I've been swamped after IPROF and before leaving town
for four days.  Last week Frank Nagy said:

> I miss the following features from QUERY

> 1) A completion bar or a message appearing at every
> thousandth DBGET during the [multi]find commands.

Someone was already good enough to point out that this
was added to and appeared on the current 1998 SIGIMAGE
Ballot after Frank made this suggestion last year:  It is item
# 34.2:  FIND, etc:  Running count progress without CTRL-Y.
Ballot results next week after polls close..       :-)

> 2) An extension of the #LIMIT=nnn option to select not only
> the firsn nnn matches but to ignore the first mmm matches
> then to select the next nnn matches.    Example:
> FIND #IGNORE=mmm;LIMIT=nnn;FIELD = value.

>                                                       Frank
> [log in to unmask]  (36-1)-395-9123 Fax: ~395-9002

Sounds like a good idea to me.....  but in adding more
functions to QUERY, it leads me to ask the readership a
somewhat more general question:

How far down the road of adding complex selection and
reporting features to QUERY is it reasonable to try and
get HP to go ??

At our site we use QUERY mostly as an ad-hoc IMAGE
maintenance tool, that we use when possible to avoid
having to write short Transact programs to manipulate
data.  But since short Transact programs are very quick
and easy to write, we don't expect QUERY to be able to
do everything.

I realize that many sites use QUERY in much more of a
production reporting capacity than we do, and not all sites
have Transact ( <semi-smug alert> ).  But from the report
generation side there are other much more capable host-
based report generators;  i.e.:  Supertool; QueryCalc; etc.;
plus all the low-cost PC-based ODBC tools that can be
used against Image/SQL via the selection of ODBC drivers.
(of course these third-party tools are not free).

Anyway, at some point I would guess that HP would tend
to give the same answer to major functionality enhancements
to QUERY that they have given to the "bundled ODBC access
direct to TurboIMAGE" request:  These features are already
available from reputable third-party vendors (note that I have
not checked with HP at all on this one; at least not yet).

Having said all of the above, speaking just for myself I think
Frank's above suggestion fits in pretty well with the six
items that are already on the SIGIMAGE Enhancement
ballot for QUERY;  most have been estimated to be a Small
or Medium effort to implement....  except for FIND FIELDA
= FIELDB, which is classed as a large effort (and is also
by far the most popular QUERY enhancement request)....

Ken Sletten
SIGIMAGE Chair