cybernelly at gmail.com
Wed May 30 06:39:51 EDT 2007
actually, we have a bit obsolete version of Mondrian used in our
development. In our version Validator interface has convert() method which
calls FunUtil.convert(). I was wondering why convert() has real
implementation while canConvert() is a stub... In newest Mondrian there is
Original reason was the need of creating an expression containing IIF with
comparison operators at schema load time. Original implementation of simple
validator didn't allow doing it (because of ambiguity between overloaded
operations). When I had added canConvert() delegation to FunUtil it started
to work for us. I haven't touched any other methods.
Maybe I have missed some other ways to implement hardcoded MDX expressions
at startup, haven't I?
From: mondrian-bounces at pentaho.org [mailto:mondrian-bounces at pentaho.org] On
Behalf Of Julian Hyde
Sent: Tuesday, May 29, 2007 11:05 PM
To: 'Mondrian developer mailing list'
Subject: RE: [Mondrian] Util.createSimpleValidator()
That validator is basically a mock object - because a validator is
needed in order to convert a UnresolvedFunCall to a ResolvedFunCall.
It's only used in one place.
If you add a call to canConvert, I have a feeling that the validator
will need to know the types of the arguments - which means you'll need
to implement other methods in the validator.
> -----Original Message-----
> From: mondrian-bounces at pentaho.org
> [mailto:mondrian-bounces at pentaho.org] On Behalf Of Anton Nikitin
> Sent: Tuesday, May 29, 2007 6:45 AM
> To: 'Mondrian developer mailing list'
> Subject: [Mondrian] Util.createSimpleValidator()
> Are there any objective reasons why mondrian.olap.Util.
> createSimpleValidator() method returns validator which
> doesn't implement
> canConvert() properly (the same way as StackValidator does - by using
> Mondrian mailing list
> Mondrian at pentaho.org
Mondrian mailing list
Mondrian at pentaho.org
More information about the Mondrian