Java EE (Java Enterprise Edition) General Discussion

Hello,

I recently discovered a serious bug in the crimson implementation bundled with jdk 1.4.2_04 (guess it is also present in other jdk versions): If you call hasAttributes() on an element which does not have any attributes, after you called getAttributes() on that element, the call returns 'true' instead of 'false' !!!!
For example:

element.hasAttributes() // returns false
element.getAttributes()
element.hasAttributes() // returns true on the same element !!!


Does anybody know why the crimson parser is bungled with the jdk and not the xerces parser (which is, in my opinion, by far more reliable and also offers more features)? Is it just because sun donated the original crimson implementation to apache, and ibm donated xerces (or xml4j, as it was formerly named)? Why don't the sun guys see that the crimson parser is c.r.a.p?

Regards,
Dirk