none
Running files sequentially RRS feed

  • Question

  • I have one EDi files with multiple ST and SE Segment........when i use edi receive pipeline it debatches the input file depeneding uopn number of ST and SE ...

    now when it goes to orchestration... i want it to go in a sequence...it does not maater which sequence it should go...

    Is it possible.....

    Tuesday, May 22, 2012 9:50 PM

Answers

  • When I first started using BT, that is what I tried to do. We migrated from another product where breaking the claims and them putting them back together was the solution. Then we learned about BT and found that it served us MUCH better to do a map to map inside an orchestration. I am guessing that you are pre-processing claims, and you may have logic to route each claim based on billing NPI, member ID, or many other rules in place. What we found was that allowing BT to break each claim inside an orchestration and then processing each message individually worked out MUCH better. Then learned that breaking and putting back together was not nesesary after we learned BT. Sometimes you may need to put a batch file together after you break the file inside BT, and you can do that at the party, and set the batch to release with a rule. But I wanted to share with you to consider using one orchestration where you can route messages based on the rules of the company. While it is possible, it is more difficult to break the claims and put together instead of handling each messge individually.

    However, if you must break the claim and you need an ID, the way to make it work is to map the file to anther map linking all the fields minus the ST-SE. Then, use one of the REF segments in the 2300 loop that you do not need for processing, unlink it, and assign a scripting functoid that will count the links. Thart will give you the REF number in that will be sequential so you can use as the primary key for that file inside the tables. Bascially you take that output file and then pass it to the maps that will push it to the 6 tables.

    Like I said, I strongly do not recomend you break the claim. We started that path and spent a LOT of time trying to make it work. Come to find out, I think processing the claims individually is much better and now we only use one transaction table instead of the tables you need to breatk the claim.


    Please Indicate "Mark as Answer" if this Post has Answered the Question

    • Proposed as answer by Carlos T. _ Thursday, May 24, 2012 6:03 PM
    • Marked as answer by LeoTangModerator Wednesday, May 30, 2012 9:41 AM
    Wednesday, May 23, 2012 6:01 PM

All replies

  • Hi,

    The orchestration here will be subscribing to split messages, in that case you would not be able to maintain sequence. However you can do a singleton pattern where only one orchestration processes all the split messages then you will be able to achieve the desired.


    Regards,
    Bali
    MCTS: BizTalk Server 2010,BizTalk Server 2006 and WCF
    My Blog:dpsbali-biztalkweblog
    -----------------------------------------------------
    Mark As Answer or Vote As Helpful if this helps.

    Wednesday, May 23, 2012 9:35 AM
  • Just curious, why do you want the transactions to be sequential?


    Please Indicate "Mark as Answer" if this Post has Answered the Question

    Wednesday, May 23, 2012 11:54 AM
  • Carlos :i am running a stored procedure in the last which do some data look up ........when it does not run sequential my stored procedure does not give correct output...

    Bali:  Is there any document to know how to achieve singleton pattern in the orchestration...

    Wednesday, May 23, 2012 3:36 PM
  • The reason I am asking is because I am not sure that the singleton will help you. I use the singleton but it is to process one file at a time. If you want to process the files in order, you would use the ordered delivery. But it sounds like all the transactions are inside one file, so neither solution will work based on my understanding becuase once biztalk divides the file into multiple transactions, it is very dificult to figure out which one goes first, unless I missunderstood what you are doing.

    What type of transaction are you running? I have several ideas that will help. We do something similar to process subscribers and dependents, where they come inside a file, but we have to process one before the other. 


    Please Indicate "Mark as Answer" if this Post has Answered the Question

    Wednesday, May 23, 2012 4:40 PM
  • I am using Biztalk 837_I and P ...........and all the transaction are inside one file......i want something like each transaction run in sequence so that my stored procedure will give correct output.......
    Wednesday, May 23, 2012 5:15 PM
  • Do you mind sharing what you mean by output? I need a bit more info so that I can give you the correct info. You are talking about claims, and there are a lot of scenerios that can happen. What are you trying to do?

    Please Indicate "Mark as Answer" if this Post has Answered the Question

    Wednesday, May 23, 2012 5:31 PM
  • I am putting the data of transaction in database in around 6 tables.......then at last i call my stored procedure which will take the Id for first table and assign it to the all other file table.....also i m doing the same thing for file name....but when it does not run sequential it does not give correct output...

     

    Wednesday, May 23, 2012 5:36 PM
  • It sounds like you are breaking the claims and then putting them back together. Is that a fair statement?

    Please Indicate "Mark as Answer" if this Post has Answered the Question

    Wednesday, May 23, 2012 5:38 PM
  • yes...thats exactly i am doing.....also i need to associate file name with the all the six table ...and id of first table to all the five table ....
    Wednesday, May 23, 2012 5:45 PM
  • When I first started using BT, that is what I tried to do. We migrated from another product where breaking the claims and them putting them back together was the solution. Then we learned about BT and found that it served us MUCH better to do a map to map inside an orchestration. I am guessing that you are pre-processing claims, and you may have logic to route each claim based on billing NPI, member ID, or many other rules in place. What we found was that allowing BT to break each claim inside an orchestration and then processing each message individually worked out MUCH better. Then learned that breaking and putting back together was not nesesary after we learned BT. Sometimes you may need to put a batch file together after you break the file inside BT, and you can do that at the party, and set the batch to release with a rule. But I wanted to share with you to consider using one orchestration where you can route messages based on the rules of the company. While it is possible, it is more difficult to break the claims and put together instead of handling each messge individually.

    However, if you must break the claim and you need an ID, the way to make it work is to map the file to anther map linking all the fields minus the ST-SE. Then, use one of the REF segments in the 2300 loop that you do not need for processing, unlink it, and assign a scripting functoid that will count the links. Thart will give you the REF number in that will be sequential so you can use as the primary key for that file inside the tables. Bascially you take that output file and then pass it to the maps that will push it to the 6 tables.

    Like I said, I strongly do not recomend you break the claim. We started that path and spent a LOT of time trying to make it work. Come to find out, I think processing the claims individually is much better and now we only use one transaction table instead of the tables you need to breatk the claim.


    Please Indicate "Mark as Answer" if this Post has Answered the Question

    • Proposed as answer by Carlos T. _ Thursday, May 24, 2012 6:03 PM
    • Marked as answer by LeoTangModerator Wednesday, May 30, 2012 9:41 AM
    Wednesday, May 23, 2012 6:01 PM