Bugs item #2893859, was opened at 2009-11-07 16:29
Message generated for change (Comment added) made by stmane
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2893859&group_id=56967

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: SQL/Core
Group: SQL "stable"
Status: Open
Resolution: None
Priority: 9
Private: No
Submitted By: Stefan Manegold (stmane)
Assigned to: Arjen de Rijke (arjenderijke)
Summary: SQL: various VIA-M tests fail

Initial Comment:
Arjen, could you please check, whether the stable output is indeed correct?

http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_Via-m/media_s14.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_Via-m/media_s16.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_Via-m/media_s18.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_Via-m/media_s19.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_Via-m/file_s10.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Stable/sql/.mTests103/GNU.64.64.d.1-Fedora10/src_test_Via-m/file_s18.out.00.html


----------------------------------------------------------------------

>Comment By: Stefan Manegold (stmane)
Date: 2009-11-16 10:07

Message:
Since the purpose of these tests is not to test the default value of
non-initialized variables, I see no reason not to simply make them
deterministic in any case by explicitly initializing the variables.
If the default value of non-initialized variables is indeed an issue, I
propose to add a separate test for that, e.g., as simple as

declare MyVar integer;
select MyVar;


----------------------------------------------------------------------

Comment By: Stefan Manegold (stmane)
Date: 2009-11-16 09:59

Message:
FYI:
The Nov2009 branch is (used as / referred to as) the "Stable" branch.


----------------------------------------------------------------------

Comment By: Arjen de Rijke (arjenderijke)
Date: 2009-11-16 09:49

Message:
The stable output is what i now expect it to be, whether it is correct is
still debated. I already approved the stable output of these tests in the
Nov2009 branch, my suggestion is to do the same for the stable branch. Than
we can continue the discussion after the release. That has become more
complicated now it appears that assertions are involved.

----------------------------------------------------------------------

Comment By: Stefan Manegold (stmane)
Date: 2009-11-14 10:26

Message:
The first 5 of the reported 6 tests seem to fail again (or still) with
nightly Stable testing.

A small analysis reveals, that they fail only with assertions disabled (as
with nightly Stable testing and in the "production" release), but work fine
with assertions enabled.

The "failure" (i.e., difference) seems to be that with assertion enabled,
the default value of a non-initialized variable is -606348325, while with
assertions disabled, either the value is 0, or a failing function
unexpectedly triggers an 0-assignment of its return value to that
variable.

In case it's the former (i.e., different default value for non-initialized
variables), and the SQL standard does not specify the default value of a
non-initialized variable, I'd propose to make these tests deterministic by
initializing the respective variable to a value that the function cannot
return, and approve the stable output accordingly.


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2893859&group_id=56967

------------------------------------------------------------------------------
Let Crystal Reports handle the reporting - Free Crystal Reports 2008 30-Day 
trial. Simplify your report design, integration and deployment - and focus on 
what you do best, core application coding. Discover what's new with
Crystal Reports now.  http://p.sf.net/sfu/bobj-july
_______________________________________________
Monetdb-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/monetdb-bugs

Reply via email to