Helping in Designing Scenario

Hello people!
       i need some helping in designing a solution, i'm kinda new in the PI world, and i never made BPM stuff and so...
      Here is my situation, i have 2 Webservices that i have to consume in PI
     R/3 will send information to PI
      in PI i have to Consume the first Webservice with some information and that will Response a Token,
then i have to Merge this Token with another XML message and send to another Webservice, the i get the Response Message of this webservice and send back to R/3
    whats the best way of doing this? is it using BPM?
this is the two Webservices i will have to use:
http://siatepqa.suseso.cl:8888/Siatep/WSToken?wsdl
http://siatepqa.suseso.cl:8888/Siatep/WSIngreso?wsdl
Can anyone help me?
Thanks!
Edit:
i've already read alot of BPM topics and i think this is the way... but i'm not sure
i want to use abap_proxy conection in r/3 > PI and the webservice connections will use SOAP/WS
Edited by: pitoshi on Jun 9, 2010 10:57 PM

Liang Ji,
thanks very mutch for your answer, this is very helpfull,
But in:
"R/3 send message to PI via Outbound proxy (using XI Adapter) ---> PI receive Message and execute Mapping program --> during the mapping program, you can have SOAP lookup to call first web service to get token, then you can use lookup value in your mapping, ---> Use SOAP receiver adapter to call first web service, then PI get response from it, then pass to R/3. "
You mean " ---> Use SOAP receiver adapter to call first web service, then PI get response from it, then pass to R/3. ""
call the Second Webservice?
in spite of my difficulties with BPM, is this solution is better than using BPM?
Edit: * and if any of these processes do not respond? how I treat it? *
anyone else have any other option or tips to this scenario?
Edited by: pitoshi on Jun 10, 2010 12:34 AM

Similar Messages

Need help in design scenario

Hi experts,
I have a scenario in which XI put an xml file in a destination server (very simple scenario, SAP>Proxy>File receiver CC).
The new request is: to make a call to SAP after having put file in destination.
In this call I have to write to SAP that file has been written successfully to destination.
My collegues told me to add a RFC call to SAP in my scenario, is it the correct way?
I'm not skilled about that.
Thanks
Fabio Boni

Hi Fabio,
If you want to achive this scenarion without BPM then you need to create 2 diffrent scneraio,.
!) Your first scnario:
SAP -Proxy-File receiver( It will write the file in Folder A)
2) Second scenario:
File sender to RFC receiver
File sender will poll the folder A for files.If file is available in Folder A then this 2nd scenarion will trigger.
Create some constant mapping in the target RFC structure. and make a call to RFC using receiver RFC adapter.
So here If you pick the file from Folder A then you need to delete it or Archive it from the source folder.
Sender file adapter has both the options of delete or Archive.
If you archive the file to folder B in sender file adapter itself then your receiver RFC can find the File from Folder B.

Help in designing SQL

Hi All,
I need little help in designing SQL. My input and output data is given below:
Input
Key
Date_1
Date_2
Key1
2/14/2011 5:22:35 PM
2/22/2011 3:05:13 PM
Key1
2/24/2011 1:48:07 PM
5/24/2011 4:50:16 PM
Key1
2/24/2011 1:48:59 PM
4/25/2011 3:36:19 PM
Key1
5/24/2011 5:18:19 PM
6/16/2011 3:17:40 PM
Key1
6/16/2011 3:25:42 PM
6/24/2011 1:33:13 PM
Key1
6/17/2011 1:12:54 PM
6/25/2011 5:55:55 PM
Output
Key
Date_1
Date_2
Key1
2/14/2011 5:22:35 PM
2/22/2011 3:05:13 PM
Key1
2/24/2011 1:48:07 PM
5/24/2011 4:50:16 PM
Key1
5/24/2011 5:18:19 PM
6/16/2011 3:17:40 PM
Key1
6/16/2011 3:25:42 PM
6/25/2011 5:55:55 PM
Here out of 2nd and third record of input, output is 2nd record of output and 5th and 6th record of input would come out as 4th record of output. Basically, if we overlapping duration between Date_1 and Date_2, we need to take min of Date_1 and max of Date_2 for that overlapping records. In case of no overlapping, we would simply output input record like record 1 and 4 of input.
Out of
Key1
2/24/2011 1:48:07 PM
5/24/2011 4:50:16 PM
Key1
2/24/2011 1:48:59 PM
4/25/2011 3:36:19 PM
2/24/2011 1:48:07 PM was min of above 2 and 5/24/20114:50:16 was max of above 2.
Similarly in other overlapping scenario:
Key1
6/16/2011 3:25:42 PM
6/24/2011 1:33:13 PM
Key1
6/17/2011 1:12:54 PM
6/25/2011 5:55:55 PM
6/16/2011 3:25:42 PM was min of above 2 and 6/25/2011 5:55:55 PM was max of above 2.
Pls help.
Thanks.

