[Isomorphic]

Once again - any third-party JS?

You previously mentioned that the ResultSet appeared to be crashing while "creating an auto-operation" - actually there's no indication of this - this log:

15:01:36.086:TMR8:INFO:ResultSet:isc_ResultSet_0 (created by: undefined):creating auto-operation for operationType: fetch

.. is the normal last log you would see from a ResultSet.

To disambiguate where the crash is:

1. java.lang.Exception.printStackTrace() prints the stack to stderr, you could find this log in JBoss or do the following silly dance that all Java programmers eventually write:

public static String getStackTrace(Throwable t)
{
StringWriter sw = new StringWriter();
t.printStackTrace(new PrintWriter(sw));
try { sw.close(); } catch (Exception e) {}
return sw.toString();
}

2. just add another logWarn() after the getRange() to call in the else case to clarify whether it returns successfully

[jpappalardo]

Although I have given you this class in its entirety, there seems to be a great deal of misunderstanding how this class functions. Without deferring you back to the beginning of the thread I will go over the high points.

Constructor


My code does not do anything until ResultSet.allRowsCached is marked as true.

on construct: DataArrivedHandler is created on each result
on init: fetch data
onDataArrived: create DataChangedHandler, process cached data
onDataChanged: process cached data

[Isomorphic]

Yes, we know, but we can't run the code so we're dependent on you for diagnostics.

Can you please:

1. answer the question about 3rd party JS (this is the 4th time we've asked)

2. obtain a stack trace for the error, or let us know if something is stopping you

[Isomorphic]

Hang on, we may have this now, Googling ArrayStoreException has revealed a bunch of GWT issues here - looks like this may be the result of three (or more) interlocking GWT (not SmartGWT) bugs - one that causes a bogus exception in GWT's JDK emulation code, one to cause the bogus Exception not to have the right details, and one to cause the Exception not be reported in compiled mode.

As far as we can tell, this exception is basically being thrown spuriously after everything has already gone correctly inside SmartGWT, so it just needs to be ignored - a valid workaround is to put a try..catch block around any call to getRange() that is really just intended to trigger a fetch.

[jpappalardo]

So I'm not crazy...good to know

[Isomorphic]

Didn't say that ;)

This is still proving elusive - can we have a couple more details: Eclipse version, OS you're developing on, OS you're deploying too if different?

[jpappalardo]

IntelliJ 11.1.1

Developing on Win XP SP 2

Deploying to JBoss AS 7.1.0 Final on Win XP SP 2 and RHEL 6.1

[Isomorphic]

Ok we believe this is now resolved.
Please try the next nightly build, dated April 26 or greater (the fix will be in both the 3.0p and 3.1d branches).

Thanks
Isomorphic Software

[jpappalardo]

Looks good