Post Children Before Parent

First, I just want to apologize if this isn't appropriate here. Feel free to lock/delete/move my post elsewhere :)
I'm having a BC4J related problem in the OA Framework on JDeveloper 9.0.3.5 and Oracle EBS 11.5.10.
I have two EOs that are linked by composition. My business requirement / process is as follows:
1. Delete child record.
2. Do any more child records exist?
2.a. No, delete parent.
2.b. Yes, update parent.
However, I can't seem to get this to work because the parent's changes are always posting before the child's which means I'm getting an ORA-02292: integrity constraint <constraint name> violated - child record found constraint violation.
What's the correct way to implement this requirement?

Timo Hahn wrote:
Check out the following links
Re: ADF BC - unable to control posting order
http://jobinesh.blogspot.com/2010/08/customizing-order-of-posting-changes-to.html
TimoI tried the solution that was mentioned in the forum thread and I'm still getting the same error.
The OA Framework Developer's Guide says that in a composition association the parent's always post before the children. Does this mean I can't use a composition association?
I also wanted to add that I tried an association that was a non-composition and I still get the same error.
Edited by: Centinul on Sep 14, 2011 7:39 AM

Similar Messages

Inventory: Movement after count with post date before the count date.

Hi,
We have stock movements all the days of the week.
We conduct physical inventories on Sunday mornings. There are not stock movements on Sunday mornings (stock movements on Sunday starts at night).
Usually we create the SAP Physical Inventory Document on Fridays, assigning Saturday as the Planned Count Date.
We enter the SAP Physical Inventory Count on Sunday (at the end of the count) or on Monday.
How you recommend us to use SAPu2019s Physical Inventory feature in order to calculate the Book Inventory considering all stock movements with posting date less than or equal to Saturday?
Please, suggest us a solution that If we register a movement after we enter the count but with post date before the count date (and the count is not posted yet), the Book inventory is modified accordingly.
Thanks in advance for your support!!

SAP takes the book inventory in the moment you enter the count.
Best you set block movements , this block will be removed automatically when you enter the count. So you can prevent users from posting until you have entered the count.
If you set the freeze book inventory, then SAP takes the book inventory in the moment you create the document.
This is not a good way if you still have movements after inventory document creation and actual count.
You have to make sure - organisational wise, outside SAP - that all movements that happened before inventory count are posted before you start entering the count. There is nothing in SAP that can help you in this case.

Child dialog set visible before parent dialog how to avoid this

I m using JDialog an for some validation showing JOptionPaneDialog but parent dialog get visible after some processing that depands on some thread execution so before parent dialog get visible the JOptionPaneDialog get visible how to resolve this problem

I m using JDialog an for some validation showing JOptionPaneDialog but parent dialog get visible after some processing that depands on some thread execution so before parent dialog get visible the JOptionPaneDialog get visible how to resolve this problem

Flash CS4 - how to create event flow from children to parents?

In my opinion natural event flow direction is from children
to its parants.
E.g. when in dialog box buttons are pressed it is natural to
inform only
this dialog box about these actions. When dialog box can't
handle particular
event then passes it to its parent and so on .
Unfortunately this direction is not well supported by Flash.
To achieve
support for this event direction it is necessary to call
parent.dispatchEvent() with bubbling option off. This
solution is
inconvenient because sometimes I can't predict what event
types will be
triggered inside dialog components. Most likely there is no
function to
catch all events which have no defined handlers in dialog in
order to be
possible to pass them one level higher.
Have you got any experience with building communication model
between
objects in Flash applications?
Regards,
Marek

XML shema is basically an XML file. So u need to know how to create an XML,
provided u know how the shema file should be.
Creating an XML :
http://forum.java.sun.com/thread.jspa?threadID=5181031&messageID=9705786#9705786

Can any one tell me ...how to check my old posts ie before dec 2007..

Hi ,
After long time I am back to check my posts ..I didn't find..
Is there any way that I can check my past posts ...which is before dec 2007.
Thanks
M.V

Or I could be having brain fade (polite term, there) of the highest order -- how about right at the top of the forum -- the link that says "Hyperion Developer Network (HDN) Forums Archive" with the big yellow asterisk? Perhaps you too were looking within a post and couldn't see it? That's my excuse and I'm sticking with it. :)
Or are you asking for something else?
Regards,
Cameron Lackpour

Question about creating child domains "before" parent

Ok, this is most probably a noob question.
My company (A.com), has a DNS zone in a linux server with records pointing to different web pages, example: radio.a.com, www.a.com, webmail.a.com, etc..
For a specific service, I need an Active directory domain called: daas.A.com
Thing is, I don't have an AD domain A.com in place. So my question is:
Do I need to create an AD domain for A.com before creating the one that I need?
If I don't, and I just create the daas.A.com as a new forest, will I be able to add an A.com domain in the future as parent?
As per DNS records in the linux box, I guess I would only need a NS pointing to the DNS of the new domain, and an A record resolving that to an IP. or Something like that..
Thanks!