Hi Frank.,
The above logic with NVL is not working.
My input sample is:
Key
Date_1
Date_2
Key1
2/14/2011 5:22:35 PM
2/22/2011 3:05:13 PM
Key1
2/24/2011 1:48:07 PM
5/24/2011 4:50:16 PM
Key1
2/24/2011 1:48:59 PM
Key1
5/24/2011 5:18:19 PM
6/16/2011 3:17:40 PM
Key1
6/16/2011 3:25:42 PM
6/24/2011 1:33:13 PM
Key1
6/17/2011 1:12:54 PM
6/25/2011 5:55:55 PM
I am giving create and insert statement below for creation of above data. Please see if you can help me:
CREATE TABLE test1
KEY1          NUMBER     ,
DATE_1               DATE ,
date_2              DATE
insert into test1
values(1, TO_TIMESTAMP ('2/14/2011 5:22:35 PM', 'MM/DD/YYYY HH:MI:SS AM'),TO_TIMESTAMP ('2/22/2011 3:05:13 PM', 'MM/DD/YYYY HH:MI:SS AM') ) ;
values(1, TO_TIMESTAMP ('2/24/2011 1:48:07 PM', 'MM/DD/YYYY HH:MI:SS AM'),TO_TIMESTAMP ('5/24/2011 4:50:16 PM', 'MM/DD/YYYY HH:MI:SS AM') ) ;
values(1, TO_TIMESTAMP ('2/24/2011 1:48:59 PM', 'MM/DD/YYYY HH:MI:SS AM'), '' ) ;
values(1, TO_TIMESTAMP ('5/24/2011 5:18:19 PM', 'MM/DD/YYYY HH:MI:SS AM'),TO_TIMESTAMP ('6/16/2011 3:17:40 PM', 'MM/DD/YYYY HH:MI:SS AM') ) ;
values(1, TO_TIMESTAMP ('6/16/2011 3:25:42 PM', 'MM/DD/YYYY HH:MI:SS AM'),TO_TIMESTAMP ('6/24/2011 1:33:13 PM', 'MM/DD/YYYY HH:MI:SS AM') ) ;
values(1, TO_TIMESTAMP ('6/17/2011 1:12:54 PM', 'MM/DD/YYYY HH:MI:SS AM'),TO_TIMESTAMP ('6/25/2011 5:55:55 PM', 'MM/DD/YYYY HH:MI:SS AM') ) ;
commit;
Expected output would be:
Output
Key
Date_1
Date_2
Key1
2/14/2011 5:22:35 PM
2/22/2011 3:05:13 PM
Key1
2/24/2011 1:48:07 PM
Key1
5/24/2011 5:18:19 PM
6/16/2011 3:17:40 PM
Key1
6/16/2011 3:25:42 PM
6/25/2011 5:55:55 PM
Thanks....

Help in designing a Query

Hello Everyone,
I want you help in designing a query that involve OACT,OJDT,JDT1,OBGT and BGT1
Actually I want a report that shows accounts Budget quarterly and its expenditure quarterly as well..
Like shown below
Account Name
Budget from Jan to March
Budget from April to June
Budget from July to Sept
Budget from Oct to Dec
Expenditure from Jan to Mar
Expenditure from Apr to June
Expenditure from July to Sept
Expenditure from Oct to Dec
Parameters would be Fiscal Year and Date range and Account name
Can anyone please help me out in that one
Thanks in advance

Looking at the test data I have in these tables, it would appear impossible to join them all in one query without using UDFs...
They don't share any columns.

How can I turn off the pop-up "Help HP Design Products You Want"? (8600 printer)

The pop-up window "Help HP Design the Products you want" keeps returning every 5 minutes all day long
regardless of how I respond to it! Can this be turned Off?
This question was solved.
View Solution.

What printer are you using?
What OS do you have?
Does it give a program name when it pops up, or does it just give you that message and a couple of options to choose from?
-------------How do I give Kudos? | How do I mark a post as Solved? --------------------------------------------------------

Digital Publishing Suite Help | Layout design overview

This question was posted in response to the following article: http://helpx.adobe.com/digital-publishing-suite/help/layout-design-overview.html

Great Video.
Can you have two 15 page docs for Horz and Vert formats marry them in folio, so that article 1 has a Horz and Vert orientation,article 2 the same and so on and so on right up to article 15?
Right now it seems like I would have to break the two docs into 15 individual docs each to accomplish the Horz and Vert orientation.

