|
Most questions can be answered by reading the
Documentation.
How do I best install your product
Since ATG is written purely in Java you can install it on various
Operation Systems. You can always run as many Gateways on each systems as
you want, just ensure that every Gateway has its own directory. This even
works on Windows as the Java applications are not entered in the system
registry. If you want to run
Accent Transformation Gateway under Windows you can use
AtYourService (http://www.prismmicrosys.com/ atyourservice/atyourservice-index.htm) to run
it as a service that will automatically start with the server. This
option will also allow you to catch the error if an error occurs, isolate
the wrong file and restart the service. This has been used by many of our
customers to ensure a flexible SLA approach.
If your question however relates on the subject how to manipulate and
transform an EDIFACT message with our tool we would like to provide you with
an example.
Let’s take an example with the two invoices EDI (EDIFACT) below.
UNA:+.7'
UNB+UNOA:1+126123+126121+930430:1409+REF800'
BGM+380+1300+930315'
NAD+BY+126321:91++Hacker PLC+4 JONES WAY+SPRINGFIELD+80000'
NAD+SE+126123:92++ABC PLC+4 ADAMS WAY + SPRINGFIELD+80000'
UNS+D'
LIN++5050+10:21:PC+1500:CA:1+10+15000'
LIN++4750+5:21:PC+500:CA:1+5+2500'
UNS+S'
TMA+17500'
UNT+8+INV001'
UNZ+1+REF800'
UNA:+.7'
UNB+UNOA:1+126123+126121+930430:1409+REF800'
BGM+380+1301+930315'
NAD+BY+126321:91++XYZ PLC+16 PORTLAND AVE+SPRINGFIELD+80000'
NAD+SE+126123:92++ABC PLC+4 ADAMS WAY + SPRINGFIELD+80000'
UNS+D'
LIN++5050+10:21:PC+1500:CA:1+10+15000'
LIN++1100+5:21:PC+500:CA:1+5+2500'
UNS+S'
TMA+17500'
UNT+8+INV001'
UNZ+1+REF800'
_____________________________________________________________________
Here are the corresponding invoices as an illustration for those not
familiar with this format .
ABC PLC
4 ADAMS WAY
80000 SPRINGFIELD
Hacker PLC
4 JONES WAY
80000
SPRINGFIELD
INVOICE No. 1300
No ITEM AMOUNT PRICE TOTAL
5050 PC Pentium Xeon 10 1500.-
15000.-
4750 LCD Monitor 5 500.-
2500.-
17500.-
ABC PLC
4 ADAMS WAY
80000 SPRINGFIELD
XYZ PLC
16 PORTLAND AVE
80000
SPRINGFIELD
INVOICE No. 1301
No ITEM AMOUNT PRICE TOTAL
5050 PC Pentium Xeon 10 1500.- 15000.-
1100 Transformation Gateway 5 500.- 2500.-
17500.-
Lets say these invoices are part of a batch run of several thousand
invoices per day and we want to upload only the receiver of the goods, the
invoice number, the item No., the amount and the total cost via a SQL loader
that requires a CSV file format with following format (any other format is
possible as well:
Hacker PLC,4 JONES WAY,80000,SPRINGFIELD,1300,5050,10,4750,5,17500
XYZ PLC,16 PORTLAND AVE,80000,SPRINGFIELD,1301,5050,10,1100,5,17500
and so on.
To archive this transformation you will first need to start the Accent
Transformation Gateway and enter all the formatting information’s about
the incoming and about the outgoing file. This is done via the very easy GUI
that requires no programming efforts at all. Once this is done you will then
be able to start the Gateway as such in the background that will then
automatically transform any incoming file (with as many data sets as you
like) in the desired outcome.
In the following we guide you through the GUI using the example above.
To start the GUI go on the Command Line to the directory where the ATG was
installed. Then enter the following:
java atgfactory13/ProviderATG
After having entered your Product Licence Key in the Welcome Screen and
accepted the Licence Agreement with Accent Communications Ltd. the Project
Information screen requires Project information.