Yes, even if initially it is an empty root, otherwise the child domain will become the forest root.
So the A.com domain will need to duplicate those records that Linux currently hosts otherwise AD clients will start having name resolution issues. If you just need to stand up AD for one application then I would go ahead and standup the empty root, add the
child domain, install the service that needs AD and go from there. That is if you think at some point you will use the a.com domain.
Active Directory will actually create 2 zones (A.com and _msdcs.A.com)
Daas.A.com will have 1 dns zone daas.a.com and will also utilize the _msdcs.a.com
So I have been in environments which have had both a windows dns server and a Linux dns server, eventually after enough duplication of records in both areas and the pain points that caused, we have retired the Linux dns servers and just used the AD DNS servers
(with the exception of DNS servers that were internet facing which we kept on Linux) Primarily the ease of administration and the fact that AD and DNS are tightly coupled.
Brad Held http://windorks.wordpress.com

Children and parent in different CDM packages?

Hi,
I wonder if there is a way to use inheritance and packages, in a conceptual data model (CDM), so that the parent and the children are not in the same packages?
I have first created entity A, B and C in a Package P1. Then, I have set an inheritance, having A as parent, and B and C as children. Then I have created a Package P2 and I have moved the entity B into it.
That way, if I open the entity B, I can still see A as its parent. However, in the P1 diagram, I cannot see B the inheritance anymore. And this, not graphically and not also in the child list when I open the inheritance. Also, if I try to add a new child in the inheritance, I can only choose among the entities included in P1.
In the P2 diagram, I can create a shortcut for the A parent entity but I cannot create an inheritance between that shortcut and B. Same thing if I create a shortcut of the existing inheritance in the P2 diagram. It will show with its link with the A entity but will not allow to add a link with B.
In conclusion, it is like I have kept half of the inheritance definition but lost the other half. Is there a way to keep everything? Is that a bug?
Thanks
Daniel

Hi,
if I understand you right - I had the same problem.
Look here:
Inheritance between two packages
Best,
Ahmet

Posting MIRO before MIGO

Dear Experts,
When a PO is raised for 100no's with each piece costing 10. Before posting a MIGO a MIRO is posted for 50 no's for 600.
Then the MIGO is posted for 100 no's.
what price will be taken at the time of MIGO?
what account get generated when the material is maintained in Standard Price as well MVP.
Regards,
Nani.

hi
>Is the value taken from MIRO for MVP or Standard Price.
it will take from MIRO only
>In this two cases what will be the accounting entries and G/L account hits?
with MAP
At the time of LIV (MIRO);
Vendor A/c - Cr
GR/IR Clearing A/c - Dr
At the time of GR (MIGO);
Stock A/c - Dr
GR/IR Clearing A/c - Cr
WITH STD
At the time of LIV (MIRO);
Vendor A/c - Cr
GR/IR Clearing A/c - Dr
PRD -Dr
At the time of GR (MIGO);
Stock A/c - Dr
GR/IR Clearing A/c - Cr
PRD-Cr
regards
kunal

Child Records getting inserted before Parent

HI All,
I have an Entity object (Vendors) that has a defined relationship to payments (12 months) when I create a new vendor record I'm creating the 12 child records that all works fine but when I try to issue a commit the child records get inserted first creating a constraint error. If I turn the constraint off and using the "Jbo.debugoutput=console" I can see the 12 months being inserted followed by the vendor record. The payment records vendor_id is being set to the correct parent but the insert is happing in the wrong order does anybody know how to control the order of the insert. There's a Association set between the Vendor entity and the payment entity. If I move the code to a view object it seems to work fine. But in this case we have multiply view objects on these entity objects and I'm trying to prevent having to write the same code in every view Object.
Here's the code for the Create on the Vendors EntityImpl
protected void create(AttributeList attributeList) {
super.create(attributeList);
// Set the default values and the Seq number for this new record
SequenceImpl seqNoImpl = new SequenceImpl("VENDORS_SEQ",getDBTransaction());
this.setVendorId( seqNoImpl.getSequenceNumber());
Number zeroAmt = new Number(0);
RowIterator paymentsIter = this.getVendorPayments();
// When creating a new Vendor create the 12 months of payment records
for (int i = 1; i < 13; i++) {
Row newRow = paymentsIter.createRow();
newRow.setAttribute("Month",i);
newRow.setAttribute("Amount",zeroAmt);
paymentsIter.insertRow(newRow);
Dose anybody have any ideas....

