On 05 Jun 2013 12:34, Needles,Stephen J wrote:
The goal is to convert existing DDS-defined DB objects to DDL.
I can pretend to infer there are specific reasons for effecting such
a conversion for each chosen file, even while knowing how unlikely that
is [esp. with an implication of "all"]. Yet, I must ask rhetorically...
Why do so many choose to waste so much time and effort when there are
certainly more productive activities; e.g. code enhancements :-(
Hopefully asked often enough, some will stop and ask themselves "Why?"
to think upon their choice, before possibly effecting nothing more than
a change for the sake of change. Amazing in contrast, how many would
allow code to remain unchanged for decades because there is nothing
inherently wrong with leaving the source and programs unchanged, yet for
some reason [even those same] people are motivated to make the change
from DDS to SQL with no logic supporting their actions.
And I must remind... Always make the change to SQL for programs
*before* changing from DDS to DDL, to best avoid problems; for one, the
mindset of ordered data outside of the ORDER BY in the SELECT query
should already have faded. Irrespective of implications by some [snake
oil salesmen], the transition to DDL from DDS has *nothing* to do with
/modernization/ of applications, except when an unsupported data type or
feature of the DDL is required for the design which probably has already
changed to use the SQL as part of its transition to no longer depend on
the tight coupling of program to the file layout in the apparently
/legacy/ RLA ways.