[
https://issues.apache.org/jira/browse/TAP5-1963?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Ben Dotte updated TAP5-1963:
----------------------------
Description:
advise() in CommitAfterWorker surrounds the method invocation and session
commit in a try/catch and runs manager.abort() if a RuntimeException is caught
and then rethrows the RuntimeException. The problem is, depending on the
original problem, it is not unlikely that manager.abort() itself could throw an
exception (Transaction Not Started, Transaction Already Started, etc.). When
that happens, the original RuntimeException is lost and all we get is the
useless exception generated by the call to manager.abort().
I think we should just throw away any exceptions generated by manager.abort()
so that we always retain the original RuntimeException. Something like this:
try
{
invocation.proceed();
// Success or checked exception:
manager.commit();
}
catch (RuntimeException ex)
{
try
{
manager.abort();
}
catch (Exception e)
{
// throw away; we want the real "ex" exception to get rethrown
}
throw ex;
}
was:
advise() in CommitAfterWorker surrounds the method invocation and session
commit in a try/catch and runs manager.abort() if a RuntimeException is caught
and then rethrows the RuntimeException. The problem is, depending on the
original problem, it is not unlikely that manager.abort() itself could throw an
exception (Transaction Not Started, Transaction Already Started, etc.). When
that happens, the original RuntimeException is lost and all we get is the
useless exception generated by the call to manager.abort().
I think we should just throw away any exceptions generated by manager.abort()
so that we always retain the original RuntimeException. Something like this:
{code}
try
{
invocation.proceed();
// Success or checked exception:
manager.commit();
}
catch (RuntimeException ex)
{
try
{
manager.abort();
}
catch (Exception e)
{
// throw away; we want the real "ex" exception to get rethrown
}
throw ex;
}
{code}
> Original exception lost in CommitAfterWorker upon abort
> -------------------------------------------------------
>
> Key: TAP5-1963
> URL: https://issues.apache.org/jira/browse/TAP5-1963
> Project: Tapestry 5
> Issue Type: Bug
> Components: tapestry-hibernate
> Affects Versions: 5.3.3
> Reporter: Ben Dotte
> Priority: Minor
>
> advise() in CommitAfterWorker surrounds the method invocation and session
> commit in a try/catch and runs manager.abort() if a RuntimeException is
> caught and then rethrows the RuntimeException. The problem is, depending on
> the original problem, it is not unlikely that manager.abort() itself could
> throw an exception (Transaction Not Started, Transaction Already Started,
> etc.). When that happens, the original RuntimeException is lost and all we
> get is the useless exception generated by the call to manager.abort().
> I think we should just throw away any exceptions generated by manager.abort()
> so that we always retain the original RuntimeException. Something like this:
> try
> {
> invocation.proceed();
> // Success or checked exception:
> manager.commit();
> }
> catch (RuntimeException ex)
> {
> try
> {
> manager.abort();
> }
> catch (Exception e)
> {
> // throw away; we want the real "ex" exception to get rethrown
> }
> throw ex;
> }
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
