Re: [Repast-developer] Getting rid of warnings? Also Repast S javadoc?

[Miles P. <mil...@gm...>]

Oh and not to be defensive, but to amplify the point about generics  
below, I am sure you understand that there is a significant amount of  
new work here including exploration of various levels of  
abstraction...it seems to me that the team has put a lot of effort  
into that, and that iun terms of code quality (which I think will be  
very high, if I do say so for others if not myself) the important  
thing right now is to get the basic abstractions and collaborations  
right -- we will only have one chance to do some of this stuff right  
--  but as the apis are not near been frozen in the current release  
(the score stuff in particular should be regarded as exploratory/ 
pilot level) it sometimes isn't efficient to sort through the various  
fiddly bits about generics -- the case you note below being a simple  
example, but when one find oneself looking at stuff like

		List<Pair<Method, Watch>> list = watchMap.get(className);
		if (list == null) {
			list = new ArrayList<Pair<Method, Watch>>();
			watchMap.put(className, list);
		}
		list.add(new Pair<Method, Watch>(method, watch));


	protected <X> Iterable<X> getContextFilter(Context<? extends X>  
context,
	                                           Predicate<X> filter) {
		return new IterableAdaptor<X>(new FilterIterator(context.iterator 
(), filter));
	}

certainly not the most involved examples -- one begins to wonder  
wether generics are on balance worth it..I suppose we've decided more  
or less that they are, but may be backing off of some of it a bit.  
Feedback most welcome but again prob. too early in the present  
release for you to have a strong sense of where we are headed w/ some  
low-level implementation details.

On Jan 29, 2007, at 9:48 AM, Miles Parker wrote:

>
> Hi Pimin,
>
> A bit tongue in cheek, but perhaps you should simply set your  
> generics warnings to "ignore"-- that's what I've done which I'm  
> sure gives you enormous confidence in our code quality. ;)  
> Seriously, leaving aside the issues involved in maintaining Java  
> implementation of generics, remember that what is our there is  
> early alpha. I am sure we will make a pass through all the warnings  
> as we approach our release candidate -- we will also certainly do  
> some major QA work, including increasing unit testing coverage,  
> providing more automated builds, untangling remaining code and  
> plugin dependencies, significant refactorings and so on...whew!
>
> best,
>
> Miles
>
> On Jan 29, 2007, at 9:10 AM, Pimin Konstantin Kefaloukos wrote:
>
>> Hi developers,
>>
>> This might not have top priority, but couldn't we get rid of some of
>> the many warnings that are generated by the current code?
>>
>> And example is the following source-code:
>>
>>
>> package repast.context;
>> public class DefaultContextFactory<T> implements ContextFactory<T> {
>> ...
>>  public Context<T> createContext(Object contextId) {
>>   // why not: return new DefaultContext<T>(contextId); ?
>>   return new DefaultContext(contextId);
>>  }
>> }
>>
>> Could we not include the type-parameter and remove some of the  
>> "noise"
>> from the compiling process.
>>
>> Also I can't find a javadoc for Repast S, has one not been made?
>>
>>
>>
>>
>> Best regards,
>>
>>
>>
>> Pimin Konstantin Kefaloukos,
>> Computer science department, University of Copenhagen.
>> Anders Henriksens Gade 3, 2. t.h.
>> 2300 Copenhagen, Denmark.
>>
>> --------------------------------------------------------------------- 
>> ----
>> Take Surveys. Earn Cash. Influence the Future of IT
>> Join SourceForge.net's Techsay panel and you'll get the chance to  
>> share your
>> opinions on IT & business topics through brief surveys - and earn  
>> cash
>> http://www.techsay.com/default.php? 
>> page=join.php&p=sourceforge&CID=DEVDEV
>> _______________________________________________
>> Repast-developer mailing list
>> Rep...@li...
>> https://lists.sourceforge.net/lists/listinfo/repast-developer
>