On Tue, 16 Dec , JEFF VANCE wrote:

As you know we are implementing the "CI Functions" SIB request... it is
better from an eliminating regression failures point of view if we do NOT
change the evaluator parsing rules... that would preclude a CI function
from being qualified... would the restriction of DISallowing qualified user
function names be a problem for you...

I've seen the discussions of 'CI Functions' but I haven't focused much
attention on them.  It seems that a CI Function is essentially a command
file with optional parameters that returns a value compatible with a CI
variable.  In keeping with the general elegance of MPE, it would be nice to
offer the greatest flexiblity in specifying the function file but the
evaluator's parsing rules preclude fully qualified file names.  Perhaps one
of the following thoughts might work:

File Equations - I'm sure that back referenced file equations would
conflict with the multiply operator, but it might be possible to call my
user defined function of X by coding:
  :file FofX=myfunc/lockword.group.account
  :setvar a FofX(x)

A new built-in CI function - CALL( "<user CI function>" )
  :setvar a CALL("/account/group/myfunc(x)")

Just my thoughts,

Tom Root

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *