Ciao Martin,
please, read below...

On 8/31/07, Martin Desruisseaux <[EMAIL PROTECTED]> wrote:
> Hello Simone
>
>
> Simone Giannecchini a écrit :
> > my fault I started a philosophycal discussion with these thread,
>
> Don't worry, I was not silent because I was upset but only because I was
> finishing a work in progress. It is normal to have different views in an open
> source project, and occasional philosophical discussions are probably part of
> the game for a long-term project. I'm trying myself to learn which beliefs I
> should stand on, and which ones are exageration of mine that I should give up.
>
>
> >>   Boolean requireGeophysicsType
> >>
> >> is really an implementation trick; a public API should rather use an
> >> enumeration (this work was started with ViewType but not finished).
> >
> > I agree, but this is the implementation we have right now even if it
> > may change.
> > This field could be pretty useful for subclasses which may want to change 
> > their
> > behaviour or do some preprocessing before doing the actual work. One 
> > example is
> > applying operations that may want to expand paletted coverage. Sometimes we 
> > can
> > apply them on the non-geo view some other on the geo view depending on 
> > various
> > factors like interpolations and the like. Conclusion is I would leave
> > it protected.
>
> Is the current state "This could be useful" or is it "We need it right now for
> XYZ implementation"? If the need is not yet there and just a possibility that
> someone may need it, maybe we could wait for the first of the following event 
> to
> occur: 1) The need occured, or 2) We replaced the implementation trick by a 
> API.
>
I ma pretty sure this would be useful as soon as I start refactoring
scale, subsample and the like as subclasses of OperationJAI. I will
need the possibility to adapt/decorate JAI params as well as these
facilities.


> We could do the following: if we need 'extractSources' and 'postProcessResult'
> righ now, we could change the return type from Boolean to ViewType.
>

Go ahead on that, so that we can start talking about the ViewType work..

> > <Adding @deprecated Decorators>
> > I would not use the deprecation tag until we have  a replacement for
> > these classes
>
> I'm going to take a coffee and look immediately when I'm back if we could 
> merge
> the decorator with their parent class.
>
>        Martin
>


Thanks,
Simone.

-- 
-------------------------------------------------------
Eng. Simone Giannecchini
President /CEO GeoSolutions S.A.S.
Via Carignoni 51
55041  Camaiore (LU)
Italy

phone: +39 0584983027
fax:      +39 0584983027
mob:    +39 333 8128928


http://www.geo-solutions.it

-------------------------------------------------------

-------------------------------------------------------------------------
This SF.net email is sponsored by: Splunk Inc.
Still grepping through log files to find problems?  Stop.
Now Search log events and configuration files using AJAX and a browser.
Download your FREE copy of Splunk now >>  http://get.splunk.com/
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to