After looking into the debug code from vm .... it look like the problem may be existing around the zip loading code... Not 100% sure yet.. but when I swap the bootpath sequence e.g kissme:classpath will have different error then classpath:kissme.
Question1 -bootclasspath should be included kissme/classes.zip and classpath/glibj.zip correct? Alex Alex Lau: >OK this is my first try... let me know what's right and what's wrong. > >I ran my little script to test the java-bean test cases in mauve with >kissme, here is the debug. > ><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< >gnu.testlet.java.beans.Introspector.jdk11 >---- >PASS: gnu.testlet.java.beans.Introspector.jdk11: decapitalize (number 1) >PASS: gnu.testlet.java.beans.Introspector.jdk11: decapitalize (number 2) >PASS: gnu.testlet.java.beans.Introspector.jdk11: decapitalize (number 3) >PASS: gnu.testlet.java.beans.Introspector.jdk11: decapitalize (number 4) > >error initialising class gnu/java/beans/info/ComponentBeanInfo, >exception object 0x40b96ac0, type java/lang/UnknownError >error initialising class gnu/java/beans/info/ComponentBeanInfo, >exception object 0x40b96ac0, type java/lang/UnknownError, wrapped >exception 0x40b96cb8 ><<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<< > >Look into the classpath code, I find out the >gnu.java.beans.info.ComponentBeanInfo >extends the java.beans.SimpleBeanInfo which implemented the interface >java.beans.BeanInfo. However, java.beans.SimpleBeanInfo is missing the >constructer >and gnu.java.beans.info.ComponentBeanInfo is calling super(). And real >question is >why would jikes pass that? is that Correct? that a class implement an >interface doesn't >need a constructer? > >Alex > > >_______________________________________________ >Classpath mailing list >[EMAIL PROTECTED] >http://mail.gnu.org/mailman/listinfo/classpath > > _______________________________________________ Classpath mailing list [EMAIL PROTECTED] http://mail.gnu.org/mailman/listinfo/classpath