I am making a Conference/Event program book. Are there mac programs or apps that could help me design this?

I am making a Conference/Event program book. Are there mac programs or apps that could help me design this?

I can see that this issue has been quite extensive, and frustrating, and I am so sorry for all that has happened societygirl! I would like to help you work this issue out. Please follow & send me a Direct Message, so I can get your account specifics and help finally bring this to a resolution.
Thank you,
MichelleH_VZW
Follow us on Twitter @VZWSupport

Help me design certificate services

Hi i am planning ADCertificate Services with PKI and we are going to come up with 2 data centers. Each data center has 2 domain controllers. I would like to come up with Two-Tier CA Hierarchy. Is this the right approach if certificate services
will be hosted on three machines, the offline Root CA wll be a physical machine in datacenter2 with one second tier CA in each data center as a virtual machine. please help me design certificates services if i am going wrong.

first of all thnx for ur reply! actually you are some what right, but what i want is that a client comes and right like this for e.g; sqr(log(sin(n)))
then this will first go to sqr server , then to log and at the end to the sin server which means that sin server will first return the value of the n to log and after the log is computed on the value which was return by the sin server, it will go to the log server and in the end the log server returns the log of that value to the client, as this phenomenon is called grid computing. this is the what i want...
now, where i m stuck is that i know how to make a server and a client .. but i dont know how will i interact one server to the other and returning back the value ... pls help me with some coding of this or any tutorial if u have .. i will be vary glad if u help me in this i will really appreciate your corporation!
thank you! waiting for ur reply
Danish

Please help in designing...

Hey Folks,
A new bee to Flex...Want to design and develop a flexible GUI in flex..Can any one help in designing the GUI?
Requirements
Database
I have four tables namely TableId, ColumnName, ColumnValue, PartyColumnValue
TableId
tableId int
tableName varchar
eg :-
10
Test
ColumnName
columnId int
tableId int
columnName Varchar
eg:-
1001
10
Col1
ColumnValue
valueId int
columnId int
tableId int
valueName Varchar
eg:-
100101
1001
10
val1
PartyColumnValue
id int
valueId int
columnId int
tableId int
In the above manner I am storing the table names, column Names and column values
Flex Requirement(GUI)
On GUI there will be one combobox in which user has the option of selecting the table name
Once table is selected
1. Dynamic Table(DATAGRID in flex) should be displayed on GUI in which the columns will be the column name from the ColumnTable and each row should be the combobox in which the values will be multiple pertaining to the ColumnId and tableId.
2. Each row should be having the option to "save" individual row means the last column will be save button, it will get persisted in the PartyColumnValue table.
3. There should be option to save all the rows means "Save All" button to save all the rows it will get persisted in the PartyColumnValue table.
4. Once i saved the invidual row or all rows and then again if user selects the table from the combobox he should be able to see the persisited values.
5. There should be one "Add Column" button which will add the column by opening the popup and asking for the default values.
Can anyone help me in acheving this functionality?
Thanks In Advance!!!

Can anyone help without money?

Help needed to create Drop down help through Design layer for attribute

Hi All,
Has anybody created a Dropdown help through Design layer for standard field, there we have check Box ............
Please Help me to know this ASAP....
Edited by: Pankaj Gupta on Feb 26, 2010 10:52 AM

Hi,
Here is the exampel Program, i am sending the code for a Paramter, If you want this for the Select-options, then you need to have the paramer name as PS_PARM-LOW and you need to write the smae logic again for PS_PARM-HIGH to get the List box for the select-option
REPORT ZLIST.
TYPE-POOLS: VRM.
DATA: NAME TYPE VRM_ID,
LIST TYPE VRM_VALUES,
VALUE LIKE LINE OF LIST.
PARAMETERS: PS_PARM(10) AS LISTBOX VISIBLE LENGTH 10.
AT SELECTION-SCREEN OUTPUT.
NAME = 'PS_PARM'.
VALUE-KEY = '1'.
VALUE-TEXT = 'LINE 1'.
APPEND VALUE TO LIST. VALUE-KEY = '2'.
VALUE-TEXT = 'LINE 2'.
APPEND VALUE TO LIST.
CALL FUNCTION 'VRM_SET_VALUES' EXPORTING ID = NAME VALUES = LIST.
START-OF-SELECTION.
WRITE: / 'PARAMETER:', PS_PARM.

