[xsd-users] Re: xsd-users Digest, Vol 35, Issue 9 Re: Code generation fails for schemas with large maxOccurs values (Boris Kolpackov)

[Gordon Kramer]

Hi Boris,

Am Mittwoch, den 14.05.2008, 09:22 +0200 schrieb Boris Kolpackov:

> While this is definitely an option, I think it is better to fix the 
> bug in Xerces-C++ since all XML Schema validation (for the C++/Tree 
> mapping) is done there. Doing some of it in Xerces-C++ and some of it
> in the generated code will make it cumbersome to enable/disable
> validation as well as report validation errors consistently.

I agree with both statements. Is it possible to validate the object
model using Xercesc facilities, without having to parse the xml again?

> http://www.codesynthesis.com/pipermail/xsd-users/2008-January/001443.html
> 
thanks for the insightful link.
answer to question 3:

> 3. If error correction by the application is hard/impossible then
>    what is the use of in-memory validation other then to know
>    whether the object model is valid/invalid?

The only application i have for validation at the moment is to check
whether the generated xml document is actually valid. 
This is enforced using serialisation followed by de-serialisation.
For my application this is fast enough at the moment. So in-memory
validation is not really an issue here. 
It would just be nice to call validate() before serialisation instead of
parsing the serialised xml again.
Also i can imagine using validate for simple types to reject input from
unreliable sources.

Maybe it would be an option to let xsd generate a Validator class which
provides functions to validate all types and global elements, maybe also
global attributes, groups ... in the schema using xercesc library (if
requested '--with-inmem-validator'). 
I don't actually know whether this is possible though.

Best Regards,
Gordon