Yes, I was able to remove the need for custom XSLT processing by being a
bit smarter about the XML output plug-in so that it has enough
flexibility to produce a file that the entity-engine should be able to read.
One can configure the root element name and the data element name so one
produce the XML required when the data is output from the datastore's
tabular structure.
This means that data importing/conversion can be done without knowing
any Java or XSLT.
The downside is that it produces a file for each entity type and they
have to be imported in the right sequence.
However each file is easy to check visually since it will consist of
record of a single entity type and the process can spit out tablular
reports of the attributes as columns in parallel with the xml. This will
help in conversion projects where thousands of customers, supplier or
products have to be added and verified.
That should help make it easier for non-programmers to generate test
data and demo data.
One still has to know the entity definitions and the relationships
between entities but not XSLT or Java.
Is my interpretation of the Party model as it applies to Customers accurate?
Ruth gave me some valuable hints and wrote a short document that was
very helpful and I found some recent articles and a book extract by Len
Silverston that were very instructive.
I still had to make some assumptions about the business case (Does the
use case allow for multiple contacts at the same customer? I originally
thought "no" but changed it so that it will and assigned each of the
initial contacts as "Buyers".
Probably could assign more roles but that is easy to do if required.
Ron
On 28/01/2015 1:11 PM, Jacques Le Roux wrote:
I just read the previous version. So I will have to read the update.
The XSLT part is missing though, normal?
Jacques
Le 28/01/2015 18:01, Ron Wheeler a écrit :
http://adtransform.artifact-software.com/?p=16 is a writeup about how
to use ADTransform to create the entity-engine-xml files to load
customers.
I would appreciate any comments from strategy to clarity of the
explanation.
I hope that this will help people with initial data loading by
allowing implementation teams to start with denormalized data
exported from other automated or manual systems and have it
automatically validated and turned into the correct structure for the
entity-engine.
It should also make it easier to create more extensive demo data and
test data without any programming skills.
Ron
--
Ron Wheeler
President
Artifact Software Inc
email: rwhee...@artifact-software.com
skype: ronaldmwheeler
phone: 866-970-2435, ext 102
