User vs Instrinsic Functions (was: String Functions in MF

"Chuck Stevens" <charles.stevens@unisys.com> wrote in message
news:cjf9g5$dnr$1@si05.rsvl.unisys.com...
> ... which is precisely why I suggested this methodology be in a user-defin
ed
> Function whose arguments and behavior are well-described ...  ;-)
>
> Someone could also propose a standard Function TRIM, but in similar
> circumstances J4 has been reluctant to specify a *standard* function where
 a
> *user-defined* one would serve as well.  The only cases I can think of whe
re
> functions of such straightforward logic as this have been proposed have be
en
> those in which the standardization of the output format was deemed desirab
le
> (as for example, for the proposed COMBINED-DATETIME function as part of th
e
> ISO Date/Time proposal).  I don't want to get in the way of a Wily
> Entrepreneur marketing a library of such functions to the COBOL user
> community, and I think Function TRIM (however named) would fall into the
> category of potentially-marketable functions.
>
>    -Chuck Stevens
>

Point of clarification (possibly just personal opinion),

The distinction that Chuck is making is a POST-2002 Standard distinction.
Although the 2002 Standard introduced "user-defined functions", the followin
g
INTRINSIC funcitons were introduced at the same time.  These are just the on
es
that I think would have been easy to create "user-defined" functions for - b
ut
instead intrinsic ones were provided:

DATE-TO-YYYYMMDD
DAY-TO-YYYYDDD
E
EXP
EXP10
FRACTION-PART
NUMVAL-F
PI
SIGN
YEAR-TO-YYYY


--
Bill Klein
wmklein <at> ix.netcom.com

Yes, I agree, and presuming the *prior* presence of user-defined intrinsic
functions I'd personally have been uncomfortable with adding some (though
not necessarily all) of these.

-Chuck Stevens

"William M. Klein" <wmklein@nospam.netcom.com> wrote in message
news:YbG6d.2339590$6p.396746@news.easynews.com...
> "Chuck Stevens" <charles.stevens@unisys.com> wrote in message
> news:cjf9g5$dnr$1@si05.rsvl.unisys.com... 
user-defined 
where a 
where 
been 
desirable 
the 
>
> Point of clarification (possibly just personal opinion),
>
> The distinction that Chuck is making is a POST-2002 Standard distinction.
> Although the 2002 Standard introduced "user-defined functions", the
following
> INTRINSIC funcitons were introduced at the same time.  These are just the
ones
> that I think would have been easy to create "user-defined" functions for -
but
> instead intrinsic ones were provided:
>
> DATE-TO-YYYYMMDD
> DAY-TO-YYYYDDD
> E
> EXP
> EXP10
> FRACTION-PART
> NUMVAL-F
> PI
> SIGN
> YEAR-TO-YYYY
>
>
> --
> Bill Klein
>  wmklein <at> ix.netcom.com
>
>