Re: Monitoring IFS objects for deletion at V5R4

Good point. However if IBM wanted us to use it as a general function they
could have gotten over their fixation with 10,0 packed numbers and went to
a bigger field and make it a one field selection instead of having to
calculate the size. Who uses those fields individually anyway? I've only
seen them used to multiply each other. Granted, x*y is not that big of a
... breaker but it just seems silly to have to do it. Besides, if you
want to create an index by size descending it gets tricky.

Actually making one field would have taken less bytes of disk space. For
example, the maximum you could get by the above is
9999999999 * 9999999999 or 99,999,999,980,000,000,000. The maximum by the
following
Data Field Buffer Buffer
Field Type Length Length Position
ODSIZU PACKED 10 0 6 1
ODBPUN PACKED 10 0 6 7
NEWSIZE PACKED 21 0 11 13
is
999,999,999,999,999,999,999
or 10 times the above plus change. You could have even went out an extra
two digits and still just have taken the same amount of disk space (and
covered even larger object size).

Using odsizu * odbpun may be no big deal to you (and honestly there have
been numerous times that I've had to do that) but if it was easier it
would have caught on faster.


Rob Berendt