Yes, the grammar is incorrect. Please file a bug (use you email text as
problem description) so we don't loose track of the problem and can
refer to the bug id in the release notes.
Thanks,
Lars
Peter Dobratz wrote:
>Peter Dobratz writes:
> >
> > Does anyone know what the LITERAL_THREADSAFE token is for? From the
> > name it would seem like it is a keyword called threadsafe, but there
> > is no such keyword. Am I missing something here?
>
>I couldn't find any reference to a threadsafe keyword in the JLS first
>and second editions. It's also not in the Oak spec. In order to
>investigate further, I constructed the following class:
>
>public class T
>{
> int threadsafe = 0;
>}
>
>This compiles on my Java 1.4.1 compiler from Sun. However, this
>generates a parser error while creating the AST.
>
>Figuring that threadsafe was intended to be some sort of modifier, I
>constructed the following example:
>
>public class S
>{
> public threadsafe void meth()
> {
>
> }
>}
>
>This does not compile, but it can be parsed into an AST. This would
>seem to indicate that the Java grammar is incorrect, although this
>problem probably won't come up in practice (standard naming
>conventions would make the identifier be threadSafe or ThreadSafe).
>
>--Peter
>
>
>
|