[ monetdb-Bugs-2038906 ] XML attribute generation

[SourceForge.net]

Bugs item #2038906, was opened at 2008-08-05 15:42
Message generated for change (Settings changed) made by stmane
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2038906&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 0.24
Status: Open
>Resolution: Fixed
Priority: 5
Private: No
Submitted By: Henning Rode (hrode)
Assigned to: Stefan Manegold (stmane)
Summary: XML attribute generation

Initial Comment:
the following xquery constantly fails on my system.

xquery:

let $rscs := doc("c.xml")//resource
for $c in $rscs//concept
return <concept score="{$c//score/text()}">{$c/text()}</concept>


output:
.
.
.
<concept score="0.0844787">sports</concept>,
<concept score="0.0836574">face</concept>,
<concept score="0.032173">soccer</concept>,
<concept score="0.021382">grass</concept>,
<concept!ERROR: XML Generation: tmp_6313 BAT does not have a 73530 head.
MAPI  = monetdb@localhost:50000
QUERY = let $rscs := doc("/export/scratch0/henning/c.xml")//resource
ERROR = !ERROR:
        !ERROR: XML Generation: NULL BAT has a 73530 head, but tail is NULL.
        !ERROR:
        !ERROR: xquery_print_result_main: operation failed.

strange enough, it works fine on smaller instances of c.xml (if i only use half of the data for instance). the complete document c.xml is about 16MB (1MB compressed) and i can provide it for testing.




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

>Comment By: Stefan Manegold (stmane)
Date: 2008-10-11 12:34

Message:
fixed in CVS:

there was a major design flaw in the false-hit removal in module mkey's
ds_link() MIL proc

(
could not port fix to M5, as the ds_link MIL proc
apparently has not been ported/propagated to C code in
MonetDB5/src/modules/kernel/mkey.mx
)

THANKS to Jan for the detailed analysis to locate the origin of this bug!

Still need to adjust / trim-down Hennings test case to turn it into a
suitable test.


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

Comment By: Stefan Manegold (stmane)
Date: 2008-10-09 10:43

Message:
Henning,

could you provide me with (access to) you "c.xml" to reproduce, locate and
hopefully fix this bug?

Thanks!

Stefan


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

Comment By: Henning Rode (hrode)
Date: 2008-09-05 11:45

Message:
Logged In: YES
user_id=909189
Originator: YES

i told it stefan already, but just to keep it recorded here... the bug is
not yet solved. i still get the exactly same error.

-henning

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

Comment By: Stefan Manegold (stmane)
Date: 2008-09-05 11:25

Message:
Logged In: YES
user_id=572415
Originator: NO

Jan, Henning,

the (maybe not that) related
http://sourceforge.net/tracker/index.php?func=detail&aid=2071638&group_id=56967&atid=482468
just works for me.

I have not yet been able to locate the problem in ds_link() reported
here.

Of course, a double-check is always welcome...

Stefan


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

Comment By: Jan Rittinger (tsheyar)
Date: 2008-09-05 10:58

Message:
Logged In: YES
user_id=993208
Originator: NO

It seems Stefan fixed the related bug 2071638. Henning could you verify if
your problem here is fixed as well.

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

Comment By: Stefan Manegold (stmane)
Date: 2008-08-24 17:21

Message:
Logged In: YES
user_id=572415
Originator: NO

*might* be related to
[ 2071638 ] mclient output is limited?
http://sourceforge.net/tracker/index.php?func=detail&aid=2071638&group_id=56967&atid=482468


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

Comment By: Jan Rittinger (tsheyar)
Date: 2008-08-06 12:10

Message:
Logged In: YES
user_id=993208
Originator: NO

It turned out to be an overflow in ds_link() (which the twig construction
uses for the correct handling of attributes). As soon as more than 65536
results (should) appear ds_link() produces garbage.

