Bugs item #1928084, was opened at 2008-03-28 17:11
Message generated for change (Comment added) made by stmane
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1928084&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: PF/runtime
Group: Pathfinder CVS Head
>Status: Closed
Resolution: Fixed
Priority: 6
Private: No
Submitted By: Jan Rittinger (tsheyar)
Assigned to: Peter Boncz (boncz)
Summary: XQ: Ancestor step with debug=10 broken
Initial Comment:
The following query:
doc("foo.xml")//text()/ancestor::b
does return the correct result (<b>textnode</b>) if run with --debug=0. With
--debug=10 as during testing no results are returned (-- see also
tests/XQuery/step_attr_nametest).
The problem arises in loop_lifted_ancestor_step_with_nsloc_test().
$ cat foo.xml
<a foo="bar"><b>textnode</b><d><e/><f/></d><c foo2="bar"/></a>
----------------------------------------------------------------------
>Comment By: Stefan Manegold (stmane)
Date: 2008-06-03 13:52
Message:
Logged In: YES
user_id=572415
Originator: NO
MPS error message has been approved.
Closing.
----------------------------------------------------------------------
Comment By: Stefan Manegold (stmane)
Date: 2008-05-09 11:52
Message:
Logged In: YES
user_id=572415
Originator: NO
According to the TestWeb, the test now works fine with the ALG back-end.
With the (soon to be) deprecated MPS back-end, it says:
"
ERROR = !ERROR: path steps are only supported starting from non-attribute
nodes
"
which IMHO is correct, since MPS did, does and will not support path steps
starting from attribute nodes, right?
If so, we should consider approving the correct/expected error message as
MPS-specific output and close this bug report (again).
----------------------------------------------------------------------
Comment By: Stefan Manegold (stmane)
Date: 2008-05-08 23:45
Message:
Logged In: YES
user_id=572415
Originator: NO
I guess, I found the bug:
The same ATOMcmp(TYPE_oid,x,y) fix that Peter applied for NIL-aware
comparisons of OIDs in BATfetchjoin also needs to be applied a couple of
lines further on the gdk_relop.mx .
If SF allows me to check my fix in before testing starts (and SF allows
testings to checkout the latest version) the TestWeb will tell us more
tomorrow morning ...
----------------------------------------------------------------------
Comment By: Stefan Manegold (stmane)
Date: 2008-05-08 23:07
Message:
Logged In: YES
user_id=572415
Originator: NO
Re-opening, as the test apparently (according to nightly testing logs)
never worked correctly (with the default GDKdebug=10) ...
----------------------------------------------------------------------
Comment By: Stefan Manegold (stmane)
Date: 2008-05-08 18:22
Message:
Logged In: YES
user_id=572415
Originator: NO
As Jan just enlightened me via pm:
The original bug report already said "see also
tests/XQuery/step_attr_nametest" ...
However, the repective test still (again?) fails, both with the MPS and
the ALG beack-end;
cf.,
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTestsG103/GNU.64.64.d-Fedora8/tests_XQuery/step_attr_nametest.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora8/tests_XQuery/step_attr_nametest.out.00.html
http://monetdb.cwi.nl/testing/projects/monetdb/Current/pathfinder/.mTests103/GNU.64.64.d-Fedora8/tests_XQuery/step_attr_nametest.err.00.html
...
----------------------------------------------------------------------
Comment By: Stefan Manegold (stmane)
Date: 2008-05-08 16:14
Message:
Logged In: YES
user_id=572415
Originator: NO
Is this problem covered by any existing test?
Otherwise, we should consider adding one to monitor the fix ...
----------------------------------------------------------------------
Comment By: Peter Boncz (boncz)
Date: 2008-03-30 01:58
Message:
Logged In: YES
user_id=591107
Originator: NO
thanks for the bug. turned out to be a very deep MonetDB problem coming
from oids being stored as either int/lng but being treated as unsigned.
Especially nil may cause surprises, since as an oid it seems > 0, however
as a signed integer it is < 0.
This particular code happened to produce a bat with an oid column starting
with nil. With debug=10, the sorting order would be detected on this
column, such that in the leftfetchjon code a wrong direct oid comparison on
the values was done. One should use ATOMcmp(TYPE_oid,x,y) instead.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1928084&group_id=56967
-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Monetdb-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/monetdb-bugs
