Davanum Srinivas wrote:
Nice...So what is the delta required ("required changes are few")?
The required functionality requires two parts:
a) Create a class "BeanReader", or whatever, which uses JaxMeJS
reflection to inspect a given Java bean class. The bean readers
purpose is the creation of an equivalent XML schema, which can
then be used as input for JaxMe.
If we restrict the first version to those classes, which are
already supported by JaxMe (in other words: Java primitive types,
the corresponding object types, Calendar, lists, arrays, ...
as opposed to arbitrary collections and instances of Map), then
a first working version should be ready within 8 or 10 hours of
work. The other data types could be added later without too much
effort.
The created XML schema would also need to contain special
annotations, which advice JaxMe that
1) no interface is being used for the created classes and
2) rather than creating own bean classes, the original bean
class is being used
b) The rest of the job is adding support for 1) and 2). Setting up
an example, running the generator on it and looking for errors
in the compiled classes would easily show, what changes are
required. I believe, that this is more work than a), but another
20 or 25 hours should really do.
The main question is for me to find the time. However, I was looking
forward into work on JaxMe anyways (with different priorities, though),
since my proposal for XML-RPC 3 is out. In other words, about 30-40
hours of work is definitely possible within the next four weeks or so.
(In particular, because this job is far away from being as complex as
the support for arbitrary nested groups, which I planned to work on.)
Jochen
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
