Tom,
Rather than go through all that work to build temp tables, why not just
create a UNION SELECT view with aliases to join the like data for your
scrolling region?
Example:
CREATE VIEW masterview (columnA, columnB, columnC) AS +
SELECT firstcolumn, secondcolumn, thirdcolumn +
FROM firsttable +
WHERE condition +
UNION ALL +
SELECT fourthcolumn, fifthcolumn, sixthcolumn +
FROM secondtable +
WHERE condition2 +
UNION ALL +
SELECT seventhcolumn, eighthcolumn, ninethcolumn +
FROM thirdtable +
WHERE condition3
This only requires that the columns being matched have the same data type.
Your scrolling region then references the aliases specified in the CREATE
VIEW.
Emmitt Dove
Manager, DairyPak Business Systems
Evergreen Packaging, Inc.
[EMAIL PROTECTED]
[EMAIL PROTECTED]
(203) 643-8022
From: [email protected] [mailto:[EMAIL PROTECTED] On Behalf Of Tom
Frederick
Sent: Wednesday, February 06, 2008 7:53 PM
To: RBASE-L Mailing List
Subject: [RBASE-L] - Appending unrelated data and scrolling edits
I have received a system request that I like, but I am not sure what
problems I will run into. I have separate tables for unrelated Events like
behavior, seizures, or Occ Med injury. Right now, a table may have up to 120
columns of data with virtually no shared columns and has its own form. Each
table has an add/edit/browse menu that selects a person's records, scrolls
by date to allow the user to pick a specific event to review with the right
form. Works very well. The suggestion has come to blend these unrelated,
separate tables/events into one summarized format, sort by date, and allow
the user to select an event with the correct form to review the real data.
Each table has an integer column called "ZID" which ties back to an autonum
integer "ZID" column in the identification table. All other column names are
specific for each table such as "BehavDate", "OMDate", "SeizDate", etc, but
if "ZID" works why wouldn't other shared names also work? Each line in each
table also gets an autonum integer ID specific to its table. If I had common
columns in each table ( say "ZID", "Date", "Time", "LineID", "EventName",
and "BriefDescription", each with the appropriate type/length), I should be
able to:
1. Project these columns based on "ZID" and a Date range to a temp
table,
2. Append those temp tables into a master temp table because the column
name/types would match (and not have to change column names from different
tables),
3. Use ORDER to scroll the master temp table records by "Date" and
4. Then by selecting a scroll line, use EDIT USING in a bit button EEP,
with WHERE commands for "EventName" and "LineID", to go to the specific row
in a specific table using the required form.
This type of system would avoid going to several menus and visually show
behavior trend summaries that would not be obvious by using multiple menus.
It sounds feasible and seems clearer the more I think about it. Still, for
me, it is a major jump on Razzak's "Keep Learning" mantra. What problems
will I run into or is there just a much simpler method to accomplish this
task? I have looked at the sales transaction examples in SAT and the sample
databases. It seems that as long as columns like "Date" or "LineID" do not
require distinct values, I should be able to append a series of columns with
other identically named columns from other tables as long as the types/names
match and, through scrolling, let the user pick exactly what they want.
Sounds like a simple process. Feels like a John Wayne movie where the scene
is "too quiet" before the guns start blazing.
Tom Frederick
Elm City Center
1314 W Walnut
Jacksonville, IL 62650
Off - 217-245-9504
Fax - 217-245-2350
Email - [EMAIL PROTECTED]
Web - www.elmcity.org
