Bug 247691 - Transaction.VALIDATE_EDIT option prevents transaction reuse
Summary: Transaction.VALIDATE_EDIT option prevents transaction reuse
Status: VERIFIED FIXED
Alias: None
Product: EMF Services
Classification: Modeling
Component: Transaction (show other bugs)
Version: unspecified   Edit
Hardware: PC Windows XP
: P3 major
Target Milestone: ---   Edit
Assignee: Christian Damus CLA
QA Contact:
URL:
Whiteboard:
Keywords: contributed
Depends on:
Blocks:
 
Reported: 2008-09-17 13:30 EDT by Linda Damus CLA
Modified: 2017-02-24 15:10 EST (History)
1 user (show)

See Also:

Attachments
Proposed patch and JUnit test (3.84 KB, patch)
2008-09-17 14:04 EDT, Linda Damus CLA 		give.a.damus: iplog+
Details | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Linda Damus CLA 2008-09-17 13:30:33 EDT 
The Transaction.OPTION_VALIDATE_EDIT option is not inherited by children of a
CompositeEMFOperation.  This causes the child operations to execute in their
own new transactions, even when reuseParentTransaction is set to 'true',
because the options of the child operation are always found to be different
from those of the active transaction (which has the validate_edit option set).

Bug 245446 was opened to request a general enhancement to allow options that don't affect the behaviour of a transaction (like
validate_edit) to be excluded from the AbstractEMFOperation#optionsDiffer
check, allowing clients to reuse the active write transaction in this case.  This will likely require new API.

In the mean time, I have opened this bug to track a fix that will address the specific effect of the Transaction.VALIDATE_EDIT option on transaction reuse.  This fix will not require new API and should be suitable for the maintenance branch.

I have set the severity to 'major' because we have found that these additional transactions have a significant negative impact on performance (see bug 141051).
Comment 1 Linda Damus CLA 2008-09-17 14:04:23 EDT 
Created attachment 112801 [details]
Proposed patch and JUnit test
Comment 2 Christian Damus CLA 2008-09-17 14:11:45 EDT 
Cool, thanks!  With a JUnit test!

Let's see whether this can be squeaked into the Gany SR1, today.  Looks like a pretty bad performance regression.
Comment 3 Christian Damus CLA 2008-09-17 15:14:49 EDT 
Committed the patch to the 1.2.2 branch.  Thanks for the contribution!
Comment 4 Christian Damus CLA 2008-09-18 12:51:06 EDT 
Fix available in R1_2_maintenance: 1.2.2 (R200809171545).