[Mondrian] RE: Default Measure Docs

Julian Hyde julianhyde at speakeasy.net
Fri Apr 20 13:49:41 EDT 2007


I second Will's comment.

Also, you use member.getName().equalsIgnoreCase() in a couple of places.
I think you should use unique name, and use case-sensitive match similar
to Util.equals(String, String).

Is it possible to use a calculated member which does not belong to the
measures level as the default member? E.g. [Measures].[Foo].[Bar]. I
think so, and for this reason, the measure name in the schema file
should be fully-qualified and of the correct case according to their
case-sensitivity settings.

Can you also add a couple of negative tests: what happens if the defn of
the calc member is cyclic; where the default member of a cube or virtual
cube does not exist; and where the default member is of the wrong case.

For the latter, PropertiesTest has a good pattern to follow:

        if (MondrianProperties.instance().CaseSensitive.get()) {
            // should fail
        } else {
            // same testcase should succeed

All in all, this looks like a very useful change. Thank you!


> -----Original Message-----
> From: Will Gorman [mailto:wgorman at pentaho.org] 
> Sent: Friday, April 20, 2007 6:34 AM
> To: Thiyagu Palanisamy
> Cc: Julian Hyde
> Subject: Default Measure Docs
> Thiyagu,
> Would it be possible to add a sentence or two in the 
> schema.html file on
> why a user would want to change the default measure?
> Thanks!
> Will

More information about the Mondrian mailing list