Pieter,
Thanks for the information!!
Using it I was able to get the process working. There where some interesting side affects. If I just turned on the "Composition Association" then no payment records got inserted into the database even though they where in the entity and view objects. I took the second recommendation of adding code to the PostChanges event and was able to get the parent and children inserted in the correct order.
Thanks again for the idea and information

Return STO able to post GR before GI

Hi,
We have a Return STO flow. The PO has the check-box "return PO" marked, the MIGO do the goods issue and the delivery do the goods receipt.
System allow execute the GR on the delivery before the MIGO make the GI. It create a negative stock in transit. How can we avoid it?
Issue is specific to the above mentioned not for the general flow of Return STO.
Thanks in advanced.
RG.

Hi
Issue is not about the Negative stock it is a consequence only.
Main Concern here is, for Return STO, GR is able to post before the GI (means before the part return).
Have read through the Notes: 25220,361014,1499411. but not helpful.
regards
RG

Posts Written Before Registering Are Just Deleted?...

Hi ... Just curious if this is somehow me, or if this inane design flaw really slipped through the cracks here.
I had a questions re Skype. The website provides no means of actually contacting someone, insofar as i can tell, so i gave up and came to the community.
I entered the community, and found a bunch of boards listed. I couldn't find one that fit my question, but i went for my best guess. I took the time to write up a detailed post with my question. I hit "Post".
The site then prompted me to log in. So i did. Then it asked me to create a community name to protect my Skype login. So i did. Then it returned me to the community.
So what happened to my post? No clue! The board doesn't show it. My profile doesn't show it. I tried to hit "Back" in my browser to recover it, but returning to the page reloaded it as if i wanted to start over fresh. It's gone!
Is this _really_ how the board is designed to work? The site didn't tell me i had to log in or register before posting, so i figured it would either ask later or let me post as a guest. ... but then the act of registering somehow makes the system forget about my post? Did no one think of this? Please tell me i'm missing something!!
Solved!
Go to Solution.

See this:
http://community.skype.com/t5/Suggestions-for-this-Community/Post-content-disapears-when-trying-to-p...

Material consumption posting date before MIGO

Hi,
I have one issue related to Material Consumption....
If material received and posted on 25.12.2011 then how come system is allowing to consume ( Mtype 261) on 12.12.2011. As material stock on 12.12.2011 is Zero. We have not allowed Negative Stock Posting for Plant or in Material.
We are consuming material for Production Order with in CO11N with Mtype 261. Material phisically received on 12.12.2011 and we used it but due to some problem we are not able to do MIGO. Now problem solved and MIGO posted on 25.12.2011, so we entered confirmation of production order with posting date 12.12.2011 and system is allowed to consume the material and now account reports shows nigetive values on 12.12.2011.
can anybody focus on this issue......

SAP checks only if there is enough stock in the period (month) not at a certain day (for performance reasons)
Dont blame SAP, Why has your user posted the goods issue with an earlier date?
Usually they do it because it was the date the goods issue happened physically, I have rarely seen users that user changing posting dates to fraud the system.
was the GR eventually posted with a wrong date?

One-to-many relation sometimes INSERTs child before parent

Hi,
I have a one-to-many relationship, say Company-Employee, that I want to be
managed by the parent entity (Company), so that when I make persistent the
Company, its Employees are also made persistent, and when I delete the
Company, its Employees are deleted. In my Java code, Company has an
ArrayList of Employees, and Employee has a Company member variable.
My metadata looks like this:
<class name="Company" identity-type="application">...
     <field name="employees">
          <collection element-type="Employee"/>
          <extension vendor-name="kodo" key="inverse-owner" value="company"/>
          
          <extension vendor-name="kodo" key="element-dependent" value="true"/>
     </field>
</class>
<class name="Employee" identity-type="application">...
     <field name="company"/>
</class>
I have a test that is creating a Company with an Employee and persisting
it. It then deletes the Company. The test is passing some of the time and
failing some of the time. It seems to not be consistent with the ordering
of the INSERTs - sometimes it inserts Company then Employee, but sometimes
it first tries to insert into Employee, which fails with a foreign key
constraint to the parent table.
How can I force it to insert into the parent table first?
I notice that the documentation says you need the field of the child
entity to be the inverse-owner. Is it not possible for the parent entity
to own the relationship?
Many thanks,
Alex

