Bugs item #2876948, was opened at 2009-10-12 11:07
Message generated for change (Settings changed) made by boncz
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2876948&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 general
Group: Pathfinder "stable"
>Status: Closed
Resolution: Fixed
>Priority: 1
Private: No
Submitted By: Maurice van Keulen (mvankeulen)
Assigned to: Jan Rittinger (tsheyar)
Summary: PF: error "We do not support the value NaN"
Initial Comment:
("Aug2009" stable, Linux)
Elements without textnodes are not treated correctly in predicates. The query
below gives the error "We do not support the value NaN.". If you modify the
query to return $noerror or $guarded, it works fine.
declare function ltnum($e as element(), $num as xs:double) as xs:boolean
{ if (empty($e/text())) then false() else ($e > $num) };
let $elems := (<a/>,<b/>,<c>20</c>)
let $error := $elems[. > 5]
let $noerror := for $e in $elems return if (empty($e/text())) then () else $e[.
> 5]
let $guarded := $elems[ltnum(.,5)]
return $error
----------------------------------------------------------------------
Comment By: Jan Rittinger (tsheyar)
Date: 2009-10-12 18:20
Message:
I just checked in a possible fix in the CVS HEAD (-- in my eyes its more an
extension than a fix). If testing shows no problems we might think about
back-porting it to the Stable branch.
----------------------------------------------------------------------
Comment By: Jan Rittinger (tsheyar)
Date: 2009-10-12 14:27
Message:
We cannot neither handle NaN values in the algebra nor on the MIL level. On
the MIL level we have no runtime support for Nan. On the algebra level this
would require that the following snippets 'not(23 > <a/>)' and '23 < </a>'
result in 'true' and 'false', respectively. For the algebra both snippets
result in the same plan and we give an error at runtime if we see such
behavior.
BTW: All your variants fail if you replace '<a/>' by '<a>no number</a>'.
To avoid this problem we added a rewrite that can cope with
fn:number()---the replacement for the atomization---on equality predicates.
----------------------------------------------------------------------
You can respond by visiting:
https://sourceforge.net/tracker/?func=detail&atid=482468&aid=2876948&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