print(a0000.slice(0,10), a0008);
#-------------------------#
# h     t       t         # name
# void  oid     oid       # type
#-------------------------#
[ 0@0,    1@0,    0@0     ]
[ 1@0,    2@0,    0@0     ]
[ 2@0,    3@0,    0@0     ]
[ 3@0,    4@0,    0@0     ]
[ 4@0,    5@0,    0@0     ]
[ 5@0,    6@0,    0@0     ]
[ 6@0,    7@0,    0@0     ]
[ 7@0,    8@0,    0@0     ]
[ 8@0,    9@0,    0@0     ]
[ 9@0,    10@0,   0@0     ]
[ 10@0,   11@0,   0@0     ]
print(a0007.fetch(0).slice(0,20), a0007.fetch(1));
#-------------------------#
# h     t       t         # name
# void  oid     oid       # type
#-------------------------#
[ 0@0,    1@0,    0@0     ]
[ 1@0,    1@0,    1@0     ]
[ 2@0,    2@0,    0@0     ]
[ 3@0,    2@0,    1@0     ]
[ 4@0,    3@0,    0@0     ]
[ 5@0,    3@0,    1@0     ]
[ 6@0,    4@0,    0@0     ]
[ 7@0,    4@0,    1@0     ]
[ 8@0,    5@0,    0@0     ]
[ 9@0,    5@0,    1@0     ]
[ 10@0,   6@0,    0@0     ]
[ 11@0,   6@0,    1@0     ]
[ 12@0,   7@0,    0@0     ]
[ 13@0,   7@0,    1@0     ]
[ 14@0,   8@0,    0@0     ]
[ 15@0,   8@0,    1@0     ]
[ 16@0,   9@0,    0@0     ]
[ 17@0,   9@0,    1@0     ]
[ 18@0,   10@0,   0@0     ]
[ 19@0,   10@0,   1@0     ]
[ 20@0,   11@0,   0@0     ]

a0016 := ds_link (a0000.slice(0,10), a0007.fetch (0).slice(0,20),
a0008.slice(0,10), a0007.fetch (1).slice(0,20));
print (a0016.count());
[ 11 ]
a0016 := ds_link (a0000.slice(0,65000), a0007.fetch (0).slice(0,130000),
a0008.slice(0,65000), a0007.fetch (1).slice(0,130000));
print (a0016.count());
[ 65001 ]
a0016 := ds_link (a0000.slice(0,70000), a0007.fetch (0).slice(0,140000),
a0008.slice(0,70000), a0007.fetch (1).slice(0,140000));
print (a0016.count());
[ 78931 ]
a0016 := ds_link (a0000, a0007.fetch (0), a0008, a0007.fetch (1));
print (a0016.count());
[ 168899 ]


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

Comment By: Stefan Manegold (stmane)
Date: 2008-08-05 16:28

Message:
Logged In: YES
user_id=572415
Originator: NO

Might be a bug in the element construction in the Algebra back-end ---
assigning to Jan R.


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

Comment By: Henning Rode (hrode)
Date: 2008-08-05 16:25

Message:
Logged In: YES
user_id=909189
Originator: YES

yes, I tested the MPS back-end as well, and also there it works fine.

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

Comment By: Stefan Manegold (stmane)
Date: 2008-08-05 16:17

Message:
Logged In: YES
user_id=572415
Originator: NO

Assuming that you used the (default) "Algebra" back-end, did you by any
chance also try the (old) "milprint_summer" back-end?
If so, what was the effect?
If not, could you please try (simply run your query with `mclient -lx
-g`), and report on the effect?


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

Comment By: Henning Rode (hrode)
Date: 2008-08-05 15:47

Message:
Logged In: YES
user_id=909189
Originator: YES

I tested also to remove either the score attribute construction or the
element content construction. In both cases the error does *not* occur.

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

You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2038906&group_id=56967

-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Monetdb-bugs mailing list
Monetdb-bugs@...
https://lists.sourceforge.net/lists/listinfo/monetdb-bugs