Godmar Back wrote:
>
> A tool that's based on reflection seems pretty useless as a javap
> replacement. (It's useful for a lot of the things javap is
> used for, such as reverse engineering etc.) As a debugging tool,
> however, javap's purpose is to examine the actual content of a .class
> file, not necessarily the result of what was in a well-formed class file
> that passes the class loading process.
However, it's very useful for what the classpath project needs - that
is, a tool to compare the (public and protected) class and method
signatures of our libraries to the JDKs. Admittedly it isn't a full
replacement for javap, but it provides the (very important) subset of
the functionality that classpath needs.
Btw, is this tool going to provide the basis of the equivalent to sun's
"signature test" (that microsoft notoriously failed) in the clean-room
JCK equivalent?
Stuart.
