On Tue, 26 Nov 1996 17:38:07, "Kishore Kumar.M" <[log in to unmask]> writes:

> *  Would the proposed  intrinsics meet your  requirements  to a large
>    extent?

Yes, I think they are a welcome addition to the Intrinsic library.

> *  While  converting  the dates from one format to another,  when the
>    destination  format  supports a four  digit/four  field year, what
>    should be the century portion of the converted date?  Should it be
>    zeros or should it use the proposed HPYYYY CI variable?

Please do *NOT* leave the century portion of the converted date zeroes unless
the century cannot be determined with any reliability (for instance, YYMMDD
converted to YYYYMMDD).  Also, I don't think the HPYYYY CI variable will be
any help here (see Adager's method of handling dates using a "cutoff year"
to determine if the the year is 19xx or 20xx).

> *  Is an  intrinsic  routine,  HPDATECOMPARE,  to  compare  two dates
>    required?  Such an  intrinsic  compares  two  dates  in one of the
>    standard   formats   described   above.  It  would  report   their
>    relationship (less than, equal, or greater).

Not really needed.  Once the dates are converted to a sort-friendly format,
the comparison is easily done.

> *  How useful would it be to have an  intrinsic to check the validity
>    of a given date with respect to one of the Standard Date Formats?

This is a much-needed intrinsic.  I vote for it.

As always, just my $.02 worth.

Jim Phillips                            Manager of Information Systems
E-Mail: [log in to unmask]      Therm-O-Link, Inc.
Phone: (330) 527-2124                   P. O. Box 285
  Fax: (330) 527-2123                   Garrettsville, Ohio  44231