I'm complaining about roundtrip issues (as I usually do), but I'm also tired of axis-user folks constantly running into this problem. It would be nicer for them if we at least provided a mapping, even if we don't go all the way.
So if you DO do this work, please create ANOTHER bug detailing the shortcomings.
Russell Butek
[EMAIL PROTECTED]
Please respond to [EMAIL PROTECTED]
To: "'[EMAIL PROTECTED]'" <[EMAIL PROTECTED]>
cc:
Subject: Schema types
There is a bug outstanding (9966) that we do not support any of the unsigned XML Schema types. I was thinking of adding these in to the default type mapping.
What do you think is the lesser of two evils:
1. Add the types as their signed Java counter parts. i.e. xsd:unsignedShort would map to short, positiveInteger would map to int, etc.
2. Just leave them unsupported
Supporting them fully would enter a more complex area: We would have to emit a Bean for these types and enforce the value restrictions in the setter functions. This seems like a great deal of work for just a small amount of benefit.
Do we want to prevent users from consuming a WSDL that has some of these simple types because we can't enforce the value restrictions?
Opinions?
--
Tom Jordahl
Macromedia Server Development
