That mapping is supposed to be part of the JAX-RPC standard. Read 'Appendix: Mapping of XML Names' on page 141-143 of the spec - you're not supposed to get control of the mapping.
However it looks like the mapping you've seen is a bug! It appears to be treating '_' as a punctuation char when its actually an identifier char. The bug is in 'JavaUtils.isPunctuation()' which has '_' identified as punctuation - its not in the table at the bottom of page 141 of the spec (in my copy anyway, which I believe is the final release?) Want to raise this in Bugzilla? -Baz [EMAIL PROTECTED] wrote: > Hello. > > I have the name FOO_Bar in typeMaping section. > something like that: > > .... > <complexType abstract="true" name="FOO_Bar"> > <sequence> > <element name="name" nillable="true" type="soapenc:string"/> > <element name="attribute" type="xsd:long"/> > <element name="valid" type="xsd:boolean"/> > </sequence> > </complexType> > .... > > During genereation process this name is converted to FOOBar. ( Both > generated java file(s) and deploy.wsdd). > > I figured out, that javifyNames() function JavaGeneratorFactory (that call > JavaUtils.xmlNameToJava()) > convert the xml QName FOO_Bar to FOOBar. Here is the stack trace: > > xmlNameToJava():649, JavaUtils.java > xmlNameToJava():325, Utils.java > xmlNameToJavaClass():333, Utils.java > getJavaName():369, Emitter.java > javifyNames():386, JavaGeneratorFactory.java > generatorPass():177, JavaGeneratorFactory.java > generate():266, Parser.java > access$000():80, Parser.java > run():235, Parser.java > run():484, Thread.java > > Why? I didn't find any command line paramter, to control this. > Any comments? > Thank you. > > David Ostrovsky > >
