When it comes to MDX queries, you might get more help in the newsgroup
called microsoft.public.sqlserver.olap.
What are the names /hierarchies of the dimensions you want to return?
What measures do you want to see?
Do you want to see number of Fatal Injuries per weather?
From your example, it looks to me like you just want something like
SELECT { [Measures].[Fatal Crashes] } ON COLUMNS,
{
[CrashPerson].[Person Injury Class].[Person Injury Class].ALLMEMBERS
} DIMENSION PROPERTIES
MEMBER_CAPTION, MEMBER_UNIQUE_NAME ON ROWS FROM [SDMTest]
Then you use a matrix, and create a column group for the CrashPerson field.
That should give you
...A-injury...|..B-injury..|..C-injury..|
...20...........|..20..........|..10...........|
If you want to see injuries by weather and category, you can add the weather
dimension and create a row group for that.
Kaisa M. Lindahl Lervik
[quoted text, click to view] "Tenchy" <Tenchy@discussions.microsoft.com> wrote in message
news:A8449F38-B807-4430-AF4C-10207DC3E5ED@microsoft.com...
> Thats my problem, i dont know what the mdx code would be to do that.
> Because
> i can do mdx in reporting services in the dataset as a calculated member.
> But
> the code i put in there gives me all kinds of syntax errors. This is the
> code
> i have right now.
>
> SELECT { [Measures].[Vehicle Count], [Measures].[Injury Crashes],
> [Measures].[Non-Fatal Injuries], [Measures].[Fatal Injuries],
> [Measures].[CrashCount], [Measures].[Fatal Crashes] } ON COLUMNS, {
> ([Crash].[Weather].[Weather].ALLMEMBERS * [CrashPerson].[Person Injury
> Class].[Person Injury Class].ALLMEMBERS ) } DIMENSION PROPERTIES
> MEMBER_CAPTION, MEMBER_UNIQUE_NAME ON ROWS FROM [SDMTest] CELL PROPERTIES
> VALUE, BACK_COLOR, FORE_COLOR, FORMATTED_VALUE, FORMAT_STRING, FONT_NAME,
> FONT_SIZE, FONT_FLAGS
>
> I just want to have the person injury classes, in seperate columns,
> instead
> of the total of all of them together, that way when i put them in my
> table,
> the user can see the (A-Injuries) fatal crashes (b-Injury) Fatal
> crashes...
> and so on. And i dont have to have 15 different datasets with the same
> fields and just one different Filter. It has to be possible.