Design Scenario Question

Hello, I have a question about how to architect a certain part of my code and I will try to formulate an example here:
Say I have a class Parent that should only be constructed with variables a and b. A no-arg constructor does not make logical sense for Parent, as it should not exist without the a and b instance variables having a value.
Parent has two subclasses, ChildOne and ChildTwo which are also constructed using the a and b variables in Parent. The a and b attributes in Parent should be used across the board for all Parent�s children to do class-specific processing. For instance, ChildOne will use Parent�s a and b attribute to derive ChildOne-specific information.
Parent also has a method, getDataFromChildren(int id) which does something similar to:
if (id == 1) {
     ChildOne one = new ChildOne(this.a, this.b);
     one.getChildData();
} else if (id==2) {
     ChildTwo two = new ChildTwo(this.a, this.b);
     two.getChildData();
}     I understand this may violate OO polymorphism principals, which will be one of my two questions listed below:
1. My first problem is that I understand to be correct, this getDataFromChildren method should accept a Parent parameter and then simply invoke a method polymorphically, such as:
getDataFromChildren(Parent p) {
     p.getChildData();
}Then from other parts of my application, one of the child classes will be passed to this method. But, my other parts of the application know nothing about the Parent, ChildOne and ChildTwo classes. All they know is an ID. So, how can they possibly pass a Parent object to this method? Furthermore, this getDataFromChildren method exists within Parent itself. That doesn�t sound correct either.
2.     My second problem involves constructors. Within my children, I have constructors to match the Parent constructor. I also have explicit calls to the super(a, b) constructor in Parent. So, when a child class is instantiated as say ChildOne(a, b), it makes an explicit call to super(a, b). That creates a new instance of parent with a and b, when I already had one in the first place. To make matters worse, my original Parent object contained some other information specific to that instance that the child classes need. When they are instantiated and make the call to super(a, b), they create new instances of Parent completely separate from my original, relevant one.
I suspect that what I need to do here is have some type of intermediate class that takes the ID from another part of my system and interprets it there. I considered that my child classes should accept a Parent object in their constructor, but that doesn�t seem to make sense.
Please help, sorry for being so verbose. Hopefully this made sense.

Learn about object-oriented design patterns a bit. It will help you design applications a bit better and come up with working solutions for isues such as the one you describe.
if (id == 1) {
     ChildOne one = new ChildOne(this.a, this.b);
     one.getChildData();
} else if (id==2) {
     ChildTwo two = new ChildTwo(this.a, this.b);
     two.getChildData();
}Considering the fragment above, you should check out the Abstract Factory design pattern. You could have a factory return the appropriate object based on the logic in the conditional structure. Also, coding to an interface is much more flexible that what you have right now. Below is a example:
Child child = null;
if (id == 1) {
     child = ChildFactory.getChild("1", this.a, this.b);
} else if (id==2) {
     child = ChildFactory.getChild("2", this.a, this.b);
child.getChildData();
...The goal is to reduce the number of hard-coded class names in your code. The only place where there are hard-coded class names is inside the factory. When you want to introduce a new behavior and a new class type, all you need is to add it to the factory.

Need help in design the scenario sending data to web service

Hello All,
I need to send data to web service and get the response back.
i want to use HTTP to soap Scenario.Please let me know the necessary steps in the scenario.
Thanks and Regards,
Chinna

Hi Chinna,
> i want to use HTTP to soap Scenario.Please let me know the necessary steps in the scenario.
Split this into 2 scenarios
1. HTTP to SOAP (Request) treated as Asynchronous.
2. SOAP to HTTP (Responce) treated as Asynchronous.
For HTTP to SOAP synchronous scenario
Here the receiver side request and responce structures no need to create, the receiver client will provide you the XSD file, on that XSD file having the request and responce message structures, so you have to import that to XI server as External definition.
For sender HTTP you have to create 2 data types, 2 message types as well as 2 message interfaces, 2 message mappings for request and responce messages and 2 service/Message interfaces (HTTP is outbound and SOAP is inbound).
Configuration objects are same.
Regards
Ramesh

Help on complex scenario

Hi all
I need help ih the following scenario.
the sender system sendes the invoice to PI , PI has to give the successful receiving answer back
to the sender , and continue to the mapping activities , where the invoice has to be mapped to a PDF image,
PI then call the SAP system(RFC) which in turn calls the content server where the image has to be stored ,
and then we get back to PI , with a successful answer , from there PI calls again the SAP system(RFC) to start a workflow.
Sender -> PI -> sender
PI -> SAP -> contentserver -> SAP -> PI -> SAP.
what is the bedst aprocah to soleve this scenario.
any help will be higlly appreciated.

