-
1. Re: Teiid 8.5 Build fails
shawkins Sep 27, 2013 8:25 AM (in response to rakeshsagar)The build is fine for me, but there could be some unanticipated locale/platform variation in effect. What is your locale and what vm version are you using?
Steve
-
2. Re: Teiid 8.5 Build fails
rakeshsagar Sep 29, 2013 5:47 AM (in response to shawkins)My locale is en and I am building the Hotspot VM on Windows 7 64 Bit version.
Java HotSpot(TM) 64-Bit Server VM (build 22.1-b02, mixed mode)
-
3. Re: Teiid 8.5 Build fails
shawkins Sep 30, 2013 10:07 AM (in response to rakeshsagar)Rakesh,
Can you provide the full failures from the surefire files? That may help me see what is going on before having to track down a similar windows vm.
Steve
-
4. Re: Teiid 8.5 Build fails
rakeshsagar Sep 30, 2013 11:05 AM (in response to shawkins)Attached are the sure fire files for the failed test cases.
-
5. Re: Teiid 8.5 Build fails
shawkins Sep 30, 2013 1:54 PM (in response to rakeshsagar)At least for the parsing tests the issue is our use of the line.separator property in the QueryParser method that is creating the string message. For consistency it should be using just the escaped new line instead.
As for the time issue, while your language is English you must be in a locale that has a timezone with a non-hour aligned offset from GMT and no (or at least a zero value for the DST offset) DST. From what I can see the logic doesn't correctly calculate the diff for calendar based hours/days/weeks in that circumstance for all time values.
Can you log something?
Steve
-
6. Re: Teiid 8.5 Build fails
rakeshsagar Sep 30, 2013 11:48 PM (in response to shawkins)Opened a JIRA issue for this problem.
[TEIID-2681] Test Case failures in teiid-engine - JBoss Issue Tracker
-
7. Re: Teiid 8.5 Build fails
shawkins Oct 1, 2013 3:33 PM (in response to rakeshsagar)Can you confirm the fix?
-
8. Re: Teiid 8.5 Build fails
rakeshsagar Oct 2, 2013 12:45 AM (in response to shawkins)The fix works fine and the test cases pass now.
But there are more test failures in the test-integration-common module.
Attached are the surefire reports.
-
SureFireReports.zip 199.1 KB
-
-
9. Re: Teiid 8.5 Build fails
shawkins Oct 2, 2013 10:01 AM (in response to rakeshsagar)Yes, those are also due to newline character handling. The test logic now no longer uses a PrintWriter to prevent the results from being platform dependent.
-
10. Re: Teiid 8.5 Build fails
rakeshsagar Oct 3, 2013 12:00 AM (in response to shawkins)Should I open a new JIRA issue for this?
-
11. Re: Teiid 8.5 Build fails
shawkins Oct 3, 2013 6:41 AM (in response to rakeshsagar)No, it's already been changed.
-
-
13. Re: Teiid 8.5 Build fails
rakeshsagar Oct 4, 2013 3:48 AM (in response to shawkins)Thanks Steve. All the test cases are fixed. There is one more test case failing.
test_pg_client_encoding of org.teiid.transport.TestODBCSocketTransport.
Do we need to set any env parameter for the encoding?
Attached is the sure fire report.
-
14. Re: Teiid 8.5 Build fails
shawkins Oct 4, 2013 6:05 AM (in response to rakeshsagar)That seems like an odd platform variation. UNICODE and UTF8 are equivalent for postgresql, so there isn't a concern. I'll debug a little in their driver.
Steve