Lloyd, the issue is that too many fragments are being kept in scope and thus
held in the expanded-tree cache. If you are not doing anything with the
results that requires the full document (and it doesn't look like you are,
since you are outputting exclusion elements), you can create a copy of the
reduced element to reduce your cache usage:
for $node in /clinical_study/eligibility/exclusion
return element {fn:node-name($node)} {$node/(@*|node())}
This will load a fragment to produce each copied element; after the element
is created, the fragment falls out of scope and does not need to be kept in
the tree cache.
Another option, which may be faster than creating copies but would require
multiple queries to MarkLogic Server, is to paginate your results:
(/clinical_study/eligibility/exclusion)[1 to 100]
James
_____ From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Lloyd Harding Sent: Monday, June 02, 2008 4:37 AM To: [email protected] Subject: [MarkLogic Dev General] XDMP-EXPNTREECACHEFULL error I'm trying to using Oxygen extract all data from our corpus. Code: (/clinical_study/eligibility /exclusion) Error: Description: XDMP-EXPNTREECACHEFULL: Expanded tree cache full on host xxxxxxxxxx <http://server2.ctrx.org/> How can I fix this? I do not think this is a volume issue. Suspect its more of a too many nodes issue. lloyd -- "Technical skill is mastery of complexity while creativity is mastery of simplicity "E.C. Zeeman, mathematician, b.1923 Thank You Lloyd Harding 218-962-3202 [EMAIL PROTECTED]
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ General mailing list [email protected] http://xqzone.com/mailman/listinfo/general
