An MDX (a language dealing with multidimensional data, in particularly dealing with the cube metaphor) statement can be translated to an SQL (a well-known language for the relational model) statement - Mondrian is one example solution which provides such translation. I was wondering if there are any available solutions to provide a translation in the other direction, i.e. from SQL to MDX. If not, what would be involved in creating such a functionality?

In a [Mondrian](http://community.pentaho.com/projects/mondrian/) Schema, I have a fact table for Sales, a dimension table Time and a dimension table Client. I need to create a [Saiku](http://www.meteorite.bi/products/saiku) report where i can show how many clients joined per year. I have a the Time dimension and a join_date in Client (which DO NOT link to the Time Dimension key, is just a date field),. If I create a cube adding Join_date as a Level of the Client Table, I wouldn't know how to filter that latter in the Saiku. Does anyone know how is the best way to add JOIN_DATE on client? I just need a report: Clients that have joined to the company by year. And how should ` work in cubes? When i tested using this `, it was not visible in Saiku...

How do i filter, in a Mondrian MDX Query, just the FACTS that have ocurred in a Range of Months over last 3 years? I need to make it dynamic so i can use parameters latter, so i need the following: Month Range over last 3 years, Where StartMonth and EndMonth is given by user by parameters. .[January], .[February], .[March] .[January], .[February], .[March] .[January], .[February], .[March] Time dim have Year, Half Year, Quarter, Month, Week and Day level. I have tryed this, but i get an error Mondrian Error:Cannot deduce type of set WITH set [MesesFiltrados] as DESCENDANTS( FILTER(LastPeriods (3, StrToMember("[Data].[" || TRIM(Str(Int(YEAR(NOW())))) || "]") ), [Data].[Mes].currentmember in ([Data].Mes.[${StartMonth}]:[Data].Mes.[${StartMonth}]) ) ) Select {[Escritorio]} on columns, { MesesFiltrados } on rows From Desempenho Tks