Let’s assume the ATG was installed on a WINDOWS platform on path c:\AccentETL,
a project name InvoiceData, a directory location of the incoming file c:\AccentETL\InputFolder,
an incoming file EDIInvoiceData.txt, a directory location for the outgoing
file c:\AccentETL\OutputFolder and an outgoing file with the name
ExtractInvoice.txt.
According to this information the “Project Information” screen needs to look
as
follows:
Please be aware that the directories have to be entered with the
directory/tree break according to your Operating System!!!
Given the above EDI example the ADVANCED tool has to be used since the
datasets in the incoming file are multilined and the data is seperated by
delimiters from each other. Delimiters in this example are “:”, “+” and the
“’” for the indication of the line end.
This example was choosen to give an illustration how the GUI can be set up
with a little more complicated data. The “easier” examples are to be found
in the general documentation.
On the following screen, the “Data Description – Advanced” screen, there
is the complete set up for the above EDI input sample and the coma delimited
output.
The descriptors choosen are sample descriptors and can of course be
different as long as they comply with the following rules:
can contain letters, numbers (thus the inclusion of &<”’^|~\... is
not allowed)must not start with a number or consist of a number onlycannot
contain spacesmust not start with the letters xml (or XML or Xml)
In order to make it easy there are descriptor names choosen like
Line_1_Entry_1, Line_1_Entry_2 because it is not always clear to the person
working with the data what each individual data stands for.
Since the screen does not show the whole set up at a glance in the
following are all the descriptors and delimiters listed for both the
incoming file and the result file.
INCOMING FILE
DESCRIPTOR DELIMITER
LINE_1_ENTRY_1 :
LINE_1_ENTRY_2 +
LINE_1_ENTRY_3 ‘
LINE_1_ENTRY_4 cr
LINE_2_ENTRY_1 +
LINE_2_ENTRY_2 :
LINE_2_ENTRY_3 +
LINE_2_ENTRY_4 +
LINE_2_ENTRY_5 +
LINE_2_ENTRY_6 :
LINE_2_ENTRY_7 +
LINE_2_ENTRY_8 ‘
LINE_2_ENTRY_9 cr
LINE_3_ENTRY_1 +
LINE_3_ENTRY_2 +
LINE_3_ENTRY_3 +
LINE_3_ENTRY_4 ‘
LINE_3_ENTRY_5 cr
LINE_4_ENTRY_1 +
LINE_4_ENTRY_2 +
LINE_4_ENTRY_3 :
LINE_4_ENTRY_4 +
LINE_4_ENTRY_5 +
LINE_4_ENTRY_6 +
LINE_4_ENTRY_7 +
LINE_4_ENTRY_8 +
LINE_4_ENTRY_9 ‘
LINE_4_ENTRY_10 cr
LINE_5_ENTRY_1 +
LINE_5_ENTRY_2 +
LINE_5_ENTRY_3 :
LINE_5_ENTRY_4 +
LINE_5_ENTRY_5 +
LINE_5_ENTRY_6 +
LINE_5_ENTRY_7 +
LINE_5_ENTRY_8 +
LINE_5_ENTRY_9 ‘
LINE_5_ENTRY_10 cr
LINE_6_ENTRY_1 +
LINE_6_ENTRY_2 ‘
LINE_6_ENTRY_3 cr
LINE_7_ENTRY_1 +
LINE_7_ENTRY_2 +
LINE_7_ENTRY_3 +
LINE_7_ENTRY_4 :
LINE_7_ENTRY_5 :
LINE_7_ENTRY_6 +
LINE_7_ENTRY_7 :
LINE_7_ENTRY_8 :
LINE_7_ENTRY_9 +
LINE_7_ENTRY_10 +
LINE_7_ENTRY_11 ‘
LINE_7_ENTRY_12 cr
LINE_8_ENTRY_1 +
LINE_8_ENTRY_2 +
LINE_8_ENTRY_3 +
LINE_8_ENTRY_4 :
LINE_8_ENTRY_5 :
LINE_8_ENTRY_6 +
LINE_8_ENTRY_7 :
LINE_8_ENTRY_8 :
LINE_8_ENTRY_9 +
LINE_8_ENTRY_10 +
LINE_8_ENTRY_11 ‘
LINE_8_ENTRY_12 cr
LINE_9_ENTRY_1 +
LINE_9_ENTRY_2 ‘
LINE_9_ENTRY_3 cr
LINE_10_ENTRY_1 +
LINE_10_ENTRY_2 ‘
LINE_10_ENTRY_3 cr
LINE_11_ENTRY_1 +
LINE_11_ENTRY_2 +
LINE_11_ENTRY_3 ‘
LINE_11_ENTRY_4 cr
LINE_12_ENTRY_1 +
LINE_12_ENTRY_2 +
LINE_12_ENTRY_3 ‘
LINE_12_ENTRY_4
Now, why is there apparently always one entry too many in each line? This
has to do with this special EDI format which uses a line break delimiter
“’”.
