Re: More SQL precompiler

John Carr wrote:

> "That's the disconnect - Rochester's assumption is that they don't support
> all of the high-level languages by default, so they are looking for
> customers to define requirements."

Huh??  Ignoring for a moment all of the bandwith that has been dedicated to
this topic on mailing lists, newsgroups, and web forums for several years
already, let's focus our attention on just how much IBM actually knows about
our requirements:

On June 7th, 2001, Kent Milligan (of the UDB/400 group) posted a
solicitation on an internet newsgroup for feedback regarding SQL Precompiler
requirements - an excerpt of which is provided below:

-------------------
VOTE NOW! HELP DECIDE NEW FEATURES FOR THE ILE RPG SQL PRECOMPILER
IBM is planning which features to add to the SQL/400 precompiler for
ILE RPG. The precompiler is part of the DB2 Query Manager and SQL
Development Kit licensed program product. The precompiler translates
RPG IV source code members that contain embedded SQL statements as the
first stage in creating a module or program object. If you use SQL
with ILE RPG, you have a major interest in what enhancements IBM
includes in future releases.

Here are the enhancements currently under consideration:

a. Full support for properly parsing RPG subprocedures
b. Support for using a local variable in a subprocedure as a host
variable in an SQL statement
c. Full support for the V5 qualified subfield names based on the new
RPG IV D-spec Qualified keyword
d. Conditional precompilation: /If, /Define, and other directives
e. Nested /Copy
f. Support for /Include
g. Support for host structure in Update statements
h. Support for a structure of host indicator variables
i. A precompiler directive to generate an array (or structure) of
indicator variables from a data structure

-----------------------

The full text can be read at:

http://groups.google.ca/groups?q=sql+precompiler+rpg&hl=en&ie=ISO-8859-1&oe=
ISO-8859-1&selm=3B1FF82B.2C1117DD%40us.ibm.com&rnum=1

The results of the survey were as follows (reprinted from
http://www.iseriesnetwork.com/as400surveys/SQLPrecompiler/):

Items were rated on a scale of 1 to 10 and are listed below from highest to
lowest average rating. There were 230 survey responses. The number of
respondents who rated each item is shown with the average rating.

  1.. (b) Support for using a local variable in a subprocedure as a host
variable in an SQL statement.
  Rated 7.65 by 227 respondents.

  2.. (g) Support for host structure in Update statements.
  Rated 7.38 by 227 respondents.

  3.. (a) Full support for properly parsing RPG subprocedures.
  Rated 7.37 by 223 respondents.

  4.. (c) Full support for the V5 qualified subfield names based on the new
RPG IV D-spec QUALIFIED keyword.
  Rated 7.09 by 222 respondents.

  5.. (d) Conditional precompilation: /If, /Define, and other directives.
  Rated 6.56 by 223 respondents.

  6.. (h) Support for a structure of host indicator variables.
  Rated 6.38 by 226 respondents.

  7.. (i) A precompiler directive to generate an array (or structure) of
indicator variables from a data structure.
  Rated 6.37 by 224 respondents.

  8.. (e) Nested /Copy.
  Rated 6.04 by 222 respondents.

  9.. (f) Support for /Include.
  Rated 5.92 by 223 respondents.

What exactly is still unclear about the customer requirements?

The problem isn't that they don't know what we need; it's that they are
unwilling, or unable, to provide it. From that perspective, discussing the
issue with their developers or the UDB architect is pointless, because in
spite of their good intentions, they are powerless to do anything about it.
We need to reach the decision maker(s).

Sadly, I can't do this effectively, and neither can 99% of the people who
post to this list. We simply don't have the clout. This is a job better
suited to the journalists at the trade mags. They should be on top of issues
like this that are a concern to their customers! Find the decision makers,
interview them, and let them go on record with their responses to these
problems. (Jill, are you listening?)

So the bottom line is that "official channels" and DCR's amount to nothing
more than stonewalling. They know very well what we need, and asking us to
repeat it for the upteenth time is simply wasting your time and mine.


John Taylor