public class QueryTag
extends javax.servlet.jsp.tagext.BodyTagSupport
QueryTag
creates a ResultCache
object and initializes
it with the MDX query. Example:Attributes are<query name="query1" resultCache="true"> select {[Measures].[Unit Sales], [Measures].[Store Cost]} on columns, CrossJoin( { [Promotion Media].[Radio], [Promotion Media].[TV], [Promotion Media].[Sunday Paper], [Promotion Media].[Street Handout] }, [Product].[Drink].children) on rows from Sales where ([Time].[1997]) </query>
name
,
resultCache
.Constructor and Description |
---|
QueryTag() |
Modifier and Type | Method and Description |
---|---|
int |
doAfterBody() |
String |
getName() |
boolean |
isResultCache() |
void |
setName(String newName)
Sets string attribute
name , which identifies this query
within its page. |
void |
setResultCache(boolean newResultCache)
Sets boolean attribute
resultCache ; if true, the query is
parsed, executed, and converted to an XML document at most once. |
doEndTag, doInitBody, doStartTag, getBodyContent, getPreviousOut, release, setBodyContent
findAncestorWithClass, getId, getParent, getValue, getValues, removeValue, setId, setPageContext, setParent, setValue
public int doAfterBody() throws javax.servlet.jsp.JspException
doAfterBody
in interface javax.servlet.jsp.tagext.IterationTag
doAfterBody
in class javax.servlet.jsp.tagext.BodyTagSupport
javax.servlet.jsp.JspException
public void setName(String newName)
name
, which identifies this query
within its page. The <transform
query>
attribute uses this.public String getName()
public void setResultCache(boolean newResultCache)
resultCache
; if true, the query is
parsed, executed, and converted to an XML document at most once. This
improves performance and consistency, but the results may become out of
date. We also need a way to prevent the cache using too much memory.public boolean isResultCache()
Copyright © 2019 Hitachi Vantara. All rights reserved.