We also had a policy of adopting GeoAPI interfaces when they were 
available; this habit broken when GeoAPI moved too quickly for us to 
keep up when the go-1 renderer was introduced. I would try and reword 
this to adopt official GeoAPI interfaces (ie that have been through the 
review process); this gives us an "out" when the experts (hi Bryce) 
point out something we obviously got wrong etc...

Jody
> Andrea Aime a écrit :
>> I guess you're referring to Sun's way of doing things, is it documented
>> anywhere? Or is it just:
>> * mark operations as deprecated for one release before removing;
>> * mark release an API has been introduced into using @since
>> * make sure interfaces do not change over time (something we may not
>>   be able to follow in practice, and we haven't followed so far either)
>
> Sun do all the above, except that I'm not aware of any deprecated 
> method they removed. So yes, this is the policy that I suggest. The 
> main difference compared to what Sun do is that we are much more 
> aggresive on removing deprecated methods.
>
> Note that even Sun does change interfaces in some occasion. For 
> example the interfaces in java.sql package changed in every new major 
> JDK release (more methods added, but no change in existing methods). 
> In the particular case of Geotools, I agree that we sometime need to 
> change existing methods.
>
>     Martin


-------------------------------------------------------------------------
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
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel

Reply via email to