Alex,
If you have non-deferred foreign keys between the tables, and you turn
on Kodo's foreign key reordering algorithm, Kodo will ensure that all
your foreign key constraints are met, regardless of the ordering of the
operations in your Java code.
See http://docs.solarmetric.com/ref_guide_dbsetup_fks.html for details
about this.
Regarding inverse-owner specification -- is there any reason why you
want to specify this on the other side? inverse-owner basically defines
which side of a shared relationship defines the mappings and should be
monitored for changes; it does not have any impact on the mappings
themselves.
-Patrick
Alex Robbins wrote:
Hi,
I have a one-to-many relationship, say Company-Employee, that I want to be
managed by the parent entity (Company), so that when I make persistent the
Company, its Employees are also made persistent, and when I delete the
Company, its Employees are deleted. In my Java code, Company has an
ArrayList of Employees, and Employee has a Company member variable.
My metadata looks like this:
<class name="Company" identity-type="application">...
     <field name="employees">
          <collection element-type="Employee"/>
          <extension vendor-name="kodo" key="inverse-owner" value="company"/>
          
          <extension vendor-name="kodo" key="element-dependent" value="true"/>
     </field>
</class>
<class name="Employee" identity-type="application">...
     <field name="company"/>
</class>
I have a test that is creating a Company with an Employee and persisting
it. It then deletes the Company. The test is passing some of the time and
failing some of the time. It seems to not be consistent with the ordering
of the INSERTs - sometimes it inserts Company then Employee, but sometimes
it first tries to insert into Employee, which fails with a foreign key
constraint to the parent table.
How can I force it to insert into the parent table first?
I notice that the documentation says you need the field of the child
entity to be the inverse-owner. Is it not possible for the parent entity
to own the relationship?
Many thanks,
Alex

To find all the child records are processed before parent records

Hi,
When i am trying to process consignment pulls and my client is using third party tool to insert the data into interface tables. My logic is written in such a way that it should pick parent record associated with child but there is some delay in child records and not all the child records are being pulled with parent.
In a brief, i need to write a condition in such a way that my parent cursor should pick all the child records with max(last_updated_date) and then it should be processed.
eg: I have a family and need to catch bus with all my family members. I need to check now for the last family member who get into and should wait for atleast 2mins so as to ensure all my family members are in. I am unable to achieve this with count. Please do revert back soon.
Details are below:
CURSOR c_pull (p_pull_id NUMBER) IS----------------------------------parent cursor
SELECT ecp.ROWID ROW_ID
,ecp.*
FROM consignment_pulls ecp
WHERE ecp.process_flag IN ( 2, 3 )
AND ecp.pull_id > NVL(p_pull_id,0)
ORDER BY ecp.pull_id
FOR UPDATE OF ECP.PROCESS_FLAG NOWAIT;
CURSOR c_line (p_pull_id NUMBER) IS----------------------------------child cursor
SELECT ecpi.ROWID ROW_ID
,ecpi.*
FROM consignment_pull_items ecpi
WHERE ecpi.process_flag IN ( 2, 3 )
AND ecpi.pull_id = p_pull_id
FOR UPDATE OF ecpi.process_flag;
Edited by: Prathyusha on Apr 5, 2012 12:07 AM
Edited by: jerry on Apr 5, 2012 12:10 AM

Hi Helios,
Thanks for the info. But its not a concurrent request something like we can filter with start date and completion date. Its all about selection of records from headers and lines table and not able to make those selection of header records with a delay of say some 2minutes until all the child records are inserted.
Thanks,
Prathyusha.

Thinkpad T400 reboot after post (right before boot)

After update bios to 2.03 i noticed, that sometimes my thinkpad reboots after post (right after system booting). First I ignored this issue, because i thought it's caused by vista cd in drive, so i was removing it and everything seemd to be fine. But now i know that this is compleatly cd-independed. It happens in almost every second boot, and is independed on boot priority or boot media (cd, usb, hdd). I tried to downgrade bios to 1.20 and 1.23, but it still happening.
If anyone know what is going on, please let me know. I'm little stressed.
I noticed one more think. After change to "diagnostic boot" in bios, after reboot I hear beep, which is clearly longer, than when reboot not occurs.
Message Edited by lucks on 03-02-2009 02:10 AM

This is still happening... but I think I've noticed some things.
I think what the problem is, and I'm sure this'll sound retarded, is Arch is restarting "too quickly". I say this because there are some times when reboot works properly. When this happens, I notice all of the hardware shut completely down for about a second. I hear sort of a "click" indicating to me that discs have stopped and power is off, like the very last sound you hear when you completely shut down a computer. Instead of a blank screen, I can tell the monitor is completely shut off. After about a second of this, everything comes back on, the Thinkpad splash screen appears, and then goes away and I am taken to GRUB. Everything is peachy.
When the reboot fails to work properly, however, I don't get that total shutdown "feeling".. no click sound, and my screen just goes blank, but not totally off. The Thinkpad splash screen reappears sooner, but this is where I hang and eventually get the error I reported in the original post.
I hope this additional info helps, as this problem is really annoying.

Post Children Before Parent

Similar Messages

Maybe you are looking for