Re: [Geotools-devel] gt-process: adding multiple algorithms with a single factory

2009-06-15 Thread Michael Bedward
2009/6/16 Graham Davis wrote: > While I'm listed as the maintainer of this module, no major work has > been done in it for about year.  I don't think many people are using > this API yet so changing it to improve it probably won't affect many people. > Except me :-) I've been (very slowly) workin

Re: [Geotools-devel] gt-process: adding multiple algorithms with a single factory

2009-06-15 Thread Graham Davis
I would go with the first for the same reasons already listed. The API was looking for feedback from others on how it would be used, so it's not really nailed down. While I'm listed as the maintainer of this module, no major work has been done in it for about year. I don't think many people a

Re: [Geotools-devel] gt-process: adding multiple algorithms with a single factory

2009-06-15 Thread Justin Deoliveira
> The solution suggested in the thread was to make a factory > that can bear multiple algorithms. I see two paths ahead: > - change the whole ProcessFactory definition so that >it returns a list of process names, and takes a name >parameter in each call (api break) > - create a new Composi

Re: [Geotools-devel] gt-process: adding multiple algorithms with a single factory

2009-06-13 Thread Jody Garnett
> The solution suggested in the thread was to make a factory > that can bear multiple algorithms. I see two paths ahead: > - change the whole ProcessFactory definition so that >   it returns a list of process names, and takes a name >   parameter in each call (api break) This is fine; the api is n