Jody Garnett a écrit :
> Andrea Aime wrote:
>> Again, I don't know how many people will be affected
>> by changes in the changes of the metadata module, so I may be
>> overestimating the issue.
>>
> People do use the metadata classes directly; and have to make use of
> constructors to do it - so this change *will* break some users. However
> the groups I know making use of these interfaces forked away from
> GeoTools in order to make a copy of these classes backed on to Hibernate.
In GeoTools code base, the main metadata user is the referencing module, which
instantiate metadata objects for "domain of validity", "accuracy",
"identifiers"
and a few others.
Outside GeoTools, there is some codebase I'm involved with but we will take
care
of it. I'm not aware of other extensive usage.
But Jody is right, we need a factory. I commited myself to write it. However
the
drawback is that I do not expect a factory like the referencing or geometry
one,
with explicit "createCitation(...)" and ton of arguments, because we have ~100
metadata interfaces and ~400 attributes. A factory interface providing
type-safe
methods for all of them would be a monster. I rather expect a single method
with
a signature like this one:
<T> T create(Class<T> type, Map<String,?> attributes);
i.e. attributes values are given in a map. They would be dispatched to setter
methods using reflections.
Comments or suggestions welcome.
Martin
-------------------------------------------------------------------------
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services for
just about anything Open Source.
http://sourceforge.net/services/buy/index.php
_______________________________________________
Geotools-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-devel
