserve extended
precision. Is there a pattern for
x dyad y
preserves extended precision if exactly one of x or y is extended?
On 10/11/2017 08:00 AM, [email protected] wrote:
I would be genuinely interested in reading about real world appllications of J.
Regards, Rob Burns.
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
