mondrian schema学习过程
===== 维度的排序通过 ordinalColumn
<Level>
=====> <MemberFormatter>
<Script> return member.getName().substr(member.getName().lastIndexOf('_')+1);
</Script>
</MemberFormatter>
</Level>
===== 级别 <level>元素中包含了一个> Degenerate dimensions
===== 内连表
<Dimension> <Hierarchy hasAll="true" primaryKey="is_trade_repeat" allLevelName="(汇总)" allMemberName="汇总数据">
<InlineTable> <ColumnDefs>
<ColumnDef> <ColumnDef name="is_trade_repeat_name" type="String" />
</ColumnDefs>
<Rows>
<Row>
<Value> <Value column="is_trade_repeat_name">新客户</Value>
</Row>
<Row>
<Value> <Value column="is_trade_repeat_name">老客户</Value>
</Row>
</Rows>
</InlineTable>
<Level> </Hierarchy>
</Dimension>
===== view 举例
<Dimension name="query" caption="搜索关键字" foreignKey="query">
<Hierarchy> <View alias="dim_query">
<SQL> select DISTINCT `query` as `query`
from> </Hierarchy>
</Dimension>
=====
format
http://ask.analytical-labs.com/questions/2855/using-formatstring-strange-results
<CellFormatter>
<Script> <![CDATA[var n = value,
> d = ",",
> s = n < 0 ? "-" : "",
> j = (j = i.length) > 3 ? j % 3 : 0;
return> </Script>
</CellFormatter>
format cell 是不能用的
Datatype 属性有 String、Integer、Numberic、Boolean、Date、Timestamp 等值,默认的是
Numberic ,count 和 distinct-count 度量的默是 Integer 类型。
可以选择该属性的值为 true 或 false(默认值)
素的操作将不会被执行。
The <Level> element allows specifying the optional attribute "approxRowCount". Specifying approxRowCount can improve performance by reducing the need to determine level, hierarchy, and dimension cardinality. This can have a significant impact when connecting to Mondrian via XMLA.
<CalculatedMember name="Profit" dimension="Measures">
<Formula>[Measures].[Store Sales] - [Measures].[Store Cost]</Formula>
<CalculatedMemberProperty> </CalculatedMember>
<CalculatedMember name="NetSalesPercent"
dimension="Measures"> <CalculatedMemberProperty name="FORMAT_STRING" value="0.0%"/>
<Dimension name="time" caption="时间" >
<Hierarchy> <Table name="dim_thedate"/>
<Level> <Level name="month" caption="年月" column="month" type="String" uniqueMembers="false"/>
<Level> </Hierarchy>
<Hierarchy> <Table name="dim_thedate"/>
<Level> <Level name="week" caption="年周" column="week" type="String" uniqueMembers="true" />
<Level> </Hierarchy>
</Dimension>
衍生度量不能提前生成./的
自己度量/本身度量的一定要对应到店铺.可以放在御膳房中
直接在返回值里面改
<CalculatedMember name="auction_num_rate" caption="销售数量占比" dimension="Measures" formula="([Measures].[same_day_auction_num] /sum([Measures].[same_day_auction_num],shop.[shop_id].CurrentMember))">
<CalculatedMemberProperty> </CalculatedMember>
all交叉没作用
度量type默认为number sum
打包:addr:10.200.187.68
/ucenter/deploy
sh package.sh
sh start.sh
日志:/data/ucenter/cs/resin-goods-web/log/
最终读:/data/ucenter/cs/Goods.xml
http://svn.yunat.com/dev2/goods/goods-web/trunk
搜索的话: 轴建模是mdx的函数的事情,搜索的话就是 :mdx row percenty
aggruate 的用处,你妹的
defualtmesure 的用处
formatString
allMemberName
allLevelName
type
property是不显示的
看东西一定要静下心,特别是新东西
measuresCaption?
approxRowCount?
highCardinality?
MondrianException: Mondrian Error:No> =====: 对于衍生维度
exsample:
<Measure> <MeasureExpression>
<SQL> case when `tb_auction_source`.`iuv`=0 then 0 else `tb_auction_source`.`alipay_winner_num`/`tb_auction_source`.`iuv` end
</SQL>
</MeasureExpression>
</Measure>
<CalculatedMember> <CalculatedMemberProperty name="FORMAT_STRING" value="Percent" />
</CalculatedMember>
讲解:
formula解决的是在维度选择过程中生成的度量.
sql解决的是一开始就确定的度量.
- 顶
- 0
- 踩
- 0
- 上一篇解决问题的思路
- 下一篇sql 几个细节