Configuring the underlying Java SDO parser

[Guy Pardon <gu...@atomikos.com>]

Hi,

Is there a way (when iterating) to avoid that the entire XML file is  
read in RAM at once? We want to read a large XML file with efficient  
memory use (like Stax should allow).

Thanks
Guy

---------------------------------------------------------------------
To unsubscribe, e-mail: tuscany-user-unsubscribe@ws.apache.org
For additional commands, e-mail: tuscany-user-help@ws.apache.org

Re: Configuring the underlying Java SDO parser

[kelvin goodson <ke...@thegoodsons.org.uk>]

Guy,

  no I'm afraid we don't have any lazy parsing capabilities at the moment.
Are you accessing the elements of the file through the data graph in the
order they appear in the instance document?  There is a way behind the
scenes that I think we could begin to provide this kind of capability,  by
creating an implementation of the InternalEObject.EStore  interface
http://download.eclipse.org/modeling/emf/emf/javadoc/2.2.3/org/eclipse/emf/ecore/InternalEObject.EStore.html
so that gets and sets on a DataObject  using such a store based
implementation could delegate operations to its backing store, thereby
permitting some level of control of the timings of Property value
instantiation.  This would be a good chunk of work to investigate, design
and develop.  Is this something you'd be interested in helping with -- or
anyone else for that matter :-)

Kelvin

On 21/01/2008, Guy Pardon <gu...@atomikos.com> wrote:
>
> Hi,
>
> Is there a way (when iterating) to avoid that the entire XML file is
> read in RAM at once? We want to read a large XML file with efficient
> memory use (like Stax should allow).
>
> Thanks
> Guy
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: tuscany-user-unsubscribe@ws.apache.org
> For additional commands, e-mail: tuscany-user-help@ws.apache.org
>
>