Bugs item #1763495, was opened at 2007-07-30 12:44
Message generated for change (Comment added) made by stmane
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1763495&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 0.18
>Status: Open
>Resolution: None
Priority: 5
Private: No
Submitted By: Wouter Alink (vzzzbx)
Assigned to: Wouter Alink (vzzzbx)
Summary: XQ: illegal offsets in swizzle (2)
Initial Comment:
(today's nightly build from the stable branch, which includes sjoerd's recent
fix for a similar problem)
given the attached script, the error:
!ERROR: [swizzle]: tmp_2750 contains illegal offsets
is returned.
the gzipped version of the attached script including source docs is too large
for sourceforge (>256kb). i've mailed it to sjoerd.
----------------------------------------------------------------------
>Comment By: Stefan Manegold (stmane)
Date: 2007-08-03 13:28
Message:
Logged In: YES
user_id=572415
Originator: NO
re-opend.
we should indeed try to either return an empty result or issue a proper
error message in such cases.
----------------------------------------------------------------------
Comment By: Wouter Alink (vzzzbx)
Date: 2007-08-03 12:13
Message:
Logged In: YES
user_id=621590
Originator: YES
what sjoerd concluded is correct. i made a mistake in reporting this bug.
i am able to reproduce the bug, but the script that i've attached was meant
to use correct nids. it was perhaps caused by not starting with an empty
dbfarm.
however, i do think that the behaviour of monetdb is counter intuitive.
the query 'doc("x.xml")/id("1234567890")' gives me an empty result (not a
swizzle error). in case when this is the 'into' part of an update
statement, monetdb could return the in my opinion correct error:
exactly-one() expected exactly one result.
----------------------------------------------------------------------
Comment By: Stefan Manegold (stmane)
Date: 2007-08-03 09:48
Message:
Logged In: YES
user_id=572415
Originator: NO
Apparently, this turned out to be a pilot error rather than a bug.
Hence, no (additional) test required.
Closing.
----------------------------------------------------------------------
Comment By: Sjoerd Mullender (sjoerd)
Date: 2007-08-03 09:37
Message:
Logged In: YES
user_id=43607
Originator: NO
What do you want tested here? As far as I can see, the bug is the use of
pf:nid() that refers to a NID which is not actually in the document.
I could add Wouter's test, but it is more-or-less the same as the previous
swizzle bug which I fixed and which does have a test.
----------------------------------------------------------------------
Comment By: Stefan Manegold (stmane)
Date: 2007-08-02 23:13
Message:
Logged In: YES
user_id=572415
Originator: NO
As usually, we should add a test for this one.
Wouter? Peter? Stefan?
----------------------------------------------------------------------
Comment By: Peter Boncz (boncz)
Date: 2007-08-02 21:02
Message:
Logged In: YES
user_id=591107
Originator: NO
can you close this now, Wouter?
----------------------------------------------------------------------
Comment By: Peter Boncz (boncz)
Date: 2007-08-02 21:01
Message:
Logged In: YES
user_id=591107
Originator: NO
can you close this now, Wouter?
----------------------------------------------------------------------
Comment By: Sjoerd Mullender (sjoerd)
Date: 2007-07-31 17:35
Message:
Logged In: YES
user_id=43607
Originator: NO
The reason for the [swizzle] failing seems to be that the last document
that is used in scenario.xq, xiraf32017.xml, internally known as
prepared32015.data, in the very first <item> already, has nid="31351", a
NID which does not occur in the wrappers.xml document.
When you use the id() function, you have to be very careful.
For now I would call this a Pilot Error.
----------------------------------------------------------------------
Comment By: Sjoerd Mullender (sjoerd)
Date: 2007-07-31 09:18
Message:
Logged In: YES
user_id=43607
Originator: NO
That's indeed how it starts. But after that it has a newline and
<device type="unknown" name="pc01"><image type="LocalFolder"
location="/scratch/
weka-3-4-10" xstart="0" xend="31417045">
----------------------------------------------------------------------
Comment By: Wouter Alink (vzzzbx)
Date: 2007-07-31 08:21
Message:
Logged In: YES
user_id=621590
Originator: YES
sorry, my fault. i might have given the wrong initial document (it looks
like it was the document after the updates instead of before the updates).
document xiraf31973 should read:
<?xml version="1.0" encoding="utf-8"?>
<XIRAF version="TEST-VERSION" timestamp="Mon Jul 30 11:04:39 CEST 2007"
revision="TEST"/>
----------------------------------------------------------------------
Comment By: Sjoerd Mullender (sjoerd)
Date: 2007-07-30 16:49
Message:
Logged In: YES
user_id=43607
Originator: NO
Hmm, maybe the input you sent me is not correct:
The first insert query inserts a node <device name="pc01" type="unknown"/>
into the document. Then the next insert query inserts into
doc("wrappers.xml")/XIRAF/[EMAIL PROTECTED]"pc01"] . I get an error there
because the original document already contained such a node, and the first
insert inserted another such node, so now there are two that match.
----------------------------------------------------------------------
Comment By: Sjoerd Mullender (sjoerd)
Date: 2007-07-30 16:36
Message:
Logged In: YES
user_id=43607
Originator: NO
Sorry, I hadn't seen the attachment. I've got it now.
----------------------------------------------------------------------
Comment By: Wouter Alink (vzzzbx)
Date: 2007-07-30 16:27
Message:
Logged In: YES
user_id=621590
Originator: YES
the subject reads 'swizzle 2', and you already replied to it on 12:37
local time.
----------------------------------------------------------------------
Comment By: Sjoerd Mullender (sjoerd)
Date: 2007-07-30 15:16
Message:
Logged In: YES
user_id=43607
Originator: NO
I haven't seen that promised mail...
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=1763495&group_id=56967
-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems? Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >> http://get.splunk.com/
_______________________________________________
Monetdb-bugs mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/monetdb-bugs