With BPM
S --> Sender; SAP --> SAP; ContentServer --> CS
Async call between PI and SAP (RFC):
Receive (from S)--> Transformation --> AsyncSend(SAP_RFC) -->AsyncReceive(SAP_Resp) --> Transformation --> AsyncSend(To Start WorkFlow)
In the above design approach:
1) The AsyncReceive Step should be enclosed within a Block with Deadline branch defined so that the BPM does not remain open for long.
2) Transformation step can be avoided and mapping can be done in Interface Determination.
3) Correlation needs to be defined
Sync Call between PI and SAP (RFC):
Receive(fromS) -->Transformation --> SyncSend(SAP_RFC) --> Transformation --> AsyncSend(To Start WorkFlow)
Personally i would prefer Async Call between BPM (PI) and SAP, so that the BPM wont go into error if the ContentServer takes more time for storing the image and giving back the response.
Without BPM
Scenario will be splitted into two Async interfaces:
1) Sender --> SAP PI --> SAP
Not to worry -
> SAP --> CS --> SAP ....let it happen at its own pace
2) This will happen once SAP gets response from CS
SAP --> SAP PI --> SAP (to start workflow)
Why do you need to send the message to PI and then again to SAP to start the workflow......can you not trigger a job within SAP itself which will start the workflow once it receives the response from CS?
Regards,
Abhishek.

Help! Designer out of his depths developing JAVA mobile game ...

Hi, I'm a desiger challenged with designing a JAVA based, multiplayer game for mobile.
Feeling a little exposed, as it's a more limited design environment than I'm used to.
Key page I'm struggling with is the profile page, on which we show:
your accumulated score over time
your rank (the higher your accumulated score, the higher your rank)
your average score.
have found an interesting reference that XBOX are using for their connected game. nice and simple:
http://news.bbc.co.uk/1/hi/technology/4534356.stm
does anyone have any similar or better references for me please?
Many, many thanks for your help.
Rich

There is an end-to-end tutorial that is available at http://www.adobe.com/go/learn_lc_firstApplication_82 that steps you through creating an application. The scenario in the tutorial shows you how to create a form design and a process where users interact with it in Workspace ES. You will also need to download the assets require to step through the tutorial located at http://www.adobe.com/go/learn_lc_firstApplicationZip_82. Go to the provided URLs to get the most updated versions.or
To start this tutorial, it is recommended that you install a JBoss Turnkey installation of LiveCycle ES Update1 (server) and Workbench ES. I presume you have done this, but if you require more information, please let me know and I'll provide more information.
Hope that helps!
...Gil

Need Help: Query Designer

Hi experts,
In Query designer, is it possible to create an IF statement for "IRRATIONAL NUMBERS"?
Lets say 10.03, 1.26, 3.1 and so on...
The statement would be as follow:
If Val1==N.x then 0 else Val1.
To summary, if the value is N.x (N for rational numbers 1,2,3,... and .x for decimal values like .23, .44, .01 and so on...),
then the value N.x will be changed or converted into "0".
Example: If the value of FieldA on the report is 10.3, it then should be converted into Zero "0".
another is if value is 1.03, then the report should show 0 instead of 1.03
I arrived to a solution wherein I have to create three logic expression.
First is to convert first the Irrational Number into Zero "0".
Second is to create new selection in KF wherein it will pull the data containing the converted "N.x to Zero".
And third is to create the IF statement logic that will satisfy my expected value.
I'm having this problem on the first one as I have no idea how to convert irrational number into zero "0" value.
the second and the last part would be fine as my main problem here is how can I convert the irrational number in to zero?
Any suggestions?

Hi,
Please try the below.. Hope this will solve your issue
Eg. of data
Dist. Channel -
Value
Dom1----100.5
exp1----200.4
dom2---50
exp2---75
Expected out put
Dist. Channel -
Value
Dom1----0
exp1----0
dom2---50
exp2---75
Steps1
Drag and drop Dist.Channel in row and value in colum
create a formula1 using ur kf value
CEIL(value) --- which will provide us integior part with out decimal
Steps2
create another fomula2
using boolean operator compare the value
(formula2==value)*value
this formula compare our two values and will produce a boolean value ie 1 or 0
: Eg 100.5 ceil(100.5) = 100 Now in our formula1 we are comparing these two values are equal or not ie 100.5 - 100 so out come will be 0 then 0100.5 = 0 always if our condition is true then 1100 = 100
Related Links :
Boolean Operator
http://help.sap.com/saphelp_nw04/helpdata/EN/71/0aec3b6e011341e10000000a114084/frameset.htm
Thanks
BVR

Helping in Designing Scenario

Similar Messages

Maybe you are looking for