8

The LA Fox Developer Newsletter
June 1994
Books and Toys....
Saved by the Junction
by Barry R. Lee

[A continuing series featuring add-on products and
books supporting FoxPro.]

Recently, I was retained by a medical clinic in
Anaheim to convert a DataEase application that
they had been running for the last six years to a
FoxPro app. This app is still under development
with a promised delivery date in late August.
Having had this system for so long, the primary
objective was to develop a very efficient system
and still maintain the years of historical data the
clinic had accumulated.

Data Ease, for those not familiar with it, is a sort of
“screen painter’ that will generate applications and
input screens once the screens/input fields are
defined. It reminded me of a sophisticated version
of the old application generator in dBase III+ One
of the confusing aspects of using it, though, is that
DataEase assigns its own naming convention to
data tables, screens, indexes, etc. In other words,
if you paint a screen, name it “Patient Input”,
DataEase will assign some non-meaningful name
to the actual table (such as “PATINAAA.DBM”),
another to any indexes (such as “PATINAAF.IOA”,
“PATINAAD. 101”, “PATI NAAB. 125”, etc.), and still
another to the input form (“PATINAAB.DBA”), with
no apparent pattern to the naming conventions. Its
only saving grace is that the system will document
itself, in terms of tying its generated names to the
actual data tables, ff you can afford to take the
hours and hours it takes to wade through the
documentation to arrive at how to do it. (In this
case, I didn’t have a choice.)

To further complicate matters, it doesn’t seem to
export to any standard file formats that FoxPro can
read directly (even though it says it does). The
exported data always seemed to be “one byte off’.
(And trying to read the files with XTree only locked
up the system.) I finally found a time-consuming
way to do it, but the thought of going through this
process for about 60 tables wasn’t too thrilling.
The results weren’t always consistant and
DataEase International certainly wasn’t going to be
much help.
So I started looking around for data conversion tools.
About two months ago, I remembered receiving a
mailing from Tools & Techniques about a product
named Data Junction. And I also remembered
receiving a free data conversion tool (which I hadn’t
used) from Pinnacle Publishing when I subscribed to
FoxTalk, Pinnacle’s monthly FoxPro newsletter. Imag-
ine my surprise when I found out the Pinnacle disk
was actually a “sampler’ of some of the conversion
routines from Data Junction. To make circumstances
even more fortunate, there was a routine for convert-
ing Data Ease data tables.

After a very short (less than one hour) learning curve, I
was able to begin conversion of the troublesome
DataEase tables. Data Junction was able to go into
the DataEase tables find and “interpret back” the
original screen names associated with the tables. This
proved to be a real life-saver. What could have been
days spent in using my kluged-together solution was
reduced to a few hours. The data came across 100%
with no exceptions. The only adjustment that had to
be made was to rename the fields in FoxPro since
DataEase allows up to 20 characters for field-naming.

Data Junction claims to have an installed base of
over 30,000 users, and is also available for Unix and
0S2. It features a “hub-and-spoke” design that lets
you filter, edit, and convert many different types of
data files. It offers five basic levels and pricings for its
product. As an example, the Standard Edition ($99)
features conversion routines for dBase, FoxPro,
ASCII, Lotus, DIF, and many others. The Professional
Edition ($199) will convert all the formats in the Stan-
dard Edition plus R:Base, Paradox, DataFlex, Clarion,
DataEase, ACT!, Peachtree, and more. The Ad-
vanced Edition sells for $299, while the LAN and Unix
Editions are priced at $499.

So, if you received the same mailing I did, dig it out of
the trash or go find it. This is one tool that can save
your b--- and your client’s data.

But wait, there’s more


(Con’t, Page 9)
Page 8

8