On Wed, 2003-07-09 at 07:45, Andrew Stevens wrote: > I had another thought as to why hyphens are now a problem - could it be > related to the metadata JSR? More specifically, are hyphens valid > characters in java identifiers? I never use them myself, but can't > remember whether the Java Language Spec actually prohibits them (and > don't have it to hand at the moment);
Okay, now I do. According to section 3.8 (http://java.sun.com/docs/books/jls/second_edition/html/lexical.doc.html#40625) identifiers must consist of "Java Letters", i.e. characters for which Character.isJavaIdentifierPart returns true. That's not the case for hyphens, so they can't be used in identifiers. > if so, and the metadata JSR is > planning to use @tags for attributes, and they're going to force > metadata attribute @tags to be valid identifiers (same as > class/package/variable names), then perhaps javadoc is enforcing it now > so that people get used to it? > > Of course, that's a lot of 'if's, and could be absolutely nothing to do > with it. Does anyone know what's going on with the metadata JSR? Is it > in public review yet, or is it still only the expert group that know > what's planned? Although JSR 175's on the Java 1.5 roadmap (http://developer.java.sun.com/developer/technicalArticles/RoadMaps/J2SE_1.5/j2se_1_5.html), there's no public draft available for it yet. So it's still only the expert group that know anything. However, there's a couple of hints - the JSR itself (http://www.jcp.org/en/jsr/detail?id=175) says "The exact syntax will need to be determined by the expert group. There appear to be a number of possibilities, including (but not limited to!) using a doc comment tag @meta or adding a new Java programming language keyword meta." and the New Language Features interview (http://java.sun.com/features/2003/05/bloch_qa.html) with Joshua Bloch (the JSR's spec lead) has a sample that reads "public class CoffeeOrder { @Remote public Coffee [] getPriceList() { ... } @Remote public String orderCoffee(String name, int quantity) { ... } }" If that turns out to be accurate (i.e. they're going for extra keywords rather than doc comment tags), there's no reason to suppose javadoc @tags have to be identifiers and hence nothing to stop them containing hyphens. Guess I'm worrying about nothing, and we just have to wait for Sun to fix javadoc again :-( Still, at least we don't need to rewrite great chunks of the templates :-) Andrew. ------------------------------------------------------- This SF.Net email sponsored by: Parasoft Error proof Web apps, automate testing & more. Download & eval WebKing and get a free book. www.parasoft.com/bulletproofapps _______________________________________________ xdoclet-user mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/xdoclet-user
