Elena,

[EMAIL PROTECTED] wrote:

>Simon,

>[EMAIL PROTECTED] wrote:
>>  We   want  to  be  able  to  programatically  analyse the content model
of a
>>  schema,  with  a view to creating e.g. an empty form where the user can
enter
>> a new instance of the document.
>> So  far,  the only route we have seen which looks likely to provide that
is
>> the DOM Level 3 Abstract Schema specification.

>Yes, this is correct.

>> The ASModel implementation in Xerces seems only just started.

>Yes, in Xerces we've only implemented a small portion of AS
>specification that allows preloading of XML Schemas.
>The reason, we have not started to implement AS-EDIT part is because we
>feel that AS model specification is not stable yet. Plus, at the moment
>we are more interested in exposing a read-only XML Schema components, to
>provide complete PSVI information.
>We plan to start working on implementing read-only access to schema
>components next month.

>> If so, there is the
>>  possibility    that   we   could   contribute   to  the  completion  of
the
>> implementation.

>People are always welcome to contribute! :)
>In this case, however, we definitely need to work together, since this
>implementation will involve changes to DOM and XML Schema/DTD
>implementations. We are still not sure whether we need to modify the
>core dom implementation or we can extend it and implement Abstract
>Schemas in a different package (available by option).

>Are you interested in read-only or read/write access? Have you started
>analyzing Xerces code?

It's  read-only  access  we're  interested  in at the moment.  Functions to
create schemas programatically are interesting for the medium term, but the
immediate need is read-only.
We  haven't  started  analysing  Xerces  code.   Do  you  have any expected
timescale  for  getting  something  up  and  running?   I would say we need
something  within 3 or 4 months.  There are realistically only two of us of
the  level  to  contribute I think.  If necessary, I could recommend a plan
committing  a  pecentage of one person's time for a certain duration, but I
would  have  to  work  hard  to  justify  that (my problem!).  Perhaps more
importantly,  the question is whether that would end up productive for you,
or  we  have  to spend too much time acclimatising to the Xerces codebase /
co-ordinating etc.

If  you  have  a  certain  timescale already in mind, that would affect the
variables  on  my side.  I like the idea of contributing, as I am a big fan
of  the  Jakarta project, and we get significant benefit, and we deploy the
software.  But while it may be a nice idea, I can see it may not overall be
the best solution, depending on a few factors.
I'm thinking about internal justification options anyway.

Best Regards,

Simon Godden
Research and Development Manager
DCS Transport and Logistics Solutions.


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to