This has to be treated as a normal delimiter which results in the fact that
there is still an entry assumed after that delimiter, therefore each line
needs to have one entry too many to construct this case.
The reason for this is that the GUI is set up for the majority of formats
that do not have a line end delimiter.
Please see the documentation for examples where this additional
(fictitious) entry per line is not necessary.
Notice: The last Descriptor of a data set has no Delimiter.
An empty line between every data set will be assumed when a file has
multilined data sets, which the GUI recognices in case the “cr” as a
delimiter to indicate line breaks was used, indicating that it is multilined.
If it is a single lined data set the ATG assumes no empty line between the
data sets.
RESULT DATA
DESCRIPTOR DELIMITER
LINE_4_ENTRY_6 ,
LINE_4_ENTRY_7 ,
LINE_4_ENTRY_9 ,
LINE_4_ENTRY_8 ,
LINE_3_ENTRY_3 ,
LINE_7_ENTRY_3 ,
LINE_7_ENTRY_10 ,
LINE_8_ENTRY_3 ,
LINE_8_ENTRY_10 ,
LINE_10_ENTRY_2
After the Save and then the Save All Button is now clicked the
“Congratulations” Screen appears on which a click on the “OK” button closes
the GUI application.
Now everything is ready for the ATG to do its work. To start the Gateway go
on the Command Line to the directory where the ATG was installed and enter
the following (according to the sample project data entered in the GUI as
per Figure1):
java –classpath c:\AccentETL AccentETL InvoiceData c:\AccentETL\
Note: This line needs to be entered in one line!
This line looks confusing since in this example the ATG was installed in the
directory c:\AccentETL and the ATG’s name is AccentETL as well. For other
examples on how to start the Gateway please see the product documentation.
The general command of the above line is as follows:
java –classpath [directory location of ATG] AccentETL [Project Name]
[directory location of ATG]
WARNING: Please enter the directory according to your OS. Also make sure
that the directory location of ATG ends with the tree/directory break, for
Windows e. g. c:\Install\Accent\ (NOT for the classpath, see above).
Please note, if you have a directory consisting of more than one word
like “Program Files” then you have to enter the directory location of ATG
included in marks (see second example above). Note as well, in this case the
second time you enter the directory location of ATG after the Project Name,
after the tree/directory break an empty space is necessary or you have to
enter twice the backslash followed by the marks (see third example above).
If the directory is not entered correctly as described, the program will not
work! The Project Name has to be the same than the Project Name which was
entered in the GUI part of the ATG!
After completion of the installation process please insert the path
variable of the Java Virtual Machine. For Windows access the Control Panel,
choose System, Advanced, click on Environment Variables, choose Path from
the System variables and insert e. g. c:\Program Files\accent\jre\bin.
If you should have any questions please contact your local administrator.
REMARK: Please note, when starting the Gateway, the first time a file is
read in the process takes longer than the subsequent times.
To view the .xml files, the .dtd files as well as the .xsl file please look
in the directory where the ATG was installed.
If you however have more questions please send them to
support@accent-ltd.com
and indicate if we can use your question for a FAQ.
To be even more proactive we would like to invite anyone who needs a
transformation done to send us the structure of the incomming and the
outgoing data with one example, so that we can show on this page how easy
the transaction can be done.
|
